Reimbursement Review

Avapritinib (Ayvakyt)

Sponsor: Medison Pharma Canada Inc.

Therapeutic area: Advanced systemic mastocytosis

This multi-part report includes:

Clinical_Review

Pharmacoeconomic_Review

Clinical_Review

Abbreviations

Executive Summary

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

Table 1: Background Information on Application Submitted for Review

AdvSM = advanced systemic mastocytosis; ASM = aggressive systemic mastocytosis; MCL = mast cell leukemia; NOC = Notice of Compliance; SM-AHN = systemic mastocytosis with an associated hematologic neoplasm.

Introduction

Systemic mastocytosis (SM) is a heterogenous group of rare disorders caused by a clonal, neoplastic proliferation of abnormal mast cells that accumulate typically in bone marrow and other extracutaneous tissues.^1-3 Symptoms of SM are related to the release of mast cell mediators and mast cell tissue infiltration, which can vary widely from isolated symptoms to a constellation of symptoms, commonly including cutaneous involvement (e.g., skin flushing, pruritus, itching, hives, skin rash), wheezing and shortness of breath, dizziness, cardiovascular symptoms (e.g., rapid heart rate, chest pain, low blood pressure), gastrointestinal symptoms (e.g., diarrhea, nausea, vomiting, abdominal pain), fatigue, musculoskeletal symptoms (e.g., bone and/or muscle pain), and neuropsychiatric symptoms (e.g., headache, brain fog, cognitive dysfunction, anxiety, depression).⁴

SM is classified into distinct subtypes in order of increasing disease burden: indolent SM and bone marrow mastocytosis, smouldering SM, aggressive systemic mastocytosis (ASM), systemic mastocytosis with an associated hematologic neoplasm (SM-AHN), and mast cell leukemia (MCL).^3-5 Advanced systemic mastocytosis (AdvSM) includes the disease variants of ASM, SM-AHN, and MCL.¹ The prevalence rate of SM is estimated at 1 per 10,000 people of all ages.⁶ Based on estimates from the Danish incidence and German prevalence of adults with AdvSM, the estimated AdvSM incidence rate in Canada is 0.06 cases per 100,000 adults and the prevalence rate is 5.2 cases per million adults.^7,8 Median overall survival (OS) has been estimated at 41 months for ASM,^4,9 11 months to 42 months for SM-AHN depending on the type of AHN,^4,10 and 2 months to 19.2 months for MCL.^3,4

Cytoreduction is the principal treatment for AdvSM; it may improve quality of life, reverse or prevent organ damage, and prolong survival.⁴ Current available cytoreductive options in Canada for AdvSM include midostaurin, cladribine, interferons (e.g., peginterferon alfa-2a), and imatinib. The 2024 National Comprehensive Cancer Network (NCCN) clinical practice guidelines recommend enrolment in clinical trial, or KIT inhibitors — midostaurin or avapritinib — as first-line treatment for AdvSM.⁴

According to the clinical experts consulted for this review, there is a major treatment gap for patients with AdvSM in Canada. The off-label treatments currently available (cladribine and interferon) have low or unpredictable response rates, have a response of short duration, and may cause significant toxicity. Imatinib is suitable for a small minority of patients with AdvSM who do not have the KIT D816V mutation or have an unknown KIT mutational status. The targeted therapy, midostaurin, is approved for the treatment of AdvSM in Canada but is not publicly funded and therefore is inaccessible.

The objective of this report is to review and critically appraise the clinical evidence submitted by the sponsor on the beneficial and harmful effects of avapritinib 25 mg, 50 mg, 100 mg, and 200 mg oral tablets in the treatment of adult patients with AdvSM, including ASM, SM-AHN, and MCL. The recommended starting dose of avapritinib is 200 mg daily.

Perspectives of Patients, Clinicians, and Drug Programs

The information in this section is a summary of input provided by the patient and clinician groups who responded to the call by Canada’s Drug Agency (CDA-AMC) for input and from clinical experts consulted for this review.

Patient Input

CDA-AMC received 2 patient group submissions from Heal Canada and the Leukemia & Lymphoma Society of Canada (LLSC). Heal Canada is a not-for-profit organization that aims to empower patients, provide patient education and awareness, improve health care outcomes, and advocate for equitable access to quality health care. LLSC is a national charitable organization dedicated to finding a cure and improving quality of life for people and their families affected by blood cancers through research, educational resources, services, and support. Heal Canada conducted an online survey of patients living with blood cancer (from February to May 2024); however, no patient with AdvSM was recruited and no patient was identified to have had experience with avapritinib, so the submitted information was based on the Mast Cell Connect (MCC) patient registry data and its publications. LLSC conducted 1-on-1 interviews with 3 patients with SM (1 patient each with AdvSM, indolent SM, and an unknown SM subtype) and 1 caregiver whose father had AdvSM. One interviewed patient with ASM and a caregiver of a patient with ASM reported on experience with avapritinib. The caregiver reported that their father’s skin issues and itching dissipated and his quality of life improved while receiving avapritinib, with no major adverse effects. The patient who received avapritinib via compassionate care access expressed its life-changing impact in alleviating symptom burden and mental strain due to the disease, with no observable adverse effects.

Both Heal Canada and LLSC reported SM to be a rare disease with a complex and variable clinical presentation that can lead to misdiagnosis or delays in diagnosis. Patients with AdvSM frequently reported symptoms of fatigue, concentration difficulties, body pain, sleep disturbances, nausea, vomiting, skin irritations (e.g., rashes, itching, hives), and anxiety and depression. The onset of symptoms is unpredictable, and may be triggered by temperature, stress, exercise, food, medication, and other factors. Patients are at risk of life-threatening anaphylaxis due to SM. The frequency and intensity of symptoms can vary widely, with some patients chronically disabled while others may lead relatively normal lives.

There is a lack of accessible and effective treatments for patients with AdvSM in Canada. Patients seek better treatments that address the underlying disease, provide symptom relief, have tolerable adverse effects, allow the restoration of daily activities, and improve physical and mental well-being.

Clinician Input

Input From Clinical Experts Consulted by CDA-AMC

According to the clinical experts consulted for this review, there is a major treatment gap for patients with AdvSM in Canada. The off-label treatments currently available have low or unpredictable response rates, have a response of short duration, and may cause significant toxicity. The clinical experts stated that avapritinib would be used as first-line monotherapy in patients with AdvSM, except for patients who present very acutely and need rapid debulking with cladribine or those patients with a platelet count of less than 50 multiplied by 10⁹/L. In these patients, avapritinib may be used as second-line therapy, after debulking or once platelet counts have increased. The experts identified that the patients most suitable for treatment are those who meet the WHO diagnostic criteria for AdvSM, who are treatment-naive, or who have received prior therapy for AdvSM. The experts anticipated that patients with all subtypes of AdvSM (ASM, SM-AHN, and MCL) would benefit from treatment with avapritinib monotherapy.

According to the clinical experts consulted, assessment of a clinically meaningful response requires the integration of patients’ goals of treatment with clinical and histopathological factors. Improvement in a patient’s symptoms is a critical part of the response assessment. Treatment response also requires a reduction in abnormal mast cell burden in bone marrow, and improvement in clinical signs of organ damage due to infiltration by neoplastic mast cells. This may include a normalization of complete blood counts and liver function enzymes, a reduction in spleen or liver volume, a reduction or an absence of transfusion requirements, and a reduction in the need for diuretics and/or therapeutic paracentesis.

As per the pivotal trials and input from the clinical experts, avapritinib should be discontinued in patients who are no longer getting clinical benefit from a symptom or quality-of-life perspective, in patients who experience intracranial bleeding or have a platelet count less than 50 multiplied by 10⁹/L, in patients with persistent severe treatment-related adverse events (AEs) that cannot be managed with dose interruptions or dose reduction, in patients where there is evidence of progressive disease of either the SM or associated hematologic neoplasm (AHN) disease component, or in patients who are pregnant.

Clinician Group Input

Two clinician groups provided input for this review: the Ontario Health (Cancer Care Ontario) Hematology Cancer Drug Advisory Committee (based on 2 clinicians) and the LLSC Clinician Network and Myeloproliferative Neoplasms Canada Clinician Group (based on 6 clinicians).

In general, the clinician groups’ input was consistent with the input provided by the clinical experts consulted for this review. The clinician groups agreed that there is a significant unmet need for patients with AdvSM in Canada who have poor outcomes and a high symptomatic burden. The experts anticipated that avapritinib would be used as first-line monotherapy in Canada for adults with all subtypes of AdvSM. The clinician groups agreed that patients with AdvSM should be managed by hematologists or medical teams with expertise in diagnosis, treatment, and response evaluation. Input from clinician groups indicated that standardized response criteria are evolving and may be used in conjunction with evaluations of clinical benefit, including patient-reported symptoms and health-related quality of life. The clinician groups agreed that treatment with avapritinib should be discontinued among patients whose health-related quality of life is impacted by a lack of clinical benefit, and those with a platelet count of less than 50 multiplied by 10⁹/L, detectable disease progression, or significant adverse effects.

Drug Program Input

The drug programs identified issues related to relevant comparators and generalizability. For more information, refer to Table 4.

Clinical Evidence

Systematic Review

Description of Studies

Two open-label, single-arm clinical trials provided data on the efficacy and safety of avapritinib in adults with AdvSM. Eligible patients were aged 18 years or older with an adjudicated diagnosis of either ASM, SM-AHN, or MCL according to the WHO diagnostic criteria. The phase I EXPLORER study enrolled 86 patients, including 69 patients with AdvSM, who received avapritinib in either the dose escalation phase (part 1) or the extension phase (part 2). In the dose escalation phase, patients received avapritinib 30 mg to 400 mg daily and in the extension phase, the avapritinib starting dose was 300 mg or 200 mg daily. The phase II PATHFINDER study was ongoing at the time of this review and provided results for 62 patients at the first data cut-off date (the planned interim analysis), and for 105 patients at a second data cut-off date. Patients in the PATHFINDER study received an avapritinib starting dose of 200 mg daily. The key efficacy outcomes were overall response rate (ORR), OS, and change in patient-reported symptom severity, measured using the Advanced Systemic Mastocytosis Symptom Assessment Form (AdvSM-SAF) total symptom score (TSS).

The results presented are from the final data cut-off date of the EXPLORER study (January 19, 2023) as well as the first data cut-off date (June 23, 2020) and second data cut-off date (September 9, 2022) of the ongoing PATHFINDER study, which had a mean treatment duration of ████ ██████ ███ █ ██████ ███ ██████ ███ █ █████ ███ ████ ██████ ███ █ █████, respectively, in the safety population. The median OS follow-up duration was ████ ███████ ███ ██████ and 26.3 months in the EXPLORER study (in the final data cut-off) and the PATHFINDER study’s first and second data cut-offs, respectively.

The mean age of patients enrolled was 65.0 (standard deviation [SD] = 11.2) years and 67.5 (SD = 11.0) years in the EXPLORER and PATHFINDER studies, respectively, 41% and 45% of patients were female, respectively, and 59% and 55% of patients were male, respectively. In the EXPLORER and PATHFINDER studies, the most common AdvSM subtype was SM-AHN (70% and 69%, respectively), followed by MCL (19% and 16%, respectively), and ASM (12% and 15%, respectively). In the EXPLORER and PATHFINDER studies, most patients had an Eastern Cooperative Oncology Group Performance Status (ECOG PS) score of 0 or 1 (70% and 69%, respectively), with 20% and 23% of patients rated as having an ECOG PS score of 2, respectively, and 10% and 8% of patients rated as having a score of 3, respectively. In the EXPLORER study, 59% of patients had received prior antineoplastic therapy compared with 68% of patients in the PATHFINDER study (in the first data cut-off).

Efficacy Results

The ORR was the primary outcome in the PATHFINDER study and a secondary outcome in the EXPLORER study. Response was based on the centrally adjudicated modified International Working Group - Myeloproliferative Neoplasms Research and Treatment - European Competence Network on Mastocytosis (IWG-MRT-ECNM) criteria in the Response Assessment Committee–Response Evaluable (RAC-RE) population. In both studies, overall response was defined as patients with complete remission (CR), patients with complete remission with partial hematological recovery (CRh), patients with partial remission (PR), and patients with clinical improvement. The observed ORR was █████ ████ ██████████ ████████ █████ █████ ██ ██████ | | ███ in the EXPLORER study, 75.0% ███████ ███ █████ ██ ██████ | | ███ in the first data cut-off of the PATHFINDER study, and 74.1% (95% confidence interval [CI], 63.1% to 83.2%; N = 81) in the second data cut-off. According to the statistical analysis plans, the ORR was tested versus the 28% null value, which was the post hoc estimate of the ORR for midostaurin. In the EXPLORER study and in the first data cut-off of the PATHFINDER study, the P value was less than 0.0001 based on a 1-sided test.

OS was an exploratory outcome in the EXPLORER study and a secondary outcome in the PATHFINDER study, and was reported for the safety population. The median survival was not reached for either study, as ██ ██ ██ ██ ████████ ███████ were alive at the end of the EXPLORER study, and 84 of 105 (80.0%) patients were alive at the second data cut-off of the PATHFINDER study. In the EXPLORER study, the Kaplan-Meier estimates for the proportion of patients alive at | █████ ███ █████ ████ ███ █████ ██ ██████ ███ ██ █ █████ ███ █████ ████ ███ █████ ██ ██████. At the interim analysis of the PATHFINDER study, ██████ ██████ ███ █████ of patients were alive at 6 months, 12 months, and 18 months, respectively. As of the PATHFINDER study second data cut-off date, the Kaplan-Meier estimate for OS was 79.0% (95% CI, 70.8% to 87.3%) at 2 years.

The AdvSM-SAF TSS captures the severity of 8 symptoms (abdominal pain, nausea, vomiting, diarrhea, spots, itching, flushing, and fatigue) and is scored from 0 (no symptoms) to 80 points (worst imaginable), based on the average daily score over the prior week. Using anchor-based methods, the estimated within-person minimal important difference (MID) was 9 points to 14 points for the TSS.¹ In the EXPLORER study, the AdvSM-SAF questionnaire was completed during part 2 only (safety population), with 40 (74%) patients reporting a baseline score. At baseline, the mean TSS was 19.1 (SD = 12.2) points, with a mean change from baseline of █████ ██████ ████ ██ █████ ██ █████ | | ███ at cycle 11, day 1.

In the PATHFINDER study, 56 of 62 (90%) patients reported a baseline AdvSM-SAF TSS at the first data cut-off date, and 91 of 105 (87%) patients reported a score at the second data cut-off date (safety population). For the first data cut-off, the baseline mean TSS was 18.3 (SD = 12.5) points (N = 56), with a mean change from baseline of –9.8 points (95% CI, –14.9 points to –4.6 points; N = 22) at cycle 11, day 1. For the second data cut-off, the baseline TSS was not reported and the mean change from baseline was –6.4 (SD = 9.7) points (N = 91) at cycle 11, day 1.

Among patients who met the overall response criteria, the median duration of response was ████ ██████ ████ ███ ████ ███████ ███ ██████████ in the EXPLORER study (N = 44). The median duration of response was not reached by the second data cut-off date of the PATHFINDER study (N = 60). At 24 months, █████ ████ ███ █████ ██ ██████ of responders in the EXPLORER study and 88.8% (95% CI, 80.4% to 97.3%) of responders in the PATHFINDER study had maintained the response. At 36 months, the proportion of patients was █████ ████ ███ █████ ██ ██████ and 84.6% (95% CI, 73.2% to 96.0%) in the EXPLORER and PATHFINDER trials, respectively. The median time to response was ███ ██████ ████ ███ ███ ██ █████ in the EXPLORER study, 2.0 months (95% CI, 0.3 months to 12.2 months) in the first data cut-off of the PATHFINDER study, and 2.2 months (95% CI, 0.3 months to 15.0 months) in the PATHFINDER study’s second data cut-off.

Progression-free survival (PFS) was an exploratory outcome in the EXPLORER study and a secondary outcome in the PATHFINDER study. In the EXPLORER study, the median PFS was 49.0 months (95% CI, 31.2 months to not estimable), and at 24 months the Kaplan-Meier PFS estimate was █████ ████ ███ █████ ██ ██████. The median PFS had not been reached in the ongoing PATHFINDER study, which reported a 24-month PFS survival estimate of 76.5% (95% CI, 66.9% to 86.0%) in the second data cut-off.

Harms Results

All patients in the EXPLORER and PATHFINDER trials reported at least 1 AE. The most common AEs were periorbital edema (69% and 41%), anemia (57% and 51%), diarrhea (49% and 31%), thrombocytopenia (41% and 43%), and peripheral edema (41% and 47%) in the EXPLORER and PATHFINDER studies (second data cut-off), respectively.

Serious adverse events (SAEs) were reported ███ and 51% of patients, and █████ and 23.8% of patients stopped treatment due to AEs in the EXPLORER and PATHFINDER studies (second data cut-off), respectively. In the EXPLORER study, the most common SAEs were ███████ ███████ █████████ ██████████ ████████ █████████ ████████████ ███████████ █████ ███████ █████████ ███ ███████. Limited information on specific AEs was available for the PATHFINDER study. █████ ████████ ██████ in the EXPLORER study and | ████████ ██████ in the PATHFINDER study (second data cut-off) died due to AEs.

Intracranial bleeding was identified as an AE of special interest by the sponsor and by the clinical experts who were consulted for this review. In the EXPLORER study ██ ████████ ███████ experienced intracranial bleeding. In the PATHFINDER study, 1 (1.6%) patient in the first data cut-off and 4 (3.7%) patients in the second data cut-off experienced intracranial bleeding.

Cognitive AEs were common and were reported by █████ and 27.6% of patients in the EXPLORER and PATHFINDER trials, respectively. These events included ██████ ███████████ █████████ █████████ ███████████ ██████ ████████ ██████████████ ███ ██████████.

Critical Appraisal

Both pivotal trials were open-label, single-arm studies, and thus provided no direct evidence on comparative efficacy or safety. The lack of controlled trials has implications for the overall strength and interpretability of the results. With single-arm studies, there is an increased risk of bias in the estimation of treatment effects due to the potential for confounding related to natural history and prognostic factors. Moreover, the extent of any selection bias is difficult to ascertain. The clinical experts emphasized that AdvSM is a heterogeneous disease, and prognosis varies substantially based on the disease subtype and other factors. The primary outcome (ORR) and other response-related outcomes were analyzed in a subset of patients enrolled in the studies, not in the entire population with AdvSM, which is another potential source of selection bias.

While the lack of a comparator group in the pivotal evidence limits the overall interpretation of the results, the feasibility of conducting a randomized controlled trial was low, given the rarity of AdvSM, and potential ethical issues were raised by the clinical experts consulted, due to the efficacy and safety of the available comparators.

The primary outcome was based on ORR according to the modified IWG-MRT-ECNM criteria. The clinical experts noted that response criteria used in clinical trials and in practice are evolving to best capture clinical benefit and to better define long-term outcomes, given the availability of targeted therapies. While the clinical experts consulted agreed that the International Working Group (IWG) criteria used in the trials was acceptable, there is no clear data to suggest which response criteria perform better in terms of predicting long-term outcomes like survival.

The pivotal trials were open-label, whereby the investigator and study participants were aware of their treatment status, potentially increasing the risk of detection bias and performance bias. As such, the open-label trial design limits the interpretability of the subjective study outcomes, such as AdvSM-SAF, and AEs, and may impact some components of the IWG-MRT-ECNM criteria. The AdvSM-SAF was further limited by the extent of missing data; ███ ███ ███ of patients were excluded from the analysis at baseline and due to attrition, ███ ███ ███ of patients had missing data at cycle 11 in the EXPLORER and PATHFINDER studies (in the first data cut-off), respectively. In the PATHFINDER study, the use of the last observation carried forward imputation method for patients with missing data also may have biased the findings.

Most of the time-to-event outcomes were considered immature, as the median OS was not met for either study, and the median PFS and duration of response were not met for the PATHFINDER trial. Additionally, comparative OS and PFS cannot be adequately assessed in a single-arm trial because all patients receive the same treatment. The FDA 2021 medical review report for Ayvakyt (avapritinib) oral (application number 212608) states that the effect of avapritinib on OS cannot be interpreted due to the single-arm, open-label design of the studies, which can return biased results.¹¹

With regard to external validity, the results predominantly reflect patients with SM-AHN with an ECOG PS score of 0 or 1 and who had received prior systemic therapy for AdvSM. In Canada, SM-AHN is the most common type of AdvSM, which is consistent with the studies. However, clinical experts consulted noted that the proportion of patients with high ECOG PS scores was lower than expected in the trials. As a result, the study patients may have been less ill than patients who receive avapritinib in clinical practice. Both trials excluded patients with comorbidities such as seizure disorder, uncontrolled cardiovascular disease, and reduced renal and hepatic function, and those patients at higher risk of intracranial bleeding; thus, the safety and efficacy of avapritinib in these patients is unknown. Approximately three-quarters of patients in the EXPLORER study did not receive the Health Canada recommended starting dose, which may impact the generalizability of the findings to clinical practice, particularly for safety, as the sponsor identified dose-related toxicities.

GRADE Summary of Findings and Certainty of the Evidence

For pivotal studies identified in the sponsor’s systematic review, the Grading of Recommendations Assessment, Development and Evaluation (GRADE) tool was used to assess the certainty of the evidence for outcomes considered most relevant to inform the expert committee deliberations, and a final certainty rating was determined as outlined by the GRADE Working Group.^12,13

Although GRADE guidance is not available for noncomparative studies, the review team assessed pivotal single-arm trials for study limitations (which refers to internal validity or risk of bias), inconsistency across studies, indirectness, the imprecision of effects, and publication bias to present these important considerations. Because the lack of a comparator arm does not allow for a conclusion to be drawn on the effect of the intervention versus any comparator, the certainty of evidence for single-arm trials started at very low certainty.

For the GRADE assessments, findings from the EXPLORER and PATHFINDER studies were considered together and summarized narratively per outcome because these studies were similar in population, interventions, design, and outcome measures.

The selection of outcomes for GRADE assessment was based on the Sponsor Summary of Clinical Evidence,¹⁴ consultation with clinical experts, and input received from patient and clinician groups and public drug plans.

Table 2: Summary of Findings for Avapritinib for Adults With AdvSM

AdvSM = advanced systemic mastocytosis; AdvSM-SAF = Advanced Systemic Mastocytosis Symptom Assessment Form; CI = confidence interval; IWG-MRT-ECNM = International Working Group - Myeloproliferative Neoplasms Research and Treatment - European Competence Network on Mastocytosis; NR = not reported; ORR = overall response rate; OS = overall survival; PFS = progression-free survival; RAC-RE = Response Assessment Committee–Response Evaluable; SAE = serious adverse event; SM = systemic mastocytosis; TSS = total symptom score.

Note: All serious concerns with study limitations (which refers to internal validity or risk of bias), inconsistency across studies, indirectness, the imprecision of effects, and publication bias are documented in the Table 2 footnotes.

^aIn the absence of a comparator group, conclusions about efficacy or safety relative to any comparator cannot be drawn and the certainty of evidence begins at very low. In addition, all outcomes were rated down 1 level for serious study limitations. The efficacy results are based on the small sample sizes of the EXPLORER and PATHFINDER studies and it is unclear if results would be replicable in a larger sample. All outcomes were rated down 1 level for indirectness as 70% or more of patients in the EXPLORER study did not receive the recommended avapritinib starting dose of 200 mg daily.

^bThe RAC-RE population included all patients with AdvSM who received at least 1 dose of avapritinib, were deemed evaluable per modified IWG-MRT-ECNM criteria at baseline as assessed by a Study Steering Committee review, and had 1 of the following conditions: had 2 or more complete postbaseline bone marrow assessments and had been in the study for at least 6 cycles (6 × 28 days), or had an end-of-study visit. The safety population included all patients who received at least 1 dose of avapritinib. For efficacy outcomes in the EXPLORER study, the safety population only included patients with AdvSM (69 of 86 patients enrolled), but for adverse event outcomes, data from all patients were reported, including 17 patients with indolent or smouldering SM. Results from the EXPLORER study were based on patients who received any starting dose of avapritinib (30 mg to 400 mg daily); the PATHFINDER study first data cut-off included patients who had received a starting dose of 100 mg daily (2 patients) or 200 mg daily (60 patients), and the second data cut-off included patients who had received a starting dose of 200 mg daily (105 patients).

^cOverall response was defined according to the modified IWG-MRT-ECNM criteria and included patients who experienced a best response of complete remission, complete remission with partial recovery of peripheral blood counts, partial remission, or clinical improvement.

^dThe median follow-up was ████ months for the EXPLORER study and 10.2 months for the planned interim analysis of the PATHFINDER study.

^eAs per the planned interim analysis of the PATHFINDER study, the ORR was tested at a 1-sided alpha of 0.00625; thus, the 98.75% CI was listed.

^fMedian follow-up for survival was ████ months for the EXPLORER study and 26.3 months for the second data cut-off of the PATHFINDER study.

^gRated down 1 level due to risk of bias, as the second data cut-off of the PATHFINDER study was not a preplanned interim analysis, according to the study protocol, and as such should be interpreted as supportive data.

^hRated down 2 levels for risk of bias due to missing data and open-label design. Not all patients enrolled provided results at baseline (missing ███ ███ ███ of patients), with further attrition over time (███ ███ ███ of patients missing from the cycle 11 time point in the EXPLORER and PATHFINDER studies, respectively). The use of the last observation carried forward method in the PATHFINDER study may bias the results, given the magnitude of missing data at cycle 11. The open-label study design and patients’ and assessors’ knowledge of assigned treatment may lead to biased estimates of subjective outcomes.

ⁱMedian follow-up was ████ months for the EXPLORER study and ████ months for the second data cut-off of the PATHFINDER study.

^jMean avapritinib exposure duration was ████ months in the final data cut-off of the EXPLORER study, and ████ months in the second data cut-off of the PATHFINDER study.

^kRated down 1 level because of risk of bias due to the open-label study design. Patients’ and assessors’ knowledge of assigned treatment may lead to biased estimates of subjective outcomes and harms. Due to the lack of a control group, the proportion of adverse events that is attributable to avapritinib versus the disease or other factors is unclear.

Sources: Final Clinical Study Report for the EXPLORER study,¹⁵ interim Clinical Study Report for the PATHFINDER study,¹⁶ PATHFINDER study clinical summary document,¹⁷ and the Sponsor Summary of Clinical Evidence.¹⁴ Additional data supplied by the sponsor (June 17, 2024).¹⁸

Long-Term Extension Studies

No long-term extension studies were submitted.

Indirect Comparisons

Description of Studies

The sponsor submitted 1 indirect treatment comparison and 1 observational comparison that evaluated the efficacy of avapritinib versus available treatments for patients with AdvSM. The indirect treatment comparison was based on a published matching-adjusted indirect comparison (MAIC) comparing avapritinib with midostaurin on OS, ORR, and CR. The individual patient-level data (IPD) observational comparison was based on inverse probability of treatment weighting (IPTW) methods comparing avapritinib with midostaurin, and with real-world best available therapy (BAT), on OS and the duration of treatment.

Efficacy Results

Overall Survival

In the MAIC of avapritinib (the pooled EXPLORER and PATHFINDER studies safety population) versus the pooled midostaurin cohort (the pooled D2201 and A2213 trials’ full analysis set), the OS hazard ratio (HR) was 0.42 (95% CI, 0.25 to 0.71), favouring avapritinib. Follow-up in the MAIC for OS was median 22.9 months and median 7.0 months in the EXPLORER and PATHFINDER studies for avapritinib, respectively; the follow-up duration was median 124 months and median 26 months in the A2213 and D2201 trials for midostaurin, respectively.

In the IPTW-weighted observational comparison of avapritinib first-line therapy (the PATHFINDER study safety population) versus BAT first-line therapy with follow-up of mean ████ ██████ ██████ ████ ████ ██████, respectively, the OS HR was ████ ████ ███ ████ ██ █████, favouring avapritinib first-line therapy.

In the IPTW-weighted observational comparison of avapritinib second-line or later-line therapy (the PATHFINDER study safety population) versus BAT second-line or later-line therapy with follow-up of mean ████ ██████ ██████ ████ ████ ██████, respectively, the OS HR was ████ ████ ███ ████ ██ █████, favouring avapritinib second-line or later-line therapy.

In the IPTW-weighted observational comparison of avapritinib first-line therapy (the PATHFINDER study safety population) versus in the midostaurin first-line therapy with follow-up of mean ████ ██████ ██████ ████ ████ ██████, the OS HR was ████ ████ ███ ████ ██ █████, favouring avapritinib first-line therapy.

Duration of Treatment

In the IPTW-weighted observational comparison of avapritinib first-line therapy (the PATHFINDER study safety population) versus BAT first-line therapy with follow-up of median ████ ██████ ██████ ██████ ███ ██████, respectively, the HR for the duration of treatment was ████ ████ ███ ████ ██ █████, favouring avapritinib first-line therapy.

In the IPTW-weighted observational comparison of avapritinib second-line or later-line therapy (the PATHFINDER study safety population) versus BAT second-line or later-line therapy with follow-up of median ████ ██████ ██████ ██████ ███ ██████, respectively, the HR for the duration of treatment was ████ ████ ███ ████ ██ █████, favouring avapritinib second-line or later-line therapy.

In the IPTW-weighted observational comparison of avapritinib first-line therapy (the PATHFINDER study safety population) versus in the midostaurin first-line therapy with follow-up of median ████ ██████ ██████ ██████ ████ ██████, the OS HR was ████ ████ ███ ████ ██ █████, favouring avapritinib first-line therapy.

Overall Response Rate

In the MAIC of avapritinib (the pooled EXPLORER and PATHFINDER studies RAC-RE population) versus the midostaurin cohort (the D2201 trial primary efficacy population), the odds ratio for ORR was 4.06 (95% CI, 3.09 to 5.33), favouring avapritinib.

Complete Remission

In the MAIC of avapritinib (the pooled EXPLORER and PATHFINDER studies RAC-RE population) versus midostaurin (the D2201 trial primary efficacy population), CR was attained by 10 of 79 (12.66%) patients and 1 of 89 (1.12%) patients, respectively, for an odds ratio of 9.56 (95% CI, 0.97 to 93.81), favouring avapritinib.

Harms Results

No comparative safety data were available in the indirect evidence.

Critical Appraisal

In the indirect treatment comparison (MAIC), methods for study selection were poorly reported. No information was provided on details of the literature search, study selection process, and data extraction. No information including rationale was provided for not assessing the quality of the included studies. Across the included cohorts, trial start dates were heterogenous and notably older in the midostaurin trials (2005 and 2008 for the A2213 and D2201 trials, respectively) than in the avapritinib trials (2016 and 2018 for the EXPLORER and PATHFINDER trials, respectively). The MAIC reported limited details regarding patients enrolled in the trials. Across the studies, differences were observed in the dosing of avapritinib (between the EXPLORER and PATHFINDER studies) and response evaluation (criteria across the 4 trials), and no information on time points used in evaluating response was provided for the trials. Prognostic factors associated with poor outcomes such as AdvSM subtypes of SM-AHN (particularly for type of myeloid neoplasm) and MCL, the KIT D816V variant allele frequency, and the type of gene mutation (e.g., SRSF2) were not included in the MAIC. The selection of prognostic factors used for matching was based on an arbitrary P value of less than 0.1 from exploratory subgroup analyses that contained the same data used for assessment, which is not consistent with recommended approaches in the National Institute for Health and Care Excellence (NICE) Decision Support Unit Technical Support Document 18.¹⁹ Information on patients who received prior systemic therapy in the midostaurin trials was not available. Patients in the avapritinib and midostaurin trials were matched on baseline imbalances that differed by outcome and by analysis populations. Similarities and differences between the populations of analyses were not detailed in the MAIC, making it challenging to determine the comparability of treatment groups and the interpretation of findings. Findings for OS and ORR after weighting resulted in reduced effective sample sizes to suggest incomplete overlap between the avapritinib and midostaurin populations and that results may be driven by a subset of the sample from the index trials that was not representative of the entire sample. The exclusion of data from the A2213 trial and the exploratory nature of the response analysis increase the potential for prognostic imbalance and the risk of type I error. In the MAIC, avapritinib 200 mg (the dose recommended by Health Canada for the indicated population) versus midostaurin was based on sensitivity analyses using the pooled the EXPLORER and PATHFINDER studies RAC-RE population. There is uncertainty in these results at least in part due to the small sample sizes in the avapritinib 200 mg cohort (44 patients and 42 patients for OS and ORR, respectively), which is reflected in the wide CI that crosses the null for survival and in findings that are driven by a reduced sample of the overall population.

No study protocol, statistical analysis protocol, or study report was provided for the observational comparison using IPTW analysis that was based on an updated data cut-off date of September 2022 for the PATHFINDER study. Information presented for the methods of the observational comparison were limited to the sponsor-provided observational comparison report and publication based on earlier analyses (data cut-off date of April 2021), where several inconsistencies and gaps in information were found. Four subgroup analyses that were specified in the sponsor-submitted observational comparison report were reported in the publication but did not match those reported in the updated analyses. Sensitivity analyses of OS described in the sponsor’s observational comparison report were not included in the submission. There were no sensitivity analyses reported to evaluate the potential impact of bias due to informative censoring on effect estimates in patients who were censored due to a new primary malignancy after the index date or due to avapritinib initiation in the BAT cohort. Patients in the avapritinib cohort were enrolled from March 2016 to March 2020 in the EXPLORER study and from November 2018 to June 2020 in the PATHFINDER study. Real-world patients with AdvSM who received BAT were enrolled from January 2009 to October 2021 and included as controls. Contextual information such as standards of care at a specified time point and across time was not directly captured. For patients in the external control group who received BAT as first-line therapy and then went on to receive avapritinib as second-line or later-line therapy either by enrolling in a trial (in the EXPLORER or PATHFINDER studies) or via compassionate program access, follow-up was censored at avapritinib initiation. No further information was detailed regarding how patients who received avapritinib in second-line therapy were included and/or analyzed in the observational comparison. Follow-up duration was not specified for patients included from the PATHFINDER study. The baseline period differed between the comparative cohorts (defined as the 8-week period up to the index date for avapritinib and the 12-week period up to the index date for BAT); no rationale was provided for the different time periods used to ascertain baseline characteristics. While AdvSM subtype was diagnosed based on the WHO criteria for all patients, the evaluation was confirmed by the RAC for the avapritinib cohort and based on local clinician assessment for the BAT cohort. There may be a greater risk of bias among patients who were diagnosed in the BAT cohort due to the retrospective nature of chart review and the lack of information on the assessors. Imbalances in covariates that were found to persist after IPTW weighting (including imbalances that increased for some covariates) for region, ECOG PS scores, anemia, thrombocytopenia, leukocyte counts, and serum tryptase concentrations. These imbalances suggested that there was a lack of sufficient overlap between the cohorts (i.e., the cohorts may have been meaningfully different). While findings for OS and duration of treatment were presented for both the safety and RAC-RE populations of analyses, imbalances in covariates differed for the safety and RAC-RE populations such that it was challenging to meaningfully assess how such differences may have translated to adjusted results (after IPTW weighting) and the comparability of the adjusted results between the analyzed populations. An analysis in the overall sample (avapritinib from the EXPLORER and PATHFINDER studies versus BAT from real-world patients regardless of lines of therapy) was not submitted by the sponsor for updated analyses (for the September 2022 data cut-off). Rather, 3 analyses with longer follow-up were submitted that appeared to be subgroup or post hoc analyses, given that the analysis comparing avapritinib with exclusively midostaurin was not prespecified in the sponsor’s observational comparison report or the publication. The small sample sizes, both overall and reduced for the avapritinib cohort compared to the BAT cohort, make it difficult to ensure prognostic matching was appropriate in the analyses. Median OS had not been reached in the avapritinib cohort in any line of therapy, indicating that OS data were immature. In the observational comparison, the avapritinib 200 mg dose was a subgroup analysis of the PATHFINDER study with small sample sizes. Variations in the timing of assessments and follow-up of patients who received BAT in the real-world setting may not fully match patients who received avapritinib in the EXPLORER and PATHFINDER trials; given the absence of information on follow-up duration (other than at least 3 months of follow-up in the BAT cohort) and this absence’s potential to create prognostic imbalance between the avapritinib and BAT cohorts, there is an unknown direction and magnitude of impact on the duration of treatment and survival. Several methods of imputations for missing data (e.g., ECOG PS score, serum tryptase) were at risk of underestimating disease severity among included patients, although the direction and magnitude of potential bias cannot be determined since the proportion of patients with missing data for the avapritinib cohort were not reported.

The findings from the indirect treatment comparison and observational comparison suggested a benefit of avapritinib (the point estimate and lower and upper bounds of the CIs suggested benefit) when indirectly compared to currently available treatments for ORR, the duration of treatment, and OS. Due to the substantial limitations identified in the analyses, there remains significant uncertainty in the magnitude of the benefit with avapritinib compared to currently available treatments. However, it appears unlikely that the benefit seen with avapritinib is solely explained by the noted limitations and sources of uncertainty of these comparisons. Therefore, while it is not possible to ascertain what the true effect is between the comparisons, it is likely to be in favour of avapritinib.

Studies Addressing Gaps in the Evidence From the Systematic Review

No studies addressing gaps were submitted by the sponsor.

Conclusions

Input from patient groups and clinicians highlighted that AdvSM is a rare, severe, and heterogeneous disease with poor prognosis. Patients have currently limited treatment options. It is estimated that the AdvSM incidence rate in Canada is 0.06 cases per 100,000 adults with presently no access to publicly funded targeted treatment options.

Two single-arm, open-label clinical trials (phase I of the EXPLORER study and the ongoing phase II PATHFINDER study) provided evidence of the efficacy and safety of avapritinib in adults with AdvSM. The studies showed that 75% of the patients who received avapritinib reported an adjudicated overall response, which the clinical experts consulted for this review considered to be clinically relevant and superior to currently available treatments. Due to the lack of a control group, the certainty of evidence was rated as very low, as the possibility of selection bias cannot be ruled out, particularly given the small sample size of the trials and the heterogeneity in patients with AdvSM.

Although OS and PFS were evaluated in the study, the single-arm design and the immaturity of the data limit the ability to attribute the study results to treatment with avapritinib. Patient-reported symptoms were identified as an important outcome but because of the noncomparative design, open-label study design, and high patient attrition rates in the trials, the effect of avapritinib on symptom severity remains inconclusive.

The safety data from the clinical trials were limited by the small sample size (a total of 191 patients) and treatment duration (a mean of ████ ██ ████ ██████). AEs, including SAEs, and AEs requiring dose reductions or discontinuation were common in the studies. However, due to the lack of a control group, the proportion of AEs that were attributable to avapritinib versus the disease or other factors is unclear. Intracranial bleeding and cognitive AEs were identified AEs of special interest for avapritinib. While the sponsor identified thrombocytopenia as a risk factor for intracranial bleeding and implemented patient selection and monitoring criteria to reduce the frequency of events, the risk of intracranial bleeding associated with avapritinib use in clinical practice is unclear.

The noncomparative design of the EXPLORER and PATHFINDER trials precludes the ability to assess the relative therapeutic benefit or safety of avapritinib in Canadian clinical practice. To address the evidence gap, the sponsor submitted indirect evidence on the comparative efficacy of avapritinib versus other treatments for AdvSM. The indirect evidence suggests that relative to midostaurin or a basket of treatments, avapritinib may be associated with improved response, duration of treatment, and OS. It appears unlikely that the benefit seen with avapritinib is solely explained by the limitations and sources of uncertainty that were identified in the comparative analyses. However, there was significant uncertainty in the magnitude of the comparative benefit with avapritinib due to substantial limitations, including heterogeneity in the data sources and patient characteristics, missing or unmeasured prognostic factors and effect modifiers, small sample sizes, few events, imbalanced follow-up times, the immaturity of survival data, and the exploratory nature of post hoc and subgroup analyses. Since there were no data on patient-reported outcomes or comparative safety in the indirect evidence, the relative effect of avapritinib versus currently available treatments on symptoms, health-related quality of life, or safety is unknown.

Introduction

The objective of this report is to review and critically appraise the evidence submitted by the sponsor on the beneficial and harmful effects of avapritinib oral tablets in the treatment of adult patients with AdvSM, including those with ASM, SM-AHN, and MCL.

Disease Background

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

SM is a heterogenous group of rare disorders caused by a clonal, neoplastic proliferation of abnormal mast cells that accumulate typically in bone marrow and other extracutaneous tissues.^1-3 Symptoms of SM are related to the release of mast cell mediators and mast cell tissue infiltration, which can vary widely from isolated symptoms to a constellation of symptoms, commonly including cutaneous involvement (e.g., skin flushing, pruritus, itching, hives, skin rash), wheezing and shortness of breath, dizziness, cardiovascular symptoms (e.g., rapid heart rate, chest pain, low blood pressure), gastrointestinal symptoms (e.g., diarrhea, nausea, vomiting, abdominal pain), fatigue, musculoskeletal symptoms (e.g., bone and/or muscle pain), and neuropsychiatric symptoms (e.g., headache, brain fog, cognitive dysfunction, anxiety, depression).⁴ Additionally, the symptoms of SM occur either spontaneously or in response to various triggers of mast cell activation (e.g., sunlight, heat, cold or sudden temperature changes, physical and emotional stress, food, alcohol consumption, insect stings, infections, drugs or medications, contrast dyes, surgery, other clinical procedures).⁴

The criteria for diagnosing SM per the WHO Classification^2,6 and International Consensus Classification (ICC)^6,20 requires the presence of either 1 major criterion and 1 minor criterion, or at least 3 of 4 minor criteria in the absence of the major criterion.^3,6 The major criterion requires a biopsy to detect multifocal dense aggregates of mast cells in the bone marrow and/or other extracutaneous organs.⁶ The minor criteria include the presence of more than 25% of mast cells that are spindle-shaped or have atypical immature morphology in a bone marrow biopsy or in a section of other extracutaneous organs, mast cells in bone marrow, peripheral blood, or other extracutaneous organs expressing CD25, CD2, and/or CD30 markers in addition to mast cell markers, KIT D816V mutation, or another KIT mutation detected in bone marrow, peripheral blood, or other extracutaneous organs, and an elevated basal serum tryptase level that is persistently higher than 20 ng/mL.^3,6 The clinical expert consulted for this review noted that due to the heterogenous clinical presentation and the lack of awareness of this disease among clinicians and pathologists, AdvSM may be underdiagnosed.

The WHO and ICC classify SM into distinct subtypes in order of increasing disease burden: indolent SM and bone marrow mastocytosis, smouldering SM, ASM, SM-AHN, and MCL.^3-5 The classification system defines B-findings (indicative of high mast cell burden) and C-findings (indicative of SM induced organ damage) that are used to help establish the subtype of SM in addition to other histopathological findings such as the number of mast cells in the aspirate (for MCL) and the presence of an AHN (for SM-AHN). B-findings indicate a higher burden of SM, and include extensive bone marrow, spleen, or liver involvement without dysfunction.⁵ C-findings, commonly in patients with AdvSM, are defined by 1 or more signs of organ damage due to infiltration by neoplastic mast cells, including cytopenia(s) (i.e., an absolute neutrophil count < 1 × 10⁹/L, a hemoglobin level < 10 g/dL, and/or a platelet count < 100 × 10⁹/L due to bone marrow dysfunction), palpable splenomegaly with hypersplenism, skeletal involvement with large osteolysis (≥ 2 cm) with or without pathologic fractures, palpable hepatomegaly with impaired liver function and/or ascites and/or portal hypertension, and malabsorption with weight loss due to gastrointestinal mast cell infiltrates.⁴

AdvSM includes the disease variants of ASM, SM-AHN, and MCL.¹ Diagnosis of ASM is based on meeting the general criteria for SM (but not for MCL or SM-AHN), and the presence of 1 or more C-findings.⁴ ASM is an uncommon but clinically severe form of SM.⁵ Diagnosis of SM-AHN is based on meeting the general criteria for SM and simultaneously meeting diagnostic criteria for another hematologic neoplasm according to WHO diagnostic criteria (most commonly myelodysplastic or myeloproliferative neoplasms, or an overlap of both diseases), with or without C-findings.^4,5 In the ICC criteria, SM-AHN was revised to SM with an associated myeloid neoplasm (SM-AMN), given that AHNs of myeloid lineage are present in most patients (about 90%).^3,4 Patients with SM-AHN comprised about 70% of patients with AdvSM in KIT inhibitor trials.⁴ The diagnosis of MCL is based on the presence of 20% or greater neoplastic mast cells on a bone marrow aspirate smear; MCL can present as a de novo disease or transform from ASM, SM-AHN, or indolent SM (in rare cases).⁴ MCL is the rarest and most aggressive subtype of SM.⁵

Several prognostic scoring tools have been developed for risk stratification of patients with SM.³ The Mayo Alliance Prognostic System and the Mutation-Adjusted Risk Score combine clinical variables and high molecular-risk mutations for risk stratification. The International Prognostic Scoring System for Mastocytosis is based on clinical variables only.⁴ The Global Prognostic Score for Mastocytosis is based on clinical variables that are prognostic for OS and PFS.⁴ Across the scoring systems, factors associated with poor prognosis among patients with AdvSM include age (≥ 60 years), tryptase level (≥ 125 ng/mL), leukocyte level (≥ 16 × 10⁹/L), hemoglobin level (≤ 11 g/dL), platelet level (≤ 100 × 10⁹/L), skin involvement, and the presence of high-risk mutations (i.e., SRSF2, ASXL1, and/or RUNX1).⁴

The prevalence rate of SM is estimated at 1 per 10,000 people of all ages.⁶ Based on estimates from the Danish incidence and German prevalence of adults with AdvSM, the estimated AdvSM incidence rate in Canada is 0.06 cases per 100,000 adults and the prevalence rate is 5.2 cases per million adults.^7,8 Median OS has been estimated at 41 months for ASM,^4,9 11 months to 42 months for SM-AHN depending on the type of AHN,^4,10 and 2 months to 19.2 months for MCL.^3,4

Standards of Therapy

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

According to the clinical experts consulted for this review, the primary goals of treatment for patients with AdvSM are to improve health-related quality of life, improve the ability to work and maintain independence, prevent life-threatening anaphylaxis, reverse or delay the progression of end organ damage, and prolong survival.

The 2024 NCCN guidelines recommend a stepwise treatment approach to managing symptoms from mast cell mediator release.⁴ Antihistamines, cromolyn sodium, leukotriene receptor antagonists, and corticosteroids may be used to control skin, gastrointestinal, neurologic, cardiovascular, and other symptoms.⁴ Omalizumab or venom immunotherapy is recommended to prevent life-threatening anaphylaxis.⁴ The clinical expert consulted for this review stated that anaphylaxis is far less common among patients with advanced disease compared to nonadvanced SM, so for the population under review, anaphylaxis is not a major driver of morbidity and mortality.

Cytoreduction is the principal treatment for AdvSM; it may improve quality of life, reverse or prevent organ damage, and prolong survival.⁴ In Canada, cytoreductive treatment options for AdvSM include midostaurin, cladribine, interferons (e.g., peginterferon alfa-2a), and imatinib. According to the NCCN guidelines, enrolment in a clinical trial or a KIT inhibitor, such as midostaurin or avapritinib, are recommended as first-line treatment for AdvSM.⁴ Midostaurin is an oral multikinase inhibitor that is approved for the treatment of AdvSM in Canada. In 2020, CADTH issued a negative recommendation for midostaurin, and currently it is not reimbursed by public drug plans in Canada. Due to its high costs and the lack of compassionate programs, the clinical experts consulted for this review indicated that midostaurin in largely inaccessible in Canada.

Cladribine is not approved by Health Canada for SM but is used off-label for all variants of AdvSM. It is recommended if rapid disease debulking is required, which may be needed when patients present very acutely at diagnosis.⁴ The clinical experts indicated that ORRs and the duration of response with cladribine are variable. Infectious complications and myelosuppression are significant and frequent AEs that limit the long-term use of this medication, especially in patients who are very frail, which the experts stated is common at disease presentation.⁴

Interferons (e.g., peginterferon alfa-2a) have been used off-label in Canada for patients with AdvSM. Based on NCCN guidelines, they may be considered for patients with ASM with very slowly progressive disease, patients with severe refractory bone issues that have not responded to other lines of treatment, and patients with SM-AHN (typically when the AHN component requires treatment).⁴ Interferons are not an appropriate treatment for patients who present acutely and require immediate disease control to prevent life-threatening organ failure as response rates and response times are variable and unreliable. In clinical practice, the clinical experts stated interferon is almost never the preferred or adequate front-line treatment for AdvSM.

Imatinib is approved by Health Canada for AdvSM in those without a KIT D816V mutation or with an unknown KIT mutational status. Since more than 90% of patients with SM have a KIT D816V mutation, imatinib has a very limited role in the treatment of AdvSM. Allogeneic hematopoietic stem cell transplant may also be considered for patients with AdvSM.⁴

Drug Under Review

Key characteristics of avapritinib are summarized in Table 3, with other treatments available for AdvSM.

Avapritinib is a type 1 kinase inhibitor that binds to the active conformation and inhibits a broad range of KIT and PDGFRA mutant kinases at clinically relevant concentrations, including KIT D816V mutants and PDGFRA D842 mutants as well as multiple KIT exon 11, 11/17, and 17 mutants, sparing activity on a range of other kinases, including VEGFR2. The constitutive activation of KIT and PDGFRA receptor tyrosine kinases have been implicated in the pathogenesis of several malignancies and rare hematologic diseases, including SM and AdvSM. KIT D816V mutations are observed in about 95% of AdvSM cases.

Avapritinib was approved by Health Canada for the treatment of adult patients with AdvSM. AdvSM includes patients with ASM, SM-AHN, and MCL. The reimbursement request aligns with the indication proposed to Health Canada. Avapritinib is available as 25 mg, 50 mg, 100 mg, and 200 mg oral tablets. The recommended dosage for avapritinib is 200 mg as an oral tablet once daily until disease progression or unacceptable toxicity. Treatment with avapritinib is not recommended in patients with a platelet count of less than 50 multiplied by 10⁹/L.

The US FDA approved avapritinib for the treatment of adults with AdvSM, including patients with ASM, SM-AHN, and MCL, indolent SM, and unresectable or metastatic gastrointestinal stromal tumour harbouring a PDGFRA exon 18 mutation, including PDGFRA D842V mutations.²¹

The European Medicines Agency (EMA) authorized avapritinib as monotherapy for the treatment of adult patients with ASM, SM-AHN, or MCL, after at least 1 systemic therapy. Additionally, avapritinib was authorized by the EMA for the treatment of adult patients with indolent SM with moderate to severe symptoms inadequately controlled on symptomatic treatment.²² The EMA also authorized avapritinib for use in the European Union as monotherapy for the treatment of adult patients with unresectable or metastatic gastrointestinal stromal tumour harbouring the PDGFRA D842V mutation.

Table 3: Key Characteristics of Avapritinib, Midostaurin, Cladribine, Interferon, and Imatinib

AdvSM = advanced systemic mastocytosis; ANC = absolute neutrophil count; ASM = aggressive systemic mastocytosis; LVEF = left ventricular ejection fraction; MCL = mast cell leukemia; QTc = corrected QT interval; SM = systemic mastocytosis; SM-AHN = systemic mastocytosis with an associated hematologic neoplasm; SM-AHNMD = systemic mastocytosis with an associated clonal hematologic non–mast cell disorder.

^aHealth Canada–approved indication.

Sources: Product monograph for avapritinib,²³ product monograph for midostaurin,²⁴ product monograph for cladribine,²⁵ product monograph for peginterferon alfa-2a,²⁶ product monograph for imatinib,²⁷ and Sponsor Summary of Clinical Evidence.¹⁴

Perspectives of Patients, Clinicians, and Drug Programs

Patient Group Input

This section was prepared by the CDA-AMC review team based on the input provided by patient groups. The full patient and clinician group submissions received by CDA-AMC are available in the consolidated patient and clinician group input document for this review on the project website.

CDA-AMC received 2 patient group submissions from Heal Canada and the LLSC. Heal Canada is a not-for-profit organization that aims to empower patients, provide patient education and awareness, improve health care outcomes, and advocate for equitable access to quality health care. LLSC is a national charitable organization dedicated to finding a cure and improving quality of life for people and their families affected by blood cancers through research, educational resources, services, and support.

Heal Canada conducted an online survey of patients living with blood cancer (from February to May 2024) and attempted to conduct interviews of patients with AdvSM via outreach to Mastocytosis Society Canada and The Mast Cell Disease Society. Since no patient with AdvSM was recruited and no patient was identified to have had experience with avapritinib, the patient group submission was based on the MCC patient registry data and its publications. According to the MCC patient registry publications, SM is a rare disease with complex and variable clinical presentation, an unpredictable onset of symptoms with varying triggers, and diagnosis difficulties such as delays or misdiagnosis. Patients with AdvSM (n = 13) reported a median time of 3 years between symptom onset and the diagnosis. Nearly one-half of the patients consulted 3 to 6 physicians while seeking a diagnosis and only 40% of patients recalled undergoing a bone marrow biopsy, which is included in SM diagnostic recommendations. Based on the MCC patient registry data, 5% to 10% of patients with SM are diagnosed with advanced disease. Among the 13 patients with AdvSM in the MCC patient registry, the most common moderately to severely bothersome symptoms were fatigue (85%), difficulty concentrating (85%), nausea (77%), abdominal pain (69%), difficulty sleeping (69%), nonabdominal pain (62%), anxiety (62%), vomiting (54%), itching (54%), and depression (54%). The MCC patient registry data reported that patients with AdvSM experienced impacts on activities of daily living (92%), feelings of worry (92%), and family and social activities (85%). Quality of life was moderately or extremely severely reduced for 64% of patients.

LLSC conducted 1-on-1 interviews with 3 patients with SM (1 patient each with AdvSM, indolent SM, and unknown SM subtype) and 1 caregiver whose father had AdvSM. One interviewed patient with ASM and a caregiver of a patient with ASM reported on their experience with avapritinib. Patients and the caregiver described SM as a rare disease with symptoms that vary in manifestation (skin irritation, abdominal pain, body aches, diarrhea) and severity (from mild to debilitating), which can impact accurate and timely diagnosis. One caregiver shared that their father’s symptoms, which were initially self-diagnosed as allergies, led to multiple referrals to dermatologists with no clear diagnosis, concluding with an official diagnosis of mastocytosis at aged 92 years. A patient highlighted their lengthy journey that involved various health care providers (i.e., a clinician at a walk-in clinic, a dermatologist, a naturopath, and a rare disease clinician) and diagnoses (i.e., celiac disease and allergy to corn), and treatment resulting in AEs, all of which concluded with a confirmation of cancer after 5 years.

According to the LLSC, patients commonly described symptoms they experienced related to skin issues (e.g., rashes, itching, flushing, hives), body pain (e.g., relentless pain in bones, joints, connective tissues, and/or muscles), sleep disturbances (e.g., problems falling or staying asleep due to symptoms), and fatigue, as highlighted by a patient who reported waking up in the night with hives and itching, which were often combined with pain. According to LLSC, the accumulation of mast cells containing histamine, which are released in response to triggers (e.g., allergic reactions to food or medication) places patients with SM at a greater risk of unexpected anaphylactic shock; as such, patients are sometimes advised to carry an EpiPen and may be prescribed daily antihistamines for prophylaxis. Additional symptoms reported by patients included itchy and dry eyes, fever, and brain fog. In addition to the variability and severity in symptoms, their unpredictability in the location of presentation (e.g., face, extremities), timing (e.g., sudden onset), and acuity (e.g., life-threatening anaphylaxis) significantly impacts patients’ activities of daily living, their ability to work, and their quality of life. Burden of care was expressed as a concern by both patients and caregivers due to limitations in daily activities imposed by the symptoms of SM.

LLSC expressed a significant unmet need for treatment options among patients with AdvSM; available treatments are limited in effectiveness and access, with intolerable adverse effects. LLSC reported that 1 patient’s experience with midostaurin included nausea, vomiting, and diarrhea, such that additional treatment with alternatives (e.g., antinausea medication, cannabis products) were needed to offset the debilitating adverse effects to maintain food intake. Patients also reported experiences with other treatments targeting their symptoms, including pain medication, compound creams, antihistamines, and UV light therapy. One caregiver shared that their father was able to access avapritinib through a clinical trial. Avapritinib was able to alleviate their father’s skin issues and intense itching to provide the father with better quality of life, according to the caregiver, who also reported the treatment regimen to be straightforward and easy for both the caregiver and their father. The caregiver reported that during treatment with avapritinib, their father experienced positive results with limited adverse effects, and survived 3 years after being diagnosed with AdvSM. One patient with AdvSM experienced severe adverse effects with imatinib (Gleevec) treatment (e.g., vomiting, hives) and cladribine chemotherapy (e.g., sweating, sleep disturbance, gastrointestinal problems, bone marrow toxicity) before treatment with avapritinib. Via compassionate care access, the patient was able to undergo treatment with avapritinib for 2 years at the time of the patient group submission, expressing immense gratitude for its life-changing impact in alleviating symptom burden and mental strain due to the disease, with no observable adverse effects.

There is a lack of accessible and effective treatments for patients with AdvSM in Canada. Patients seek better treatments that address the underlying disease, provide symptom relief, have tolerable side effects, allow the restoration of daily activities, and improve physical and mental well-being.

Clinician Input

Input From Clinical Experts Consulted by CDA-AMC

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

Unmet Needs

The primary goals of treatment for patients with AdvSM are to improve health-related quality of life, improve the ability to work and maintain independence, prevent life-threatening anaphylaxis, reverse or delay progression of end organ damage, and prolong survival.

According to the clinical experts, there is a major treatment gap for patients with AdvSM in Canada. The off-label treatments currently available have low or unpredictable response rates, have a response of short duration, and may cause significant toxicity. The targeted therapy, midostaurin, is approved for the treatment of AdvSM in Canada but is not publicly funded and therefore is inaccessible. The clinical experts noted that internationally, KIT inhibitors are recommended by major guidelines as first-line treatment for patients with AdvSM.⁴ Most patients living in Canada with AdvSM do not have access to recommended first-line or second-line treatment, which represents a substantial care gap and equity issue; only patients who live near academic centres with the opportunity to participate in clinical trials or those with significant financial security may have access to recommended treatments.

Place in Therapy

The clinical experts stated that avapritinib would be used as first-line monotherapy in patients with AdvSM, except for patients who present very acutely and need rapid debulking with cladribine. In these cases, often a single cycle of cladribine could be offered for immediate disease control, followed by monotherapy with avapritinib. Avapritinib should not be used in patients with a platelet count of less than 50 multiplied by 10⁹/L. A short course of midostaurin or cladribine may be used as initial cytoreduction to allow the platelet count to recover to greater than 50 multiplied by 10⁹/L, then transition to avapritinib.

According to the clinical experts, avapritinib will be the second treatment approved to address the underlying disease process in patients with AdvSM and would cause a shift in the current treatment paradigm. The first drug approved that impacts the disease process, midostaurin, is currently inaccessible to patients in Canada.

Patient Population

As per the pivotal trials and the clinical expert input, patients most suitable for treatment with avapritinib include those who meet the WHO diagnostic criteria for AdvSM who are treatment-naive or who have received prior therapy for AdvSM. All subtypes of AdvSM (ASM, SM-AHN, and MCL) are predicted to benefit from avapritinib monotherapy, and the experts stated that in particular, patients with life-threatening organ damage and debilitating symptoms should be considered for treatment.

There is no companion diagnostic test needed other than the investigations outlined in the WHO diagnostic criteria, fifth edition. One expert indicated that there is some potential for the misclassification of patients with SM, such as those with indolent SM being classified as ASM. No other issues related to diagnosis or misdiagnosis were raised by the clinical experts. Patients with other mast cell disorders, such as mast cell activation syndrome, should not receive avapritinib.

Identifying which patients will have an optimal treatment response is not well defined and more research is needed in this area.

Assessing the Response Treatment

The clinical experts stated that the response criteria used in clinical trials continue to be frequently modified to best capture clinical benefit, better define long-term outcomes in the tyrosine kinase inhibitor era, and standardize response evaluation across clinical trials so that different drugs can be compared. Major response criteria that have been used in clinical trials and in clinical practice include the Valent criteria (plus its modified version), the Mayo criteria, the IMWG-MRT-ECNM criteria (plus its modified version), and more recently, the pure pathologic response criteria and the European Competence Network on Mastocytosis–American Initiative in Mast Cell Diseases criteria. The modified IWG-MRT-ECNM criteria used in the avapritinib trials are very specific and defining clinical responses to treatment can be complex due to the rigid criteria itself, drug adverse effects, and concomitant AHN, which makes it challenging to implement in a real-world setting. Many patients will have a partial response as per the aforementioned criteria but will have tremendous clinical improvement after starting treatment, which isn’t always captured by these response criteria. At this time, there are no clear data to suggest which response criteria perform better in terms of predicting long-term outcomes like survival.

According to the clinical experts consulted, the assessment of a clinically meaningful response requires the integration of patients’ goals of treatment with clinical and histopathological factors. From a clinical perspective, improvement in a patient symptom score is a critical part of the response assessment. Improvement in patients’ symptom scores often translates to improved quality of life and patient fitness. These, of course, are important factors for all patients but are additionally important if the goal is to improve fitness to proceed with an allogeneic hematopoietic stem cell transplant as patients with AdvSM can present with severe cachexia and poor performance status. This can limit the eligibility to proceed with an allogeneic stem cell transplant, which is the only curative treatment for AdvSM.

From a laboratory perspective, the normalization of complete blood counts and the reduction or absence of transfusion requirements not only improve patients’ quality of life but also liberates them and their caregivers from the time and financial burden of constantly needing to travel for count checks and transfusions. In those patients with liver involvement, the normalization of elevated liver enzymes, the reduction in liver size, and the reduction in the need for diuretics and/or therapeutic paracentesis have similar benefits on quality of life and reduced burden on the patient, caregivers, and health care system. Patients also commonly have splenomegaly and avapritinib can reduce spleen size, which is a key treatment target before an allogeneic hematopoietic stem cell transplant. The routine monitoring of basal serum tryptase levels is also an important surrogate for disease response.

From a histopathological perspective, a reduction in abnormal mast cell burden in bone marrow is a critical part of treatment response as well as treatment response to an AHN component such as improvement in blast count and the restoration of effective hematopoiesis. From a molecular perspective, the reduction in variant allele frequency of KIT variants is becoming an increasingly important marker for response to treatment, though 1 clinical expert stated that funding for this is not widely available in Canada.

Clinical and laboratory monitoring as summarized previously should be done at every visit, which is more frequent at the initiation of treatment. Outside of clinical trials, specific time points for the histopathological assessment of disease response are not well defined in the literature. Histopathological assessment should be done at 3 months after the initiation of avapritinib treatment and then can be considered every 3 months thereafter for the first 12 cycles of therapy and/or if there is evidence of loss of response or disease progression (e.g., worsening cytopenias or mastocytosis, increased peripheral blasts, increasing tryptase).

Discontinuing Treatment

As per the pivotal trials and input from the clinical experts, avapritinib should be discontinued in patients who are no longer getting clinical benefit from a symptom or quality of life perspective, in patients who experience intracranial bleeding, in patients with a platelet count lower than 50 multiplied by 10⁹/L, in patients with persistent severe treatment-related AEs that cannot be managed with dose interruptions or dose reduction, in patients who are pregnant, or in patients where there is evidence of progressive disease of either the SM or AHN disease component.

Prescribing Considerations

Based on NCCN guidelines⁴ and clinical expert input, patients with AdvSM should be treated by medical teams with expertise in the diagnosis of, treatment of, and response assessment in AdvSM. For patients who live far from a centre, a shared care model with local partners should be considered.

Clinician Group Input

This section was prepared by the CDA-AMC review team based on the input provided by clinician groups. The full original clinician group input received by CDA-AMC is available on the CDA-AMC website.

Two clinician groups provided input for this review: Ontario Health (Cancer Care Ontario) Hematology Cancer Drug Advisory Committee (based on 2 clinicians) and the LLSC Clinician Network and Myeloproliferative Neoplasms Canada Clinician Group (based on 6 clinicians).

In general, the clinician groups’ input was consistent with the input provided by the clinical experts consulted for this review. The clinician groups agreed that there is a significant treatment gap for patients with AdvSM in Canada, who have poor outcomes with high symptomatic burden and a need for reduced transfusions, improved treatment administration, and affordable therapies. Goals of treatment for patients with AdvSM include controlling the disease, improving health-related quality of life, preventing life-threatening anaphylaxis, reversing end organ damage, minimizing severe adverse effects, and prolonging survival.

The clinician groups agreed that patients with AdvSM should be managed by hematologists or medical teams with expertise in diagnosis, treatment, and response evaluation. Input from clinician groups indicated that standardized response criteria are evolving and may be used in conjunction with evaluations of clinical benefit, including patient-reported symptoms and health-related quality of life, laboratory measurements (e.g., complete blood count and tryptase levels, spleen volume), and histopathological measurements (e.g., bone marrow measurements of abnormal mast cell burden, blast count).

Neither Ontario Health (Cancer Care Ontario) nor the LLSC group of clinicians reported experience with avapritinib; however, they anticipated that avapritinib would be used as first-line monotherapy and eligible for use in any line of therapy among patients with AdvSM in Canada. Ontario Health (Cancer Care Ontario) indicated that patients with ASM would be best suited for treatment with avapritinib since the driver mutation is present in 95% of these patients. The group of clinicians anticipated that patients with all subtypes of AdvSM would benefit from avapritinib, particularly among those with life-threatening organ damage and debilitating symptoms. However, they acknowledged that identifying which patients would have an optimal response is unclear and requires further research. The clinician groups agreed that treatment with avapritinib should be discontinued among patients whose health-related quality of life is impacted by a lack of clinical benefit, a platelet count of less than 50 multiplied by 10⁹/L, detectable disease progression, or significant adverse effects.

Drug Program Input

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

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

AdvSM = advanced systemic mastocytosis.

Clinical Evidence

The objective of the CDA-AMC Clinical Review Report is to review and critically appraise the clinical evidence submitted by the sponsor on the beneficial and harmful effects of avapritinib oral tablets in the treatment of adult patients with AdvSM, including ASM, SM-AHN, and MCL. The focus will be placed on comparing avapritinib to relevant comparators and identifying gaps in the current evidence.

A summary of the clinical evidence included by the sponsor in the review of avapritinib is presented in 2 sections, with the CDA-AMC critical appraisal of the evidence included at the end of each section. The first section, the systematic review, includes pivotal studies and other studies that were selected according to the sponsor’s systematic review protocol. The CDA-AMC assessment of the certainty of the evidence in this first section using the GRADE approach follows the critical appraisal of the evidence. The second section includes indirect evidence from the sponsor. No long-term extension studies or studies addressing gaps in the evidence were submitted by the sponsor.

Included Studies

Clinical evidence from the following is included in the CDA-AMC review and appraised in this document:

• 2 pivotal studies identified in the systematic review

• 2 studies with indirect evidence (1 indirect treatment comparison, and 1 observational comparison).

Systematic Review

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

Description of Studies

Characteristics of the included studies are summarized in Table 5.

Two single-arm, open-label studies met the inclusion criteria for the systematic review: EXPLORER and PATHFINDER studies.

The EXPLORER study was a phase I study to evaluate the safety, tolerability, pharmacokinetics, pharmacodynamics, and efficacy of avapritinib. Eligible patients were adults with AdvSM and relapsed or refractory myeloid malignancies. The EXPLORER study was completed in 2 parts.

Part 1 was a dose escalation study to determine the maximum tolerable dose and recommended phase II dosage. Part 1 used a 3 plus 3 design, with the first patients receiving avapritinib 30 mg daily, escalating to a maximum of 400 mg per day (N = 32). The primary outcomes were safety, maximum tolerable dose, and recommended phase II dosage.

Table 5: Details of Studies Included in the Systematic Review

AdvSM = advanced systemic mastocytosis; AdvSM-SAF = Advanced Systemic Mastocytosis Symptom Assessment Form; AHN = associated hematologic neoplasm; ALT = alanine aminotransferase; AML = acute myeloid leukemia; ASM = aggressive systemic mastocytosis; AST = aspartate aminotransferase; BM = bone marrow; ECOG PS = Eastern Cooperative Oncology Group Performance Status; EORTC QLQ-C30 = European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire Core 30; ISM = indolent systemic mastocytosis; IWG-MRT-ECNM = International Working Group - Myeloproliferative Neoplasms Research and Treatment - European Competence Network on Mastocytosis; MC = mast cell; MCL = mast cell leukemia; ORR = overall response rate; PGI-S = Patient Global Impression–Severity; RAC = Response Assessment Committee; RAC-RE = Response Assessment Committee–Response Evaluable; SM-AHN = systemic mastocytosis with an associated hematologic neoplasm; SSM = smouldering systemic mastocytosis; ULN = upper limit of normal; VAF = variant allele frequency.

Note: C-findings include thrombocytopenia, anemia, and neutropenia due to BM involvement; splenomegaly with hypersplenism; and elevated AST and ALT, hypoalbuminemia, and ascites due to liver infiltration.

^aBased on protocol amendment 7, patients subsequently enrolled in cohort 1 of the EXPLORER study received a starting dose of 200 mg daily.

^bResponse-based outcomes were evaluated in the RAC-RE population.

^cPatients from both cohorts from part 2 of the EXPLORER study.

Sources: Clinical Study Reports for the EXPLORER study,^15,28 interim Clinical Study Report for the PATHFINDER study,¹⁶ clinical summary for the PATHFINDER study,¹⁷ and Sponsor Summary of Clinical Evidence.¹⁴ Additional data supplied by the sponsor (June 17, 2024).¹⁸

Part 2 of the EXPLORER study was an open-label extension to further evaluate the safety, tolerability, pharmacokinetics, and efficacy of avapritinib. For part 2, patients with a local diagnosis of AdvSM (N = 54) were enrolled and received an avapritinib starting dose of 300 mg daily. Protocol amendments were implemented during the EXPLORER study and with protocol amendment 6, a second cohort was added that enrolled patients who were evaluable according to the modified IWG-MRT-ECNM criteria (i.e., required to have at least 1 measurable C-finding attributable to SM, except for patients with a diagnosis of MCL). Patients in cohort 2 received a starting dose of 200 mg per day. Those who were not evaluable according to the modified IWG-MRT-ECNM criteria continued to be enrolled in cohort 1 and received a starting dose of 300 mg daily, until the implementation of protocol amendment 7, when the starting dose was reduced to 200 mg daily.

A total of 86 patients were enrolled from the US and UK, including 32 patients who entered part 1 and an additional 54 patients who entered part 2. No patients from part 1 continued into part 2 of the study. The first patient in the EXPLORER study was enrolled on March 10, 2016, an initial data cut-off was made on May 27, 2020, and a final data cut-off was made on January 19, 2023. In total, there were 4 data cut-off dates in support of regulatory requirements. In this report, data from the final data cut-off of January 19, 2023, have been summarized. The study schematic is shown in Figure 1. Outcome data from patients who received any dose of avapritinib (30 mg daily to 400 mg daily) were included in this report; however, the recommended starting dose is 200 mg daily according to the product monograph. Data from the avapritinib 200 mg subgroup (N = 21) have also been presented.

Figure 1: Study Schematic for the EXPLORER Study

AdvSM = advanced systemic mastocytosis; ASM = aggressive systemic mastocytosis; IWG-MRT-ECNM = International Working Group - Myeloproliferative Neoplasms Research and Treatment - European Competence Network on Mastocytosis; MCL = mast cell leukemia; RAC-RE = Response Assessment Committee–Response Evaluable; SM = systemic mastocytosis; SM-AHN = systemic mastocytosis with an associated hematologic neoplasm.

Sources: Created by CDA-AMC based on data from the final Clinical Study Report for the EXPLORER study¹⁵ and additional data supplied by the sponsor (June 17, 2024).¹⁸

The PATHFINDER study is an ongoing, single-arm, phase II study (Figure 2). Adults with AdvSM (ASM, myeloid SM-AHN, or MCL) were enrolled in 2 cohorts based on the presence of evaluable AdvSM C-findings at baseline. The objective of the study is to determine the adjudicated ORR according to the modified IWG-MRT-ECNM criteria. The study was to continue until 63 patients had the opportunity to receive 10 cycles, with a planned interim analysis when 32 patients were evaluable for response. The study includes patients from Canada, the US, and Europe. The first patient was enrolled on November 21, 2018, with an initial data cut-off on June 23, 2020 (62 patients) and 3 other data cut-offs for outcome adjudication and publication purposes (April 2021, September 2022, and September 2023). This report has summarized the data from the June 23, 2020, data cut-off date (the first data cut-off and planned interim analysis, with 62 patients) and the September 9, 2022, data cut-off date — hereafter referred to as the second data cut-off (with 105 patients). According to the sponsor, the final Clinical Study Report is expected to be available in the first quarter of 2025.

Figure 2: Study Schematic for the PATHFINDER Study

AdvSM = advanced systemic mastocytosis; AdvSM-SAF = Advanced Systemic Mastocytosis Symptom Assessment Form; ASM = aggressive systemic mastocytosis; CDA-AMC = Canada’s Drug Agency; ECOG = Eastern Cooperative Oncology Group; IWG-MRT-ECNM = International Working Group - Myeloproliferative Neoplasms Research and Treatment - European Competence Network on Mastocytosis; MCL = mast cell leukemia; ORR = overall response rate; QD = once a day; SM-AHN = systemic mastocytosis with an associated hematologic neoplasm; SSC = Study Steering Committee; TSS = total symptom score.

Note: The sponsor stated that this figure is based on the interim Clinical Study Report for the PATHFINDER study; 2 additional patients (to a total of 105) were enrolled after that cut-off and are accounted for in the clinical summary document for the PATHFINDER study. The CDA-AMC reviewer noted that the number of patients in cohort 1 and cohort 2 are inconsistent with the additional data supplied by the sponsor (June 17, 2024),¹⁸ which states 83 patients were included in cohort 1 and 24 patients were included in cohort.2.

Source: Interim Clinical Study Report for the PATHFINDER study.¹⁶

Populations

Inclusion and Exclusion Criteria

For the EXPLORER and PATHFINDER trials, eligible patients were aged 18 years or older and had 1 of the following diagnoses (based on WHO diagnostic criteria): ASM, SM-AHN, or MCL. For patients with SM-AHN, the hematologic neoplasm must have been myeloid or incidental indolent, low-grade, lymphoid AHNs (e.g., chronic lymphocytic leukemia) not requiring treatment. C-findings included thrombocytopenia, anemia, and neutropenia due to bone marrow involvement; splenomegaly with hypersplenism; elevated aspartate aminotransferase and alanine aminotransferase due to liver infiltration; hypoalbuminemia; and symptomatic ascites or pleural effusion requiring medical intervention.

In the EXPLORER study, patients with histologically or cytologically confirmed myeloid malignancy that was relapsed or refractory to standard treatments were eligible for enrolment to part 1, the dose escalation phase, but not part 2. In part 2, patients were eligible if they had a local diagnosis of AdvSM (whose subtypes are ASM, SM-AHN, or MCL), including those patients with or without measurable C-findings at baseline attributed to SM (according to the modified IWG-MRT-ECNM criteria). With protocol amendment 6, a second cohort was added that included modified IWG-MRT-ECNM–evaluable patients only (i.e., must have had at least 1 measurable C-finding per modified IWG-MRT-ECNM criteria at baseline, attributed to SM, unless the diagnosis was MCL, which did not require a C-finding).

Patients in the PATHFINDER study were required to have a bone marrow biopsy within 56 days of the first dose of avapritinib and be on stable doses of non-antineoplastic SM therapies and corticosteroids (as supportive care) for 14 days or more before screening. The PATHFINDER study enrolled adults with AdvSM into 2 cohorts.

• Cohort 1 — This cohort consisted of patients with 1 or more modified IWG-MRT-ECNM criteria for evaluable disease (i.e., had severe and quantifiable organ damage [an evaluable C-finding] or had MCL [regardless of C-findings]) as confirmed by the Study Steering Committee.

• Cohort 2 — This cohort consisted of patients who were not considered eligible for an adjudicated response and were confirmed centrally to have ASM or SM-AHN but were lacking an evaluable C-finding as determined by the Study Steering Committee.

The sponsor stated that patients were assessed by the Study Steering Committee throughout the study on whether they continued to meet the criteria for each cohort and could transition between cohorts, if their eligibility status changed.³³

In the PATHFINDER study, enrolment of the SM-AHN subgroup was capped at approximately 70% of the planned 63-patient sample size of cohort 1 (i.e., a maximum of 45 patients), which the sponsor stated was to ensure that the study population reflects the general AdvSM patient population.

Patients were excluded from the EXPLORER and PATHFINDER studies if they had liver, renal, or hematological abnormalities (low platelet or neutrophil count), eosinophilia and positivity for the FIP1L1-PDGFRA fusion, a diagnosis of high-risk or very high-risk acute myeloid leukemia, myelodysplastic syndrome, or Philadelphia chromosome–positive malignancy. Patients were also excluded from the EXPLORER and PATHFINDER studies if they underwent antineoplastic therapy less than 14 days before bone marrow assessment or had radiotherapy or major surgical procedures less than 14 days before the first dose of the study drug.

Interventions

In both studies, open-label avapritinib treatment was administered orally once daily in the morning from day 1 to day 28 in 28-day cycles. Dosing was continuous with no intercycle rest periods. The drug was supplied as capsules in the EXPLORER study and as tablets in the PATHFINDER study.

In part 1 of the EXPLORER trial, the first patients were administered avapritinib 30 mg once daily and in subsequent cohorts, the dose increased to 60 mg, 100 mg, 130 mg, 200 mg, 300 mg, and 400 mg daily. Dose escalation proceeded at increments of up to 100% until 1 or more patients treated at a given dose had at least a grade 2 nonhematologic AE or a grade 4 hematologic AE and the AE (nonhematologic or hematologic) was not clearly attributable to a cause other than avapritinib, or the dose exceeded the highest dose determined to be safe in the EXPLORER study. Based on part 1, the recommended phase II dose was set as 300 mg daily, and patients enrolled in part 2 initially received this dose. According to the sponsor, further data revealed that most patients who started at the 300 mg dose reduced their dose to 200 mg daily, most often due to cytopenia, fluid retention, and/or gastrointestinal or cognitive AEs. The study protocol was amended to add a second cohort that included only modified IWG-MRT-ECNM–evaluable patients who received a starting dose of 200 mg daily. Nonevaluable patients continued to enrol in cohort 1 at 300 mg daily; however, with protocol amendment 7 (dated April 17, 2019), enrolment at 300 mg daily ended and all patients enrolled at 200 mg daily.

In the EXPLORER study, intrapatient dose modification was allowed for patients in part 1 to minimize the number of patients treated at potentially inactive doses. Patients who completed 2 or more cycles without experiencing a grade 3 or higher nonhematologic toxicity or a grade 4 hematologic toxicity could have their dose escalated (up to the maximum tolerated dose). Patients who experienced dose-limiting toxicity during cycle 1 of part 1 had avapritinib treatment interrupted until the AE resolved (defined as ≤ grade 1 or the patient’s baseline value), with platelet counts of at least 25 multiplied by 10⁹/L and an absolute neutrophil count of at least 0.5 multiplied by 10⁹/L. After resolution of the AE, patients could resume avapritinib with a reduction of 1 dose level, or at the previous dose with the sponsor’s approval. If the patients required a dose interruption of more than 3 weeks due to an AE that was considered related to avapritinib, the treatment was discontinued.

During part 2 of the EXPLORER study (and after the dose escalation phase of part 1), dose modifications and re-escalation were allowed. Specific criteria were outlined for patients who developed thrombocytopenia, cognitive adverse effects, or another toxicity, with stepwise dose reductions (≥ 200 mg to 100 mg, then 50 mg, and 25 mg) and with treatment stopped if the AE occurred at the 25 mg daily dose level. After resolution of the AE, stepwise re-escalation up to the starting dose was allowed, with each dose level administered for 1 cycle before moving to the next higher dose. Dose interruptions of up to 56 days were allowed, and any longer interruptions had to be approved by the medical monitor. The study drug was discontinued if the patient experienced intracranial bleeding of any grade.

In the PATHFINDER study, the avapritinib starting dose was 200 mg daily. To manage thrombocytopenia and bleeding risk, the starting dose was reduced to 100 mg daily for patients with platelet counts of between 25 multiplied by 10⁹/L and 50 multiplied by 10⁹/L at baseline (protocol amendment 3). Subsequently, based on protocol amendment 5, patients with platelet counts of less than 50 multiplied by 10⁹/L at baseline were excluded from the enrolment. During the trial, dose reductions for thrombocytopenia or another toxicity were permitted, with patients first dropping to 100 mg daily, then 50 mg daily, and 25 mg daily. Patients on the 25 mg dose who required a dose reduction either had treatment interrupted or permanently discontinued. Doses could be interrupted for up to 56 days, but if nonhematologic avapritinib-related toxicity did not resolve to grade 2 or lower, or had not returned to baseline, then the sponsor’s medical monitor was contacted. Stepwise re-escalation from a reduced dose up to the initial starting dose was allowed, with each dose increase administered for 1 cycle before the next dose increase started. Dose escalation was allowed if the prior dose was well tolerated (e.g., no grade 3 or grade 4 AEs). After 8 weeks of treatment, the investigator was permitted to increase the dose to 300 mg daily (in the absence of toxicity) for suboptimal response. Suboptimal response was defined as no better than stable disease by day 1 of cycle 3 or later, according to the modified IWG-MRT-ECNM response criteria. If a patient experienced study drug-related intracranial bleeding of any grade, the study drug was permanently discontinued.

In the PATHFINDER study, palliative and supportive care for disease-related symptoms were permitted. This included histamine blockers, proton pump inhibitors, osteoclast inhibitors, leukotriene inhibitors, corticosteroids (not to exceed 20 mg of prednisone or an equivalent per day), cromolyn sodium or another mast cell stabilizer, and omalizumab. Antiemetics and hematopoietic growth factors could be used at the investigator’s discretion and in accordance with clinical guidelines. Similar criteria for concomitant medications were applied during the EXPLORER study.

Patients were prohibited from receiving strong CYP3A4 inhibitors, or strong or moderate CYP3A4 inducers, as well as any other antineoplastic treatments (excluding local radiotherapy to treat localized bone lesions) in both studies. Patients were withdrawn from the studies if they experienced study drug-related intracranial bleeding of any grade, pregnancy, death, confirmed SM progressive disease (4 weeks after initial documentation), confirmed AHN progressive disease requiring the initiation of other cytoreductive therapy, SM or AHN clinical progression (i.e., not meeting the modified IWG-MRT-EDNM criteria) requiring alternative cytoreductive treatment, the initiation of other cytoreductive therapy, the withdrawal of consent, or as the result of an investigator or sponsor decision. Patients may have been withdrawn due to AEs, nonadherence, protocol deviation, investigator decision, or loss to follow-up.

Outcomes

A list of efficacy end points assessed in this Clinical Review Report is provided in Table 6, followed by descriptions of the outcome measures. Summarized end points are based on outcomes included in the Sponsor Summary of Clinical Evidence as well as any outcomes identified as important to this review, according to the clinical experts consulted for this review and input received from patient and clinician groups and public drug plans. Using the same considerations, the review team selected end points that were considered the most relevant to inform the expert committee deliberations and finalized this list of end points in consultation with members of the expert committee. All summarized efficacy end points were assessed using GRADE. Select notable harms outcomes considered important for informing the expert committee deliberations were also assessed using GRADE.

Table 6: Outcomes Summarized From the Studies Included in the Systematic Review

AdvSM-SAF = Advanced Systemic Mastocytosis Symptom Assessment Form; ClinI = clinical improvement; CR = complete remission; CRh = complete remission with partial hematological recovery; IWG-MRT-ECNM = International Working Group - Myeloproliferative Neoplasms Research and Treatment - European Competence Network on Mastocytosis; ORR = overall response rate; OS = overall survival; PFS = progression-free survival; PR = partial remission; SAE = serious adverse event; TSS = total symptom score.

Note: One cycle is 28 days in length.

^aThere was a planned interim analysis of ORR once 32 patients enrolled in cohort 1 in the PATHFINDER study were evaluable for response. The null hypothesis was to be rejected if the 1-sided P value was less than 0.00625, and then the interim analysis would be used to support a marketing application.

^bAnalysis of AdvSM-SAF TSS at cycle 11, day 1, was a key secondary end point and analyzed in a sequential manner if the primary end point in the PATHFINDER study was met to control the studywise type I error rate.

Sources: Clinical Study Report for the EXPLORER study¹⁵ and interim Clinical Study Report for the PATHFINDER study.¹⁶

Based on input received, overall response rate, OS, duration of response, time to response, and patient-reported symptoms were identified as important outcomes. The clinical experts consulted for this review selected these end points based on their relevance to patients and their use in making treatment decisions. PFS was also considered relevant by the clinical experts and was used to inform the pharmacoeconomic analysis. While the patient input stated that adverse effects of treatment were a significant concern, SAEs, cognitive AEs, and intracranial bleeding were selected as the key harms for avapritinib due to their potential to cause substantial morbidity. The clinical experts discussed the heterogeneity among patients with AdvSM, and thus identified AdvSM subtype as subgroups of interest. Data for subgroups based on prior antineoplastic therapy were also included in the Clinical Review Report to inform the pharmacoeconomic analysis (specifically PFS). Based on input from the clinical experts, the GRADE assessment focused on the overall AdvSM population in the EXPLORER study regardless of the starting dose received, as the mean daily dose was similar to that of the PATHFINDER study, which used the Health Canada recommended starting dose of 200 mg daily.

At the request of the sponsor, data for the European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire Core 30 (EORTC QLQ-C30) global health score were added to Appendix 1 but were not evaluated using GRADE.

Overall Response

The primary end point for the PATHFINDER study, and a secondary end point for the EXPLORER study, was adjudicated ORR assessed by the modified IWG-MRT-ECNM criteria (Table 7), which the sponsor stated were modified from the original IWG-MRT-ECNM criteria in consultation with AdvSM experts, regulatory authorities, and authors who published the initial criteria, to address some issues and challenges that were noted in the EXPLORER trial.¹⁵ The key modification was to add a subcategory for patients with CRh. The modified criteria also allowed for a 4-week confirmation of progressive disease, rather than the 8-week time frame in the original criteria. Changes were also made to the criteria for patients with splenomegaly to remove criteria requiring the resolution of difficult to measure or subjective symptoms.¹⁵

Assessments of response were based on bone marrow biopsy and aspirate, serum tryptase, organomegaly on imaging and physical examination, and the assessment of ascites and pleural effusions, which were performed at baseline and repeated at cycle 3, cycle 5, cycle 7, cycle 11, and cycle 18, and then every 6 months thereafter (a bone marrow biopsy was not performed in cycle 5). Investigators were also asked for their subjective assessment if worsening in these findings was attributed to SM. Bone marrow samples, radiographs, and laboratory samples were evaluated centrally by independent specialists. The Response Assessment Committee (RAC) adjudicated response during response assessment meetings, held twice a year, using available data from all visits. The RAC reviewed central pathology assessments, tryptase levels, and imaging, together with local data to assess C-finding changes.

The ORR was defined as the proportion of patients in the RAC-RE population who had an adjudicated best response of CR, CRh, PR, or clinical improvement based on the modified IWG-MRT-ECNM criteria shown in Table 7. Patients meeting response criteria were confirmed 12 or more weeks after the initial response was documented.

Overall Survival

For both studies, OS was defined as the time from the start of treatment to the date of death. Patients who died before or on the data cut-off date were considered to have had an OS event. All patients who did not have a death record before or on the cut-off date were censored at the last date known alive. Survival follow-up occurred every 3 months until death or closure of the study by the sponsor.

Advanced Systemic Mastocytosis Symptom Assessment Form

The AdvSM-SAF is a patient-reported, 10-item symptom questionnaire that can be scored at the item level and domain level, and summed for a total daily score. It includes the severity of 8 symptoms over a 24-hour recall period (abdominal pain, nausea, vomiting, diarrhea, spots, itching, flushing, and fatigue) that were each scored on a numeric rating scale from 0 (indicating no symptoms) to 10 (indicating worst symptoms imaginable) and the frequency of vomiting and diarrhea using a discrete numerical value.³⁴ At the domain level, a gastrointestinal symptom score comprised 4 items (abdominal pain, nausea, vomiting, and diarrhea; the daily domain score was 0 to 40) and a skin symptom score comprised 3 items (spots, itching, and flushing; the daily domain score was 0 to 30).³⁴ The daily TSS was calculated by summing all items (excluding the frequency items) for a total score of 0 to 80.³⁴

The change from baseline in the AdvSM-SAF TSS was a secondary end point in the EXPLORER study part 2 and a key secondary end point in the PATHFINDER study. Patients in both studies completed the AdvSM-SAF daily using an electronic diary from 7 days before the first dose of the study drug to cycle 12, day 28, in the EXPLORER study, and up to cycle 17, day 28, in the PATHFINDER study. The individual symptom, domain, or TSSs are determined by the 7-day average for the prior week, where at least 4 of the 7 days have a score available.

The sponsor developed the AdvSM-SAF to address the lack of patient-reported tools validated among patients with AdvSM. The sponsor designed the AdvSM-SAF using concept elicitation interviews with clinician experts and patients with AdvSM for evaluating symptoms in patients with AdvSM.³⁴ Based on patients with AdvSM enrolled in cohort 1 (300 mg; n = 35) and cohort 2 (200 mg; planned n = 20) of the expansion phase of the EXPLORER study, 31 patients were included in the evaluation of the AdvSM-SAF using the reference measure of the EORTC QLQ-C30, the Patient Global Impression–Severity (PGI-S), and the ECOG PS.³⁴ Patients had a mean age of 63.7 (SD = 10.3) years and consisted of 51.6% females and 48.4% males, and most had an ECOG PS score of 1 (35.5%), a KIT D816V mutation (87.1%), and a median tryptase level of 217.8 ng/mL (range, 12.8 ng/mL to 765.3 ng/mL).³⁴ Construct validity evaluated using Spearman correlation coefficients indicated strong correlations for the TSS with the PGI-S (r = 0.614) and the EORTC QLQ-C30 on fatigue (r = 0.710), pain (r = 0.752), and nausea and vomiting (r = 0.811).³⁴ Among patients with a baseline and at least 1 follow-up visit score, internal consistency reliability was met (Cronbach alpha > 0.7) for all AdvSM-SAF items and the TSS.³⁴ Among patients with AdvSM-SAF scores at baseline who did not have a change in their ECOG PS score from baseline to day 18 of cycle 1, test-retest reliability between day 1 and day 8 of cycle 1 was met (intraclass correlation > 0.7) for all AdvSM-SAF items and the TSS.³⁴ Change scores in the gastrointestinal symptom score were moderately to strongly correlated with change scores in the PGI-S, serum tryptase levels, and EORTC QLQ-C30 items and domains (r = 0.240 to 0.697).³⁴ Change scores in the skin symptom score were moderately correlated with the change scores in ECOG PS, and the EORTC QLQ-C30 items of dyspnea, insomnia, and fatigue (r = 0.341 to 0.433).³⁴ Change scores in the TSS were moderately to strongly correlated with change scores in the PGI-S, ECOG PS, and EORTC QLQ-C30 items and domains (r = 0.306 to 0.812).³⁴ Using anchor-based methods, within-person MIDs were estimated as 6 to 9 for the gastrointestinal symptom score, 1 to 4 for the skin symptom score, and 9 to 14 for the TSS.³⁴ Using distribution-based methods, between-group differences were estimated as 2 to 4 for the gastrointestinal symptom score, 2 to 3 for the skin symptom score, and 4 to 7 for the TSS.³⁴

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

EORTC QLQ-C30 is a 30-item questionnaire that includes 5 functional domains (physical, cognitive, role, emotional, and social), 3 symptom scales (fatigue, nausea and vomiting, and pain), a global health status or quality of life scale, and 6 single items (dyspnea, insomnia, appetite loss, constipation, diarrhea, and financial difficulties). Each subscale and single item are evaluated on a standardized scale of 0 to 100; higher scores indicate better functioning (functional scales), better quality of life (the global health status scale), or poorer symptoms (symptom scales and single items). Patients completed the EORTC QLQ-C30 on day 1 of each cycle in the PATHFINDER study, and day 1 of cycle 1 to cycle 12 of part 2 of the EXPLORER study. The sponsor was not aware of a reported MID for the EORTC QLQ-C30 specifically in AdvSM but cited 2 publications of studies in other cancers. In patients with multiple myeloma, the estimated MID was 8 points to 12 points.³⁵ A review of 21 studies in 9 different cancer types (brain, breast, colorectal, head or neck, lung, melanoma, ovarian, and prostate) reported that most anchor-based MID estimates ranged from 5 points to 10 points but varied by scale, improvement or deterioration, within or between comparisons, and by cancer type.³⁶ The anchor-based MID for the within-group change in the global health status score ranged from 4 to 14 for improvement and −11 to −5 for deterioration.³⁶

Duration of Response

For the EXPLORER and PATHFINDER trials, duration of response was defined as the time from the first documented overall response to the date of first documented progressive disease, loss of response, or death due to any cause, whichever occurred first. Responses were determined by modified IWG criteria. Patients who were still in response at the data cut-off date were censored at their last valid assessment. Patients without confirmed response were excluded from this analysis.

Time to Response

In both studies, time to response was defined as the time from the start of treatment to the time an overall response by modified IWG-MRT-ECNM criteria was first met. Patients without confirmed response were excluded from this analysis.

Progression-Free Survival

In the EXPLORER and PATHFINDER trials, PFS was defined as the time from the start of treatment to the date of first documented progressive disease or death due to any cause, whichever occurred first. For patients without an event, PFS was censored at the date of the last valid assessment that noted loss of response or better. Development of acute myeloid leukemia was considered disease progression. Progressive disease was confirmed 4 weeks after initial documentation as per the modified IWG-MRT-ECNM criteria.

Safety

The sponsor identified 2 AEs of special interest, cognitive effects and intracranial bleeding; AEs related to these events were aggregated. In both studies, cognitive effects consisted of 4 preferred terms: cognitive disorder, confusional state, encephalopathy, and memory impairment, and intracranial bleeding consisted of 3 preferred terms: cerebral hemorrhage, hemorrhage intracranial, and subdural hematoma.

Treatment-emergent AEs were defined as any AE occurring during or after administration of the first dose of avapritinib through to 30 days after the last dose, and any event considered study drug related regardless of the start date. It also included events present at baseline that worsened or subsequently were considered to be treatment-related by the investigator. Disease progression was not reported as an AE. SAEs were any event that was fatal or life-threatening, or led to inpatient hospitalization or prolongation of hospitalization, persistent or significant disability or incapacity, or a congenital anomaly or birth defect, or was an important medical event that jeopardized the patient and required intervention to prevent 1 of the SAEs described earlier. Intracranial bleeding was considered an SAE.

Table 7: Modified IWG-MRT-ECNM Consensus Response Criteria in Advanced Systemic Mastocytosis

ANC = absolute neutrophil count; BM = bone marrow; ClinI = clinical improvement; CR = complete remission; CRh = complete remission with partial hematological recovery; Hb = hemoglobin; IWG-MRT-ECNM = International Working Group - Myeloproliferative Neoplasms Research and Treatment - European Competence Network on Mastocytosis; LOR = loss of response; MC = mast cell; PD = progressive disease; PR = partial remission; PRBC = packed red blood cell; SM = systemic mastocytosis.

Note: Guidelines for assessing response are only disease-related grade 2 or higher organ damage is evaluable as a primary end point; response assessments of CR, CRh, PR, SD, PD, and LOR should only be applied to these grade 2 or higher organ damage findings in the context of trials; disease status at the time of patient removal from the study singularly relates to the updated status of initial grade 2 or higher organ damage finding(s); the exclusion of drug-related toxicity and/or other clinical issues (e.g., gastrointestinal tract bleeding in the case of worsening anemia and/or transfusion dependence) should be undertaken before assigning the designation PD or LOR in a patient with a worsening of baseline grade 2 or higher organ damage.

^aResponses that are not maintained for a period of at least 12 weeks do not fulfill criteria for CR or CRh, PR, or ClinI; however, both maintained and unmaintained responses (< 12 weeks’ duration) should be recorded in the electronic case report form each time they are observed to measure the duration of response.

^bOnly valid as a response criterion if the pretreatment serum tryptase level is 40 ng/mL or more (i.e., if pretreatment serum tryptase is < 40 ng/mL, it will not be considered as a criterion in the evaluation of response).

^cBiopsy of organ(s) in addition to the BM to evaluate for SM-related organ damage may be considered.

^dOnly valid as a response criterion if the pretreatment BM MCs are 5% or more (i.e., if pretreatment BM MCs are < 5%, BM MCs will not be considered as a criterion in the evaluation of response).

^eThe preservation of at least 1 ClinI finding permits a patient to maintain the response of ClinI if 1 or more ClinI findings are lost but none meet criteria for PD. However, if 1 or more of the ClinI findings become PD, then the ClinI finding assignment is lost and the patient meets criteria for PD. The baseline value for evaluating PD is the pretreatment measurement(s). The PD findings must be considered related to the underlying disease and not to other clinical factors. Progression of an underlying chronic myeloid neoplasm to acute myeloid leukemia is also considered PD.

Source: Final Clinical Study Report for the EXPLORER study.¹⁵

Statistical Analysis

EXPLORER Study

The ORR was estimated using frequency, percentage, and 95% CI based on the exact binomial distribution (Clopper-Pearson) and Wald test P value (1-sided alpha = 0.025) in the RAC-RE population. Statistical testing was conducted versus the null value of 28%, which was based on the midostaurin registration trial (the D2201 study) response data that were calculated post hoc using the original IWG-MRT-ECNM criteria.³⁷ An alternate response rate (defined as CR, CRh, and PR by modified IWG criteria) was statistically tested versus the null of 17% (Wald test P value, 1-sided alpha = 0.025). The 17% ORR was based on a US FDA post hoc assessment using the IWG criteria for CR plus PR for midostaurin.³⁸ According to the statistical analysis plan, response rates using other criteria (e.g., CR + CRh) were reported descriptively with no statistical testing.

AdvSM-SAF data were collected during part 2 of the EXPLORER study. One-way analysis of a variance model was used to estimate the change from baseline in the all-AdvSM population that reported AdvSM-SAF data at baseline. A paired t test was performed at day 1 of cycle 3, cycle 7, and cycle 11 to explore whether the mean change was significantly different from 0. There was no imputation for missing data. Summary statistics for the change from baseline in the EORTC QLQ-C30 scores were reported, with no imputation for missing data.

Time-to-event analyses were estimated using Kaplan-Meier methods as per the PATHFINDER study. There was no control for multiplicity mentioned in the statistical analysis plan. No interim analysis was planned for this study.

PATHFINDER Study

The PATHFINDER study estimated the ORR using frequency, percentage, and 95% CI based on the exact binomial distribution (Clopper-Pearson) and Wald test P value in the RAC-RE population. Statistical testing for efficacy was conducted versus the assumed null value of 28%.³⁷ There was 1 interim analysis planned once 32 patients enrolled in cohort 1 were evaluable for response. The null hypothesis was to be rejected if the 1-sided P value was less than 0.00625, and then the interim analysis would be used to support a marketing application. If the 1-sided P value was 0.00625 or more at the interim analysis, the final analysis would be used to support a marketing application. The final analysis of the primary efficacy end point is planned to occur after 63 patients (with the SM-AHN subgroup capped at approximately 70%) are enrolled in cohort 1 and have had the opportunity to receive avapritinib treatment for at least 10 cycles or discontinued treatment earlier. The final primary efficacy analysis in this ongoing study will be tested at a 1-sided alpha level of 0.02178.

Statistical testing was planned for the proportion of patients who attained a best response of CR plus CRh plus PR against a null of 17%, using similar statistical methods as for the primary end point. However, this end point was not part of the statistical testing hierarchy used to control the familywise type I error rate. Other response categories were reported descriptively with no statistical testing, based on methods similar to those of ORR.

The key secondary outcome was the change from baseline in the AdvSM-SAF TSS. The change in AdvSM-SAF TSS was calculated from baseline to cycle 11, day 1, for patients on treatment for more than 10 cycles. The last observation carried forward method was used for patients who ended treatment before completing 10 cycles. The mean change was calculated for all patients in the safety population and tested against the null hypothesis of greater than or equal to 0 in the mean change of the TSS. The null hypothesis was rejected if the 1-sided 1-sample t test P value was less than 0.025. To control the studywise type I error rate, testing for this key secondary end point was to proceed only if the null hypothesis for the primary objective was rejected. A sensitivity analysis was planned for the change from baseline to cycle 7, day 1, for patients on treatment for more than 6 cycles, using the same statistical methods.

Summary statistics for the change from baseline in the EORTC QLQ-C30 scores were reported, with no imputation for missing data.

Time-to-event analyses (i.e., PFS, OS, duration of response, and time to response) were conducted using the Kaplan-Meier method. Median survival with a 2-sided 95% CI was estimated, as well as event rates at specific time points (e.g., 3 months, 6 months, 9 months).

Sample Size and Power Calculation

The PATHFINDER study was estimated to have 93.5% power to detect a difference for the adjudicated ORR versus the null hypothesis ORR of 28%. The power calculations were based on a sample size of 63 patients in the modified IWG-MRT-ECNM–evaluable cohort (cohort 1), using the exact 1-sample binomial test and a 1-sided type I error rate of 0.025. A sample size of 63 patients was estimated to have greater than 90% power to test the key secondary outcome, the change from baseline in the AdvSM-SAF TSS. This was based on an assumed null hypothesis of a mean change in the TSS of 0 or higher versus the alternative hypothesis of a mean change in the TSS of less than or equal to –10 with avapritinib treatment, an SD of a mean change in TSS of 20, and a 1-sided type I error rate of 0.025, in all treated patients in cohort 1 and cohort 2.

In the EXPLORER study, an estimated 25 patients would be enrolled in the dose-escalating part 1 of the trial. In part 2, the study planned to enrol approximately 35 patients (including 15 patients with ASM, 15 patients with SM-AHN, and 5 patients with MCL) in cohort 1 with a starting dose of 300 mg per day, and 20 patients in cohort 2 (with a starting dose of 200 mg per day).

Subgroup Analyses

Both studies analyzed ORR for the subgroups of age (< 65 years or ≥ 65 years), sex (male or female), region (North America or Europe), baseline SRSF2, ASXL1, and RUNX1 genotype (positive or negative), prior treatment with midostaurin (yes or no), and prior antineoplastic therapy (yes or no). The PATHFINDER study also analyzed PFS and OS for subgroups based on baseline SRSF2, ASXL1, and RUNX1 genotype, and prior antineoplastic or midostaurin use. Results were also presented according to AdvSM subtype in both studies (ASM, SM-ADH, or MCL) and by avapritinib dose in the EXPLORER study.

Of the subgroup analyses planned, prior antineoplastic treatment and AdvSM subtype were subgroups of interest to this review, based on clinical expert input and on informing the pharmacoeconomic analysis.

Table 8: Statistical Analysis of Efficacy End Points for the EXPLORER and PATHFINDER Studies

AdvSM-SAF = Advanced Systemic Mastocytosis Symptom Assessment Form; CI = confidence interval; IWG-MRT-ECNM = International Working Group - Myeloproliferative Neoplasms Research and Treatment - European Competence Network on Mastocytosis; LOCF = last observation carried forward; OS = overall survival; PFS = progression-free survival, RAC-RE = Response Assessment Committee–Response Evaluable; SSC = Study Steering Committee; TSS = total symptom score.

Sources: Final Clinical Study Report for the EXPLORER study and interim Clinical Study Report for the PATHFINDER study.^15,16

Analysis Populations

The primary efficacy outcome (ORR) and other response-related end points in both studies were based on the RAC-RE population (Table 9). This included all patients who had received at least 1 dose of avapritinib and were evaluable according to the modified IWG-MRT-ECNM criteria, had either completed at least 2 postbaseline bone marrow assessments and had been in the study for at least 6 cycles, or had an end-of-study visit. OS, AdvSM-SAF, and safety outcomes were reported for the safety population, which included all patients who had received at least 1 dose of avapritinib. For the EXPLORER study, only patients with AdvSM were included in the efficacy analyses.

Table 9: Analysis Populations of the EXPLORER and PATHFINDER Studies

AdvSM = advanced systemic mastocytosis; DOR = duration of response; EOS = end of study, ISM = indolent systemic mastocytosis; IWG-MRT-ECNM = International Working Group - Myeloproliferative Neoplasms Research and Treatment - European Competence Network on Mastocytosis; PFS = progression-free survival; RAC-RE = Response Assessment Committee–Response Evaluable; SSC = Study Steering Committee; SSM = smouldering systemic mastocytosis.

Sources: Final Clinical Study Report for the EXPLORER study,¹⁵ interim Clinical Study Report for the PATHFINDER study,¹⁶ and Sponsor Summary of Clinical Evidence.¹⁴

Results

Patient Disposition

In total, 86 patients were treated with avapritinib in part 1 or part 2 of the EXPLORER study, including 69 patients with a diagnosis of AdvSM, and 57 patients who met the criteria for the RAC-RE population (part 1 = 18 patients; part 2 = 39 patients). A total of 21 patients received an avapritinib starting dose of 200 mg per day in either part 1 or part 2 of the trial. At the end of the EXPLORER study, all patients had stopped treatment with avapritinib. In the safety population, the most common reasons for discontinuing treatment were ███████ █████████ █████ ███ ██████ ██████████ ████ ███ ████████ ██ █████ ███ █████ ████████ ███████ ███████████ ████████ ███████ ██████ ████████ ██████████ ██ ███████ ██ ██████████████ ████████ █████ ███ (Table 10). Of the ██ patients who progressed during the study, ||████████ ████ progressed due to █████ ███████ ████████. Other reasons for progression were ███ ██ ██████████ ███████ ██████████████ ████████ ██ ██████████ ███ █ ███████ ████ ███ ██ ████ ████ ███████ ███ █████████████ ███████████████ █████████ ████ ███ █████████ ████████. The most common reasons for discontinuation from the study were ███████ ████████ ███████ ██ █████ ███████.

At the first data cut-off date in the PATHFINDER study, 62 patients were included in the analysis, of whom 32 patients from cohort 1 met the criteria for the RAC-RE population. In the safety population, 10 (16.1%) patients discontinued treatment and | ████████ ███████ had discontinued from the study. The most common reasons for stopping treatment were AEs (9.7%) and disease progression (4.8%). The reasons for stopping the study were death ██████ or AEs ██████ (Table 11).

At the second data cut-off date, a total of 105 patients had received avapritinib at a starting dose of 200 mg daily, and 81 patients were included in the RAC-RE population (cohort 1). In the safety population, 47 (44.8%) patients had stopped treatment, due mainly to AEs ████████ disease progression ███████ or sponsor decision ██████. A total of 34 (32.4%) patients discontinued the study because of █████ ████████ ██████████ ██ ███████ ██████ ██ █████ ███████ ██████.

The number of patients screened for inclusion in the trials was not reported in either study.

In both studies, the primary efficacy outcome was based on the RAC-RE population, which is a subset of the overall study population. The reasons for exclusion from the RAC-RE population are summarized in Table 12.

Table 10: EXPLORER Study Patient Disposition (Final Data Cut-Off at January 19, 2023)

███ █ █████ ███████ █████████ ██████ █ ████████ ██████████ █████████ ████████████████████

^a███ ██████ ██████████ ████████ ██ ████████ ████████ ██ ████ █ ███ ██ ████████ ████████ ██ ████ ██ ███ ███████████ ████████ ████ ███ █ ███ ██ ██ ██████████ ███ ██ ███ ██████████ ██ █ ███ ██ ███ ██████████

^b███ ██████ ██████████ ████████ ██ ████████ ████████ ██ ████ █ ███ ██ ████████ ████████ ██ ████ ██ ███ ███████████ ████████ ████ ███ █ ███ ██ ███ ██████████ ███ ██ ███ ██████████ ██ █ ███ ██ ███ ███████████

^c███ ███████ ███ ████████████ ██ █ █████████ █████████ ██████ ███ ████████ █████████ ███^|

Sources: Final Clinical Study Report for the EXPLORER study¹⁵ and Sponsor Summary of Clinical Evidence.¹⁴ Additional data supplied by the sponsor (June 17, 2024).¹⁸

Table 11: PATHFINDER Study Patient Disposition

AML = acute myeloid leukemia; PPRE = pure pathologic response evaluable; RAC-RE = Response Assessment Committee–Response Evaluable.

^aIncludes 32 patients from cohort 1 and 30 patients from cohort 2.

^bExcludes 2 patients who received avapritinib at a starting dose of 100 mg daily.

^cIncludes 81 patients from cohort 1 and 24 patients from cohort 2.

Sources: Interim Clinical Study Report for the PATHFINDER study,¹⁶ the PATHFINDER study clinical summary document,¹⁷ and Sponsor Summary of Clinical Evidence.¹⁴ Additional data supplied by the sponsor (June 17, 2024).¹⁸

Table 12: Reasons for Exclusion From the RAC-RE Population (EXPLORER and PATHFINDER Studies)

█████ █ ████████ ████████ █████████████ ███ █ ██████████ ████████ █████████████ ████ █████████ █████████ ████████████ ███ █ ███ ██ █████████████████████ █ █████████████ ███████ ████████████████████████ █████████ ████████ ███ █████████ ███ ████████ ██████████ ███████ ██

Source: Additional data supplied by the sponsor (July 15, 2024).³³

Baseline Characteristics

The baseline characteristics outlined in Table 13, Table 15, and Table 16 are limited to those that are most relevant to this review or were felt to affect the outcomes or interpretation of the study results.

Table 13 shows the baseline characteristics of the different analysis populations in the EXPLORER study. In this trial, 69 of the 86 (80%) patients enrolled were adjudicated as having AdvSM, 14 (16%) patients had indolent SM, 2 (2%) patients had smouldering SM, and 1 (1%) patient had chronic myelomonocytic leukemia without SM. Patients with a diagnosis of nonadvanced SM or AdvSM were included in the safety population, which was used for the reporting of harms. Efficacy outcomes were based on patients with AdvSM in the RAC-RE population (i.e., response and PFS), and the all-AdvSM safety population (OS). Generally, the characteristics of the patients were similar across the different analysis populations. The mean age of patients in the all-AdvSM group was ████ █████ ███ █ █████, 41% of patients were female and 59% of patients were male. The most common AdvSM subtype was SM-AHN (70%), followed by MCL (19%) and ASM (12%). In total, 70% of patients had an ECOG PS score of 0 or 1, with 20% and 10% of patients rated as having a score of 2 or 3, respectively. Most patients had received prior antineoplastic therapy for SM (59%), including 33% of patients who had received prior midostaurin. Table 14 provides a summary of the number of prior lines of therapy for AdvSM.

Table 13: EXPLORER Study Baseline Characteristics (Final Data Cut-Off at January 19, 2023)

AdvSM = advanced systemic mastocytosis; ASM = aggressive systemic mastocytosis; BM = bone marrow; BMMC = bone marrow mast cell; BMI = body mass index; CMML = chronic myelomonocytic leukemia; dx = diagnosis; ECOG PS = Eastern Cooperative Oncology Group Performance Status; ISM = indolent systemic mastocytosis; MC = mast cell; MCL = mast cell leukemia; RAC-RE = Response Assessment Committee–Response Evaluable; SD = standard deviation; SM = systemic mastocytosis; SM-AHN = systemic mastocytosis with an associated hematologic neoplasm; SSM = smouldering systemic mastocytosis.

^aRepresents combined data for patients with starting avapritinib doses of 30 mg, 60 mg, 100 mg, 130 mg, 200 mg, 300 mg, and 400 mg.

^bThe safety population included 16 patients with ISM or SSM, and 1 patient with CMML (without SM) based on central committee adjudication.

^cAt least 20% immature MCs in BM aspirate smears.

^dMC infiltration in BM biopsy of more than 30% and serum tryptase of more than 200 ng/mL.

^eSigns of dysplasia or myeloproliferation, in non–mast cell lineage(s), but insufficient criteria for a definitive diagnosis of a hematopoietic neoplasm, with normal or only slightly abnormal blood counts.

^fOrganomegaly (without impaired organ function), including palpable hepatomegaly and/or palpable splenomegaly without hypersplenism and/or palpable lymphadenopathy (or on CT or ultrasound > 2 cm).

^gCytopenia(s), including an absolute neutrophil count of less than 1.0 multiplied by 10⁹/L, hemoglobin of less than 10 g/dL, or a platelet count of less than 100 multiplied by 10⁹/L.

^hHepatomegaly with impaired liver function — namely, elevated transaminases and/or bilirubin levels and/or hypoalbuminemia (with or without ascites or portal hypertension).

ⁱPalpable splenomegaly with hypersplenism (e.g., as documented by thrombocytopenia — namely, platelet count < 100 × 10⁹/L).

^jMalabsorption with hypoalbuminemia and/or significant weight loss defined as more than 10% weight loss over the last 6 months.

^kSkeletal involvement with large osteolytic lesions and/or pathologic fractures.

Sources: Final Clinical Study Report for the EXPLORER study¹⁵ and Sponsor Summary of Clinical Evidence.¹⁴

For the PATHFINDER study, the baseline characteristics for the RAC-RE and safety populations for the first data cut-off are shown in Table 15 and for the second data cut-off in Table 16. At the first data cut-off on June 23, 2020 (N = 62), the mean age of the safety population was 67.5 (SD = 11.0) years, 45% of patients were female, and 55% of patients were male. The highest proportion of patients were diagnosed with SM-AHN (69%), followed by MCL (16%) and ASM (15%). Most patients had an ECOG PS score of 0 or 1 (69%), with 23% and 8% of patients rated as having a score of 2 or 3, respectively. A total of 68% of patients had received prior antineoplastic therapy, including 55% of patients who had received prior midostaurin.

Table 14: EXPLORER Study Number of Prior Systemic Treatments for AdvSM (Final Data Cut-Off at January 19, 2023)

AdvSM = advanced systemic mastocytosis; RAC-RE = Response Assessment Committee–Response Evaluable.

Source: Additional data supplied by the sponsor (July 15, 2024).³³

As for the second PATHFINDER study data cut-off on September 9, 2022 (N = 105), the sponsor provided limited information on the characteristics of the patients analyzed, but based on the data available, the characteristics are generally similar to those of the initial data cut-off. In the safety population, the median age of patients was 68 years (range, 31 years to 88 years), with 42% female patients and 58% male patients. The proportion of patients with SM-AHN, ASM, and MCL was 66%, 20%, and 14%, respectively. The ECOG PS score was rated as 0 or 1 (74%), 2 (18%), and 3 (8%), and 64% of patients had received prior antineoplastic therapy (including midostaurin for 53% of patients). A summary of the number of prior lines of therapy for AdvSM is listed in Table 17.

Table 15: PATHFINDER Study Baseline Characteristics, All Doses (First Data Cut-Off at June 23, 2020)

AdvSM = advanced systemic mastocytosis; AHN = associated hematologic neoplasm; ASM = aggressive systemic mastocytosis; BMI = body mass index; CEL = chronic eosinophilic leukemia; CMML = chronic myelomonocytic leukemia; CRF = case report form; dx = diagnosis; ECOG PS = Eastern Cooperative Oncology Group Performance Status; ISM = indolent systemic mastocytosis; MC = mast cell; MCL = mast cell leukemia; MDS = myelodysplastic syndrome; MPN = myeloproliferative neoplasm; MPN-U = myeloproliferative neoplasm-unclassifiable; PPRE = pure pathologic response evaluable; RAC-RE = Response Assessment Committee–Response Evaluable; SD = standard deviation; SM = systemic mastocytosis; SM-AHN = systemic mastocytosis with an associated hematologic neoplasm; SSC = Study Steering Committee; SSM = smouldering systemic mastocytosis.

^aNote that this refers to patients who had MCL and no C-finding.

^bIncluding an absolute neutrophil count of less than 1.0 multiplied by 10⁹/L, hemoglobin of less than 10 g/dL, or a platelet count of less than 100 multiplied by 10⁹/L.

^cThat is, elevated transaminase and/or bilirubin levels and/or hypoalbuminemia (with or without ascites or portal hypertension).

^dFor example, as documented by thrombocytopenia (i.e., platelet count < 100 × 10⁹/L).

^eWeight loss defined as more than a 10% loss over the last 6 months.

Sources: Interim Clinical Study Report for the PATHFINDER study¹⁶ and Sponsor Summary of Clinical Evidence.¹⁴

Table 16: PATHFINDER Study Baseline Characteristics, Avapritinib 200 mg Dose (Second Data Cut-Off at September 9, 2022)

AdvSM = advanced systemic mastocytosis; ASM = aggressive systemic mastocytosis; BM = bone marrow; ECOG PS = Eastern Cooperative Oncology Group Performance Status; MCL = mast cell leukemia; NOS = not otherwise specified; RAC-RE = Response Assessment Committee–Response Evaluable; SM-AHN = systemic mastocytosis with an associated hematologic neoplasm; VAF = variant allele fraction.

^aPrior therapies are coded using WHO Drug Dictionary B2 Enhanced, version March 2017.

Sources: Clinical summary document for the PATHFINDER study¹⁷ and Sponsor Summary of Clinical Evidence.¹⁴

Exposure to Study Treatments

Table 18 summarizes treatment exposure for the final data cut-off of the EXPLORER study (safety population) and includes data from part 1 (dose escalation) and part 2 (extension). In part 1 (N = 32), the avapritinib starting dose was less than 200 mg, 200 mg, 300 mg, and 400 mg daily for 15 patients, 4 patients, 6 patients, and 7 patients, respectively. In the total study population (N = 86), most patients received a starting dose of 300 mg (43 [50%] patients), with 21 (24%) patients starting with 200 mg daily, 15 (17%) patients starting with less than 200 mg per day (i.e., 30 mg, 60 mg, 100 mg, or 130 mg), and 7 (8%) patients starting with 400 mg daily. The distribution of patients who received an avapritinib starting dose of less than 200 mg, 200 mg, or greater than 200 mg was ████ ████ ███ ███ in the all-AdvSM population (total N = 69), and ████ ████ ███ ███ in the RAC-RE population (total N = 57), respectively.

Table 17: PATHFINDER Study Number of Prior Systemic Treatments for AdvSM (First and Second Data Cut-Offs)

AdvSM = advanced systemic mastocytosis; RAC-RE = Response Assessment Committee–Response Evaluable.

Source: Additional data supplied by the sponsor (July 15, 2024).³³

In the EXPLORER study, the mean duration of treatment was ████ ██████ ███ █ █████ with a mean daily dose of █████ ██ ███ █ █████ for the safety population. For the 69 patients with AdvSM, the mean treatment duration was ████ ██████ ███ █ █████, and in the RAC-RE population (N = 57), the mean treatment duration was ████ ██████ ███ █ █████.¹⁸ Dose reductions due to AEs were common and dose-related, with ████ ████ ███ ███ of patients in the safety population reducing their dose at least once in the less than 200 mg, 200 mg, and 300 mg starting dose groups, respectively. Most patients reported ██ █████ | dose interruption due to AEs (███ ██ ███), with no clear dose relationship.

In the PATHFINDER study first data cut-off (on June 23, 2020), the mean treatment duration was ███ ██████ ███ █ ████ in the RAC-RE population, and ███ ██████ ███ █ ████ in the safety population. The mean daily dose was █████ ██ ███ █ █████ (Table 19). ███ ██████ ██ ████████ had at least 1 dose reduction and ███ had a dose interruption due to AEs. As of the September 9, 2022, data cut-off date, among patients who received a 200 mg avapritinib starting dose, the mean treatment duration was ████ ██████ ███ █ █████ in the RAC-RE population and ████ ██████ ███ █ █████ in the safety population.¹⁸ The median average daily doses of avapritinib were █████ ██ ███████ ████ ██ ██████ in the subgroup of patients who were previously treated and █████ ██ ███████ ████ ██ ██████ in the subgroup of patients who were previously untreated.

During the EXPLORER (for the final data cut-off) and PATHFINDER (for the first data cut-off) studies, ███ ███ ███ of patients were receiving drugs for peptic ulcers, ███ ███ ███ were receiving antihistamines, ███ ███ ███ received diuretics, ███ ███ ███ received corticosteroids, and ███ ███ ███ received opioids, respectively (Table 20). Concomitant medications were not reported in the sponsor-provided clinical summary document for the second PATHFINDER study data cut-off. Neither study provided information on subsequent therapies received.

Table 18: EXPLORER Study Patient Exposure Based on Avapritinib Starting Dose (Safety Population — Data Cut-Off at January 19, 2023)

AE = adverse event; SD = standard deviation.

^aDuration of treatment is defined as ([treatment end date – treatment start date] + 1) ÷ 30.4375.

^bCumulative dose is defined as the sum of all doses actually taken.

^cAverage daily dose: Cumulative dose divided by the number of days actually dosed.

Sources: Final Clinical Study Report for the EXPLORER study¹⁵ and Sponsor Summary of Clinical Evidence.¹⁴

Table 19: PATHFINDER Study Patient Exposure Based on Avapritinib Starting Dose (Safety Population — First Data Cut-Off at June 23, 2020)

AE = adverse event; SD = standard deviation.

^aDefined as ([treatment end date – treatment start date] + 1) ÷ 30.4375.

^bDefined as the sum of all doses actually taken.

^cDefined as the cumulative dose divided by the number of days actually dosed.

Sources: Interim Clinical Study Report for the PATHFINDER study¹⁶ and Sponsor Summary of Clinical Evidence.¹⁴

Table 20: Concomitant Therapy (Safety Population)

Sources: Final Clinical Study Report for the EXPLORER study¹⁵ and interim Clinical Study Report for the PATHFINDER study.¹⁶

Efficacy

Overall Response Rate

Table 21 provides a summary of the adjudicated best response according to the modified IWG-MRT-ECNM criteria in the RAC-RE population in both pivotal trials. The ORR was defined as patients with a best response of CR, CRh, PR, or clinical improvement, which was the primary outcome in the PATHFINDER study and a secondary outcome in the EXPLORER study. The observed ORR was similar in the final data cut-off of the EXPLORER study ██████ ████ ███ █████ ██ ███████ | | ███ as in the first data cut-off of the PATHFINDER study (75.0% ███████ ███ █████ ██ ███████ | | ███ and the second data cut-off at 74.1% (95% CI, 63.1% to 83.2%; N = 81).

According to the statistical analysis plans, the ORR was tested versus the 28% null value in all patients with AdvSM in the RAC-RE population who received any dose of avapritinib in part 1 or part 2 of the EXPLORER study (N = 57) and any dose of avapritinib in the preplanned interim analysis (i.e., first data cut-off) of the PATHFINDER study (N = 32). In both studies, the P value was less than 0.0001 based on a 1-sided test (alpha = 0.025 for EXPLORER; alpha = 0.00625 for PATHFINDER interim analysis).

CR was attained by █████, 0%, and 13.6% of patients in the EXPLORER and PATHFINDER studies’ first and second data cut-offs, respectively. Data for other response categories are shown in Table 21.

The ORR for the subgroup with ASM was ████, 100%, and 76.9% in the EXPLORER and PATHFINDER studies’ first and second data cut-offs, respectively (Table 41, Table 42, and Table 43 in Appendix 1). For patients with SM-AHN, the ORR was █████, 80.8%, and 75.5%, and for patients with MCL the ORR was █████, 25.0%, and 66.7% in the EXPLORER and PATHFINDER studies’ first and second data cut-offs, respectively. For patients with and without prior antineoplastic therapy, the ORR was █████ ██████ █████ in the EXPLORER study, 73.9% versus 77.8% in the PATHFINDER study’s first data cut-off, and 64.7% versus 90.0% in the PATHFINDER study’s second data cut-off, respectively (Appendix 1, Table 46).

Overall Survival

OS was an exploratory outcome in the EXPLORER study and a secondary outcome in the PATHFINDER study. The median survival was not reached for either study, as ██ ██ ██ ██ ████████ ███████ were alive at the end of the EXPLORER study, ██ ██ ██ ████████ ███████ were alive at the interim analysis of the PATHFINDER study, and 84 of 105 patients were alive at the PATHFINDER study’s second data cut-off (80.0%).

Table 21: Adjudicated Best Response by Modified IWG-MRT-ECNM Criteria (RAC-RE Population)

AdvSM = advanced systemic mastocytosis; CI = confidence interval; ClinI = clinical improvement; CR = complete remission; CRh = complete remission with partial hematological recovery; IWG-MRT-ECNM = International Working Group - Myeloproliferative Neoplasms Research and Treatment - European Competence Network on Mastocytosis; ORR = overall response rate; PR = partial remission; RAC-RE = Response Assessment Committee–Response Evaluable.

^aIncluded all starting doses of avapritinib (EXPLORER: 30 mg to 400 mg daily; PATHFINDER: 100 mg or 200 mg daily). In the EXPLORER study, the proportion of patients who received an avapritinib starting dose of less than 200 mg, 200 mg, or greater than 200 mg was 14%, 30%, and 56%, respectively, in the RAC-RE population. In the PATHFINDER study (first data cut-off), 1 (3%) patient received a starting dose of 100 mg; all other patients received avapritinib 200 mg.

^bStatistical test on binomial proportion of ORR against a null hypothesis of 28%, 1-sided alpha of 0.025, Wald test P value, for the all-AdvSM patient population (any dose of avapritinib in part 1 and part 2).

^cStatistical test on binomial proportion against a null of 28% was performed using a 1-sided alpha of 0.025. Null was rejected at the interim analysis as the 1-sided P value was less than 0.00625 (Wald test P value).

^dStatistical test on binomial proportion of ORR against a null hypothesis of 17%, 1-sided alpha of 0.025, Wald test P value, for the all-AdvSM patient population (any dose of avapritinib). The P value was not controlled for multiple testing.

Sources: Final Clinical Study Report for the EXPLORER study,¹⁵ interim Clinical Study Report for the PATHFINDER study,¹⁶ the PATHFINDER study clinical summary document,¹⁷ and Sponsor Summary of Clinical Evidence.¹⁴ Additional data supplied by the sponsor (June 17, 2024).¹⁸

In the EXPLORER study, the Kaplan-Meier estimates for the proportion of patients alive at 12 months, 24 months, 36 months, 48 months, and 66 months were ██████ ██████ ██████ ██████ ███ █████, respectively (Table 22). At the interim analysis of the PATHFINDER study, ██████ ██████ ███ █████ of patients were alive at 6 months, 12 months, and 18 months, respectively. As of the second data cut-off date, the Kaplan-Meier estimate for OS was 79.0% (95% CI, 70.8% to 87.3%) at 24 months (200 mg dose group; N = 105).

The Kaplan-Meier OS curve for the EXPLORER and PATHFINDER studies (second data cut-off, 200 mg dose group) for the safety study population and AdvSM subtypes are shown in Figure 3 and Figure 4. Data for OS in subgroups with and without prior antineoplastic therapy are shown in Appendix 1, Table 47 and Figure 9. The OS estimates for the RAC-RE population (N = 57) were similar to those in the all-AdvSM population (N = 69) in the EXPLORER study, and in the 2 populations based on the first data cut-off of the PATHFINDER study (RAC-RE population, N = 32; all-AdvSM population, N = 62).

Table 22: EXPLORER and PATHFINDER Studies Overall Survival (Safety Population)

AdvSM = advanced systemic mastocytosis; CI = confidence interval; NE = not estimable.

^aIncluded all starting doses of avapritinib (EXPLORER: 30 mg to 400 mg daily; PATHFINDER: 100 mg or 200 mg daily). In the EXPLORER study, the proportion of patients who received an avapritinib starting dose of less than 200 mg, 200 mg, or greater than 200 mg was 13%, 29%, and 58%, respectively, in the all-AdvSM population. In the PATHFINDER study (in the first data cut-off), 2 (3%) patients received a starting dose of 100 mg; all other patients received avapritinib 200 mg.

Sources: Final Clinical Study Report for the EXPLORER study,¹⁵ interim Clinical Study Report for the PATHFINDER study,¹⁶ the PATHFINDER study clinical summary document,¹⁷ and Sponsor Summary of Clinical Evidence.¹⁴

Figure 3: EXPLORER Study Overall Survival, All Doses (Safety Population — Final Data Cut-Off at January 19, 2023) [Redacted]

AdvSM = advanced systemic mastocytosis; ASM = aggressive systemic mastocytosis; ISM/SSM = indolent systemic mastocytosis/smouldering systemic mastocytosis; MCL = mast cell leukemia; SM-AHN = systemic mastocytosis with an associated hematologic neoplasm.

Source: Final Clinical Study Report for the EXPLORER study.¹⁵

Figure 4: PATHFINDER Study Overall Survival, Avapritinib 200 mg Dose (Safety Population — Second Data Cut-Off at September 9, 2022)

AdvSM = advanced systemic mastocytosis; ASM = aggressive systemic mastocytosis; MCL = mast cell leukemia; SM-AHN = systemic mastocytosis with an associated hematologic neoplasm.

Source: PATHFINDER study clinical summary document.¹⁷

Patient-Reported Outcomes

The AdvSM-SAF TSS captures the severity of 8 symptoms (abdominal pain, nausea, vomiting, diarrhea, spots, itching, flushing, and fatigue) and is scored from 0 (no symptoms) to 80 points (worst symptoms imaginable) based on the average daily score over the prior week. Using anchor-based methods, the estimated within-person MID was 9 points to 14 points for the TSS.¹

In the EXPLORER study, the AdvSM-SAF questionnaire was completed during part 2 only. At baseline, ██ ██ ██ ████████ █████ enrolled in part 2 had reported an AdvSM-SAF TSS, which dropped to ██ ████████ █████ at cycle 11. At baseline, the mean TSS was 19.1 (SD = 12.2) points (N = 40), with a mean change from baseline of ████ ██████ ████ ███ █████ ██ █████ | | ███ on cycle 3, day 1, and █████ ██████ ████ ███ █████ ██ █████ | | ███ at cycle 11, day 1 (Table 23).

In the PATHFINDER study, 56 of 62 (90%) patients reported a baseline TSS at the first data cut-off, and 91 of 105 (87%) patients reported a score at the second data cut-off. At baseline, the mean TSS was 18.3 (SD = 12.5) points (N = 56) at the first data cut-off date, with a mean change from baseline of –7.1 (SD = 9.9) points (N = 51) at cycle 3, day 1, and –9.8 points (95% CI, –14.9 points to –4.6 points; N = 22) at cycle 11, day 1. For the second data cut-off, the baseline TSS was not reported. The mean change from baseline was –5.4 (SD = 9.1) points (N = 90) at cycle 3, day 1, and –6.4 (SD = 9.7) points (N = 91) at cycle 11, day 1.

Table 23: Change From Baseline in AdvSM-SAF TSS (Safety Population — Patients With AdvSM)

AdvSM = advanced systemic mastocytosis; AdvSM-SAF = Advanced Systemic Mastocytosis Symptom Assessment Form; CI = confidence interval; GI = gastrointestinal; NR = not reported; SD = standard deviation; TSS = total symptom score.

^aIncluded all starting doses of avapritinib (EXPLORER: 30 mg to 400 mg daily; PATHFINDER: 100 mg or 200 mg daily). The distribution of doses received was not reported for this outcome.

^bTwo-sided P value of paired t test. The null hypothesis was that the mean change from baseline was 0.

^cThe P value was not controlled for multiple testing.

^dUnobserved patients were imputed using the last observation carried forward method.

^eAnalysis of AdvSM-SAF TSS at cycle 11, day 1, was a key secondary end point in the PATHFINDER study and was analyzed in a sequential manner if the primary end point was met to control the studywise type I error rate.

Sources: Final Clinical Study Report for the EXPLORER study,¹⁵ interim Clinical Study Report for the PATHFINDER study,¹⁶ the PATHFINDER study clinical summary document,¹⁷ and Sponsor Summary of Clinical Evidence.¹⁴

Duration of Response

Duration of response was a secondary outcome in both studies.

In the EXPLORER study, the median estimated duration of response for all 44 patients with AdvSM who met the overall response criteria in the RAC-RE population was ████ ██████ ████ ███ ████ ███████ ███ ██████████. The proportions of patients with AdvSM who maintained their response were █████ at 12 months, ████ | at 24 months, and █████ at 36 months (Table 24). Among patients with AdvSM who received a 200 mg starting dose (N = 11), █████ ███ █████ maintained their response at 12 months and 24 months, respectively.

Among patients who attained an overall response in the PATHFINDER study, the median estimated duration of response had not been reached at either the first data cut-off date (June 23, 2020) or the second data cut-off date (September 9, 2022). No patients had an event at the first data cut-off date (N = 24), whereas | ███████ of 60 patients had an event at the second data cut-off date. In the second data cut-off, the Kaplan-Meier estimates of ongoing response among all responders were 94.6% at 12 months, 88.8% at 24 months, 84.6% at 36 months, and 70.5% at 42 months (Table 24).

Time to Response

Time to response was an exploratory outcome in the EXPLORER study and a secondary outcome in the PATHFINDER study.

The median time to response was ███ ██████ ████ ███ ███ ██ █████ in the EXPLORER study, 2.0 months (95% CI, 0.3 months to 12.2 months) in the first data cut-off of the PATHFINDER study, and 2.2 months (95% CI, 0.3 months to 15.0 months) in the second data cut-off of the PATHFINDER study (Table 25).

Table 24: Adjudicated Duration of Responses by Modified IWG-MRT-ECNM Criteria (RAC-RE Population — Responders)

AdvSM = advanced systemic mastocytosis; CI = confidence interval; ClinI = clinical improvement; CR = complete remission; CRh = complete remission with partial hematological recovery; IWG-MRT-ECNM = International Working Group - Myeloproliferative Neoplasms Research and Treatment - European Competence Network on Mastocytosis; NE = not estimable; PR = partial remission; RAC-RE = Response Assessment Committee–Response Evaluable.

Note: Duration of response is defined as the time in months from the first documented response (CR, CRh, PR, or ClinI) to the date of the first documented progressive disease or loss of response or death due to any cause, whichever occurs first. Patients without confirmed response were excluded from this analysis. Patients who were still responding at the time of the data cut-off date were censored at their last valid assessment.

^aAll starting doses of avapritinib (EXPLORER: 30 mg to 400 mg daily; PATHFINDER: 100 mg or 200 mg daily). The distribution of doses received was not reported for this outcome.

Sources: Final Clinical Study Report for the EXPLORER study,¹⁵ interim Clinical Study Report for the PATHFINDER study,¹⁶ the PATHFINDER study clinical summary document,¹⁷ and Sponsor Summary of Clinical Evidence.¹⁴

Table 25: Adjudicated Time to Response by Modified IWG-MRT-ECNM Criteria (RAC-RE Population — Responders)

AdvSM = advanced systemic mastocytosis; ClinI = clinical improvement; CR = complete remission; CRh = complete remission with partial hematological recovery; IWG-MRT-ECNM = International Working Group - Myeloproliferative Neoplasms Research and Treatment - European Competence Network on Mastocytosis; PR = partial remission; RAC-RE = Response Assessment Committee–Response Evaluable; SD = standard deviation.

Note: Time to response is defined as the time in months from the start of treatment to the time that criteria for response (CR, CRh, PR, or ClinI) are first met. Patients without confirmed response were excluded from this analysis.

^aAll starting doses of avapritinib (EXPLORER: 30 mg to 400 mg daily; PATHFINDER: 100 mg or 200 mg daily). The distribution of doses received was not reported for this outcome.

Sources: Final Clinical Study Report for the EXPLORER study,¹⁵ interim Clinical Study Report for the PATHFINDER study,¹⁶ the PATHFINDER study clinical summary document,¹⁷ and Sponsor Summary of Clinical Evidence.¹⁴

Progression-Free Survival

PFS was an exploratory outcome in the EXPLORER study and a secondary outcome in the PATHFINDER study (Table 26).

At the final data cut-off date of the EXPLORER study, ██ ██ ██ ████████ ███████ reported an event and ██ ████████ were censored. The median PFS was ████ ██████ ████ ██ █████ ███ ██████████ among patients in the RAC-RE population who received any dose of avapritinib. The estimated PFS rates at 12 months, 24 months, 36 months, and 66 months were ██████ ██████ ██████ ███ █████, respectively.

At the interim analysis of the PATHFINDER study, | ██ ██ ████████ ███████ in the RAC-RE population had an event and ██ patients were censored, and at the second data cut-off date, 20 of 81 (24.7%) patients had events and 61 patients were censored. The median PFS was not estimable at either time point. For the second data cut-off, the estimated PFS rates at 12 months, 24 months, and 36 months were 83.5%, 76.5%, and 73.4%, respectively.

PFS Kaplan-Meier survival curves for both studies are available in Appendix 1. PFS according to prior antineoplastic therapy is shown in Appendix 1, Table 48, for the PATHFINDER study.

Table 26: Adjudicated Progression-Free Survival by Modified IWG-MRT-ECNM Criteria (RAC-RE Population)

AdvSM = advanced systemic mastocytosis; ASM = aggressive systemic mastocytosis; CI = confidence interval; IWG-MRT-ECNM = International Working Group - Myeloproliferative Neoplasms Research and Treatment - European Competence Network on Mastocytosis; MCL = mast cell leukemia; NE = not estimable; PFS = progression free survival; RAC-RE = Response Assessment Committee–Response Evaluable; SM-AHN = systemic mastocytosis with an associated hematologic neoplasm.

^aIncluded all starting doses of avapritinib (EXPLORER: 30 mg to 400 mg daily; PATHFINDER: 100 mg or 200 mg daily). In the EXPLORER study, the proportion of patients who received an avapritinib starting dose of less than 200 mg, 200 mg, or greater than 200 mg was 14%, 30%, and 56%, respectively, in the RAC-RE population. In the PATHFINDER study (the first data cut-off), 1 (3%) patient received a starting dose of 100 mg; all other patients received avapritinib 200 mg.

Sources: Final Clinical Study Report for the EXPLORER study,¹⁵ interim Clinical Study Report for the PATHFINDER study,¹⁶ and the PATHFINDER study clinical summary document.¹⁷

EORTC QLQ-C30 Global Health Score

Figure 12 and Figure 13 in Appendix 1 show the change from baseline over time for the EORTC QLQ-C30 global health status score. Both studies reported an increase in the mean global health score ranging from ████ ██████ ███ █ █████ ██ ████ ██████ ███ █ █████ at cycle 2, day 1 (N = 40), and cycle 12, day 1 (N = 21), respectively, of the EXPLORER study (final data cut-off date of January 19, 2023), and from 13.3 (SD = 24.3) points at cycle 2, day 1 (N = 80), 18.7 (SD = 28.0) points at cycle 11, day 1 (N = 62), and 20.9 (SD = 28.5) points at cycle 17, day 1 (N = 51), in the PATHFINDER study (second data cut-off date of September 9, 2022). At baseline, ██ ██ ██ ████████ ████) with AdvSM in the EXPLORER study provided EORTC QLQ-C30 results, with ███ ███ ███ of patients reporting data at cycle 2 and cycle 12, respectively. In the PATHFINDER study (second data cut-off), baseline data were available for 97 of 105 (92%) patients, 59% of patients at cycle 11 and 49% of patients at cycle 17.

Harms

Refer to Table 27 for harms data for the EXPLORER study final data cut-off, and the PATHFINDER study’s first and second interim data cut-offs, which had a mean treatment duration of ████ ██████ ███ █ ██████ ███ ██████ ███ █ █████ ███ ████ ██████ ███ █ █████, respectively.

Adverse Events

All patients in the EXPLORER and PATHFINDER studies reported at least 1 AE. The most commonly reported AEs in the EXPLORER and PATHFINDER studies (the second data cut-off) were periorbital edema (69% and 41%, respectively), anemia (57% and 51%, respectively), diarrhea (49% and 31%, respectively), thrombocytopenia (41% and 43%, respectively) and peripheral edema (41% and 47%, respectively).

In the EXPLORER study, ███ of patients experienced at least 1 AE that was at least grade 3 in severity, among which hematologic AEs were common, including ████████████████ ██████ ██████ █████ ███ ███████████ █████. Grade 3 or more severe AEs were reported by 68% of patients in the PATHFINDER study (in the first data cut-off). The most common events were anemia (16%), neutropenia ██████ █████████ ██████████ █████ ██████ ████████ █████ █████████ █████ and thrombocytopenia ████. For the second data cut-off, 83% of patients had experienced at least 1 grade 3 or higher event. ██████ ██████ ███████████ ██████ ███ ████████████████ █████ were reported most commonly.

Serious Adverse Events

SAEs were reported ████ ███ and 51% of patients in the EXPLORER study (in the final data cut-off) and the PATHFINDER study (in the first and second data cut-offs), respectively. In the EXPLORER study, the most common SAEs were ███████ ███████ █████████ ██████████ ████████ █████████ ████████████ ███████████ █████ ███████ █████████ ███ ███████. Limited data on specific SAEs were reported for the PATHFINDER study.

Withdrawals Due to Adverse Events

The proportion of patients who stopped treatment due to AEs was ██████ █████ and 23.8% in the EXPLORER study and in the first and second data cut-offs of the PATHFINDER study, respectively. ████████████ ███████████ █████████ █████████ ███████ ███ █████ ███████ ████████ were the most common reasons reported in the EXPLORER study. Limited information was available for the PATHFINDER study.

Mortality

█████ ████████ ██████ died due to AEs in the EXPLORER study. In the PATHFINDER study, 3 (4.8%) patients and ██ ████████ ██████ died due to AEs in the first and second data cut-offs. Table 27 provides a description of the cause of death.

Table 27: Summary of Harms Results From Studies Included in the Systematic Review (EXPLORER Study Final Data Cut-off, and PATHFINDER Study First and Second Data Cut-offs)

AE = adverse event; NR = not reported; SAE = serious adverse event; SD = standard deviation.

^aIn the EXPLORER study, 15%, 24%, and 58% of patients in the safety population received an avapritinib starting dose of less than 200 mg, 200 mg, or greater than 200 mg, respectively. In the PATHFINDER study (in the first data cut-off), 2 of 62 (3%) patients received a starting dose of 100 mg; all other patients received avapritinib 200 mg.

^bAEs reported in 20% or more of patients in any study population.

Sources: Final Clinical Study Report for the EXPLORER study,¹⁵ interim Clinical Study Report for the PATHFINDER study,¹⁶ the PATHFINDER study clinical summary document,¹⁷ and Sponsor Summary of Clinical Evidence.¹⁴

Notable Harms

Intracranial bleeding was identified as an AE of special interest by the sponsor, and by the clinical experts who were consulted for this review. In the EXPLORER study ██ ████████ ███████ experienced intracranial bleeding. In the PATHFINDER study, 1 (1.6%) patient in the first data cut-off and 4 (3.7%) patients in the second data cut-off experienced intracranial bleeding.

Cognitive AEs were common and were reported by █████, 11.3%, and 27.6% of patients in the EXPLORER study and in the first and second data cut-offs of the PATHFINDER study, respectively. These events included ██████ ███████████ █████████ █████████ ███████████ ██████ ████████ ███████████████ ███ ██████████.

Critical Appraisal

Internal Validity

The sponsor submitted information on a completed phase I trial (with 86 patients) and an ongoing phase II trial (with 62 patients and 105 patients in the first and second data cut-offs, respectively). Both pivotal trials were open-label, single-arm studies, and thus provided no direct evidence on comparative efficacy or safety. The lack of controlled trials has implications for the overall strength and interpretability of the results. With single-arm studies, there is an increased risk of bias in the estimation of treatment effects due to the potential for confounding related to natural history and prognostic factors. Moreover, the extent of any selection bias is difficult to ascertain. The clinical experts emphasized that AdvSM is a heterogeneous disease, and prognosis varies substantially based on the disease subtype and other factors. In the trials, few patients had a diagnosis of ASM or MCL, which typically have the most and least favourable prognosis of the 3 subtypes, respectively. Most patients had a diagnosis of SM-AHN, which the clinical experts stated has a variable prognosis, depending largely on the type of concurrent hematologic neoplasm. The distribution of AdvSM subtypes has the potential to affect the results, but the extent of the impact is unknown. In addition, the Clinical Study Reports did not provide any information on the number of patients screened for inclusion in the trial or the reasons for exclusion, so there is no information on the pool of patients who were considered for enrolment. The primary outcome (ORR) and other response-related outcomes were analyzed in a subset of patients enrolled in the studies, not in the entire population with AdvSM. The RAC-RE population consisted of patients who, at baseline, were evaluable according to modified IWG-MRT-ECNM criteria and who met the follow-up requirements (≥ 2 complete postbaseline bone marrow assessments and had been in the study for ≥ 6 cycles, or had an end-of-study visit). The RAC-RE population consisted of 83%, 52%, and 77% of the overall AdvSM population enrolled in the EXPLORER study and in the first and second data cut-offs of the PATHFINDER study, respectively. Although the baseline characteristics of the RAC-RE population and the overall study population appeared to be similar, and the OS estimates were comparable in both populations, the analysis of a subpopulation for the response-related end points is another potential source of selection bias. It is unclear if patients who were evaluable based on the modified IWG-MRT-ECNM criteria (i.e., had 1 or more C-findings attributable to AdvSM) are prognostically similar to patients who did not meet these criteria.

While the lack of a comparator group in the pivotal evidence limits the overall interpretation of the results, the feasibility of conducting a randomized controlled trial was low, given the rarity of AdvSM, and potential ethical issues were raised by the clinical experts consulted due to the efficacy and safety of the available comparators.

The primary outcome was based on ORR according to the modified IWG-MRT-ECNM criteria. The clinical experts noted that response criteria used are evolving to best capture clinical benefit and to better define long-term outcomes, given the availability of targeted therapies. Major response criteria that have been used in clinical trials and in clinical practice include the Valent criteria (plus its modified version), the Mayo criteria, the IMWG-MRT-ECNM criteria (plus its modified version), and more recently, the pure pathologic response criteria and the European Competence Network on Mastocytosis–American Initiative in Mast Cell Diseases criteria. According to the clinical experts consulted, there are no clear data to suggest which response criteria perform better in terms of predicting long-term outcomes like survival. In the pivotal trials, overall response was defined as patients with CR, CRh, PR, and clinical improvement, which the clinical experts agreed were relevant measures. The FDA decision on avapritinib was based on patients with CR or PR, excluding those whose best response was clinical improvement. The clinical experts consulted, however, stated that clinical improvement, without a PR or CR, can have a profound impact on a patient’s quality of life and survival; thus, they advocated for using the broader definition of response.

The pivotal trials were open-label, whereby the investigator and the study participants were aware of their treatment status, potentially increasing the risk of detection bias and performance bias. As such, the open-label trial design limits the interpretability of the subjective study outcomes, such as AdvSM-SAF, and AEs. The sponsor noted that there is subjectivity in assessing some components of the modified IWG-MRT-ECNM criteria, specifically if the C-findings are attributable to SM or to some other cause. The sponsor attempted to mitigate the reporting bias by requiring response outcomes to be adjudicated by a central committee, using bone marrow samples, radiographs, and laboratory samples that were evaluated centrally by independent specialists. Sensitivity analyses were conducted for ORR using an algorithm and investigator-assessed response. In the EXPLORER study, the ORR was similar for the RAC (█████), algorithm (█████), and investigator assessment of overall response (█████) which were based on the modified IWG criteria, as well as the ORR based on the unmodified IWG-MRT-ECNM criteria (█████). However, greater variation was noted across assessment methods in the PATHFINDER study interim analysis (June 23, 2020), which reported an algorithm-assessed ORR of █████ (based on the unmodified IWG-MRT-ECNM criteria) and an investigator-assessed ORR of █████ (based on modified IWG-MRT-ECNM criteria). Both estimates were lower than the centrally assessed ORR (75.0%), which was based on the modified IWG-MRT-ECNM criteria. The sponsor attributed the lower scores to the differences in the criteria and to the investigator basing decisions on data from response assessment visits only versus all study visits for the RAC. It is possible that the methods used to assess response could bias the findings, but the magnitude of any reporting bias remains unclear.

The sponsor initially supplied results from 2 data cut-offs for the PATHFINDER study. Results from these data cut-off dates are the focus of this report. The first data cut-off date was the protocol-specified interim analysis (June 23, 2020) and according to the sponsor, the second data cut-off date (September 9, 2022) was performed to provide data to the Study Steering Committee for regular response adjudication and for publication. Two other adjudication data cut-offs were conducted, dated April 2021 and September 15, 2023. Later on during the review, in response to an additional information request from the review team, the sponsor provided limited information on the results of the September 15, 2023, data cut-off, which appeared to be supportive of the earlier results submitted by the sponsor. For a brief summary of the updated findings, refer to the Interpretation of Results section. The primary outcome was tested versus a historical null value (28%), which was based on the midostaurin registration trial data and was calculated post hoc using the original IWG-MRT-ECNM criteria.³⁷ According to the study’s protocol, if at the interim analysis the primary outcome was statistically significant versus the 28% null value, these data would be used for marketing applications. Early assessment of study outcomes may be associated with inflated findings, and in this study, the interim analysis was conducted when 32 of the 62 patients were evaluable (information fraction = 51.6%). However, considering that subsequent data cut-off dates showed results similar to the interim data, some concerns regarding inflated results are allayed. The September 2022 and September 2023 data cut-offs provided data with a larger sample size and longer follow-up duration than the interim results, but since these were not based on preplanned interim analyses, these data should be interpreted as exploratory. Moreover, the P values reported for the September 2022 and September 2023 data cut-offs were not controlled for multiplicity; thus, there is an inflated familywise risk of type I error. There was no control of type I error in the EXPLORER study.

Patient input expressed how the symptoms of AdvSM may have a substantial impact on their quality of life. The clinical experts noted that changes in symptoms may not be captured in response criteria; consequently, an assessment of patient-reported symptoms was an important outcome in the clinical trials. The studies reported data for the AdvSM-SAF questionnaire, which the sponsor developed and evaluated using data from 31 patients from the EXPLORER study. Based on their analysis, the TSS showed good construct validity and reliability, and moderate responsiveness. The AdvSM-SAF results were not available for all patients, with data missing for ███ ███ ███ of patients at baseline, and for ███ ███ ███ of patients at cycle 11, for the EXPLORER and PATHFINDER studies (in the first data cut-off), respectively. There was no imputation for missing data in the EXPLORER study but in the PATHFINDER study, the last observation carried forward method was used to impute data for patients who stopped treatment before completing 10 cycles. Given the extent of missing data, this imputation method will overestimate the precision of the estimates, with 95% CIs that are too narrow relative to the amount of data actually observed. The AdvSM-SAF results may be biased due to the extent of missing data and the imputation methods used, limiting the interpretability of these data.

The trials reported Kaplan-Meier estimates for the time-to-event outcomes. OS, PFS, and time to response were exploratory end points in the EXPLORER study and secondary outcomes in the PATHFINDER study. The median OS was not met for either study, and the median PFS was not met for the PATHFINDER trial; thus, these data were considered immature. Additionally, comparative OS and PFS cannot be adequately assessed in a single-arm trial because all patients receive the same treatment. The FDA 2021 medical review report for Ayvakyt (avapritinib) oral (application number 212608) states that OS and patient-reported outcomes cannot be interpreted due to the absence of a preplanned comparator in the trials and the open-label design of the studies; therefore, according to the FDA report, the presentation of OS results had an exploratory purpose.¹¹

According to the EXPLORER study’s statistical analysis plan, efficacy outcomes were based on patients who received any dose of avapritinib. Only 21 of 86 (24%) patients received the Health Canada recommended starting dose of 200 mg daily. One-half of the patients enrolled received a 300 mg per day starting dose, 17% of patients received less than 200 mg per day (i.e., 30 mg, 60 mg, 100 mg, or 130 mg), and 8% of patients started with 400 mg daily. The dose distribution was similar for the RAC-RE population (████ ███ ███ ███ received < 200 mg, 200 mg, and > 200 mg starting doses, respectively). Despite the different starting doses, the average mean daily dose was comparable in the 200 mg subgroup (█████ ██) to the overall safety population (█████ ██). The results reported for the 200 mg subgroup were generally consistent with those of the overall population but should be interpreted with caution considering the limited sample size (17 patients and 20 patients for ORR and OS, respectively). It is unclear how variation in starting doses received impacted the efficacy findings. However, according to the sponsor, some AEs were associated with higher starting doses. Based on the data from the EXPLORER trial, the dosing and patient enrolment criteria were modified for the PATHFINDER study to improve safety.

Overall, the sample size of the trials was limited, with a total of 86 patients, 69 patients, and 57 patients providing data on safety, OS, and response end points, respectively, in the EXPLORER trial. For the PATHFINDER study, 62 patients provided safety and OS data and 32 patients provided response data in the first data cut-off. A total of 105 patients provided safety data and 86 patients provided response data in the second data cut-off. The magnitude of the treatment effect estimates observed in a relatively small study sample may not be replicable in a larger study sample. Interpretation of the safety data should consider the limited sample size and the mean exposure duration (████ ██████ ███ ████ ██████ for the EXPLORER and PATHFINDER studies, respectively). Although results were reported for subgroups by AdvSM subtypes, the ASM and MCL subgroups were particularly small (4 patients to 15 patients, depending on the outcome and data cut-off); thus, the results should be interpreted with caution. Results for the subgroups with and without prior treatments for SM may be confounded due to differences between groups at baseline and should be interpreted as exploratory.

The sponsor-submitted materials for both studies lacked clarity for aspects of the trial methods, particularly the disposition of patients in the different cohorts and data cut-offs, and the statistical analysis of the results. Further, the data supplied for the PATHFINDER study’s second data cut-off were incomplete and based on a summary document rather than on a Clinical Study Report. Several errors and omissions were found in the Sponsor Summary of Clinical Evidence template that was received for this review.

External Validity

The studies enrolled similar patient populations of adults with a confirmed diagnosis of AdvSM according to the WHO diagnostic criteria. The EXPLORER study also enrolled 20% of patients with other forms of SM, who were not consistent with the indicated population. Patients with nonadvanced forms of SM contributed to the safety data but not for efficacy outcomes in the EXPLORER trial. The average age of patients enrolled was late 60s, and most patients had a diagnosis of SM-AHN and an ECOG PS score of 0 or 1. In Canada, SM-AHN is the most common type of AdvSM, which is consistent with the studies. The majority of study patients had received systemic treatments for AdvSM, including 33% to 53% of patients who had received midostaurin. This treatment pattern is not consistent with clinical practice, as midostaurin is largely unavailable to patients with AdvSM in Canada. However, the experts consulted for this review did not believe that prior exposure to midostaurin would significantly impact response to avapritinib. The clinical experts consulted noted that the proportion of patients with high ECOG PS scores was lower than expected; thus, the study patients may have been less ill than patients who receive avapritinib in clinical practice. The trials excluded patients with comorbidities such as seizure disorder, long QT syndrome, HIV or active viral hepatitis, clinically significant uncontrolled cardiovascular disease, and reduced renal or hepatic function. In addition, patients with a higher risk of intracranial bleeding were excluded; thus, the safety and efficacy of avapritinib in these patients is unknown.

As noted earlier, only one-quarter of patients in the EXPLORER study received the recommended starting dose of avapritinib (200 mg), with most patients receiving a higher-than-recommended initial dose. This may impact the generalizability of the findings to the real-world population, particularly for safety, as the sponsor identified dose-related toxicities.

The outcomes reported in the trial were identified as important in the patient input submitted. While the clinical experts expressed that the components of the response criteria used in the trial to evaluate treatment response (e.g., mast cell burden in bone marrow; hematologic values and tryptase levels; the need for transfusions, diuretics, or paracentesis; a reduction in liver or spleen size and improvement in liver function tests) were relevant and appropriate, they noted that the detailed and specific nature of the IWG response criteria can make it time-consuming to use in clinical practice. Patients’ perception of the severity of symptoms and their impact on quality of life was also identified as important in the patient input submitted. While patient-reported symptoms were assessed in the trials, these data had limited internal validity and, therefore, their external validity cannot be assessed.

GRADE Summary of Findings and Certainty of the Evidence

Methods for Assessing the Certainty of the Evidence

For pivotal studies identified in the sponsor’s systematic review, GRADE was used to assess the certainty of the evidence for outcomes considered most relevant to inform the CDA-AMC expert committee deliberations, and a final certainty rating was determined as outlined by the GRADE Working Group.^12,13