CADTH Reimbursement Review

Ripretinib (Qinlock)

Sponsor: Medison Pharma Canada Inc.

Therapeutic area: Gastrointestinal stromal tumours

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

Gastrointestinal stromal tumours (GISTs) are rare soft-tissue sarcomas of the gastrointestinal tract.¹ GISTs are typically characterized by primary gain-of-function mutations in CD117/c-KIT (80% to 85% of GISTs)² and platelet-derived growth factor alpha (PDGFRA).³ Small and/or slow-growing GISTs may be clinically more benign, while tumours that grow significantly outward from the bowel wall can cause dysphagia, bleeding, abdominal pain/discomfort, fatigue, vomiting, loss of appetite, and other gastrointestinal issues.⁴ Both disease symptoms and the side effects of therapy severely impact health-related quality of life (HRQoL). At diagnosis, approximately half of patients with GISTs are eligible for potentially curative surgical resection.³ Among patients undergoing resection, disease will recur in about half within 5 years.⁵ Patients with advanced disease are transferred to the care of a medical oncologist for systemic therapy with palliative intent. According to the clinical experts consulted by CADTH for this review, in these patients, chemotherapy and radiotherapy are ineffective,^1,6 and the mainstay of therapy is sequential treatment with the tyrosine kinase inhibitors (TKIs) imatinib, sunitinib, and regorafenib. Responses to each line of therapy after imatinib are generally short-lived, and resistance and progression develop in most patients within months. Following progression on third-line regorafenib, there are no standard therapy options beyond best supportive care (BSC). Patient outcomes are unfavourable: further progression and death can be expected within a few months.

Published data on the prevalence, incidence, and survival of advanced GIST in Canada are unavailable. Based on an estimated 500 GIST cases diagnosed per year in Canada,⁷ with 75% representing advanced GIST, and assuming 80%, 70%, and 60% failure/progression rates on imatinib, sunitinib, and regorafenib, respectively, the sponsor estimated a target population of 62 to 86 patients with advanced GIST per year from 2023 to 2025 in Canada (outside of Quebec) who have received prior treatment with imatinib, sunitinib, and regorafenib and would be eligible for fourth-line ripretinib.⁸

Ripretinib is a TKI administered at a dosage of 150 mg orally (three 50 mg tablets) once daily. Ripretinib is indicated “for the treatment of adult patients with advanced gastrointestinal stromal tumour (GIST) who have received prior treatment with imatinib, sunitinib, and regorafenib.” The objective of this report was to perform a systematic review of the beneficial and harmful effects of ripretinib for the treatment of adult patients with advanced GIST who have received prior treatment with imatinib, sunitinib, and regorafenib.

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 clinical experts consulted by CADTH for the purpose of this review.

Patient Input

Two patient groups submitted patient input for this review: the CanCertainty Coalition and the GIST Sarcoma Life Raft Group Canada (LRGC). CanCertainty raised concerns about the financial and administrative barriers to accessing cancer treatments such as ripretinib in some Canadian jurisdictions (Ontario and the Atlantic provinces). Young patients (younger than 65 years) who require take-home cancer treatment such as ripretinib and do not have private or automatic public prescription drug coverage may incur significant deductibles or co-payments from their personal savings. These costs can become a financial burden and may lead to distress and hardship for Canadian patients with GIST younger than 65 years without private drug coverage (estimated at approximately 5 patients per year).

In September and October 2021, LRGC conducted telephone interviews with 11 patients with advanced GIST (5 Canadian and 6 from the US) who had experience with ripretinib. All respondents were either initially or eventually diagnosed with metastatic GIST, and many experienced delays in diagnosis due to nonspecific symptoms. Patients highlighted the negative impacts of advanced GIST on HRQoL, including symptoms of vomiting, abdominal pain/discomfort, and bowel issues, including diarrhea, severe fatigue, black stools, and loss of appetite. Patients had received 1 line to 4 lines of therapy before ripretinib, and several recounted their rapid progression and sometimes severe side effects during treatment with prior TKIs. Most patients conveyed that ripretinib was generally more tolerable than other TKIs, with milder and acceptable side effects that included hair loss, cramping in body extremities, nausea, fatigue, hand-and-foot syndrome, foot calluses, and curly/kinky hair regrowth. More than half of patients reported improved HRQoL during ripretinib treatment compared with prior TKIs.

Patients with advanced GIST identified an unmet need for novel therapies that can stabilize or enhance HRQoL while effectively reducing disease progression for several years. In addition to improved survival, patients desired access to new drugs with that have improved toxicity profiles and longer-term effectiveness, and that can target specific GIST mutations.

Clinician Input

Input From Clinical Experts Consulted by CADTH

Two clinical experts with expertise in the diagnosis and management of patients with advanced GIST were consulted for this review. According to the clinical experts, not all patients with metastatic GIST respond to available treatments (TKIs), and responses are generally short-lived, especially in later lines of therapy. Following exhaustion of available TKIs (imatinib, sunitinib, and regorafenib), there are no standard treatment options available in Canada, and additional lines of therapy are required to fulfill the unmet needs of these patients. According to the clinical experts, the goals of fourth-line treatment of advanced GIST following progression/intolerance to imatinib, sunitinib, and regorafenib are prolonging survival, delaying disease progression, palliating symptoms, and preventing new symptom development. The clinical experts stated that, based on currently available evidence, ripretinib would be used for fourth-line monotherapy after progression on or intolerance to imatinib, sunitinib, and regorafenib and would not result in a treatment paradigm shift but rather would provide an additional option for later-line therapy in patients with no other good options. According to the clinical experts, there are no established biomarkers of response to ripretinib, and all patients with advanced GIST who experienced progression or intolerance on imatinib, sunitinib, and regorafenib with adequate performance status (PS), organ function, and hematological function would be candidates for ripretinib, irrespective of tumour mutational status. Patients with poor PS, limited organ/hematological function, significant comorbidities (especially cardiac problems), central nervous system metastases, and problems taking or absorbing oral medications would be least suitable for ripretinib treatment.

The clinical experts stated that treatment with ripretinib would be initiated either immediately following progression on third-line treatment (regorafenib) or after symptoms worsen follow discontinuation of third-line treatment. According to the clinical experts, response to ripretinib treatment would be assessed by clinical evaluation, in conjunction with imaging scans, every 2 months to 4 months. Clinically meaningful responses to therapy would be reflected by restricted tumour growth, prolongation of overall survival (OS) and progression-free survival (PFS), maintained or improved HRQoL, and stabilization or reduction of symptom severity. Treatment would be discontinued because of disease progression, significant adverse events (AEs), persistent treatment intolerance despite dosage reductions, or patient preference. The clinical experts also noted the convenience of ripretinib as an oral drug that can be self-administered at home in this advanced disease setting.

Clinician Group Input

One group of 7 Canadian medical oncologists who treat patients with advanced GIST provided input for this review; some of the oncologists are medical advisors to LRGC. No major contrary views were presented. The clinicians echoed the absence of fourth-line treatment options for patients after available TKIs (imatinib, sunitinib, and regorafenib) have been exhausted and the poor outcomes in these patients. Minor discrepancies were noted between the clinical experts and the clinician group input in the frequency of response assessment by imaging scans (2 to 3 months versus 3 to 4 months), possibly due to jurisdictional variation.

Drug Program Input

The Provincial Advisory Group (PAG) identified the following jurisdictional implementation issues: relevant comparators, considerations for continuation or renewal of therapy, considerations for discontinuation of therapy, considerations for prescribing of therapy, and care provision issues. The clinical experts consulted by CADTH for this review weighed evidence from the included study and other clinical considerations to provide responses to PAG’s drug program implementation questions (Table 4).

Clinical Evidence

Pivotal Studies and Protocol-Selected Studies

Description of Studies

INVICTUS was a phase III, double-blind (DB), placebo-controlled multi-centre randomized controlled trial (RCT; N = 129) with an open-label (OL) period of active treatment.^9-11 The primary objective of the study was to assess the efficacy of ripretinib in prolonging PFS per independent radiologic review (IRR) in patients with advanced GIST who had received prior anticancer therapies, including imatinib, sunitinib, and regorafenib. Secondary objectives included comparing objective response rate (ORR) per IRR (hierarchically tested), OS (hierarchically tested), patient-reported changes in disease symptoms and HRQoL from baseline to the start of cycle 2 (hierarchically tested), and other efficacy outcomes, including duration of response (DOR) between the ripretinib and placebo arms. Patients were enrolled at 29 sites in 12 countries (1 site in Toronto, Canada) and randomized to DB treatment with either ripretinib 150 mg orally once daily plus BSC or placebo orally once daily plus BSC. Following initial objective progression per IRR, patients and investigators were unblinded to treatment allocation, and patients could choose either to receive OL ripretinib at the same dosage (150 mg once daily) or to escalate the dosage to OL ripretinib 150 mg twice daily. Following treatment discontinuation, patients entered survival follow-up.

Adult patients with unresectable advanced GIST who had progressed on or developed intolerance to imatinib, sunitinib, and regorafenib were eligible for the study if they had Eastern Ontario Oncology Group (ECOG) PS 0, 1, or 2 and did not have active central nervous system metastases, clinically significant cardiac conditions or other comorbidities, or gastrointestinal problems preventing their taking or absorbing oral medication. The mean age of participants was approximately 60 years, approximately 57% were men, approximately 75% were White, and approximately 47% were enrolled at sites in the US. The most common tumour site was gastric (45.0%), and the most common location of primary tumour mutations was KIT exon 11 (58.1%). Approximately 60% of patients had received 3 prior lines of therapy, while approximately 40% had received 4 or more prior lines of therapy. Baseline demographic and disease characteristics were generally well balanced between study arms, apart from minor imbalances of potential prognostic relevance in age, ECOG PS, and gastric tumour site. The clinical experts consulted by CADTH for this review did not feel that any of these imbalances would be likely to affect the study results.

Efficacy Results

Key efficacy results of the INVICTUS study are summarized in Table 2. Importantly, the OS analysis did not account for crossover from placebo to ripretinib following initial objective progression, post-progression ripretinib treatment, or post-progression dosage escalation to the non–Health Canada–approved dosage of 150 mg twice daily. At the time of the primary analysis (database lock May 31, 2019), median OS was 28.6 weeks (95% confidence interval [CI], 17.9 to 50.4 weeks) in patients originally randomized to the placebo arm and 65.6 weeks (95% CI, 53.6 to 65.6 weeks) in patients originally randomized to the ripretinib arm. The hazard ratio (HR) for OS comparing ripretinib to placebo was 0.36 (95% CI, 0.21 to 0.62). A post hoc subgroup analysis of OS by combined treatment assignment in both the DB and OL periods showed the following results: DB placebo, no crossover, median OS 7.9 weeks (95% CI, 3.7 to 19.6 weeks); DB placebo with crossover to OL ripretinib 150 mg once daily, median OS 30.1 weeks (95% CI, 12.4 weeks to not calculable); and ||| |||||||||| ||| || || |||| || || |||||||||| |||||| || |||| ||||| |||| ||| |||| ||||| || ||| |||||||||||| ||| |||| ||| |||||||||| ||| || || |||| || |||||||||| ||| || ||| |||||| || ||||| ||||| |||| ||| |||| ||||| || ||||||||||||||||. During DB treatment, no objective tumour responses occurred in the placebo arm, while the ORR at the time of the primary analysis was 9.4% (95% CI, 4.2% to 17.7%) in the ripretinib arm (P = 0.0504). Among patients responding to ripretinib, median DOR was not estimable at the time of the primary analysis (May 31, 2019, data cut) but was 14.5 months (95% CI, 3.7 weeks to not estimable) at the more recent data cut of January 15, 2021. At the time of the primary analysis, median PFS during DB treatment was 4.1 weeks (95% CI, 4.0 to 7.3 weeks) in the placebo arm and 27.6 weeks (95% CI, 20.0 to 29.9 weeks) in the ripretinib arm (P < 0.0001). The HR for PFS comparing ripretinib to placebo was 0.15 (95% CI, 0.09 to 0.25). OS, ORR, and PFS results for the most recent cut (January 15, 2021) were similar. The results of the primary PFS analysis were statistically and clinically significant, according to the clinical experts consulted by CADTH for this review. Comparisons of OS and ORR (despite their descriptive nature and non-statistical significance, respectively, and despite the complexities of the OS analysis) were viewed by the clinical experts as supportive of the PFS findings and were judged to be potentially clinically important, given that the population is affected by advanced disease and has no other available treatment options.

HRQoL indicators (European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire for Cancer 30 item [EORTC QLQ-C30] role and physical function, EQ-5D-5L usual activities and pain/discomfort, EQ-5D-5L utility index scores, and EQ-visual analogue scale [VAS]) were similar at baseline in the ripretinib and placebo arms. Interpretation of changes from baseline to cycle 2 day 1 was limited by several factors (refer to Critical Appraisal).

Harms Results

Key harms results of the INVICTUS study are summarized in Table 2. AEs occurred in most patients treated with placebo (97.7%) and ripretinib (98.8%). Serious AEs occurred in larger proportions of patients receiving placebo (44.2%) than ripretinib (30.6%). Withdrawals due to AEs occurred in 11.6% of placebo-treated patients and 8.2% of ripretinib-treated patients. Deaths were more frequent among placebo-treated patients (23.3%) than among ripretinib-treated patients (5.9%), primarily due to disease progression.

Among protocol-specified AEs of special interest, squamous cell carcinoma of the skin occurred in no patients who received placebo and 2 patients (2.4%) who received ripretinib. Actinic keratosis occurred in 1 patient (2.3%) who received placebo and 5 patients (5.9%) who received ripretinib. All notable harms specified in the CADTH review protocol occurred more frequently in patients who received ripretinib than in those who received placebo: cardiac dysfunction, cardiac ischemic events, hypertension, cutaneous malignancies, palmar-plantar erythrodysesthesia syndrome, arthralgia, myalgia, and increased bilirubin. The most common AEs by preferred term were peripheral edema (ripretinib versus placebo: 16.5% versus 7.0%), hypertension (14.1% versus 4.7%), palmar-plantar erythrodysesthesia syndrome (21.2% versus 0%), arthralgia (17.6% versus 4.7%), myalgia (31.8% versus 11.6%), and increased bilirubin (16.5% versus 0%).

Critical Appraisal

The major limitation of the INVICTUS trial was its small size and associated uncertainty, although this was expected for a study of a rare disease. Minor baseline differences between arms (in favour of ripretinib) in age, ECOG PS, and tumour site were not considered likely to affect the study results by clinical experts consulted by CADTH for this review. Because of elective crossover of patients from DB placebo to OL ripretinib 150 mg once daily, as well as elective intra-patient post-progression dosage escalation to 150 mg twice daily (a dosage not approved by Health Canada), the relative impacts of ripretinib versus placebo treatment, pre- versus post-progression treatment, and ripretinib dose on OS could not be ascertained from the study data. Early failure of the statistical hierarchy at the time of the primary analysis precluded testing of OS and HRQoL differences between arms. Analyses of HRQoL outcomes were further limited by missing data and uncertainty regarding the measurement properties or minimal important differences (MIDs) of the instruments used in GIST patients. Changes in GIST symptoms were not directly assessed in the study.

The demographic and disease characteristics of the INVICTUS study population were considered broadly reflective of the Canadian population with advanced GIST who would be eligible for ripretinib, according to the clinical experts consulted by CADTH for this review. There were no major generalizability concerns to smaller subgroups (e.g., tumour mutational status). However, the descriptive OS results should not be generalized to clinical practice, due to inability to account for patient crossover from placebo to ripretinib, post-progression OL treatment, and dosage escalation during OL treatment to 150 mg twice daily. The impact of crossover from placebo to OL ripretinib would be expected to bias OS comparisons against ripretinib, while the impacts of post-progression treatment and intra-patient dosage escalation were uncertain.

Indirect Comparisons

No indirect evidence was identified for this review.

Other Relevant Evidence

No other relevant evidence was identified for this review.

Conclusions

Evidence from the INVICTUS study suggested that administration of ripretinib in patients with advanced GIST who had previously received imatinib, sunitinib, and regorafenib was associated with statistically significant and clinically meaningful prolongation of PFS compared with placebo. Administration of ripretinib also resulted in numerically higher ORRs compared with placebo, although this difference was not statistically significant at the 0.05 level at the time of the primary analysis. OS was numerically longer in patients randomized to receive ripretinib compared with those randomized to receive placebo. However, differences in OS between patients randomized to receive ripretinib versus placebo were not tested statistically due to early failure of the statistical hierarchy, precluding definitive conclusions. Changes in patient-reported HRQoL (EORTC QLQ-C30, EQ-5D-5L) following ripretinib administration were difficult to interpret due to absence of formal statistical testing, missing data, wide variation in estimates, and uncertainty regarding HRQoL measurement properties in GIST patients. Ripretinib was generally well tolerated in most patients, and its notable harms were considered expected and acceptable by patients and clinicians. The observed PFS benefits, consistent numeric improvements in other efficacy outcomes, and acceptable toxicity profile in the study were aligned with outcomes identified as important to patients with advanced GIST who currently have no treatment options available.

Introduction

Disease Background

Gastrointestinal stromal tumours (GISTs) are the most common soft-tissue sarcomas (rare cancers of mesenchymal cells) of the gastrointestinal tract.¹ GISTs occur more often in older individuals; they are found equally in both genders.¹⁴ Tumours arise in interstitial cells of Cajal of the stomach, or less commonly, the small intestine, esophagus, or other locations in the gastrointestinal tract (e.g., the rectum, colon, or mesentery without bowel wall primary).¹⁵ GISTs are typically characterized by primary gain-of-function mutations in CD117/c-KIT (80% to 85% of GISTs), especially exons 9 and 11²; mutations in PDGFRA are less frequent (5% to 10% of GISTs).³ Small and/or slow-growing GISTs may be benign, while tumours that grow significantly outward from the bowel wall can cause dysphagia, bleeding, abdominal pain/discomfort, fatigue, vomiting, loss of appetite, and other gastrointestinal issues.⁴ Both disease symptoms and the side effects of therapy severely affect HRQoL.

Data on the prevalence, incidence, and survival of advanced GIST in Canada are unavailable. At diagnosis, approximately half of patients with GISTs are eligible for potentially curative surgical resection.³ Among patients undergoing resection, 5-year survival is approximately 54% and disease-free survival is approximately 45%.⁵ Based on an estimated 500 GIST cases diagnosed per year in Canada,⁷ with 75% representing advanced GIST (50% at diagnosis and 25% recurrent disease following resection) and assuming 80%, 70%, and 60% failure/progression rates on imatinib, sunitinib, and regorafenib, respectively, the sponsor estimated a target population of 62 to 86 patients with advanced GIST per year from 2023 to 2025 in Canada (outside of Quebec) who have received prior treatment with imatinib, sunitinib, and regorafenib and would be eligible for fourth-line ripretinib.⁸ Following progression or failure on third-line regorafenib, patient outcomes are dismal: further progression and death can be expected within several months.

Diagnosis of GIST is typically made by a gastroenterologist or surgeon based on endoscopic biopsy, pathology, and imaging findings. Few tumours are identified incidentally, and screening programs are rare in North America.¹⁶ Patients with advanced disease are transferred to the care of a medical oncologist for systemic therapy with palliative intent. Tumour mutational testing may be conducted at diagnosis or following progression after earlier lines of therapy (depending on the centre), which may influence treatment sequencing (e.g., avapritinib first line for patients with PDGFRA D842V).

Standards of Therapy

Clinical experts consulted by CADTH for this review stated that localized GIST can be treated by resection followed by adjuvant imatinib in high-risk patients,¹⁷ while locally advanced disease may require a neo-adjuvant approach. In the approximately 50% of patients with metastatic/unresectable GIST at diagnosis, as well as the 25% of GIST patients who undergo resection and subsequently experience recurrence, radiotherapy and chemotherapy are ineffective.^1,6 Four TKIs are indicated for the treatment of patients with advanced/unresectable GIST: imatinib, sunitinib, regorafenib, and ripretinib.^1,6 Another TKI, avapritinib, is not approved in Canada but may be obtained through special access programs for early-line (typically first) therapy of patients with a rare PDGFRA D842V mutation that is found in approximately 5% of GISTs. Imatinib (approved in 2003) is not generally curative but is associated with high ORRs^3,18; progression typically occurs following secondary mutations in the KIT kinase domain.¹⁹ Sunitinib (approved in 2006) is used in patients with GIST who progressed on or were intolerant to imatinib; although the drug offers clinical benefit for the subset of patients with imatinib-induced secondary KIT mutations that do not confer sunitinib resistance, most patients will relapse within 6 months to 1 year following additional KIT mutations.²⁰ Regorafenib (approved in 2013) is used as third-line therapy for patients who progressed on or were intolerant to imatinib and sunitinib.^21,22 Despite providing clinical benefit for patients with certain secondary KIT mutations, responses are generally brief, and relapse generally occurs within 6 months. Additional KIT secondary mutations occur following third-line regorafenib treatment (particularly in exons 13 and 17) and contribute to resistance.²¹ The clinical experts consulted by CADTH stated that, in most patients, metastatic disease would be treated sequentially with imatinib, sunitinib, and regorafenib; disease control following TKI administration ameliorates symptoms, including pain, obstruction, and bleeding.

Following development of resistance to imatinib, sunitinib, and regorafenib, rechallenge with previously failed TKIs may offer minor survival benefits,²³ but rechallenge is not currently a fourth-line treatment option in Canada. Ripretinib (approved in 2020) is not currently funded in Canadian jurisdictions except through special access programs. Thus, experimental drugs and/or BSC (e.g., analgesics, laxatives, antidiarrheals, antiemetics, and antibiotics) are the only options remaining to patients. According to the clinical experts, the goals of fourth-line treatment of advanced GIST following progression/intolerance to imatinib, sunitinib, and regorafenib are prolonging survival, delaying disease progression, palliating symptoms, or preventing new symptom development.

Drug

Ripretinib is a switch-control TKI that broadly inhibits signalling of KIT, PDGFRA, and other kinases (wild-type and multiple primary and secondary mutations) by binding to both the switch pocket and the activation loop, locking the kinase in an inactive state. Key characteristics of ripretinib are listed in Table 3. The drug is administered at a dosage of 150 mg orally (three 50 mg tablets) once daily. Ripretinib is indicated “for the treatment of adult patients with advanced gastrointestinal stromal tumour (GIST) who have received prior treatment with imatinib, sunitinib, and regorafenib.”²⁴ The drug has not been previously reviewed by CADTH. The sponsor’s reimbursement request is aligned with the Health Canada–approved indication. The drug underwent expedited review by Health Canada under Project Orbis.

Stakeholder Perspectives

Patient Group Input

This section was prepared by CADTH staff based on the input provided by patient groups. The original patient group submission can be found in Appendix 1.

Two patient groups submitted patient input for this review: the CanCertainty Coalition and the GIST Sarcoma Life Raft Group Canada (LRGC). CanCertainty raised concerns about the financial and administrative barriers to accessing cancer treatments such as ripretinib in some Canadian jurisdictions (Ontario and the Atlantic provinces). Young patients (under the age of 65) who require take-home cancer treatment such as ripretinib and do not have private or automatic public prescription drug coverage may incur significant deductibles or co-payments from their personal savings. These costs can become a financial burden and may lead to distress and hardship for Canadian GIST patients under 65 without private drug coverage (estimated at approximately 5 patients per year).

In September and October 2021, LRGC conducted telephone interviews of 11 patients with advanced GIST (5 Canadian and 6 from the US) who had experience with ripretinib. All respondents were either initially or eventually diagnosed with metastatic GIST and many had experienced delays in diagnosis due to nonspecific symptoms. Patients highlighted the negative impacts of advanced GIST on HRQoL, including symptoms of vomiting, abdominal pain/discomfort, and bowel issues, including diarrhea, severe fatigue, black stools, and loss of appetite. Patients had received 1 to 4 lines of therapy before ripretinib, and several recounted their rapid progression and sometimes severe side effects during treatment with prior TKIs. Most patients conveyed that ripretinib was generally more tolerable than other TKIs, with milder and acceptable side effects that included hair loss, cramping in body extremities, nausea, fatigue, hand-and-foot syndrome, foot calluses, and curly/kinky hair regrowth. More than half of patients reported improved HRQoL during ripretinib treatment compared with prior TKIs.

Patients with advanced GIST identified an unmet need for novel therapies that can stabilize or enhance HRQoL while effectively reducing disease progression for several years. In addition to improved survival, patients desired access to new drugs that have improved toxicity profiles and longer-term effectiveness, and that can target specific GIST mutations.

Clinician Input

Input From Clinical Experts Consulted by CADTH

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

Unmet Needs

According to the clinical experts consulted by CADTH, not all patients with metastatic GIST respond to available treatments (TKIs). The clinical experts noted that, among responding patients, responses are generally short-lived and become shorter with advancing lines of therapy (e.g., 7 months in second line, 5 months in third line). Eventually, almost all patients will develop refractory disease. Following exhaustion of available TKIs (imatinib, sunitinib, and regorafenib), there are no standard treatment options available in Canada. An additional line of therapy is required to fulfill the unmet needs of these patients.

Place in Therapy

The clinical experts consulted by CADTH for this review stated that, based on currently available evidence, ripretinib would be used as per the Health Canada indication for fourth-line monotherapy after progression on or intolerance to imatinib, sunitinib, and regorafenib. The clinical experts noted that, although use of avapritinib might, in theory, be an option for later-line treatment of patients with PDGFRA D842V mutations, it would be much more common for these patients to receive avapritinib in earlier lines of therapy. According to the clinical experts, ripretinib might theoretically be efficacious in earlier lines of therapy, but this has yet to be determined; trials evaluating ripretinib for second-line therapy are under way. The clinical experts noted that the mechanism of kinase inhibition by ripretinib is unique compared with that of other TKIs (binding to the switch conformation area rather than the adenosine triphosphate binding pocket). Based on the currently available data, clinical experts did not expect that funding of ripretinib would result in a treatment paradigm shift but rather would provide an additional option for later-line therapy that could be offered to patients who currently have no efficacious treatment available to them.

Patient Population

According to the clinical experts consulted by CADTH for this review, there are no established biomarkers of response to ripretinib. Thus, all patients with advanced GIST who experienced progression or intolerance on imatinib, sunitinib, and regorafenib with ECOG PS 0 to 2 and adequate organ and hematological function would be candidates for ripretinib. The clinical experts noted that most patients evaluated in the pivotal trial for ripretinib approval had KIT mutations, while only a few had PDGFRA mutations or wild-type tumours. However, patients would be candidates for ripretinib regardless of mutational status, and mutational analysis would not be a requirement for treatment. Similarly, the clinical experts stated that patients would be candidates for ripretinib regardless of tumour location. Patients with poor PS (e.g., ECOG PS > 2) or inadequate organ/hematological function, patients with significant comorbidities (e.g., class 2 to class 4 heart failure, cerebrovascular accident within 6 months, or venous thromboembolism within 3 months); patients who cannot take or absorb oral medications (e.g., due to bowel obstruction); and patients with central nervous system metastases would be least suitable for treatment with ripretinib treatment. The clinical experts stated that diagnosis of GIST through biopsy and pathologic examination is straightforward, and that misdiagnosis or diagnostic delays are not major issues. The clinical experts noted that GIST patients with progressive disease (PD) who are candidates for later-line therapy would generally be symptomatic. However, if patients remain asymptomatic following failure of prior TKIs, either immediate ripretinib treatment or a break in treatment until symptoms recur would be reasonable options.

Assessing Response to Treatment

According to clinical experts consulted by CADTH for this review, response to ripretinib treatment would be assessed via clinical evaluation (every 2 weeks to 4 weeks) and by CT or MRI scans (every 2 weeks to 4 months). Clinically meaningful responses to therapy would be reflected by prolonged OS and PFS, maintained or improved HRQoL and PS, and stabilized or reduced symptom severity (e.g., pain, bowel transit problems, liver/biliary obstruction, and bleeding).

Discontinuing Treatment

According to the clinical experts consulted by CADTH for this review, ripretinib treatment would be discontinued because of disease progression (either clinical symptomatic progression or radiographic progression), significant AEs (e.g., Grade ≥ 3 elevated liver enzymes), persistent treatment intolerance despite dosage reductions, or patient preference.

Prescribing Conditions

According to the clinical experts consulted by CADTH for this review, GIST diagnosis is typically made by a gastroenterologist or surgeon, but treatment with ripretinib in the fourth-line setting would be handled by a medical oncologist. Appropriate treatment setting would include outpatient oncology clinics (community or academic setting).

Additional Considerations

The clinical experts consulted by CADTH for this review noted the convenience of ripretinib as an oral drug that can be self-administered at home, minimizing clinic visits and potential hospitalizations in this advanced disease setting, and emphasized the absence of alternative viable options for fourth-line GIST therapy.

Clinician Group Input

This section was prepared by CADTH staff based on the input provided by clinician groups. The original clinician group submission can be found in Appendix 2.

One group of 7 Canadian medical oncologists who treat patients with advanced GIST provided input for this review; some of the oncologists are medical advisors to LRGC. No major contrary views were presented. The clinicians echoed the absence of fourth-line treatment options for patients after available TKIs (imatinib, sunitinib, and regorafenib) have been exhausted and the poor outcomes in these patients. Minor discrepancies were noted between the clinical experts and the clinician group input in the frequency of response assessment by imaging scans (2 to 3 months versus 3 to 4 months), possibly due to jurisdictional variation.

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.

Clinical Evidence

The clinical evidence included in the review of ripretinib is presented in a single section (the Systematic Review) that 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. No indirect evidence met the inclusion criteria for this review. No additional relevant studies were identified that were considered to address important gaps in the evidence included in the systematic review.

Systematic Review (Pivotal and Protocol-Selected Studies)

Objectives

To perform a systematic review of the beneficial and harmful effects of ripretinib (150 mg orally once daily) for the treatment of adult patients with advanced GIST who have received prior treatment with imatinib, sunitinib, and regorafenib.

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.²⁵

Published literature was identified by searching the following bibliographic databases: MEDLINE All (1946—) via Ovid and Embase (1974—) via Ovid. The search strategy comprised both controlled vocabulary, such as the National Library of Medicine’s MeSH (Medical Subject Headings), and keywords. The main search concepts were Qinlock/ripretinib. Clinical trials registries were searched: the US National Institutes of Health’s clinicaltrials.gov, WHO’s International Clinical Trials Registry Platform (ICTRP) 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. See Appendix 3 for the detailed search strategies. The initial search was completed on November 12, 2021. Regular alerts updated the search until the meeting of the CADTH pan-Canadian Oncology Drug Review Expert Committee (pERC) on March 9, 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.²⁶ 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. Refer to Appendix 3 for more information on the grey literature search strategy.

A focused literature search for network meta-analyses dealing with Qinlock/ripretinib was run in MEDLINE All (1946–) on November 12, 2021. No limits were applied.

These searches were supplemented by reviewing bibliographies of key papers and through contacts with appropriate experts. Two CADTH clinical reviewers independently selected studies for inclusion in the review based on titles and abstracts, according to the predetermined protocol. Full-text articles of all citations considered potentially relevant by at least 1 reviewer were acquired. Reviewers independently made the final selection of studies to be included in the review, and differences were resolved through discussion.

Findings From the Literature

Four reports of a single study^9-12 were identified from the literature for inclusion in the systematic review (Figure 1). The included study is summarized in Table 6.

Figure 1: Flow Diagram for Inclusion and Exclusion of Studies

Description of Studies

INVICTUS was a phase III, DB, placebo-controlled multi-centre RCT (N = 129) with an OL period of active treatment.^9-11 The study was funded by Deciphera Pharmaceuticals; an exclusive agreement was subsequently reached between Deciphera and the sponsor, Medison Biopharma Canada Inc., to commercialize and distribute ripretinib in multiple regions, including Canada. The primary objective of the study was to assess the efficacy of ripretinib in prolonging PFS per IRR in patients with advanced GIST who had received prior anticancer therapies, including imatinib, sunitinib, and regorafenib. Secondary objectives included comparing ORR per IRR (hierarchically tested to control for type I error), OS (hierarchically tested), patient-reported changes in disease symptoms and HRQoL from baseline to the start of cycle 2 (hierarchically tested), time to response (TTR) per IRR, DOR per IRR, and time to progression per IRR between the ripretinib and placebo arms. A summary of the design of the INVICTUS study is shown in Figure 2. Patients aged ≥ 18 years with inoperable advanced GIST who had progressed or become intolerant to all 3 prior TKIs and ECOG PS 0 to 2 were enrolled from 27 February 2018 until 15 November 2019 at 29 sites in 12 countries (including 1 site in Toronto, Canada); patients with active central nervous system metastases, clinically significant cardiac conditions or other comorbidities, and gastrointestinal problems preventing absorption of medication were excluded. Patients were screened for eligibility within 4 weeks of starting protocol therapy.

Patients were randomized 2:1 using an interactive response technology system to receive either: (i) ripretinib 150 mg orally once daily plus BSC or (ii) placebo orally once daily plus BSC, both on 28-day cycles. The randomization algorithm/procedure was not explicitly stated. Randomization was stratified by number of prior lines of therapy (3 versus ≥ 4; enrolment of patients who had received ≥ 4 prior lines of therapy was capped at 40%) and ECOG PS (0 versus 1 or 2). During the DB period, patients were treated until the first designation of PD by investigator assessment and/or IRR, unacceptable toxicity, withdrawal by patient or physician, or death, whichever came first. Following initial objective PD per IRR, patients and investigators were unblinded to treatment allocation, patients randomized to receive ripretinib 150 mg once daily during the DB period could choose from the following OL treatment options: continue ripretinib 150 mg once daily, escalate the dosage to ripretinib 150 mg twice daily, or discontinue ripretinib. Following initial progression, patients randomized to receive placebo during the DB period could choose from the following OL treatment options: (i) crossover to ripretinib 150 mg once daily or (ii) discontinue the study. Patients initially randomized to the placebo arm who chose to cross over to ripretinib 150 mg following initial progression and thereafter were designated with a second objective PD per IRR could choose from the following OL treatment options: continue ripretinib 150 mg once daily, escalate the dosage to ripretinib 150 mg twice daily, or discontinue ripretinib. The 150 mg twice daily dosage is not aligned with the Health Canada–approved dosage, and data for patients receiving this dosage are not presented in this report.

Following treatment discontinuation, patients entered survival follow-up (every 3 months until study withdrawal, death, or data cut-off, whichever came first). The database was closed on May 31, 2019, and these data were used for regulatory approval. Data for a second database lock of January 15, 2021, were made available in the CADTH review submission.

Figure 2: Design of the INVICTUS Study

BID = twice daily; ECOG PS = Eastern Cooperative Oncology Group performance status; QD = once daily; vs = versus.

Source: INVICTUS Clinical Study Report.¹²

Populations

Inclusion and Exclusion Criteria

Key inclusion and exclusion criteria for the INVICTUS study are summarized in Table 6. Adult patients (age ≥ 18 years) with inoperable, advanced, histologically confirmed GIST and ECOG PS 0 to 2 were eligible if they had experienced PD on imatinib, sunitinib, and regorafenib or had documented intolerance to these TKIs. Patients had to have ≥ 1 measurable lesion by modified Response Evaluation Criteria in Solid Tumours (mRECIST) version 1.1 – GIST-specific, adequate hematological and organ function, and have recovered from any toxicity (to Grade ≤ 1) from prior therapies. Patients previously treated with ripretinib and patients with active central nervous system metastases, cardiac conditions (e.g., class II to IV heart disease or active ischemia), recent arterial thrombotic or embolic events, recent venous thrombotic events or pulmonary arterial events, electrocardiogram abnormalities, reduced ejection fraction, clinically significant comorbidities, or active bleeding were excluded. Use of proton pump inhibitors or other drugs that increase gastric pH was prohibited, as was use of drugs or substances with known drug-drug, drug-herb, or drug-food interactions with TKIs. Patients with gastrointestinal abnormalities that would impair taking or absorbing oral medications were excluded.

Baseline Characteristics

The baseline demographic characteristics of patients in the INVICTUS study are shown in Table 7. In the placebo and ripretinib arms, 59.1% and 55.3% of patients were men, respectively, and the mean ages were 62.0 and 59.1 years, respectively. A higher proportion of patients in the ripretinib arm were younger than 65 years (67.1%) compared with those in the placebo arm (50.0%). Approximately 75% of patients were White, approximately 90% were non-Hispanic, and approximately half were from the US. The proportions of patients with ECOG PS 0, 1, or 2 were 38.6%, 54.5%, 6.8%, respectively, in the placebo arm, and 43.5%, 47.1%, or 9.4%, respectively, in the ripretinib arm. Approximately 60% of patients had received 3 prior lines of therapy, while approximately 40% had received 4 or more prior lines of therapy. Baseline demographic characteristics were generally well balanced between study arms during the DB period, apart from minor imbalances in age and ECOG PS.

The baseline disease characteristics of patients in the INVICTUS study are shown in Table 8. The most common site of the primary tumour was gastric (40.9% of the placebo arm and 47.1% of the ripretinib arm). The most common location of primary tumour mutations was KIT exon 11 (63.6% of the placebo arm and 55.3% of the ripretinib arm). Few patients had PDGFRA mutations (none in the placebo arm and 3.5% of the ripretinib arm), and a small subset had wild-type KIT and PDGFRA genes (6.8% of the placebo arm and 8.2% of the ripretinib arm). Approximately two-thirds of patients had stage IV disease at diagnosis. The most common histologic type at diagnosis was spindle cell (70.5% of the placebo arm and 43.5% of the ripretinib arm). The mean time elapsed since diagnosis was approximately 7 years. Approximately 80% of patients had undergone prior surgery for GIST, while approximately 20% had received radiotherapy. Among patients who had received 4 or more prior lines of therapy, the most common drugs (other than imatinib, sunitinib, and regorafenib) were sorafenib, pazopanib, and nilotinib. Baseline disease characteristics were generally well balanced between study arms, apart from minor imbalances in tumour site and histology.

Interventions

During the DB period, patients were randomized 2:1 to receive either ripretinib 150 mg orally once daily plus BSC or placebo orally once daily plus BSC, both on repeated 28-day cycles. Both ripretinib and placebo were supplied as identically sized, shaped, and coloured tablets (ripretinib: 50 mg per tablet). Concomitant use of medications for symptomatic relief (e.g., analgesics, laxatives, antiemetics) was permitted. Medications that increased gastric pH (e.g., antacids) other than proton pump inhibitors were permitted provided they were not administered within 2 hours before or after administration of study drug. Proton pump inhibitors, strong or moderate inhibitors or inducers of cytochrome P450 (CYP)3A4, grapefruit or grapefruit juice, substrates or inhibitors of breast cancer resistance protein, and other systemic anticancer therapies (including investigational therapies) were prohibited. Patients were cautioned against taking strong or moderate inhibitors or inducers of CYP2D6, CYP2C8, or CYP2E1; substrates or inhibitors of P-glycoprotein 1; substrates of organic anion transporter polypeptides 1B1 and 1B3; and medications dependent on CYP2C8, CYP2C9, CYP2C19, or CYP2D6 for their metabolism. Patients taking any of these drugs were closely monitored for drug-drug interactions.

Treatment was continued until initial PD (either investigator assessed and/or per IRR), unacceptable toxicity, withdrawal by patient or physician, death, pregnancy, loss to follow-up, or nonadherence to study drug. Treatment was discontinued for the following toxicities: Grade 4 dermatologic toxicities or arthralgia/myalgia (unless not life threatening), Stevens-Johnson syndrome of any grade, Grade 4 hypertension, and clinically significant Grade 3 or higher laboratory AEs, including creatine phosphokinase and lipase elevation. In addition, treatment could be interrupted for no more than 1 cycle (28 days) and/or the dosage could be reduced stepwise (first reduction: 100 mg once daily; second reduction: 50 mg once daily) for the following toxicities: any Grade ≥ 3 toxicity, Grade 2/3 dermatologic toxicities or arthralgia/myalgia, symptomatic Grade 3 hypertension, and asymptomatic/not clinically significant Grade 3 or higher laboratory AEs that persist for more than 10 days. Per protocol, attempts were made to re-escalate to the dosage level at which toxicity occurred. If the AE returned to Grade 1 or baseline after dosage reduction, the patient could restart at the next higher step and remain at this dosage level for 1 cycle without interruption before escalating to the starting dosage level (if required). If treatment was delayed longer than 28 days or patients required dosages lower than 50 mg once daily, treatment was discontinued.

Following initial PD, patients randomized to receive placebo during the DB period could choose to cross over and receive ripretinib 150 mg once daily plus BSC. Patients randomized to receive ripretinib 150 mg once daily during the DB period, as well as the subset of patients initially randomized to receive DB placebo who crossed over to OL ripretinib 150 mg once daily and developed subsequent PD per IRR, were eligible to escalate the dosage to 150 mg twice daily plus BSC (at the first and second designation of PD, respectively).

Outcomes

A list of efficacy and safety end points identified in the CADTH review protocol that were assessed in INVICTUS study is provided in Table 9. These end points are further summarized in this section.

OS, ORR, DOR, TTR, and PFS are standard and accepted outcome measures in oncology trials. A detailed discussion and critical appraisal of the HRQoL measures used in the INVICTUS study (EORTC QLQ-C30²⁸ and EQ-5D-5L²⁹) is provided in Appendix 2. Neither the measurement properties nor the MID of either instrument have been specifically evaluated in patients with GIST. Among all cancer patients, the measurement properties of these instruments have been investigated. Differences of approximately 10 points in EORTC QLQ-C30 individual items and scale scores (range: 0 to 100, with higher scores on function scales reflecting better function and higher scores on symptom scales reflecting increased symptoms), approximately 7 to 12 points in EQ-VAS scores (range: 0 to 100, with 0 and 100 representing “worst imaginable health” and ”best imaginable health,” respectively), and approximately 0.07 to 0.12 (using the UK algorithm) in EQ-5D-5L utility index scores (range: < 0 to 1, with 0 and 1 representing the health states “dead” and ”perfect health,” respectively) are typically considered significant. MIDs for EQ-5D-5L descriptive system dimension scores (range: 1 to 5, representing no problems, slight problems, moderate problems, severe problems, and extreme problems, respectively) are uncertain.

During treatment, tumour response was assessed by CT or MRI at screening (within 21 days of cycle 1 day 1) and then on day 1 of each cycle until cycle 4. Subsequently, pelvic and abdominal scans were performed every other cycle, and chest scans were performed only for patients with lung metastases at baseline or with lung symptoms. The same imaging modality was used for each patient throughout the study. Copies of all scans were sent for blinded IRR by an independent radiologist.

Tumour response was assessed using mRECIST version 1.1 – GIST-specific. No lymph nodes were chosen as target lesions, and enlarged lymph nodes were followed as nontarget lesions. No bone lesions were chosen as target lesions. Partial response (PR) was defined as at least a 30% decrease in the sum of diameters of target lesions, taking as reference the baseline sum diameters. Complete response (CR) was defined as disappearance of all target lesions with reduction of the short axis of any pathological lymph nodes to less than 10 mm. PD was defined as a predefined increase ( + 20%), taking as reference the smallest sum on study, in the sum of target lesions or the appearance of new nontarget lesions; the sum must also demonstrate an absolute increase of at least 5 mm. A progressively growing new tumour nodule within a pre-existing tumour mass was considered unequivocal evidence of progression if the lesion was at least 2 cm in size and definitively a new active GIST lesion or the lesion was expanding on at least 2 sequential imaging studies. Stable disease was defined as neither sufficient shrinkage (compared to baseline) to qualify for PR nor sufficient increase (taking as reference the smallest sum diameters while on study) to qualify for PD.

An initial indication of PR or CR per investigator assessment was confirmed 4 or more weeks later, following IRR. The investigator decided whether to discontinue protocol therapy due to PD based on local imaging scans and clinical evaluation in conjunction with IRR. If the IRR confirmed no disease progression, the patient continued to receive study drug unless there was a medical need (i.e., rapid progression or clinical deterioration requiring treatment discontinuation). For investigator-determined progression based on clinical deterioration, a scan was performed and sent for IRR to determine PD. Following confirmation of PD by IRR, the patient’s treatment assignment was unblinded, and the patient’s dosage could be escalated or the patient could cross over.

During treatment, patient-reported HRQoL instruments (EORTC QLQ-C30 and EQ-5D-5L) were completed on days 1 and 15 of cycle 1, on day 1 of each cycle thereafter, and at the end-of-treatment visit. Following treatment discontinuation, an end-of-treatment visit within 7 days of the final dose included clinical evaluation, imaging scans, and completion of HRQoL instruments. Thereafter, patients entered survival follow-up every 3 months (by phone).

All efficacy outcomes, except OS and PFS, were evaluated during the DB treatment period. OS was censored at the date patients were last known alive but not for patients who crossed over from placebo to OL ripretinib and/or who had a dosage escalation to 150 mg twice daily during the OL treatment period. For patients randomized to receive placebo during the DB period who had objective PD per IRR and chose to cross over to ripretinib, second PFS was defined as the interval between the date of the first ripretinib dose and the first subsequent objective PD per IRR or death, whichever came first. See the footnotes to Table 9 for PFS censoring rules.

Harms outcomes included treatment-emergent AEs, serious AEs, AEs requiring dosage interruption or reduction, withdrawals due to AE, and AEs of special interest (AESIs). AESIs were squamous cell carcinoma, actinic keratosis, and keratoacanthoma. AEs that began or worsened on or after the start of protocol therapy until 30 days after the last dose of study drug were captured. AEs were defined as any untoward medical occurrence and were coded according to Medical Dictionary for Regulatory Activities (MedDRA) version 21.1³⁰ and graded according to the National Cancer Institute Common Terminology Criteria for Adverse Events (NCI-CTCAE) version 4.03.³¹

Statistical Analysis

Statistical analysis of efficacy outcomes in the INVICTUS study is summarized in Table 10. No interim analyses were planned or conducted. The trigger for the final primary analysis (May 31, 2019, database lock) was the occurrence of 90 PFS events, which was expected to occur approximately 6 months after the last patient was enrolled in the trial. An unplanned analysis of a more recent data cut (January 15, 2021) was also made available for CADTH review.

The planned sample size of 120 patients (80 ripretinib and 40 placebo) was based on the primary efficacy analysis of PFS per IRR during the DB treatment period, as well as the key secondary outcome of ORR per IRR and safety assessment, assuming 15% patient dropout. In the resulting final sample of 105 patients (n = 70 ripretinib, n = 35 placebo), it was assumed that 90 PFS events would occur (55 ripretinib, 35 placebo) in the final analysis. This would yield at least 90% power to detect a difference in PFS (by log-rank test, refer to following discussion), assuming median PFS of 4.5 months for ripretinib and 1 month for placebo, 9 months of uniform recruitment, and 6 months of additional follow-up (maximum patient follow-up 15 months). This sample size was expected to yield approximately 80% power to detect a 20% difference in ORR (by Fisher’s exact test, refer to following discussion), assuming an ORR of 22% for ripretinib and 2% for placebo.

Type I error was controlled using a hierarchical testing strategy. Hypothesis tests for treatment differences were performed at a 2-sided alpha significance level of 0.05 sequentially, in the following order: (1) primary analysis of PFS per IRR, (2) key secondary analysis of ORR per IRR, (3) analysis of OS, and (4) analysis of HRQoL reflected in changes from baseline to cycle 2 day 1 in EORTC QLQ-C30 role function and physical function (each at a 0.025 significance level). Once a hypothesis test was nonsignificant at the 0.05 level, the remaining analyses were viewed as descriptive.

For the primary PFS analysis, PFS per IRR was compared in the intention-to-treat (ITT) population during the DB period between the ripretinib and placebo arms using a 2-sided stratified log-rank test. The HR comparing treatment arms was calculated using stratified Cox proportional hazard models with treatment and randomization stratification factors as fixed factors. Stratification factors for these analyses were those used for randomization (3 versus ≥ 4 prior systemic therapies and ECOG PS 0 versus 1 or 2). 95% CIs for the HR were calculated using the Wald method. The PFS function was estimated using the Kaplan-Meier (KM) product limit method. PFS time (25th percentile, median, 75th percentile) and PFS rates at pre-specified time points (26, 39, and 52 weeks), each with 2-sided 95% CIs, were derived from KM analysis. Sensitivity analyses for the primary PFS analysis included analysis using randomization stratification factor values collected on the electronic case record form instead of the interactive response technology, analysis in the per-protocol and safety populations instead of the ITT population, and analysis of PFS per investigator assessment.

For the key secondary ORR analysis, ORR per IRR was calculated in the ITT population during the DB period in each treatment arm along with the exact binomial 95% CI. ORRs were compared between treatment arms using 2-sided Fisher’s exact test. The ORR difference between treatment arms was calculated by simple subtraction and a 95% Newcombe score CI was constructed. Sensitivity analyses of the key secondary ORR analysis included analysis in the per-protocol population instead of the ITT population and analysis of ORR per investigator assessment.

For the secondary analysis of changes from baseline to cycle 2 day 1 of the DB period in EORTC QLQ-C30 role and physical function in the ITT population (included in the statistical hierarchy), a stratified analysis of covariance (ANCOVA) model was used to calculate adjusted mean changes from baseline and to compare changes from baseline between treatment arms. The ANCOVA model included randomization stratification factors as factors (3 versus ≥ 4 prior systemic therapies and ECOG PS 0 versus 1 or 2). Analysis of changes from baseline to cycle 2 day 1 of the DB period in other HRQoL outcomes outlined in the remainder of this paragraph were not adjusted for multiplicity. For changes from baseline in EQ-5D-5L pain/discomfort and usual activities, a Cochran-Mantel-Haenszel test was used to assess differences in response distribution between treatment arms. For index utility scores, a stratified ANCOVA model was used to calculated adjusted mean changes from baseline and to compare changes from baseline between treatment arms, as previously described. For changes from baseline in EQ-VAS scales, a t-test was used to assess differences in mean change from baseline between treatment arms.

Analysis of OS and time to progression as secondary outcomes during the DB period was performed in the same manner as the primary PFS analysis, as was analysis of second PFS during the OL treatment period. TTR was assessed using descriptive and summary statistics. DORs and their 95% CIs in each treatment arm were calculated using KM methodology. Of the analyses described in this paragraph, only the OS analysis was adjusted for multiplicity.

Subgroup analyses of PFS and ORR per IRR during the DB period were conducted for pre-specified subgroups (by age, gender, race, region, ECOG PS, and number of prior therapies), as per the primary and key secondary efficacy analysis, but in exploratory fashion. A post hoc subgroup analysis of ORR per IRR by tumour mutational status was also conducted. Data for a post hoc subgroup analysis of OS by treatment assignment in both the DB and OL periods was made available for CADTH review. The study was not specifically powered to evaluate outcomes in individual strata. Safety data were tabulated and presented using descriptive statistics (frequencies).

For time-to-event analyses (PFS, OS, DOR), missing data were accounted for by censoring. TTR was evaluated among responders and, thus, missing data were not possible. For analysis of ORR, patients without tumour response assessment were classified as nonresponders. For analysis of EORTC QLQ-C30 role and physical function, missing data for cycle 2 day 1 were replaced with data from the end of study treatment during the DB period.

Analysis Periods

The DB treatment analysis period was defined as 1 of the following intervals:

• from the randomization date to the last follow-up date if a patient did not have disease progression based on IRR, if a placebo patient did not cross over to ripretinib treatment, or if a patient was initially treated with ripretinib and did not continue to receive ripretinib at 150 mg once daily or escalate to 150 mg twice daily after disease progression based on IRR

• from the randomization date to the first disease progression date if a patient was initially treated with ripretinib and continued to receive ripretinib at 150 mg once daily after disease progression based on IRR

• from the randomization date to the day immediately before the first dosage of ripretinib 150 mg twice daily if a patient’s dosage was escalated to ripretinib 150 mg twice daily after disease progression based on IRR

• from the randomization date to the day immediately before the first dose of ripretinib 150 mg once daily if a patient crossed over from placebo to receive ripretinib at 150 mg once daily after disease progression based on IRR.

The OL treatment analysis period was defined as 1 of the following intervals:

• from the day immediately after the first disease progression based on IRR to the last follow-up date if a patient was initially treated with ripretinib and continued to receive ripretinib 150 mg once daily after disease progression based on IRR

• from the first dose date of ripretinib 150 mg twice daily to the last follow-up date if a patient was initially treated with ripretinib and the patient’s dosage was escalated to ripretinib 150 mg twice daily after disease progression based on IRR

• from the first dose date of ripretinib 150 mg once daily to the last follow-up date if a patient crossed over from placebo to receive ripretinib 150 mg once daily after disease progression based on IRR.

The OL treatment analysis period was further subdivided into 2 subperiods: before intra-patient dosage escalation and following intra-patient dosage escalation.

All efficacy outcomes were evaluated in the DB treatment period, with the following exceptions. OS was evaluated during the entire study period. Among patients who were randomized to receive placebo in the DB period and subsequently crossed over to ripretinib in the OL period following initial progression per IRR, second PFS was evaluated in the period before intra-patient dosage escalation.

Safety outcomes were evaluated separately during each analysis period (DB period and OL period). For the OL period, safety analyses were conducted separately for the 2 subperiods (before and following intra-patient dosage escalation).

Analysis Populations

The ITT population was defined as all patients who provided informed consent and were randomized. The per-protocol population was defined as randomized patients who did not have important protocol deviations that were expected to compromise efficacy and/or safety assessments, as follows: (i) patients with pre-specified inclusion/exclusion criteria deviations, (ii) patients who received the wrong treatment, (iii) patients who received an incorrect dose, and (iv) patients who received pre-specified prohibited medications. The safety population was defined as all patients who received at least 1 dose of study drug, and data were analyzed according to the treatment the patient actually received.

Results

Unless otherwise noted, data presented are for the primary analysis (May 31, 2019, database lock).

Patient Disposition

Patient disposition during the DB period of the INVICTUS study is summarized in Table 11. For disposition during the OL period, refer to Appendix 3. A total of 154 patients were screened, of whom 25 (16.2%) were screen failures. The remaining 129 patients were randomized to the ripretinib (n = 85) and placebo arms (n = 44) for DB treatment. As of the May 31, 2019, data cut, approximately 30% and 20% of patients in the placebo and ripretinib arms, respectively, had discontinued treatment. The most common reasons for treatment discontinuation were PD and death. Few patients (n = 2, 4.7% in the placebo arm and n = 3, 3.5% in the ripretinib arm) discontinued protocol therapy due to AEs. Approximately 32% and 18% of patients in the placebo and ripretinib arms, respectively, had discontinued the study. The most common reason for study discontinuation was death. Attrition due to losses to follow-up or withdrawal of consent were minimal.

As of the May 31, 2019, data cut, a total of 71 patients (55.0%), including 29 (67.4%) patients randomized to the placebo arm and 42 (49.4%) of patients randomized to the ripretinib 150 mg once daily arm for DB treatment, had entered the OL treatment period. Among the 29 patients initially randomized to the placebo arm for DB treatment who crossed over to ripretinib 150 mg once daily, 10 (22.7%) subsequently had an escalation in dosage to 150 mg twice daily, while 19 (43.2%) continued to receive the 150 mg once daily dosage. Among the 42 patients initially randomized to the ripretinib 150 mg once daily arm for DB treatment who chose to receive OL ripretinib, 31 (36.5%) had an escalation in dosage to 150 mg twice daily, while 11 (12.9%) continued to receive the 150 mg once daily dosage.

Table 11: Patient Disposition in the INVICTUS Study (DB Period)

DB = double-blind; IC = informed consent; IRR = independent radiological review; ITT = intention to treat; OL = open label; PD = progressive disease; PP = per protocol.

Note: Data are from the May 31, 2019, database lock.

^aDenominator is the safety population.

^bDenominator is the ITT population.

^cFor the OL period, all patients received assigned treatments per protocol.

Source: INVICTUS Clinical Study Report.¹²

Important protocol deviations during the DB treatment period of the INVICTUS study as of the May 31, 2019, data cut are summarized in Table 12. Important protocol deviations occurred in 2 (4.5%) patients randomized to the placebo arm and 4 (4.7%) patients randomized to the ripretinib 150 mg once daily arm.

Exposure to Study Treatments

Treatment exposure in the INVICTUS study is summarized in Table 13. Treatment adherence was assessed by site personnel at each site visit, both verbally and by ongoing study drug count. Patients were instructed to return all unused, partially used, and used study drug bottles to the site at each visit. As of the May 31, 2019, data cut, the mean treatment duration was shorter in patients who received placebo compared with those who received ripretinib 150 mg once daily during the DB period (8.25 and 24.44 weeks, respectively). On mean, patients who received ripretinib 150 mg once daily during the DB period had completed 6.11 cycles and achieved 96.5% dose intensity with 97.74% treatment adherence. Dose was modified in 21 (24.7%) of patients treated with ripretinib in the DB period, most commonly dose was interrupted (n = 18, 21.2%).

Table 13: Treatment Exposure in the INVICTUS Study (Safety Population)

DB = double-blind; OL = open label; q.d. = once daily; SD = standard deviation.

Note: Data are from the May 31, 2019, database lock.

^aCalculated as (date of last treatment – date of first treatment + 1)/7. For patients who entered the OL period, the end date of the DB treatment period is used as the date of last treatment for calculation.

^bCalculated as (date of last treatment – date of first treatment + 1)/28. Date of last treatment is defined as in footnote a.

^cCalculated as total dose (mg) / total planned dose (mg) × 100.

^dCalculated as total number of days dosed / treatment duration in days × 100. Treatment duration in days calculated as (date of last treatment – date of first treatment + 1) with date of last treatment defined as in footnote a.

Source: INVICTUS Clinical Study Report.¹²

Efficacy

Only those efficacy outcomes and analyses of subgroups identified in the review protocol are reported in this section. Refer to Appendix 3 for detailed efficacy data (sensitivity and subgroup analyses; second PFS among patients randomized to receive DB placebo who crossed over to ripretinib 150 mg once daily following initial progression per IRR).

Overall Survival

OS results are shown in Table 14, Figure 3, and Figure 4. This outcome was positioned in the statistical hierarchy after a nonsignificant result for a prior outcome (ORR) at the time of the primary analysis; the analysis was viewed as descriptive. OS was analyzed for the study as a whole and reflects both the DB and OL treatment periods, including patients who crossed over from placebo to ripretinib and patients who had an escalation in dosage to 150 mg twice daily for OL treatment. As of the May 31, 2019, database lock and in the ITT population, OS events had occurred in 26 (59.1%) of patients originally randomized to the placebo arm and 26 (30.6%) of patients originally randomized to the ripretinib arm. Median OS was 28.6 weeks (95% CI, 17.9 to 50.4 weeks) for patients originally randomized to the placebo arm and 65.6 weeks (95% CI, 53.6 to 65.6 weeks) for patients originally randomized to the ripretinib arm. The HR for OS comparing ripretinib with placebo was 0.36 (95% CI, 0.21 to 0.62). Results were similar for the January 15, 2021, database lock.

Figure 3: Kaplan-Meier Plot of OS in the INVICTUS Study (ITT Population; Data Cut: May 31, 2019)

CI = confidence interval; DCC-2618 = ripretinib; ITT = intention to treat; OS = overall survival; QD = once daily.

Note: Plus symbols represent censored observations.

Source: INVICTUS Clinical Study Report.¹²

Figure 4: Kaplan-Meier Plot of OS in the INVICTUS Study (ITT Population; Data Cut: January 15, 2021)

CI = confidence interval; DCC-2618 = ripretinib; ITT = intention to treat; OS = overall survival.

Note: Circle and plus symbols represent censored observations.

Source: Sponsor’s additional information.¹³

Refer to Appendix 3 for subgroup analyses of OS. A post hoc subgroup analysis of OS by combined treatment assignment in both the DB and OL periods showed the following results: DB placebo, no crossover (n = 14), median OS 7.9 weeks (95% CI, 3.7 to 19.6 weeks); DB placebo with crossover to OL ripretinib 150 mg once daily (n = 14), median OS 30.1 weeks (95% CI, 12.4 weeks to not calculable); and ||||| ||| |||||||||| ||| || || |||| || || ||||||||| ||||||| |||||| || |||| ||||| |||| ||| |||| ||||| || ||| |||||||||||| ||| |||| ||| |||||||||| ||| || || |||| || |||||||||| ||| || || ||||||| |||||| || ||||| ||||| |||| ||| |||| ||||| || ||| ||||||||||||

Health-Related Quality of Life

Changes in EORTC QLQ-C30 role and physical functioning scores from baseline to cycle 2 day 1 by treatment arm in the ITT are shown in Table 15. This outcome was positioned in the statistical hierarchy after a nonsignificant result for a prior outcome (ORR) at the time of the primary analysis; the analysis was viewed as descriptive. Baseline role functioning (placebo: mean = 73.8; ripretinib: mean = 69.4) and physical functioning (placebo: mean = 76.0; ripretinib: mean = 75.7) were similar in both treatment arms. The adjusted mean changes from baseline in role functioning were –17.1 (standard error [SE] = 5.0) in the placebo arm and 3.5 (SE = 3.5) in the ripretinib arm; the difference in adjusted mean changes from baseline comparing ripretinib with placebo was 20.61 (95% CI, 8.58 to 32.63). The adjusted mean changes from baseline in physical functioning were –8.9 (SE = 3.0) in the placebo arm and 1.6 (SE = 2.1) in the ripretinib arm; the difference in adjusted mean changes from baseline comparing ripretinib with placebo was 10.48 (95% CI, 3.37 to 17.59).

Table 15: EORTC QLQ-C30 Scores in the INVICTUS Study (ITT Population, DB Period)

CI = confidence interval; DB = double-blind; EORTC QLQ-C30 = European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire for Cancer 30 item; ITT = intention to treat; NA = not applicable; SD = standard deviation; SE = standard error.

Note: EORTC QLQ-C30 role and physical functioning are scored from 0 (worse HRQoL) to 100 (better HRQoL). Data are for the May 31, 2019, database lock.

^aEstimated from an ANCOVA model that included factors for study treatment, number of prior anticancer treatment, and ECOG status at baseline as fixed effects. Testing occurred after failure of the statistical hierarchy; therefore, the P value should be considered descriptive.

Source: INVICTUS Clinical Study Report.¹²

Changes in EQ-5D-5L usual activities and pain/discomfort domains, and index utility scores, from baseline to cycle 2 day 1 by treatment arm in the ITT are shown in Table 16. These outcomes were outside the statistical hierarchy and not adjusted for multiplicity. Baseline scores were generally similar between study arms, although the proportion of patients in the placebo arm who responded 1 (“no problems”) was higher in the placebo arm (47.7%) compared with the ripretinib arm (35.3%). The proportions of patients reporting improvements in usual activities were 6.8% in the placebo arm and 23.5% in the ripretinib arm; the proportions of patients reporting no change were 50.0% in the placebo arm and 47.1% in the ripretinib arm. The proportions of patients reporting improvements in pain/discomfort were 15.9% in the placebo arm and 17.7% in the ripretinib arm; the proportions of patients reporting no change were 36.4% in the placebo arm and 42.4% in the ripretinib arm. The adjusted mean changes from baseline in utility index scores were –0.0606 (SE = 0.02796) in the placebo arm and –0.0094 (SE = 0.01957) in the ripretinib arm; the difference in adjusted mean changes from baseline comparing ripretinib with placebo was 0.05 (95% CI, –0.02 to 0.12).

Table 16: EQ-5D-5L Scores in the INVICTUS Study (ITT Population, DB Period)

CI = confidence interval; DB = double-blind; ITT = intention to treat; SD = standard deviation; SE = standard error.

Note: Usual activities and pain/discomfort are scored on a 1 to 5 scale representing, no problems, slight problems, moderate problems, severe problems, and unable to/extreme problems, respectively. The index utility score ranges from < 0 (worse than dead) to 1 (perfect health). Data are for the May 31, 2019, database lock.

^aP value from Cochran-Mantel-Haenszel test.

^bDifference of placebo arm – ripretinib arm.

^cEstimated from an ANCOVA model that included factors for study treatment, number of prior anticancer treatment, and ECOG status at baseline as fixed effects. Analysis not adjusted for multiplicity.

Source: INVICTUS Clinical Study Report.¹²

Changes in EQ-VAS scores from baseline to cycle 2 day 1 by treatment arm in the ITT are shown in Table 17. This outcome was outside the statistical hierarchy, and the analysis was not adjusted for multiplicity. Baseline EQ-VAS scores were similar between treatment arms. The mean change from baseline in EQ-VAS score was –8.9 (standard deviation [SD] = 19.31) in the placebo arm and 3.7 (SD = 20.36) in the ripretinib arm. The difference in change from baseline between treatment groups was not reported.

Objective Response Rate

ORR (best overall response of CR or PR) per IRR in the ITT population is shown in Table 18. At the time of the primary analysis (as of the May 31, 2019, database lock) and in the ITT population, no patients in the placebo arm and 8 patients (9.4%) in the ripretinib arm had achieved objective responses (P = 0.0504). The ORR difference between arms was 9.4% (95% CI, 0.2% to 17.5%). The proportions of patients in the placebo and ripretinib arms with stable disease were 20.5% and 65.9%, respectively. Results were similar for the January 15, 2021, database lock.

Refer to Appendix 3 for sensitivity and subgroup analyses of ORR. Sensitivity analyses of ORR were consistent with the primary analysis. A planned subgroup analysis of ORR by number of lines of prior therapy was suggestive of similar response rates among patients in the ripretinib arm who had received 3 (5/54, 9.3%) and 4 or more (3/31, 9.7%) lines of prior therapy. A post hoc subgroup analysis of ORR by tumour mutational status was uninformative as all responses occurred in patients with mutations in KIT exon 11, the most common primary mutational type among patients in the study.

Duration of Response

DOR among patients responding to ripretinib is shown in Table 19. As of the 31 May 2019 database lock, only 1 responding patient in the ripretinib arm had progressed per IRR and the median DOR was not estimable. As of the January 15, 2021, database lock, median DOR in the ripretinib arm was 14.5 months (95% CI, 3.7 months to not estimable).

Time to Response

TTR among patients responding to ripretinib is shown in Table 20. As of the May 2019 database lock, the mean TTR in the ripretinib arm was 9.1 (SD = 5.83) weeks.

Progression-Free Survival

The results of the primary PFS analysis are shown in Table 21, Figure 5, and Figure 6. As of the May 31, 2019 database lock and in the ITT population, PFS events had occurred in 37 (84.1%) of patients randomized to the placebo arm and 51 (60.0%) of patients randomized to the ripretinib arm. Median PFS was 4.1 weeks (95% CI, 4.0 to 7.3 weeks) for patients randomized to the placebo arm and 27.6 weeks (95% CI, 20.0 to 29.9 weeks) for patients randomized to the ripretinib arm (P < 0.0001). The HR for PFS comparing ripretinib with placebo was 0.15 (95% CI, 0.09 to 0.25). Results were similar for the January 15, 2021, database lock.

Analysis of second PFS in the OL period (for patients who crossed over from placebo to ripretinib 150 mg once daily) is presented in Appendix 3. Among patients who crossed over from placebo to ripretinib 150 mg once daily following initial designation of objective PD per IRR (before any dosage escalation; n = 29), subsequent PFS events (second PFS) occurred in 13 (44.8%) patients. Median second PFS was 20.0 weeks (95% CI, 8.0 weeks to not estimable) in this group of patients.

Figure 5: Kaplan-Meier Plot of PFS in the INVICTUS Study (ITT Population; Data Cut: May 31, 2019)

CI = confidence interval; DCC-2618 = ripretinib; ITT = intention to treat; PFS = progression-free survival; QD = once daily.

Note: Symbols represent censored observations.

Source: INVICTUS Clinical Study Report.¹²

Figure 6: Kaplan-Meier Plot of PFS in the INVICTUS Study (ITT Population; Data Cut: January 15, 2021)

CI = confidence interval; DCC-2618 = ripretinib; ITT = intention to treat.

Note: Symbols represent censored observations.

Source: Sponsor’s additional information.¹³

Refer to Appendix 3 for sensitivity and subgroup analyses of PFS. Sensitivity analyses of PFS were consistent with the primary analysis. A planned subgroup analysis of PFS by number of lines of prior therapy was suggestive of similar effect sizes of ripretinib on PFS among patients who had received 3 (HR 0.15; 95% CI, 0.08 to 0.25) and 4 or more (HR 0.24; 95% CI, 0.12 to 0.51) lines of prior therapy. No subgroup analysis of PFS by tumour mutational status was conducted.

Symptom Severity

Symptom severity was not specifically assessed in the INVICTUS study, other than via the symptom scales of generic patient-reported HRQoL instruments (EORTC QLQ-C30, EQ-5D-5L).

Harms

Only those harms identified in the review protocol are reported in this section. See Table 22 for detailed harms data.

Adverse Events

Almost all patients who received placebo (97.7%) and ripretinib (98.8%) experienced at least 1 AE during the DB treatment period. Common AEs that occurred more frequently in ripretinib-treated patients than in placebo-treated patients included alopecia (51.8% versus 4.7%), fatigue (42.4% versus 23.3%), nausea (38.8% versus 11.6%), constipation (34.1% versus 18.6%), myalgia (31.8% versus 11.6%), diarrhea (28.2% versus 14.0%), and vomiting (21.2% versus 7.0%).

Serious Adverse Events

Serious adverse events occurred in 44.2% of patients who received placebo and in 30.6% of patients who received ripretinib during the DB treatment period.

Adverse Events Leading to Dosage Modification

AEs leading to dosage reduction occurred in 1 (2.3%) patient who received placebo and 6 (7.1%) patients who received ripretinib during the DB treatment period. AEs leading to dosage interruption occurred in 9 (20.9%) patients who received placebo and 20 (23.5%) patients who received ripretinib during the DB treatment period.

Withdrawals Due to Adverse Events

Five patients (11.6%) who received placebo and 7 patients (8.2%) who received ripretinib withdrew from the protocol therapy due to AEs during the DB treatment period.

Mortality

Ten patients (23.3%) who received placebo and 5 patients (5.9%) who received ripretinib died within 30 days of the last dose of study drug during the DB treatment period. The primary cause of death in most patients was PD.

Notable Harms

Among protocol-specified AESIs, squamous cell carcinoma of the skin occurred in no patients who received placebo and 2 patients (2.4%) who received ripretinib during the DB treatment period. Actinic keratosis occurred in 1 patient (2.3%) who received placebo and 5 patients (5.9%) who received ripretinib during the DB treatment period. No patients experienced keratoacanthoma during the study.

All notable harms specified in the CADTH review protocol occurred more frequently in patients who received ripretinib than in those who received placebo during the DB treatment period: cardiac dysfunction, cardiac ischemic events, hypertension, cutaneous malignancies, palmar-plantar erythrodysesthesia syndrome, arthralgia, myalgia, and increased bilirubin. The most common AEs by preferred term were peripheral edema (ripretinib versus placebo: 16.5% versus 7.0%), hypertension (14.1% versus 4.7%), palmar-plantar erythrodysesthesia syndrome (21.2% versus 0%), arthralgia (17.6% versus 4.7%), myalgia (31.8% versus 11.6%), and increased bilirubin (16.5% versus 0%).

Critical Appraisal

Internal Validity

INVICTUS was a relatively small, phase III DB multi-centre RCT with an OL period of active treatment (N = 129) conducted in patients with advanced GIST who had prior progression or intolerance on imatinib, sunitinib, and regorafenib.^9-11 Randomization appeared generally successful in balancing baseline demographic and disease characteristics between study arms. However, as expected for a smaller study of a rare condition, small baseline imbalances were present between study arms, several of which were of potential prognostic significance. The younger age, higher proportion of patients with ECOG PS 0, and higher proportion of patients with gastric tumours in the ripretinib arm would be expected to potentially favour ripretinib; however, according to the clinical experts consulted by CADTH for this review, these slight imbalances would not limit the interpretation of study results.

Many of the outcomes used in the INVICTUS study (OS, ORR, DOR, TTR, PFS) are standard in oncology trials, and tumour responses were objectively evaluated using mRECIST version 1.1 – GIST-specific per IRR. According to the clinical experts consulted by CADTH for this review, the definition of stable disease used in the study (6 weeks) was arbitrary. Although there is no standardized definition of stable disease in GIST patients, the clinical experts stated that this interval is shorter than what most clinicians would consider stable, especially as scans would typically be conducted every 2 to 4 months in clinical practice. Few patients in the study had important deviations (< 5%) and there was very good adherence and dose intensity (> 95%) for patients in the ripretinib arm. Withdrawal of consent and losses to follow-up were both very low. However, there were several limitations of the basic design of the study. First, only 44 patients were randomized to receive placebo, many of whom progressed very rapidly, resulting in unblinding of patient and investigator. Development of characteristic AEs, including AESIs, may also have led to partial unblinding. Thus, the extent to which investigators remained blinded over the course of the study was unclear, although the clinical experts consulted for this review did not expect that any resulting bias (direction uncertain) would limit interpretation of the study results. Second, and more importantly, interpretation of OS results was limited by elective crossover of patients from DB placebo to OL ripretinib 150 mg once daily and by intra-patient post-progression dosage escalation to 150 mg twice daily, a dosage not approved by Health Canada. Crossover, post-progression treatment, and dosage escalation were not accounted for by censoring or other techniques. Thus, the relative effects of ripretinib versus placebo treatment, pre- versus post-progression treatment, and ripretinib dosage on OS could not be easily ascertained from the available data. The lack of adjustment for crossover would be expected to bias the analysis of OS differences toward the null (against ripretinib). The magnitudes and directions of biases in the OS analysis attributable to post-progression treatment and dosage escalation were uncertain. However, based on the similar proportions of patients in each arm who received OL post-progression treatment and escalated dosage, these biases may have been nondirectional, although this interpretation was associated with substantial uncertainty.

Several statistical issues should be considered when interpreting the results of the INVICTUS study. Overall statistical tests were appropriate, and the results of the primary PFS analysis and key secondary ORR analysis were robust to several sensitivity analyses. Although not explicitly tested, the proportional hazards assumption appeared likely to be satisfied for Cox models based on KM survival curves. Multiplicity was controlled using an appropriate hierarchical testing strategy. However, the second step in the hierarchy was the hypothesis test of ORR, which had only 80% power to detect a difference in ORR between the ripretinib and placebo arms of 20%. According to the clinical experts consulted by CADTH for this review, ORRs of this magnitude were unlikely to occur in this patient population. Because of the modest power to detect differences in ORR of magnitudes that, according to the clinical experts, would be somewhat optimistic in this advanced disease setting, statistical testing of ORR returned a nonsignificant result at the time of the primary analysis, precluding subsequent testing of important outcomes to clinicians and patients (HRQoL and OS). Only 1 of the subgroup analyses of interest for this review was specified a priori and was based on a stratification factor (number of prior lines of therapy). Subgroup analysis by tumour mutational status was conducted only for ORR, in an exploratory fashion. The study was not specifically powered to evaluate strata among subgroups; there were no tests for differences among subgroups; and subgroup analyses were not controlled for multiplicity.

Analyses of HRQoL data (EORTC QLQ-C30 role and physical function; EQ-5D-5L usual activities and pain/discomfort dimensions; EQ-5D-5L utility index scores; EQ-VAS) were limited by several factors, including absence of formal statistical testing (due to failure of the statistical hierarchy at the time of the primary analysis at an earlier stage or hypothesis tests being outside the hierarchy). Several HRQoL outcomes (e.g., EQ-5D-5L dimensions) were affected by high rates of missing data for cycle 2 day 1. The strategy of replacing missing data with those from the end of treatment (e.g., for EORTC QLQ-C30 functional scales), which may have been significantly earlier or later than cycle 2 day 1, was not clearly justified. Importantly, the measurement properties of both HRQoL instruments have not been investigated, and MIDs have not been identified specifically for GIST patients. The degree to which these generic HRQoL instruments capture the symptoms of GIST patients, which were considered highly important outcomes by patients and clinicians, was unclear. Moreover, limited evidence is available documenting the responsiveness to change of the EORTC QLQ-C30 and the reliability and responsiveness to change of the EQ-5D-5L among all cancer patients. In addition, the clinical experts consulted by CADTH for this review expressed uncertainty as to whether it is plausible that HRQoL would improve owing to tumour response during the first cycle of ripretinib. Analysis of longer-term trends in HRQoL data were not conducted in the study. Safety outcomes were generally well captured, although the small size of the study may have reduced the likelihood of detecting very rare harms.

External Validity

According to the clinical experts consulted by CADTH for this review, the demographic and disease characteristics of patients enrolled in the INVICTUS study^9-11 reflect the Canadian population of advanced GIST patients they would treat in their clinical practice. Similarly, the clinical experts felt that the study eligibility criteria would be expected to result in recruitment of a patient population that reflects Canadian practice. As with most oncology trials, however, the study likely enrolled a healthier cross-section of patients with better PS who were more likely to tolerate and respond to therapy than the general population of GIST patients. The clinical experts confirmed that exclusion criteria related to cardiac problems would eliminate, because of safety concerns, a small percentage of patients who would not be good candidates for ripretinib in clinical practice. The clinical experts noted that few patients with ECOG PS 2 were enrolled in the study and that generalizability to patients with ECOG PS greater than 2 was even less certain. The clinical experts stated that, although most patients in the INVICTUS study were White and had gastric tumours with primary KIT exon 11 mutations, this would not limit generalizability to other patients (e.g., the smaller numbers of patients with other tumour sites and mutations).

The clinical experts consulted by CADTH for this review confirmed that placebo plus BSC was an appropriate comparator for ripretinib, as there are currently no options for fourth-line therapy following failure of imatinib, sunitinib, and regorafenib. Dosages of ripretinib administered during the DB period of the INVICTUS study (150 mg once daily) were aligned with the Health Canada—approved dosage and with clinical practice. However, dosage during the OL period (patients could choose to receive ripretinib beyond initial PD and to escalate the dosage to 150 mg twice daily) was not aligned with Health Canada–approved dosage or an accepted standard of clinical practice. Thus, generalizability of the OS results of the INVICTUS study to Canadian patients is uncertain.

Several of the outcomes assessed in the INVICTUS study, including OS, HRQoL, and PFS, were identified as clinically important by both patients and clinical experts. Patients and clinicians alike indicated that improved survival is the most important outcome of treatment but that stabilization or improvement of HRQoL is also critical. According to the clinical experts, the generic HRQoL instruments use in the INVICTUS study are research tools that are not used in clinical practice. Moreover, improvement in GIST symptoms was identified as a critical outcome of treatment but was not directly assessed in the study; instead, it was assessed indirectly using generic HRQoL instruments. The duration of follow-up was adequate for assessment of the primary and secondary efficacy outcomes and safety outcomes, especially for the unplanned analysis from the recent database lock of January 15, 2021.

Patients in the INVICTUS study underwent imaging scans more frequently (every cycle; 28 days) and may have had better access to the treating clinician and team via participation in the trial compared with the expected treatment setting in Canada. However, the clinical experts consulted by CADTH for this review did not feel this would limit generalizability of the study findings.

Indirect Evidence

No indirect evidence was identified for this review.

Other Relevant Evidence

No other relevant evidence was identified for this review.

Discussion

Summary of Available Evidence

One phase III, DB multi-centre RCT with an OL period of active treatment (INVICTUS, N = 129)^9-11 contributed evidence to this report. The study enrolled patients with advanced GIST who had previously progressed on or developed intolerance to imatinib, sunitinib, and regorafenib. Patients were randomized 2:1 to receive either ripretinib 150 mg once daily or placebo during the DB period until the first objective PD per IRR or unacceptable toxicity. The primary outcome was PFS per IRR, and hierarchically tested secondary outcomes were ORR per IRR, change from baseline to cycle 2 day 1 in EORTC QLQ-C30 role function and physical function, and OS. Other secondary outcomes assessed outside of the statistical hierarchy included TTR, DOR, and change from cycle 2 day 1 in EQ-5D-5L usual activities and pain/discomfort dimensions, EQ-5D-5L utility index score, and EQ-VAS.

According to the clinical experts consulted by CADTH for this review, the baseline characteristics of the INVICTUS study population were broadly representative of Canadian GIST patients who have progressed on or developed intolerance to imatinib, sunitinib, and regorafenib and would be candidates for fourth-line ripretinib. Most patients were White and from the US, with a mean age of approximately 60 years, and most had ECOG PS 0/1. The most common tumour site was gastric; the most common location of primary tumour mutations was KIT exon 11; and patients were roughly evenly split between having received 3 versus 4 or more prior lines of therapy. A major limitation of the included study was uncertain interpretation of changes due to ripretinib treatment in 2 outcomes identified as critically important by both patients and clinicians (OS and HRQoL). The OS analysis did not account for crossover, post-progression treatment, and dosage escalation to a not approved by Health Canada. Analyses of HRQoL data were limited by absence of formal statistical testing, missing data (up to 25% of the randomized population), and lack of instrument validation specifically in patients with GIST. Changes in patient symptoms, which were noted as highly important by both patients and clinicians, were not directly measured in the study.

Interpretation of Results

Efficacy

Patients originally randomized to the ripretinib arm in the INVICTUS study had numerically improved OS (median 65.6 weeks) compared with those originally randomized to the placebo arm (median = 28.6 weeks). However, the statistical hierarchy failed at the time of the primary analysis before testing of this outcome, precluding definitive conclusions. Interpretation of the OS data was additionally limited by complexities associated with patient crossover, post-progression treatment, and intra-patient dosage escalation, which were not adjusted for in the analysis. The impact of crossover from placebo to OL ripretinib would be expected to bias OS comparisons toward the null (i.e., bias against ripretinib), while the impacts of post-progression treatment and intra-patient dosage escalation were uncertain but possibly nondirectional. Nonetheless, the clinical experts consulted by CADTH judged that the difference in OS between patients originally randomized to DB ripretinib or placebo was potentially clinically relevant in this patient population that is affected by advanced disease and has no other treatment options. Numerical differences in OS were also observed for the subgroup of patients who received only DB treatment with ripretinib versus placebo before initial objective PD (without any OL treatment), as well as for patients who crossed over from placebo to OL ripretinib 150 mg once daily versus patients randomized to the placebo arm who did not cross over.

The clinical experts consulted by CADTH for this review felt that the difference in ORR per IRR during the DB treatment period of the INVICTUS study was encouraging and potentially clinically relevant for fourth-line treatment of patients with advanced GIST. However, the comparison of ORR between the ripretinib and placebo arms did not reach statistical significance at the time of the primary analysis, precluding definitive conclusions. The clinical experts stated that the relatively low ORR in the ripretinib arm was expected and that, in tandem with the higher proportion of patients in the ripretinib arm with stable disease (65.9% versus 20.5%), the ORR findings supported and were consistent with the OS and PFS results.

Administration of ripretinib during the DB treatment period of the INVICTUS study resulted in statistically significant prolongation of PFS (median = 27.6 weeks) compared with placebo (median = 4.1 weeks; P < 0.0001), which the clinical experts consulted by CADTH for this review judged as highly clinically meaningful. An unplanned analysis of mature PFS data from a more recent data cut (January 15, 2021) were consistent with the primary PFS analysis.

Patient input identified HRQoL and symptom relief as very important outcomes for patients with GIST. Unfortunately, assessment of changes in HRQoL from baseline was limited by absence of formal statistical testing, missing data, large variability in estimates, and lack of HRQoL instrument validation specifically in patients with GIST. Consistent numeric differences in the INVICTUS study provided potential signals of HRQoL stabilization or improvement in the ripretinib arm versus placebo arm across all outcomes assessed. However, these differences in HRQoL outcomes were not easily interpretable, and conclusions were uncertain. Validated tools for studying HRQoL and symptoms in GIST patients are needed.

Harms

According to input from the clinical experts consulted by CADTH for this review, as well as clinician and patient groups, the safety profile of ripretinib reflected in the INVICTUS study was acceptable and manageable, especially considering the seriousness of the underlying disease and the lack of other treatment options. Clinicians and patients both noted that the tolerability and toxicity profile of ripretinib appeared to be improved compared with prior lines of therapy for GIST, including other TKIs. Similar and low proportions of patients treated with ripretinib and placebo arms withdrew from the protocol therapy due to AEs, and ripretinib dosage reductions were required in relatively few (approximately 7%) patients receiving the drug. SAEs occurred less often in patients receiving ripretinib compared to placebo, potentially due to delayed disease progression. Although the notable harms of ripretinib are varied and, in many cases, serious, the clinical experts consulted by CADTH for this review emphasized that most can be managed through careful patient selection and close monitoring, without posing undue risk to patients.

Conclusions

Evidence from the INVICTUS study suggested that administration of ripretinib in patients with advanced GIST who had previously received imatinib, sunitinib, and regorafenib was associated with statistically significant and clinically meaningful prolongation of PFS compared with placebo. Administration of ripretinib also resulted in numerically higher ORR compared with placebo, although this difference was not statistically significant at a P value of less than 0.05 at the time of the primary analysis. OS was numerically longer in patients randomized to receive ripretinib compared with those randomized to receive placebo. However, differences in OS between patients randomized to receive ripretinib versus placebo were not tested statistically due to early failure of the statistical hierarchy, precluding definitive conclusions. Changes in patient-reported HRQoL (EORTC QLQ-C30, EQ-5D-5L) following ripretinib administration were difficult to interpret due to absence of formal statistical testing, missing data, wide variation in estimates, and uncertainty regarding HRQoL measurement properties in GIST patients. Ripretinib was generally well tolerated in most patients, and its notable harms were considered expected and acceptable by patients and clinicians. The observed PFS benefits, consistent numeric improvements in other efficacy outcomes, and acceptable toxicity profile in the study were aligned with outcomes identified as important to patients with advanced GIST who currently have no treatment options available.

References

1.Aubin F, Blanke CD. Metastatic gastrointestinal stromal tumors. Cancer Chemother Pharmacol. 2011;67(Suppl 1):S9-14. PubMed

2.Emile JF, Brahimi S, Coindre JM, et al. Frequencies of KIT and PDGFRA mutations in the MolecGIST prospective population-based study differ from those of advanced GISTs. Med Oncol. 2012;29(3):1765-1772. PubMed

3.Kee D, Zalcberg JR. Current and emerging strategies for the management of imatinib-refractory advanced gastrointestinal stromal tumors. Ther Adv Med Oncol. 2012;4(5):255-270. PubMed

4.Caterino S, Lorenzon L, Petrucciani N, et al. Gastrointestinal stromal tumors: correlation between symptoms at presentation, tumor location and prognostic factors in 47 consecutive patients. World J Surg Oncol. 2011;9:13. PubMed

5.Maleddu A, Pantaleo MA, Nannini M, Biasco G. The role of mutational analysis of KIT and PDGFRA in gastrointestinal stromal tumors in a clinical setting. J Transl Med. 2011;9:75. PubMed

6.Eisenberg BL, Smith KD. Adjuvant and neoadjuvant therapy for primary GIST. Cancer Chemother Pharmacol. 2011;67(Suppl 1):S3-8. PubMed

7.About GIST. Cobourg (ON): GIST Sarcoma Life Raft Group Canada; 2020: https://liferaftgroup.ca/about-gist/. Accessed 2022 Jan 1.

8.Drug Reimbursement Review sponsor submission: Qinlock (ripretinib) 50 mg oral tablets [internal sponsor's package]. Toronto (ON): Medison; 2021 Oct 15.

9.Bauer S, Heinrich MC, George S, et al. Clinical activity of ripretinib in patients with advanced gastrointestinal stromal tumor harboring heterogeneous KIT/PDGFRA mutations in the phase III INVICTUS study. Clin Cancer Res. 2021;27(23):6333-6342. PubMed

10.Blay JY, Serrano C, Heinrich MC, et al. Ripretinib in patients with advanced gastrointestinal stromal tumours (INVICTUS): a double-blind, randomised, placebo-controlled, phase 3 trial. Lancet Oncol. 2020;21(7):923-934. PubMed

11.Zalcberg JR, Heinrich MC, George S, et al. Clinical benefit of ripretinib dose escalation after disease progression in advanced gastrointestinal stromal tumor: an analysis of the INVICTUS study. Oncologist. 2021;26(11):e2053-e2060. PubMed

12.Clinical Study Report: DCC-2618-03-001. A phase 3, interventional, double-blind, placebo-controlled study to assess the safety and efficacy of DCC-2618 in patients with advanced gastrointestinal stromal tumors who have received treatment with prior anticancer therapies (INVICTUS) [internal sponsor's report]. Waltham (MA): Deciphera Pharmaceuticals; 2019.

13.Medison Pharma Canada Inc. response to 16 December 2021 DRR request for additional information regarding ripretinib DRR review [internal additional sponsor's information]. Toronto (ON): Medison Pharma Canada; 2021.

14.Corless CL. Gastrointestinal stromal tumors: what do we know now? Mod Pathol. 2014;27(Suppl 1):S1-16. PubMed

15.Miettinen M, Lasota J. Gastrointestinal stromal tumors: pathology and prognosis at different sites. Semin Diagn Pathol. 2006;23(2):70-83. PubMed

16.Nishida T, Blay JY, Hirota S, Kitagawa Y, Kang YK. The standard diagnosis, treatment, and follow-up of gastrointestinal stromal tumors based on guidelines. Gastric Cancer. 2016;19(1):3-14. PubMed

17.Laurent M, Brahmi M, Dufresne A, et al. Adjuvant therapy with imatinib in gastrointestinal stromal tumors (GISTs)-review and perspectives. Transl Gastroenterol Hepatol. 2019;4:24. PubMed

18.Blanke CD, Rankin C, Demetri GD, et al. Phase III randomized, intergroup trial assessing imatinib mesylate at two dose levels in patients with unresectable or metastatic gastrointestinal stromal tumors expressing the kit receptor tyrosine kinase: S0033. J Clin Oncol. 2008;26(4):626-632. PubMed

19.Care RS, Valk PJ, Goodeve AC, et al. Incidence and prognosis of c-KIT and FLT3 mutations in core binding factor (CBF) acute myeloid leukaemias. Br J Haematol. 2003;121(5):775-777. PubMed

20.Corless CL, Barnett CM, Heinrich MC. Gastrointestinal stromal tumours: origin and molecular oncology. Nat Rev Cancer. 2011;11(12):865-878. PubMed

21.Demetri GD, Reichardt P, Kang YK, et al. Efficacy and safety of regorafenib for advanced gastrointestinal stromal tumours after failure of imatinib and sunitinib (GRID): an international, multicentre, randomised, placebo-controlled, phase 3 trial. Lancet. 2013;381(9863):295-302. PubMed

22.George S, Wang Q, Heinrich MC, et al. Efficacy and safety of regorafenib in patients with metastatic and/or unresectable GI stromal tumor after failure of imatinib and sunitinib: a multicenter phase II trial. J Clin Oncol. 2012;30(19):2401-2407. PubMed

23.Kang YK, Ryu MH, Yoo C, et al. Resumption of imatinib to control metastatic or unresectable gastrointestinal stromal tumours after failure of imatinib and sunitinib (RIGHT): a randomised, placebo-controlled, phase 3 trial. Lancet Oncol. 2013;14(12):1175-1182. PubMed

24.PrQinlockTM (ripretinib): 50mg oral tablets [product monograph]. Waltham (MA): Deciphera Pharmaceuticals; 2020 Jun 19.

25.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

26.Grey matters: a practical tool for searching health-related grey literature. Ottawa (ON): CADTH; 2019: https://www.cadth.ca/grey-matters. Accessed 2021 Nov 9.

27.Deciphera Pharmaceuticals completes enrollment in the INVICTUS pivotal phase 3 clincial study of DCC-2618 in patients with advanced gastrointestinal stromal tumours [press release]. San Francisco (CA): Business Wire; 2018: https://www.businesswire.com/news/home/20181115006037/en/Deciphera-Pharmaceuticals-Completes-Enrollment-in-the-INVICTUS-Pivotal-Phase-3-Clinical-Study-of-DCC-2618-in-Patients-with-Advanced-Gastrointestinal-Stromal-Tumors. Accessed 2022 Jan 1.

28.EORTC Questionnaires: CORE. 2021: https://qol.eortc.org/core/. Accessed 2021 Nov 24.

29.EuroQol Group. EuroQol--a new facility for the measurement of health-related quality of life. Health Policy. 1990;16(3):199-208. PubMed

30.Medical Dictionary for Regulatory Activities (MedDRA) version 21.1. Geneva (CH): International Council for Harmonisation of Technical Requirements for Pharmaceuticals for Human Use; 2018: https://www.meddra.org. Accessed 2022 Jan 1.

31.Common Terminology Criteria for Adverse Events (CTCAE v4.0: CTCAE v4.03. Bethesda (MD): National Cancer Institute; 2010: https://ctep.cancer.gov/protocoldevelopment/electronic_applications/ctc.htm. Accessed 2022 Jan 1.

32.Luo N, Fones CS, Lim SE, Xie F, Thumboo J, Li SC. The European Organization for Research and Treatment of Cancer Quality of Life Questionnaire (EORTC QLQ-c30): validation of English version in Singapore. Qual Life Res. 2005;14(4):1181-1186. PubMed

33.Davda J, Kibet H, Achieng E, Atundo L, Komen T. Assessing the acceptability, reliability, and validity of the EORTC Quality of Life Questionnaire (QLQ-C30) in Kenyan cancer patients: a cross-sectional study. J Patient Rep Outcomes. 2021;5(1):4. PubMed

34.Naamala A, Eriksson LE, Orem J, Nalwadda GK, Kabir ZN, Wettergren L. Psychometric properties of the EORTC QLQ-C30 in Uganda. Health Qual Life Outcomes. 2021;19(1):131. PubMed

35.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

36.Snyder CF, Blackford AL, Sussman J, et al. Identifying changes in scores on the EORTC-QLQ-C30 representing a change in patients' supportive care needs. Qual Life Res. 2015;24(5):1207-1216. PubMed

37.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

38.Richardson J, Iezzi A, Khan MA, Chen G, Maxwell A. Measuring the sensitivity and construct validity of 6 utility instruments in 7 disease areas. Med Decis Making. 2016;36(2):147-159. PubMed

39.Pickard AS, Neary MP, Cella D. Estimation of minimally important differences in EQ-5D utility and VAS scores in cancer. Health Qual Life Outcomes. 2007;5:70. PubMed

40.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

41.EORTC Quality of life: FAQ. Brussels (BE): EORTC; 2021: https://qol.eortc.org/faq/. Accessed 2021 Nov 24.

42.EORTC QLC-C30 scoring manual. Brussels (BE): EORTC; 2001: https://www.eortc.org/app/uploads/sites/2/2018/02/SCmanual.pdf. Accessed 2021 Nov 24.

43.Brooks R. EuroQol: the current state of play. Health Policy. 1996;37(1):53-72. PubMed

44.EQ-5D-5L user guide: basic information on how to use the EQ-5D-5L instrument. Rotterdam (NL): EuroQol Research Foundation; 2019: https://euroqol.org/wp-content/uploads/2021/01/EQ-5D-5LUserguide-08-0421.pdf. Accessed 2021 Nov 24.

45.Ameri H, Yousefi M, Yaseri M, Nahvijou A, Arab M, Akbari Sari A. Mapping the cancer-specific QLQ-C30 onto the generic EQ-5D-5L and SF-6D in colorectal cancer patients. Expert Rev Pharmacoecon Outcomes Res. 2019;19(1):89-96. PubMed

46.Sinnott PL, Joyce VR, Barnett PG. Preference measurement in economic analysis. Guidebook. Menlo Park (CA): Health Economics Research Center; 2007: https://www.herc.research.va.gov/files/BOOK_419.pdf. Accessed 2021 Nov 24.

Appendix 1: Literature Search Strategy

Note 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: November 12, 2021

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 23: Syntax Guide

Multi-Database Strategy

1. (Qinlock* or ripretinib* or DCC2618 or DCC-2618 or 9XW757O13D).ti,ab,ot,kf,hw,nm,rn.

2. 1 use medall

3. *ripretinib/ or (Qinlock* or ripretinib* or DCC2618 or DCC-2618).ti,ab,kf,dq.

4. 3 use oemezd

5. 4 not (conference abstract or conference review).pt.

6. 2 or 5

7. remove duplicates from 6

Clinical Trials Registries

ClinicalTrials.gov

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

Search terms – Qinlock/ripretinib

WHO ICTRP

International Clinical Trials Registry Platform, produced by WHO. Targeted search used to capture registered clinical trials.

Search terms – Qinlock/ripretinib

Health Canada’s Clinical Trials Database

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

Search terms – Qinlock/ripretinib

EU Clinical Trials Register

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

Search terms – Qinlock/ripretinib

Grey Literature

Search dates: November 2 to 8, 2021

Keywords: Qinlock, ripretinib

Limits: None

Updated: Search updated before 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)

Appendix 2: Description and Appraisal of Outcome Measures

Note this appendix has not been copy-edited.

Aim

To describe the following outcome measures and their measurement properties (validity, reliability, responsiveness to change, and MID):

• EORTC QLQ-C30

• EQ-5D-5L and EQ-VAS

Findings

A focused literature search was conducted to evaluate the psychometric properties and the MID of each of these outcome measures. The findings on reliability, validity, responsiveness, and the MID of each outcome measure are summarized in Table 24.

Table 24: Summary of Outcome Measures and Their Measurement Properties

ECOG PS = Eastern Cooperative Oncology Group performance status; EORTC QLQ-C30 = European Organization for Research and Treatment of Cancer Quality of Life Questionnaire-Core 30; GIST = gastrointestinal stromal tumour; HRQoL = health-related quality of life; MID = minimal important difference; SF-36 = 36 item short-form survey; VAS = visual analogue scale.

No literature was identified that assessed validity, responsiveness, or reliability of either of these HRQoL instruments in patients with GIST. No MID information for either HRQoL instrument was identified in patients with GIST.

Studies assessing the psychometric properties of the instruments were only summarized if the assessment was done for the English version and in a sample of mixed (more than 1 type) cancer patients. No literature was identified regarding responsiveness to change of the instruments based on these criteria.

European Organization for Research and Treatment of Cancer Quality of Life Questionnaire-Core 30 (EORTC QLQ-C30)

Description

The EORTC Quality of Life Questionnaire-Core 30, (EORTC QLQ-C30), is 1 of the most commonly used patient-reported outcome measures in oncology clinical trials.^28,40 It is a multi-dimensional, cancer-specific, evaluative measure of HRQoL. It was designed specifically for the purpose of assessing changes in participants’ HRQoL in clinical trials in response to treatment.⁴¹ The core questionnaire of the EORTC QLQ-C30 consists of 30 questions across 5 multi-item functional scales, 3 multi-item symptom scales, 6 single-item symptom scales, and a 2-item global HRQoL scale (Table 25). Version 3.0 of the questionnaire, used in the INVICTUS study, is the most current version and has been in use since December of 1997.⁴² The instrument is available more than 110 languages²⁸ and is intended for use in adult populations only.³³ Notably, the global HRQoL scale is also known as global health status which was an outcome in the INVICTUS study.³³ Patients in the INVICTUS study self-reported their responses to the EORTC QLQ-C30 on a tablet computer.

Table 25: Scales of the EORTC QLQ-C30

Scoring

The EORTC QLQ-C30 uses a 1-week recall period in assessing function and symptoms. Most questions have 4 response options (“not at all,” “a little,” “quite a bit,” “very much”), with scores on these items ranging from 1 (not at all) to 4 (very much).⁴² For the 2 items that form the global quality of life scale, however, responses are on a 7-point Likert-type scale, with anchors of 1 (very poor) and 7 (excellent).⁴²

Raw scores for each scale are computed as the average of the items that contribute to a particular scale. This scaling approach is based upon the assumption that it is appropriate to provide equal weighting to each item that comprises a scale. There is also an assumption that, for each item, the interval between response options is equal (for example, the difference in score between “not at all” and “a little” is the same as “a little” and “quite a bit”). Each raw scale score is converted to a standardized score that ranges from 0 to 100 using a linear transformation, with a higher score reflecting better function on the function scales, higher symptoms on the symptom scales, and better HRQoL on the global quality of life scale. Thus, a decline in scores on symptom scales would reflect an improvement, whereas an increase in scores on function and quality of life scales would reflect an improvement.⁴²

Psychometric Properties

Validity

Three studies with a sample of mixed (more than 1 type) cancer patients were identified that evaluated the psychometric properties of the English version of the EORTC QLQ-30.

One cross-sectional study aimed to validate the EORTC QLQ-30 in a convenience sample of 57 cancer patients (mean 43 [range 15 to 79] years, 44% male) receiving chemotherapy in Singapore.³² Most patients had breast and colorectal cancer (n = 32, 56%), but leukemia, lung cancer, lymphoma, germ cell tumour, and other cancers were also reported. Construct validity was assessed by cross-sectional correlational evidence and discriminative evidence. First, convergent validity was assessed using Spearman’s correlations between 8 pairs of QLQ-30 and Short Form-36 (SF-36) scales (which measures HRQoL within the past 4 weeks), hypothesizing moderate to strong correlation (defined as correlation coefficients of 0.35 to 0.5, and > 0.5, respectively) between scales of these 2 instruments measuring similar dimensions of HRQoL. Moderate to strong correlations were observed between the QLC-30 and SF-36 scales, ranging from 0.35 (comparison of QLQ-C30 role functioning and SF-36 role-emotional scales) to 0.67 (comparison of QLQ-C30 pain and SF-36 bodily pain scales). Next, the known-groups approach was used to compare 6 QLQ-30 scale scores between patients reporting mild and severe symptoms, as well as by stage of disease and presence of comorbid conditions. With the exception of emotional functioning, the remaining 5 scales showed better scores in patients with mild symptoms than those with severe symptoms. There were no significant differences found in QLQ-30 scores between patients in early stages of cancer (or with no comorbid conditions) compared to those in advanced disease stages (or with comorbid conditions).³² This may be attributable to the small sample size of the study which may not have been sufficiently powered to detect a significant difference.

A recent cross-sectional study in Kenya was conducted to evaluate the psychometric properties of the EORTC QLQ-C30, using the English or Kiswahili version in 100 patients with cancer (median 53.5 [range 18 to 83] years, 41% male).³³ Most (40%) patients had breast cancer, followed by prostate cancer, Kaposi sarcoma, lung cancer, and other cancers. Convergent validity was assessed by examining Pearson’s intra-scale correlations among the subscales of the EORTC QLQ-C30. Regarding convergent validity, items showed at least moderate correlation with their own scales (r ≥ 0.4) with higher correlations with their own scales than with other scales. This included strong correlations between physical and role functioning (r = 0.66), emotional and cognitive functioning (r = 0.58), pain and fatigue (r = 0.73), and sleep disorders and fatigue (r = 0.68). Furthermore, with the exception of cognitive functioning, emotional functioning, nausea and vomiting, dyspnea, appetite loss, constipation, and diarrhea, global health status correlated moderately with the remaining subscales (r ≥ 0.30).³³

Another recent cross-sectional study in Uganda evaluated the English version of EORTC QLQ-C30 among 168 adult patients with cancer (37% male).³⁴ A total of 168 patients with cancers (cervical, breast, Kaposi sarcoma, leukemia, and other cancers) completed the English version of the EORTC QLQ-C30. The study assessed the presence of floor and ceiling effects that were defined as acceptable if they did not exceed 15%. Subscales that exceeded the acceptable floor effect included role and social functioning as well as nausea/vomiting. Subscales exceeding the acceptable ceiling effect included role, emotional, and cognitive functioning, as well as pain. This may be explained by the high number of patients with advanced stages of cancer with reduced role and social functioning due to fatigue and pain. Confirmatory factor analysis using the maximum likelihood method was conducted to assess construct validity of the EORTC QLQ-C30, with coefficients of ≥ 0.4 considered acceptable. Satisfactory correlations were observed ranging from 0.46 to 0.97, suggesting strong construct validity. Further evidence of construct validity of the instrument was demonstrated using the known-groups comparisons approach by examining the capacity to discriminate between patients differing in disease stage (stages I/II versus stages III/IV). Effect sizes (ES) were calculated and 0.2, 0.5 to 0.8, and > 0.8 were classified as small, moderate, and large differences, respectively. There were statistically significant differences for all subscales except for cognitive function (ES = 0.23) and emotional function (ES = 0.21). Small effect sizes (ranging from 0.14 for social functioning to 0.46 for fatigue) were identified for all scales except for physical function which displayed a moderate ES (0.70). These results demonstrated that the instrument was able to discriminate between early and late stages of cancer among the surveyed patients.³⁴ Criterion validity was assessed by examining associations between 2 EORTC QLQ-C30 subscales (global quality of life and physical function) and the Karnofsky Performance Scale (KPS). Positive correlations were observed between KPS and the 2 subscales, with r = 0.72 and r = 0.76 for each of global quality of life and physical function, respectively.

Reliability

The Singaporean cross-sectional study also assessed internal consistency reliability by calculating Cronbach alpha for all EORTC QLQ-C30 scales. Cronbach alpha was ≥ 0.70 for 6 of the 9 assessed QLQ-30 scales; cognitive functioning, physical functioning, and nausea and vomiting had Cronbach alpha ranging from 0.19 to 0.68.³²

Both the Kenyan and Ugandan studies previously described assessed the internal consistency of each scale of the questionnaire using Cronbach alpha-coefficients. In the Kenyan study, with the exception of the cognitive function scale, all of the scales had Cronbach alpha ≥ 0.70.³³ In the Ugandan study, Cronbach alpha ranged from 0.50 (cognitive function scale) to 0.96 (global quality of life scale).³⁴

Studies evaluating the responsiveness to change of the English version of the EORTC QLQ-C30 among mixed cancer patients were not identified.

MID

For use in clinical trials, scores on the EORTC QLQ-C30 can be compared between different groups of patients or within a group of patients over time. One study of patients in 2 randomized trials of chemotherapy (1 for breast cancer and 1 for small-cell lung cancer) conducted in 1998 estimated a clinically relevant change in score on any scale of the EORTC QLQ-C30 to be 10 points.³⁵ The study used an anchor-based approach to estimate the MID in which patients who reported “a little” change (for better or worse) on the subjective significance questionnaire had corresponding changes on a function or symptom scale of the EORTC QLQ-C30 of approximately 5 to 10 points. Participants who reported a “moderate” change had corresponding changes in the EORTC QLQ-C30 of about 10 to 20 points, and those who reported having “very much” changed HRQoL had corresponding changes of more than 20 points.³⁵

More recently in 2015, a Canadian study estimated the MIDs of EORTC QLQ C-30 scales using data from 193 newly diagnosed breast and colorectal cancer patients who had recently undergone surgery (mean = 60 [range = 22 to 88] years, 20% male).³⁶ The Supportive Care Needs Survey-Short Form-34 (SCNS-SF34) was used as an anchor; mean changes in EORTC QLQ C-30 scales associated with improvement, worsening, and no change in supportive care needs based on the SCNS-SF34 were then calculated. MIDs were assessed for the following scales: physical function, role function, emotional function, global health status, pain, and fatigue. For improvement, MIDs associated with improved supportive care needs ranged from 10 to 32 points. For worsening, MIDs associated with a worsening of supportive care needs ranged from 9 to 21 points. The range for unchanged supportive care needs was from 1-point worsening to 16-point improvement in EORTC QLQ-C30 score.³⁶ Based on this, the authors suggested that a 10-point change in EORTC QLQ C-30 score represented changes in supportive care needs, and therefore should be considered for clinical use.³⁶

In 2014, another Canadian study estimated the MID for EORTC QLQ-C30 among 369 patients with advanced cancer (mean = 57.7 [SD = 12.8] years, 22% male) who completed the questionnaire at baseline and 1-month post-radiation.³⁷ The most common cancer types were breast cancer (n = 270, 70%), followed by lung, prostate, gastrointestinal, renal cell, and other cancers. MID was estimated using both anchor- and distribution-based methods for improvement and deterioration. Two anchors of overall health and overall HRQoL were used, both taken directly from the EORTC QLQ-C30 (questions 29 and 30) where patients rated their overall health and HRQoL themselves. The MIDs were determined by calculating the difference in mean change in scores between patients with improved versus unchanged overall health and between patients with deteriorated versus unchanged overall health. Improvement and deterioration were categorized as an increase or decrease of 2 units to account for the natural fluctuation of patient scoring (i.e., assuming that 1 unit may not be sensitive). With these 2 anchors, the estimated MIDs (95% CI) across all EORTC QLQ-C30 scales ranged from 9.1 (1.4 to 16.7) to 23.5 (31.9 to 15.2) units for improvement, and from 7.2 (0.2 to 14.2) to 13.5 (3.7 to 23.3) units for deterioration. For the overall health anchor, none of the symptom scales showed a significant MID at the 1-month follow-up which may be due to differing symptoms among the patient groups examined. Distribution-based estimates were closest to 0.5 SD.

EQ-5D-5L

Scoring

The EQ-5D is a generic HRQoL instrument that may be applied to a wide range of health conditions and treatments.^29,43 The first of 2 parts of the EQ-5D is a descriptive system that classifies respondents (aged ≥ 12 years) based on the following 5 dimensions: mobility, self-care, usual activities, pain/discomfort, and anxiety/depression. In 2005, updates were made to the EQ-5D-3L to create the EQ-5D-5L which included creating 5 response levels of severity (no problems, slight problems, moderate problems, severe problems, unable to/extreme problems) in each of the 5 existing dimensions.⁴⁴ Respondents are asked to choose the level that reflects their health state for each of the 5 domains resulting in 3,125 possible health states.⁴⁵ A scoring function can be used to assign a value to self-reported health states from a set of population-based preference weights.^29,43 The second part is a 20 cm visual analogue scale (EQ-VAS) that has end points labelled 0 and 100, with respective anchors of “worst imaginable health state” and “best imaginable health state.” Respondents are asked to rate their health by drawing a line from an anchor box to the point on the EQ-VAS which best represents their health on that day.

The EQ-5D index score is generated by applying societal preference weights for the various aforementioned health states.⁴⁶ Health state index scores less than 0 represent health states that are valued by society as being worse than dead, while scores of 0 and 1.00 are assigned to the health states ‘dead’ and ‘perfect health,’ respectively.

Psychometric Properties

Validity

Richardson et al. (2016)³⁸ examined various instruments, including the EQ-5D-5L, among respondents who were healthy and who had a chronic disease (i.e., arthritis, asthma, cancer, depression, diabetes, hearing loss, and heart disease) through an online survey in Australia, Canada, Germany, Norway, the UK, and the US (total N = 7,933; cancer N = 772). For discriminant validity, the mean EQ-5D-5L utility score of healthy respondents was compared to patients with cancer (and other chronic diseases). The mean utility scores differed in these subgroups (0.88 in healthy individuals, 0.18 in patients with cancer). Regarding convergent validity, the EQ-5D-5L was compared to other related scales using Pearson’s correlations. The EQ-5D-5L was strongly correlated with the physical component of the SF-36 in cancer patients (r = 0.66), moderately correlated with the psychosocial content of the mental component of the SF-36, the Capabilities Instrument, and the Subjective Well-Being Instrument of the UK Office of National Statistics (r = 0.50), and moderately correlated with preference measures of VAS and time trade-off on own health state (r = 0.43).³⁸

Studies evaluating the reliability and responsiveness of the English version of instrument among mixed cancer patients were not identified.

MID

Pickard et al. (2007)³⁹ estimated the MID of the EQ-5D VAS based on cross-sectional data collected from 534 patients (mean 59 [SD = 12] years, 52% male) in the US with advanced (stage III or IV) cancer of the bladder, brain, breast, colon or rectum, head or neck, liver or pancreas, kidney, lung, lymphoma, ovary, or prostate.³⁹ EQ-5D index-based utility (UK and US) scores were estimated using both anchor and distribution-based approaches. Groups were anchored by ECOG PS and then distribution approaches including the standard error of the mean and 0.5 SD were applied to each anchor-based category. MID estimates were similar across all cancers. For utility scores, MID estimates for all cancers and the lung cancer subgroup ranged from 0.10 to 0.12 for UK scores and 0.07 to 0.09 for US scores. MIDs for the EQ-5D VAS ranged from 8 to 12 based on the ECOG PS, and from 7 to 10 based on Functional Assessment of Cancer Therapy HRQoL questionnaire quintiles.³⁹

Appendix 3: Detailed Outcome Data

Note this appendix has not been copy-edited.

Table 26: Patient Disposition in the INVICTUS Study (OL Period)

IC = informed consent; IRR = independent radiological review; ITT = intention to treat; NA = not applicable; OL = open label; PD = progressive disease; PP = per protocol; q.d. = once daily.

Note: Data are from the May 31, 2019, database lock.

^aDenominator is the safety population.

^bDenominator is the ITT population.

^cFor the OL period, all patients received assigned treatments per protocol.

Source: INVICTUS Clinical Study Report.¹²

Figure 7: Kaplan-Meier Plot of OS in the INVICTUS Study by Subgroups of DB and OL Treatment (ITT Population; Data Cut: January 15, 2021)

BD = twice daily; CI = confidence interval; DB = double-blind; DCC = ripretinib; ITT = intention to treat; NC = not calculable; OL = open label; OS = overall survival; PLA = placebo; QD = once daily.

Note: symbols represent censored observations.

Source: Sponsor’s additional information.¹³

Table 27: Sensitivity Analyses of ORR in the INVICTUS Study

BOR = best overall response; CI = confidence interval; IRR = independent radiological review; ITT = intention to treat; NA = not applicable; ORR = objective response rate.

^aFrom exact binomial CI.

^bP value from Fisher’s exact test.

^cNewcombe Score 95% CI of the difference in ORR between treatment arms.

^dDefined as absence of PD for at least 6 weeks.

Source: INVICTUS Clinical Study Report.¹²

Table 28: Subgroup Analysis of ORR per IRR in the INVICTUS Study

CI = confidence interval; IRR = independent radiological review; NE = not estimable; ORR = objective response rate.

^aNewcombe Score 95% CI of the difference in ORR between treatment arms.

Source: INVICTUS Clinical Study Report.¹²

Table 29: Sensitivity Analyses of PFS in the INVICTUS Study

CI = confidence interval; eCRF = electronic case record ford; HR = hazard ratio; IRR = independent radiological review; IRT = interactive response technology; ITT = intention to treat; NE = not estimable; NR = not reported; PFS = progression-free survival; PP = per protocol.

^aFrom KM analysis.

^bP value from 2-sided stratified log-rank test. Stratification factors were the same as those applied to randomization (number of prior anticancer treatment and ECOG status at baseline).

^cHR from Cox proportional regression model with treatment and randomization stratification factors as fixed factors.

Source: INVICTUS Clinical Study Report.¹²

Table 30: Subgroup Analysis of PFS in the INVICTUS Study

CI = confidence interval; HR = hazard ratio; PFS = progression-free survival.

^aHR from Cox proportional regression model with treatment and randomization stratification factors as fixed factors.

Source: INVICTUS Clinical Study Report.¹²

Table 31: PFS per IRR in the INVICTUS Study (ITT Population, OL Period)

CI = confidence interval; HR = hazard ratio; IRR = independent radiological review; ITT = intention to treat; NE = not estimable; NR = not reported; OL = open label; PFS = progression-free survival.

Note: Data are from the May 31, 2019, database lock.

^aFrom KM analysis.

^bP value from 2-sided stratified log-rank test. Stratification factors were the same as those applied to randomization (number of prior anticancer treatment and ECOG status at baseline).

^cHR from Cox proportional regression model with treatment and randomization stratification factors as fixed factors.

Source: INVICTUS Clinical Study Report.

Pharmacoeconomic Review