CADTH Reimbursement Review

Trifluridine-Tipiracil (Lonsurf)

Sponsor: Taiho Pharma Canada, Inc.

Therapeutic area: Metastatic colorectal cancer

This multi-part report includes:

Clinical Review

Pharmacoeconomic Review

Stakeholder Input

Clinical Review

Executive Summary

An overview of the submission details for the drug under review is provided in Table 1.

Table 1: Background Information of Application Submitted for Review

Anti-EGFR = anti–epidermal growth factor receptor; anti-VEGF = anti–vascular endothelial growth factor; NOC = Notice of Compliance.

Introduction

Colorectal cancer (CRC) collectively refers to malignant tumours that develop in the epithelial lining of the rectum or colon from polyps that progress into cancer.¹ CRC is the third most prevalent cancer² and the second leading cause of cancer-related death (11% of all cancer deaths) in Canada.³ It is estimated that the Canadian (excluding Quebec) 10-year prevalence of CRC in both sexes of all ages is 343.5 cases per 100,000 population in 2018 (or 97,755 cases in total).⁴ Metastatic colorectal cancer (mCRC) indicates that the cancer has spread beyond the primary tumour site to other organs of the body (i.e., stage IV disease), where the most common location of metastases are the liver, lung, peritoneum, and distant lymph nodes.⁵ The stage of CRC at diagnosis is strongly associated with survival.⁶ Patients with early CRC are usually asymptomatic, whereas patients with advanced disease experience varying symptoms depending on the location of metastasis, including upper-right quadrant pain, abdominal distention, early satiety, supraclavicular adenopathy, and periumbilical nodules.⁷ Right-sided (proximal) tumours rarely present with obvious rectal bleeding as the blood becomes admixed with the stool. Left-sided (distal) tumours are more likely to present with bright red blood per rectum and symptoms of bowel obstruction.^8,9 The majority of patients with mCRC have unresectable (inoperable) disease, for which the mainstay of treatment is systemic multidrug chemotherapy.^10,11 The choice of treatment is dependent on a number of factors, including a patient’s fitness (e.g., performance status), organ function, and comorbidities, in addition to tumour characteristics (e.g., tumour location [right versus left], presence of primary tumour, mutation status for RAS and BRAF, presence of deficient mismatch repair [dMMR] or microsatellite instability–high [MSI-H]), type and timing of prior therapy, and toxicity profiles of constituent drugs.^5,10,12-15 Trifluridine-tipiracil (Lonsurf) and regorafenib (Stivarga) are approved in Canada for the treatment of patients who have been previously treated with fluoropyrimidine-, oxaliplatin-, and irinotecan-based chemotherapies, anti–vascular endothelial growth factor (anti-VEGF) biological agents, and, if RAS wild-type, anti–epidermal growth factor receptor (anti-EGFR) agents;^16,17 however, these treatments are not publicly funded in Canada, except in Quebec.^18,19 Following treatment with standard cytotoxic chemotherapy backbone regimens, patients are usually treated with best supportive care (BSC).

The objective of CADTH’s Clinical Review Report is to review and critically appraise the clinical evidence submitted by the sponsor on the beneficial and harmful effects of trifluridine-tipiracil (15 mg trifluridine and 6.14 mg tipiracil [as tipiracil hydrochloride] and 20 mg trifluridine and 8.19 mg tipiracil [as tipiracil hydrochloride]) plus bevacizumab for the treatment of mCRC in adults who have been previously treated with, or are not candidates for, available therapies, including fluoropyrimidine-, oxaliplatin-, and irinotecan-based chemotherapies, anti-VEGF biological agents, and, if RAS wild-type, anti-EGFR agents, which is an unlabelled indication. Trifluridine-tipiracil alone was previously reviewed by CADTH.

Stakeholder Perspectives

The information in this section is a summary of input provided by the patient and clinician groups who responded to CADTH’s call for input and from clinical experts consulted by CADTH for the purpose of this review.

Patient Input

CADTH received 2 patient group submissions from Colorectal Cancer Resource and Action Network (CCRAN) and Colorectal Cancer Canada. CCRAN used a multifaceted outreach approach by emailing clinicians who treat advanced CRC to help recruit patients or caregivers with experience with Lonsurf (in combination with bevacizumab) and via an online survey of patients’ experience of mCRC and prior drug therapies, resulting in 77 survey respondents (including 60 patients, 13 caregivers, and 4 patients who were also caregivers). Colorectal Cancer Canada conducted an online survey of 23 respondents (22 patients and 1 caregiver). Most patients reported that fatigue/weakness, bloody stools, diarrhea, and abdominal cramping/gas/feeling bloated and abdominal pain are common symptoms they experienced and that they felt were important to control. Symptoms of CRC affected quality of life for patients and their families, limiting the ability to work, to exercise, to participate in social activities, or to perform daily tasks. According to both patient groups, it is very important for a new therapy to bring about improvement to patients’ physical condition (e.g., tumour shrinkage, tumour stability, reduced pain, improved breathing) and quality of life (e.g., improved mobility, improved sense of wellness, relief from side effects). Patients would take a new therapy to bring about improvement in their quality of life even if it does not extend overall survival (OS) (e.g., at a modest 3 months to 4 months of survival, 53% of respondents were willing to tolerate significant side effects including nausea, anemia, and neutropenia). Moreover, patients prefer a drug therapy that is convenient (e.g., orally administered, either at home or with a short infusion duration or chair time at a cancer centre). CCRAN believes that if publicly funded, trifluridine-tipiracil plus bevacizumab would be an extremely important third-line and beyond therapy for patients whose disease has been deemed to be refractory or ineligible for standard of care (SOC) therapies. Colorectal Cancer Canada noted that, given that Lonsurf alone is currently reimbursed only in Quebec, there is a strong need for equity of access for patients located elsewhere in Canada. Both patient groups strongly agree that trifluridine-tipiracil aligns well with the identified patient and caregiver need for a new effective treatment option that is capable of prolonging life and maintaining quality of life.

Clinician Input

Input From Clinical Experts Consulted by CADTH

Two clinical experts with expertise in the diagnosis and management of mCRC reported that the cornerstone of treatment for patients with mCRC involves sequential use of the best available systemic therapies. SOC first-line treatment in Canada includes pembrolizumab immunotherapy (for patients with dMMR or MSI-H mCRC); chemotherapy with a regimen of folinic acid, fluorouracil, and oxaliplatin (FOLFOX) or a regimen of folinic acid, fluorouracil, and irinotecan (FOLFIRI) in combination with an epidermal growth factor receptor (EGFR) inhibitor (for patients with left-sided, extended RAS wild-type CRC); and chemotherapy with FOLFOX or FOLFIRI in combination with bevacizumab (for patients with right-sided or extended RAS-mutated CRC). Patients who progress on or within 6 months of adjuvant therapy (e.g., cancer growth while on adjuvant therapy or within 6 months of adjuvant FOLFOX) would be considered to have experienced progression on first-line treatment. Following disease progression on first-line therapy, the clinical experts consulted by CADTH indicated that SOC second-line systemic treatment in Canada includes encorafenib plus cetuximab (for patients with BRAF V600E mutations) or the switching of the backbone chemotherapeutic regimen (for patients without a BRAF V600E mutation) such that patients who were initially treated with FOLFOX would then be switched to FOLFIRI, for example. Antiangiogenic therapies added to the chemotherapy backbone (e.g., bevacizumab, aflibercept, ramucirumab) for patients without BRAF V600E mutation or dual immunotherapy (e.g., nivolumab plus ipilimumab) for patients with dMMR or MSI-H molecular marker are routinely offered to patients with CRC and recommended in guidelines for CRC, according to the clinical experts consulted by CADTH. The clinical experts consulted by CADTH noted that following disease progression on 2 lines of prior therapy, a single-drug EGFR inhibitor (cetuximab or panitumumab) or cetuximab plus irinotecan as SOC treatment in Canada is an option for patients with extended RAS wild-type, whereas regorafenib monotherapy or trifluridine-tipiracil is SOC in Canada for patients without the extended RAS wild-type marker (among patients with access through private insurance or who pay out of pocket). Importantly, there exists a significant unmet need for effective treatment options for patients with mCRC who experience disease progression following 2 lines of anticancer therapy, according to the clinical experts consulted by CADTH.

The clinical experts consulted by CADTH considered trifluridine-tipiracil plus bevacizumab to represent a new SOC treatment for patients with unresectable CRC after progression on 2 prior lines of anticancer therapy. According to the clinical experts consulted by CADTH, eligible patients should be able to tolerate both trifluridine-tipiracil (i.e., able to safely swallow pills, have normal bowel transit, have an Eastern Cooperative Oncology Group Performance Status [ECOG PS] score of 0 to 1, and have adequate hematologic, hepatic, and renal function) and bevacizumab (i.e., without absolute contraindication to the use of a vascular endothelial growth factor [VEGF] inhibitor, including but not limited to uncontrolled hypertension, in situ colonic stent, recent surgery, a high risk for bleeding, and a risk for or presence of fistula or gastrointestinal tract perforation). The clinical experts consulted by CADTH outlined the following hierarchy for determining treatment response: 1) patient-reported symptoms or side effects, as determined by clinician assessment of patient treatment history, 2) examination and selective use of clinical instruments to evaluate symptoms (e.g., Edmonton Symptom Assessment System, EQ-5D), and 3) cross-sectional imaging (e.g., CT scan, MRI) and tumour markers (e.g., carcinoembryonic antigen [CEA], CA 19 to 9). Patients should be assessed after every 2 cycles to 3 cycles of treatment (and more frequently with bothersome symptoms or adverse events [AEs]), with tumour markers completed at least once every 4 weeks and CT scans conducted every 2 months to 3 months, according to the clinical experts consulted by CADTH. The clinical experts consulted by CADTH highlighted OS, symptom control, and quality of life as clinically meaningful end points. Side effects or toxicity were key determinants for discontinuing treatment with trifluridine-tipiracil plus bevacizumab, according to the clinical experts consulted by CADTH, particularly for discontinuing bevacizumab in the event of the development of an absolute contraindication to further therapy with a VEGF inhibitor. The clinical experts consulted by CADTH highlighted the importance of shared and fully informed decision-making with patients that includes discussions regarding treatment effectiveness and symptoms or AEs that significantly impact quality of life.

Clinician Group Input

CADTH received 2 clinician group submissions from the Canadian Gastrointestinal Oncology Evidence Network (CGOEN) with the Medical Advisory Board of Colorectal Cancer Canada (and other Colorectal Cancer Canada treating physicians) and the Ontario Health (Cancer Care Ontario) (OH-CCO) Gastrointestinal Cancer Drug Advisory Committee. CGOEN gathered data and information based on personal experience in treating patients with mCRC and expert evidence-based review by gastrointestinal cancer specialists in Canada of the following information presented at international oncology meetings and subsequently published in The New England Journal of Medicine; OH-CCO’s Drug Advisory Committees gathered information through videoconferencing and email communication. Both clinician groups highlighted that trifluridine-tipiracil would be placed as a further line of therapy and would be used in patients who received current SOC options and have experienced disease progression, have experienced intolerance, or have chosen to stop for personal reasons. This combination would also be used for those with medical contraindications to earlier-line SOC therapies. CGOEN stated that trifluridine-tipiracil is currently Health Canada–approved but received a do not reimburse recommendation in August 2019 from CADTH due to the fact that the magnitude of benefit was felt to be too small to warrant approval, despite being recognized as addressing the needs of a population with an unmet need. It is currently funded in Quebec, having received a reimburse recommendation from the Institut national d’excellence en santé et en services sociaux. Outside Quebec, patients have been able to apply to the sponsor for access to the drug under review through private insurance or direct user pay. Therefore, the majority of patients with mCRC in Canada do not have access to a publicly funded version of the drug under review, according to CGOEN. OH-CCO’s Drug Advisory Committees also echoed this concern highlighted by CGOEN. Therefore, CGOEN felt that findings from the original trial of trifluridine-tipiracil alone compared to BSC should be considered in the current review of trifluridine-tipiracil plus bevacizumab, given the current landscape in Canada.

Drug Program Input

Input was obtained from the drug programs that participate in the CADTH reimbursement review process. The following were identified as key factors that could potentially affect the implementation of a CADTH recommendation for trifluridine-tipiracil plus bevacizumab: relevant comparators, generalizability, a funding algorithm, care provision issues, and system and economic issues.

The clinical experts consulted by CADTH provided advice on the potential implementation issues raised by the drug programs. Refer to Table 5 for more details.

Clinical Evidence

Systematic Review

Description of Studies

One randomized, phase III, open-label, multicentre study (the SUNLIGHT trial) evaluated the efficacy and safety of trifluridine-tipiracil plus bevacizumab versus trifluridine-tipiracil alone. The SUNLIGHT study enrolled 492 adults with advanced mCRC who had received up to 2 previous chemotherapy regimens and demonstrated progressive disease or intolerance to their last regimen, and randomized patients to each group with stratification by geographic region (North America, European Union, the rest of the world), time since first metastasis diagnosis (< 18 months, ≥ 18 months), and RAS status (wild-type, mutant). The primary objective of the SUNLIGHT study was to demonstrate the superiority of OS and the key secondary objective was to estimate investigator-assessed progression-free survival (PFS). Additional secondary end points included health-related quality of life (HRQoL) assessed with the European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire Core 30 (EORTC QLQ-C30) and 5-Level EQ-5D (EQ-5D-5L), and treatment-emergent adverse events (TEAEs).

Patients had a mean age of 61.7 years (standard deviation [SD] = 11.1 years), and most were enrolled from the European Union (64.0%). Most patients had a primary diagnosis of colon cancer (73%) and stage IV disease (66%), and a primary tumour located on the left side (72%). The time from the diagnosis of the first metastasis until randomization was 18 months or longer in 57.5% of the patients, and 30.7% of patients had RAS wild-type disease. Most patients (92.1%) had received 2 previous treatment regimens for metastatic disease, 2.6% of patients had received more than 2 prior regimens, and 5.3% of patients had received 1 previous treatment regimen. All patients had received previous fluoropyrimidine-based therapy, 72.0% of patients had received previous anti-VEGF therapy (47.8% of patients had received bevacizumab as part of their first regimen, 43.9% of patients had received bevacizumab as part of their second regimen, and 20.3% of patients had received bevacizumab as part of both their first and second regimens), and 93.7% of the patients with RAS wild-type disease had received previous anti-EGFR therapy. Demographic characteristics were generally similar between the trifluridine-tipiracil plus bevacizumab group and the trifluridine-tipiracil alone group, with notable (> 5%) between-group differences for patients aged 65 years and older (41% versus 48%, respectively), with primary tumour located on the right side (25% versus 31%, respectively), and with primary tumour located on the left side (75% versus 69%, respectively).

Efficacy Results

The key efficacy results from the SUNLIGHT trial are summarized in Table 2, based on the data cut-off date of July 5, 2022, for clinical (nonsurvival) data and July 19, 2022, for survival data.

Overall Survival

At the survival cut-off date of July 19, 2022, the median follow-up was 14.2 months (interquartile range = 12.6 months to 16.4 months) in the trifluridine-tipiracil plus bevacizumab group and 13.6 months (interquartile range = 12.7 months to 15.9 months) in the trifluridine-tipiracil alone group. OS at 6 months among patients in the full analysis set (FAS) population was 0.77 (95% confidence interval [CI], 0.72 to 0.82) and 0.61 (95% CI, 0.55 to 0.67) for the trifluridine-tipiracil plus bevacizumab group and the trifluridine-tipiracil alone group, respectively. OS at 12 months was 0.43 (95% CI, 0.36 to 0.49) and 0.30 (95% CI, 0.24 to 0.36) for the trifluridine-tipiracil plus bevacizumab group and the trifluridine-tipiracil alone group, respectively. The median OS was 10.78 months (95% CI, 9.36 months to 11.83 months) in the trifluridine-tipiracil plus bevacizumab group and 7.46 months (95% CI, 6.34 months to 8.57 months) in the trifluridine-tipiracil alone group. The hazard ratio (HR) in the FAS population was 0.61 (95% CI, 0.49 to 0.77; P < 0.001) for the trifluridine-tipiracil plus bevacizumab group when compared with the trifluridine-tipiracil alone group.

Progression-Free Survival

PFS at 3 months among patients in the FAS population was 0.73 (95% CI, 0.67 to 0.78) in the trifluridine-tipiracil plus bevacizumab group versus 0.45 (95% CI, 0.39 to 0.51) in the trifluridine-tipiracil alone group. PFS at 6 months was 0.43 (95% CI, 0.37 to 0.49) and 0.16 (95% CI, 0.11 to 0.21) for the trifluridine-tipiracil plus bevacizumab group and the trifluridine-tipiracil alone group, respectively. Median PFS was 5.6 months (95% CI, 4.50 months to 5.88 months) in the trifluridine-tipiracil plus bevacizumab group and 2.4 months (95% CI, 2.07 months to 3.22 months) in the trifluridine-tipiracil alone group. The HR for PFS was 0.44 (95% CI, 0.36 to 0.54; P < 0.001) for the trifluridine-tipiracil plus bevacizumab group when compared with the trifluridine-tipiracil alone group.

Health-Related Quality of Life

In the SUNLIGHT trial, analyses for the EORTC QLQ-C30 and EQ-5D-5L were performed in patients from the FAS with at least 1 questionnaire item at baseline and during the study period. Higher scores in the EORTC QLQ-C30 global health status and EQ-5D- 5L utility and the EQ visual analogue scale (EQ VAS) indicated better HRQoL, with positive change from baseline indicating benefit and negative change from baseline indicating deterioration.

In the global health status score, the least squares mean (LSM) change from baseline was –2.85 (95% CI, –5.92 to 0.22) for the trifluridine-tipiracil plus bevacizumab group and –6.62 (95% CI, –10.36 to –2.88) for the trifluridine-tipiracil alone group. The LSM difference in change from baseline for global health status was 3.77 (95% CI, 0.22 to 7.32; P = 0.038) in favour of the trifluridine-tipiracil plus bevacizumab group. The number of patients in the FAS population with 10 points or greater definitive deterioration was 62 (25.2%) patients and 72 (29.3%) patients in the trifluridine-tipiracil plus bevacizumab group and the trifluridine-tipiracil alone group, respectively. The median time until definitive deterioration in the global health status was 8.54 months (95% CI, 7.49 months to 10.94 months) in the trifluridine-tipiracil plus bevacizumab group and 4.70 months (95% CI, 4.01 months to 5.78 months) in the trifluridine-tipiracil alone group (P < 0.001).

In the EQ-5D-5L utility, the LSM change from baseline was –0.01 (95% CI, –0.03 to 0.01) for the trifluridine-tipiracil plus bevacizumab group and –0.03 (95% CI, –0.06 to –0.01) for the trifluridine-tipiracil alone group. The LSM difference in change from baseline for the EQ-5D-5L utility was 0.02 (95% CI, 0.00 to 0.05; P = 0.070). In the EQ VAS, the LSM change from baseline was –0.87 (95% CI, –3.74 to 2.00) for the trifluridine-tipiracil plus bevacizumab group and –5.34 (95% CI, –8.75 to –1.92) for the trifluridine-tipiracil alone group. The LSM difference in change from baseline for the EQ VAS was 4.46 (95% CI, 1.11 to 7.81; P = 0.009).

Harms Results

The analysis population for harms included all patients who received at least 1 dose of trifluridine-tipiracil, with patients grouped according to the treatment received. Safety data were performed using the clinical data cut-off date of July 5, 2022.

In the SUNLIGHT study, the number of patients reporting any TEAEs was 98.0% for the trifluridine-tipiracil plus bevacizumab group and 98.0% for the trifluridine-tipiracil alone group. The most common TEAEs occurring in at least 20% of patients in the trifluridine-tipiracil plus bevacizumab group and the trifluridine-tipiracil alone group were neutropenia (62.2% versus 51.2%, respectively), nausea (37.0% versus 27.2%, respectively), anemia (28.9% versus 31.7%, respectively), asthenia (24.4% versus 22.4%, respectively), fatigue (21.5% versus 16.3%, respectively), diarrhea (20.7% versus 18.7%, respectively), and decreased appetite (20.3% versus 15.4%, respectively).

The proportion of patients who experienced at least 1 serious adverse event (SAE) was 24.8% in the trifluridine-tipiracil plus bevacizumab group and 31.3% in the trifluridine-tipiracil alone group. SAEs occurring in at least 2% of patients in the trifluridine-tipiracil plus bevacizumab group and the trifluridine-tipiracil alone group were intestinal obstruction (2.8% versus 2.0%, respectively), malignant neoplasm progression (2.4% versus 4.5%, respectively), COVID-19 (2.0% versus 2.4%, respectively), anemia (0.4% versus 3.3%, respectively), febrile neutropenia (0.4% versus 2.4%, respectively), jaundice (0.8% versus 2.0%, respectively), and hepatic failure (0 versus 2.0%, respectively). The proportion of patients who experienced AEs of grade 3 or greater were 72.4% in the trifluridine-tipiracil plus bevacizumab group and 69.5% in the trifluridine-tipiracil alone group. The most common AEs of grade 3 or greater occurring in at least 5% of patients in the trifluridine-tipiracil plus bevacizumab group and the trifluridine-tipiracil alone group were neutropenia (43.1% versus 32.1%, respectively), anemia (6.1% versus 11.0%, respectively), decreased neutrophil count (8.9% versus 5.3%, respectively), and hypertension (5.7% versus 1.2%, respectively).

A total of 12.6% of patients experienced TEAEs that led to treatment withdrawal in each treatment group. Withdrawals due to AEs occurring in at least 1 patient in the trifluridine-tipiracil plus bevacizumab group and the trifluridine-tipiracil alone group were asthenia (3.3% versus 0.4%, respectively), jaundice (0.8% versus 0.8%, respectively), decreased appetite (0.8% versus 0.4%, respectively), fatigue (0.4% versus 0.8%, respectively), anemia (0.4% versus 0.8%, respectively), intestinal obstruction (0.4% versus 0.8%, respectively), malignant neoplasm progression (0.4% versus 0.8%, respectively), biliary dilation (0.8% versus 0, respectively), increased blood bilirubin (0.8% versus 0, respectively), pain (0.8% versus 0, respectively), and metastases to central nervous system (0 versus 0.8%, respectively).

|| ||| |||||||| ||||||| |||||||||| || ||| |||||||| ||| ||||| ||||||||| ||||| || |||||||| || ||| |||||||||||||||||||||| |||| ||||||||||| ||||| ||| ||||| || |||||||| || ||| |||||||||||||||||||||| ||||||| |||||||||| || || |||||||||||| ||| ||||||||| |||||| |||||||| || || |||||| |||||||| ||| || |||||| |||||||| || ||| |||||||||||||||||||||| |||| ||||||||||| ||| |||||||||||||||||||||| ||||||| |||||| ||||||||||||| |||||| ||||| |||||||| |||||| |||||||||||| |||||| |||||| ||| |||||| |||| |||||| ||| || ||||||||||| ||||||| || ||| |||||||||||||||||||||| |||| ||||||||||| ||| |||||||||||||||||||||| ||||||| |||||| ||||||||||||.

Notable Harms

Analyses of notable harms in the SUNLIGHT trial were conducted post hoc using lists of predefined preferred terms with similar medical concepts to define the overall terms. The proportion of patients who experienced bone marrow suppression events was 80.9% in the trifluridine-tipiracil plus bevacizumab group and 73.2% in the trifluridine-tipiracil alone group, including neutropenia (62.2% versus 51.2%, respectively), anemia (28.9% versus 31.7%, respectively), thrombocytopenia (17.1% versus 11.4%, respectively), and leukopenia (6.5% versus 8.5%, respectively). The proportion of patients who experienced at least 1 TEAE related to infections was 30.9% in the trifluridine-tipiracil plus bevacizumab group and 23.2% in the trifluridine-tipiracil alone group. Infections of grade 3 or higher were reported for 7.7% of patients and 7.3% of patients in the trifluridine-tipiracil plus bevacizumab group and the trifluridine-tipiracil alone group, respectively. The proportion of patients who experienced gastrointestinal symptoms was 48.4% in the trifluridine-tipiracil plus bevacizumab group and 41.1% in the trifluridine-tipiracil alone group, including nausea (37.0% versus 27.2%, respectively), diarrhea (20.7% versus 18.7%, respectively), and vomiting (18.7% versus 14.6%, respectively). Gastrointestinal symptoms of grade 3 or higher were reported for 2.0% of patients and 4.9% of patients in the trifluridine-tipiracil plus bevacizumab group and the trifluridine-tipiracil alone group, respectively, including nausea (1.6% versus 1.6%, respectively), diarrhea (0.8% versus 2.4%, respectively), and vomiting (0.8% versus 1.6%, respectively). The proportion of patients who experienced hypertension was 10.2% in the trifluridine-tipiracil plus bevacizumab group and 2.0% in the trifluridine-tipiracil alone group. Hypertension events of grade 3 or higher were reported for 5.7% of patients and 1.2% of patients in the trifluridine-tipiracil plus bevacizumab group and the trifluridine-tipiracil alone group, respectively.

Critical Appraisal

The SUNLIGHT study was a phase III, open-label, randomized controlled trial (RCT) that used stratified randomization that appeared to be appropriate as patients were generally balanced between treatment groups for key prognostic factors, disease characteristics, and prior chemotherapy regimens. The open-label study design has the potential to impact HRQoL, for which knowledge of the assigned treatment may bias reporting in favour of the intervention (trifluridine-tipiracil plus bevacizumab) group. Trifluridine-tipiracil alone was the comparator used in the SUNLIGHT trial. Trifluridine-tipiracil is approved and available in Canada but is not universally publicly funded, so the majority of patients must gain access via private drug coverage or out-of-pocket costs. OS as the primary end point and PFS as a key secondary end point were included in statistical hierarchical testing and were appropriate key end points according to treatment guidelines and outcomes identified as important by patients and clinicians. Findings for OS and PFS demonstrated a benefit for patients treated with trifluridine-tipiracil plus bevacizumab; the proportional hazards assumption was likely valid based on Schoenfeld residuals testing and visual inspection of the Kaplan-Meier (KM) curves and log(-log) curves showing crossover early during treatment but clear separation thereafter. For HRQoL, MIDs were identified in the literature among patients with cancer and with mCRC for the cancer-specific EORTC QLQ-C30 tool, and among patients with cancer for the generic preference-based EQ-5D-5L tool. It was unclear whether significant missing data for HRQoL by cycle 3 to cycle 4 could impact findings. Longer treatment duration and a higher mean dose of trifluridine-tipiracil in the trifluridine-tipiracil plus bevacizumab group may not be fully explained by the relatively small difference in treatment discontinuations between groups and it is unknown whether the open-label study design may have impacted patients’ adherence to assigned treatment.

The enrolled population in the SUNLIGHT trial was generally aligned with patients seen in clinical practice according to the clinical experts consulted by CADTH, despite there being no patients in Canada enrolled in the trial. Patients who were excluded from eligibility (those with more than 2 prior chemotherapy regimens, those who had prior treatment with trifluridine-tipiracil, or those with an ECOG PS score greater than 1) were considered by the clinical experts consulted by CADTH to be eligible for treatment with trifluridine-tipiracil plus bevacizumab. The clinical experts consulted by CADTH also considered patients with small bowel or appendiceal adenocarcinoma as eligible to be treated with trifluridine-tipiracil plus bevacizumab based on the small number of patients that precludes a trial enrolling patients exclusively in this subpopulation. While the clinical experts consulted by CADTH noted a higher proportion of patients with RAS status expressing mutations (compared with wild-type), the key prognostic indicators (i.e., age, number of metastatic sites, number of prior chemotherapy regimens, sidedness of tumour, and ECOG PS) appeared to be reflective of patients in clinical practice. The intervention in the SUNLIGHT trial is for an unlabelled indication, as trifluridine-tipiracil alone was approved by Health Canada for adult patients with mCRC but is not universally publicly funded. Acknowledging that this treatment is only available to a small patient population with access (via private insurance or self-funding) among other treatment options (including BSC and regorafenib, the latter of which is available via compassionate access), the clinical experts consulted by CADTH emphasized that trifluridine-tipiracil alone is the most relevant comparator for trifluridine-tipiracil plus bevacizumab. Outcomes included in the SUNLIGHT trial were identified as important to patients and clinicians, including survival, HRQoL, and TEAEs. OS at 6 months and 12 months was highlighted by the clinical experts consulted by CADTH as important for assessing the effects of treatment. Furthermore, PFS (at 3 months and 6 months) was an appropriate end point as supportive evidence for OS. Findings may be limited in generalizability to patients with mCRC in Canada for EQ-5D-5L health utility values derived using a French value set and in the absence of patients enrolled from sites in Canada. A higher proportion of patients who discontinued treatment in the trifluridine-tipiracil alone group was not concerning to the clinical experts consulted by CADTH as they noted that the proportions were low with similar between-group rates for discontinuations due to AEs and deaths.

GRADE Summary of Findings and Certainty of the Evidence

For pivotal studies and RCTs identified in the sponsor’s systematic review, the Grading of Recommendations Assessment, Development and Evaluation (GRADE) tool was used to assess the certainty of the evidence for outcomes considered most relevant to inform CADTH’s expert committee deliberations, and a final certainty rating was determined as outlined by the GRADE Working Group.^20,21 Following the GRADE approach, evidence from RCTs started as high-certainty evidence and could be rated down for concerns related to study limitations (which refers to internal validity or risk of bias), inconsistency across studies, indirectness, the imprecision of effects, and publication bias.

The selection of outcomes for GRADE assessment was based on the sponsor’s Summary of Clinical Evidence, on consultation with clinical experts, and on input received from patient and clinician groups and public drug plans. The following list of outcomes was finalized in consultation with expert committee members: survival (OS and PFS), HRQoL (measured as LSM change from baseline and the proportion of patients with a 10-point or greater deterioration from baseline in the EORTC QLQ-C30 global health status, and LSM change from baseline in the EQ-5D-5L utility score and EQ VAS), and harms (bone marrow suppression, infections, gastrointestinal symptoms, and hypertension).

When possible, the certainty was rated in the context of the presence or absence of an important (nontrivial) treatment effect; if this was not possible, certainty was rated in the context of the presence of any treatment effect (i.e., the clinical importance was unclear). In all cases, the target of the certainty of evidence assessment was based on the point estimate and where it was located relative to the threshold for a clinically important effect (when a threshold was available) or to the null. The target of the certainty of evidence assessment was the presence or absence of an important effect based on thresholds informed by the clinical experts consulted for this review for survival (OS and PFS), HRQoL (EORTC QLQ-C30 and EQ-5D-5L), and harms (bone marrow suppression, infections, gastrointestinal symptoms, and hypertension).

Table 2: Summary of Findings for Trifluridine-Tipiracil Plus Bevacizumab Versus Trifluridine-Tipiracil Alone for Patients With mCRC

CI = confidence interval; EORTC QLQ-C30 = European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire Core 30; EQ-5D-5L = 5-Level EQ-5D; EQ VAS = EQ visual analogue scale; HRQoL = health-related quality of life; LSM = least squares mean; mCRC = metastatic colorectal cancer; MID = minimal important difference; NA = not applicable; NR = not reported; RCT = randomized controlled trial; RR = risk ratio; vs. = versus.

Notes: Details included in Table 2 were provided from the sponsor in response to additional data request.²³

Study limitations (which refer to internal validity or risk of bias), inconsistency across studies, indirectness, the imprecision of effects, and publication bias were considered when assessing the certainty of the evidence. All serious concerns in these domains that led to the rating down of the level of certainty are documented in the table footnotes.

^aRated down 1 level for serious imprecision. There is no established MID. The clinical experts consulted by CADTH indicated that a between-group difference of 10% to 20% was clinically important. The lower bound of the 95% CI for difference between groups did not reach the identified threshold.

^bThere is no established MID. The clinical experts consulted by CADTH indicated that a between-group difference of 20% was clinically important.

^cRated down 1 level for serious imprecision. There is no established MID. The clinical experts consulted by CADTH indicated that a between-group difference of 20% was clinically important. The lower bound of the 95% CI for difference between groups did not reach the identified threshold.

^dA mixed model of repeated measures that included terms for treatment, baseline stratification factors, baseline score, time to visit before any procedure (at each cycle, including the withdrawal visit), and treatment groups by time to visit interaction was used to compare change from baseline subscale scores longitudinally (cycle 1 to cycle 10) over time between treatment groups.

^eRated down 2 levels for very serious study limitations. The open-label study design and patients’ and caregivers’ knowledge of assigned treatment may have biased reporting of HRQoL questionnaires. There was substantial missing data from treatment cycle 1 to cycle 10 that may have impacted the prognostic balance of the treatment groups. Rated down 1 level for serious imprecision. An MID of 5.53 to 6.36 (weighted = 5.86) for improvement, and –9.21 to –6.81 (weighted = –8.13) for deterioration was identified in the literature. The point estimate suggests little to no difference and the 95% CI included the possibility of important benefit. Statistical testing for EORTC QLQ-C30 was not conducted; therefore, results are considered as supportive evidence.

^fRated down 2 levels for very serious study limitations. The open-label design and patients’ and caregivers’ knowledge of assigned treatment may have biased reporting of HRQoL questionnaires. There was substantial missing data from cycle 1 to cycle 10 that may have impacted the prognostic balance of treatment groups. Rated down 1 level for serious imprecision. The clinical experts consulted by CADTH indicated that a between-group difference of 10% was clinically important. The lower bound of the 95% CI for difference between groups included possible important benefit. Statistical testing for EORTC QLQ-C30 were not conducted; therefore, results are considered as supportive evidence.

^gRated down 2 levels for very serious study limitations. The open-label study design and patients’ and caregivers’ knowledge of assigned treatment may have biased the reporting of HRQoL questionnaires. There was substantial missing data across and up to treatment cycle 10 that may have impacted the prognostic balance of the treatment groups. Rated down 1 level for serious indirectness due to utility values that were derived from a French population set. No MID was identified in the literature for patients with mCRC. An MID of 0.08 based on literature for patients with cancer was identified by the clinical experts consulted by CADTH. Statistical testing for EQ-5D-5L was not conducted; therefore, results are considered as supportive evidence.

^hRated down 2 levels for serious study limitations. The open-label study design and patients’ and caregivers’ knowledge of assigned treatment may have biased the reporting of HRQoL questionnaires. There was substantial missing data across and up to treatment cycle 10 that may have impacted the prognostic balance of the treatment groups. No MID was identified in the literature for patients with mCRC. An MID of greater than 7 points to 10 points was identified by the clinical experts consulted by CADTH, based on literature for patients with cancer. Statistical testing for EQ-5D-5L was not conducted; therefore, results are considered as supportive evidence.

ⁱRated down 1 level for post hoc analyses of adverse events of risk of bias in the selection of outcomes reported in the results. Rated down 1 level for serious imprecision. The clinical experts consulted by CADTH indicated that a between-group difference of 10% was clinically important. The point estimate suggests little to no difference and the 95% CI included the possibility of important harm.

Source: SUNLIGHT Clinical Study Report.²²

Long-Term Extension Studies

No long-term extension studies were submitted in the systematic review evidence.

Indirect Comparisons

Description of Studies

The sponsor submitted a systematic review and indirect treatment comparison (ITC) report where trifluridine-tipiracil plus bevacizumab was compared to BSC, regorafenib, and trifluridine-tipiracil alone among patients with mCRC who had been previously treated with, or were not considered candidates for, available therapies, including fluoropyrimidine-, oxaliplatin-, and irinotecan-based chemotherapies, anti-VEGF agents, and anti-EGFR agents.²⁴

In this ITC, OS, PFS, and treatment-related AEs (defined as any AEs that were considered related to treatment received) were assessed. The network meta-analyses (NMAs) were conducted within a Bayesian framework.

In total, 10 RCTs were included and contributed evidence. These studies were conducted in Asia, North America, South America, and Europe. There was no information as to whether patients living in Canada were enrolled. The mean age of patients ranged from 55.5 years to 67 years. The proportion of male patients ranged from 48.5% to 64.8%. These studies were published between 2007 and 2023. The included RCTs evaluated the efficacy and safety of the following therapies, which are relevant to this review: trifluridine-tipiracil plus bevacizumab in 2 studies, BSC alone in 7 studies, regorafenib in 2 studies, and trifluridine-tipiracil alone in 6 studies.

Efficacy Results

Based on the results of the sponsor-submitted ITC, the treatment of trifluridine-tipiracil plus bevacizumab may be associated with prolonged OS and PFS in patients with mCRC compared to other treatments such as BSC, regorafenib, or trifluridine-tipiracil alone.

Harms Results

The treatment of trifluridine-tipiracil plus bevacizumab may be associated with an increased risk of treatment-related AEs in patients with mCRC compared to other treatments such as BSC, regorafenib, or trifluridine-tipiracil alone. However, results of the NMA for treatment-related AEs were imprecise with wide credible intervals (CrIs).

Critical Appraisal

In the sponsor-submitted ITC, based on the data presented, potential sources of heterogeneity with respect to the patients’ characteristics were identified, such as ECOG PS (the proportion of patients with an ECOG PS score of 0 ranged from 22% to 64%) and RAS status (the proportion of patients with RAS-positive status ranged from 27% to 70%) at baseline. Heterogeneities in trial characteristics were observed in study design (such as blinding, the definition of BSC across trials, and prior lines of therapies). Despite various statistical models being employed to lessen the impact of potential clinical heterogeneity on the estimated comparative treatment effect of trifluridine-tipiracil plus bevacizumab, there remains significant uncertainty in the ITC results. In addition, given the lack of closed loops in any of the networks, consistency in the ITC analyses could not be tested. All comparisons were therefore informed only by indirect evidence, which increased the level of uncertainty.

Some important patient characteristics in the included trials were not reported in this ITC, such as treatment duration, the timing of study end point evaluation, the use of subsequent therapies after disease progression, and the length of follow-up. Therefore, adjustment for their potential treatment-effect modification was not feasible, and it is likely that the transitivity assumption was not met (the transitivity assumption is the assumption that if treatment A is preferred to treatment B and treatment B is preferred to treatment C, then treatment A is preferred to treatment C). Furthermore, it is unclear whether the results can provide insight into the long-term effect of the study drug for patients with mCRC due to a lack of data regarding the length of trial follow-up.

Outcomes other than OS and PFS that are important to the patients and clinicians (e.g., HRQoL) were not analyzed in the ITC. A more comprehensive assessment on the safety profile of the study drug is desired.

Studies Addressing Gaps in the Evidence From the Systematic Review

No additional studies addressing important gaps in the systematic review evidence were submitted.

Conclusion

The SUNLIGHT study was a randomized, phase III, open-label trial in adults with mCRC who had been previously treated with, or were not candidates for, available therapies, including fluoropyrimidine-, oxaliplatin-, and irinotecan-based chemotherapies, anti-VEGF biological agents, and, if RAS wild-type, anti-EGFR agents. Evidence from the SUNLIGHT study demonstrated that trifluridine-tipiracil plus bevacizumab likely results in a clinically important increase in the proportion of patients with OS at 6 months and 12 months (moderate certainty). The evidence demonstrated that trifluridine-tipiracil plus bevacizumab results in a clinically important increase in the proportion of patients with PFS at 3 months (high certainty) and likely results in a clinically important increase in the proportion of patients with PFS at 6 months (moderate certainty) when compared with trifluridine-tipiracil alone. Trifluridine-tipiracil plus bevacizumab may result in little to no clinically important difference in the change from baseline in EQ VAS (low certainty) when compared with trifluridine-tipiracil alone. However, the evidence is very uncertain about the effect of trifluridine-tipiracil plus bevacizumab in the change from baseline in EORTC QLQ-C30 global health status score, the proportion of patients with at least a 10-point deterioration from baseline in the global health status score, and the change from baseline in EQ-5D-5L health utilities (very low certainty) when compared with trifluridine-tipiracil alone. Trifluridine-tipiracil plus bevacizumab may result in little to no clinically important difference in bone marrow suppression, infections, gastrointestinal symptoms, and hypertension (low certainty) when compared with trifluridine-tipiracil alone. Patients treated with trifluridine-tipiracil plus bevacizumab experienced similar frequencies of TEAEs, SAEs, treatment discontinuations due to AEs, and deaths, and no new safety signals were identified. Confidence in the effect estimates in the SUNLIGHT trial were limited due to imprecision from wide CIs for survival and harms outcomes, the potential of the open-label design to bias patient-reported outcomes, and significant missing data for HRQoL.

There is a lack of head-to-head direct evidence between trifluridine-tipiracil plus bevacizumab and other treatments for advanced mCRC. A sponsor-submitted ITC indicated that the treatment of trifluridine-tipiracil plus bevacizumab may be associated with prolonged OS and PFS, but with an increased risk of treatment-related AEs in patients with mCRC compared to other treatments such as BSC, regorafenib, or trifluridine-tipiracil alone. However, there was substantial uncertainty due to limitations, including significant between-trial heterogeneity, potential intransitivity, and the absence of HRQoL outcomes of importance to patients and clinicians.

Introduction

The objective of this report is to review and critically appraise the evidence submitted by the sponsor on the beneficial and harmful effects of trifluridine-tipiracil (15 mg/6.14 mg and 20 mg/8.19 mg) oral tablets in combination with bevacizumab in the treatment of mCRC in adults who have been previously treated with, or are not candidates for, available therapies, including fluoropyrimidine-, oxaliplatin-, and irinotecan-based chemotherapies, anti-VEGF biological agents, and if RAS wild-type, anti-EGFR agents. The submission is not a Health Canada–approved indication; rather, it has been accepted under review as an unlabelled indication.

Disease Background

Content in this section has been informed by materials submitted by the sponsor and clinical expert input. The following has been summarized and validated by the CADTH review team.

CRC collectively refers to malignant tumours that develop in the rectum or colon, due to their similar tissue type and continuity. Most CRC cases develop in the epithelial lining of the colon or rectum from polyps that progress into cancer.¹ Metastatic CRC indicates that the cancer has spread beyond the primary tumour site to other organs of the body (i.e., stage IV disease), where the most common locations of metastases are the liver, lung, peritoneum, and distant lymph nodes.⁵

Patients with early CRC are usually asymptomatic.⁷ In patients with advanced disease, common clinical presentation includes iron-deficiency anemia, rectal bleeding, abdominal pain, change in bowel habits, and intestinal obstruction or perforation. Right-sided (proximal) tumours rarely present with obvious rectal bleeding as the blood becomes admixed with the stool.⁹ Left-sided (distal) tumours are more likely to present with bright red blood per rectum and symptoms of bowel obstruction.^8,9 In mCRC, symptoms may vary depending on the location of metastasis, and may include upper-right quadrant pain, abdominal distention, early satiety, supraclavicular adenopathy, and periumbilical nodules.⁷ In its early stages, CRC is a subtle disease; resultantly, most CRCs are diagnosed after symptom onset, although population-based screening is increasing the number of asymptomatic cases identified.⁷

The cause of CRC is dependent on a number of factors. People with a family history of CRC (first-degree relatives) are at a higher risk for CRC.^1,7,11 Lynch syndrome is the most common hereditary factor in CRC predisposition, accounting for 2% to 4% of all CRC cases.^1,7,11,25 People with inflammatory bowl disease are also at an increased risk for CRC.^1,7,11 Other risk factors include smoking, dietary factors (consumption of red and processed meats), heavy alcohol consumption, diabetes mellitus, low levels of physical activity, metabolic syndrome, and obesity and/or high body mass index.^1,7,11

CRC is the third most prevalent cancer in Canada overall and the second most prevalent cancer by sex, accounting for 12.7% and 10.0% of all 25-year prevalent cancers among males and females, respectively.² It is estimated that the Canadian (excluding Quebec) 10-year prevalence of CRC in both sexes of all ages was 343.5 cases per 100,000 population in 2018 (or 97,755 cases in total).⁴ CRC is also the fourth most commonly diagnosed cancer (10% of all new cancer cases) as well as the second leading cause of cancer-related death (11% of all cancer deaths) among Canadians.³ In 2022, it was estimated that 13,500 men and 10,800 women were diagnosed with CRC and 5,200 men and 4,200 women died from it.³

Data from the Canadian Cancer Registry database at Statistics Canada shows that approximately one-fifth (19.9%) of CRC cases are stage IV at diagnosis. The stage of CRC at diagnosis is strongly associated with survival.⁶ The 5-year relative survival for colon cancer is estimated to be 92% for cancers diagnosed at stage I compared with only 11% for those diagnosed at stage IV disease; similarly, the 5-year survival for rectal cancer is estimated to be 87% for stage I compared with only 12% for those diagnosed at stage IV disease.⁶

Diagnostic testing is not required for the indication under review, as patients eligible for trifluridine-tipiracil will have already been diagnosed with metastatic disease (stage IV) and would be considered refractory to all existing treatment options. As previously noted, trifluridine-tipiracil plus bevacizumab is an unlabelled indication, and while the approved indication is for trifluridine-tipiracil alone, no specific diagnostic technology is recommended in the product monograph.¹⁷ In general, diagnostic testing is required to confirm a diagnosis of CRC and pathologic confirmation of CRC is required before treatment. Typically, colonoscopy-guided biopsy confirms the primary cancer, and biopsy of the liver, lung, or lymph node confirms metastases.^11,12,26 Diagnostic staging should include contrast-enhanced CT imaging of the chest, abdomen, and pelvis.^11,12,26 At the time of diagnosis, patients will also undergo testing for activating mutations of RAS (KRAS and NRAS), BRAF, and the evaluation of microsatellite instability or mismatch repair deficiency in tumour tissue, which guides the overall treatment plan for an earlier line of therapy.²⁶ Additional baseline imaging is appropriate if symptoms are suggestive of metastases (e.g., CT head, bone scan). While on systemic therapy, imaging should be done every 2 months to 3 months depending on the clinical scenario.²⁶

Standards of Therapy

Content in this section has been informed by materials submitted by the sponsor and clinical expert input. The following has been summarized and validated by the CADTH review team.

Place in Therapy and Comparators

Current Treatment Paradigm

For patients with advanced CRC or mCRC, the main goals of treatment are to improve quality of life by controlling or delaying the onset of tumour-related symptoms and, if possible, to prolong life;^26,27 treatment intent is palliative rather than curative.²⁷ The majority of patients with mCRC have unresectable (inoperable) disease, for which the mainstay of treatment is systemic multidrug chemotherapy.^10,11 The choice of treatment is dependent on a number of factors, including a patient’s fitness (e.g., performance status), organ function, and comorbidities, in addition to tumour characteristics (e.g., tumour location [right versus left], presence of primary tumour, mutation status [RAS, BRAF], presence of dMMR or MSI-H), type and timing of prior therapy, and toxicity profiles of constituent drugs.^5,10,12-15 External beam radiation therapy — either conventional or stereotactic — may be also offered to suitable patients with mCRC to palliate symptoms or control disease, according to the clinical expert consulted by CADTH.

Management of mCRC is increasingly driven by tumour biology and gene expression analysis of individual tumours.^10,27 Multiple different classes of drugs with antitumour activity in mCRC may be appropriate, although the optimal combination and sequence of available drugs is evolving and is focused on individualized treatment along a continuum of care.¹⁰ Systemic fluorouracil-based chemotherapy, especially when combined with regimens containing irinotecan or oxaliplatin, produce meaningful improvements in survival now consistently approaching 2 years.²⁷ FOLFIRI and FOLFOX are the standard initial chemotherapy regimens.^11,27 Anti-VEGF and anti-EGFR drugs can be added to chemotherapy regimens to further increase survival for patients with mCRC. Anti-EGFR drugs only benefit a subpopulation of patients with mCRC with wild-type RAS tumours (approximately 52% of patients with CRC^28,29).^5,11,13-15 There is also a small subset of patients with mCRC with MSI-H status (approximately 3.5% to 6.5% of patients²⁷) for whom pembrolizumab (an immune checkpoint inhibitor) would be an initial treatment option. For a small population of patients with BRAF mutations (approximately 5% to 12% of patients with mCRC²⁷), encorafenib (a kinase inhibitor) is funded for use in combination with cetuximab or panitumumab (an anti-EGFR drug) after disease progression on initial chemotherapy with or without bevacizumab.^{5,10,11,13-15}

The duration of chemotherapy for unresectable mCRC is dependent on the patient and regimen used, where intermittent therapy rather than continuous therapy has the potential to reduce treatment-related toxicity (e.g., cumulative neurotoxicity has been observed in oxaliplatin-containing regimens).²⁷ Importantly, the initiation of chemotherapy before patients become symptomatic is ideal.

Response to chemotherapy is typically assessed by a periodic assay of serum CEA levels, if initially elevated, and radiographic evaluation (every 8 weeks to 12 weeks, or as prompted by rising CEA levels), quantified using Response Evaluation Criteria in Solid Tumours.²⁷

Regimens containing fluoropyrimidine-, oxaliplatin-, and irinotecan-based chemotherapies are broadly funded and comprise the most frequently used cytotoxic regimens for patients with mCRC.^5,11,13-15 In Canada, oncologists often follow well-accepted international treatment guidelines, including the National Comprehensive Cancer Network Clinical Practice Guidelines in Oncology, the American Society of Clinical Oncology guidelines, and the European Society for Medical Oncology Clinical Practice Guidelines.^{5,11,12,14,15} Across Canada, provincial cancer agencies have also developed treatment guidelines that align with international guidelines.^26,30,31

Following treatment with standard cytotoxic chemotherapy backbone regimens, there are few efficacious options in later lines of therapy. Both trifluridine-tipiracil (Lonsurf) and regorafenib (Stivarga) are approved in Canada for the treatment of patients who have been previously treated with fluoropyrimidine-, oxaliplatin-, and irinotecan-based chemotherapies, anti-VEGF biological agents, and, if RAS wild-type, anti-EGFR agents;^16,17 however, these treatments are not publicly funded in Canada (except in Quebec)^18,19 Patients with mCRC who have already been treated with fluoropyrimidine-, oxaliplatin-, and irinotecan-based chemotherapies, anti-VEGF biological agents, and, if RAS wild-type, anti-EGFR agents, they are usually treated with BSC, for whom median survival is approximately 5 months to 6 months.²⁷

The overall treatment algorithm for mCRC is summarized in Figure 1.

Figure 1: Metastatic Colorectal Cancer Funding Algorithm (Current)

1L = first-line; 2L = second-line; 3L = third-line; 4L = fourth-line; EGFR = epidermal growth factor receptor; EGFRi = epidermal growth factor receptor inhibitor; ESMO = European Society for Medical Oncology; FOLFIRI = folinic acid, fluorouracil, and irinotecan, FOLFOX = folinic acid, fluorouracil, and oxaliplatin; FOLFOXIRI = folinic acid, fluorouracil, oxaliplatin, and irinotecan; NCCN = National Comprehensive Cancer Network; WT = wild-type.

Note: Figure 1 presents information from the CADTH provisional algorithm,¹³ Cancer Care Alberta guidelines,²⁶ NCCN guidelines,^11,12 and ESMO guidelines.³² The clinical experts consulted by CADTH generally agreed with the algorithm submitted by the sponsor; however, they noted that among patients who are treated with EGFRi (e.g., FOLFIRI plus panitumumab) in the 1L setting, there is limited to no funding to rechallenge with this therapy in the 3L setting.

^a Multidrug chemotherapy consists of doublet therapy (FOLFOX, FOLFIRI, or capecitabine + oxaliplatin) and triplet therapy (FOLFOXIRI).

^b EGFRi includes cetuximab and panitumumab, where available.

^c If had not received EGFR in previous lines.

Source: Sponsor’s Summary of Clinical Evidence.³³

Drug Under Review

Key characteristics of trifluridine-tipiracil (Lonsurf) plus bevacizumab are summarized in Table 3 with other treatments available for the treatment of adult patients with mCRC who have been previously treated with, or are not candidates for, available therapies, including fluoropyrimidine-, oxaliplatin-, and irinotecan-based chemotherapies, anti-VEGF biological agents, and, if RAS wild-type, anti-EGFR agents.

Table 3: Key Characteristics of Trifluridine-Tipiracil Plus Bevacizumab, Lonsurf Alone, and Regorafenib

Anti-EGFR = anti–epidermal growth factor receptor; anti-VEGF = anti–vascular endothelial growth factor; mCRC = metastatic colorectal cancer; NA = not applicable; RTK = receptor tyrosine kinase; VEGF-A = vascular endothelial growth factor A.

^aHealth Canada–approved indication.

Source: Product monographs for Stivarga, Lonsurf, and Avastin.^16,17,34

Dosing and Administration

Note: As this is an unlabelled indication, the individual dosing of trifluridine-tipiracil and bevacizumab have been taken from their respective product monographs for their CRC indications.^17,34 Both dosing regimens align with the dosing used in the SUNLIGHT trial.²²

Trifluridine-Tipiracil Dosing

The recommended starting dose of trifluridine-tipiracil for adults is 35 mg/m² per dose administered orally with water, twice daily, within 1 hour after completion of morning and evening meals, on day 1 to day 5 and day 8 to day 12 of each 28-day cycle. This treatment cycle is repeated every 4 weeks as long as benefit is observed or until unacceptable toxicity occurs. Refer to Table 4 for dose calculation based on body surface area (BSA). The dosage must not exceed 80 mg per dose based on the trifluridine component.¹⁷

Dosing adjustments may be required based on individual safety and tolerability. A maximum of 3 dose reductions is permitted to a minimum dose of 20 mg/m² twice daily. Dose escalation is not permitted after it has been reduced.¹⁷

In the event of hematologic and/or nonhematologic toxicities, patients should follow the dose interruption, resumption, and reduction criteria, as detailed in the product monograph.¹⁷

Bevacizumab Dosing

As per the product monograph for the mCRC indication, the recommended dose of bevacizumab is 5 mg/kg of body weight given once every 14 days as an IV infusion.³⁴

In a 28-day treatment cycle with trifluridine-tipiracil, bevacizumab would be administered on day 1 and day 15.²²

Mechanism of Action

Trifluridine-tipiracil comprises an antineoplastic thymidine-based nucleoside analogue, trifluridine, and the thymidine phosphorylase inhibitor tipiracil (as tipiracil hydrochloride).²²

Bevacizumab is a recombinant humanized monoclonal antibody that selectively binds to and neutralizes the biologic activity of human VEGF. Bevacizumab contains human framework regions with antigen binding regions of a humanized murine antibody that binds to VEGF. Bevacizumab is produced by recombinant DNA technology in a Chinese hamster ovary mammalian cell expression system and is purified by a process that includes specific viral inactivation and removal steps. Bevacizumab consists of 214 amino acids and has a molecular weight of approximately 149,000 daltons.³⁴

Combination regimens of different chemotherapeutic drugs (e.g., fluorouracil, oxaliplatin, irinotecan, molecularly targeted drugs) have prolonged survival in patients with mCRC compared to treatment with fluorouracil alone. Blood vessels in tumour tissues, including mCRC, tend to be poorly organized as well as hyperpermeable, resulting in a diminished blood supply, which may limit the accumulation of trifluridine-tipiracil in tumours. Since bevacizumab inhibits angiogenesis and normalizes tumour vasculature, tumour blood supply is improved, allowing an increase in the accumulation of trifluridine-tipiracil in the tumour and subsequent phosphorylation of trifluridine in the tumour. The phosphorylated forms of trifluridine are active components, responsible for the inhibitory effect on tumour DNA.^35,36

Prescribing

Both trifluridine-tipiracil and bevacizumab should only be prescribed by a doctor with experience in the use of anticancer drugs for mCRC.^17,34

Table 4: Trifluridine-Tipiracil Starting Dose Calculation According to Body Surface Area

BSA = body surface area.

Sources: Lonsurf product monograph¹⁷ and SUNLIGHT Clinical Study Report.²²

Stakeholder Perspectives

Patient Group Input

This section was prepared by the CADTH review team based on the input provided by patient groups. The full original patient input received by CADTH has been included in the Stakeholder section of this report.

CADTH received 2 patient group submissions from CCRAN and Colorectal Cancer Canada. CCRAN is a national, not-for-profit patient advocacy group championing the health and well-being of Canadians touched by CRC and those at risk of developing the disease, by providing support, education, and advocacy to help improve patient outcomes by way of longevity and quality of life. Colorectal Cancer Canada is a not-for-profit organization for patients with CRC dedicated to CRC awareness and education, supporting patients and their caregivers and advocating on their behalf.

CCRAN used a multifaceted outreach approach by emailing clinicians who treat advanced CRC to help recruit patients or caregivers with experience with Lonsurf (plus bevacizumab) and via an online survey of patients’ experience of mCRC and prior drug therapies. Five patients who had experience with the therapy under review participated in a telephone interview from June 12 to July 26, 2023, and 77 survey respondents (including 60 patients, 13 caregivers, and 4 patients who were also caregivers) provided input from June 13 to August 5, 2023. Colorectal Cancer Canada conducted an online survey of 23 respondents (22 patients and 1 caregiver) from July 19 to August 17, 2023.

Most patients reported that fatigue/weakness, bloody stools, diarrhea, and abdominal cramping/gas/feeling bloated, and abdominal pain are common symptoms they experienced and that they felt were important to control. Symptoms of CRC affected the quality of life for patients and their families, limiting the ability to work, to exercise, to participate in social activities, or to undertake daily tasks. Furthermore, patients cited feeling consistently worried, nervous, or uneasy and with a persistent fear of the cancer getting worse or recurring (coming back) as the most common psychological impacts. For some patients with mCRC, there were no symptoms before diagnosis and the diagnosis was the result of incidental findings. Caregivers also reported challenges with caring for patients with CRC, including loss of lifestyle, inability to work outside the home, difficulty managing treatment-induced side effects, loss of income and/or financial strain, time spent at medical appointments, and feelings of inadequacy or helplessness.

In terms of current therapy options, FOLFOX, FOLFIRI, capecitabine, and bevacizumab were cited most frequently. According to Colorectal Cancer Canada’s survey, 5% of respondents said “no,” 45% said “somewhat,” and 50% said “yes” when asked whether these drug therapies had been effective at controlling the progression of the disease. Moreover, when patients and caregivers were asked whether they believed their needs were not being met by current drugs available to treat the cancer, 30% of respondents replied “yes.” In general, all patients from both patient groups reported debilitating side effects while undergoing treatments for their mCRC, which compromised their quality of life. According to CCRAN’s survey, all interviewed patients and 1 caregiver reported debilitating side effects while undergoing treatments for mCRC, which compromised their quality of life. While patients may have derived a clinical benefit in terms of response, that response was accompanied by incapacitating side effects such as fatigue, nausea, lack of energy, diarrhea, neuropathy, skin rash, lethargy, and flu-like symptoms that prevented them from engaging in life on any meaningful level. Patients were quite emphatic about their experience with combination chemotherapies that compromised their well-being some or most of the time, resulting in the need to take time off work, the inability to care for children, a lack of self-care, and less time spent enjoying life in general. Normal daily activities could not be resumed nor could quality time with friends and family be spent, permitting them the freedom to “live life again”.

According to both patient groups, it is very important for a new therapy to bring about improvement to patients’ physical condition (e.g., tumour shrinkage, tumour stability, reduced pain, improved breathing) and quality of life (e.g., improved mobility, improved sense of wellness, relief from side effects) such that patients would take a new therapy to bring about improvement in their quality of life even if it does not extend OS (e.g., at a modest 3 months to 4 months of survival, 53% of respondents were willing to tolerate significant side effects, including nausea, anemia, and neutropenia). Moreover, patients preferred a drug therapy that is convenient (e.g., orally administered, either at home or with a short infusion duration or chair time at a cancer centre).

A total of 5 respondents from CCRAN’s survey and 1 respondent from Colorectal Cancer Canada’s survey indicated having experience with the drug under review, and the primary method of access to the drug under review was through private insurance and the sponsor’s Patient Support Program. According to the CCRAN survey, all 5 patients demonstrated excellent tolerance with Lonsurf plus bevacizumab, with a reduced incidence of AEs, and maintained quality of life despite having experienced significant side effects with SOC therapies. One respondent from Colorectal Cancer Canada’s survey rated trifluridine-tipiracil plus bevacizumab as being similar to their experiences with other drugs in effectiveness (i.e., able to partially shrink or control their CRC and/or metastasis) and side effects (i.e., neutropenia, loss of appetite, diarrhea, and nausea), being easy to administer or receive, and enabling them to continue daily activities while undergoing treatment.

CCRAN believes that if publicly funded, trifluridine-tipiracil plus bevacizumab would be an extremely important third-line and beyond therapy for patients whose disease has been deemed to be refractory or ineligible for SOC therapies. Colorectal Cancer Canada noted that given that Lonsurf alone is currently reimbursed only in Quebec, there is a strong need for equity of access for patients located elsewhere in Canada. Both patient groups strongly agree that this combination under review aligns well with the identified patient and caregiver need for a new effective treatment option that is capable of prolonging life and maintaining quality of life.

Clinician Input

Input From Clinical Experts Consulted by CADTH

All CADTH review teams include at least 1 clinical specialist with expertise regarding the diagnosis and management of the condition for which the drug is indicated. Clinical experts are a critical part of the review team and are involved in all phases of the review process (e.g., providing guidance on the development of the review protocol, assisting in the critical appraisal of clinical evidence, interpreting the clinical relevance of the results, providing guidance on the potential place in therapy). The following input was provided by 2 clinical specialists with expertise in the diagnosis and management of mCRC.

Unmet Needs

The cornerstone of treatment for patients with mCRC involves the sequential use of the best available systemic therapies, according to the clinical experts consulted by CADTH. For patients with dMMR or MSI-H mCRC, SOC first-line treatment in Canada includes pembrolizumab immunotherapy, based on findings of the KEYNOTE-177 phase III RCT according to input from clinical experts consulted by CADTH.³⁷ For patients with left-sided, extended RAS wild-type CRC, SOC first-line treatment in Canada includes chemotherapy with FOLFOX or FOLFIRI in combination with an EGFR inhibitor (panitumumab or cetuximab), based on the findings of the PARADIGM phase III RCT according to the clinical experts consulted by CADTH.³⁸ For patients with right-sided or extended RAS-mutated CRC, the clinical experts consulted by CADTH reported that SOC first-line treatment in Canada is chemotherapy with FOLFOX or FOLFIRI in combination with bevacizumab, based on results from a retrospective analysis of 6 RCTs.³⁹ Patients who progress on or within 6 months of adjuvant therapy (e.g., cancer growth while on adjuvant therapy or within 6 months of adjuvant FOLFOX) would be considered to have experienced progression on first-line treatment.

Following disease progression on first-line therapy, the clinical experts consulted by CADTH indicated that SOC second-line systemic treatment in Canada for patients with BRAF V600E mutations includes encorafenib plus cetuximab, based on findings of the BEACON phase III RCT,⁴⁰ whereas patients without the BRAF V600E mutation are switched on the backbone chemotherapeutic regimen (e.g., patients initially treated with FOLFIRI would be switched to FOLFOX, patients initially treated with FOLFOX would be switched to FOLFIRI), based on findings of the GERCOR phase III RCT.⁴¹ According to the clinical experts consulted by CADTH, other therapies in addition to the chemotherapy backbone are not routinely used in Canada; conversely, the addition of antiangiogenic therapies to the chemotherapy backbone (e.g., bevacizumab, aflibercept, ramucirumab) for patients without a BRAF V600E mutation or dual immunotherapy (e.g., nivolumab plus ipilimumab) for patients with a dMMR or MSI-H molecular marker are routinely offered to patients with CRC outside Canada and are listed in the National Comprehensive Cancer Network and European Society for Medical Oncology guidelines, respectively.^11,12,15

Following disease progression on 2 lines of prior therapy, patients with extended RAS wild-type CRC who have not been previously treated with an EGFR inhibitor are eligible for a single-drug EGFR inhibitor (cetuximab or panitumumab) or cetuximab plus irinotecan as SOC treatment in Canada, based on the findings of several phase III RCTs,^42,43 according to the clinical experts consulted by CADTH. The clinical experts noted that other patients (without an extended RAS wild-type marker) are treated with either regorafenib monotherapy or trifluridine-tipiracil as SOC treatment in Canada, based on the results of the CORRECT and RECOURSE phase III RCTs, respectively.^44,45

According to the clinical experts consulted by CADTH, the selection of systemic therapy in all lines of treatment is dependent on the patient’s performance status, the sidedness (for treatment decision-making in the first-line setting) and molecular features of the tumour, symptoms, values, and preferences. Notably, the clinical experts consulted by CADTH reported that BSC without anticancer therapy is a treatment option in all lines of treatment and is often given higher consideration, taking into account the limited efficacy of therapies after progression on second-line treatments. The clinical experts consulted by CADTH indicated that for patients with mCRC who experience disease progression following 2 lines of anticancer therapy, there exists a significant unmet need for effective treatment options that improve OS, decrease symptom burden, and improve quality of life. Additionally, given that there are patients diagnosed with de novo CRC presenting with more aggressive disease, and that there are higher rates of CRC diagnoses in a younger demographic, the clinical experts consulted by CADTH highlighted an important need for novel and effective treatment options.

Place in Therapy

The clinical experts consulted by CADTH considered trifluridine-tipiracil plus bevacizumab to represent a new SOC treatment for patients with mCRC after disease progression on 2 prior lines of anticancer therapy.

Patient Population

According to the clinical experts consulted by CADTH, trifluridine-tipiracil plus bevacizumab would be an appropriate treatment option for patients with unresectable CRC who experience disease progression on 2 prior lines of anticancer systemic therapy. Among patients with mCRC, the clinical experts consulted by CADTH noted that those eligible for treatment should be able to tolerate both trifluridine-tipiracil (i.e., able to safely swallow pills, have normal bowel transit, have an ECOG PS score of 0 to 1, and have adequate hematologic, hepatic, and renal function) and bevacizumab (i.e., without absolute contraindication to the use of a VEGF inhibitor, including but not limited to uncontrolled hypertension, in situ colonic stent, recent surgery, a high-risk for bleeding, and a risk for or presence of fistula or gastrointestinal tract perforation). The clinical experts consulted by CADTH indicated that a companion diagnostic is not required for treatment with trifluridine-tipiracil plus bevacizumab.

Assessing the Response Treatment

According to the clinical experts consulted by CADTH, 3 key factors are used to determine response to treatment among patients with mCRC, in the following hierarchy: 1) patient-reported symptoms or side effects, as determined by clinician assessment of patient treatment history, 2) the examination and selective use of clinical instruments to evaluate symptoms (e.g., Edmonton Symptom Assessment System, EQ-5D), and 3) cross-sectional imaging (e.g., CT scan, MRI) and tumour markers (e.g., CEA, CA 19 to 9). Patients should be assessed after every 2 cycles to 3 cycles of treatment (and more frequently with bothersome symptoms or AEs), with tumour markers completed at least once every 4 weeks and CT scans conducted every 2 months to 3 months, according to the clinical experts consulted by CADTH. Overall, the clinical experts consulted by CADTH emphasized that OS, symptom control, and quality of life are clinically meaningful end points; all other end points (e.g., response rate, CEA response, PFS) may be considered surrogates if they predict better OS or quality of life.

Discontinuing Treatment

Patient-reported symptoms and well-being, including side effects or toxicity, were key determinants for discontinuing treatment with trifluridine-tipiracil plus bevacizumab, according to the clinical experts consulted by CADTH. In the event of the development of an absolute contraindication to further therapy as noted previously (e.g., the development of uncontrolled hypertension, the requirement for surgery or having had recent surgery, uncontrollable bleeding, the risk for or presence of fistula or gastrointestinal perforation), the clinical experts consulted by CADTH indicated that treatment with bevacizumab should be discontinued. Overall, the clinical experts highlighted the importance of shared and fully informed decision-making with patients that includes discussions regarding treatment effectiveness (e.g., incorporating results of imaging and serum tumour markers for context) and symptoms or AEs that significantly impact quality of life.

Prescribing Considerations

The clinical experts consulted by CADTH agreed that trifluridine-tipiracil plus bevacizumab should be prescribed by or be administered under the supervision of an oncologist with expertise in the prescription and management of chemotherapy-related side effects in an inpatient clinic or an outpatient setting.

Clinician Group Input

This section was prepared by the CADTH review team based on the input provided by clinician groups. The full original clinician group input received by CADTH has been included in the Stakeholder section of this report.

CADTH received 2 clinician group submissions from the CGOEN with the Medical Advisory Board of Colorectal Cancer Canada (and other Colorectal Cancer Canada treating physicians) and the OH-CCO Gastrointestinal Cancer Drug Advisory Committee. CGOEN is a virtual and inclusive network of gastrointestinal oncology clinicians in Canada who contribute to the knowledge of gastrointestinal cancer and its treatments, including by participating in clinical trials, by conducting observational research, and by being involved in local, provincial, and national clinical guideline developments and Health Technology Assessments. OH-CCO’s Drug Advisory Committees provide timely evidence-based clinical and health system guidance on drug-related issues in support of OH-CCO’s mandate, including the Provincial Drug Reimbursement Programs and the Systemic Treatment Program.

CGOEN gathered data and information based on personal experience in treating patients with mCRC and expert evidence-based review by gastrointestinal cancer specialists in Canada of the following information presented at international oncology meetings and subsequently published in The New England Journal of Medicine; OH-CCO’s Drug Advisory Committees gathered information through videoconferencing and email communication.

Both CGOEN and OH-CCO’s Drug Advisory Committees emphasized that the treatment of mCRC is limited to fluoropyrimidine plus irinotecan and fluoropyrimidine plus oxaliplatin chemotherapy backbones with the use of bevacizumab and anti-EGFR monoclonal antibody therapy (either panitumumab or cetuximab) in tumours without an RAS mutation. Treatment of MSI-H mCRC includes the use of checkpoint inhibitor pembrolizumab. In tumours that have a BRAF V600E variant, treatment would include the use of encorafenib with anti-EGFR monoclonal antibody therapy.

CGOEN stated that trifluridine-tipiracil is currently Health Canada–approved but received a do not reimburse recommendation in August 2019 from CADTH due to the fact that the magnitude of benefit was felt to be too small to warrant approval, despite being recognized as addressing the needs of a population with an unmet need. It is currently funded in Quebec, having received a reimburse recommendation from the Institut national d’excellence en santé et en services sociaux. Outside Quebec, patients have been able to apply to the sponsor for access to the drug under review through private insurance or direct user pay. Therefore, the majority of patients with mCRC in Canada do not have access to a publicly funded version of the drug under review, according to CGOEN. OH-CCO’s Drug Advisory Committees also echoed this concern highlighted by CGOEN. Therefore, CGOEN felt that findings from the original trial of trifluridine-tipiracil compared to BSC should be considered in the current review of trifluridine-tipiracil plus bevacizumab, given the current landscape in Canada.

CGOEN indicated that there is an unmet need for patients with advanced CRC whose disease progresses on currently available and/or funded therapies. Many of these patients have a reasonable ECOG PS score of 0 or 1. Moreover, the primary goal for therapy in this population is OS. Response rates and quality of life are also goals for minimizing symptoms and the ability to maintain independence.

CGOEN expressed that a clinical assessment would be done before each cycle, with radiographic restaging generally performed every 8 weeks to 12 weeks with the primary goal being to extend survival. The goal is to improve current symptoms by shrinking tumour burden and delaying further progression and the associated decrease in quality of life and increased symptom burden that comes with that. It is also important to focus on patient functioning and impact on caregivers.

Both clinician groups highlighted that the drug under review would be placed as a further line of therapy and would be used in patients who received current SOC options and have experienced disease progression, have experienced intolerance, or have chosen to stop for personal reasons. This combination would also be used for those with medical contraindications to earlier-line SOC therapies. Patients suitable for treatment would include those who are refractory or intolerant of fluoropyrimidine, irinotecan, oxaliplatin, bevacizumab, EGFR monoclonal antibody therapy (if KRAS wild-type), encorafenib (if BRAF V600E), and immunotherapy (if dMMR or MSI-H). CGOEN stated patients should have an ECOG PS score of 0 or 1 and adequate hematologic and biochemical parameters for therapy; there is no biomarker for this therapy and therefore a companion diagnostic is not required. There is not a subgroup that did not show a benefit; it should be noted that the majority of patients had previously received bevacizumab and a benefit is seen with re-treatment with bevacizumab in combination with the drug under review. Both groups agreed that treatment would be discontinued due to disease progression, toxicity, clinician discretion, or patient request.

CGOEN mentioned that it would be reasonable for trifluridine-tipiracil plus bevacizumab to be given in any centre and by any specialist who is currently giving systemic therapy for patients with mCRC. OH-CCO’s Drug Advisory Committees noted that this treatment should be given in OH-CCO–approved cancer facilities that have qualified staff, lab facilities, and infusion facilities (level 1 to level 4 cancer centres).

Drug Program Input

The drug programs provide input on each drug being reviewed through CADTH’s reimbursement review processes by identifying issues that may impact their ability to implement a recommendation. The implementation questions and corresponding responses from the clinical experts consulted by CADTH are summarized in Table 5.

Table 5: Summary of Drug Plan Input and Clinical Expert Response

dMMR = deficient mismatch repair; ECOG PS = Eastern Cooperative Oncology Group Performance Status; MSI-H = microsatellite instability–high; PAG = Provincial Advisory Group; pERC = CADTH pan-Canadian Oncology Drug Review Expert Review Committee; PFS = progression-free survival.

Clinical Evidence

The objective of CADTH’s Clinical Review Report is to review and critically appraise the clinical evidence submitted by the sponsor on the beneficial and harmful effects of trifluridine-tipiracil (15 mg trifluridine and 6.14 mg tipiracil [as tipiracil hydrochloride] and 20 mg trifluridine and 8.19 mg tipiracil [as tipiracil hydrochloride]) plus bevacizumab in the treatment of mCRC in adults with mCRC who have been previously treated with, or are not candidates for available therapies, including fluoropyrimidine-, oxaliplatin-, and irinotecan-based chemotherapies, anti-VEGF biological agents, and, if RAS wild-type, anti-EGFR agents. The focus will be placed on comparing trifluridine-tipiracil plus bevacizumab to relevant comparators and identifying gaps in the current evidence.

A summary of the clinical evidence included by the sponsor in the review of trifluridine-tipiracil plus bevacizumab is presented in 4 sections with CADTH’s critical appraisal of the evidence included at the end of each section. The first section, the systematic review, includes pivotal studies and RCTs that were selected according to the sponsor’s systematic review protocol. CADTH’s assessment of the certainty of the evidence in this first section using the GRADE approach follows the critical appraisal of the evidence. The third section includes indirect evidence from the sponsor. There were no long-term extension studies or additional studies that were considered by the sponsor to address important gaps in the systematic review evidence.

Included Studies

Clinical evidence from the following is included in the CADTH review and appraised in this document:

• One pivotal study identified in the systematic review

• One ITC.

Systematic Review

Content in this section has been informed by materials submitted by the sponsor. The following has been summarized and validated by the CADTH review team.

Description of Studies

Characteristics of the included studies are summarized in Table 6.

Table 6: Details of Studies Included in the Systematic Review

ALT = alanine aminotransferase; ANC = absolute neutrophil count; anti-EGFR = anti–epidermal growth factor receptor; anti-VEGF = anti–vascular endothelial growth factor; AST = aspartate aminotransferase; CNS = central nervous system; CR = complete response; CTCAE = Common Terminology Criteria for Adverse Events; ECOG PS = Eastern Cooperative Oncology Group Performance Status; EORTC QLQ-C30 = European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire Core 30; EQ-5D-5L = 5-Level EQ-5D; EU = European Union; HBsAg = hepatitis B surface antigen; HBV = hepatitis B virus; HCV = hepatitis C virus; INR = international normalized ratio; NYHA = New York Heart Association; PR = partial response; RECIST 1.1 = Response Evaluation Criteria in Solid Tumours Version 1.1; RNA = ribonucleic acid; SGOT = serum glutamic-oxaloacetic transaminase; SGPT = serum glutamic pyruvic transaminase; ULN = upper limit of normal; VEGF = vascular endothelial growth factor.

Note: Details included in Table 6 are from the sponsor’s Summary of Clinical Evidence.³³

Source: SUNLIGHT Clinical Study Report.²²

The SUNLIGHT trial was a randomized, double-arm, phase III, open-label, multicentre study that evaluated the safety and efficacy of trifluridine-tipiracil plus bevacizumab versus trifluridine-tipiracil alone among adult patients with refractory mCRC who had received no more than 2 previous chemotherapy regimens. The primary objective was to estimate OS of patients in the trifluridine-tipiracil plus bevacizumab group compared with trifluridine-tipiracil alone group. The key secondary objective was to estimate PFS of patients in the trifluridine-tipiracil plus bevacizumab group compared with trifluridine-tipiracil alone group, based on investigator assessment. Following screening, patients were enrolled from November 25, 2020, to February 18, 2022, at 87 study sites (there were no study sites in Canada).

A total of 492 patients were randomized 1:1 via an Interactive Web Response System (IWRS) using central randomization to the trifluridine-tipiracil plus bevacizumab group or the trifluridine-tipiracil alone group. Randomization was stratified by geographic region (North America, European Union, the rest of the world), time since first metastasis diagnosis (< 18 months, ≥ 18 months), and the RAS status (wild-type, mutant). The data cut-off dates were July 5, 2022, for clinical data (i.e., nonsurvival) and July 19, 2022 (occurrence of the 331st death), for survival data.

Populations

Inclusion and Exclusion Criteria

The study included adult patients (aged 18 years or older) with refractory mCRC who had received no more than 2 previous chemotherapy regimens and had progressive disease or whose last regimen had caused unacceptable adverse effects. Previous treatment must have included fluoropyrimidine, irinotecan, oxaliplatin, an anti-VEGF monoclonal antibody (not necessarily bevacizumab), or an anti-EGFR monoclonal antibody (for patients with RAS wild-type disease), and the treatment could have included neoadjuvant or adjuvant chemotherapy if disease had recurred during treatment or within 6 months after the last administration of neoadjuvant or adjuvant therapy. Eligibility required knowledge of RAS status. Patients had to have adequate organ function and an ECOG PS score of 0 or 1.

Interventions

Pretreatment Phase

Blood and urine tests are required before starting trifluridine-tipiracil. Blood and urine tests are routine procedures that are ordered by a physician and are typically performed at an outpatient laboratory clinic. Complete blood cell counts must be obtained before the initiation of therapy with trifluridine-tipiracil and as needed to monitor blood counts, but at a minimum, they must be obtained before each treatment cycle. Treatment with trifluridine-tipiracil must not be started if the absolute neutrophil count is below 1.5 multiplied by 10⁹/L, if the platelet counts are below 75 multiplied by 10⁹/L, or if the patient has unresolved grade 3 or grade 4 nonhematologic clinically relevant toxicity from prior therapies and/or prior trifluridine-tipiracil cycles. Monitoring of proteinuria by dipstick urinalysis is recommended before starting therapy and during therapy.

Treatment Phase

Trifluridine-tipiracil is administered orally and can be taken at home or in a community setting. As noted earlier, complete blood cell counts must be obtained before each treatment cycle along with the monitoring of proteinuria by dipstick urinalysis.

Trifluridine-tipiracil was administered orally, twice daily, at a starting dose of 35 mg/m² of BSA, on day 1 through day 5 and on day 8 through day 12 every 28 days. If doses were missed or held on those days, the patient was not to make up for missed doses. The extension of study treatment into day 6 to day 7 or into the rest period (day 13 through day 28) was not permitted. The patient’s weight at screening (baseline) was used to calculate the initial dose (based on BSA); however, if at the beginning of a new treatment cycle, the patient’s weight changed by at least 10%, the dose was recalculated based on the updated weight and the adjusted dose was provided to the study site by the IWRS. Table 7 shows the number of tablets that are needed per calculated BSA for a dose of 35 mg/m² and for a reduced dose (30 mg/m², 25 mg/m², and 20 mg/m²).

Table 7: Number of Tablets of Trifluridine-Tipiracil per Dose (Reduced Dose)

BSA = body surface area.

^aAt a total daily dose of 50 mg, patients should take one 20 mg/8.19 mg tablet in the morning and two 15 mg/6.14 mg tablets in the evening.

Source: SUNLIGHT Clinical Study Report.²²

Bevacizumab was administered intravenously by study site personnel at a dose of 5 mg/kg of body weight on day 1 and day 15. At each visit, the IWRS provided the study site with the adjusted bevacizumab dose based on the patient’s actual weight. The study site could round this dose at the nearest 0.5 mL. The initial dose was to be delivered over 90 minutes as an IV infusion. If the first infusion was well tolerated, the second infusion could be administered over 60 minutes. If the 60-minute infusion was well tolerated, all subsequent infusions could be administered over 30 minutes. If a medical condition developed that required bevacizumab to be permanently withdrawn, the patient could continue with trifluridine-tipiracil alone. Bevacizumab was not to be administered alone in case of dose delay due to trifluridine-tipiracil toxicities. The cycle was to be delayed and bevacizumab was to be restarted at the same time as trifluridine-tipiracil. Bevacizumab infusions should be prepared by a health care professional using an aseptic technique and would be administered at an outpatient infusion clinic or outpatient hospital setting.

There are no concomitant drugs required or recommended for patients receiving trifluridine-tipiracil plus bevacizumab. However, as trifluridine-tipiracil has emetogenic potential, a single antiemetic drug, such as dexamethasone, a 5-HT3 receptor antagonist, or a dopamine receptor antagonist such as metoclopramide hydrochloride may be considered for prophylaxis. For patients who experience anticipatory nausea and vomiting, benzodiazepines are recommended.

Post-Treatment Phase

After treatment, medications may be used for the management of AEs and can include the following.

• Loperamide hydrochloride may be used for diarrhea at an initial dose of 4 mg followed by 2 mg every 4 hours or after every unformed stool (not to exceed 16 mg per day). For patients with complicated diarrhea, treatment includes IV fluids, octreotide at a starting dose of 100 mcg to 150 mcg, subcutaneously 3 times daily or IV (25 mcg per hour to 50 mcg per hour) if the patient is severely dehydrated, with dose escalation up to 500 mcg until diarrhea is controlled, plus the administration of antibiotics (e.g., fluoroquinolone).

• Sodium bicarbonate mouthwash, anesthetic mouthwashes, topical analgesic or anti-inflammatory drugs, topical anesthetics, and alternative mouthwashes may be used for mucosal injury.

• Hematologic support may be used as medically indicated (e.g., blood transfusions, granulocyte colony-stimulating factor [G-CSF], erythropoietin) for febrile neutropenia and anemia.

The treatment schema for the SUNLIGHT trial is depicted in Figure 2. Each treatment cycle was 28 days in duration for each group.

Figure 2: Schematic of Study Design for SUNLIGHT Study

D = day; QoL = quality of life; w = week; IMP = investigational medicinal product.

Source: SUNLIGHT Clinical Study Report.²²

One treatment cycle of the trifluridine-tipiracil plus bevacizumab group consisted of the following:

• day 1 to day 5 — oral intake of trifluridine-tipiracil and bevacizumab IV infusion on day 1

• day 6 to day 7 — rest

• day 8 to day 12 — oral intake of trifluridine-tipiracil

• day 13 to day 14 — rest

• day 15 — bevacizumab IV infusion

• day 16 to day 28 — rest.

One treatment cycle of the trifluridine-tipiracil alone group consisted of the following:

• day 1 to day 5 — oral intake of trifluridine-tipiracil

• day 6 to day 7 — rest

• day 8 to day 12 — oral intake of trifluridine-tipiracil

• day 13 to day 28 — rest.

The 28-day treatment cycles continued until disease progression occurred, unacceptable toxic effects occurred, or consent was withdrawn. Patients were considered to be receiving treatment for as long as they continued to receive trifluridine-tipiracil; bevacizumab monotherapy was not allowed.

Dose Modifications of Trifluridine-Tipiracil Due to Toxicities

All toxicities related to trifluridine-tipiracil had to be resolved to grade 1 or baseline before the start of a new treatment cycle. In the event of hematologic and/or nonhematologic toxicities, rules for trifluridine-tipiracil dose interruption and resumption, and rules for dose modifications are provided (refer to Appendix 1, Figure 13). A new treatment cycle with trifluridine-tipiracil could be started only if the neutrophils count was equal to or greater than 1.5 multiplied by 10⁹/L and the platelets count was equal to or greater than 75 multiplied by 10⁹/L.

If the patient recovered from toxicities requiring treatment interruption, the following occurred.

• During the 2-week active treatment intake period of a cycle (treatment day 1 to day 12):

  • if no dose reduction was required, trifluridine-tipiracil could be resumed during that cycle

  • if a dose reduction was required, trifluridine-tipiracil had to be resumed at the start of the next cycle at the appropriate dose level.

• During the rest period (day 13 to day 28):

  • the next cycle was to be started on schedule at the appropriate trifluridine-tipiracil dose level.

If the toxicities did not resolve during the given cycle to grade 1 or baseline, the start of the next cycle had to be delayed for a maximum of 28 days from the scheduled start date of the next cycle. If more than 28 days were needed to recover, the patient had to be withdrawn from treatment.

Dose Discontinuation of Bevacizumab Due to Toxicities

Dose reduction for adverse reactions was not recommended. If indicated, therapy had either to be permanently discontinued or suspended temporarily. If a medical condition developed that required bevacizumab to be permanently withdrawn, the patient could continue with trifluridine-tipiracil alone.

Concomitant Treatments

Concomitant treatments that were permitted during the study treatment included palliative radiotherapy (irradiation of target lesions was to be avoided), effective contraception methods, and the management of toxicities of the study treatment (e.g., antiemetic regimen, antidiarrheal therapy, G-CSF use). Concomitant G-CSF treatment was permitted for primary prophylaxis (i.e., G-CSF at cycle 1, with or without neutropenia at cycle 1), secondary prophylaxis (i.e., G-CSF at any subsequent cycle to cycle of neutropenia), and therapeutic use (i.e., G-CSF on or 1 day after neutropenia onset date); patients with both secondary prophylaxis and therapeutic use of G-CSF were defined only as therapeutic use. Patients treated with anti–vitamin K were strongly recommended to be switched to low–molecular weight heparin. Caution was directed when using drugs that were human thymidine kinase substrates (e.g., zidovudine), which could have competed with the effector, trifluridine, for the activation of thymidine kinases.

Intercurrent Events

The following were considered intercurrent events in the SUNLIGHT study: the administration of further anticancer therapy, treatment discontinuation, or treatment switch (from trifluridine-tipiracil plus bevacizumab to trifluridine-tipiracil alone, and vice versa).

Withdrawal Criteria

Patients could discontinue treatment for the following reasons: AEs that were incompatible with the continuation of study treatment as judged by the investigator, protocol deviation that interfered with study evaluations and/or the patient’s safety, radiologic progressive disease documented by CT scan or MRI, clinical progressive disease evident by symptomatic deterioration, a nonmedical reason (e.g., consent withdrawal, patient’s removal), and other reasons, including physician decision. A withdrawal visit was to take place within 4 weeks after the date of study treatment withdrawal and before the start of a new anticancer therapy.

Outcomes

Summarized end points are based on outcomes included in the sponsor’s Summary of Clinical Evidence as well as any outcomes identified as important to this review, according to the clinical experts consulted by CADTH and stakeholder input from patient and clinician groups and public drug plans. Using the same considerations, the CADTH review team selected end points that were considered the most relevant to inform CADTH’s expert committee deliberations and finalized this list of end points in consultation with members of the expert committee. All summarized efficacy end points in Table 8 were assessed using GRADE. Select notable harms outcomes considered important for informing CADTH’s expert committee deliberations were also assessed using GRADE.

Overall Survival

In the SUNLIGHT trial, the primary efficacy end point was OS, defined as the time elapsed between the date of randomization and the date of death due to any cause. For patients without documentation of death, the OS was censored at the last contact date the patient was known to be alive. Based on the consensus from Colorectal Cancer Canada, an improvement of 2 months or more in median OS or an HR for survival of 0.75 or lower (or both) are the threshold for a clinically meaningful benefit.⁴⁶

Progression-Free Survival

In the SUNLIGHT trial, the key secondary end point was PFS based on investigator assessment. PFS was defined as the time elapsed between the date of randomization and the date of radiologic tumour progression according to Response Evaluation Criteria in Solid Tumours Version 1.1 (RECIST 1.1) or death from any cause.

Health-Related Quality of Life

In the SUNLIGHT trial, HRQoL was measured using the EORTC QLQ-C30 and the EQ-5D-5L.

The EORTC QLQ-C30 is a multidimensional, cancer-specific, self-administered measure of HRQoL consisting of 30 questions across 5 multi-item functional scales, 3 multi-item symptom scales, 6 single-item symptom scales, and a 2-item global health status scale. Each item is evaluated using 4-point and 7-point Likert scales, raw scores for each scale are computed as the average of the items that contribute to a particular scale, and each raw scale score is converted to a standardized score that ranges from 0 to 100 using a linear transformation. Higher scores in the global health status and functioning scales indicate better HRQoL, whereas higher scores in the symptom scales indicate poorer HRQoL.^47-50 The change in score from baseline in the global health score was the primary HRQoL end point. In the SUNLIGHT trial, an absolute change from baseline greater than 10 points in the EORTC QLQ-C30 was considered a clinically meaningful difference.^22,51,52 A study of 21 European Organisation for Research and Treatment of Cancer phase III trials of patients across 9 cancer types estimated that for CRC (1,491 patients), most MIDs were approximately 5 points to 10 points, with larger MIDs for improvements than for deterioration across most scales in within-group and between-group analyses.⁵³ A between-group difference of 10 points was identified by the sponsor to be a MID.^52,54

The EQ-5D-5L is a generic, preference-based measure of HRQoL consisting of 5 domains: mobility, self-care, usual activities, pain/discomfort, and anxiety/depression; index scores range from 0 to 1, with 0 and 1 representing the health states “dead” and “perfect health,” respectively, while EQ VAS scores range from 0 to 100, with 0 and 100 representing “worst imaginable health” and ”best imaginable health,” respectively.⁵⁵ No MID for the EQ-5D-5L utility or EQ VAS was identified by the sponsor.

In the SUNLIGHT trial, HRQoL data (i.e., EORTC QLQ-C30 and EQ-5D-5L) were collected at baseline (within 7 days before randomization), at every cycle before any study procedure, and at the withdrawal visit. Questionnaires were completed on an electronic patient-reported outcomes tablet by patients (independent of study personnel) or by a caregiver who read the questions to the patient and recorded responses. No MID for the EQ-5D-5L utility or EQ VAS was identified by the sponsor.

Harms

In the SUNLIGHT trial, safety end points included AEs, SAEs, concomitant treatments, the assessment of laboratory tests, physical examinations, vital signs, and the time from randomization to worsening of the ECOG PS score from 0 or 1 to 2 or more. AEs were defined as any untoward medical occurrence in a patient, whether or not there was a causal relationship with the study treatments and/or experimental procedures, occurring or detected from the date the patient signed the information and consent form, irrespective of the period of the study. An AE could include any unfavourable and unintended sign (e.g., abnormal finding from an additional examination such as lab tests, X-rays, electrocardiogram) deemed clinically relevant by the investigator, any symptom or disease, and any worsening during a study visit at an additional examination or that had occurred since the previous study visit. SAEs included any AE that at any dose resulted in death, was life-threatening, required inpatient hospitalization, or required the prolongation of existing hospitalization, was medically significant, resulted in persistent or significant disability or incapacity, or was a congenital anomaly or birth defect. A data safety monitoring board was responsible for reviewing the safety data on a regular basis and providing written recommendations to the sponsors regarding the conduct of the study.

Table 8: Outcomes Summarized From the Studies Included in the Systematic Review

EORTC QLQ-C30 = European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire Core 30; EQ-5D-5L = 5-Level EQ-5D; NA = not applicable; RECIST 1.1 = Response Evaluation Criteria in Solid Tumours Version 1.1.

Note: Details included in Table 8 are from the sponsor’s Summary of Clinical Evidence.³³

^aStatistical testing for these end points was adjusted for multiple comparisons (e.g., hierarchical testing).

^bThis outcome was identified as important by input from patient groups and clinical experts consulted by CADTH but was not measured in the SUNLIGHT trial.

Source: SUNLIGHT Clinical Study Report.²²

Considerations that informed the selection of efficacy outcomes to be summarized and assessed using GRADE included the following.

• Survival outcomes were identified by patient input (as OS, including tumour shrinkage and tumour stability), clinician group input (as extending survival, delaying further progression, and shrinking tumour burden), clinical experts consulted by CADTH (as OS at 6 months and 12 months, and PFS at 3 months and 6 months), and in the sponsor’s pharmacoeconomic model (as OS and PFS). OS and PFS were also identified as relevant in a previous CADTH review of trifluridine-tipiracil.⁵⁶ PFS was acknowledged by the clinical experts to provide supportive evidence of OS as a primary end point in the SUNLIGHT trial. The clinical experts identified an MID of 10% between treatment groups to be clinically important.

• HRQoL measured using the EORTC QLQ-C30 was identified by patients (as improved mobility, improved sense of wellness), clinician group input (as minimizing symptoms and symptom burden and being able to maintain independence), and clinical experts consulted by CADTH, and was identified as relevant in a previous CADTH review of trifluridine-tipiracil.⁵⁶ In the EORTC QLQ-C30, functioning scales (physical and emotional) and symptom scales (fatigue, nausea and vomiting, pain, dyspnea, appetite loss, and diarrhea) were identified by patient input and clinical experts consulted by CADTH to be important. Given the breadth of subscales, the CADTH review team focused on the global health status scale, to be assessed using GRADE to encompass the broad range of functioning and symptoms identified by stakeholders. Anchor-based MIDs for between-group differences in change on the global health status were estimated to range from 5.53 to 6.36 (weighted value of 5.86) for improvement and range from –9.21 to –6.81 (weighted value of –8.13) for deterioration among patients with advanced CRC treated with chemotherapy.⁵⁷

• HRQoL measured using the EQ-5D-5L was a key input in the sponsor’s pharmacoeconomic model and was identified as relevant in a previous CADTH review of trifluridine-tipiracil.⁵⁶ Acknowledging the lack of an MID for EQ-5D-5L in the published literature for patients with mCRC, the clinical experts identified a greater than 0.08-point difference between groups for the EQ-5D-5L utility score and a greater than 7-point difference between groups for the EQ VAS score based on MIDs reported for patients with cancer.⁵⁸

• Caregiver burden was identified as important by patient input and the clinical experts consulted by CADTH. No data on caregiver burden were collected in the SUNLIGHT trial.

• Harms of treatment were identified as important by patient input, clinician group input, and the clinical experts consulted by CADTH. Bone marrow suppression was identified by patient input (as nausea, anemia, neutropenia), by the clinical experts consulted by CADTH, and by the product monograph¹⁷ for trifluridine-tipiracil (including grade 3 or grade 4 neutropenia, leukopenia, anemia, thrombocytopenia, and febrile neutropenia). Gastrointestinal symptoms (including nausea, vomiting, and diarrhea) were identified by patient input, by the clinical experts consulted by CADTH (including gastrointestinal perforation), and by the product monograph¹⁷ for trifluridine-tipiracil. Infections were also identified as important by clinical experts consulted by CADTH. Hypertension was identified as important by the CADTH pan-Canadian Oncology Drug Review Expert Review Committee. The clinical experts identified an MID of 10% between treatment groups to be clinically important for bone marrow suppression, gastrointestinal symptoms, infections, and hypertension.

Statistical Analysis

Sample Size and Power Calculation

The sample size considerations for OS were based on a maximum of 331 events (deaths for any cause) for the primary analysis to detect an HR of 0.70 with 90% power using a log-rank test at a 1-sided cumulative 2.5% level of significance. Based on the data from the RECOURSE study,⁴⁴ the median OS in the control group was expected to be around 7.1 months. An HR of 0.70 translates into a 3-month increase of the median OS in the experimental group (10.1 months) compared to the control group. Based on the assumption that enrolment would have continued for approximately 12 months, and that an estimated 5% per year of the patients would drop out, a total of 490 patients randomized in a 1:1 ratio was needed to observe the 331st OS event approximately 9 months after the last patient randomization.

No interim analyses were planned for the primary end point of OS.

Statistical Testing

Statistical analysis of efficacy outcomes in the SUNLIGHT trial were controlled for overall type I error rate by using a hierarchical testing strategy. PFS was to be statistically evaluated and interpreted only if the primary efficacy end point of OS was significantly different between the 2 treatment groups.

Analysis of OS

The primary efficacy end point was the effect of the randomized treatments on OS in all patients regardless of whether or not intercurrent events occurred. In line with the intention-to-treat principle, further anticancer therapies were assumed to represent clinical practice and therefore, intercurrent events were considered to be part of the treatments being compared. All data collected during the trial, regardless of occurrence of an intercurrent event, were used.

The distribution of OS was compared between the 2 treatment groups using a stratified log-rank test at a 1-sided 2.5% level of significance (stratification factors were based on IWRS data). OS for each group was estimated using KM curves and further characterized in terms of the median and survival probabilities at 6 months, 12 months, and 18 months with the corresponding 2-sided 95% CIs. The HR of OS with its 95% CI was estimated with a stratified Cox proportional hazards model (stratification factors were based on IWRS data). Underlying assumptions of proportional hazards were checked using a Schoenfeld residuals test and graphical methods (i.e., log cumulative hazard curve): if the curve of the estimated treatment effect at each time and its CI deviate away from the horizontal, a time dependent effect is present. A statistical test using the Grambsch and Therneau (1994){GRAMBSCH, 1994 #119} test based on Schoenfeld residuals was performed. If the P value of the test was below 0.05, the null hypothesis (proportional hazards) was rejected and the proportional hazards assumption was not validated.

Analysis of PFS

The key secondary end point was the effect of the randomized treatments on PFS in all patients regardless of whether or not intercurrent events occurred (treatment policy strategy). The PFS was based on the investigator’s judgment and was defined as the time elapsed between the date of randomization and the date of radiologic tumour progression (according to RECIST 1.1) or death from any cause. All data collected during the trial, regardless of the occurrence of an intercurrent event, were used.

A hierarchical testing strategy, where PFS was to be statistically evaluated and interpreted only if the primary efficacy estimand OS was significantly different between the 2 treatment groups, was used to control the overall type I error rate. The distribution of PFS was compared between the 2 treatment groups using a stratified log-rank test at a 1-sided 2.5% level of significance (stratification factors were based on IWRS data). PFS for each group was estimated using KM curves and further characterized in terms of median and survival probabilities at 3 months, 6 months, 9 months, and 12 months with the corresponding 2-sided 95% CIs. The HR of PFS with its 95% CI was estimated with a stratified Cox proportional hazards model (stratification factors were based on IWRS data).

Censoring

For missing OS data not linked to intercurrent events (i.e., patients without documentation of death [lost to follow-up, withdrawal of consent, or administrative end of study]), OS was censored on the last contact date the patient was known to be alive or the survival cut-off date, whichever was earlier.

For missing PFS data not linked to intercurrent events (i.e., patients who were lost to follow-up or who had withdrawn their consent without radiologic progression or had reached the time point of analysis without a known record of death or radiologic progression), PFS was censored at the date of the last evaluable tumour assessment or the survival cut-off date, whichever was earlier.

Subgroup Analyses

Subgroup analyses for OS and PFS were performed in the FAS using an unstratified Cox-regression model with treatment group as predictor variable fitted separately for each subgroup category. The HR for treatment and associated 95% CIs were reported. No statistical testing was conducted to evaluate differences in effects between subgroups, and analyses were not adjusted for multiplicity. Prespecified subgroups included the following:

• region (North America, European Union, the rest of the world)

• time since diagnosis of first metastasis (< 18 months, ≥ 18 months)

• RAS status (wild-type, mutant)

• location of primary disease (right, left)

• ECOG PS (0, ≥ 1)

• sex (female, male)

• age (< 65, ≥ 65 years)

• prior surgical resection (yes, no)

• number of metastatic sites (1 to 2, ≥ 3)

• neutrophils to lymphocytes ratio (< 3, ≥ 3)

• number of prior metastatic drug regimens (1, ≥ 2)

• BRAF status (wild-type, mutant)

• microsatellite instability status (MSI-H, microsatellite stable, or microsatellite instability–low)

• prior bevacizumab (yes, no)

• subsequent regorafenib (yes, no).

Sensitivity Analyses

The following sensitivity analyses of OS were performed.

• The distribution of OS was compared between the treatment groups using an unstratified log-rank test and the HR (together with associated 95% CI) resulting from an unstratified Cox model.

• The distribution of OS excluding patients not fulfilling 1 of the following criteria was compared between the treatment groups using a stratified log-rank test (stratification factors were based on IWRS data) and the HR (together with associated 95% CI) resulting from a stratified Cox proportional hazards model (stratification factors were based on IWRS data):

  • has histologically confirmed unresectable adenocarcinoma of the colon or rectum

  • has received a maximum of 2 prior chemotherapy regimens for the treatment of advanced CRC and had demonstrated progressive disease or intolerance to their last regimen

  • has measurable or nonmeasurable metastatic lesion(s) as defined by RECIST 1.1

  • has an ECOG PS of 1 or less

  • has more than 2 prior chemotherapy regimens for the treatment of advanced CRC

  • has previously received trifluridine-tipiracil

  • was not undergoing or has not received anticancer therapies within 4 weeks before randomization.

• The restricted mean survival time for OS was reported with its 95% CI in each treatment group. The difference in restricted mean survival time between treatment groups was tested.

The following sensitivity analyses of PFS were performed.

• The distribution of PFS was compared between treatment groups using an unstratified log-rank test and the HR (with associated 95% CI) resulting from an unstratified Cox model.

• Analysis was conducted including clinical progression and the administration of further anticancer therapy as PFS events in addition to the radiologic progression or death.

• Analysis was conducted censoring the PFS duration at the date of last evaluable tumour assessment before or on the day of administration of further anticancer therapy and censoring patients with radiologic progression or who die after 2 or more consecutive missing radiological assessments at the latest evaluable assessment before the missing assessments. Based on the time interval between the last evaluable tumour assessment date and the event date, if the interval was greater than (2 multiplied by 8) plus 2 equals 18 weeks (protocol-specified threshold), then the analysis would assume 2 missing assessments.

Secondary End Points

No non–key secondary end points were adjusted for multiplicity.

Health-Related Quality of Life

European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire Core 30: Patients in the FAS who have completed at least 1 questionnaire item at baseline and during the study period were analyzed in the group they were assigned at randomization. The raw data were scored according to the European Organisation for Research and Treatment of Cancer scoring manual. Descriptive statistics (number of patients, mean, SD, median, range) were used to summarize values at baseline and at postbaseline, and the change from baseline in the global health status and subscale scores at each scheduled assessment time point and by classes (benefit [large, moderate, small], no change, and deterioration [small, moderate, large]) for each treatment group. A mixed model of repeated measures that included terms for treatment, baseline stratification factors, baseline score, time to visit before any procedure (at each cycle, including the withdrawal visit), and treatment groups by time to visit interaction was used to compare change from baseline subscale scores longitudinally over time between treatment groups. The most suitable covariance structure was chosen according to Akaike’s information criterion and Schwartz’s Bayesian Criterion between unstructured, compound symmetric, and autoregressive of order 1.

Time until definitive deterioration was defined as the interval between baseline and the first worsening in HRQoL score of 10 points or greater compared to the baseline HRQoL score with no later improvement above this threshold observed during the study. Time until definitive deterioration of 10 points for each subscale was compared between the treatment groups using the stratified log-rank test at a 2-sided 5% level of significance. Distributions were described using KM curves, including the median time to a definitive 10 points’ deterioration with 95% CIs. Death was considered as deterioration of HRQoL. Patients who received any further antitumoural treatment before worsening were censored at the date of their last HRQoL assessment before starting such therapy. Patients who did not definitively worsen as of the cut-off date for the analysis were censored at the date of their last assessment before the cut-off date. Patients with a missing scale score at baseline were censored at the randomization date plus 1 day.

5-Level EQ-5D: Patients in the FAS with an evaluable EQ-5D-5L assessment at baseline and at least 1 evaluable assessment postbaseline were analyzed in the group they were assigned at randomization. For data from all countries, the EQ-5D-5L index utility score was derived according to the French value set. Descriptive statistics (number of patients, mean, SD, median, range) were used to summarize values at baseline and the change from baseline in EQ-5D-5L index utility and EQ VAS by treatment group at each scheduled assessment time point before any study procedure. A mixed model of repeated measures that included terms for treatment, baseline stratification factors, baseline score, time to visit before any procedure (at each cycle, including the withdrawal visit), and treatment groups by time to visit interaction was used to compare change from baseline scores between treatment groups. The most suitable covariance structure was chosen according to Akaike’s information criterion and Schwartz’s Bayesian Criterion between unstructured, compound symmetric, and autoregressive of order 1.

Safety Analyses

Descriptive statistics (number of events, number and percentage of patients reporting at least 1 event) were summarized for each treatment group for emergent AEs, SAEs, withdrawals due to AEs, and deaths. Emergent AEs were to be described for each treatment group according to worst grade, severity, relationship to treatment (related to trifluridine-tipiracil, related to bevacizumab, or related to trifluridine-tipiracil plus bevacizumab), relationship to disease progression, whether they led to death, or whether they required added therapy. The seriousness and relationship with trifluridine-tipiracil and/or bevacizumab were based on investigator opinion and included sponsor decision to upgrade seriousness and/or relationship. Reasons for deaths were to be reported during the treatment and follow-up periods.

Analysis Populations

Analysis sets included in the SUNLIGHT trial are summarized in Table 10.

Table 9: Statistical Analysis of Efficacy End Points

CI = confidence interval; EORTC QLQ-C30 = European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire Core 30; EQ-5D-5L = 5-Level EQ-5D; EQ VAS = EQ visual analogue scale; HR = hazard ratio; KM = Kaplan-Meier; MMRM = mixed model of repeated measures; PFS = progression-free survival; QoL = quality of life; RMST = restricted mean survival time.

Note: Details included in Table 9 are from the sponsor’s Summary of Clinical Evidence.³³

Source: SUNLIGHT Clinical Study Report.²²

Table 10: Analysis Populations in the SUNLIGHT Trial

FAS = full analysis set; IWRS = Interactive Web Response System; SAS = safety analysis set.

Note: Details included in Table 10 are from the sponsor’s Summary of Clinical Evidence.³³

Source: SUNLIGHT Clinical Study Report.²²

Results

Patient Disposition

A summary of patient disposition is presented in Table 11.

Of 659 patients screened, 167 (25.3%) patients did not pass screening, resulting in 492 patients being randomized into the SUNLIGHT trial. Most (24.3%) patients did not pass screening due to eligibility criteria not being met. By the cut-off date of July 5, 2022, for clinical (nonsurvival) data, all randomized patients had received study treatment; of these, 456 (92.7%) patients had discontinued treatment and 36 (7.3%) patients were still on treatment. The proportion of patients who discontinued randomized treatment was lower in the trifluridine-tipiracil plus bevacizumab group (87.0%) compared with the trifluridine-tipiracil alone group (98.4%). The most common reason for patients to discontinue study treatment was clinical and/or radiological progressive disease in the trifluridine-tipiracil plus bevacizumab (77.6%) group and the trifluridine-tipiracil alone (88.6%) group. Reasons for study discontinuation occurred with similar frequencies between groups, except for lower frequencies in the trifluridine-tipiracil plus bevacizumab group versus the trifluridine-tipiracil alone group for radiological and clinical progressive disease (10.6% versus 21.1%, respectively). The number of patients in the FAS and safety analysis set were identical (246 patients each in the trifluridine-tipiracil plus bevacizumab group and the trifluridine-tipiracil alone group).

Table 11: Summary of Patient Disposition in the SUNLIGHT Trial — Full Analysis Set

FAS = full analysis set; NR = not reported; SAS = safety analysis set.

Source: SUNLIGHT Clinical Study Report.²²

Baseline Characteristics

The baseline characteristics outlined in Table 12 are limited to those that are most relevant to this review or were felt to affect the outcomes or interpretation of the study results.

Table 12: Summary of Baseline Characteristics in the SUNLIGHT Trial — Full Analysis Set

Anti-EGFR = anti–epidermal growth factor receptor; anti-VEGF = anti–vascular endothelial growth factor; ECOG PS = Eastern Cooperative Oncology Group Performance Status; IWRS = Interactive Web Response System; MMR = mismatch repair; MSI = microsatellite instability; NR = not reported; SD = standard deviation.

Note: Details included in Table 12 are from the sponsor’s Summary of Clinical Evidence.³³

Source: SUNLIGHT Clinical Study Report.²²

The demographic characteristics were similar between the trifluridine-tipiracil plus bevacizumab group and the trifluridine-tipiracil alone group. However, there was a higher proportion of patients aged younger than 65 years in the trifluridine-tipiracil plus bevacizumab group compared with the trifluridine-tipiracil alone group (59% versus 52%, respectively). More patients were aged 65 years and older in the trifluridine-tipiracil alone group (48%) versus the trifluridine-tipiracil plus bevacizumab group (41%). A slightly higher proportion of patients had a primary tumour located on the right side in the trifluridine-tipiracil alone group (31%) compared with the trifluridine-tipiracil plus bevacizumab group (25%). Conversely, more patients had a left-sided tumour in the trifluridine-tipiracil plus bevacizumab group compared with the trifluridine-tipiracil alone group (75% versus 69%, respectively).

The mean age was 61.7 years (SD = 11.1 years) with 217 of 492 (44.1%) patients aged 65 years and older and 58 (11.8%) patients aged 75 years and older. Overall, 52.0% of patients were males and most patients enrolled were from the European Union (64.0%) and white (95.2%). Most patients had a primary diagnosis of colon cancer (73%) and stage IV disease (66%). The time from the diagnosis of the first metastasis until randomization was 18 months or longer in 57.5% of the patients, and 30.7% of the patients had RAS wild-type disease. Most patients (92.1%) had received 2 previous treatment regimens for metastatic disease; however, 4.5% of the patients in the trifluridine-tipiracil plus bevacizumab group and 6.1% of the patients in the trifluridine-tipiracil alone group had received only 1 first-line triplet regimen, and 2.6% of the patients in the trial had received 3 or more previous drug regimens for metastatic disease. All patients had received previous fluoropyrimidine-based therapy, 72.0% of patients had received previous anti-VEGF therapy (47.8% of patients had received bevacizumab as part of their first regimen, 43.9% of patients as part of their second regimen, and 20.3% of patients as part of both their first and second regimens), and 93.7% of patients with RAS wild-type disease had received previous anti-EGFR therapy.

Exposure to Study Treatments

Treatment exposure in the SUNLIGHT trial is summarized in Table 13. The median duration of treatment was 5.0 months (range = 0.1 months to 18.5 months) in the trifluridine-tipiracil plus bevacizumab group and 2.1 months (range = 0.6 months to 14.3 months) in the trifluridine-tipiracil alone group. ||| |||||||||| |||| |||| |||| || |||||||||||||||||||||| ||| |||||| || ||| ||||||||||| ||||| ||||| ||||||| ||| |||||||| |||||||||||||||||||||||||| ||||| ||||| ||||||| |||| In the trifluridine-tipiracil plus bevacizumab group, the median relative dose intensities of trifluridine-tipiracil was 88.3% and the median relative dose intensities of bevacizumab was 87.6%. The median relative dose intensity of trifluridine-tipiracil in the trifluridine-tipiracil alone group was 90.4%.

Table 13: Summary of Patient Exposure in the SUNLIGHT Trial

SD = standard deviation.

Note: Details included in Table 13 are from the sponsor’s Summary of Clinical Evidence.³³

^aTreatment adherence was assessed using the relative dose intensity, defined as the ratio of the dose intensity to the planned starting dose intensity at inclusion.

Source: SUNLIGHT Clinical Study Report.²²

Concomitant Medications and Cointerventions

A summary of concomitant medications by pharmacological class is presented in Table 14.

In the FAS, 84.3% of patients reported at least 1 concomitant treatment at inclusion, with similar frequency in the 2 treatment groups. Most patients (95.5%) received at least 1 concomitant treatment during the treatment period. The most frequent pharmacological classes (≥ 30%) were analgesics (49.0%), drugs for acid-related disorders (38.4%), and drugs acting on the renin-angiotensin system (35.2%). Among other concomitant treatments, the frequency of the therapeutic class “Other viral vaccines” was higher in the trifluridine-tipiracil plus bevacizumab group than in the trifluridine-tipiracil group (28.0% versus 11.8%, respectively); all were COVID-19 vaccines. Of note, 72 (29.3%) patients in the trifluridine-tipiracil plus bevacizumab group and 48 (19.5%) patients in the trifluridine-tipiracil group received at least 1 concomitant G-CSF, including for primary prophylaxis (0.8% and 1.2%, respectively), secondary prophylaxis (22.0% and 9.8%, respectively), therapeutic use (i.e., neutropenia) (16.3% and 11.8%, respectively), and for another indication (3.3% and 2.9%, respectively). There was no relevant between-group difference for the other classes of concomitant treatments at inclusion.

Table 14: Concomitant Medications in the SUNLIGHT Trial

Source: SUNLIGHT Clinical Study Report.²²

Subsequent Treatment

|||| ||||||||||| |||||||||| || ||||||||||||||| ||||| ||| ||||||||| || ||||| ||| |||| |||||||| ||||| ||||||| ||| |||||||||||| ||||||| ||||| |||||||| ||| |||||||||||||| || ||||||| ||||||||||| ||||||| ||| ||| ||||||||| |||||| |||| |||||||||||||||||||||| || ||||||||||| |||| ||||||||||| || |||||||||||||||||||||| ||||||| ||| |||| |||||||||||||||||||||| ||||||| || |||||||||||||||||||||| || ||||||||||| |||| ||||||||||| |||||| ||||||| ||||| ||||||||| |||| |||||| |||||||.

Table 15: Summary of Subsequent Treatment in the SUNLIGHT Trial

Note: Details included in Table 15 are from the sponsor’s Summary of Clinical Evidence.³³

^aIncludes detoxifying drugs for antineoplastic treatment (including any of folinic acid, calcium folinate, calcium levofolinate, levofolinate sodium, levofolinic acid, or sodium folinate).

^bNot specified.

Source: SUNLIGHT Clinical Study Report.²²

Efficacy

Findings for key efficacy outcomes in the SUNLIGHT trial are summarized in Table 16.

Overall Survival

At the survival cut-off date of July 19, 2022, the median follow-up was 14.2 months (interquartile range = 12.6 months to 16.4 months) in the trifluridine-tipiracil plus bevacizumab group and 13.6 months (interquartile range = 12.7 months to 15.9 months) in the trifluridine-tipiracil group (Figure 3). The number of patients who had died was 148 of 246 (60.2%) patients in the trifluridine-tipiracil plus bevacizumab group and 183 of 246 (74.4%) patients in the trifluridine-tipiracil alone group. The number of patients who were censored (alive at the time of data cut-off) was 98 (39.8%) patients and 63 (25.6%) patients in the trifluridine-tipiracil plus bevacizumab group and the trifluridine-tipiracil alone group, respectively. OS at 6 months among patients in the FAS population was 0.77 (95% CI, 0.72 to 0.82) and 0.61 (95% CI, 0.55 to 0.67) for the trifluridine-tipiracil plus bevacizumab group and the trifluridine-tipiracil alone group, respectively. OS at 12 months was 0.43 (95% CI, 0.36 to 0.49) and 0.30 (95% CI, 0.24 to 0.36) for the trifluridine-tipiracil plus bevacizumab group and the trifluridine-tipiracil alone group, respectively. OS at 18 months was 0.28 (95% CI, 0.19 to 0.37) and 0.15 (95% CI, 0.09 to 0.22) for the trifluridine-tipiracil plus bevacizumab group and the trifluridine-tipiracil alone group, respectively. The median OS was 10.78 months (95% CI, 9.36 months to 11.83 months) in the trifluridine-tipiracil plus bevacizumab group and 7.46 months (95% CI, 6.34 months to 8.57 months) in the trifluridine-tipiracil alone group. The HR was 0.61 (95% CI, 0.49 to 0.77), corresponding to a 39% reduction (P < 0.001) in the risk of death for trifluridine-tipiracil plus bevacizumab when compared with trifluridine-tipiracil alone.

Sensitivity Analyses

||||||||||| |||||||| || || ||||| ||| ||| |||||||||| ||||| || |||||||||||| |||||||| ||||||||||||| |||||| ||| ||| |||| || |||||||||| ||| ||||||||| |||||||| ||| ||| ||| ||||||| ||| || ||| ||||||||||| |||||||| ||||||| ||| ||| |||| ||||||||||||| |||| |||||||||| |||| ||| ||||||| |||||||| || || || ||| |||| ||||||||||| |||||||| ||| |||| |||||||| |||| |||||||||||||| ||||||| ||| || |||||||||||||||||||| || |||| |||| ||| |||| || |||||||||| ||||||||||||||||||||||||| |||| ||||||||||| |||| |||||||| |||| |||||||||||||||||||||| |||||.

Subgroup Analyses

A forest plot of the HRs and associated 95% CIs for subgroup analyses of OS are presented in Figure 4 and Figure 5. The CADTH review focused on the subgroups of location of primary disease, ECOG PS, age, number of metastatic sites, and number of prior metastatic drug regimens.

The HRs for primary OS favoured treatment with trifluridine-tipiracil plus bevacizumab compared with trifluridine-tipiracil alone among patients with:

• left-sided primary disease (n = 108 of 184 versus n = 120 of 169, respectively; unstratified HR = 0.65 [95% CI, 0.50 to 0.85]) and right-sided primary disease (n = 40 of 62 versus n = 63 of 77, respectively; unstratified HR = 0.59 [95% CI, 0.40 to 0.87])

• ECOG PS of 1 or greater (n = 78 of 127 versus n = 109 of 140, respectively; unstratified HR = 0.54 [95% CI, 0.41 to 0.73])

• age younger than 65 years (n = 89 of 146 versus n = 94 of 129, respectively; unstratified HR = 0.65 [95% CI, 0.48 to 0.87]) and age 65 years and older (n = 59 of 100 versus n = 89 of 117, respectively; unstratified HR = 0.59 [95% CI, 0.42 to 0.81])

• One to 2 metastatic sites (n = 83 of 152 versus n = 97 of 141, respectively; unstratified HR = 0.62 [95% CI, 0.46 to 0.83]) and 3 or more metastatic sites (n = 65 of 94 versus n = 86 of 105, respectively; unstratified HR = 0.66 [95%CI, 0.47 to 0.91])

• Two or more prior metastatic drug regimens (n = 142 of 235 versus n = 172 of 231, respectively; unstratified HR = 0.64 [95% CI, 0.51 to 0.80]).

Additional Analyses

|||||||||| |||||||| ||||||||| ||| |||||| || ||||| ||||||||| ||||||| |||||| |||| ||||||| ||| ||||||||| ||||||| || ||| |||||||||||||| || ||| ||||||||||| |||||||| |||||| |||| |||||||| |||||| ||||||| ||||||||||| ||||||| ||| || ||||||| |||||||| || ||| |||||||||||||||||||||| |||| ||||||||||| ||||| ||| ||| ||||||| |||||||| || ||| |||||||||||||||||||||| ||||| |||||| ||| |||||| || ||| |||| |||||| |||||| ||| |||||| ||| |||||||||||||||||||||| |||| ||||||||||| |||||| |||||||||||||||||||||| |||||| |||||||||||| |||||| ||| ||| |||| || |||||||||| ||||||||||| || || |||||||| || ||||||| ||| || |||||| ||||||| |||||||||| ||| ||| ||||||||||| ||||||| |||| |||||||||| |||| ||| ||||||| |||||||| || || || ||| ||| ||||||||||.

Figure 3: Kaplan-Meier Curves of Overall Survival in the SUNLIGHT Trial, Data Cut-Off Date of July 19, 2022 — Full Analysis Set

OS = overall survival; S95005 = trifluridine-tipiracil.

Source: SUNLIGHT Clinical Study Report.²²

Progression-Free Survival

At the survival cut-off date of July 19, 2022, the number of patients with events (those who had radiological progressive disease or had died) were 206 of 246 (83.7%) patients and 236 of 246 (95.9%) patients in the trifluridine-tipiracil plus bevacizumab group and the trifluridine-tipiracil alone group, respectively. The following are the number of patients who were censored in the trifluridine-tipiracil plus bevacizumab group and the trifluridine-tipiracil alone group, respectively, due to being alive without radiological disease progression (38 [15.5%] patients versus 5 [2.0%] patients, respectively), lost to follow-up (2 [0.8%] patients versus 4 [1.6%] patients, respectively), or without baseline or postbaseline tumour assessment (0 versus 1 [0.4%] patient, respectively). PFS among patients in the FAS at 3 months was 0.73 (95% CI, 0.67 to 0.78) in the trifluridine-tipiracil plus bevacizumab group versus 0.45 (95% CI, 0.39 to 0.51) in the trifluridine-tipiracil alone group. PFS at 6 months was 0.43 (95% CI, 0.37 to 0.49) and 0.16 (95% CI, 0.11 to 0.21) for the trifluridine-tipiracil plus bevacizumab group and the trifluridine-tipiracil alone group, respectively. PFS at 9 months was 0.28 (95% CI, 0.22 to 0.34) and 0.05 (95% CI, 0.03 to 0.09) for the trifluridine-tipiracil plus bevacizumab group and the trifluridine-tipiracil alone group, respectively. PFS at 12 months was 0.16 (95% CI, 0.12 to 0.21) and 0.01 (95% CI, 0.00 to 0.03) for the trifluridine-tipiracil plus bevacizumab group and the trifluridine-tipiracil alone group, respectively. Median PFS was 5.6 months (95% CI, 4.50 months to 5.88 months) in the trifluridine-tipiracil plus bevacizumab group and 2.4 months (95% CI, 2.07 months to 3.22 months) in the trifluridine-tipiracil alone group (Figure 6). The HR for PFS was 0.44 (95% CI, 0.36 to 0.54) for the trifluridine-tipiracil plus bevacizumab group when compared with the trifluridine-tipiracil alone group (P < 0.001).

Figure 4: Forest Plot of Hazard Ratios for Treatment Effect on Overall Survival by Selected Subgroups in the SUNLIGHT Trial, Data Cut-Off Date of July 19, 2022 — Full Analysis Set

CI = confidence interval; Dx = diagnosis; ECOG = Eastern Cooperative Oncology Group; IWRS = Interactive Web Response System; mets = metastasis; S95005 = trifluridine-tipiracil.

Source: SUNLIGHT Clinical Study Report.²²

Figure 5: Forest Plot of Hazard Ratios for Treatment Effect on Overall Survival by Selected Subgroups in the SUNLIGHT Trial, Data Cut-Off Date of July 19, 2022 — Full Analysis Set (Continued) [Redacted]

Source: SUNLIGHT Clinical Study Report.²²

Figure 6: Kaplan-Meier Curves of PFS in the SUNLIGHT Trial, Data Cut-Off Date of July 19, 2022 — Full Analysis Set

PFS = progression-free survival; S95005 = trifluridine-tipiracil.

Source: SUNLIGHT Clinical Study Report.²²

Sensitivity Analyses

||||||||||| |||||||| ||| ||| ||||| ||| ||| |||||||||| ||||| || |||||||||||| |||||||| |||||||| ||| ||| |||| || |||||| ||||||||| |||||||| ||||||||||| || ||| ||||||||||| ||||||| || ||| |||||| |||||| ||| ||| |||| || |||||| ||||||||||| ||| ||||||||||| ||||||| || |||| ||||||||||| ||||||| |||||| ||||||||||| || ||| ||||||| |||||| ||| ||| |||| || ||||| |||| |||||||||| |||| ||| ||||||| ||| ||||||||.

Subgroup Analyses

A forest plot of the HRs and associated 95% CIs for subgroup analyses of PFS are presented in Figure 7 and Figure 8.

The HRs (95% CIs) for primary PFS favoured treatment with trifluridine-tipiracil plus bevacizumab compared with trifluridine-tipiracil alone among patients with:

• left-sided disease (n = 155 of 184 versus n = 160 of 169, respectively; unstratified HR = 0.44 [95% CI, 0.35 to 0.56]) and right-sided primary disease (n = 51 of 62 versus n = 76 of 77, respectively; unstratified HR = 0.44 [95% CI, 0.31 to 0.64])

• ECOG PS of 0 (n = 96 of 119 versus n = 103 of 106, respectively; unstratified HR = 0.37 [95% CI, 0.27 to 0.49]) and ECOG PS of 1 or greater (n = 110 of 127 versus n = 133 of 140, respectively; unstratified HR = 0.50 [95% CI, 0.39 to 0.65])

• age younger than 65 years (n = 124 of 146 versus n = 122 of 129, respectively; unstratified HR = 0.41 [95% CI, 0.31 to 0.53]) and age 65 years and older (n = 82 of 100 versus n = 114 of 117, respectively; unstratified HR = 0.46 [95% CI, 0.34 to 0.62)]

• One to 2 metastatic sites (n = 121 of 152 versus n = 137 of 141, respectively; unstratified HR = 0.39 [95% CI, 0.31 to 0.51]) and 3 or more metastatic sites (n = 85 of 94 versus n = 99 of 105, respectively; unstratified HR = 0.54 [95% CI, 0.40 to 0.72])

• One prior metastatic drug regimen (n = 8 of 11 versus n = 13 of 15, respectively; unstratified HR = 0.37 [95% CI, 0.15 to 0.93]) and 2 or more prior metastatic drug regimens (n = 198 of 235 versus n = 223 of 231, respectively; unstratified HR = 0.44 [95% CI, 0.36 to 0.54]).

Figure 7: Forest Plot of Hazard Ratios for Treatment Effect on Progression-Free Survival by Selected Subgroups in the SUNLIGHT Trial, Data Cut-Off Date of July 19, 2022 — Full Analysis Set

CI = confidence interval; Dx = diagnosis; ECOG = Eastern Cooperative Oncology Group; IWRS = Interactive Web Response System; S95005 = trifluridine-tipiracil.

Source: SUNLIGHT Clinical Study Report.²²

Figure 8: Forest Plot of Hazard Ratios for Treatment Effect on Progression-Free Survival by Selected Subgroups in the SUNLIGHT Trial, Data Cut-Off Date of July 19, 2022 — Full Analysis Set (Continued) [Redacted]

Source: SUNLIGHT Clinical Study Report.²²

Time to Worsening of ECOG PS

In the FAS population, the number of patients with worsening of ECOG PS of 2 or greater postbaseline (from 0 to 1 at baseline) was 31 of 246 (20.0%) patients in the trifluridine-tipiracil plus bevacizumab group compared with 43 of 246 (23.4%) patients in the trifluridine-tipiracil alone group, including death for 124 (80.0%) patients and 141 (76.6%) patients in the trifluridine-tipiracil plus bevacizumab group and the trifluridine-tipiracil alone group, respectively. The median time to worsening of ECOG PS of 2 or greater was 9.3 months (95% CI, 8.34 months to 10.61 months) and 6.3 months (95% CI, 5.55 months to 7.23 months) in the trifluridine-tipiracil plus bevacizumab group and the trifluridine-tipiracil alone group, respectively.

Table 16: Summary of Primary and Key Efficacy Results in the SUNLIGHT Trial, Data Cut-Off Date of July 19, 2022 — Full Analysis Set

CI = confidence interval; HR = hazard ratio; IWRS = Interactive Web Response System.

Note: Details included in Table 16 are from the sponsor’s Summary of Clinical Evidence.³³

^aKaplan-Meier estimate.

^bMethodology of Brookmeyer and Crowley.

^cStratified log-rank test at 1-sided 2.5% level of significance using IWRS stratification factors (geographic region, time since diagnosis of first metastasis, and RAS status).

^dP value has been adjusted for multiple testing.

^eUsing the log-log transformation methodology of Kalbfleisch and Prentice.

^fStratified Cox proportional hazards model using IWRS stratification factors (geographic region, time since diagnosis of first metastasis, and RAS status).

Source: SUNLIGHT Clinical Study Report.²²

Health-Related Quality of Life

Findings for HRQoL outcomes in the SUNLIGHT trial at the data cut-off date of July 5, 2022, are summarized in Table 17.

European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire Core 30

EORTC QLQ-C30 and EQ-5D-5L analyses were performed in patients from the FAS with at least 1 questionnaire item at baseline and during the study period. In the EORTC QLQ-C30, higher scores in the global and functioning scales indicated better HRQoL, with positive change from baseline indicating benefit and negative change from baseline indicating deterioration. Conversely, for the symptom scales, lower scores indicated better HRQoL, with positive change from baseline indicating deterioration, and negative change from baseline indicating benefit. In the EQ-5D-5L utility and EQ VAS, higher scores indicated better HRQoL, with positive change from baseline indicating benefit and negative change from baseline indicating deterioration.

Global Health Status: The LSM change from baseline in the global health status was –2.85 (95% CI, –5.92 to 0.22) for trifluridine-tipiracil plus bevacizumab and –6.62 (95% CI, –10.36 to –2.88) for trifluridine-tipiracil alone. The LSM difference in change from baseline for global health status was 3.77 (95% CI, 0.22 to 7.32; P = 0.038) in favour of trifluridine-tipiracil plus bevacizumab. The number of patients with 10 points or greater definitive deterioration was 62 (25.2%) patients and 72 (29.3%) patients in the trifluridine-tipiracil plus bevacizumab group and the trifluridine-tipiracil alone group, respectively. The number of patients who died was 58 (23.6%) patients and 69 (28.1%) patients in the trifluridine-tipiracil plus bevacizumab group versus the trifluridine-tipiracil alone group, respectively. A total of 126 of 246 (51.2%) patients and 105 of 246 (42.7%) of patients in the trifluridine-tipiracil plus bevacizumab group versus the trifluridine-tipiracil alone group, respectively, were censored, including patients with the following: new anticancer therapy (83 [33.7%] patients versus 89 [36.2%] patients, respectively); 10 points or greater deterioration followed by an improvement, fewer than 10 points deterioration from baseline, or no deterioration in 10 points or greater (40 [16.3%] patients versus 13 [5.3%] patients, respectively); and being alive without a baseline or postbaseline evaluable questionnaire (3 [1.2%] patients versus 3 [1.2%] patients, respectively). Median survival based on time until definitive deterioration in the global health status was 8.54 months (95% CI, 7.49 months to 10.94 months) in the trifluridine-tipiracil plus bevacizumab group and 4.70 months (95% CI, 4.01 months to 5.78 months) in the trifluridine-tipiracil alone group (P < 0.001).

Functioning: ||| ||| |||||| |||| |||||||| |||| ||| || |||||||| ||||||||||| ||| ||||| |||||| || ||||| ||| |||||||||||||||||||||| |||| ||||||||||| ||| ||||| ||||||| || ||||| ||| |||||||||||||||||||||| |||||| ||| ||| |||||||||| |||| ||| || |||||| |||| |||||||| ||| |||||||| ||||||||||| ||| |||| ||||| || ||||||||| || |||||| || |||||||||||||||||||||| |||| |||||||||||| ||| |||||| || |||||||| || ||| ||| |||||||||| |||| || |||||| || ||||||| |||||||||| ||||||||||||| || |||||||| ||||||||||| |||| || ||||||| ||| || ||||||| || ||| |||||||||||||||||||||| |||| ||||||||||| ||| |||||||||||||||||||||| ||||||| |||||| ||||||||||||| |||||| |||| |||| ||| ||||| |||||||||| ||||||||||||| || |||||||| ||||||||||| ||| |||| |||||| ||||| || |||||| || ||| |||||||||||||||||||||| |||| ||||||||||| ||||| ||| |||| |||||| ||||| || ||||| || ||| |||||||||||||||||||||| ||||||| ||||| ||||||||||| ||| |||||| |||| |||||||| |||| ||| || ||||||||| ||||||||||| ||| |||| |||||| || ||||| ||| |||||||||||||||||||||| |||| ||||||||||| ||| ||||| |||||| || ||||| ||| |||||||||||||||||||||| |||||| ||| ||| |||||||||| |||| ||| || |||||| |||| |||||||| |||||||||||| ||||||||||| ||| |||| |||||| || |||||||||| ||| |||||| || |||||||| || ||| ||| |||||||||| |||| || |||||| || ||||||| |||||||||| ||||||||||||| || ||||||||| ||||||||||| |||| || ||||||| ||| || ||||||| || ||||||||||||||||||||||||| |||| ||||||||||| ||| |||||||||||||||||||||| ||||||| |||||| ||||||||||||| |||||| |||| |||| ||| ||||| |||||||||| ||||||||||||| || ||||||||| ||||||||||| ||| |||| |||||| ||||| || |||||| || ||| ||||||||||||||||||||| |||| ||||||||||| ||||| ||| |||| |||||| ||||| || ||||| || ||| |||||||||||||||||||||| ||||||| ||||||||||||||||||||| |||||||||| ||| |||||| |||| |||||||| |||| ||| || ||||||| ||| |||| |||||| || ||||| ||| |||||||||||||||||||||| |||| ||||||||||| ||| |||| ||||| || ||||| ||| |||||||||||||||||||||| |||||| ||| ||| |||||||||| |||| ||| || |||||| |||| |||||||| ||| ||||||| ||| ||||| |||||| || |||||||||||| || |||||| || |||||||||||||||||||||| |||| |||||||||||| ||| |||||| || |||||||| || ||| ||| |||||||||| |||| || |||||| || ||||||| |||||||||| ||||||||||||| || ||||||| |||| || ||||||| ||| || ||||||| || ||| |||||||||||||||||||||| |||| ||||||||||| ||| |||||||||||||||||||||| ||||||| |||||| ||||||||||||| |||||| |||| |||| ||| ||||| |||||||||| ||||||||||||| || ||||||||| ||||||||||| ||| |||| |||||| ||||| || ||||| || ||| |||||||||||||||||||||| |||| ||||||||||| ||||| ||| |||| |||||| ||||| || ||||| || ||| |||||||||||||||||||||| ||||||| ||||| ||||||||||| ||| |||||| |||| |||||||| |||| ||| || |||||| ||| |||||||| ||| |||| ||||| || ||||| ||| |||||||||||||||||||||| |||| ||||||||||| ||| |||| ||||| || ||||| ||| |||||||||||||||||||||| |||||| ||| ||| |||||||||| |||| ||| || |||||| |||| |||||||| ||| |||||| ||| |||||||| ||| |||| |||||| || ||||||||||| ||| |||||| || |||||||| || ||| ||| |||||||||| |||| || |||||| || ||||||| |||||||||| ||||||||||||| || |||||| ||| |||||||| |||| || ||||||| ||| || ||||||| || ||| |||||||||||||||||||||| |||| ||||||||||| ||| |||||||||||||||||||||| ||||||| |||||| ||||||||||||| |||||| |||| |||| ||| ||||| |||||||||| ||||||||||||| || |||||| ||| |||||||| ||| |||| |||||| ||||| || |||||| || ||| |||||||||||||||||||||| |||| ||||||||||| ||||| ||| |||| |||||| ||||| || ||||| || ||| |||||||||||||||||||||| ||||||| ||||| |||||||||||| ||| |||||| |||| |||||||| |||| ||| || |||| ||| |||| |||||| || ||||| ||| |||||||||||||||||||||| |||| ||||||||||| ||| |||| ||||| || |||||| ||| |||||||||||||||||||||| |||||| ||| ||| |||||||||| |||| ||| || |||||| |||| |||||||| ||| |||| ||| ||||| |||||| || ||||||||||| || |||||| || |||||||||||||||||||||| |||| ||||||||||| ||| |||||| || |||||||| || ||| ||| |||||||||| |||| || |||||| || ||||||| |||||||||| ||||||||||||| || |||| ||| || ||||||| ||| || ||||||| || ||| |||||||||||||||||||||| |||| ||||||||||| |||||||||||||||||||||| ||||||| |||||| ||||||||||||| |||||| |||| |||| ||| ||||| |||||||||| ||||||||||||| || |||| ||| |||| |||||| ||||| || |||||| || ||| |||||||||||||||||||||| |||| ||||||||||| ||||| ||| |||| |||||| ||||| || ||||| || ||| |||||||||||||||||||||| ||||||| ||||| ||||||||||| ||| |||||| |||| |||||||| |||| ||| || ||||||| ||| |||| |||||| || ||||| ||| |||||||||||||||||||||| |||| ||||||||||| ||| |||| ||||| || ||||| ||| |||||||||||||||||||||| |||||| ||| ||| |||||||||| |||| ||| || |||||| |||| |||||||| ||| ||||||| ||| ||||| |||||| || |||||||||| ||| |||||| || |||||||| || ||| ||| |||||||||| |||| || |||||| || ||||||| |||||||||| ||||||||||||| || ||||||| ||| || ||||||| ||| || ||||||| || ||| |||||||||||||||||||||| |||| ||||||||||| ||| |||||||||||||||||||||| ||||||| |||||| ||||||||||||| |||||| |||| |||| ||| ||||| |||||||||| ||||||||||||| || ||||||| ||| |||| |||||| ||||| || |||||| || ||| |||||||||||||||||||||| |||| ||||||||||| ||||| ||| |||| |||||| ||||| || ||||| || ||| |||||||||||||||||||||| ||||||| ||||| |||||||||||| ||| |||||| |||| |||||||| |||| ||| || |||||||| |||| ||| |||| ||||| || |||||| ||| |||||||||||||||||||||| |||| ||||||||||| ||| ||||| ||||| || |||||| ||| |||||||||||||||||||||| |||||| ||| ||| |||||||||| |||| ||| || |||||| |||| |||||||| ||| |||||||| |||| ||| ||||| |||||| || ||||||||||| ||| |||||| || |||||||| || ||| ||| |||||||||| |||| || |||||| || ||||||| |||||||||| ||||||||||||| || ||||||| ||| || ||||||| ||| || ||||||| || ||| |||||||||||||||||||||| |||| ||||||||||| ||| |||||||||||||||||||||| ||||||| |||||| ||||||||||||| |||||| |||| |||| ||| ||||| |||||||||| ||||||||||||| || ||||||| ||| |||| |||||| ||||| || |||||| || ||| |||||||||||||||||||||| |||| ||||||||||| ||||| ||| |||| ||||| || ||||| || ||| |||||||||||||||||||||| ||||||| ||||| ||||||||||| ||| |||||| |||| |||||||| |||| ||| || |||||||| ||| |||| |||||| || ||||| ||| |||||||||||||||||||||| |||| ||||||||||| ||| |||| |||||| || ||||| ||| |||||||||||||||||||||| |||||| ||| ||| |||||||||| |||| ||| || |||||| |||| |||||||| ||| |||||||| ||| ||||| |||||| || |||||||||| ||| |||||| || |||||||| || ||| ||| |||||||||| |||| || |||||| || ||||||| |||||||||| ||||||||||||| || |||||||| ||| || ||||||| ||| || ||||||| || ||| |||||||||||||||||||||| |||| ||||||||||| ||| |||||||||||||||||||||| ||||||| |||||| ||||||||||||| |||||| |||| |||| ||| ||||| |||||||||| ||||||||||||| || |||||||| ||| ||||| |||||| ||||| || |||||| || ||| |||||||||||||||||||||| |||| ||||||||||| ||||| ||| |||| |||||| ||||| || ||||| || ||| |||||||||||||||||||||| ||||||| ||||| |||||||||||||||||| ||| |||||| |||| |||||||| |||| ||| || ||| |||||||| ||||||| ||| ||||| |||||| || ||||| ||| |||||||||||||||||||||| |||| ||||||||||| ||| ||||| |||||| || |||||| ||| |||||||||||||||||||||| |||||| ||| ||| |||||||||| |||| ||| || |||||| |||| |||||||| ||| |||||||| ||||||| ||| |||| ||||| || ||||||||||| ||| ||| |||||| |||| |||||||| |||| ||| || ||| |||||||| ||| ||| ||||| |||||| || ||||| ||| |||||||||||||||||||||| |||| ||||||||||| ||| ||||| |||||| || |||||| ||| |||||||||||||||||||||| |||||| ||| ||| |||||||||| |||| ||| || |||||| |||| |||||||| ||| |||||||| ||| ||| |||| ||||| || ||||||||||||

Caregiver Burden

Caregiver burden was not assessed in the SUNLIGHT trial.

Table 17: Summary of Health-Related Quality of Life Results in the SUNLIGHT Trial, Data Cut-Off Date of July 5, 2022 — Full Analysis Set

CI = confidence interval; EORTC QLQ-C30 = European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire Core 30; EQ-5D-5L = 5-Level EQ-5D; EQ VAS = EQ visual analogue scale; IWRS = Interactive Web Response System; LSM = least squares mean; SD = standard deviation.

^aA mixed model of repeated measures that included terms for treatment, baseline stratification factors, baseline score, time to visit before any procedure (at each cycle, including the withdrawal visit), and treatment groups by time to visit interaction was used to compare change from baseline subscale scores longitudinally over time between treatment groups.

^bKaplan-Meier estimate.

^cMethodology of Brookmeyer and Crowley.

^dStratified log-rank test using IWRS stratification factors (geographic region, time since diagnosis of first metastasis, and RAS status).

Source: SUNLIGHT Clinical Study Report.²²

Harms

Harms data for the SUNLIGHT trial were based on the clinical data cut-off date of July 5, 2022, and are summarized in Table 18.

Adverse Events

The number of patients reporting any TEAEs was 241 of 246 (98.0%) patients for the trifluridine-tipiracil plus bevacizumab group and 241 of 246 (98.0%) patients for the trifluridine-tipiracil alone group. The most common TEAEs occurring in at least 20% of patients in the trifluridine-tipiracil plus bevacizumab group and the trifluridine-tipiracil alone group were neutropenia (62.2% versus 51.2%, respectively), nausea (37.0% versus 27.2%, respectively), anemia (28.9% versus 31.7%, respectively), asthenia (24.4% versus 22.4%, respectively), fatigue (21.5% versus 16.3%, respectively), diarrhea (20.7% versus 18.7%, respectively), and decreased appetite (20.3% versus 15.4%, respectively).

Serious Adverse Events

The proportion of patients who experienced at least 1 SAE was 24.8% in the trifluridine-tipiracil plus bevacizumab group and 31.3% in the trifluridine-tipiracil alone group. SAEs occurring in at least 2% of patients in the trifluridine-tipiracil plus bevacizumab group and the trifluridine-tipiracil alone group were intestinal obstruction (2.8% versus 2.0%, respectively), malignant neoplasm progression (2.4% versus 4.5%, respectively), COVID-19 (2.0% versus 2.4%, respectively), anemia (0.4% versus 3.3%, respectively), febrile neutropenia (0.4% versus 2.4%, respectively), jaundice (0.8% versus 2.0%, respectively), and hepatic failure (0 versus 2.0%, respectively).

The proportion of patients who experienced AEs of grade 3 or greater were 72.4% in the trifluridine-tipiracil plus bevacizumab group and 69.5% in the trifluridine-tipiracil alone group. The most common AEs of grade 3 or greater occurring in at least 5% of patients in the trifluridine-tipiracil plus bevacizumab group and the trifluridine-tipiracil alone group were neutropenia (43.1% versus 32.1%, respectively), anemia (6.1% versus 11.0%, respectively), decreased neutrophil count (8.9% versus 5.3%, respectively), and hypertension (5.7% versus 1.2%, respectively).

Withdrawals Due to Adverse Events

A total of 12.6% of patients experienced TEAEs that led to treatment withdrawal in each treatment group. Withdrawals due to AEs occurring in at least 1 patient in the trifluridine-tipiracil plus bevacizumab group and the trifluridine-tipiracil alone group were asthenia (3.3% versus 0.4%, respectively), jaundice (0.8% versus 0.8%, respectively), decreased appetite (0.8% versus 0.4%, respectively), fatigue (0.4% versus 0.8%, respectively), anemia (0.4% versus 0.8%, respectively), intestinal obstruction (0.4% versus 0.8%, respectively), malignant neoplasm progression (0.4% versus 0.8%, respectively), biliary dilation (0.8% versus 0, respectively), increased blood bilirubin (0.8% versus 0, respectively), pain (0.8% versus 0, respectively), and metastases to the central nervous system (0 versus 0.8%, respectively).

Mortality

|| ||| |||||||| ||||||| ||||||||| || ||| |||||||| ||| ||||| ||||||||| ||| ||||||| |||||||| || ||| |||||||||||||||||||||| |||| ||||||||||| ||||| ||| ||| ||||||| |||||||| || ||| |||||||||||||||||||||| ||||||| |||||||||| || || |||||| |||||| ||| ||||||||| |||||| |||||||| || || |||||| |||||||| ||| || |||||| || ||| |||||||||||||||||||||| |||| ||||||||||| ||| |||||||||||||||||||||| ||||||| |||||| ||||||||||||| |||||| ||||| |||||||| |||||| ||| ||||||||| |||||| |||||| ||| |||||| |||| |||||| ||| || ||||||||||| ||||||| || ||| |||||||||||||||||||||| |||| ||||||||||| ||| |||||||||||||||||||||| ||||||| |||||| |||||||||||||

Notable Harms

Notable harms in the SUNLIGHT trial were conducted post hoc using lists of predefined preferred terms with similar medical concepts to define the overall terms for trifluridine-tipiracil (bone marrow suppression, infections, and gastrointestinal symptoms) and bevacizumab (hypertension, proteinuria, bleeding or hemorrhage, embolic and thrombotic events, gastrointestinal perforation, and wound healing complication). Peripheral neuropathy was identified by patient group input to be an important and difficult-to-tolerate side effect.

Bone Marrow Suppression Events

The proportion of patients who experienced bone marrow suppression events was 80.9% in the trifluridine-tipiracil plus bevacizumab group and 73.2% in the trifluridine-tipiracil alone group, including neutropenia (62.2% versus 51.2%, respectively), anemia (28.9% versus 31.7%, respectively), thrombocytopenia (17.1% versus 11.4%, respectively), and leukopenia (6.5% versus 8.5%, respectively).

Infections

The proportion of patients who experienced at least 1 TEAE related to infections was 30.9% in the trifluridine-tipiracil plus bevacizumab group and 23.2% in the trifluridine-tipiracil alone group. Infections of grade 3 or higher were reported for 7.7% of patients and 7.3% of patients in the trifluridine-tipiracil plus bevacizumab group and the trifluridine-tipiracil alone group, respectively.

Gastrointestinal Events

The proportion of patients who experienced gastrointestinal events was 48.4% in the trifluridine-tipiracil plus bevacizumab group and 41.1% in the trifluridine-tipiracil alone group, including nausea (37.0% versus 27.2%, respectively), diarrhea (20.7% versus 18.7%, respectively), and vomiting (18.7% versus 14.6%, respectively). Gastrointestinal events of grade 3 or higher were reported for 2.0% of patients and 4.9% of patients in the trifluridine-tipiracil plus bevacizumab group and the trifluridine-tipiracil alone group, respectively, including nausea (1.6% versus 1.6%, respectively), diarrhea (0.8% versus 2.4%, respectively), and vomiting (0.8% versus 1.6%, respectively).

Hypertension

The proportion of patients who experienced hypertension was 10.2% in the trifluridine-tipiracil plus bevacizumab group and 2.0% in the trifluridine-tipiracil alone group. Hypertension events of grade 3 or higher were reported for 5.7% of patients and 1.2% of patients in the trifluridine-tipiracil plus bevacizumab group and the trifluridine-tipiracil alone group, respectively.

Proteinuria

The proportion of patients who experienced proteinuria was 6.1% in the trifluridine-tipiracil plus bevacizumab group and 1.2% in the trifluridine-tipiracil alone group. Proteinuria events of grade 3 or higher were reported for 0.8% of patients in the trifluridine-tipiracil plus bevacizumab group.

Hemorrhage

The proportion of patients who experienced at least 1 TEAE related to hemorrhage was 11.8% in the trifluridine-tipiracil plus bevacizumab group and 3.7% in the trifluridine-tipiracil alone group. Hemorrhage events of grade 3 or higher were reported for 1.2% of patients and 0.8% of patients in the trifluridine-tipiracil plus bevacizumab group and the trifluridine-tipiracil alone group, respectively.

Embolic and Thrombotic Events

The proportion of patients who experienced embolic and thrombotic events was 4.9% in the trifluridine-tipiracil plus bevacizumab group and 3.7% in the trifluridine-tipiracil alone group. Embolic and thrombotic events of grade 3 or higher were reported for 1.6% of patients and 3.3% of patients in the trifluridine-tipiracil plus bevacizumab group and the trifluridine-tipiracil alone group, respectively.

Gastrointestinal Perforation

Among the 6 (2.4%) patients in the trifluridine-tipiracil plus bevacizumab group who experienced gastrointestinal perforation, 4 (1.6%) patients experienced gastrointestinal perforation of grade 3 or higher.

Peripheral Neuropathy

Three (1.2%) patients and 1 (0.4%) patient experienced peripheral sensory neuropathy in the trifluridine-tipiracil plus bevacizumab group and the trifluridine-tipiracil alone group, respectively. One patient experienced peripheral neuropathy in the trifluridine-tipiracil plus bevacizumab group.

Table 18: Summary of Harms Results in the SUNLIGHT Trial, Data Cut-Off Date of July 5, 2022 — Safety Analysis Set

AE = adverse event; AESI = adverse event of special interest; CI = confidence interval; NR = not reported; SAE = serious adverse event; TEAE = treatment-emergent adverse event.

Note: Details included in Table 18 are from the sponsor’s Summary of Clinical Evidence.³³

^aPatients who stopped study treatment due to a TEAE corresponded to trifluridine-tipiracil withdrawal since bevacizumab monotherapy was not permitted.

^bThe follow-up period as defined for safety analysis occurred more than 30 days after the last study treatment intake.

^cPercentages are based on the number of deaths during the follow-up period (132 deaths vs. 149 deaths).

^dAnalyses of AESIs were conducted post hoc. Lists of predefined preferred terms of similar medical concept were used to define the overall AESI terms.

Source: SUNLIGHT Clinical Study Report.²²

Critical Appraisal

Internal Validity

The SUNLIGHT study was a phase III, randomized, open-label RCT. Patients were randomized 1:1 to trifluridine-tipiracil plus bevacizumab or trifluridine-tipiracil alone based on stratification factors (geographic region, time since first metastasis diagnosis, and RAS status) using a centralized IWRS. Baseline characteristics were generally balanced between treatment groups for key prognostic factors, disease characteristics (e.g., tumour biology, gene expression, number of and site of the primary tumour), and prior chemotherapy regimens. However, the open-label study design has the potential to impact patient-reported outcomes — including HRQoL, for which knowledge of the assigned treatment may bias reporting in favour of the intervention group (trifluridine-tipiracil plus bevacizumab).

Trifluridine-tipiracil alone was the comparator used in the SUNLIGHT trial. Trifluridine-tipiracil is approved and available in Canada but is not universally publicly funded; therefore, the majority of patients must obtain access via private drug coverage or out-of-pocket costs.

The duration of treatment for patients in the trifluridine-tipiracil plus bevacizumab group was a median of 5.0 months whereas the duration of treatment among patients in the trifluridine-tipiracil alone group was a median of 2.1 months. The mean (SD) cumulative dose was 3,907 mg/m² versus 2,247 mg/m² in the trifluridine-tipiracil plus bevacizumab group versus the trifluridine-tipiracil alone group, respectively. It is important to note that the differential treatment duration and dose may not be fully explained by the relatively small difference (11%) in treatment discontinuations between groups. Additionally, in the absence of data, it is unknown whether the open-label nature of the trial may have impacted patients’ adherence to treatment assignment. A significant proportion of patients had missing data (without imputation) for HRQoL by cycle 3 of treatment in the trifluridine-tipiracil plus bevacizumab group and the trifluridine-tipiracil alone group (68% versus 41%, respectively) and cycle 4 of treatment (60% versus 33%, respectively); it is unclear how the missing data may impact findings for HRQoL. Moreover, it is unknown whether there is an impact of higher rates of G-CSF use in the trifluridine-tipiracil plus bevacizumab group on HRQoL.

A significant number of censored patients and differences in censoring between groups was observed relatively early for both OS and PFS. For OS, the number of patients who were at risk in the trifluridine-tipiracil plus bevacizumab group (n = 246) versus the trifluridine-tipiracil alone group (n = 246) was 230 patients versus 205 patients at 3 months, respectively, 183 patients versus 143 patients at 6 months, respectively, and 88 patients versus 63 patients at 12 months, respectively. For PFS, the number of patients who were at risk in the trifluridine-tipiracil plus bevacizumab group versus the trifluridine-tipiracil alone group was 179 patients versus 109 patients at 3 months, respectively, and 99 patients versus 36 patients at 6 months, respectively. There were more patients who were censored due to being alive without radiological disease progression in the trifluridine-tipiracil plus bevacizumab group compared with the trifluridine-tipiracil alone group (15.5% versus 2.0%, respectively). A positive correlation has been associated between early censoring and the experience of the event of interest, leading to an overestimation of the survival probability in both study groups.⁵⁹ The SUNLIGHT trial did not provide data on the number of patients with missing follow-up data who were censored, so it is unknown whether this may have impacted survival findings.

OS was a primary end point in the SUNLIGHT study and statistical analyses employed a hierarchical testing strategy with the aim to demonstrate superiority of trifluridine-tipiracil plus bevacizumab over trifluridine-tipiracil alone. Analyses met the prespecified sample size and statistical significance in the primary estimand to subsequently assess the effect of treatment on PFS. OS and PFS were estimated using stratification factors employed at randomization, which is appropriate for minimizing bias due to potential effect modifiers. The stratified Cox proportional hazards model was checked for underlying assumptions; KM and log(-log) curves (crossing over early in the first month of treatment but with clear separation thereafter) suggested that the proportional hazards assumption likely holds for OS or PFS. Additionally, P values (0.0883 for OS and 0.568 for PFS) from a Schoenfeld residuals test did not appear to have been violated based on a prespecified P value of lower than 0.05 to suggest rejection of the null hypothesis, according to data submitted from the sponsor.²³

While appropriate statistical testing was conducted to adjust for multiplicity for the primary and key secondary end points, HRQoL was a secondary end point that was not adjusted for multiple testing and, therefore, is at risk of inflated type I error. MIDs were identified in the literature among patients with cancer and with mCRC for the cancer-specific EORTC QLQ-C30 tool, and among patients with cancer for the generic preference-based EQ-5D-5L tool.

External Validity

Approximately 24% of patients did not pass screening due to eligibility not being met. This is indicative of the challenge in selecting suitable patients for treatment, suggesting the possibility of a highly selected population who may derive the greatest benefit from treatment and who may not be fully representative of patients in clinical practice. However, the enrolled population in the SUNLIGHT trial was generally aligned with patients seen in clinical practice according to the clinical experts consulted by CADTH. Patients with unresectable adenocarcinoma of the colon and rectum were included; the clinical experts agreed that patients with metastatic disease and unresectable CRC were considered eligible for treatment with trifluridine-tipiracil, noting that patients with resectable disease undergo a different treatment pathway. Although patients with more than 2 prior chemotherapy regimens and those with prior treatment with trifluridine-tipiracil were excluded, the clinical experts consulted by CADTH considered these patients to be eligible for treatment with trifluridine-tipiracil plus bevacizumab. Patients who had an ECOG PS score of greater than 1 and those with small bowel or appendiceal adenocarcinoma were also considered to be eligible for treatment with trifluridine-tipiracil plus bevacizumab, according to the clinical experts consulted by CADTH. Among patients randomized in the SUNLIGHT study, the clinical experts consulted by CADTH noted a higher proportion of patients with RAS status expressing mutations (compared with wild-type) than would be expected in the population in Canada; however, the clinical experts acknowledged that subgroup analyses on RAS status did not appear to demonstrate an impact on treatment effects. Overall, the key prognostic indicators outlined by the clinical experts consulted by CADTH (i.e., age, number of metastatic sites, number of prior chemotherapy regimens, sidedness of the tumour, and ECOG PS) appeared to be reflective of patients in clinical practice. This included some with between-group imbalances of approximately 5%, which may have been due to chance rather than systematic differences based on appropriate randomization via IWRS. Since a small proportion of the patients enrolled were from North America (3.3%) with no patients enrolled from sites in Canada, the clinical experts consulted by CADTH considered the findings to be mainly applicable to the included population and limited in generalizability to patients with mCRC in Canada. Notably, the low number of non-white patients (e.g., Black, Asian) who comprised fewer than 2% of the trial population and the absence of Indigenous Peoples signify an important gap in the population and limited the applicability of findings.

The clinical experts consulted by CADTH noted that most patients had discontinued treatment in the SUNLIGHT study, with a greater proportion who stopped treatment due to clinical and/or progressive disease in the trifluridine-tipiracil alone group. This was unsurprising as many patients with advanced disease were likely to have exhausted 2 or more prior lines of therapies and the experts consulted did not consider the imbalance to be concerning given that a low rate of discontinuations due to AEs (6.5%) occurred at the same frequency in both groups and the proportion of patients who discontinued the study due to death was balanced between groups.

The intervention in the SUNLIGHT trial is for an unlabelled indication, as trifluridine-tipiracil alone was approved by Health Canada for adult patients with mCRC who have previously been treated with, or are not candidates for, available therapies, including fluoropyrimidine-, oxaliplatin-, and irinotecan-based chemotherapies, anti-VEGF biological agents, and, if RAS wild-type, anti-EGFR agents. The clinical experts consulted by CADTH considered trifluridine-tipiracil to be the most relevant comparator (among other treatment options, including BSC and regorafenib) for trifluridine-tipiracil plus bevacizumab, while acknowledging that this treatment is only available to a small patient population with access as it is not currently universally funded in Canada.

Approximately 29% of patients in the trifluridine-tipiracil plus bevacizumab group and 20% of patients in the trifluridine-tipiracil alone group received at least 1 concomitant G-CSF, for primary and secondary prophylaxis, therapeutic use (i.e., management of neutropenia) (16% and 12%, respectively), and for another indication (3.3% and 2.9%, respectively). The clinical experts consulted by CADTH commented that G-CSF prophylaxis was employed variably across clinical settings in Canada. According to the clinical experts, all patients treated with trifluridine-tipiracil (with or without bevacizumab) at their institution are given prophylactic G-CSF and that the proportions of patients in the SUNLIGHT trial who received G-CSF appeared to be reasonable.

Outcomes included in the SUNLIGHT study were identified as important to patients, including survival, HRQoL, and TEAEs. The clinical experts considered OS to be the key efficacy end point, which was examined as a primary outcome. The trial also included PFS as a key secondary end point, which the clinical experts consulted by CADTH identified as less desirable than OS as the presence or absence of measurable disease was not considered to be directly associated with patient survival. According to the clinical experts consulted by CADTH, PFS was appropriately included as supportive evidence for OS. Patients were followed for a median of 14.1 months in the trifluridine-tipiracil plus bevacizumab group and a median of 13.6 months in the trifluridine-tipiracil alone group. This duration appeared to be sufficient according to the clinical experts consulted by CADTH, who indicated that patients with advanced mCRC have limited efficacious treatment options and that OS at 6 months and 12 months is important for assessing the effects of treatment. EQ-5D-5L utility values derived according to a French value set (not calculated using a population health dataset for North America or Canada) was highlighted by the clinical experts consulted as being a limitation for the generalizability of these results to patients in Canada. Caregiver burden was identified as important by patient input and clinical experts consulted by CADTH; however, this was not assessed in the SUNLIGHT trial.

GRADE Summary of Findings and Certainty of the Evidence

Methods for Assessing the Certainty of the Evidence

For pivotal studies and RCTs identified in the sponsor’s systematic review, GRADE was used to assess the certainty of the evidence for outcomes considered most relevant to inform CADTH’s expert committee deliberations, and a final certainty rating was determined as outlined by the GRADE Working Group.^20,21

• High certainty: “We are very confident that the true effect lies close to that of the estimate of the effect.”

• Moderate certainty: “We are moderately confident in the effect estimate — The true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different. We use the word ‘likely’ for evidence of moderate certainty (e.g., ‘X intervention likely results in Y outcome’).”

• Low certainty: “Our confidence in the effect estimate is limited — The true effect may be substantially different from the estimate of the effect. We use the word ‘may’ for evidence of low certainty (e.g., ‘X intervention may result in Y outcome’).”

• Very low certainty: “We have very little confidence in the effect estimate — The true effect is likely to be substantially different from the estimate of effect. We describe evidence of very low certainty as ‘very uncertain’.”

Following the GRADE approach, evidence from RCTs started as high-certainty evidence and could be rated down for concerns related to study limitations (which refers to internal validity or risk of bias), inconsistency across studies, indirectness, the imprecision of effects, and publication bias.

When possible, certainty was rated in the context of the presence of an important (nontrivial) treatment effect; if this was not possible, certainty was rated in the context of the presence of any treatment effect (i.e., the clinical importance is unclear). In all cases, the target of the certainty of evidence assessment was based on the point estimate and where it was located relative to the threshold for a clinically important effect (when a threshold was available) or to the null. The target of the certainty of evidence assessment was the presence or absence of a clinically important effect for survival rates (OS, PFS), HRQoL (EORTC QLQ-C30, EQ-5D-5L), and notable harms (bone marrow suppression, infections, gastrointestinal symptoms, and hypertension) based on a threshold informed by the clinical experts consulted by CADTH for this review.

Results of GRADE Assessments

Table 2 presents the GRADE summary of findings for trifluridine-tipiracil plus bevacizumab versus trifluridine-tipiracil alone.

Long-Term Extension Studies

No long-term extension studies were submitted in the systematic review evidence.

Indirect Evidence

Content in this section has been informed by materials submitted by the sponsor. The following has been summarized and validated by the CADTH review team.

Objectives for the Summary of Indirect Evidence

As there was limited direct evidence comparing trifluridine-tipiracil plus bevacizumab versus other relevant comparators for the treatment of adult patients with mCRC who have been previously treated with, or are not considered candidates for, available therapies, including fluoropyrimidine-, oxaliplatin-, and irinotecan-based chemotherapies, anti-VEGF agents, and anti-EGFR agents, a review of indirect evidence was undertaken and submitted by the sponsor.²⁴

The objective of this section is to summarize and critically appraise the sponsor-submitted ITC, and to inform the pharmacoeconomic model.

Description of Sponsor-Submitted ITC

Objectives

The objective of this ITC was to compare the treatment efficacy and safety of trifluridine-tipiracil plus bevacizumab relative to currently existing therapies for the treatment of mCRC.

The ITC included a systematic review of the literature to identify trials investigating trifluridine-tipiracil plus bevacizumab or comparator interventions in patients with mCRC who have been previously treated with, or are not considered candidates for, currently available therapies; a feasibility assessment of conducting an NMA of trifluridine-tipiracil plus bevacizumab versus other relevant comparators in mCRC; and an NMA that compared trifluridine-tipiracil plus bevacizumab to other relevant treatments on both efficacy and safety outcomes. The other treatments examined in this ITC were trifluridine-tipiracil alone, chemotherapy, EGFR inhibitors or anti-VEGF drugs, BRAF inhibitors, immune checkpoint inhibitors, regorafenib, and BSC.

Study Selection Methods

Multiple databases were searched for relevant clinical trials in this ITC with a time frame of 2004 to the present. Studies were selected by 2 independent reviewers according to prespecified criteria for patient population, intervention, comparator, outcome measures, and study design. Any disagreements were resolved by discussion between reviewers, including a third researcher if needed. RCTs were included. Efficacy outcomes, safety outcomes, and HRQoL outcomes relevant to the treatment of mCRC were evaluated in this ITC. Data were extracted by a single reviewer and independently validated by a second reviewer. Any discrepancies were resolved by discussion between reviewers, including a third researcher if needed. All included studies in the ITC were evaluated according to a Cochrane risk-of-bias tool for RCTs.

Details of patient population, intervention, comparator, outcome measures, and study design for study selection in the systematic literature review of the ITC are provided in Table 19.

Table 19: Study Selection Criteria and Methods for ITC Submitted by the Sponsor

1L = first-line; 2L = second-line; 3L = third-line; 4L = fourth-line; AE = adverse event; anti-EGFR = anti–epidermal growth factor receptor; anti-VEGF = anti–vascular endothelial growth factor; ASCO = American Society of Clinical Oncology; ASCO GI = American Society of Clinical Oncology Gastrointestinal; BSC = best supportive care; CDSR = Cochrane Database of Systematic Reviews; CENTRAL = Cochrane Central Register of Controlled Trials; DCR = disease control rate; EGFR = epidermal growth factor receptor; ESMO = European Society for Medical Oncology; EU CTR = European Union Clinical Trials Register; HRQoL = health-related quality of life; HTA = Health Technology Assessment; ICTRP = International Clinical Trials Registry Platform; IQWiG = Institute for Quality and Efficiency in Health care; mCRC = metastatic colorectal cancer; ORR = objective response rate; OS = overall survival; PFS = progression-free survival; RCT = randomized controlled trial; wt = wild-type.

Note: Details included in Table 19 are from the sponsor’s Summary of Clinical Evidence.³³

^aIn studies where outcomes are not stratified by line of treatment (2L vs. 3L vs. 4L+), a population threshold will be applied to determine if the majority of patients represent the 3L or beyond indication. A threshold of 80% is considered acceptable by other HTA bodies, such as IQWiG.⁶¹

Source: Sponsor-submitted indirect treatment comparison.²⁴

ITC Analysis Methods

After conducting the feasibility assessment, 4 outcomes were deemed feasible for the NMA, from both a methodological and clinical standpoint: OS, PFS, disease control rate, and treatment-related AEs. Note that for the outcome of any AEs, 4 studies did not report the incidence of patients with at least 1 AE. In 1 study, the BSC arm was an outlier, with only 51.9% of patients experiencing AEs, versus more than 85% of patients for BSC arms from 2 other studies. In addition, multiple studies reported a rate of 100% AEs, leading to issues of modelling underlying any ITC and the associated uncertainty of estimates. Therefore, an NMA on any AEs was not performed. Authors of this ITC stated that treatment-related AEs instead of any AEs was considered to be more helpful in modelling and allowed for a better estimation of the associated uncertainty in the safety profile.

All NMA models were conducted using a Bayesian framework. All univariate analyses involved a 10,000 run-in iteration phase and a 10,000-iteration phase for parameter estimation. All calculations were performed via the program Just Another Gibbs Sampler 3.2.3 using 3 chains initiated under 3 main paradigms: “negative,” meaning that treatment is better than BSC; “equal,” meaning that treatment is equivalent to BSC; and “positive,” meaning that treatment is worse than BSC.⁶² Fixed-effects and random-effects models were conducted for each outcome of interest and per analysis type (base-case or sensitivity analysis). The random-effects NMAs used a vaguely informative normal prior with a uniform SD, the value of which was equivalent to 1.0 for all binary outcomes, and U(0,2X) for continuous outcomes, where X was the median SD of the outcome within the study. The choice of prior was somewhat arbitrary by nature, and in these cases was chosen to allow for a moderate to large amount of random-effects variation and to discourage the estimation of extreme values of random-effects variation. Sensitivity to priors was investigated in sensitivity analyses, generally by shortening the internal by 50%. Unconstrained parameters, such as treatment effects, had a prior of N (0, 10,000), though this was decreased to N (0, 1,000) in cases where convergence was imperfect. For each model, model fit was assessed based on deviance, the effective number of parameters, deviance information criterion, and an estimate of random-effects dispersion for random-effects models only. The best model was chosen based on the values of deviance information criterion.

The assessment of consistency was not performed due to the absence of a closed loop in the available networks. Convergence was confirmed through inspection of the ratios of Monte Carlo error to the SDs of the posteriors.

Due to the differences in study design across all included studies, such as population (mCRC or refractory mCRC), blinding (open-label or double-blind), specific ethnic groups (multinational, Asian countries, or European countries), and the definition of BSC, sensitivity analyses were conducted to account for these heterogeneities. These analyses included only multinational trials for each outcome of interest.

Detailed statistical methods of ITC for OS, PFS, and treatment-related AEs are provided in Table 20.

Results of Sponsor-Submitted ITC

Summary of Included Studies

Among the 15 RCTs identified from the systematic review, 5 RCTs were excluded (the reasons for exclusion were not provided) and 10 RCTs were included in the NMA. All 10 studies were RCTs in patients with mCRC or refractory mCRC. The included RCTs evaluated the efficacy and safety of the following therapies, which are relevant to this review: trifluridine-tipiracil plus bevacizumab in 2 studies, BSC alone in 7 studies, regorafenib in 2 studies, and trifluridine-tipiracil alone in 6 studies. Other active treatments included panitumumab, cetuximab, panitumumab with BSC, cetuximab with BSC, and trifluridine-tipiracil plus panitumumab.

Table 20: ITC Analysis Methods

AE = adverse event; BSC = best supportive care; CET = cetuximab; DCR = disease control rate; DIC = deviance information criterion; HR = hazard ratio; IRC = independent review committee; ITC = indirect treatment comparison; NICE TSD 2 = National Institute for Health and Care Excellence Technical Support Document 2; NMA = network meta-analysis; ORR = objective response rate; OS = overall survival; PAN = panitumumab; Pd = effective number of parameters; PFS = progression-free survival; RCT = randomized controlled trial; RE = random effect; RECIST 1.1 = Response Evaluation Criteria in Solid Tumours Version 1; REGO = regorafenib; SD = standard deviation; TAE = treatment-related adverse event; vs. = versus.

Note: Details included in Table 20 are from the sponsor’s Summary of Clinical Evidence.³³

Source: Sponsor-submitted indirect treatment comparison.²⁴

Across the included studies, patients’ baseline characteristics were similar in age (mean age = 61.6 years [range = 56 years to 67 years]) and the proportion of colon as the primary site of mCRC. The dosages, frequency of administration, and dosage cycles of commonly investigated drugs (trifluridine-tipiracil plus bevacizumab, trifluridine-tipiracil alone, and regorafenib) were consistent across the included studies and reflective of the clinical practice. Definitions of OS, PFS, and treatment-related AEs were similar across the included studies.

Significant heterogeneities were observed across the included studies for ECOG PS (ECOG PS of 0 ranged from 22.2% to 64.3%) and the incidence of RAS wild-type tumour (range = 26.5% to 69.5%).

Patients in 1 study (the SUNLIGHT trial) had received no more than 2 prior lines of therapy while patients in 4 studies had received at least 2 prior lines of therapy and patients in another 5 studies had received at least 3 prior lines of therapy for mCRC.

In the majority of the included studies, BSC was defined as clinically indicated interventions that were required to provide palliation of symptoms and improve quality of life as needed, such as hematologic support and the management of gastrointestinal symptoms, excluding antineoplastic drugs. There were still differences in the definition of BSC across all included studies, though. For example, BSC was defined as any interventions excluding other investigational antitumour drugs or antineoplastic chemotherapy, hormonal therapy, or immunotherapy in 3 trials. The quality of the included studies was evaluated using the Cochrane risk-of-bias tool for randomized trials as described in the Cochrane Handbook for Systemic Reviews of Interventions. In general, most studies were considered to have a low risk of bias for “deviations from intended interventions,” “missing outcome data,” “measurement of the outcome,” “selection of the reported result,” and “overall bias;” however, 6 of 10 RCTs were considered to have some concerns related to “randomization process.”

A description of important differences across trials for key characteristics is provided in Table 21.

The current review presents the results of OS, PFS, and treatment-related AEs, which were identified as the most relevant clinical outcomes for patients with mCRC by the patients and the clinical experts consulted by CADTH.

Table 21: Assessment of Homogeneity for Sponsor-Submitted ITC

Anti-EGFR = anti–epidermal growth factor receptor; BSC = best supportive care; CTCAE v5.0 = Common Terminology Criteria for Adverse Events Version 5.0; ECOG PS = Eastern Cooperative Oncology Group Performance Status; ITC = indirect treatment comparison; NMA = network meta-analysis; OS = overall survival; PFS = progression-free survival; RCT = randomized controlled trial; RECIST 1.1 = Response Evaluation Criteria in Solid Tumours Version 1.1; TAE = treatment-related adverse event; vs. = versus.

Note: Details included in Table 21 are from the sponsor’s Summary of Clinical Evidence.³³

Source: Sponsor-submitted indirect treatment comparison.²⁴

Figure 9 to Figure 11 present the networks of evidence for base-case analysis of OS, PFS, and treatment-related AEs in the included studies.

Figure 9: Network Diagram for OS

BEV = bevacizumab; BSC = best supportive care; CET = cetuximab; FTD/TPI = trifluridine-tipiracil; OS = overall survival; PAN = panitumumab; REGO = regorafenib.

Source: Sponsor-submitted indirect treatment comparison.²⁴

Figure 10: Network Diagram for PFS

BEV = bevacizumab; BSC = best supportive care; CET = cetuximab; FTD/TPI = trifluridine-tipiracil; PAN = panitumumab; PFS = progression-free survival; REGO = regorafenib.

Source: Sponsor-submitted indirect treatment comparison.²⁴

Figure 11: Network Diagram for Treatment-Related AEs

AE = adverse event; BEV = bevacizumab; BSC = best supportive care; FTD/TPI = trifluridine-tipiracil; REGO = regorafenib.

Source: Sponsor-submitted indirect treatment comparison.²⁴

Results

Detailed efficacy and safety results are presented in Table 22. Only the results for trifluridine-tipiracil plus bevacizumab versus relevant comparators for this review (trifluridine-tipiracil alone, regorafenib, and BSC) are summarized in this section.

Overall Survival

In total, 9 studies were included in the network for OS. All studies presented mature OS data. All studies but 1 reported OS HR or a KM curves plot. Trifluridine-tipiracil plus bevacizumab was compared with 4 other active treatments: BSC, panitumumab, cetuximab, regorafenib, and trifluridine-tipiracil alone. The fixed-effects model was found to be the best fitting model for the analysis of OS.

Overall, results from fixed-effects models showed that the HR for OS was 0.42 (95% CrI, 0.33 to 0.55), 0.60 (95% CrI, 0.44 to 0.81), and 0.60 (95% CrI, 0.49 to 0.73) for the comparisons of trifluridine-tipiracil plus bevacizumab versus BSC, regorafenib, and trifluridine-tipiracil alone, respectively, in favour of trifluridine-tipiracil plus bevacizumab. A larger treatment effect was observed for trifluridine-tipiracil plus bevacizumab versus BSC, compared to the other 2 comparators.

A sensitivity analysis for OS was carried out that included 4 studies to investigate the findings in a network on only multinational trials. Results from fixed-effects models were consistent with those in the primary analysis: HR was 0.42 (95% CrI, 0.32 to 0.55) versus BSC, 0.55 (95% CrI, 0.39 to 0.77) versus regorafenib, and 0.61 (95% CrI, 0.49 to 0.77) versus trifluridine-tipiracil alone.

Progression-Free Survival

PFS was defined consistently across all included studies. Ten studies were included in the network for PFS. All studies presented mature PFS data. Trifluridine-tipiracil plus bevacizumab was compared with trifluridine-tipiracil alone, BSC, cetuximab, regorafenib, panitumumab, and trifluridine-tipiracil plus panitumumab. The fixed-effects model was found to be the best fitting model for the analysis of PFS.

Results from fixed-effects models showed that the HR for PFS was 0.20 (95% CrI, 0.16 to 0.25), 0.44 (95% CrI, 0.34 to 0.58), and 0.44 (95% CrI, 0.36 to 0.53) for the comparisons of trifluridine-tipiracil plus bevacizumab versus BSC, regorafenib, and trifluridine-tipiracil alone, respectively, in favour of trifluridine-tipiracil plus bevacizumab. A larger treatment effect was observed for trifluridine-tipiracil plus bevacizumab versus BSC compared to the other 2 comparators.

A sensitivity analysis for PFS was carried out that included 4 studies to investigate the findings in a network on only multinational trials. Results from fixed-effects models were consistent with those in the primary analysis: HR was 0.21 (95% CrI, 0.16 to 0.27) versus BSC, 0.43 (95% CrI, 0.32 to 0.58) versus regorafenib, and 0.44 (95% CrI, 0.36 to 0.54) versus trifluridine-tipiracil alone.

Treatment-related AEs

Seven studies were included in the network for the incidence of patients with at least 1 treatment-related AE. Trifluridine-tipiracil plus bevacizumab was compared to trifluridine-tipiracil alone, BSC, and regorafenib. Treatment-related AEs were defined consistently across all included studies using the National Cancer Institute’s Common Terminology Criteria for Adverse Events, Version 5.0. Based on a deviance information criterion value of 58.3 in the random-effects model and 78.3 in the fixed-effects model, the random-effects model was determined to be the most suitable model for the analysis of treatment-related AEs. However, considering that the prior given to the between-study heterogeneity significantly impacted its posterior distribution and the range of treatment effects’ posterior distribution was heavily influenced by the between-study heterogeneity, which spanned from 0.01 to 100, the fixed-effects model was chosen to facilitate evidence-driven decision-making.

Results from fixed-effects models showed that the odds ratio (OR) was 11.0 (95% CrI, 5.8 to 21.3), 1.1 (95% CrI, 0.5 to 2.3), and 2.22 (95%CrI, 1.3 to 3.9) for the comparisons of trifluridine-tipiracil plus bevacizumab versus BSC, regorafenib, and trifluridine-tipiracil alone, respectively, in favour of the comparators. A larger effect was observed for trifluridine-tipiracil plus bevacizumab versus BSC compared to the other 2 comparators.

A sensitivity analysis for the incidence of treatment-related AEs was carried out that included 3 studies to investigate the findings in a network on only multinational trials. Results from fixed-effects models were consistent with those in the primary analysis: OR of 1.45 (95% CrI, 0.70 to 3.12) versus regorafenib, OR of 2.22 (95% CrI, 1.32 to 3.81) versus trifluridine-tipiracil alone, and OR of 11.15 (95% CrI, 5.95 to 20.98) versus BSC.

Table 22: Summary of Efficacy and Safety Outcome Measures in the Sponsor-Submitted ITC

AE = adverse event; BEVA = bevacizumab; BSC = best supportive care; CrI = credible interval; HR = hazard ratio; ITC = indirect treatment comparison; OR = odds ratio; OS = overall survival; PFS = progression-free survival; vs. = versus.

Source: Sponsor-submitted indirect treatment comparison.²⁴

Critical Appraisal of Sponsor-Submitted ITC

In the ITC, studies were identified by searching multiple databases based on prespecified inclusion and exclusion criteria. The error and bias in the study selection process are minimized. The reviewers of this ITC used appropriate methods for study selection and data extraction. The quality of the included studies was assessed using a validated tool. However, there was no discussion on how any potential biases in the included trials could have an impact on the validity of the ITC and there were no sensitivity analyses being conducted to assess the impact of trials with poor quality. In this ITC, results were derived from random-effects or fixed-effects models. Reasons for model selection were justified by the authors and based on statistical and clinical considerations.

A significant concern with the ITC presented is that the trials included in the analyses were highly heterogeneous in terms of both study design and patient characteristics. Among the 10 RCTs, 5 were open-label trials and 5 were blinded. However, OS is an objective measurement of the treatment effect. Having or not having the knowledge of treatment allocation (blinding or not) is less likely to have a large impact on OS among patients with cancer.⁶³ The SUNLIGHT trial enrolled patients who had received no more than 2 previous chemotherapy regimens while the other trials enrolled patients who had received 2 or more lines of therapy for mCRC. Significant heterogeneities in patient characteristics included ECOG PS (the proportion of patients with an ECOG PS score of 0 ranged from 22% to 64%) and RAS status (the proportion of patients with RAS-positive status ranged from 27% to 70%) at baseline. Despite various statistical models being employed to lessen the impact of potential clinical heterogeneity on the estimated treatment effect of trifluridine-tipiracil plus bevacizumab, there is still significant uncertainty in the ITC results. In addition, given the lack of closed loops in any of the networks, consistency in the ITC analyses could not be tested. All comparisons were therefore informed only by indirect evidence, which increases the level of uncertainty.

The efficacy outcomes and safety outcomes were defined consistently across the included trials. BSC was defined somewhat differently. According to the clinical experts consulted by CADTH, BSC generally includes any interventions other than chemotherapy or immunotherapy — for example, medications for gastrointestinal symptoms and hematologic support, blood transfusion, or radiation for palliative care. The experts noted that it is less likely that the varied definitions for BSC would have an impact on study result interpretation.

Some important patient characteristics in the included trials were not reported in this ITC, such as treatment duration, the timing of study end point evaluation, the use of subsequent therapies after disease progression, and the length of follow-up. Therefore, adjustment for their potential treatment-effect modification was not feasible, and it is likely that the transitivity assumption was not met. Furthermore, it is unclear whether the results can provide insight into the long-term effect of the study drug for patients with mCRC due to a lack of data regarding the length of trial follow-up.

In the NMA, safety data were sparse and only the analysis on treatment-related AEs was deemed feasible by the sponsor. Even though the authors of this ITC indicated that using treatment-related AEs instead of any AEs could help modelling and allow for better estimation of the uncertainty across associated rates of treatment-related AEs, it would be desirable to understand the risk of any adverse effect in patients with mCRC during the study, no matter if they were considered drug-related or not.

In this ITC, several efficacy and safety outcomes were analyzed, such as OS, PFS, and treatment-related AEs. However, other efficacy end points of interest to patients and clinicians, such as HRQoL, were not investigated. Therefore, the relative treatment effect of trifluridine-tipiracil plus bevacizumab versus other active treatments on patients’ HRQoL remains unknown.

Summary

Based on the results of the sponsor-submitted ITC, treatment of trifluridine-tipiracil plus bevacizumab may be associated with prolonged OS and PFS but also an increased risk of treatment-related AEs in patients with mCRC compared to other treatments such as BSC, regorafenib, or trifluridine-tipiracil alone. However, the magnitude of these differences was uncertain due to substantial uncertainty related to a number of limitations that impact the internal and external validity of this study, such as significant clinical heterogeneities between the included trials, potential intransitivity, reliance solely on indirect data, and a lack of other important outcomes that are important to patients and clinicians (e.g., HRQoL).

Studies Addressing Gaps in the Systematic Review Evidence

Content in this section has been informed by materials submitted by the sponsor. The following has been summarized and validated by the CADTH review team.

No additional studies addressing important gaps in the systematic review evidence were submitted.

Discussion

Summary of Available Evidence

One randomized, phase III, open-label, multicentre study evaluated the efficacy and safety of trifluridine-tipiracil plus bevacizumab versus trifluridine-tipiracil alone. The SUNLIGHT study enrolled 492 adults with advanced mCRC who had received up to 2 previous chemotherapy regimens and demonstrated progressive disease or intolerance to their last regimen, and randomized patients to each group with stratification by geographic region (North America, European Union, the rest of the world), time since first metastasis diagnosis (< 18 months, ≥ 18 months), and RAS status (wild-type, mutant). The primary objective of the SUNLIGHT study was to demonstrate superiority of OS and the key secondary objective was to estimate investigator-assessed PFS. Additional secondary end points included HRQoL assessed with the EORTC QLQ-C30 and EQ-5D-5L, and TEAEs.

In the SUNLIGHT trial, patients had a mean age of 61.7 years (SD = 11.1 years), and most patients were male (52%), white (95.2%), and enrolled from the European Union (64.0%). Most patients had a primary diagnosis of colon cancer (73%) and stage IV disease (66%). The time from the diagnosis of the first metastasis until randomization was 18 months or longer in 57.5% of the patients, and 30.7% of patients had RAS wild-type disease. Most patients (92.1%) had received 2 previous treatment regimens for metastatic disease, 2.6% of patients had more than 2 prior regimens, and 5.3% of patients had received 1 previous treatment regimen. All patients had received previous fluoropyrimidine-based therapy, 72.0% of patients had received previous anti-VEGF therapy (47.8% had received bevacizumab as part of their first regimen, 43.9% as part of their second regimen, and 20.3% as part of both their first and second regimens), and 93.7% of patients with RAS wild-type disease had received previous anti-EGFR therapy. Demographic characteristics were generally similar between the trifluridine-tipiracil plus bevacizumab group and the trifluridine-tipiracil alone group, with notable (> 5%) between-group differences for patients aged 65 years and older (41% versus 48%, respectively), for patients whose primary tumour was located on the right side (25% versus 31%, respectively), and for patients whose primary tumour was located on the left side (75% versus 69%, respectively).

One ITC was submitted by the sponsor to compare the treatment efficacy and safety of trifluridine-tipiracil plus bevacizumab with active therapies for the treatment of mCRC. Studies included relevant therapies for the drug under review, including BSC, regorafenib, and trifluridine-tipiracil alone. Outcomes included OS, PFS, and treatment-related AEs.

Interpretation of Results

Efficacy

Patients with mCRC highlighted their need for new therapy to prolong survival while maintaining or improving quality of life, with tolerable side effects. Clinicians also identified a significant need for treatment to improve OS, decrease symptom burden, and improve quality of life among patients with mCRC who experience disease progression following 2 lines of anticancer therapy. The pivotal SUNLIGHT trial assessed OS as the primary end point and PFS as a key secondary end point, which aligned with stakeholder inputs and guidelines for clinical trials. While the clinical experts consulted by CADTH considered OS to be the key efficacy outcome among patients with mCRC, they acknowledged that PFS may provide added value as being predictive of OS. HRQoL was also assessed in the SUNLIGHT trial, which aligned with stakeholder inputs, especially for patients who emphasized that they would undertake new treatments for modest gains in OS if improvements in HRQoL could be achieved to enable them to engage in normal activities of daily living. However, HRQoL outcomes related to caregiver burden identified as important by patient and clinician inputs were not captured in the SUNLIGHT trial.

Findings for OS demonstrated the superiority of trifluridine-tipiracil plus bevacizumab when compared with trifluridine-tipiracil alone. Visual inspection of the KM curves and log(-log) plots and Schoenfeld residuals testing for OS were examined to evaluate the validity of the proportional hazards assumption; the P value supported the underlying assumption whereas the curves suggested a violation of the assumption. The clinical experts consulted by CADTH agreed that a 39% reduction in the risk of death with a median OS gain of greater than 3 months was clinically meaningful in demonstrating benefit for treatment with trifluridine-tipiracil plus bevacizumab. Findings for OS at 6 months and 12 months were consistent with the primary OS results and clinically meaningful according to the clinical experts consulted by CADTH. While the effect estimates for each time point showed a direction of benefit, the lower bound of the CIs did not reach the threshold identified for clinical significance, which reduced the certainty of the magnitude of benefit.

Findings for PFS were consistent with OS for demonstrating the benefit of trifluridine-tipiracil plus bevacizumab when compared with trifluridine-tipiracil alone, and were also supported by Schoenfeld residuals testing (P value of 0.568) and the KM curves and log(-log) plots. Nevertheless, the clinical experts consulted by CADTH considered that findings for PFS at 3 months and 6 months demonstrated a clinically meaningful difference in favour of treatment with trifluridine-tipiracil plus bevacizumab. As with OS, there was reduced certainty in the magnitude of benefit due to wide CIs.

Subgroup analyses for OS and PFS were conducted for 15 subgroups; these analyses included factors important for patient prognosis of advanced mCRC as identified by the clinical experts consulted by CADTH (i.e., age, number of metastatic sites, number of prior regimens, sidedness of tumour, and ECOG PS). Acknowledging that subgroup analyses were unstratified and unadjusted for multiplicity with a relatively small number of patients and events for each subgroup category, findings are considered exploratory, especially in the absence of known treatment-effect modifiers for trifluridine-tipiracil plus bevacizumab according to the clinical experts consulted by CADTH.

Patients reported having experienced side effects of treatments for mCRC (e.g., fatigue, nausea, diarrhea, neuropathy) that significantly impacted their quality of life. In the SUNLIGHT study, HRQoL was assessed using a disease-specific tool (EORTC QLQ-C30) and a generic measure (EQ-5D-5L). Analyses of the EORTC QLQ-C30 subscales included LSM change from baseline from a mixed model of repeated measures with terms including treatment and baseline stratification factors longitudinally over time. Analyses also included time until definitive deterioration of at least 10 points or greater on the subscales. The sponsor identified a 10-point difference between groups to be important based on literature for patients with cancer.^52,54 Based on literature-identified anchor-based MIDs among patients with CRC for within-group differences (7 to 10 for improvement and –8 to –5 for deterioration) and between-group differences (6 for improvement and –9 to –7 for deterioration),⁵⁷ the clinical experts consulted by CADTH used a between-group treatment difference of 10 points (for the LSM change from baseline) and 10% (for the proportion of patients with a 10-point or greater deterioration) on the global health status subscale. Employing a relatively conservative threshold for a clinically meaningful difference between groups, effect estimates were affected by wide CIs. For patients, it may be worthwhile to note that there was a prolonged time to deterioration (of ≥ 10 points) on subscales of functioning (physical, emotional) and symptoms (fatigue, nausea and vomiting, pain, diarrhea, dyspnea, and appetite loss) of approximately 4 months when treated with trifluridine-tipiracil plus bevacizumab compared with trifluridine-tipiracil alone. Findings for EQ VAS appeared to show improved quality of life among patients treated with trifluridine-tipiracil plus bevacizumab, although the magnitude of difference when compared with trifluridine-tipiracil alone may not have been clinically meaningful according to thresholds used for patients with various cancers. According to the clinical experts consulted by CADTH, utility metrics derived and standardized for the population in Canada were an identified gap in the SUNLIGHT trial as EQ-5D-5L utility values were calculated using a French population. All HRQoL outcomes were affected by concerns related to risk of bias due to the open-label design and knowledge of assigned treatment, and due to missing data over the treatment duration that exceeded 20% at cycle 3 and 40% at cycle 4, with significant losses by cycle 10.

Overall, the patients enrolled in the SUNLIGHT study were representative of patients in clinical practice, according to the clinical experts consulted by CADTH; however, patients who were excluded (having more than 2 lines of prior therapy, an ECOG PS score of greater than 1, and/or small bowel or appendiceal adenocarcinoma) would be considered eligible for treatment with trifluridine-tipiracil plus bevacizumab. Additionally, the clinical experts consulted by CADTH outlined an unmet need for novel and effective therapies among patients with CRC diagnosed in a younger demographic or with de novo metastatic disease who have more aggressive presentations, which were not subpopulations or subgroups examined in the SUNLIGHT trial.

Input from clinician groups identified that BSC is a relevant and important comparator for trifluridine-tipiracil plus bevacizumab given that trifluridine-tipiracil alone is not universally funded or available in Canada. The clinical experts consulted by CADTH also identified BSC among other treatments but indicated trifluridine-tipiracil to be the most appropriate comparator for assessing the efficacy of trifluridine-tipiracil plus bevacizumab among patients with mCRC. While treatment of trifluridine-tipiracil plus bevacizumab may be associated with prolonged OS and PFS compared to other treatments in the sponsor-submitted ITC, there was substantial uncertainty in the magnitude of the differences due to significant heterogeneities between the included trials and an absence of evidence for HRQoL.

Harms

Overall, the safety profile of trifluridine-tipiracil plus bevacizumab showed similar proportions in patients with TEAEs, SAEs, treatment discontinuations due to AEs, and deaths. Input from a patient with experience using trifluridine-tipiracil plus bevacizumab reported side effects of neutropenia, loss of appetite, diarrhea, and nausea. The product monograph for trifluridine-tipiracil alone included serious warnings and precautions for myelosuppression (including neutropenia, leukopenia, anemia, thrombocytopenia, and febrile neutropenia) and gastrointestinal toxicity (including nausea, vomiting, and diarrhea), which were included as AEs of special interest in the SUNLIGHT trial. Patients in the SUNLIGHT trial commonly experienced neutropenia, nausea, anemia, and diarrhea. A higher proportion of patients in the trifluridine-tipiracil plus bevacizumab group compared with the trifluridine-tipiracil alone group experienced myelosuppression (80.9% versus 73.2%, respectively), including neutropenia (62.2% versus 51.2%, respectively) and thrombocytopenia (17.1% versus 11.4%, respectively). More patients in the trifluridine-tipiracil plus bevacizumab group compared with the trifluridine-tipiracil alone group experienced gastrointestinal symptoms (48.4% versus 41.4%, respectively), including nausea (37.0% versus 27.2%, respectively) and decreased appetite (20.3% versus 15.4%, respectively). Gastrointestinal symptoms of grade 3 or higher were reported for 2.0% of patients and 4.9% of patients in the trifluridine-tipiracil plus bevacizumab group and the trifluridine-tipiracil alone group, respectively, including nausea (1.6% versus 1.6%, respectively), diarrhea (0.8% versus 2.4%, respectively), and vomiting (0.8% versus 1.6%, respectively). In the trifluridine-tipiracil plus bevacizumab group compared with the trifluridine-tipiracil alone group, a higher proportion of patients also experienced infections (30.9% versus 23.2%, respectively) and hypertension (10.2% versus 2.0%, respectively). Fewer patients died in the trifluridine-tipiracil plus bevacizumab group compared with patients treated with trifluridine-tipiracil alone (5.3% versus 9.8%, respectively). Evidence from the ITC submitted by the sponsor demonstrated an increased risk of treatment-related AEs with trifluridine-tipiracil plus bevacizumab. This finding was limited by the lack of specification of AEs as well as by the distinguishing of treatment-related AEs from treatment-emergent AEs (without attribution) and the sparse safety data in the NMA.

Overall, the clinical experts consulted by CADTH expressed that the AEs reported in the SUNLIGHT trial were supported by their observations in clinical practice and noted that patients with advanced mCRC are managed with respect to toxicities experienced during treatment, including the use of G-CSF if available. The clinical experts consulted by CADTH indicated that treatment with trifluridine-tipiracil plus bevacizumab should be well informed by patients’ symptoms, side effects, and quality of life as well as treatment effectiveness.

Conclusion

The SUNLIGHT study was a randomized, phase III, open-label trial in adults with mCRC who have been previously treated with, or are not candidates for, available therapies, including fluoropyrimidine-, oxaliplatin-, and irinotecan-based chemotherapies, anti-VEGF biological agents, and, if RAS wild-type, anti-EGFR agents. Evidence from the SUNLIGHT study demonstrated that trifluridine-tipiracil plus bevacizumab likely results in a clinically important increase in the proportion of patients with OS at 6 months and 12 months (moderate certainty). The evidence demonstrated that trifluridine-tipiracil plus bevacizumab results in a clinically important increase in the proportion of patients with PFS at 3 months (high certainty) and likely results in a clinically important increase in the proportion of patients with PFS at 6 months (moderate certainty) when compared with trifluridine-tipiracil alone. Trifluridine-tipiracil plus bevacizumab may result in little to no clinically important difference in the change from baseline in EQ VAS (low certainty) when compared with trifluridine-tipiracil alone. However, the evidence is very uncertain about the effect of trifluridine-tipiracil plus bevacizumab in the change from baseline in EORTC QLQ-C30 global health status score, the proportion of patients with at least a 10-point deterioration from baseline in the global health status score, and the change from baseline in EQ-5D-5L health utilities (very low certainty) when compared with trifluridine-tipiracil alone. Trifluridine-tipiracil plus bevacizumab may result in little to no clinically important difference in bone marrow suppression, infections, gastrointestinal symptoms, and hypertension (low certainty) when compared with trifluridine-tipiracil alone. Patients treated with trifluridine-tipiracil plus bevacizumab experienced similar frequencies of TEAEs, SAEs, treatment discontinuations due to AEs, and deaths, and no new safety signals were identified. Confidence in the effect estimates in the SUNLIGHT trial was limited due to imprecision from wide CIs for survival and harms outcomes, the potential of the open-label design to bias patient-reported outcomes, and significant missing data for HRQoL.

There is a lack of head-to-head direct evidence between trifluridine-tipiracil plus bevacizumab and other treatments for advanced mCRC. A sponsor-submitted ITC indicated treatment of trifluridine-tipiracil plus bevacizumab may be associated with prolonged OS and PFS but also an increased risk of treatment-related AEs in patients with mCRC, compared to other treatments such as BSC, regorafenib, or trifluridine-tipiracil alone. However, there was substantial uncertainty due to limitations, including significant between-trial heterogeneity, potential intransitivity, and the absence of HRQoL outcomes of importance to patients and clinicians.

References

1.Canadian Cancer Society. Colorectal Cancer. Accessed 2023 Oct 16.

2.Canadian Cancer Statistics Advisory in collaboration with the Canadian Cancer Society Statistics Canada and the Public Health Agency of Canada. Canadian Cancer Statistics: a 2022 special report on cancer prevalence. Toronto (ON): Canadian Cancer Society; 2022: https://cancer.ca/Canadian-Cancer-Statistics-2022-EN. Accessed 2023 Oct 22.

3.Brenner DR, Poirier A, Woods RR, et al. Projected estimates of cancer in Canada in 2022. CMAJ. 2022;194(17):E601-e607. PubMed

4.Statistics Canada. Table: 13-10-0751-01. Number of prevalent cases and prevalence proportions of primary cancer, by prevalence duration, cancer type, attained age group and sex. 2022.

5.Cervantes A, Adam R, Roselló S, et al. Metastatic colorectal cancer: ESMO Clinical Practice Guideline for diagnosis, treatment and follow-up. Ann Oncol. 2023;34(1):10-32. PubMed

6.Canadian Cancer Statistics Advisory Committee. Canadian cancer statistics. Toronto (ON): Canadian Cancer Society 2018: https://cdn.cancer.ca/-/media/files/research/cancer-statistics/2018-statistics/canadian-cancer-statistics-2018-en.pdf. Accessed 2023 Oct 16.

7.Macrae F, Bendell J. Clinical presentation, diagnosis, and staging of colorectal cancer. In: Post TW, ed. UpToDate. Waltham (MA): UpToDate; 2023: http://www.uptodate.com/. Accessed 2023 Oct 16.

8.Georgios KS, Bampali AD. Colorectal cancer. In: Bustamante M, ed. Colonoscopy and colorectal cancer screening: future directions. London (GB): IntechOpen Limited; 2013: https://www.intechopen.com/chapters/42023. Accessed 2023 Oct 16.

9.Bourakkadi Idrissi M, El Bouhaddouti H, Mouaqit O, Ousadden A, Ait Taleb K, Benjelloun EB. Left-sided colon cancer and right-sided colon cancer: are they the same cancer or two different entities? Cureus. 2023;15(4):e37563. PubMed

10.Clark JW, Sanoff HK. Systemic therapy for nonoperable metastatic colorectal cancer: approach to later lines of systemic therapy. In: Post TW, ed. UpToDate. Waltham (MA): UpToDate; 2023: http://www.uptodate.com. Accessed 2023 Sep 18.

11.Colon cancer. Version 2. 2023. NCCN clinical practice guidelines in oncology. Plymouth Meeting (PA): National Comprehensive Cancer Network; 2023 Apr 25.

12.Rectal cancer. Version 3. 2023. NCCN clinical practice guidelines in oncology. Plymouth Meeting (PA): National Comprehensive Cancer Network; 2023 May 26.

13.CADTH reimbursement review: provisional funding algorithm for metastatic colorectal cancer. Ottawa (ON): CADTH; 2021: https://www.cadth.ca/sites/default/files/attachments/2021-11/PH0005-000_CRC-FINAL-report-Nov11.pdf. Accessed 2023 Sep 19.

14.Morris VK, Kennedy EB, Baxter NN, et al. Treatment of metastatic colorectal cancer: ASCO guideline. J Clin Oncol. 2023;41(3):678-700. PubMed

15.Yoshino T, Cervantes A, Bando H, et al. Pan-Asian adapted ESMO Clinical Practice Guidelines for the diagnosis, treatment and follow-up of patients with metastatic colorectal cancer. ESMO Open. 2023;8(3):101558. PubMed

16.Stivarga (regorafenib): 40 mg tablets [product monograph]. Mississauga (ON): Bayer Inc.; 2020 Mar 9.

17.Lonsurf (trifluridine and tipiracil tablet): 15 mg trifluridine / 6.14 mg tipiracil, 20 mg trifluridine / 8.19 tipiracil [product monograph]. Oakville (ON): Taiho Pharma Canada, Inc.; 2020 Oct 29.

18.Provincial funding summary: trifluridine and tipiracil (Lonsurf) for metastatic colorectal cancer resubmission (pCODR 10173). Ottawa (ON): CADTH; 2020: https://www.cadth.ca/sites/default/files/pcodr/pcodr_provfund-10173-trifluridine-tipiracil_lonsurf_mcrc.pdf. Accessed 2023 Sep 19.

19.Provincial funding summary: regorafenib (Stivarga) for metastatic colorectal cancer. Ottawa (ON): CADTH; 2017: https://www.cadth.ca/sites/default/files/pcodr/pcodr-provfund_regorafenib_stivarga-resub_mcrc.pdf. Accessed 2023 Sep 19.

20.Balshem H, Helfand M, Schünemann HJ, et al. GRADE guidelines: 3. Rating the quality of evidence. J Clin Epidemiol. 2011;64(4):401-406. PubMed

21.Santesso N, Glenton C, Dahm P, et al. GRADE guidelines 26: informative statements to communicate the findings of systematic reviews of interventions. J Clin Epidemiol. 2020;119:126-135. PubMed

22.Clinical Study Report: CL3-95005-007. An open-label, randomised, phase III study comparing trifluridine/tipiracil in combination with bevacizumab to trifluridine/tipiracil monotherapy in patients with refractory metastatic colorectal cancer (SUNLIGHT study) [internal sponsor's report]. Princeton (NJ): Taiho Oncology, Inc; 2023 Jan 4.

23.Taiho Pharma Canada Inc. response to September 12, 2023 request for additional information regarding Lonsurf (trifluridine-tipiracil) CADTH review [internal sponsor's report]. Oakville (ON): Taiho Pharma Canada, Inc.; 2023 Sep 19.

24.Indirect treatment comparison in metastatic colorectal cancer [internal sponsor's report]. In: Drug Reimbursement Review sponsor submission: Lonsurf (trifluridine and tipiracil), 15 mg trilfuridine/ 6.14 mg tipiracil and 20 mg trifluridine/ 8.19 mg tipiracil tablets. Oakville (ON): Taiho Pharma Canada; 2023 Aug 10.

25.Biller LH, Schrag D. Diagnosis and treatment of metastatic colorectal cancer: a review. JAMA. 2021;325(7):669-685. PubMed

26.Metastatic colorectal cancer. Clincial practice guideline GI-003 - version 13. Edmonton (AB): Alberta Health Services; 2023: https://www.albertahealthservices.ca/assets/info/hp/cancer/if-hp-cancer-guide-gi003-colorectal-metastatic.pdf. Accessed 2023 Sep 19.

27.Clark JW, Sanoff HK. Systemic chemotherapy for metastatic colorectal cancer: general principles. In: Post TW, ed. UpToDate. Waltham (MA): UpToDate; 2023: http://www.uptodate.com. Accessed 2023 Sep 18.

28.Koulouridi A, Karagianni M, Messaritakis I, et al. Prognostic value of KRAS mutations in colorectal cancer patients. Cancers (Basel). 2022;14(14). PubMed

29.Saeed O, Lopez-Beltran A, Fisher KW, et al. RAS genes in colorectal carcinoma: pathogenesis, testing guidelines and treatment implications. J Clin Pathol. 2019;72(2):135-139. PubMed