CADTH Reimbursement Review

Pralsetinib (Gavreto)

Sponsor: Hoffman-La Roche Ltd.

Therapeutic area: RET fusion–positive non–small cell lung 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.

Introduction

Lung cancer is the most frequently diagnosed cancer in Canada and the leading cause of cancer-related deaths,¹ with more than 29,600 new diagnoses and 21,000 disease-related deaths projected in 2021.¹ Lung cancers are classified into 2 types based on histology: small cell lung cancer and non–small cell lung cancer (NSCLC), the latter being the most common histology. Patients may experience worsening coughs, chest pain, hemoptysis, malaise, weight loss, dyspnea, and hoarseness at clinical presentation or upon chest imaging.^1,2 The adjusted 5-year net survival estimate in Canada for all forms of lung cancers is 22%,¹ and the anticipated 5-year survival is approximately 25% for patients with NSCLC and 7% for patients with stage IV disease.³ Unfortunately, almost 50% of NSCLC diagnoses in Canada are made at stage IV, with only about 23.1% of cases diagnosed at early-stage I.¹ Abnormal RET receptor activation by rearrangement or mutation is recognized as an oncogenic driver for many cancers, including NSCLC. These alterations are commonly associated with patients with adenocarcinoma histology, younger patients (usually ≤ 60 years), and those with non-smoking or light-smoking status.⁴

Pralsetinib is an orally available, highly selective inhibitor of the RET receptor tyrosine kinase. It is available in 100 mg capsules. Pralsetinib received a Notice of Compliance with conditions from Health Canada on June 30, 2021, for the treatment of adult patients with RET fusion–positive locally advanced unresectable or metastatic NSCLC.

The objective of this review is to perform a systematic review of the beneficial and harmful effects of pralsetinib 400 mg oral tablets for the treatment of adult patients with RET fusion–positive locally advanced unresectable or metastatic NSCLC.

Stakeholder Perspectives

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

Patient Input

One patient advocacy group, Lung Cancer Canada (LCC), provided input regarding patients’ experiences, values, and preferences related to RET fusion–positive NSCLC and its treatment. The group was able to gather information from 4 patients with RET fusion–positive NSCLC who had received pralsetinib treatment and 1 caregiver from 4 countries (Canada, the US, Ireland, and Norway) in March 2022.

Patients and caregivers emphasized the consequences of delayed diagnosis due to mild and unspecific symptoms such as lower back pain, weight loss, coughing, and shortness of breath. As a result, patients are often diagnosed at an advanced or metastatic stage for which the prognosis is relatively poor. Patients reported that chemotherapy has limited long-term effectiveness due to toxicity. Patients experienced harsh side effects, such as fatigue, hair loss, and blood clots, which have negative impacts on patients’ functionality and quality of life and create additional burdens on patients.

Patients who had experience with pralsetinib indicated that the drug was effective in shrinking tumour size, producing less-severe side effects and improving functionality. For all 4 patients, pralsetinib treatment allowed them to continue working or doing household chores and conduct their daily lives with autonomy and dignity. The most frequently reported side effect was fatigue, which happened during onboarding and the initial weeks of treatment. Patients also reported other general side effects such as dry mouth, anemia, constipation, loss of appetite, and itchiness and/or dry skin. One patient was re-hospitalized due to liver function conditions and had a severe headache. Patients reported alleviation of the side effects once their dosages were reduced.

Outcomes important to patients were treatment effectiveness in managing symptoms, stopping or delaying disease progression, settling patients into long-term remission for improved survivorship, having manageable side effects, maintaining independence and functionality that would minimize the burden on their caregivers and family members, and improved quality of life. Detailed information of the patient group input is provided in the Stakeholder Input document.

Clinician Input

Input From the Clinical Expert Consulted by CADTH

One clinical expert with experience in the diagnosis and management of NSCLC highlighted the differences in patients with RET fusion–positive NSCLC and the broader NSCLC population, with some of the key differences being that patients with RET fusion–positive NSCLC are younger and less likely to have a history of tobacco use. The expert noted that single-drug immunotherapy has limited activity in this population, and chemotherapy, while it is as effective in RET fusion patients as in the broader NSCLC population, does not have activity in the brain. Pralsetinib, a targeted oral therapy, is an option with good response rates and activity in the brain. The expert also noted the potential to reduce hospital burden through oral administration. This is in contrast to IV administration of immunotherapy and chemotherapy, which is more likely to require in-person or hospital care for adverse events (AEs). The clinical expert noted that radiographic assessments would generally be conducted every 8 to 12 weeks, with clinical assessments every 3 to 4 weeks, and patients would be discontinued from treatment in the presence of unacceptable AEs, patient preference, and symptomatic disease progression, with the exception of oligoprogression amenable to local intervention. Detailed information on the clinical expert input is provided in the Stakeholder Input section in the main body of the report.

Clinician Group Input

Clinician group input on the review of pralsetinib for the treatment of RET fusion–positive locally advanced unresectable or metastatic NSCLC was received from 2 groups: LCC and the Ontario Health – Cancer Care Ontario (CCO) Lung Cancer Drug Advisory Committee. The input was generally consistent with that provided by the clinical expert. The submission from the CCO committee suggested that patients with an Eastern Cooperative Oncology Group Performance Status (ECOG PS) of 3 or greater would be least suitable for treatment with pralsetinib, whereas the clinical expert supported extending access to pralsetinib to patients with an ECOG PS of 2 or 3. The submission from LCC highlighted pandemic considerations and the potential for a reduced patient footprint in cancer centres with an oral therapy such as pralsetinib. Detailed information of the clinician group input is provided in the Stakeholder Input document.

Drug Program Input

The drug programs provide input on each drug reviewed through CADTH’s reimbursement review process by identifying issues that may affect their ability to implement a recommendation. The drug plans identified implementation issues related to relevant comparators; considerations for initiation, prescribing, and discontinuation of therapy; generalizability; funding algorithms; care provision; system issues; and economic considerations. The clinical expert consulted by CADTH for this review weighed evidence from the included study and other clinical considerations to provide responses to the drug plan’s implementation questions. Detailed information of the drug program input is provided in the Drug Program Input section in the main body of the report.

Clinical Evidence

Pivotal Studies and Protocol-Selected Studies

Description of Studies

The ARROW trial (N = 281; safety population at the November 6, 2020, data cut-off) is an ongoing phase I and II, open-label, single-arm study of pralsetinib in RET fusion–positive locally advanced or metastatic NSCLC patients. The primary objective of the phase II portion of the study was to determine the efficacy (measured by the overall response rate [ORR]) and safety of pralsetinib 400 mg once daily. The phase II portion of the study and the 400 mg once daily dosage are the focus of this report, as they represent the Health Canada–approved indication. Intracranial ORR, duration of response (DOR), progression-free survival (PFS), overall survival (OS), and health-related quality of life (HRQoL) were secondary end points in the trial. There was no predefined duration of treatment; patients with progressive disease could remain on treatment if the investigator determined that it was in the best interest of the patient to do so. Two unplanned interim clinical data cut-offs are presented in this report. The first is the November 18, 2019, data cut-off presented in a provided clinical study report⁵ and the November 6, 2020, data cut-off that was summarized in a European Medicines Agency report.⁶ The efficacy population in both analyses was a subset of patients who had been enrolled at the time of data cut-off to allow for an appropriate amount of time for patients to achieve an ORR. The respective cut-off dates were July 11, 2019, and May 22, 2020. A safety analysis was provided for all patients who had been enrolled up to each data cut-off date. At the November 6, 2020, data cut-off, the median age was 60 years, with similar proportions of each sex (52.4% female and 47.6% male); 51.9% of patients were White and 39.5% of patients were Asian.

Efficacy Results

Key efficacy outcomes are summarized in Table 2.

Table 2: Summary of Key Results From Pivotal and Protocol-Selected Studies

AE = adverse event; CI = confidence interval; EORTC QLQ-C30 = European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire Core 30; HRQoL = health-related quality of life; NR = not reported; SAE = serious adverse event; SD = standard deviation.

Note: The November 18, 2019, data cut-off was an unplanned interim analysis with requirement for patients to have been enrolled on or before July 11, 2019. The November 6, 2020, data cut-off was an unplanned interim analysis, generated in support of the European Medicines Agency submission, with a requirement for patients to have been enrolled on or before May 22, 2020.

^a95% CI based on the Greenwood formula.

^b95% CI based on an exact binomial distribution using the Clopper-Pearson method.

^cPatients who discontinued treatment.

Source: ARROW Clinical Study Report,⁵ European Medicines Agency Pralsetinib Public Assessment Report.⁶

Overall Survival and Progression-Free Survival

At the November 18, 2019, data cut-off, the median OS follow-up time was 10.5 months (95% confidence interval [CI], 9.7 to 13.1); the median OS had not been reached. At data cut-off, 19.7% of patients had died and 80.3% were censored. Median PFS was 12.7 months (95% CI, 9.1 to not estimable) with 62.1% of patients censored at data cut-off.

At the November 6, 2020, data cut-off, the median OS follow-up time was 17.1 months (95% CI, 13.7 to 19.6); the median OS had not been reached. At data cut-off, 24.5% of patients had died and 75.5% were censored. Median PFS was 16.4 months (95% CI, 11.0 to 24.1), with 56.2% of patients censored at data cut-off.

Health-Related Quality of Life

Baseline mean global health status scores on the European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire Core 30 (EORTC QLQ-C30) was |||||||||||| recorded from |||||| total patients. At the |||||| time point, the mean global health status score, recorded from ||||||| patients was ||||||| corresponding to a mean change from baseline of ||||||||||||||, meeting the published minimally important difference (MID) for a moderate improvement. Data for HRQoL were only available at the November 18, 2019, data cut-off.

Overall Response Rate, Intracranial Overall Response Rate, and Duration of Response

At the November 18, 2019, data cut-off, the ORR was |||||||||||||||||||||. Among patients who achieved a response (|||||||), the median DOR ||||||||||||||||||||||||||||||||||||||||||||||||||||||||. In ||||||| patients with measurable intracranial lesions, the ORR rate was |||||||||||||||||||||). At the November 6, 2020, data cut-off, the ORR was 64.4% (95% CI, 57.9 to 70.5). Among patients who achieved a response (150 of 233), the median DOR was 22.3 months (95% CI, 14.7 to not reported). In the 10 patients with measurable intracranial lesions, the ORR was 70.0% (95% CI, 34.8 to 93.3). Additional subgroups reported for patients who received prior systemic therapy, prior platinum therapy, prior nonplatinum therapy, and no prior systemic therapy, along with analysis of the measurable disease population, showed results similar to those of the primary analysis.

Harms Results

At the November 18, 2019, data cut-off, ||||||| of a total of ||||||| patients in the safety analysis set experienced at least 1 AE. The most common AEs were increased aspartate transaminase (AST) (|||||||), constipation (|||||||), anemia (|||||||), and increased alanine transaminase (ALT) (|||||||). At the November 6, 2020, data cut-off, 99.3% of the 281 patients in the safety analysis set experienced at least 1 AE. The most common were anemia (45.9%), increased AST (44.8%), constipation (42.0%), hypertension (34.2%), and increased ALT (32.7%).

At the November 18, 2019, data cut-off, pneumonitis was reported as a grade 3, 4, or 5 AE by ||||||| of patients and a serious adverse event (SAE) by ||||||| of patients, resulting in a dose reduction in ||||||| of patients, a dose interruption in 8.4% of patients, and treatment discontinuation by ||||||| of patients. There were |||||||||||||| attributed to pneumonitis at the November 18, 2019, data cut-off. At the November 6, 2020, data cut-off, pneumonitis was reported as a grade 3, 4, or 5 AE by 2.1% of patients and an SAE by 4.6% of patients, resulting in a dose reduction in 6.4% of patients, a dose interruption in 9.6% of patients, and treatment discontinuation by 2.5% of patients. No deaths attributable to pneumonitis were reported at the November 6, 2020, data cut-off.

Critical Appraisal

The most important limitation with the ARROW trial stems from the single-arm design. This design increased the risk of bias in estimating treatment effects due to the potential for confounding related to unidentified prognostic factors and treatment-effect modifiers that could affect the activity of the study drug. While RET fusion patients are considered rare, accounting for 1% to 2% of all NSCLC patients, a phase III randomized trial is currently being conducted for pralsetinib in this patient population.

The results for the primary end point of ORR rejected the null hypothesis for response, and the clinical expert consulted described the response rates and duration of responses as impressive. No pre-specified interim analyses were planned in the statistical analysis plan for ARROW, increasing the potential for bias and type I error with successive ad hoc data cut-off analyses.

Patients recruited to the treatment-naive group were initially required to be deemed unsuitable for standard of care (SOC) chemotherapy, but the inclusion criteria were later amended to allow all treatment-naive patients. This amendment may have biased the results against pralsetinib if the patients recruited before the amendment had a worse prognosis compared to the average first-line patient. Important protocol deviations further increased the level of uncertainty, given that 16 patients at the November 6, 2020, data cut-off did not have measurable disease at baseline and 1 had inconclusive evidence of RET fusion. Patients who did not have measurable disease would be unlikely to record a response, biasing the ORR results against pralsetinib; however, OS and PFS would be unaffected. Results of subgroup analyses of the post-eligibility revision group, as well as the measurable-disease-only group, were similar to those of the primary analysis.

Uncertainty remains regarding the long-term effects of pralsetinib on secondary outcomes such as PFS, OS, and HRQoL given the lack of a comparator and the immaturity of the survival data as the median OS was not reached. The HRQoL results, which are important to patients, appear to be positive, reaching the MID for a moderate improvement; however, the number of patients in the analysis is low because this measure was added to the protocol through an amendment after initiation of the study, and patient numbers were further reduced as the time points progressed. There is potential for selection bias over time given that long-term survivors in the trials tend to be healthier patients. The absence of a comparator arm and open-label design introduces reporting bias, and the impact of pralsetinib on patient-reported outcomes in relation to other therapies is unknown.

According to the clinical expert consulted by CADTH, the demographic and disease characteristics of the ARROW population were reflective of the Canadian population of patients with RET fusion–positive NSCLC.

Indirect Comparisons

Description of Studies

In the absence of direct comparative evidence from trials, the aim of each analysis was to compare the efficacy (OS and PFS) of pralsetinib in patients with RET fusion–positive locally advanced or metastatic NSCLC versus patients with wild-type NSCLC receiving comparators of interest. The studies identified for comparators of interest were KEYNOTE-042 (pembrolizumab monotherapy),⁷ KEYNOTE-189 (pembrolizumab plus platinum-based chemotherapy plus pemetrexed),⁸ IMpower132 (platinum-based chemotherapy plus pemetrexed),⁹ OAK (second-line docetaxel),¹⁰ CheckMate 057 (second-line nivolumab),¹¹ and GOIRC 02 to 2006¹² pooled with NVALT7¹³ (carboplatin plus pemetrexed). The IMpower132 and OAK studies were chosen due to the availability of individual patient data (IPD), allowing a propensity-score weighting method to be applied to adjust for differences in study populations for the first-line platinum-based chemotherapy plus pemetrexed and second-line docetaxel comparisons. All other comparisons were naive unadjusted analyses that did not account for differences in population characteristics.

Efficacy Results

Propensity-Score Weighted Analysis

The adjusted hazard ratio (HR) for OS in the comparison of pralsetinib versus platinum-based chemotherapy plus pemetrexed was |||||||||||||||||||||. The adjusted HR for OS in the comparison of pralsetinib versus docetaxel was ||||||||||||||||||||||||||||. The adjusted HR for PFS in the comparison of pralsetinib versus platinum-based chemotherapy plus pemetrexed was||||||||||||||||||||||||||||. The adjusted HR for PFS in the comparison of pralsetinib versus docetaxel was ||||||||||||||||||||||||||||.

Naive Comparisons

The HRs for OS and PFS for the naive comparisons of pralsetinib versus first-line pembrolizumab monotherapy, first-line pembrolizumab plus pemetrexed-platinum, second-line nivolumab, and second-line pemetrexed plus carboplatin all favoured pralsetinib.

Critical Appraisal

A key limitation of the indirect treatment comparison (ITC) submitted by the sponsor comes from the single-arm design of the ARROW study, precluding any connected network of trials and resulting in a reliance on unanchored comparisons. For 2 comparisons, first-line platinum-based chemotherapy plus pemetrexed and second-line docetaxel, the sponsor had access to IPD and was able to conduct propensity-score weighting to adjust for between-trial differences in population characteristics. The methodology for choosing the prognostic factors to adjust for relied on data availability in place of a rigorous literature search. The analysis assumed the presence of RET fusion was not a predictive factor and therefore not included in the model. While a lack of evidence available in patients with RET fusion–positive NSCLC required this assumption, patients who are RET fusion–positive tend to be younger, less likely to smoke, and have mostly nonsquamous histology. Patients who are RET fusion–positive are more likely to respond to targeted RET therapy and less likely to respond to immunotherapy. The sponsor provided evidence from Hess et al. (2021)¹⁴ suggesting that, before the introduction of RET inhibitors, there was no relationship between RET status and outcomes in an adjusted model. However, the clinical expert consulted for this review suggested that the presence of RET fusion is a positive predictor for the efficacy of RET-targeted therapy and a negative predictor for the effect of immunotherapy. A methodology to adjust for prognostic factors other than RET fusion status was used; however, it was not possible to account for all differences in patient characteristics. With regard to the naive comparisons specifically, no adjustments were made. Patients with positive or negative RET fusion status are therefore expected to respond differently to pralsetinib and it is unclear how similar the patient populations in the comparator studies are to those enrolled in the ARROW trial, despite the adjustments in propensity-score weighted analysis. Once adjusted, the trial populations were vastly reduced in size (||||||||||||||||||||| in the case of the OAK trial), likely a result of the initial imbalance in baseline covariates.

As IPD were available for only 2 comparisons in the sponsor-submitted ITC, the remaining comparisons were unadjusted naive comparisons (no adjustments for between-trial differences in population characteristics were made). This introduces major uncertainty to the results, given that the prognostic factors identified by the sponsor as having an impact on treatment effects remained heterogenous for the naive comparisons. Conclusions cannot be drawn based upon the naive comparisons and conclusions drawn from the propensity-score weighted analysis are uncertain.

With these limitations in mind, all results were directionally consistent and in line with the clinical expert’s expectations that pralsetinib is likely superior to the comparators included in the ITC analysis.

An additional ITC identified from the literature search, Popat et al. (2022),¹⁵ compared first-line patients receiving pralsetinib in the ARROW trial to synthetic control arms sourced from 3 real-world populations. The first involved patients with RET fusion–positive NSCLC receiving a basket of best alternative therapy (most commonly pembrolizumab plus chemotherapy). The remaining 2 real-world populations comprised patients with wild-type NSCLC receiving pembrolizumab monotherapy and pembrolizumab plus chemotherapy, respectively. The analysis used inverse probability weighting (IPW) where possible to adjust for differences in prognostic factors. The results indicate that patients given pralsetinib received a statistically significant benefit in OS and PFS compared to the chosen comparators, which is consistent with the expectations of the clinical expert consulted for this review; however, the same limitations are present as in the sponsor-submitted ITC. The analysis is an unanchored ITC relying on a limited number of prognostic factors and a small effective sample size (ESS) compared to the original sample sizes of the populations.

Other Relevant Evidence

Description of Studies

The CADTH review team identified an ongoing phase III, randomized, open-label study, AcceleRET-Lung, comparing pralsetinib to a physician’s choice of platinum chemotherapy–based regimen based on SOC treatments for first-line treatment of patients with RET fusion–positive metastatic NSCLC who have not previously received systemic anticancer therapy for metastatic disease. No results are currently available, as this trial is actively recruiting patients. The estimated primary completion date (on which the last participant in a clinical study will be examined or receive an intervention to collect final data for the primary outcome measure) and study completion date (on which the last participant in a clinical study will be examined or receive an intervention or treatment to collect final data for the primary outcome measures, secondary outcome measures, and AEs) are September 30, 2023, and December 31, 2024, respectively.

Conclusions

The evidence supporting the funding request for pralsetinib was derived from an ongoing phase I and II, open-label, single-arm study, ARROW. The ORR observed in the ARROW trial, based on unplanned interim analysis results, suggested favourable tumour response in both treatment-naive and treatment-experienced patients and was consistent with further follow-up analysis. The ORR and DOR, including central nervous system (CNS) ORR, were considered clinically meaningful by the clinical expert consulted for the review. The ability to draw conclusions from time-to-event end points of PFS and OS were affected by the immaturity of the data and the single-arm design of the trial. The safety profile of pralsetinib in the ARROW trial was considered by the clinical expert consulted for this review to be an improvement compared to SOC chemotherapy and immunotherapy. According to expert clinical opinion, the differences in safety profiles compared with selpercatinib highlight the benefits to patients that come with additional treatment options. The ITC submitted to inform the comparative effects of pralsetinib was associated with limitations that prevented drawing conclusions from the results, and uncertainty remains in the comparative effectiveness and safety of pralsetinib.

Introduction

Disease Background

Lung cancer is the most frequently diagnosed cancer in Canada and the leading cause of cancer-related deaths in males and females,¹ with more than 29,600 new diagnoses (12.5% of new cancer cases in males and 13.3% new cases in females) and 21,000 disease-related deaths (24.2% of cancer deaths in males and 25.8% in females) projected in 2021.¹ The adjusted 5-year net survival estimate in Canada for all forms of lung cancers is 22%¹ and the anticipated 5-year survival is approximately 25% for patients with NSCLC and 7% for patients with stage IV disease.³ Smoking is an established risk factor for developing lung cancer, accounting for more than 72% of newly diagnosed cases in Canada.^1,2

Lung cancers are classified into 2 types based on histology: small cell lung cancer and NSCLC. The latter is the most common histology, and is further categorized based on cell type: adenocarcinomas, squamous cell carcinomas, or large cell carcinomas. The clinical expert shared information from their jurisdiction’s provincial data showing that up to 80% of all NSCLC cases are classified as nonsquamous and suggested that this figure could be higher in areas with lower tobacco use.

Early diagnosis improves prognoses and offers the best chance at optimal therapy. Diagnosis is based on symptom presentation^2,16; patients may experience worsening coughs, chest pain, hemoptysis, malaise, weight loss, dyspnea, and hoarseness at clinical presentation or upon chest imaging.^1,2 In advanced or metastatic disease, patients experience additional symptom burdens, such as trouble breathing, chronic coughing and chest pain, pain in bones or the spine, yellowing of the skin or eyes, weakness or numbness of arms or legs, fatigue and unexplained weight loss depression, insomnia, and pain.^17,18 Staging at diagnosis is key in determining disease prognosis and facilitates treatment selection.^2,18 Diagnosis at an advanced stage is a significant contributing factor to early mortality and challenging for disease management in real-world practice. Unfortunately, almost 50% of NSCLC diagnoses in Canada are made at stage IV, with only about 23.1% of cases diagnosed at early-stage I.¹

The expression of oncogenic driver mutations in tumours plays a vital role in patient response to treatment when there is an accessible targeted therapy specific to that mutation.¹⁸ Several predictive driver mutations identified in recent years, including mutations in EGFR, ROS1, KRAS, ALK, BRAF V600E, and others have greatly influenced treatment strategies in practice, improved patient quality of life, and increased OS for patients.^18-20 The RET protein is a transmembrane tyrosine kinase receptor encoded by the RET gene and is known to play a substantial role in the development and maintenance of many systems, including the enteric nervous and genitourinary systems in neonates.²¹ Abnormal RET receptor activation by rearrangement or mutation was recognized as an oncogenic driver for many cancers, including NSCLC. These alterations are commonly associated with patients with adenocarcinoma histology, younger patients (usually ≤ 60 years), and those with no or minimal history of tobacco use.⁴ Prevalence estimates from studies show that only about 1% to 2% of NSCLC cases are RET fusion–positive.²² Testing for driver mutations at initial diagnosis using molecular techniques such as next-generation sequencing or polymerase chain reaction amplification is available across jurisdictions in Canada.^18,19

Standards of Therapy

The clinician experts and clinician groups consulted for this review outlined similar treatment goals for patients with advanced or metastatic disease, including improvement in median OS, rapidity and prolonged improvement in cancer-related symptoms and improvement in quality of life (given that patients with advanced and metastatic disease experience greater symptom burden), reduced treatment-related toxicity, prevention, and treatment of brain metastasis.

Expert opinion from the clinician groups and drug plans consulted during the CADTH review emphasized the importance of treatment combinations funded in practice for patients without confirmed RET fusion. For the treatment-naive population, first-line treatment combinations with platinum plus pemetrexed and pembrolizumab were identified as the most preferred in patients with programmed death ligand 1 (PD-L1) expression below 50%. For those patients with PD-L1 expression of 50% or greater, options include single-drug pembrolizumab or platinum plus pemetrexed and pembrolizumab, with the latter often favoured for patients who are nonsmokers, have high disease (or symptom) burdens, or who have a known oncogene driver mutation associated with poor outcomes when treated with immunotherapy alone. For patients who progressed on prior systemic therapy, treatment options with second-line platinum plus pemetrexed are most preferred if they had received pembrolizumab in the first-line therapy. Second-line anti–PD-L1 therapy, using pembrolizumab, nivolumab, or atezolizumab, is favoured for those who received platinum plus pemetrexed as first-line therapy, and second-line docetaxel for those who progressed on platinum plus pemetrexed and pembrolizumab.

The drug plans consulted for this review identified publicly funded options for patients with advanced unresectable or metastatic NSCLC who are treatment-naive, including pembrolizumab single-drug therapy for patients with PD-L1 expression greater than or equal to 50%, pembrolizumab plus pemetrexed plus platinum-based chemotherapy, or platinum-based chemotherapy based on histology. Nivolumab in combination with ipilimumab plus platinum doublet chemotherapy is under consideration for listing in provinces. In the second-line setting, the drug plans noted that funded options may include immune checkpoint inhibitors if no prior programmed cell death protein 1 (PD-1) inhibitor was administered to the patient (either pembrolizumab, nivolumab, atezolizumab depending on patient’s PD-L1 status), or chemotherapy if a prior PD-1 inhibitor (docetaxel or pemetrexed) was administered. The LCC clinician group pointed out that evidence from some cohort studies indicates that patients with RET fusion–positive NSCLC are sensitive to pemetrexed. The group therefore noted that, in the absence of any randomized data, a combination of pemetrexed and platinum will likely be the most efficacious therapy in patients with RET fusion–positive NSCLC who had received only pembrolizumab as first-line therapy.

The clinician expert consulted identified treatments similar to those outlined by the clinician group and drug plans. The expert mentioned that the most preferred therapy used in the first-line setting across jurisdictions in Canada is a triplet therapy of platinum, pemetrexed, and pembrolizumab regardless of the PD-L1 tumour proportion score. While they acknowledged that single-drug immunotherapies are approved and available in practice for those with a PD-L1 tumour proportion score of 50% or greater, they added that studies have reported poor response rates to immunotherapy when given alone in patients with RET fusion–positive NSCLC. The expert indicated that patients with RET fusion–positive NSCLC (most likely with adenocarcinoma histology) have been shown to respond to platinum and pemetrexed combinations, and other platinum doublets would be considered inferior for any adenocarcinoma patient.

The clinical expert added that, beyond the first line (after administration of a triple therapy), single-drug docetaxel is the typical SOC, although there is limited evidence for outcomes specific to patients with RET fusion–positive NSCLC using docetaxel. If a patient received pembrolizumab in the first line, a doublet combination of platinum and pemetrexed may be administered, and if they received platinum plus pemetrexed in the first line, they may likely receive immunotherapy (e.g., pembrolizumab, nivolumab, or atezolizumab) in the second line. However, if the patient was tested for RET fusion, they may likely be placed on docetaxel after platinum-pemetrexed doublet therapy in the second line rather than immunotherapy (based on reports of low response rates to immunotherapy in patients positive for RET fusion. The experts added that gemcitabine and vinorelbine are available as therapies in the third line and beyond. Other nonsystemic options outlined included radiation and surgical interventions employed as aggressive modalities in patients with oligometastatic disease or as palliative interventions with the goal of alleviating symptoms.

Drug

Pralsetinib is an orally available, highly selective, adenosine triphosphate–competitive, small-molecule inhibitor of RET receptor tyrosine kinase. It is available in 100 mg capsules. Pralsetinib received a Notice of Compliance with conditions from Health Canada on June 30, 2021, for the treatment of adult patients with RET fusion–positive locally advanced unresectable or metastatic NSCLC. The sponsor is requesting reimbursement for the treatment of adult patients with RET fusion–positive locally advanced unresectable or metastatic NSCLC, specifically noting that treatment should continue as long as the patient is deriving a clinical benefit from therapy or until unacceptable toxicity, in line with the Health Canada product monograph.

Pralsetinib has obtained regulatory approval, including the FDA (September 4, 2020) and the European Medicines Agency (November 18, 2021) and is marketed in other regulatory jurisdictions for indications similar to those outlined in the Canadian product monograph. Market approval in Canada was granted based on evidence generated from the ARROW trial, a phase I and II open-label trial in patients 18 years and older with RET-altered NSCLC, medullary thyroid cancer, and other RET-altered solid tumours. The recommended dose is 400 mg, taken as four 100 mg capsules, once daily.

Table 3: Key Characteristics of Pralsetinib

ATP = adenosine triphosphate; NSCLC = non–small cell lung cancer.

^aHealth Canada–approved indication

Source: Pralsetinib product monograph.²³

Stakeholder Perspectives

Patient Group Input

This section was prepared by CADTH staff based on the input provided by patient groups.

One patient advocacy group, LCC, provided input regarding patients’ experiences, values, and preferences related to RET fusion–positive NSCLC and its treatment. The group was able to gather information from 4 patients with RET fusion–positive NSCLC who had received pralsetinib treatment and 1 caregiver in March 2022; the respondents were from Canada, the US, Ireland, and Norway.

Patients and caregiver respondents emphasized the consequences of a delay in diagnosis due to mild and unspecific symptoms such as lower back pain, weight loss, coughing, and shortness of breath. As a result, patients are often diagnosed at an advanced or metastatic stage for which the prognosis is relatively poor. Patients reported that chemotherapy has limited long-term effectiveness due to toxicity. Patients experienced harsh side effects, such as fatigue, hair loss, and blood clots, which have negative effects on patients’ functionality and quality of life and create additional burdens on patients.

Patients who had experience with pralsetinib indicated that the drug was effective in shrinking the tumour size, resulted in less-severe side effects, and improved functionality. For all 5 patients, the benefits of pralsetinib treatment allowed them to continue working or doing household chores and conduct their daily lives with autonomy and dignity. The most frequently reported side effect was fatigue, which happened during onboarding and the initial weeks of treatment. Patients also reported other general side effects, such as dry mouth, anemia, constipation, loss of appetite, and itchiness and/or dry skin. One patient was re-hospitalized due to liver function conditions and had a severe headache. Patients reported alleviation of the side effects once their dosages were reduced.

Outcomes important to patients are treatment effectiveness in managing symptoms, stopping or delaying disease progression, settling patients into long-term remission for improved survivorship, having manageable side effects, maintaining patients’ independence and functionality (thereby minimizing the burden on their caregivers and family members), and improving quality of life. Detailed information of the patient group input is provided in the Stakeholder Input document.

Clinician Input

Input From Clinical Experts Consulted by CADTH

All CADTH review teams include at least 1 clinical specialist with expertise in 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, and providing guidance on the potential place in therapy). The following input was provided by a clinical specialist with expertise in the diagnosis and management of locally advanced and metastatic NSCLC.

Unmet Needs

At the time of the pralsetinib submission, the clinical expert noted that no reimbursed therapies for NSCLC that specifically target RET fusion are available in Canada . Although the clinical expert noted that selpercatinib has received a positive recommendation with conditions from CADTH, the time from a final recommendation to actual reimbursement by different provincial jurisdictions can be considerable. At present, Eli Lilly offers a patient support program to facilitate access to selpercatinib for Canadian patients.

The clinical expert explained that patients are treated with chemotherapy and immunotherapy that require IV infusions and come with significant adverse effects requiring intensive supportive care and hospital stays. Furthermore, not all patients respond to chemotherapy in combination with immunotherapy (the response rate in the KEYNOTE-189 trial with platinum-pemetrexed-pembrolizumab was 48%).⁸ Because single-drug immunotherapy has limited utility in the RET fusion population and chemotherapy does not have activity in the brain, patients receiving chemotherapy who develop brain metastasis will require brain radiation, which carries a significant risk of toxicity.

Place in Therapy

The clinical expert highlighted that pralsetinib would be given as a single drug as an alternative to selpercatinib. As targeted therapies, they would ideally be used in first-line therapy, with chemotherapy and immunotherapy shifted to the second or later line. The clinical expert reiterated that patients should receive either pralsetinib or selpercatinib, but not both. It would be inappropriate to begin patients on chemotherapy and immunotherapy, which, according to the clinical expert, would be less effective, more toxic, and carry a larger burden on the health care system given that IV therapies must be administered in clinics and are more likely to require in-person or in-hospital supportive care for AEs. Additionally, when considering attrition between lines of therapy, the clinical expert highlighted the importance of offering either pralsetinib or selpercatinib in the first line so that the largest number of patients with RET fusion–positive NSCLC are able to gain the expected benefits from the therapy.

The clinical expert explained that if patients have already received or are currently receiving treatment other than a RET-specific tyrosine kinase inhibitor due to a lack of availability of pralsetinib and selpercatinib at the time of initiation of first-line therapy, pralsetinib or selpercatinib should be used in the next line of therapy upon progression.

Patient Population

According to the clinical expert, all patients with incurable RET fusion–positive NSCLC can be expected to respond to pralsetinib. The clinical expert indicated that patients who are candidates for curative-intent therapy should not be offered pralsetinib. Multiple molecular tests are used to detect RET fusion: immunohistochemistry, fluorescence in situ hybridization, reverse transcriptase polymerase chain reaction amplification, and next-generation sequencing using either DNA or both DNA and RNA. Although testing technology has become more affordable in Canada, access to next-generation sequencing can still vary across jurisdictions, according to the clinical expert.

Assessing Response to Treatment

The clinical expert noted that radiographic assessments of patients receiving pralsetinib would generally be conducted every 8 to 12 weeks, or sooner if new symptoms or physical findings suggest progression. Clinical assessments of the presence and severity of symptoms and AEs would be conducted every 3 to 4 weeks.

According to the clinical expert, clinically meaningful responses to treatment would include improved survival, improved time to progression of disease, reduced symptom burden, increased functioning, and improved quality of life.

Discontinuing Treatment

According to the clinical expert, the factors that should be considered when deciding to discontinue treatment with pralsetinib are the presence of unacceptable adverse effects, patient preference, and symptomatic disease progression, with the exception of oligoprogression amenable to local intervention to achieve disease control or progression in the CNS amenable only to brain-targeted therapy such as radiation.

Prescribing Conditions

The clinical expert noted that the prescribing criteria should be consistent with those of selpercatinib. That is, pralsetinib should be prescribed by clinicians with expertise in NSCLC and should not be prescribed in combination with other systemic anticancer drugs.

Clinician Group Input

This section was prepared by CADTH staff based on the input provided by patient groups.

Clinician group input on the review of pralsetinib for the treatment of RET fusion–positive locally advanced unresectable or metastatic NSCLC was received from 2 groups: LCC and the CCO Lung Cancer Drug Advisory Committee. The input was generally consistent with that supplied by the clinical expert. The submission from the CCO committee suggested that patients with an ECOG PS of 3 or greater would be least suitable for treatment with pralsetinib, whereas the clinical expert supported extending access to pralsetinib to patients with an ECOG PS of 2 or 3. The submission from LCC highlighted pandemic considerations and the potential for a reduced patient footprint in cancer centres that offer an oral therapy such as pralsetinib. Details of the clinician group input is provided in the Stakeholder Input document.

Drug Program Input

The drug programs provide input on each drug being reviewed through CADTH’s reimbursement review processes by identifying issues that may affect their ability to implement a recommendation. The implementation questions and corresponding responses from the clinical experts consulted by CADTH are summarized in Table 4.

Table 4: Summary of Drug Plan Input and Clinical Expert Responses

CNS = central nervous system; ECOG PS = Eastern Cooperative Oncology Group Performance Status; NSCLC = non–small cell lung cancer; PD-1 = programmed cell death protein 1; PD-L1 = programmed death ligand 1; pERC = CADTH pan-Canadian Oncology Drug Review Expert Review Committee.

Clinical Evidence

The clinical evidence included in the review of pralsetinib is presented in 3 sections. The first section, the systematic review, includes pivotal studies provided in the sponsor’s submission to CADTH and Health Canada, as well as those studies that were selected according to an a priori protocol. The second section includes indirect evidence from the sponsor and indirect evidence selected from the literature that met the selection criteria specified in the review. The third section includes a summary of key ongoing clinical trials.

Systematic Review (Pivotal and Protocol-Selected Studies)

Objectives

To perform a systematic review of the beneficial and harmful effects of pralsetinib 400 mg oral tablets for the treatment of adult patients with RET fusion–positive locally advanced unresectable or metastatic NSCLC.

Methods

Studies selected for inclusion in the systematic review included pivotal studies provided in the sponsor’s submission to CADTH and Health Canada, as well as those meeting the selection criteria presented in Table 5. Outcomes included in the CADTH review protocol reflect outcomes considered to be important to patients, clinicians, and drug plans.

The literature search for clinical studies was performed by an information specialist using a peer-reviewed search strategy according to the PRESS Peer Review of Electronic Search Strategies checklist.²⁴

Published literature was identified by searching the following bibliographic databases: MEDLINE All (1946–) via Ovid and Embase (1974–) via Ovid. All Ovid searches were run simultaneously as a multi-file search. Duplicates were removed using Ovid deduplication for multi-file searches, followed by manual deduplication in EndNote. The search strategy comprised both controlled vocabulary, such as the National Library of Medicine’s MeSH (Medical Subject Headings), and keywords. The main search concept was Gavreto (pralsetinib). Clinical trials registries searched included the US National Institutes of Health’s clinicaltrials.gov, WHO’s International Clinical Trials Registry Platform search portal, Health Canada’s Clinical Trials Database, and the European Union Clinical Trials Register.

No filters were applied to limit the retrieval by study type. Retrieval was not limited by publication date or by language. Conference abstracts were excluded from the search results. Appendix 1 provides detailed search strategies. The initial search was completed on March 31, 2022. Regular alerts updated the search until the meeting of the CADTH pan-Canadian Oncology Drug Review Expert Committee on August 10, 2022.

Grey literature (literature that is not commercially published) was identified by searching relevant websites from the Grey Matters: A Practical Tool For Searching Health-Related Grey Literature checklist.²⁵ Included in this search were the websites of regulatory agencies (US FDA and European Medicines Agency). Google was used to search for additional internet-based materials. Appendix 1 provides for more information on the grey literature search strategy.

These searches were supplemented by reviewing bibliographies of key papers and through contacts with appropriate experts. In addition, the manufacturer of the drug was contacted for information regarding unpublished studies.

Findings From the Literature

One study identified from the literature met criteria for inclusion in the systematic review (Figure 1). The included study identified from other relevant sources is summarized in Table 6. A list of excluded studies is presented in Appendix 2.

Figure 1: Flow Diagram for Inclusion and Exclusion of Studies

Source: ARROW Clinical Study Report,⁵ European Medicines Agency Pralsetinib Public Assessment Report,⁶ Gainor et al. (2021),²⁶ Health Canada Reviewers Report.²⁷

Table 6: Details of Included Studies

CBR = clinical benefit rate; CNS = central nervous system; DCR = disease control rate; DOR = duration of response; EORTC QLQ-C30 = European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire Core 30; MTD = maximum tolerated dose; NA = not applicable; NSCLC = non–small cell lung cancer; ORR = overall response rate; OS = overall survival; q.d. = once daily; RECIST 1.1 = Response Evaluation Criteria in Solid Tumors Version 1.1; ULN = upper limit of normal.

^aSpecific to NSCLC cohorts within the ARROW trial.

Source: ARROW Clinical Study Report,⁵ Gainor et al. (2021).²⁶

Description of Studies

The ARROW trial is an ongoing phase I and II, multi-centre, multi-cohort, single-arm open-label study of pralsetinib in patients with thyroid cancer, NSCLC, and other solid tumours with oncogenic RET fusions, consisting of a dose-escalation phase (phase I) and a dose-expansion phase (phase II). The first patient was enrolled on March 17, 2012, and the trial had 53 participating centres in the Europe, the US, and Asia at the time of the November 18, 2019, data cut-off. At that point, 179 patients had been recruited specifically in the NSCLC groups, and at the time of the November 6, 2020, data cut-off, 281 patients had been recruited in the NSCLC groups. Final enrolment in all groups and all tumour types was 528 patients. A schematic of the study design of both phases is shown in Figure 2.

Phase I: Dose Escalation

The objective of the dose-escalation phase was to determine the maximum tolerated dose and the recommended phase II dose (RP2D), along with safety and efficacy. The maximum tolerated dose was determined based on isotonic regression and was the dose for which the isotonic estimate of the toxicity rate, defined as a dose-limiting toxicity (grade 3 or greater AE), was closest to the target of 30%, although the RP2D could be chosen at a lower dose if clinical data warranted. The phase I portion of the study was completed on April 3, 2018, with the RP2D determined to be 400 mg once daily. Patients included in the phase I portion who had begun treatment at 400 mg once daily and met inclusion criteria for the phase II expansion were pooled in the final analysis. The focus of this report will be on the phase II expansion at the RP2D of 400 mg once daily, given that this is the Health Canada–approved dose. Phase I results for patients receiving doses other than 400 mg will not be expanded on further.

Phase II: Dose Expansion at Recommended Phase II Dose

The phase II portion of the ARROW trial is ongoing and includes patients from phase I who had been treated at the 400 mg once daily dosage in addition to newly recruited patients. Patients were recruited into 7 groups based on type of cancer and prior treatment history, as shown in Figure 2. As only groups 1 and 2 included patients with RET fusion–positive locally advanced and metastatic NSCLC, which is the indication of interest for this review, this report focuses solely on these groups, and results relating to patients in other groups are not expanded on further.

The primary objective of the phase II expansion was to determine the clinical efficacy of pralsetinib in patients with RET fusion–positive NSCLC, as measured by ORR, and to further define the safety and tolerability of pralsetinib at the 400 mg once daily dose. Two clinical data cut-offs are presented in this report, both of which were unplanned interim analyses conducted to support regulatory approval: the November 18, 2019, data cut-off presented in a provided clinical study report⁵ and the November 6, 2020, data cut-off that was summarized in a European Medicines Agency report.⁶ The efficacy population in both analyses was a subset of patients who had been enrolled at the time of data cut-off (July 11, 2019 and May 22, 2020, respectively) to allow for an appropriate amount of time for patients to achieve an ORR. Safety analysis was provided for all patients that had been enrolled up to each data cut-off.

There was no predefined duration of treatment; patients with progressive disease could remain on treatment if the investigator determined that it was in the best interest of the patient to do so.

Figure 2: ARROW Study Schematic

BID = twice daily; BOIN = Bayesian optimal interval; MTC = medullary thyroid cancer; NSCLC = non–small cell lung cancer; QD = once daily; RP2D = recommended phase II dose; TKI = tyrosine kinase inhibitor.

Source: ARROW Clinical Study Review.⁵

At the November 6, 2020, data cut-off a total of 41 patients (17.6%) had a major protocol deviation. Of these patients, 16 did not have measurable disease at baseline according to a blinded independent central review (BICR) and 1 patient had inconclusive evidence of RET fusion. Two of these 17 patients ultimately achieved a response.

Populations

Inclusion and Exclusion Criteria

Patients in the ARROW study, specifically those in the locally advanced and metastatic NSCLC groups, were required to be 18 years of age or older with an oncogenic RET fusion or mutation (excluding synonymous, frameshift, and nonsense mutations) and pathologically documented, definitively diagnosed, locally advanced or metastatic NSCLC. Patients with prior treatment experience were required to have been treated with a platinum-based chemotherapy, while patients recruited into the treatment-naive patients could not have received any prior systemic therapy. Prior platinum chemotherapy was permitted in the neo-adjuvant or adjuvant setting provided that the last dose was 4 months or more before the first dose of pralsetinib. Prior to the July 2019 protocol amendment, patients recruited into the treatment-naive group were required to have been deemed unsuitable for systemic therapy. Following the amendment, this requirement was removed. Patients were also required to have an ECOG PS of 0 or 1, and eligibility was adjusted following a July 2018 amendment to exclude patients with an ECOG PS of 2. Patients were required to have nonresectable and measurable disease as defined by the Response Evaluation Criteria in Solid Tumors Version 1.1 (RECIST 1.1).

Patients were excluded from the ARROW trial if a primary driver mutation other than RET was known to be present or if they had a primary CNS tumour or CNS metastases associated with progressive neurologic symptoms or that required increasing doses of corticosteroids to control. Patients with uncontrolled grade 3 or 4 congestive heart failure or clinically symptomatic interstitial lung disease were also excluded from the trial.

Baseline Characteristics

Baseline characteristics are shown in Table 7. There were 132 patients in the efficacy population at the November 18, 2019, data cut-off. The median age was ||||||||||||||, with slightly more females (||||||| compared with ||||||| males). Most patients were either Asian ||||||| or White |||||||. The efficacy population at the November 6, 2020, data cut-off included 233 patients. The median age in these patients remained 60 years, with similar proportions of the sexes (52.4% female and 47.6% male). At the data cut-off, 51.9% of patients were White and 39.5% of patients were Asian.

Baseline disease characteristics for the ARROW efficacy population are shown in Table 8. Almost all patients had tumours classified as adenocarcinomas, |||||||||||||| at the November 18, 2019, and November 6, 2020, data cut-offs, respectively. At the November 18, 2019, and November 6, 2020, data cut-offs, ||||||| and 37.3% of patients had a history of or current CNS metastasis, respectively. Nonlung lesion locations were evenly distributed at both data cut-offs. Few patients at either data cut-off presented with stage III disease, |||||||||||||| and 2.5% at the November 18, 2019, and November 6, 2020, data cut-offs, respectively. Most patients at both data cut-offs were never smokers, ||||||| and 62.2% at the November 18, 2019, and November 6, 2020, data cut-offs, respectively.

A summary of prior antineoplastic therapies is shown in Table 9. At the November 18, 2019, data cut-off, ||||||| of patients had received prior chemotherapy, ||||||| had received prior PD-1 and/or PD-L1 inhibitors, and |||||||||||||| had received prior multikinase inhibitors. At the November 6, 2020, data cut-off, 59.2% of patients had received prior chemotherapy, 29.6% had received prior PD-1 and/or PD-L1 inhibitors, and 18.9% had received prior multikinase inhibitors.

Table 7: Baseline Characteristics ARROW Efficacy Population

NR = not reported; SD = standard deviation.

Note: The November 18, 2019, data cut-off was an unplanned interim analysis with requirement for patients to have been enrolled on or before July 11, 2019. The November 6, 2020, data cut-off was an unplanned interim analysis, generated in support of the European Medicines Agency submission, with a requirement for patients to have been enrolled on or before May 22, 2020.

Source: ARROW Clinical Study Review,⁵ European Medicines Agency Pralsetinib Public Assessment Report.⁶

Table 8: Baseline Disease Characteristics

CNS = central nervous system; ECOG = Eastern Cooperative Oncology Group; NR = not reported; TNM = tumour node metastasis.

^aPatient had recurrent non–small cell lung cancer and was considered unfit for surgery.

Note: The November 18, 2019, data cut-off was an unplanned interim analysis with requirement for patients to have been enrolled on or before July 11, 2019. The November 6, 2020, data cut-off was an unplanned interim analysis, generated in support of the European Medicines Agency submission, with a requirement for patients to have been enrolled on or before May 22, 2020.

Source: ARROW Clinical Study Review,⁵ European Medicines Agency Pralsetinib Public Assessment Report.⁶

Table 9: Prior Antineoplastic Therapies

MKI = multikinase inhibitor; NR = not reported; PD-1 = programmed cell death protein 1; PD-L1 = programmed death ligand 1.

Note: The November 18, 2019, data cut-off was an unplanned interim analysis with requirement for patients to have been enrolled on or before July 11, 2019. The November 6, 2020, was an unplanned interim analysis, generated in support of the European Medicines Agency submission, with a requirement for patients to have been enrolled on or before May 22, 2020.

Source: ARROW Clinical Study Review,⁵ European Medicines Agency Pralsetinib Public Assessment Report.⁶

Table 10: Medical History

NR = not reported.

Note: The November 18, 2019, data cut-off was an unplanned interim analysis with requirement for patients to have been enrolled on or before July 11, 2019. The November 6, 2020, data cut-off was an unplanned interim analysis, generated in support of the European Medicines Agency submission, with a requirement for patients to have been enrolled on or before May 22, 2020.

Source: ARROW Clinical Study Review,⁵ European Medicines Agency Pralsetinib Public Assessment Report.⁶

The medical history for the efficacy population at the November 18, 2019, data cut-off is summarized in Table 10. The most common conditions in patients’ medical histories were hypertension (34.1%), cough (26.5%), dyspnea (25.0%), and fatigue (25.8%). Medical history was not available for the November 6, 2020, data cut-off.

Interventions

Patients received pralsetinib orally as 100 mg capsules at a dosage of 400 mg once daily. As the ARROW trial was a single-arm phase I and II trial, there was no randomization and all patients enrolled into the phase II dose expansion were assigned the RP2D determined in phase I of the trial (400 mg once daily). Dose reductions by 100 mg increments, but to no lower than a 100 mg total dose, were permitted in the case of grade 3 or higher AEs. If reductions were required below this level, the patient was discontinued from treatment. Dose interruptions were permitted for 4 weeks; however, if the AE that led to discontinuation did not resolve to grade 2 or lower the patient was discontinued from treatment. Temporary interruptions up to 2 weeks were permitted for patients requiring surgery or other procedures.

Prohibited concomitant medications in the ARROW trial included strong inhibitors and inducers of CYP3A4, any investigational drug other than pralsetinib, any antineoplastic agent other than pralsetinib, and neutrophil growth factor, unless the patient experienced dose-limiting toxicity (grade 4) associated with neutropenia.

Use of concomitant medications in the ARROW trial at the November 18, 2019, data cut-off is summarized in Table 11. The most commonly reported concomitant medications were other analgesics and antipyretics |||||||||, opioids (|||||||||), drugs for constipation (|||||||||), and drugs for peptic ulcer and gastro-esophageal reflux disease (|||||||||). Concomitant medication information was not available for the November 6, 2020, data cut-off.

Outcomes

A list of efficacy end points identified in the CADTH review protocol that were assessed in the clinical trials included in this review is provided in Table 12. These end points are further summarized below. A detailed discussion and a critical appraisal of the outcome measures are provided in Appendix 4.

Table 11: Concomitant Medications

NR = not reported.

Note: The November 18, 2019, data cut-off was an unplanned interim analysis with requirement for patients to have been enrolled on or before July 11, 2019. The November 6, 2020, data cut-off was an unplanned interim analysis, generated in support of the European Medicines Agency submission, with a requirement for patients to have been enrolled on or before May 22, 2020.

Source: ARROW Clinical Study Review,⁵ European Medicines Agency Pralsetinib Public Assessment Report.⁶

Table 12: Summary of Outcomes of Interest Identified in the CADTH Review Protocol

Overall survival was defined as the time from the first dose of pralsetinib to the date of death due to any cause. Patients who were still alive or lost to follow-up at the time of data cut-off were censored according to the rules listed in Table 13 at the last known time alive.

The EORTC QLQ-C30, a 30-item, patient-reported, cancer-specific questionnaire using 4- and 7-point Likert scales was used to measure HRQoL. All scales and single-item measures are scored from 0 to 100. Most questions have 4 response options (“not at all,” “a little,” “quite a bit,” and “very much”), with scores on these items ranging from 1 to 4. For the 2 items that form the global HRQoL scale, the response format is a 7-point Likert-type scale with anchors at 1 (“very poor”) and 7 (“excellent”). Validity, responsiveness, and reliability have been demonstrated in NSCLC patients. An MID of 5 to 10 points corresponded with a small clinical change, 10 to 20 points with a moderate clinical change, and greater than 20 points with a large clinical change.

Progression-free survival was defined as the time from the first dose of pralsetinib to the date of first documented progressive disease or death due to any cause. Patients without documented progression at the time of data cut-off were censored according to the rules listed in Table 13.

The ORR was defined, according to RECIST 1.1 or a response assessment in neuro-oncology if appropriate, as the proportion of patients with a complete response (CR) or partial response (PR) for at least 2 assessments separated by at least 28 days with no documented progressive disease in between. The ORR was analyzed according to a BICR. The ORR in brain or CNS lesions were also assessed according to RECIST v1.1.

The DOR was analyzed within the population of patients with confirmed CR or PR. It was defined as the time from first documented CR or PR to the date of first documented progressive disease or death due to any cause. Patients still in response at the time of data cut-off were censored according to the rule listed in Table 13.

Table 13: Censoring Rules in ARROW

EOT = end of trial.

Statistical Analysis

Twelve protocol amendments were reported for the ARROW trial. Of these 12, the following were particularly noteworthy and associated with a consequential impact. Protocol amendment 4.1, made July 25, 2018, removed eligibility for patients with an ECOG PS of 2 to enrol in the ARROW trial. Protocol amendment 9, made July 3, 2019, removed the need for patients enrolled in the treatment-naive group to be deemed unsuitable for SOC chemotherapy.

Sample-size calculations were conducted separately for treatment-experienced and treatment-naive patient groups in the ARROW study. For treatment-experienced patients with NSCLC, approximately 80 patients would provide greater than 95% power at the 2-sided significance level of 0.05, for which the null hypothesis was an ORR of 0.23 compared to the alternative hypothesis of an ORR of 0.5. For treatment-naive patients, approximately 170 patients would provide greater than 90% power to test the null hypothesis of an ORR of 0.48 compared to the alternative hypothesis of an ORR of 0.61. No interim analyses were planned for the ARROW trial and no adjustments for multiplicity applied to multiple end points or multiple data cut-offs were analyzed.

Patients enrolled in phase I and treated at the RP2D were pooled together with the appropriate phase II patient groups for analyses. A description of the statistical analysis models used in the ARROW trial is provided in Table 14.

Table 14: Statistical Analysis of Efficacy End Points

NA = not applicable.

Analysis Populations

Efficacy population from the November 18, 2019, data cut-off (N = 132) included all patients who received a dose of pralsetinib on or before July 11, 2019. The efficacy population from the November 6, 2020, data cut-off (N = 233) included all patients who received a dose of pralsetinib on or before May 22, 2020.

The safety population included all patients who had received 1 or more doses of pralsetinib 400 mg once daily at the time of data cut off. At the November 18, 2019, data cut-off, the safety population included 179 patients; at the November 6, 2020, data cut-off, the population had grown to 281 patients.

The responder analysis set included all patients with a confirmed response within the efficacy population. The responder analysis set was used to determine a DOR. At the November 18, 2019, data cut-off set this included 75 patients; at the November 6, 2020, data cut-off, the set had grown to 150 patients.

The response-evaluable population was used for a sensitivity analysis conducted on a subset of patients who did not have major protocol violations (including incomplete baseline imaging), evidence of RET mutation, or a known primary driver other than RET. At the November 18, 2019, data cut-off, the response-evaluable population was 163 patients; at the November 6, 2020, data cut-off the population had grown to 216 patients. As this was not a predefined subgroup analysis, the results are not shown in this report.

Results

Patient Disposition

A summary of patient disposition in the ARROW trial at the November 18, 2019, and November 6, 2020, data cut-offs is presented in Table 15. At the November 18, 2019, data cut-off, 404 patients with all tumour types had been enrolled and received treatment with pralsetinib. Of these, 179 patients had RET fusion–positive NSCLC and received treatment at 400 mg once daily, representing the safety analysis population. From the safety population, 3 patients were originally enrolled in the phase I portion of the trial. The efficacy population included 132 of these patients who had received their first dose on or before July 11, 2019. At the time of data cut-off, 49.2% patients had discontinued treatment, most commonly citing disease progression; however, 18.2% of patients discontinued due to AEs. Discontinuation of the study occurred in 44.7% of patients, with death and disease progression being the most common reasons, and withdrawal of consent the next most common reason (7.6%).

At the November 6, 2020, data cut-off, 587 patients with all tumour types had been screened, and 521 enrolled and received treatment. Of these, 281 patients had RET fusion–positive NSCLC and received treatment at 400 mg once daily, representing the safety population. The efficacy population included 233 of these patients who had received treatment on or before May 22, 2020. At the time of data cut-off, 52.8% of patients had discontinued treatment, mostly due to disease progression; however, 14.6% discontinued due to AEs and 43.8% had discontinued from the study, with death and disease progression being the most common reasons, and withdrawal of consent the next most common reason (6.9%).

Exposure to Study Treatments

Treatment exposure in the ARROW trial safety population is summarized in Table 16. At the November 18, 2019, data cut-off, the median duration of exposure was ||||||||||||||||||) weeks with a median relative dose intensity, defined as actual dose divided by initial assigned dose times 100, of ||||||||||||||||||. At the November 6, 2020, data cut-off, the median duration of exposure was 7.89 (range = 0.3 to 28.4) months with a median relative dose intensity of 92.1% (range = 27% to 100%).

Table 15: Patient Disposition

NSCLC = non–small cell lung cancer; q.d. = once daily.

^aThe safety analysis includes all patients who received pralsetinib at a dosage of 400 mg once daily. Safety analysis conducted for all pralsetinib doses is not shown in this report.

Source: ARROW Clinical Study Review,⁵ European Medicines Agency Pralsetinib Public Assessment Report.⁶

Table 16: Treatment Exposure in ARROW Safety Population

Note: The November 18, 2019, data cut-off was an unplanned interim analysis with requirement for patients to have been enrolled on or before July 11, 2019. The November 6, 2020, data cut-off was an unplanned interim analysis, generated in support of the European Medicines Agency submission, with a requirement for patients to have been enrolled on or before May 22, 2020.

^aDuration of exposure measured in months.

Source: ARROW Clinical Study Review,⁵ European Medicines Agency Pralsetinib Public Assessment Report.⁶

Efficacy

Only those efficacy outcomes and analyses of subgroups identified in the review protocol are reported here. Appendix 3 provides detailed efficacy data.

Overall Survival

Overall survival in the ARROW trial at the November 18, 2019, and November 6, 2020, data cut-offs is summarized in Table 17. At the November 18, 2019, data cut-off the median OS follow-up time was 10.5 (95% CI, 9.7 to 13.1) months. In the efficacy population, 19.7% of patients had died and median OS had not been reached. Kaplan-Meier estimates of OS at |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||| respectively. Results were similar when analysis was conducted separately for treatment-experienced and treatment-naive patients.

At the November 6, 2020, data cut-off the median, the OS follow-up time was 17.1 (95% CI, 13.7 to 19.6) months. In the efficacy population, 24.5% of patients had died and the median OS had not been reached. Kaplan-Meier estimates of OS at 3, 6, 12, and 24 months were 96.0% (95% CI, 93.5 to 98.6), 87.6% (95% CI, 83.2 to 92.0), 76.0% (95% CI, 69.9 to 82.0) and 66.0% (95% CI, 57.9 to 74.1), respectively. Results were similar when treatment-experienced and treatment-naive patients were analyzed separately. A separate analysis for OS was conducted on the safety population of all patients who had received a dose of pralsetinib up to the data cut-off. In this population of 281 patients, the median follow-up time was 13.2 month (95% CI, 12.1 to 15.5) and the median OS was not reached.

Health-Related Quality of Life

The HRQoL as measured by global health status on the EORTC QLQ-C30 at the November 18, 2019, data cut-off is summarized in Table 18. Of the 150 patients included in the HRQoL analysis set, baseline results were available for |||||||||. The mean baseline value was ||||||||||||||||||. ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||. The HRQoL results from the November 6, 2020, data cut-off were not available.

Table 17: Overall Survival Analysis in the ARROW Trial

CI = confidence interval; NA = not applicable; NR = not reached.

Note: The November 18, 2019, data cut-off was an unplanned interim analysis with requirement for patients to have been enrolled on or before July 11, 2019. The November 6, 2020, data cut-off was an unplanned interim analysis, generated in support of the European Medicines Agency submission, with a requirement for patients to have been enrolled on or before May 22, 2020.

^a95% CI based on the Greenwood formula.

Source: ARROW Clinical Study Review,⁵ European Medicines Agency Pralsetinib Public Assessment Report.⁶

Table 18: EORTC QLQ-C30 Global Health Status in the ARROW Trial

HRQoL = health-related quality of life; SD = standard deviation

Note: The November 18, 2019, was an unplanned interim analysis with requirement for patients to have been enrolled on or before July 11, 2019. The November 6, 2020, data cut-off was an unplanned interim analysis, generated in support of the European Medicines Agency submission, with a requirement for patients to have been enrolled on or before May 22, 2020.

Source: ARROW Clinical Study Review.⁵

Progression-Free Survival

Progression-free survival in the ARROW trial at the November 18, 2019, and November 6, 2020, data cut-offs is summarized in Table 19. At the November 18, 2019, data cut-off, in the efficacy population, ||||||||| of patients had experienced an event and median PFS was |||||||||||||||||| months. Kaplan-Meier estimates of PFS at |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||| respectively. Results were similar when analyses were conducted separately for treatment-experienced and treatment-naive patients.

At the November 6, 2020, data cut-off, 43.8% of patients in the efficacy population had experienced an event and the median PFS was 16.4 months (95% CI, 11.0 to 24.1). Kaplan-Meier estimates of PFS at 3, 6, 12, and 24 months were 90.7% (95% CI, 86.9 to 94.5), 75.1% (95% CI, 69.3 to 80.9), 56.0% (95% CI, 48.9 to 63.1), and 42.1% (95% CI, 33.2 to 51.0), respectively. Results were similar when analyses were conducted separately for treatment-experienced and treatment-naive patients. A separate analysis for PFS was conducted on all 281 patients in the safety population who had received a dose of pralsetinib up to the data cut-off. In this population, 37.7% of patients had experienced an event and the median PFS was 16.4 (95% CI, 11.0 to 24.1) months.

Overall Response Rate

Table 20 summarizes the ORR results at the November 18, 2019, and November 6, 2020, data cut-offs from the ARROW trial. At the November 18, 2019, data cut-off, in the efficacy population, the ORR was ||||||||||||||| including ||||||||| of patients with a CR. The ORR was |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||) compared with patients |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||.

At the November 6, 2020, data cut-off, the ORR in the efficacy population was 64.4% (95% CI, 52.7 to 68.4), including 4.7% of patients with a CR. The ORR was higher in patients with no prior systemic therapy (72.0%; 95% CI, 60.4 to 81.8) compared with patients with prior systemic therapy (60.8%; 95% CI, 52.7 to 68.4). These results successfully reject the pre-specified null hypothesis for treatment effect at both data cut-offs. A summary of the maximum change from baseline in target lesions for patients who had not received prior systemic therapy is shown in Figure 3, and in Figure 4 for patients who had received prior platinum therapy.

Table 21 summarizes the intracranial ORR results at the November 18, 2019, and November 6, 2020, data cut-offs from the ARROW trial. For the 9 response-evaluable patients with measurable intracranial metastasis at baseline at the November 18, 2019, data cut-off, the ORR was ||||||||||||||||||||||||||||||||||||, including ||||||||| of these patients achieving a CR. Among the ||||||||| patients who did not have a history of CNS metastasis at study entry, ||||||||| developed CNS metastasis while on pralsetinib 400 mg once daily at the November 18, 2019, data cut-off. At the November 6, 2020, data cut-off, for the 10 patients with measurable intracranial metastasis at baseline, the ORR was 70.0% (95% CI, 34.8 to 93.3), including 30.0% of patients achieving a CR.

Table 20: Overall Response Rate Analysis in the ARROW Trial

CI = confidence interval; NE = not evaluable.

Note: The November 18, 2019, data cut-off was an unplanned interim analysis with requirement for patients to have been enrolled on or before July 11, 2019. The November 6, 2020, data cut-off was an unplanned interim analysis, generated in support of the European Medicines Agency submission, with a requirement for patients to have been enrolled on or before May 22, 2020.

^a95% CI based on exact binomial distribution using Clopper-Pearson method.

Source: ARROW Clinical Study Review,⁵ European Medicines Agency Pralsetinib Public Assessment Report.⁶

Table 21: Intracranial Overall Response Rate Analysis in the ARROW Trial

CI = confidence interval; NA = not applicable; NE = not evaluable.

Note: The November 18, 2019, data cut-off was an unplanned interim analysis with requirement for patients to have been enrolled on or before July 11, 2019. The November 6, 2020, data cut-off was an unplanned interim analysis, generated in support of the European Medicines Agency submission, with a requirement for patients to have been enrolled on or before May 22, 2020.

^a95% CI based on exact binomial distribution using the Clopper-Pearson method.

Source: ARROW Clinical Study Review,⁵ European Medicines Agency Pralsetinib Public Assessment Report.⁶

Figure 3: Maximum Change From Baseline in Patients Who Had Not Received Prior Systemic Therapy, November 6, 2020, Data Cut-Off

Source: European Medicines Agency Pralsetinib Public Assessment Report.⁶

Figure 4: Maximum Change From Baseline in Patients Who Received Prior Platinum Therapy, November 6, 2020, Data Cut-off

Source: European Medicines Agency Pralsetinib Public Assessment Report.⁶

The ORR was also reported in a post hoc subgroup of only patients with measurable disease at baseline. A further subgroup analysis of these patients was conducted to report the ORR in treatment-naive patients before the eligibility criteria were revised and after revision. These results, along with the subgroups for prior platinum therapy and prior nonplatinum therapy are presented in Appendix 3 along with DOR results. The ORR and DOR results in all patient subgroups are similar to those of the primary analysis.

The ORR results by pre-specified subgroups of ECOG PS and history of CNS metastasis from the November 6, 2020, data cut-off are summarized in Table 22. The analysis was stratified by prior platinum experience or no prior systemic therapy. In the group with prior platinum experience, the ORR was 67.6% (95% CI, 50.2 to 82.0) and 56.4% (95% CI, 45.8 to 66.6) for an ECOG PS of 0 and 1, respectively. In patients with no prior systemic therapy the ORR was 83.9% (95% CI, 66.3 to 94.5) and 62.8% (95% CI, 46.7 to 77.0) for an ECOG PS of 0 and 1, respectively. When considering the history of CNS metastasis, patients with prior platinum experience had an ORR of 55.6% (95% CI, 41.4 to 69.1) with a history of CNS metastasis and 61.0% (95% CI, 49.6 to 71.6) without a history of CNS metastasis. In patients with no prior systemic therapy, these values were 68.0% (95% CI, 46.5 to 85.1) and 74.0% (95% CI, 59.7 to 85.4), respectively.

Duration of Response

Table 23 summarizes the DOR results at the November 18, 2019, and November 6, 2020, data cut-offs for the ARROW trial. At the November 18, 2019, data cut-off, in the responder analysis set of 75 patients, the median DOR |||||||||||||||||||||||||||||||||||| with ||||||||| of patients having had experienced an event. The Kaplan-Meier estimate of DOR at 12 months was |||||||||||||||||||||||||||. At the November 6, 2020, data cut-off, the median DOR in the responder analysis set of 150 patients was 22.3 (95% CI, 14.7 to NR) months, with 32.7% of patients having had experienced an event. The Kaplan-Meier estimate of DOR at 12 months was 63.8% (95% CI, 54.5 to 73.0). The observed DOR included 32.0% of patients who had a response longer than or equal to 12 months. The Kaplan-Meier curve for DOR from the November 6, 2020, data cut-off is shown in Figure 5.

Table 22: ORR Subgroup Analysis by ECOG PS and CNS Metastasis at Baseline

CI = confidence interval; CNS = central nervous system; ECOG PS = Eastern Cooperative Oncology Group Performance Status; ORR = overall response rate.

Note: The November 6, 2020, data cut-off was an unplanned interim analysis, generated in support of the European Medicines Agency submission, with a requirement for patients to have been enrolled on or before May 22, 2020.

Source: European Medicines Agency Pralsetinib Public Assessment Report.⁶

Table 23: Duration of Response Analysis (Responder Analysis Set)

CI = confidence interval; NA = not applicable; NR = not reached.

Note: The November 18, 2019, data cut-off was an unplanned interim analysis with requirement for patients to have been enrolled on or before July 11, 2019. The November 6, 2020, data cut-off was an unplanned interim analysis, generated in support of the European Medicines Agency submission, with a requirement for patients to have been enrolled on or before May 22, 2020.

^a95% CI based on the Greenwood formula.

Source: ARROW Clinical Study Review,⁵ European Medicines Agency Pralsetinib Public Assessment Report.⁶

Figure 5: Duration of Response (Responder Analysis Set; November 6, 2020, Data Cut-Off)

NSCLC = non–small cell lung cancer.

Source: Sponsor submission.⁴

Harms

Only those harms identified in the review protocol are reported. Table 24 provides detailed harms data.

Table 24: Summary of Harms

ALT = alanine transaminase; AST = aspartate transaminase.

^aFrequency greater than 15% at either data cut-off.

^bFrequency greater than 2% at either data cut-off.

^cPatients who discontinued treatment.

^dFrequency greater than 1%.

Source: ARROW Clinical Study Review,⁵ European Medicines Agency Pralsetinib Public Assessment Report.⁶

Adverse Events

At the November 18, 2019, data cut-off, ||||||||| of a total of ||||||||| patients in the safety analysis set experienced at least 1 AE. The most common were increased AST (|||||||||), constipation (|||||||||), anemia (|||||||||), and increased ALT |||||||||. At the November 6, 2020, data cut-off, 99.3% of the 281 patients in the safety analysis set experienced at least 1 AE. The most common were anemia (45.9%), increased AST (44.8%), constipation (42.0%), hypertension (34.2%), and increased ALT (32.7%).

Serious Adverse Events

At the November 18, 2019, data cut-off, ||||||||| of patients experienced at least 1 SAE. The most common were pneumonia (||||||||| and disease progression (|||||||||). At the November 6, 2020, data cut-off, 59.1% of patients had experienced at least 1 SAE. The most common were consistent with the earlier data cut-off, with pneumonia reported in 11.7% of patients and disease progression reported in 7.5%.

Withdrawals Due to Adverse Events

At the November 18, 2019, data cut-off, ||||||||| of patients had discontinued treatment due to an AE. The most common AEs to result in treatment discontinuation were disease progression (|||||||||), pneumonia (|||||||||), and pneumonitis (|||||||||). At the November 6, 2020, data cut-off, 19.6% of patients had discontinued treatment due to an AE. The most common AEs to result in treatment discontinuation were pneumonia (3.6%), disease progression (2.5%), and pneumonitis (2.5%).

Mortality

At the November 18, 2019, data cut-off, ||||||||| of patients had died as a result of an AE. The most common AEs to result in death were disease progression (||||||||| and pneumonia (|||||||||). At the November 6, 2020, data cut-off, 12.5% of patients had died as a result of an AE. The most common AEs to result in death were disease progression (5.0%) and pneumonia (1.4%).

Notable Harms

Detailed information on notable harms was available for pneumonitis and hypertension. At the November 18, 2019, data cut-off, pneumonitis was reported as a grade 3, 4, or 5 AE by ||||||||| of patients and an SAE by ||||||||| of patients, resulting in a dose reduction in ||||||||| of patients, a dose interruption in of patients, and treatment discontinuation by ||||||||| of patients. There were |||||||||||||||||||||||||||||||||||| at the November 18, 2019, data cut-off. At the November 6, 2020, data cut-off, pneumonitis was reported as a grade 3, 4, or 5 AE by 2.1% of patients and an SAE by 4.6% of patients, resulting in a dose reduction in 6.4% if patients, a dose interruption in 9.6% of patients, and treatment discontinuation by 2.5% of patients. No deaths were attributed to pneumonitis at the November 6, 2020, data cut-off.

At the November 18, 2019, data cut-off, hypertension was reported as a grade 3, 4, or 5 AE by ||||||||| of patients and an SAE by ||||||||| of patients, resulting in a dose reduction in ||||||||| of patients, a dose interruption in ||||||||| of patients, and treatment discontinuation by ||||||||| of patients. There were ||||||||||||||||||||||||||||||||||||||||||||| at the November 18, 2019, data cut-off. At the November 6, 2020, data cut-off, hypertension was reported as a grade 3, 4, or 5 AE by 16.0% of patients and an SAE by 1.4% of patients, resulting in a dose reduction in 4.3% if patients, a dose interruption in 8.5% of patients, and treatment discontinuation by 0.4% of patients. No deaths were attributed to hypertension at the November 6, 2020, data cut-off.

Summary data for hemorrhage and hepatoxicity were not reported.

Critical Appraisal

Internal Validity

The ARROW study is an ongoing phase I and II, open-label, single-arm assessment of the clinical efficacy and safety of pralsetinib in patients with RET fusion–positive locally advanced or metastatic thyroid cancer, NSCLC, or other solid tumours. Given that ARROW is a single-arm trial, the most important limitation is the lack of a comparator arm. This design increases the risk of bias in estimating treatment effects due to the potential for confounding related to unidentified prognostic factors and treatment-effect modifiers that could affect the activity of the study drug. CADTH, as well as the clinical expert consulted for this review, acknowledge that patients with RET fusion–positive NSCLC are rare, representing 1% to 2% of all NSCLC cases, increasing the difficulty of conducting randomized controlled trials. However, the sponsor has initiated a phase III randomized controlled trial (RCT) assessing first-line pralsetinib in RET fusion–positive metastatic NSCLC against physician’s-choice platinum-based chemotherapy, with results estimated to be available for the primary end point in December 2024.

The primary objective investigated in the phase II portion of the ARROW study was the ORR as measured by RECIST 1.1. The trial used standardized imaging protocols across study sites and centralized reading to ensure quality and consistency across the trial for imaging end points. The FDA considers the ORR a surrogate measurement when assessing treatment response in advanced or metastatic NSCLC patients, and it may not correlate well with survival, unless the effect size of the ORR is large, and the responses are durable.²⁸ The ARROW trial reported a null hypothesis of an ORR of 0.23 and an alternative hypothesis of an ORR of 0.5 in treatment-experienced patients, and a null hypothesis of 0.48 and alternative hypothesis of 0.61 in treatment-naive patients, both of which were confirmed to be reasonable and clinically significant by the clinical expert consulted for this review. Although the trial’s sample size was not large enough to satisfy the reported power calculations at the time of both data cut-offs, the observed results exceeded both alternative hypotheses and the lack of power is therefore not of concern.

While there was a lack of formal statistical hypothesis testing, there was also no pre-specified interim analyses planned in the statistical analysis plan for the ARROW trial, increasing the potential for bias and type I error with successive ad hoc data cut analyses. Furthermore, extensive protocol amendments affecting the conduct of the trial after patients had first been randomized may have biased the results and increased uncertainty by increasing the heterogeneity of the patient population. Patients recruited to the treatment-naive group were initially required to be deemed unsuitable for SOC chemotherapy, but this was later amended to allow all treatment-naive patients. This may have biased the results against pralsetinib if the patients recruited before this amendment had a worse prognosis compared to the average first-line patient. Important protocol deviations further increased the uncertainty, given that 16 patients (6.9% of the efficacy population) at the November 6, 2020, data cut-off did not have measurable disease at baseline and 1 had inconclusive evidence of RET fusion. This was due to eligibility for the trial being based on investigator assessments that were ultimately not confirmed by a BICR. The sponsor did provide sensitivity analyses removing these patients from the analysis and reported results similar to those of the primary analysis; however, the ad hoc removal of patients due to protocol deviations means the results should be interpreted with caution.

Time-to-event outcomes of PFS and OS were immature and, given the lack of comparator arm, uncertainty remains regarding the long-term effects of pralsetinib on PFS and OS. The HRQoL outcome was evaluated using the EORTC QLQ-C30, which has been validated in NSCLC patients and is considered appropriate for advanced or metastatic forms of the disease. Although this was an exploratory outcome investigated in the ARROW trial, the clinician expert and clinician groups noted that findings for quality of life are clinically significant because patients with advanced disease usually experience greater symptom burdens due to disease progression. Although the HRQoL results are directionally positive, uncertainty remains regarding these findings because the number of patients who completed the questionnaires at baseline represented a smaller subset of the overall population due to the use of a protocol amendment in addition to attrition in later weeks of the trial. There is potential for selection bias over time given that long-term survivors in the trials tend to be healthier patients. In the absence of a comparator arm and an open-label design, which introduces reporting bias, the impact of pralsetinib on patient-reported outcomes in relation to other therapies is unknown.

External Validity

The outcomes assessed in the ARROW trial (ORR, DOR, OS, PFS, and HRQoL) are standard in oncology and considered clinically meaningful. According to the clinical expert consulted by CADTH, the demographic and disease characteristics of the ARROW population were reflective of the Canadian population of patients with RET fusion–positive NSCLC. Patients are more likely to have never smoked, be younger, have adenocarcinoma histology, and present with stage 4 disease. Few patients with stage III disease were included in the ARROW trial; however, the clinical expert indicated that if a patient with stage III disease was ineligible for curative therapy they would be treated and expected to respond just as well as a patient with stage IV disease. Patients with an ECOG PS of 2 were originally included in the trial population but later removed from the inclusion criteria. There are therefore few patients in the trial with an ECOG PS of 2, calling into question the generalizability of the study results to patients with an ECOG PS of 2 or greater. The clinical expert indicated that pralsetinib should be offered to patients with an ECOG PS of 2 or 3, given the likelihood of improving symptom burden and ECOG PS.

Pralsetinib dosing and drug administration interval in the trial align with Health Canada’s indication and are therefore generalizable to the Canadian setting. All outcomes evaluation in the trial and considered in this review (ORR, DOR, HRQoL, OS, and PFS) were clinically relevant, important to patients, and used in clinical practice. The duration of follow-up was sufficient for the assessment of the primary outcome ORR; however, longer-term outcomes of PFS and OS are difficult to interpret given the immaturity of the data. Although there was no predefined maximum duration of treatment in the ARROW trial, the most recent data cut-off had a median survival follow-up of 17.1 months. As such, the generalizability of long-term safety and efficacy conclusions is uncertain beyond what is seen at the November 6, 2020, data cut-off. The clinical expert consulted for this review did not anticipate that extending treatment beyond the follow-up times in the trial would present any issues of concern, noting that treatment would be discontinued in the event of clinical (not radiographic) progression, unacceptable toxicity, or patient preference. (Refer to Table 4).

Indirect Evidence

Objectives and Methods for the Summary of Indirect Evidence

This section provides a summary and appraisal of the indirect evidence submitted by the sponsor comparing pralsetinib to other treatments used in RET fusion–positive locally advanced or metastatic NSCLC.

A focused literature search for network meta-analyses dealing with NSCLC was run in MEDLINE All (1946–) on March 31, 2022. No search limits were applied.

Description of Indirect Comparison

The sponsor noted that, due to the lack of available evidence for comparators of interest in the population with RET fusion–positive NSCLC, an ITC comparing pralsetinib evidence in patients with RET fusion–positive NSCLC to evidence for comparators of interest in wild-type NSCLC was conducted. Given that the pivotal trial for pralsetinib, ARROW, is a single-arm trial, unanchored comparisons were required. For comparators for which IPD were available, propensity-score analyses were conducted to adjust for differences in population characteristics between the trials. For comparators for which IPD were not available, naive comparisons, with no adjustments for differences in population characteristics, were made.

One published ITC, Popat et al. (2022),¹⁵ compared first-line patients from the ARROW trial receiving pralsetinib against relevant comparators through the use of synthetic control arms from real-world population sets. The authors used IPW to adjust for differences in patient characteristics.

Methods of Sponsor-Submitted Indirect Treatment Comparison

Objectives

In the absence of direct comparative evidence from trials, the aim of each analysis was to compare the efficacy (OS and PFS) of pralsetinib in patients with RET fusion–positive locally advanced or metastatic NSCLC versus patients with wild-type NSCLC receiving comparators of interest.

Study Selection Methods

The index trial was based on IPD from the November 6, 2020, data cut-off of the phase I and II ARROW trial; specifically, the safety analysis set (N = 281) that included all patients who had received a dose of pralsetinib up to the data cut-off. To identify evidence for relevant comparators and identify evidence from RCTs, a systematic literature review (SLR) was conducted. The selection criteria for inclusion in the ITC are summarized in Table 25.

Table 25: Study Selection Criteria and Methods for Indirect Treatment Comparisons

CNS = central nervous system; NSCLC = non–small cell lung cancer; SLR = systematic literature review; RCT = randomized control trial.

Source: Sponsor-submitted indirect treatment comparison report.²⁹

The SLR identified 131 unique studies; the resulting list was refined to determine inclusion in the comparative-analysis ITC. For example, 112 of the 131 studies reported data for either first-line platinum-based chemotherapy plus pemetrexed or paclitaxel (68 studies) or second-line docetaxel (44 studies). The sponsor determined that, where possible, studies with IPD available would be prioritized and included in the analysis. As such, the IMpower132 and OAK trials were selected for these comparators, respectively. If no IPD were available for a comparator, studies were chosen using a sequential decision tree based on whether the population aligned with the ARROW trial in terms of histology and PD-L1 status (RET status was assumed not to be predictive); whether pooled analysis were available, and which study had the largest number of patients enrolled. A summary of the included studies is provided in Table 26.

ITC Analysis Methods

In the absences of direct trial evidence for relevant comparators or RCTs to form a connected network, 2 methods for generating comparative efficacy estimates for pralsetinib versus comparators of interest were used. Only the PFS and OS outcomes were analyzed. For comparisons for which IPD were available for the comparator trials (the IMpower132 trial for platinum-based chemotherapy plus pemetrexed in the first-line setting and the OAK trial for docetaxel in the second-line setting) a propensity-score method was used. For comparisons for which no IPD were available for the comparator trials, naive comparisons were generated for first-line pembrolizumab monotherapy, pembrolizumab in combinations with platinum-based chemotherapy plus pemetrexed, and second-line nivolumab, and pemetrexed plus carboplatin.

Propensity-Score Method

The propensity-scoring approach is based on methodology outlined in National Institute for Health and Care Excellence Decision Support Unit guidance (Technical Support Document 17).³⁰ Using IPW, the approach attempts to control for the sampling bias whereby the probability of being assigned a treatment in 1 trial versus another is not random, but a function of observable covariates. The method adjusts the population of the comparator trial based on these covariates. Appropriate covariates must be chosen as inputs for the statistical model. The sponsor consulted expert clinical input and selected the following 7 prognostic factors (from the characteristics that were reported in both comparator studies) to be included in the analysis:

• age (continuous)

• gender (male or female)

• presence of CNS metastasis (yes or no)

• ECOG PS (0 or 1)

• race (Asian or non-Asian)

• histology (adenocarcinoma or other)

• smoking status (never or current/former).

Weights were obtained by fitting a logistic regression model, with treatment assignment being a function of these 7 covariates. A Cox regression model was then fitted to the pooled IPD for OS and PFS (although there was no mention of whether the proportional hazards assumption was violated) using the weights obtained from the propensity-scoring analyses to estimate the weighted treatment effect between pralsetinib and the comparator of interest. The sponsor provided absolute mean differences of covariates pre- and post-weighting as well as the distribution of propensity-score weights as a diagnostic to check the weighting.

In contrast to the naive comparisons, for the propensity-scoring analysis PFS was based on investigator assessment, not a BICR. This was because the most recent data cut-off for the IMpower132 study only reported investigator-assessed PFS and, to be consistent, the sponsor also used investigator-assessed PFS for the ARROW and OAK trials. If patients were missing covariate data in any trial (ARROW, IMpower132, or OAK), they were removed from the analysis (9 were removed for the IMpower132 comparison and 15 removed for the OAK comparison). Additional patients were removed from analysis before propensity-score weighting, including patients in the ARROW trial with an ECOG PS of 2 (1 patient in the ARROW trial), those with ALK- or EGFR-driver mutations in the IMpower132 study (5 patients) as they would not have been eligible to enrol in the ARROW trial, and patients not treated as intended in IMpower132 (12 patients).

Naive Comparisons

In the absence of IPD for comparators of interest, naive comparisons were conducted. The was no adjustment for differences between trial populations and no selective exclusion of patients from the analyses. Data for OS and PFS were recreated from published Kaplan-Meier curves using the Guyot method,³¹ and virtual IPD were created for the comparator arms. A Cox regression model was fitted to the IPD to estimate naive HRs. General conclusions from the naive comparisons are commented on in the following section; however, given the uncertainty in conducting naive comparisons, detailed results are not provided in this report.

Results of the Sponsor-Submitted Indirect Treatment Comparison

Results of Propensity-Scoring Method

The population characteristics for the first-line comparison between pralsetinib and platinum-based chemotherapy plus pemetrexed before and after IPW adjustments conducted in the IMpower132 study are summarized in Table 27. Residual imbalances are evident in gender (||||||||||||||||||||||||||||||||||||||||||||||||||||||) and CNS metastasis (||||||||||||||||||||||||||||||||||||||||||||||||||||||).

The population characteristics for the second-line comparison between pralsetinib and docetaxel before and after IPW adjustments conducted on the OAK study are summarized in Table 28. There is a residual imbalance in ECOG PS of 1 (||||||||||||||||||||||||||||||||||||).

The adjusted and unadjusted HRs for OS results using IPW analysis for the IMpower132 and OAK studies compared with the ARROW trial are summarized in Table 29. All adjusted and unadjusted HRs favoured pralsetinib for OS. The adjusted HR for OS for the pralsetinib versus platinum-based chemotherapy plus pemetrexed comparison was |||||||||||||||||||||||||||. The adjusted HR for OS for the pralsetinib versus docetaxel comparison was |||||||||||||||||||||||||||.

The adjusted and unadjusted HRs for the PFS results using IPW analysis for the IMpower132 and OAK studies compared with ARROW are summarized in Table 30. All adjusted and unadjusted HR favoured pralsetinib for PFS. The adjusted PFS HR for the pralsetinib versus platinum-based chemotherapy plus pemetrexed comparison was |||||||||||||||||||||||||||. The adjusted PFS HR for the pralsetinib versus docetaxel comparison was |||||||||||||||||||||||||||.

Table 27: Patient Characteristics Before and After IPW With IMpower132 — First Line

CNS = central nervous system; ECOG PS = Eastern Cooperative Oncology Group Performance Status; IPW = inverse probability weighting.

Source: Sponsor-submitted indirect treatment comparison report.²⁹

Table 28: Patient Characteristics Before and After IPW With OAK — Second Line

CNS = central nervous system; ECOG PS = Eastern Cooperative Oncology Group Performance Status; IPW = inverse probability weighting.

Source: Sponsor-submitted indirect treatment comparison report.²⁹

Table 29: Adjusted and Unadjusted Overall Survival Results in the IPW Analysis

CI = confidence interval; IPW = inverse probability weighting; NE = not estimable.

Source: Sponsor-submitted indirect treatment comparison report.²⁹

Table 30: Adjusted and Unadjusted Progression-Free Survival Results in the IPW Analysis

CI = confidence interval; IPW = inverse probability weighting; NE = not estimable.

Source: Sponsor-submitted indirect treatment comparison report.²⁹

Results of the Naive Comparisons

The unadjusted naive comparisons for OS and PFS all favoured pralsetinib over first-line pembrolizumab monotherapy, pembrolizumab in combinations with platinum-based chemotherapy plus pemetrexed, and second-line nivolumab and pemetrexed plus carboplatin. Given the uncertainty in unadjusted naive comparisons across heterogeneous populations, detailed results are not shown in this report.

Critical Appraisal of Sponsor-Submitted Indirect Treatment Comparison

The sponsor submitted a single ITC report that included comparisons of interest for pralsetinib against pembrolizumab monotherapy, pembrolizumab plus platinum-based chemotherapy in combination with pemetrexed, and platinum-based chemotherapy in combination with pemetrexed in treatment-naive patients, as well as comparisons against docetaxel, nivolumab, and pemetrexed plus carboplatin in treatment-experienced patients. Comparator trials were identified via an SLR that selected publications to be included in the analysis. Given the lack of evidence for comparators of interest in patients with RET fusion–positive NSCLC specifically, the decision was made to compare with wild-type NSCLC, under the assumption that RET fusion is not a predictive or prognostic factor. The CADTH review team acknowledges that this assumption was required to provide an estimate of comparative efficacy; however, it is associated with key limitations. Patients who are RET fusion–positive tend to be younger, less likely to smoke, and have mostly nonsquamous histology. The sponsor provided evidence from Hess et al. (2021)¹⁴ suggesting that, before the introduction of RET inhibitors, there was no relationship between RET status and outcomes in an adjusted model. However, the clinical expert consulted for this review suggested that the presence of RET fusion is a positive predictor of the efficacy of RET-targeted therapy and a negative predictor for the effect of immunotherapy. A methodology to adjust for prognostic factors other than RET status was used; however, all differences in patient characteristics could not be accounted for. With regard to the naive comparisons specifically, no adjustments were made. Patients with a positive or negative RET fusion status are expected to respond differently to pralsetinib and it is unclear how similar the patient populations in the comparator studies are to those enrolled in the ARROW trial, despite the adjustments.

As the index trial for pralsetinib (ARROW) is a single-arm trial with no comparator arm, no connected network could be formed. The sponsor noted that, due to differences in population characteristics between the patients with RET fusion–positive NSCLC in ARROW, who were likely to be younger, have a nonsquamous histology, and no or minimal beforebacco use compared with the wild-type NSCLC patients in the comparator trials, a matched-adjusted indirect comparison would result in an ESS too small to draw informative conclusions from the results. The sponsor chose to selectively use studies for which they had access to IPD to conduct propensity-score weighting to adjust for between-trial differences in population characteristics. Selecting comparator studies for an arbitrary reason such as availability of IPD can introduce selection bias; however, the ability to adjust for important population characteristics adds informative power to the results in contrast to the naive comparisons also presented. Additionally, propensity-score weighting with IPD from both trials comes with an advantage over matched-adjusted indirect comparisons in that the comparator trial population is adjusted to better reflect the characteristics of the index trial, which is a better representation of the requested reimbursement population.

A key limitation with methods centred on adjusting for prognostic factors is that identification of all prognostic factors that affect treatment effects is unlikely. A rigorous SLR validated by clinical opinion is the preferred method of identifying prognostic factors. The sponsor consulted clinical opinion and selected 7 prognostic factors based on availability in the reported data, which is not a robust method of identification, although the clinical expert consulted by CADTH considered the 7 prognostic factors to be appropriate. After the comparator trials had been adjusted using the IPW methodology, the reductions in overall sample size were quite large (from 510 to 142 in the case of the OAK trial), likely due to the imbalance in baseline covariates. The distribution of propensity-scoring weights were generally concentrated around 0 and 1; however, a small number of inflated weights of 14 and 15 added to the uncertainty of the results.

As IPD were available for only 2 comparisons in the sponsor-submitted ITC, the remaining comparisons were unadjusted naive comparisons i.e., no adjustments for between-trial differences in population characteristics were made. This introduces a high risk of bias in the results, given that the important prognostic factors, identified by the sponsor as having an impact on treatment effects, remained heterogenous for the naive comparisons. Given these limitations, conclusions cannot be drawn based upon the naive comparisons, and conclusions drawn from the propensity-score weighted analysis are uncertain.

With these limitations in mind, all results were directionally consistent and in line with the clinical expert’s expectations that pralsetinib is likely a better option for patients than the comparators included in the ITC analysis.

Indirect Treatment Comparison Published by Popat et al. (2022)¹⁵

Methods

The ITC published by Popat et al. (2022)¹⁵ compared the efficacy of pralsetinib in patients with RET fusion–positive advanced NSCLC against SOC therapies. The authors analyzed the first-line cohort from the ARROW study (n = 116) in comparison to real-world data from synthetic control arms from the Flatiron Clinico-Genomic Database (CGDB) and Enhanced Datamart (EDM). Given the limited number of real-world data sources specific to patients with RET fusion–positive advanced NSCLC, only the CGDB population was RET fusion–positive (n = 10) and treated with a basket of best available therapy (the most common of which was pembrolizumab plus chemotherapy). Two Flatiron EMD populations, both with wild-type NSCLC, were considered. One population was treated with pembrolizumab monotherapy (n = 686) and the other was treated with pembrolizumab plus chemotherapy (n = 1,270). To account for the heterogeneity between the ARROW first-line population and the wild-type NCSLC patients in the Flatiron EMD, IPW adjustments were made to the EMD populations. Characteristics included in the IPW model were sex, ECOG PS, time from initial diagnosis, stage at diagnosis, age, smoking history, and race. Metastasis to the CNS was not included in the IPW adjustments due to differing methods of recording metastasis. The small sample size of the CGDB population precluded any adjustments based on the IPW methodology.

Population Characteristics

For the unadjusted comparison between the CGDB population and the ARROW first-line population, heterogeneity (standardized mean difference [SMD] > 0.6) in patient characteristics were noted in sex (20% male versus 47.4% male), ECOG PS (50% ECOG PS 0 and 30% ECOG PS 1 versus 30.2% ECOG PS 0 and 69.0% ECOG PS 1), and race (60% White versus 49.1% White). All characteristics had an SMD greater than 0.1.

Prior to IPW adjustments, there was heterogeneity between the ARROW first-line population compared to the pembrolizumab monotherapy and pembrolizumab plus chemotherapy populations, most notably with respect to smoking history (39.4% in ARROW versus 91.5% and 90.1%, respectively). Following IPW adjustments, there were residual differences in the pembrolizumab monotherapy comparison versus ARROW first-line population (SMD > 0.1), in age (48.3% younger than 65 years versus 59.6% younger than 65 years), smoking history (48.9% versus 39.4%), and race (56.7% versus 49.5%). Metastasis to the CNS was not adjusted for and therefore remained at an SMD of 0.241. The ESS following IPW adjustment was 115 from an original 686. For the comparison of pembrolizumab plus chemotherapy versus the ARROW first-line population, following IPW adjustments, all characteristics had an SMD of less than 0.1, except for CNS metastasis that was not included in the IPW model and therefore remained at an SMD of 0.383. The ESS following IPW adjustment was 217 from an original 1,270.

Results

Table 31 summarizes the HRs for PFS and OS in the comparisons of pralsetinib versus best alternative therapy in the CGDB RET fusion–positive population, pembrolizumab monotherapy in the EDM population, and pembrolizumab plus chemotherapy in the EDM population. The HRs for PFS favoured pralsetinib in all comparisons, with statistically significant findings against pembrolizumab monotherapy (HR = 0.47; 95% CI, 0.31 to 0.70) and against pembrolizumab plus chemotherapy (HR = 0.50; 85% CI, 0.36 to 0.70). The HRs for OS favoured pralsetinib in all comparisons, with statistically significant findings against pembrolizumab monotherapy (HR = 0.33; 95% CI, 0.18 to 0.61) and against pembrolizumab plus chemotherapy (HR = 0.36; 95% CI, 0.21 to 0.64).

Table 31: Comparative Efficacy Results Reported by Popat et al. (2022)

CGDB = Clinico-Genomic Database; CI = confidence interval; EDM = Enhanced Datamart; ESS = effective sample size; vs. = versus.

Source: Popat et al. (2022).¹⁵

Critical Appraisal

The ITC presented by Popat et al. (2022)¹⁵ utilizes synthetic control arms based on real-world data to estimate comparative efficacy for pralsetinib in comparison to a basket of best alternative therapy, pembrolizumab monotherapy, and pembrolizumab plus chemotherapy. Criteria for selection of real-world datasets were unclear in the publication and the presence of selection bias of available real-world populations is therefore uncertain. There are limitations to comparisons made between single-arm clinical trials and observational real-world data, 1 of which is seen in the heterogenous patient characteristics between the patients in the ARROW trial and real-world patient characteristics, particularly the lack of RET fusion–positive status in the pembrolizumab populations and increased smoking history, which was above 90% in the pembrolizumab populations and 39.4% in the ARROW population. The authors of the ITC conducted IPW adjustments to account for the differences in patient characteristics and achieved mostly balanced characteristics for the comparison against pembrolizumab plus chemotherapy (except for CNS metastasis, which was described by the clinical expert consulted for this review as an important prognostic factor in patients with RET fusion–positive NSCLC). The IPW methodology adjusts populations based on chosen prognostic factors. It is unclear in the publication how this list of prognostic factors was chosen, but the clinical expert considered the list to be appropriate. However, the inability to adjust based on CNS metastasis is a major limitation given its importance as a prognostic factor. Accordingly, IPW adjustments substantially reduced both pembrolizumab populations, producing an ESS much smaller than the original population in both cases (pembrolizumab monotherapy = 683 reduced to 115, and pembrolizumab plus chemotherapy = 1,270 reduced to 217), introducing uncertainty to the results.

The analysis conducted by Popat et al. showed statistically significant improvements in PFS and OS for patients receiving pralsetinib as compared to pembrolizumab and pembrolizumab plus chemotherapy. Pralsetinib showed a numerical benefit over a basket of best alternative therapy; however, despite the fact that this analysis was conducted in patients with RET fusion–positive NSCLC, the small sample size resulted in such wide CIs that it was difficult to draw any conclusions from the analysis.

To account for some of the limitations of the analysis, the authors conducted various sensitivity analyses. First, to assess the impact of missing ECOG PS inputs in the pembrolizumab populations, multiple imputation rules were tested with all analyses maintaining statistically significant OS and PFS results. Sensitivity analysis to include metastases in the IPW model despite differences in recording definitions maintained statistically significant OS and PFS results. Quantitative bias analysis was provided, with the authors suggesting that the results show that it is implausible for systematic differences in unmeasured prognostic factors to reverse the findings. Despite the limitations to the published ITC, the results are aligned with the expectation of the consulted clinical expert that there is likely to be a benefit for patients receiving pralsetinib over pembrolizumab monotherapy and pembrolizumab plus chemotherapy.

Other Relevant Evidence

In addition to the pivotal ARROW trial, a phase I and II study with no comparator arm, the AcceleRET-Lung study was considered as an other relevant study for this report. The CADTH review team identified AcceleRET-Lung as an ongoing phase III, randomized, open-label study that met systematic review inclusion criteria with the exception that no results are currently available. In addition, Health Canada issued a Notice of Compliance with conditions for pralsetinib and requested the final report, including datasets of an ongoing study, to verify and further characterize the clinical benefit of pralsetinib for the treatment of treatment-naive patients with RET fusion–positive metastatic NSCLC to be shared. For these reasons, the study is summarized in the following section. The study sponsor is F. Hoffmann-La Roche, Ltd.

AcceleRET-Lung

The AcceleRET-Lung trial is a phase III, randomized, multi-centre, open-label study comparing pralsetinib to a physician’s choice of platinum chemotherapy–based regimens based on SOC treatments for the first-line treatment of patients with RET fusion–positive metastatic NSCLC who have not previously received systemic anticancer therapy for metastatic disease. Patients will be enrolled in approximately 97 sites in across the Americas (including Canada), Asia, Europe, and Oceania.^32-34 Table 32 provides more details.

Table 32: Details of Other Relevant Studies — AcceleRET-Lung

CNS = central nervous system; 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; EORTC QLQ-LC13 = European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire Lung Cancer 13; NSCLC = non–small cell lung cancer; RECIST 1.1 = Response Evaluation Criteria in Solid Tumors Version 1.1.

^aThe date on which the last participant in a clinical study was examined or received an intervention to collect final data for the primary outcome measure. Whether the clinical study ended according to the protocol or was terminated does not affect this date. For clinical studies with more than a primary outcome measure with different completion dates, this term refers to the date on which data collection is completed for all the primary outcome measures.

^bThe date on which the last participant in a clinical study was examined or received an intervention or treatment to collect final data for the primary outcome measures, secondary outcome measures, and adverse events (that is, the last participant's last visit).

^cEstimated at up to 32 months.

^dBaseline, at every 21 day cycle visit until progressive disease or death (estimated 32 months).

^eFrom baseline until progressive disease or death (estimated 32 months).

Source: AcceleRET-Lung,³² Besse et al. (2021),³³ and Popat et al. (2022).³⁴

This study is currently recruiting participants, with a planned enrolment of approximately 226 patients. The estimated primary completion date (on which the last participant in a clinical study will be examined or receive an intervention to collect final data for the primary outcome measure) and study completion date (on which the last participant in a clinical study will be examined or receive an intervention or treatment to collect final data for the primary outcome measures, secondary outcome measures, and AEs) are September 30, 2023, and December 31, 2024, respectively. Patients are to be randomized in a 1:1 ratio to receive oral pralsetinib (400 mg once daily) or SOC. The SOC will be 1 of 6 platinum-based chemotherapy treatment regimens (with or without pembrolizumab) determined by the treating investigator based on histology. For nonsquamous histology, the treatment regimen will be carboplatin or cisplatin plus pemetrexed (with vitamin supplementation) with or without pembrolizumab followed by optional pemetrexed (with vitamin supplementation) maintenance with or without pembrolizumab; for squamous histology, the treatment regimen will be either carboplatin or cisplatin plus gemcitabine, or carboplatin with paclitaxel plus nab-paclitaxel and pembrolizumab followed by maintenance pembrolizumab.^32-34

Stratification factors include intended use of pembrolizumab, history of brain metastases, and ECOG PS. Key eligibility criteria include no prior systemic treatment for advanced and/or metastatic NSCLC; RET fusion–positive tumour by local or central assessment; no additional actionable oncogenic drivers; no prior selective RET inhibitor; and measurable disease as defined by RECIST 1.1. Patients with CNS metastases were included if they were asymptomatic and on a stable dose of corticosteroids. Crossing over to receive pralsetinib upon disease progression will be permitted for patients randomized to the SOC arm.^32-34

The primary end point is PFS (defined by RECIST 1.1) assessed by a BICR, whereas the secondary end points include ORR, OS, DOR, disease control rate, clinical benefit rate, time to intracranial progression, intracranial ORR, safety and tolerability, and quality of life evaluations. Identification of potential biomarkers of antineoplastic activity and resistance is an exploratory end point.^32-34

Discussion

Summary of Available Evidence

The pivotal trial submitted for this review, ARROW (N = 281; safety population at the November 6, 2020, data cut-off), is an ongoing phase I and II, open-label, single-arm study of pralsetinib in patients with RET fusion–positive locally advanced or metastatic cancer, including NSCLC. The first phase of the study was a dose-escalation study to establish the fixed dose for the second phase. The primary objective of the second phase of the study was to determine the efficacy (as measured by ORR) and safety of pralsetinib (400 mg once daily). The ORR was analyzed at the November 6, 2020, data cut-off in a subset efficacy population of 233 patients with NSCLC who had received their first dose of pralsetinib on or before May 22, 2020, with a median follow-up of 17.1 months. The average age was 59.2 years; 68% of patients had received prior systemic therapy, while 32% had not received any prior systemic therapy.

The sponsor submitted an ITC report aiming to demonstrate the efficacy of pralsetinib compared to relevant treatments. An SLR was conducted to identify evidence for comparators of interest. Given the lack of evidence for comparators of interest in patients with RET fusion–positive NSCLC, pralsetinib was compared to trials in wild-type NSCLC. For comparisons for which IPD were available to the sponsor (platinum-based chemotherapy plus pemetrexed in the first line and docetaxel in the second line), a propensity-scoring method was used to adjust for differences in identified prognostics factors across study populations. For all other comparisons for which IPD were not available to the sponsor, naive unadjusted comparisons were made for first-line pembrolizumab monotherapy, pembrolizumab in combinations with platinum-based chemotherapy plus pemetrexed, second-line nivolumab, and pemetrexed plus carboplatin. Key limitations included the unanchored nature of the indirect comparisons, the inability to adjust for all known and unknown prognostic factors, as well as the use of naive comparisons for all but 2 comparisons.

A published ITC from Popat et al. (2022)¹⁵ was identified that compared first-line patients receiving pralsetinib in the ARROW trial to synthetic control arms sourced from 3 real-world populations. In the first, patients with RET fusion–positive NSCLC received either a basket of best alternative therapy (most commonly pembrolizumab plus chemotherapy). The remaining 2 real-world populations were patients with wild-type NSCLC receiving pembrolizumab monotherapy and pembrolizumab plus chemotherapy. The analysis used IPW methodology where possible to adjust for differences in prognostic factors. The results found that patients given pralsetinib received a statistically significant benefit in OS and PFS compared to the chosen comparators, which is consistent with the expectations of the clinical expert consulted for this review. However, the same limitations are present as in the sponsor-submitted ITC. The analysis is an unanchored ITC relying on a limited number of prognostic factors and a small ESS compared to the original sample sizes of the populations.

Interpretation of Results

Efficacy

The patient input and the clinical expert consulted for this review identified several important goals of treatment for locally advanced and metastatic NSCLC, including prolonging a patient’s life, reducing symptom burden, delaying progression, and improving HRQoL. They confirmed that the outcomes (ORR, OS, PFS, DOR, CNS ORR, and HRQoL) investigated in the ARROW trial were appropriate and align with the needs of patients, caregivers, and clinicians in practice.

The clinical expert consulted for this review considered the ORR responses to be clinically meaningful to patients in practice. The expert emphasized that, based on their experience, the ORR is larger and DOR longer than what would be expected with other therapies offered as SOC. The expert noted that there is limited evidence that patients with locally advanced or metastatic RET fusion–positive NSCLC benefit from currently available single-drug immunotherapy, and not all patients respond to chemotherapy plus immunotherapy, indicating an unmet need for this patient population. The ITC was submitted to inform these comparisons of pralsetinib with SOC chemotherapy and immunotherapy; however, there were major limitations with this analysis stemming from the single-arm design of the ARROW trial, necessitating an unanchored comparison. Propensity-score weighting was used to account for between-trial differences for first-line platinum-based chemotherapy plus pemetrexed and second-line docetaxel, but naive unadjusted comparisons were made for all other comparisons. Given these limitations, conclusions cannot be drawn from the naive comparisons and should be drawn instead, but with caution, from the propensity-score weighted analysis. However, all results for PFS and OS from the ITC were directionally in favour of pralsetinib and aligned with the clinical expert’s expectation that pralsetinib is likely to be a beneficial option over SOC chemotherapy and immunotherapy for this patient population. Furthermore, the ITC published by Popat et al. (2022) reports a statistically significant benefit for OS and PFS from pralsetinib over pembrolizumab monotherapy and pembrolizumab plus chemotherapy, although this analysis comes with important limitations as well.

The median OS was not reached in the ARROW trial and is therefore considered immature. The sponsor provided analysis based on the methodology published by Anderson et al. (2008)³⁵ investigating the correlation of survival based on the ORR within the ARROW trial. The results suggest that responders in the ARROW trial were more likely to be alive at 2, 3, and 4 months. While this analysis provides important context and adds to the evidence base, it must be considered alongside systematic reviews such as those published by Haslam et al. (2019)³⁶ and Cooper et al. (2020)³⁷ that show surrogate end points, such as ORR, may be unreliable indicators of long-term OS.

The HRQoL outcomes in the ARROW trial were limited by low patient numbers due to the measure being added to the protocol as an amendment after the trial had begun and the further loss of patients over time. The results were consistent and positive, reaching the MID threshold for a moderate improvement (10 to 20 points) at some time points. Health-related quality of life is an important outcome for patients and, although the results from ARROW are promising, their interpretation is limited by the single-arm design of the trial, low patient numbers, and loss of patients to study discontinuation for reasons other than mortality follow-up.

The ability to affect CNS and/or brain metastasis outcomes is an important factor when comparing pralsetinib to SOC chemotherapy and immunotherapy. A common site for NSCLC metastasis is the CNS and/or brain, which come with burdensome symptoms for patients and worse prognosis. Although SOC chemotherapy and immunotherapy has little to no activity in the brain, given that pralsetinib can cross the blood-brain barrier, it is expected to benefit patients with CNS and/or brain metastasis. This expectation is further supported by the ARROW trial, in which 10 patients had evaluable intracranial lesions, and the ORR was 70% for these specific lesions, including 30% of patients who reported a CR. The clinical expert consulted for this review suggested that there may be a protective element to pralsetinib and the prevention of CNS and/or brain metastasis, and this was supported by the ARROW trial: of the patients treated at 400 mg once daily who did not have a history of CNS metastasis at baseline, none developed CNS metastasis throughout the trial. However, due to the single-arm nature of the trial, this should be considered speculatory.

It is noted that selpercatinib, a similar targeted therapy for patients with RET fusion–positive NSCLC, has recently received a positive recommendation with conditions from CADTH. Although there is no evidence available comparing the 2 therapies, the clinical expert consulted for this review stated that there was no evidence to suggest a significant difference between pralsetinib and selpercatinib in terms of efficacy.

Harms

In the ARROW trial, pralsetinib appeared to have a tolerable safety profile, according to the clinical expert consulted for this review. The proportion of patients who discontinued treatment due to an AE was 19.6%, although this number is likely inflated due to the fact that disease progression was characterized as an AE and was the most commonly cited AE that resulted in discontinuations. Pneumonitis, a key AE of special interest for pralsetinib, was reported as a SAE by 4.6% of patients at the November 6, 2020, data cut-off. At that same data cut-off, 2.1% of patients reported grade 3, 4, or 5 pneumonitis, and no deaths due to pneumonitis were reported. The clinical expert noted that, given the assumption of similar efficacy between pralsetinib and selpercatinib, treatment decisions for patients would likely be determined by the differing safety profile between the 2 therapies, with pneumonitis or de novo poor pulmonary reserves, for example, indicating that a patient may be better suited for selpercatinib. The submitted ITC did not conduct analysis on safety end points; however, input from patient groups and clinical opinion suggests that pralsetinib, as a targeted oral therapy, offers reduced toxicity compared to IV SOC chemotherapies and immunotherapies.

Conclusions

The evidence supporting the funding request for pralsetinib was derived from an ongoing phase I and II, open-label, single-arm study, ARROW. The ORR observed in the ARROW trial, based on unplanned interim analysis results, suggested favourable tumour response in both treatment-naive and treatment-experienced patients and was consistent with further follow-up analysis. The ORR and DOR, including CNS ORR, were considered clinically meaningful by the clinical expert consulted for this review. The ability to draw conclusions from time-to-event end points of PFS and OS was limited by the immaturity of the data and the single-arm design of the trial. The safety profile of pralsetinib in the ARROW trial was considered by the clinical expert to be an improvement compared to SOC chemotherapy and immunotherapy. In the opinion of the clinical expert, the differences in the safety profile compared with selpercatinib highlight the benefits to patients that come with additional treatment options. The ITC submitted to inform the comparative effects of pralsetinib were associated with limitations that prevented drawing conclusions from the results. Uncertainty therefore remains in the comparative effectiveness and safety of pralsetinib.

References

1.Canadian Cancer Statistics Advisory Committee. Canadian cancer statistics 2021. Toronto (ON): Canadian Cancer Society; 2021: https://cdn.cancer.ca/-/media/files/research/cancer-statistics/2021-statistics/2021-pdf-en-final.pdf. Accessed 2022 Apr 26.

2.National Cancer Institute. Non-small cell lung cancer treatment (PDQ®) – health professional version. 2022; https://www.cancer.gov/types/lung/hp/non-small-cell-lung-treatment-pdq. Accessed 2022 Apr 26.

3.Cancer.Net. Lung cancer - non-small cell: statistics. 2022; https://www.cancer.net/cancer-types/lung-cancer-non-small-cell/statistics. Accessed 2022 Apr 26.

4.Drug Reimbursement Review sponsor submission: Gavreto (pralsetinib), 100 mg capsules, oral [internal sponsor's package]. Mississauga (ON): Hoffmann-La Roche Ltd.; 2022 Mar 9.

5.Clinical Study Report: BLU-667-1101. A phase 1/2 study of the highly-selective RET inhibitor, BLU-667, in patients with thyroid cancer, non-small cell lung cancer (NSCLC) and other advanced solid tumors [internal sponsor's report]. Cambridge (MA): Blueprint Medicines; 2020 Feb 26.

6.Committee for Medicinal Products for Human Use. Assessment report: Gavreto (pralsetinib). (European public assessment report). Amsterdam (NL): European Medicines Agency; 2021: https://www.ema.europa.eu/en/documents/assessment-report/gavreto-epar-public-assessment-report_en.pdf. Accessed 2022 May 3.

7.Mok TSK, Wu Y-L, Kudaba I, et al. Pembrolizumab versus chemotherapy for previously untreated, PD-L1-expressing, locally advanced or metastatic non-small-cell lung cancer (KEYNOTE-042): a randomised, open-label, controlled, phase 3 trial. The Lancet. 2019;393(10183):1819-1830. PubMed

8.Gadgeel S, Rodríguez-Abreu D, Speranza G, et al. Updated analysis from KEYNOTE-189: pembrolizumab or placebo plus pemetrexed and platinum for previously untreated metastatic nonsquamous non–small-cell lung cancer. J Clin Oncol. 2020;38(14):1505-1517. PubMed

9.Nishio M, Barlesi F, West H, et al. Atezolizumab plus chemotherapy for first-line treatment of nonsquamous NSCLC: results from the randomized phase 3 IMpower132 trial. J Thorac Oncol. 2021;16(4):653-664. PubMed

10.Gandara DR, von Pawel J, Mazieres J, et al. Atezolizumab treatment beyond progression in advanced NSCLC: results from the randomized, phase III OAK study. J Thorac Oncol. 2018;13(12):1906-1918. PubMed

11.Borghaei H, Gettinger S, Vokes EE, et al. Five-year outcomes from the randomized, phase III trials checkmate 017 and 057: nivolumab versus docetaxel in previously treated non-small-cell lung cancer. J Clin Oncol. 2021;39(7):723-733. PubMed

12.Ardizzoni A, Tiseo M, Boni L, et al. Pemetrexed versus pemetrexed and carboplatin as second-line chemotherapy in advanced non-small-cell lung cancer: results of the GOIRC 02-2006 randomized phase II study and pooled analysis with the NVALT7 trial. J Clin Oncol. 2012;30(36):4501-4507. PubMed

13.Smit EF, Burgers SA, Biesma B, et al. Randomized phase II and pharmacogenetic study of pemetrexed compared with pemetrexed plus carboplatin in pretreated patients with advanced non-small-cell lung cancer. J Clin Oncol. 2009;27(12):2038-2045. PubMed

14.Hess LM, Han Y, Zhu YE, Bhandari NR, Sireci A. Characteristics and outcomes of patients with RET-fusion positive non-small lung cancer in real-world practice in the United States. BMC Cancer. 2021;21(1):28. PubMed

15.Popat S, Liu SV, Scheuer N, et al. Addressing challenges with real-world synthetic control arms to demonstrate the comparative effectiveness of pralsetinib in non-small cell lung cancer. Nat Commun. 2022;13(1):3500. PubMed

16.Lung Cancer Canada. Lung cancer staging in Canada. 2020; https://www.lungcancercanada.ca/Lung-Cancer/Staging.aspx. Accessed 2022 Apr 26.

17.Iyer S, Roughley A, Rider A, Taylor-Stokes G. The symptom burden of non-small cell lung cancer in the USA: a real-world cross-sectional study. Support Care Cancer. 2014;22(1):181-187. PubMed

18.NCCN clinical practice guidelines in oncology (NCCN guidelines): non-small cell lung cancer. Version 3.2022. Plymouth Meeting (PA): National Comprehensive Cancer Network (NCCN); 2022: https://www.nccn.org/. Accessed 2022 Apr 26.

19.Planchard D, Popat S, Kerr K, et al. Metastatic non-small cell lung cancer: ESMO Clinical Practice Guidelines for diagnosis, treatment and follow-up. Lugano (CH): European Society for Medical Oncology; 2020: https://www.esmo.org/content/download/347819/6934778/1/ESMO-CPG-mNSCLC-15SEPT2020.pdf. Accessed 2022 Jul 19.

20.Ellis PM, Vella ET, Ung YC. Systemic treatment for patients with advanced non-small cell lung cancer. (Guideline 7-10, version 3). Toronto (ON): Cancer Care Ontario; 2016: https://www.cancercareontario.ca/en/guidelines-advice/types-of-cancer/31811 Accessed 2022 Jul 19.

21.Mahato AK, Sidorova YA. RET receptor tyrosine kinase: role in neurodegeneration, obesity, and cancer. Int J Mol Sci. 2020;21(19):7108. PubMed

22.Cascetta P, Sforza V, Manzo A, et al. RET inhibitors in non-small-cell lung cancer. Cancers (Basel). 2021;13(17):01.

23.Gavreto (pralsetinib capsules): capsules, 100 mg, oral. Protein kinase inhibitor (L01EX) [product monograph]. Mississauga (ON): Hoffmann-La Roche Limited; 2021 Jul 12.

24.McGowan J, Sampson M, Salzwedel DM, Cogo E, Foerster V, Lefebvre C. PRESS Peer Review of Electronic Search Strategies: 2015 guideline statement. J Clin Epidemiol. 2016;75:40-46. PubMed

25.Grey matters: a practical tool for searching health-related grey literature. Ottawa (ON): CADTH; 2019: https://www.cadth.ca/grey-matters. Accessed 2022 Mar 23.

26.Gainor JF, Curigliano G, Kim DW, et al. Pralsetinib for RET fusion-positive non-small-cell lung cancer (ARROW): a multi-cohort, open-label, phase 1/2 study. Lancet Oncol. 2021;22(7):959-969. PubMed

27.Health Canada reviewer's report: Gavreto (pralsetinib) [internal sponsor's report]. In: Drug Reimbursement Review sponsor submission: Gavreto (pralsetinib), 100 mg capsules, oral. Mississauga (ON): Hoffmann-La Roche Ltd.; 2022 Mar 9.

28.Clinical trial endpoints for the approval of non-small cell lung cancer drugs and biologics: guidance for industry. Silver Spring (MD): U.S. Food and Drug Administration; 2015: https://www.fda.gov/media/116860/download. Accessed 2022 Apr 26.

29.Wild type (WT) non-small cell lung cancer (NSCLC): updated clinical systematic literature review (SLR) and comparative analyses of pralsetinib with comparators of interest [internal sponsor's report]. In: Drug Reimbursement Review sponsor submission: Gavreto (pralsetinib), 100 mg capsules, oral. Mississauga (ON): Hoffmann-La Roche Ltd.; 2022 Mar 9.

30.Faria R, Hernández Alava M, Manca A, Wailoo AJ. NICE DSU Technical Support Document 17: the use of observational data to inform estimates of treatment effectiveness in technology appraisal: methods for compariatve individual patient data. London (GB): National Institute for Health and Care Excellence; 2015: https://www.sheffield.ac.uk/media/34204/download?attachment. Accessed 2022 Jul 19.

31.Guyot P, Ades AE, Ouwens MJNM, Welton NJ. Enhanced secondary analysis of survival data: reconstructing the data from published Kaplan-Meier survival curves. BMC Med Res Methodol. 2012;12(1):9. PubMed

32.Hoffmann-La Roche. NCT04222972: A study of pralsetinib versus standard of care for first-line treatment of advanced non-small cell lung cancer (NSCLC) (AcceleRET-Lung). ClinicalTrials.gov. Bethesda (MD): U.S. National Library of Medicine; 2022: https://clinicaltrials.gov/ct2/show/study/NCT04222972. Accessed 2022 Jun 9.

33.Besse B, Felip E, Kim ES, et al. PUL01.02 AcceleRET lung: a phase 3 study of first-line pralsetinib in patients with RET-fusion+ advanced/metastatic NSCLC. J Thorac Oncol. 2021;16(1):S44-S45.

34.Popat S, Felip E, Kim ES, et al. AcceleRET Lung: A phase 3 study of first-line pralsetinib in patients with RET fusion–positive advanced/metastatic NSCLC. J Clin Oncol. 2022;40(16_suppl):TPS9159-TPS9159.

35.Anderson JR, Cain KC, Gelber RD. Analysis of survival by tumor response and other comparisons of time-to-event by outcome variables. J Clin Oncol. 2008;26(24):3913-3915. PubMed

36.Haslam A, Hey SP, Gill J, Prasad V. A systematic review of trial-level meta-analyses measuring the strength of association between surrogate end-points and overall survival in oncology. Eur J Cancer. 2019;106:196-211. PubMed

37.Cooper K, Tappenden P, Cantrell A, Ennis K. A systematic review of meta-analyses assessing the validity of tumour response endpoints as surrogates for progression-free or overall survival in cancer. Br J Cancer. 2020;123(11):1686-1696. PubMed

38.Aaronson NK, Ahmedzai S, Bergman B, et al. The European Organization for Research and Treatment of Cancer QLQ-C30: a quality-of-life instrument for use in international clinical trials in oncology. J Natl Cancer Inst. 1993;85(5):365-376. PubMed

39.Letellier ME, Dawes D, Mayo N. Content verification of the EORTC QLQ-C30/EORTC QLQ-BR23 with the International Classification of Functioning, Disability and Health. Qual Life Res. 2015;24(3):757-768. PubMed

40.Nicklasson M, Bergman B. Validity, reliability and clinical relevance of EORTC QLQ-C30 and LC13 in patients with chest malignancies in a palliative setting. Qual Life Res. 2007;16(6):1019-1028. PubMed

41.Larsson M, Ljung L, Johansson BB. Health-related quality of life in advanced non-small cell lung cancer: correlates and comparisons to normative data. Eur J Cancer Care. 2012;21(5):642-649. PubMed

42.Bedard G, Zeng L, Zhang L, et al. Minimal important differences in the EORTC QLQ-C30 in patients with advanced cancer. Asia Pac J Clin Oncol. 2014;10(2):109-117. PubMed

43.Giesinger JM, Kieffer JM, Fayers PM, et al. Replication and validation of higher order models demonstrated that a summary score for the EORTC QLQ-C30 is robust. J Clin Epidemiol. 2016;69:79-88. PubMed

44.Fayers PM, Aaronson NK, Bjordal K, et al. The EORTC QLQ-C30 Scoring Manual. 3rd ed. Brussels (BE): European Organisation for Research and Treatment of Cancer; 2001: https://www.eortc.org/app/uploads/sites/2/2018/02/SCmanual.pdf. Accessed 2022 Jul 19.

45.Osoba D, Zee B, Pater J, Warr D, Kaizer L, Latreille J. Psychometric properties and responsiveness of the EORTC quality of Life Questionnaire (QLQ-C30) in patients with breast, ovarian and lung cancer. Qual Life Res. 1994;3(5):353-364. PubMed

46.Osoba D, Rodrigues G, Myles J, Zee B, Pater J. Interpreting the significance of changes in health-related quality-of-life scores. J Clin Oncol. 1998;16(1):139-144. PubMed

47.Maringwa JT, Quinten C, King M, et al. Minimal important differences for interpreting health-related quality of life scores from the EORTC QLQ-C30 in lung cancer patients participating in randomized controlled trials. Support Care Cancer. 2011;19(11):1753-1760. PubMed

48.Revicki D, Hays RD, Cella D, Sloan J. Recommended methods for determining responsiveness and minimally important differences for patient-reported outcomes. J Clin Epidemiol. 2008;61(2):102-109. PubMed

49.Van Der Weijst L, Surmont V, Schrauwen W, Lievens Y. Real life data on patient-reported outcomes and neuro-cognitive functioning of lung cancer patients: the PRO-Long study. Front Oncol. 2021;11:685605. PubMed

50.Smit EF, van Meerbeeck JP, Lianes P, et al. Three-arm randomized study of two cisplatin-based regimens and paclitaxel plus gemcitabine in advanced non-small-cell lung cancer: a phase III trial of the European Organization for Research and Treatment of Cancer Lung Cancer Group--EORTC 08975. J Clin Oncol. 2003;21(21):3909-3917. PubMed

51.Sarna L, Swann S, Langer C, et al. Clinically meaningful differences in patient-reported outcomes with amifostine in combination with chemoradiation for locally advanced non-small-cell lung cancer: an analysis of RTOG 9801. Int J Radiat Oncol Biol Phys. 2008;72(5):1378-1384. PubMed

52.Rutkowski J, Szymanik M, Blok M, Kozaka J, Zaucha R. Prospective evaluation of anxiety, depression and quality of life in medically inoperable early stage non-small cell lung cancer patients treated with stereotactic ablative radiotherapy. Rep Pract Oncol Radiother. 2017;22(3):217-222. PubMed

53.Rittmeyer A, Gorbunova V, Vikstrom A, et al. Health-related quality of life in patients with advanced nonsquamous non-small-cell lung cancer receiving bevacizumab or bevacizumab-plus-pemetrexed maintenance therapy in AVAPERL (MO22089). J Thorac Oncol. 2013;8(11):1409-1416. PubMed

54.Ponce Aix S, Talbot D, Govindan R, et al. Quality of life with second or third line nab-paclitaxel-based regimens in advanced non-small-cell lung cancer. Fut Oncol. 2020;16(12):749-762. PubMed

55.Movsas B, Scott C, Langer C, et al. Randomized trial of amifostine in locally advanced non-small-cell lung cancer patients receiving chemotherapy and hyperfractionated radiation: radiation therapy oncology group trial 98-01. J Clin Oncol. 2005;23(10):2145-2154. PubMed

56.Mathieu D, Campeau MP, Bahig H, et al. Long-term quality of life in early-stage non-small cell lung cancer patients treated with robotic stereotactic ablative radiation therapy. Pract Radiat Oncol. 2015;5(4):e365-373. PubMed

57.Hui R, Ozguroglu M, Villegas A, et al. Patient-reported outcomes with durvalumab after chemoradiotherapy in stage III, unresectable non-small-cell lung cancer (PACIFIC): a randomised, controlled, phase 3 study. Lancet Oncol. 2019;20(12):1670-1680. PubMed

58.Hechtner M, Eichler M, Wehler B, et al. Quality of life in NSCLC survivors - a multicenter cross-sectional study. J Thorac Oncol. 2019;14(3):420-435. PubMed

59.Fiteni F, Anota A, Bonnetain F, et al. Health-related quality of life in elderly patients with advanced non-small cell lung cancer comparing carboplatin and weekly paclitaxel doublet chemotherapy with monotherapy. Eur Respir J. 2016;48(3):861-872. PubMed

Appendix 1: Literature Search Strategy

Note that this appendix has not been copy-edited.

Clinical Literature Search

Overview

Interface: Ovid

Databases

• MEDLINE All (1946-present)

• Embase (1974-present)

• Note: Subject headings and search fields have been customized for each database. Duplicates between databases were removed in Ovid.

Date of search: March 31, 2022

Alerts: Bi-weekly search updates until project completion

Search filters applied: No filters were applied to limit the retrieval by study type.

Limits:

• Conference abstracts: excluded

Table 33: Syntax Guide

Multi-Database Strategy

1. (pralsetinib* or pratinib* or gavreto* or BLU-667 or BLU667 or BLU-123244 or BLU123244 or BLU-3244 or BLU3244 or C-683 or C683 or CS-3009 or CS3009 or RG-6396 or RG6396 or X-581238 or X581238 or WHO-11004 or WHO11004 or 1WPE73O1WV).ti,ab,kf,ot,hw,rn,nm.

1. 1 use medall

2. *pralsetinib/

3. (pralsetinib* or pratinib* or gavreto* or BLU-667 or BLU667 or BLU-123244 or BLU123244 or BLU-3244 or BLU3244 or C-683 or C683 or CS-3009 or CS3009 or RG-6396 or RG6396 or X-581238 or X581238 or WHO-11004 or WHO11004).ti,ab,kf,dq.

4. 3 or 4

5. 5 use oemezd

6. (conference abstract or conference review).pt.

7. 6 not 7

8. 2 or 8

9. remove duplicates from 9

Clinical Trials Registries

ClinicalTrials.gov

Produced by the US National Library of Medicine. Targeted search used to capture registered clinical trials.

[Search terms: (Gavreto OR pralsetinib) AND RET fusion–positive non-small cell lung cancer (NSCLC)]

WHO ICTRP

International Clinical Trials Registry Platform, produced by the World Health Organization. Targeted search used to capture registered clinical trials.

[Search terms: (Gavreto OR pralsetinib) AND RET fusion–positive non-small cell lung cancer (NSCLC)]

Health Canada’s Clinical Trials Database

Produced by Health Canada. Targeted search used to capture registered clinical trials.

[Search terms: (Gavreto OR pralsetinib) AND RET fusion–positive non-small cell lung cancer (NSCLC)]

EU Clinical Trials Register

European Union Clinical Trials Register, produced by the European Union. Targeted search used to capture registered clinical trials.

[Search terms: (Gavreto OR pralsetinib) AND RET fusion–positive non-small cell lung cancer (NSCLC)]

Grey Literature

Search dates: March 25 – March 30, 2022

Keywords: (Gavreto OR pralsetinib) AND RET fusion–positive non-small cell lung cancer (NSCLC)

Limits: No limits

Updated: Search updated prior to the completion of stakeholder feedback period

Relevant websites from the following sections of the CADTH grey literature checklist Grey Matters: A Practical Tool for Searching Health-Related Grey Literature were searched:

• Health Technology Assessment Agencies

• Health Economics

• Clinical Practice Guidelines

• Drug and Device Regulatory Approvals

• Advisories and Warnings

• Drug Class Reviews

• Clinical Trials Registries

• Databases (free)

• Health Statistics

• Internet Search

• Open Access Journals

Appendix 2: Excluded Studies

Note that this appendix has not been copy-edited.

Table 34: Excluded Studies