Reimbursement Review

Brentuximab Vedotin (Adcetris)

Sponsor: BC Cancer Agency and Pediatric Oncology Group of Ontario

Therapeutic area: Hodgkin lymphoma

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 of Application Submitted for Review

HL = Hodgkin lymphoma; NA = not applicable; NOC = Notice of Compliance.

Source: Manufacturer’s Summary of Clinical Evidence¹ and brentuximab vedotin product monograph.²

Introduction

Hodgkin lymphoma (HL) is a B-cell malignancy that originates in the lymphocytes.³ Classical HL accounts for 95% of all HL cases. The estimated incidence in Canada in 2022 was 2.6 cases per 100,000.⁴ Based on 2018 Canadian Cancer Statistics, which report cancer incidence by stage, approximately 23.3% of patients in Canada presenting with HL have stage III disease and 22.7% have stage IV disease.⁵ Childhood HL represents 6% of all cancers and has an incidence rate of 12 cases per million (1.2 cases per 100,000) per year in patients aged 0 to 14 years.⁶ In 2019, a total of 25 children in Canada in this age group were diagnosed with HL.⁷ The clinical experts consulted by the review team noted that advanced-stage HL in adult patients refers to Ann Arbor stage III and IV HL. The clinical experts consulted for this review noted that, in clinical practice, pediatric patients with HL are usually classified into low-, intermediate-, and high-risk groups, and the high-risk group is generally considered equivalent to those with advanced-stage classical HL in adults. The clinical experts reported that pediatric patients with high-risk or advanced-stage HL are generally defined as having stage II with bulk tumour, stage III with B symptoms (stage IIIB), and stage IV with or without B symptoms (stage IVA and stage IVB). These patients are treated the same in clinical practice: as having advanced-stage HL. However, the clinical experts consulted by the review team also indicated that the definition of advanced-stage HL in pediatric patients is evolving and may vary by centre, with some centres defining any stage III or IV case of HL in pediatric patients as advanced-stage disease.

The goal of therapy in patients with advanced HL is curative.^8,9 The clinical experts consulted by the review team noted that patients with advanced HL are treated the same, regardless of stage. Current first-line treatment regimens for adult patients with advanced-stage HL rely on chemotherapy. For patients with stage IV HL, the preferred regimen uses brentuximab vedotin (BV) in combination with doxorubicin-vinblastine-dacarbazine (AVD). For patients with advanced HL, treatment approaches also include doxorubicin-bleomycin-vinblastine-dacarbazine (ABVD) for up to 6 cycles with PET response after 2 cycles of chemotherapy (PET2)-directed treatment adaptation, and based on upfront PET2-driven treatment adaptation with bleomycin-etoposide-doxorubicin-cyclophosphamide-vincristine-procarbazine-prednisone (BEACOPP).

For pediatric patients, the clinical experts consulted by the review team also noted that most clinical centres in Canada use doxorubicin-bleomycin-vincristine–etoposide-prednisone-cyclophosphamide (ABVE-PC) for 5 cycles with radiation therapy determined by PET2, while relatively fewer centres use vincristine-etoposide-prednisone-doxorubicin–cyclophosphamide-vincristine-prednisone-dacarbazine.

The objective of this report is to review and critically appraise the evidence submitted by the sponsor on the beneficial and harmful effects of brentuximab (50 mg per vial, IV infusion) for the treatment of previously untreated patients with advanced-stage HL, in combination with AVD. BV has been previously reviewed and recommended for reimbursement by the CDA-AMC pan-Canadian Oncology Review Expert Review Committee for the treatment of previously untreated patients with stage IV HL in combination with AVD.¹⁰

Stakeholder Perspectives

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

Patient Input

One patient group, Lymphoma Canada, provided input on the use of BV to treat previously untreated patients with advanced-stage HL, in combination with AVD. Patient input was gathered from an online anonymous patient survey from March 14 to May 2, 2023. A total of 26 responses were gathered, and 3 respondents reported receiving BV + AVD treatment.

Patients were asked questions regarding the physical and psychosocial symptoms experienced at the time of diagnosis, current quality of life, and how these symptoms affected their daily activities. At the time of their lymphoma diagnosis, most of the patients described fatigue (79%) as the most consequential symptom (5 out of 5), followed by enlarged lymph nodes (58%), shortness of breath (63%), and weight loss (47%). In addition, 74% of patients reported experiencing anxiety or worry, 68% stressing about their diagnosis, 63% difficulty in sleeping, and 58% fearing progression of their lymphoma. Regarding physical symptoms that currently affect their quality of life, out of 7 responses, fatigue (29%) and headaches (14%) were identified as the most significant factors having a negative impact on quality of life. The most psychosocial factors with the greatest impact were the stress of having cancer (71%), fear of progression (71%), anxiety or worry (71%), difficulty sleeping (43%), problems concentrating (43%), and inability to attend work or school (43%).

Among the surveyed patients, 3 reported receiving ABVD in the front-line setting, 2 were treated with other forms of chemotherapy, and 1 was treated with cyclophosphamide-doxorubicin-prednisone-rituximab-vincristine. While evaluating the importance of the outcomes of new treatments, patients from the survey prioritized the need for a novel lymphoma therapy to control disease symptoms, prolong disease remission, extend life spans, and improve quality of life.

While describing the experience with the treatment under review, 1 of the 3 patients who had received BV + AVD indicated they have been in remission for between 6 months and a year, another was in remission for longer than a year, and the other patient was in post-treatment (and unsure about their remission status). Side effects from the BV + AVD treatment reported by patients were fatigue (n = 3), neutropenia (n = 2), constipation (n = 2), joint or muscle pain (n = 2), low platelet count (n = 1), low blood pressure (n = 1), and decreased appetite (n = 1). Two patients reported experiencing financial setbacks — 1 due to absence from work and 1 due to the cost of other medications. One of these patients mentioned having a poor experience with BV, and the other 2 rated their experience as very good.

Clinician Input

Input From Clinical Experts Consulted by the Review Team

The clinical experts consulted by the review team noted that improving the proportion of patients cured with first-line treatment is an important unmet need for patients with advanced-stage HL. The clinical experts also emphasized the need to reduce treatment failure, prevent disease progression or relapse, and avoid late side effects (e.g., secondary malignancies and cardiac and pulmonary late effects) and therapies that are toxic (e.g., autologous stem cell transplant [ASCT]), particularly for younger patients diagnosed with advanced-stage HL and older patients who have poor tolerance to treatment.

The clinical experts noted that BV + AVD is considered a front-line therapy for advanced HL. The clinical experts noted that, at the time of the review, BV + AVD had been approved only for the treatment of previously untreated patients with stage IV HL, and pointed out that the use of BV + AVD in patients with stage III classical HL could shift the current treatment paradigm for those patients. The clinical experts indicated that, in pediatric patients, BV would be used in combination with a different chemotherapy backbone, namely the therapy investigated in a phase III randomized controlled trial (AHOD1331)¹¹ in pediatric patients: BV in combination with doxorubicin-vincristine-etoposide-prednisone-cyclophosphamide (AVEPC). Trials of BV + AVD in pediatric patients had not been completed at the time of this review.¹²

The clinical experts noted that any previously untreated adult patients with stage III or IV classical HL who meet the eligibility criteria of the ECHELON-1 trial are best suited for the use of BV + AVD. The clinical experts indicated that pediatric patients with advanced-stage classical HL could also be eligible for BV in combination with chemotherapy, and eligibility for therapy should be determined by the treating physician or based on the eligibility criteria of the AHOD1331 trial.

The clinical experts noted that PET scans, typically PET2 and a PET scan after all 6 cycles (end of treatment), are used to assess responses. In patients who have a complete response (CR), follow-up visits take place every 3 to 4 months for 2 years, then typically every 6 months for 3 more years. Patients with a partial response may undergo radiation therapy, and patients with refractory disease undergo further investigations (e.g., a biopsy) and treatment with a second-line regimen. One of the clinical experts, who specializes in pediatrics, noted that outcomes used in clinical practice to assess treatment response are generally aligned with outcomes typically used in adult therapeutics. The clinical experts noted that overall survival (OS) is the most clinically meaningful outcome to assess the efficacy of BV in combination with chemotherapy in patients with advanced-stage HL, and progression-free survival (PFS) is an important outcome. In pediatric patients, event-free survival (EFS) is also important.

The clinical experts agreed that discontinuation of BV + AVD is uncommon overall because unacceptable toxicity or refractory HL is not common. The clinical experts identified several situations in which BV + AVD can be discontinued, including completion of treatment, clear evidence of progression of disease, and an unacceptable adverse event (AE).

The clinical experts noted that diagnosis of the disease in adult patients must be made by an experienced pathologist. Selection of patients for BV + AVD should be made by a hemato-oncologist experienced with treating HL. Treatment can be delivered in the specialty clinics of nonacademic centres. The clinical expert specializing in pediatrics noted that all pediatric oncology patients are cared for by pediatric oncology teams at tertiary care centres. Some aspects of care may be provided at satellite centres after diagnostic and management decisions are made.

Clinician Group Input

Clinician group input on the review of BV was received from the Ontario Health (Cancer Care Ontario) (OH-CCO) Hematology Cancer Drug Advisory Committee and the Pediatric Oncology Group of Ontario (POGO). Six clinicians provided input on behalf of OH-CCO. POGO is a collaboration of Ontario’s 5 specialized childhood cancer centres. The input collected from POGO was prepared in a consultative manner, with 1 clinician discussing the indication with other members of the submission panel and seeking input from POGO’s Therapeutic and Technology Advisory Committee. POGO’s final submission was based on this process.

Input from OH-CCO emphasized the need to improve outcomes with first-line therapy to avoid the need for second-line therapy. The group noted that patients with stage III and IV disease would be best suited for current treatment. The group indicated that typical lymphoma response measures, including PET scans, are used in clinical practice to assess patients’ response to treatment. They noted they would discontinue treatment with BV + AVD in the event of significant toxicity or disease progression.

Input from POGO noted that, while a variety of chemotherapy and radiation approaches are available for use as standard of care, these vary by region and between pediatric- and adult-focused practitioners in Canada. POGO reported that, historically, the ABVD chemotherapy backbone used with BV in adult patients has not been used by pediatric oncologists to treat pediatric patients due to concerns regarding higher anthracycline (doxorubicin) and bleomycin exposure, as well as known dose-dependent cardiac and pulmonary toxicities. POGO noted that BV has been studied and used in combination with another chemotherapy regimen (AVEPC) in patients aged 2 to 21 years with previously untreated high-risk HL.¹¹ POGO indicated that this alternative chemotherapy backbone is more commonly used in the pediatric setting, and BV + AVEPC has become standard care for high-risk pediatric patients in Ontario. Regarding treatment goals in pediatric patients with HL, POGO emphasized the need to avoid disease recurrence to minimize potential late effects from subsequent chemotherapies and ASCT received at relapse, and the associated impact on health-related quality of life (HRQoL). While describing the outcomes used to determine whether a pediatric patient is responding to treatment for HL, POGO emphasized the importance of OS and EFS, considering the higher chance of experiencing late effects of therapy after treatment among the younger patient population. Like OH-COO, POGO suggested treatment be discontinued at disease progression.

Drug Program Input

Input was obtained from the drug programs that participate in our reimbursement review process. The following were identified as key factors that could potentially affect the implementation of a recommendation for BV:

• relevant comparators

• consideration for initiation of therapy

• consideration of discontinuation of therapy

• consideration for prescribing of therapy

• generalizability

• funding algorithm (oncology only)

• care provision issues.

The clinical experts consulted by the review team provided advice on the potential implementation issues raised by the drug programs (Table 4).

Clinical Evidence

Pivotal Studies and Randomized Controlled Trial Evidence

Description of Studies

One phase III, open-label, randomized, active-controlled, superiority trial (ECHELON-1, N = 1,334) was identified from a systematic literature review conducted by the sponsor.^13,14 The primary objective of the ECHELON-1 trial was to determine the efficacy of BV + AVD relative to ABVD as measured by modified progression-free survival (mPFS). The key secondary objective was to compare OS between BV + AVD and ABVD. The ECHELON-1 trial is ongoing. Data gathered at the cut-off dates of April 20, 2017, and June 1, 2021, were assessed for this review. New data from a descriptive analysis of OS conducted in response to a request for supplementary information from the European Medicines Agency (EMA) with a data cut-off date of March 11, 2023, was also included in this report.¹⁵

Participants eligible to be included in the ECHELON-1 trial were previously untreated adult patients with histologically confirmed advanced-stage classical HL, consisting of stage III and stage IV patients as determined by the Ann Arbor classification system. Patients with nodular lymphocyte-predominant HL and those with sensory or motor peripheral neuropathy were excluded. The median age of enrolled patients was 36 years (range = 18 to 83); most (66%) were younger than 45 years, and 14% were aged 60 years or older. Of the total number of patients enrolled, 58% were male and 84% were white. Notably, most patients had stage IV disease (64%), 2 or 3 (53%) had International Prognostic Factor Project (IPFP) risk factors, an Eastern Cooperative Oncology Group Performance Status (ECOG PS) of 0 (57%), extranodal involvement at diagnosis (62%), and B symptoms (59%) at baseline.

Efficacy Results

The efficacy end points that were noted to be important to patients and clinicians in stakeholder input are summarized in Table 2. Outcomes of OS, PFS (as determined by an investigator), percentage of patients alive without HL, HRQoL as measured by the European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire Core 30 (EORTC QLQ-C30), and EQ-5D-3L questionnaire were obtained from data with a cut-off date of June 1, 2021, while mPFS as determined by an independent review facility (IRF) was based on data with a cut-off date of April 20, 2017.

Overall Survival

As of the data cut-off date of June 1, 2021, the median follow-up was 73.3 months (95% confidence interval [CI], 72.61 to 74.05) in the BV + AVD group and 72.4 months (95% CI, 71.10 to 73.63) in the ABVD group. In the intention-to-treat (ITT) population, the hazard ratio (HR) for OS was 0.59 (95% CI, 0.396 to 0.879; P = 0.009), favouring BV + AVD treatment. The absolute difference in the number of OS events between the BV + AVD (6%) and ABVD (10%) arms was 4%. The median OS was not reached for patients with advanced-stage classical HL in either the BV + AVD group or the ABVD group. In a subgroup analyses by disease stage, the HR for OS was 0.863 (95% CI, 0.452 to 1.648; P = 0.654) for patients with stage III classical HL and 0.478 (95% CI, 0.286 to 0.799; P = 0.004) patients with stage IV classical HL.

As of the data cut-off date of March 11, 2023, the descriptive analysis for OS had a median follow-up of approximately 88 months for the ITT population: The median follow-up duration was 89.7 months (95% CI, 86.57 to 90.55) for the BV + AVD group and 86.3 months (95% CI, 84.53 to 89.33) in the ABVD group. This analysis included 111 OS events (deaths): 44 deaths (7%) were reported in the BV + AVD group and 67 deaths (10%) in the ABVD group. Median OS was not reached for either group. The HR for OS was 0.61 (95% CI, 0.414 to 0.892, descriptive P = 0.010). In the stage III subgroup, the median OS was not reached for either treatment arm, and the HR for OS was 1.004 (95% CI, 0.540 to1.866) for BV + AVD patients, compared with ABVD patients. In the stage IV subgroup, the median OS was not reached for either treatment arm, and the HR for OS was 0.48 (95% CI, 0.291 to 0.784; descriptive P = 0.003) for BV + AVD patients, compared with that for ABVD patients.

Alive Without HL

In the ITT population, the 3-year rates of being alive without HL were 96% (546 of 567) in the BV + AVD group and 93% (503 of 540) in the ABVD group. The 5-year rates of being alive without HL were about 94% (450 of 480) in the BV + AVD group and 92% (408 of 443) in the ABVD group. No subgroup analyses by disease stage were reported for this efficacy end point.

PFS According to Investigator

As of the data cut-off date of June 1, 2021, the median follow-up was 73.2 months (95% CI, 72.48 to 74.05) in the BV + AVD group and 71.6 months (95% CI, 70.37 to 72.87) in the ABVD group. In the ITT population, the HR of PFS according to investigator was 0.678 (95% CI, 0.532 to 0.863; P = 0.002), favouring treatment with BV + AVD. There was a 7% absolute difference in the number of PFS events between the BV + AVD group (17%) and the ABVD group (24%). The median PFS according to investigator was not reached for patients with advanced-stage classical HL for either the BV + AVD or ABVD group. In subgroup analyses by disease stage, the HRs for PFS according to investigator were 0.603 (95% CI, 0.391 to 0.930; P = 0.021) for patients with stage III classical HL and 0.715 (95% CI, 0.534 to 0.959; P = 0.024) patients with stage IV classical HL.

Modified PFS According to IRF

As of the data cut-off date of April 20, 2017, the median mPFS was not reached in either the BV + AVD group or the ABVD group. In the ITT population, the HR for mPFS according to IRF was 0.770 (95% CI, 0.603 to 0.982; P = 0.035). There was a 4% absolute difference in number of mPFS events between the BV + AVD arm and the ABVD arm, favouring BV + AVD (18% versus 22%). In subgroup analyses by disease stage, the HRs for mPFS according to IRF were 0.923 (95% CI, 0.600 to 1.420; P = 0.716) for patients with stage III classical HL and 0.712 (95% CI, 0.530 to 0.957; P = 0.024) for patients with stage IV classical HL.

Harms Results

Harms results for the safety population are summarized in Table 2. Deaths and secondary malignancies were from the data cut-off date of June 1, 2021, while the remaining data were from the data cut-off of April 20, 2017.

In the safety population, the proportions of patients experiencing treatment-emergent adverse events (TEAEs) up to 30 days after the last front-line dose were similar between patients treated with BV + AVD (99%) and those treated with ABVD (98%). Higher percentages of patients in the BV + AVD group experienced treatment-emergent serious adverse events (TESAEs) up to 30 days after the last front-line dose, compared to the percentages of patients in the ABVD group (43% versus 27%, respectively). Deaths were reported in 6% of the patients in the BV + AVD arm and 10% of the patients in the ABVD arm. Treatment discontinuation due to AEs occurred in 13% of the patients in the BV + AVD arm and 16% of those in the ABVD arm. In terms of notable harms, 67% of the patients in the BV + AVD group and 43% in the ABVD group experienced at least 1 peripheral neuropathy event. About 3% of the patients in the BV + ABVD group and 5% of the patients in the ABVD group developed secondary malignancies. The proportion of patients who experienced neutropenia as TEAEs of grade 3 or higher was higher in the BV + ABVD group than in the ABVD group (54% versus 39%, respectively). Similarly, the proportions of patients who experienced febrile neutropenia as TEAEs of grade 3 or higher were also higher in the BV + ABVD group than in the ABVD group (19% versus 8%). Fewer patients in the BV + AVD arm experienced AEs of pulmonary-related toxicity than in the ABVD arm (13% versus 25%). The most common AE of pulmonary-related toxicity for either group was dyspnea (12% versus 24%).

Table 2: Summary of Key Results From Pivotal Studies and RCT Evidence

ABVD = doxorubicin-bleomycin-vinblastine-dacarbazine; AVD = doxorubicin-vinblastine-dacarbazine; BV = brentuximab vedotin; CI = confidence interval; HL = Hodgkin lymphoma; IRF = intendent review facility; ITT = intention to treat; OS = overall survival; mPFS = modified progression-free survival; NE = not estimable; PFS = progression-free survival; RCT = randomized controlled trial.

Note: Multiplicities were only adjusted for OS and mPFS, and P values for other efficacy end points were provided for descriptive purposes only.

^aThe HR and 95% CI were based on a stratified Cox’s proportional hazard regression model with stratification factors region and number of IPFP risk factors at baseline with treatment as the explanatory variable in the model. A HR of less than 1 favours BV + AVD group.

^bThe median OS follow-up was calculated from the Kaplan-Meier method switching the OS event and censored status, i.e., OS event as censored and censored as OS event.

^cThe median PFS follow-up was calculated from the Kaplan-Meier method switching the PFS according to investigator event and censored status, i.e., PFS according to investigator event as censored and censored as PFS according to investigator event.

^dThe median mPFS follow-up was calculated from the Kaplan-Meier method switching the mPFS event and censored status, i.e., mPFS event as censored and censored as mPFS event.

^ePreferred terms of neutropenia and decreased neutrophil count are counted as neutropenia.

^fPulmonary-related toxicity include preferred terms of “dyspnea” and “hypoxia,” and all preferred terms in an interstitial lung disease standardized Medical Dictionary for Regulatory Activities query, and preferred terms with the high-level term “respiratory and pulmonary function diagnostic procedures.”

Sources: ECHELON-1 original Clinical Study Report (data cut-off: April 20, 2017),¹³ ECHELON-1 Clinical Study Report Addendum 1 (data cut-off: June 01, 2021),¹⁴ Drug Reimbursement Review sponsor submission,^16, and the sponsor’s Summary of Clinical Evidence.¹

Critical Appraisal

Generally, no serious concerns were identified in the conduct of the ECHELON-1 trial. However, the validity of the primary outcome (mPFS) is a key consideration in evaluating the evidence for BV + AVD. The mPFS was adopted in the ECHELON-1 trial to capture all events that reflect a failure of front-line chemotherapy by counting a response that was less than complete at the end of the front-line therapy as an event. The ECHELON-1 trial defined a response of less than complete as “receipt of anticancer therapy or radiotherapy for HL after completion of front-line therapy for patients who were confirmed non-complete responders.” However, the clinical experts consulted by the review team noted that this definition is not consistent with practice in defining disease progression or first-line treatment failure in advanced HL, and receipt of radiotherapy does not necessarily indicate disease progression in clinical practice. Despite the end-of-treatment PET scans conducted by the IRF, there is a concern that the results for mPFS could be biased given that the administration of new anticancer therapy was at the discretion of the treating physician. The clinical experts consulted by the review team noted that OS and PFS are more clinically relevant to assessing patient benefits from treatment, and no evidence was included in the submission to the review team that empirically validated mPFS as an outcome measure or established a correlation with OS. High percentages in loss to follow-up and withdrawal by patients were noted in both OS and PFS analyses. Although the percentages of loss to follow-up and withdrawal by patients were balanced between treatment arms, reasons for loss to follow-up and withdrawal could be differential between groups, which could lead to biased estimates of treatment effects. Moreover, no sensitivity analyses assessing the potential impact of the loss to follow-up and withdrawal on OS and PFS results were available. Subgroup analyses by HL stage signal that there might be a difference in treatment effects between patients with stage III and those with stage IV classical HL for mPFS and OS. However, the review team’s ability to make a definitive conclusion as to whether the difference between the disease stage subgroups is true was limited by several concerns, such as the possibility that the balance of known and unknown factors between treatment groups achieved by randomization was not preserved in stage III or stage IV subgroups. In addition, the trial was not specifically designed to test statistical inferences between BV + AVD and ABVD in stage III and stage IV subgroups.

All participants in the ECHELON-1 trial were required to be aged 18 years or older and diagnosed with classical HL. The ECHELON-1 trial therefore did not reflect results for pediatric patients or patients with nodular lymphocyte-predominant HL. The eligibility criteria of the ECHELON-1 trial in general were aligned with selection criteria in the Canadian settings when identifying suitable candidates for BV + AVD, according to the clinical experts consulted by the review team. However, the clinical experts noted that, in clinical practice, a small percentage of patients who were excluded from the ECHELON-1 trial, such as patients with HIV, might be eligible to receive BV + AVD, if the disease is well managed, and to patients with a borderline left ventricle ejection fraction after consultation with a cardiologist. The clinical experts also noted that patients with ECOG PS scores higher than 2 could be considered for treatment with BV + AVD on a case-by-case basis. The clinical experts noted that the doses of BV + AVD and ABVD used in the ECHELON-1 trial generally reflected the standard dose schedules used for adults in Canada. The clinical experts also confirmed that the direct comparator, ABVD up to 6 cycles (not adapted based on PET response), is a relevant therapy used in current standard of care, although, because it is not the only standard-of-care front-line therapy used in Canada, other relevant comparators are available. In addition, the clinical experts noted that the percentages of patients who received a transplant as subsequent treatment were lower in either group than they would have expected to see in clinical practice. According to the clinical experts consulted by the review team, the characteristics of the study population were generally reflective of patients who would be eligible for BV + AVD in Canadian practice. However, the experts noted that the percentage of patients with stage IV HL in the trial population and the percentage of participants who are white were higher than what would be seen in clinical practice.

Long-Term Extension Studies

No long-term extension studies were submitted by the sponsor.

Indirect Comparisons

No indirect treatment comparisons (ITCs) were submitted by the sponsor. The sponsor provided a feasibility assessment that determined it would be infeasible to conduct ITCs of BV + AVD versus other front-line therapies examined in clinical studies for advanced HL.

Studies Addressing Gaps in the Pivotal and RCT Evidence

To address gaps in the pivotal randomized controlled trial (RCT) evidence related to the use of BV in pediatric patients with advanced HL, the review team reviewed evidence from an additional phase III RCT.

Description of Studies

The AHOD1331 trial (N = 587),¹¹ published in The New England Journal of Medicine, is a phase III, multicentre, open-label, randomized active-controlled trial comparing BV + AVEPC with doxorubicin-bleomycin-vincristine-etoposide-prednisone- cyclophosphamide (ABVE-PC) in previously untreated patients aged 2 to 21 years with high-risk classical HL, defined as Ann Arbor stage IIB with bulk tumour, stage IIIB, stage IVA and stage IVB patients. The primary objective of the AHOD1331 trial was to determine the efficacy of BV + AVEPC relative to ABVE-PC as measured by EFS. Harms and OS were also reported. The AHOD1331 trial is ongoing. The final analysis of EFS was based on the database lock date of December 31, 2021.

In the AHOD1331 trial, the median age of participants was 15.6 years (range = 3.4 to 21.99); most (84.7%, 497 of 587) were aged between 12 and 22 years. Of the 587 patients enrolled, 47% (276) were female and 57.6% (338) were non-Hispanic white. The proportions of patients by disease stage were 20.6% (121) for stage IIB with bulk tumour, 19.3% (113) for stage IIIB, 28.4% (167) for stage IVA, and 31.7% (186) for stage IVB. Most of the patients had nodular-sclerosis classical HL (76.5%, 449 of 587).

Efficacy Results

The median follow-up time was 42.1 months (range = 0.1 to 80.9). In terms of the 3-year OS in the ITT population, the proportions of patients who were censored were 99.3% (95% CI, 97.3 to 99.8) in the BV + AVEPC group and 98.5% (95% CI, 96.0 to 99.4) in the ABVE-PC group. The HR for 3-year OS was not provided.

In terms of the 3-year EFS in the ITT population, the proportions of patients who were censored were 92.1% (95% CI, 88.4 to 94.7) in the BV + AVEPC group and 82.5% (95% CI, 77.4 to 86.5) in the ABVE-PC group. The HR for 3-year EFS was 0.41 (95% CI, 0.25 to 0.67; P < 0.001), favouring the BV + AVEPC arm.

Harms Results

The incidence of any AEs of grade 3 or higher was 73.5% in patients treated with BV plus AVEPC and 68.2% in patients treated with ABVE-PC. Peripheral neuropathy of grade 3 or higher occurred in 6.7% of the patients in the BV plus AVEPC arm and 5.5% % of the patients in the ABVE-PC arm. Febrile neutropenia occurred in 30.9% and 32.5% of patients in the BV plus AVEPC and ABVE-PC arms, respectively. None of the patients in the BV plus AVEPC group experienced pneumonitis compared to 1 patient in the ABVE-PC group.

Critical Appraisal

Although details about the randomization process and allocation concealment were not reported in the research protocol or the main article, the risk of bias in the AHOD1331 trial is anticipated to be low given that baseline characteristics between the treatment arms were generally similar for clinically important factors. The AHOD1331 trial was open-label but had blinded outcome assessors, the definition of EFS was aligned with accepted definitions from regulators, and treatment response was assessed via centralized review, helping reduce the risk of detection bias related to the open-label design. Although patients were aware of the treatment allocation, which may result in performance bias, this risk is considered low as the 3-year PFS in the ABVE-PC group (82.5%) and the types of AEs were generally in line with what the consulted clinical expert who specializes in pediatrics expected. Those patients who remained PET2-positive received response-adapted involved-site radiation therapy guided by blinded central assessment of PET scans. This could bias the EFS results if the radiation therapy could improve response, reduce the likelihood of relapse, and/or increase the risk of secondary malignancy. However, the risk of this potential bias was mitigated by the requirement that radiation therapy could not be administered until directed by the blinded assessment. Also, the percentages of patients who received involved-site radiation therapy were similar between the BV + AVEPC and ABVE-PC groups (53.4% versus 56.8%, respectively). Concomitant anticancer medications were not allowed. Antibiotics and supportive medications (e.g., antiemetics) were permitted as needed. Patients also received granulocyte colony-stimulating factor (G-CSF) support. None of the permitted medications would likely influence the results for either treatment group. However, after a progression event, the treating physician could treat the patient at their discretion, which may affect the longer-term OS results.

The AHOD1331 trial was appraised in this section to address an important gap with respect to the unmet needs of using BV + chemotherapy in pediatric patients with classical HL. However, several notable issues need to be considered when generalizing results from the AHOD1331 trial. First, although the chemotherapy backbone used in the AHOD1331 trial (AVEPC) is a preferred backbone for pediatric patients, according to POGO and the clinical experts consulted by the review team, it is different from the backbone used in adults (ABVD). Regarding the regulatory status of the pediatric regimen, the review team confirmed that BV is not approved for use in combination with the pediatric regimen, and the sponsor confirmed it is not planning to request Health Canada approval for BV + AVEPC. Second, the clinical experts consulted by the review team noted the definition of high-risk or advanced-stage HL in pediatric patients varies. While the AHOD1331 trial adopted the definition of advanced-stage HL in pediatric patients as stage II with bulk tumour, stage IIIB, stage IVA, and stage IVB, some medical centres may define any stage III or IV as advanced-stage disease in pediatric patients. Finally, the AHOD1331 trial involved both nonadults and young adults (up to 22 years old), while the pivotal ECHELON-1 trial enrolled patients aged 18 years and older. This created an overlap in patient age between the pivotal ECHELON-1 trial and the AHOD1331 trial. The clinical experts consulted by the review team noted that the chemotherapy backbone used in the AHOD1331 trial (AVEPC) would not typically be used in patients aged 18 years or older in Canada, and the chemotherapy backbone ABVD investigated in the pivotal ECHELON-1 trial may be used in adolescents with HL aged close to 18 years.

Conclusions

Overall, evidence from the phase III, open-label, randomized ECHELON-1 trial suggests that BV + AVD is an effective front-line treatment for previously untreated adult patients diagnosed with advanced-stage classical HL. The clinically relevant efficacy end points examined in the report (OS, PFS, and mPFS) were consistently in favour of BV + AVD compared to ABVD in the ITT population. However, the clinical significance of the magnitude of the treatment differences is uncertain, and concerns remain about the validity of the primary outcome (mPFS) as well as the high percentages in loss to follow-up and withdrawal by patients in both OS and PFS analyses. In addition, although subgroup analyses of the OS and mPFS results signalled that BV + AVD may be more effective in patients with stage IV classical HL than those with stage III HL, conclusions regarding subgroup differences between stage III versus stage IV patients are uncertain because the study was not designed to detect differences between these subgroups and patients were not stratified according to disease stage at randomization. The safety profile of BV + AVD is consistent with the known AEs for the individual components of the regimen, but more patients treated with BV + AVD experienced serious adverse events (SAEs) compared with those in the ABVD group. An evidence gap remains with respect to the clinical efficacy and safety of BV + AVD in the pediatric population as all participants in the pivotal ECHELON-1 trial were required to be aged 18 years or older. To address unmet needs in the pediatric patient population, the AHOD1331 trial was examined. Efficacy results from the AHOD1331 trial indicated that BV + AVEPC provides a clinically meaningful benefit in EFS compared to ABVE-PC in patients aged 2 to 21 years with high-risk classical HL defined as Ann Arbor stage IIB with bulk tumour, stage IIIB, stage IVA, and stage IVB HL.

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 BV (50 mg per vial, lyophilized powder for reconstitution, IV infusion) for the treatment of previously untreated patients with advanced-stage HL, in combination with AVD.

This is a submission from a tumour group. However, the review is based on a previous sponsor-initiated submission (Project Number PC0311 to 000) that was withdrawn before CADTH adopted the Grading of Recommendations Assessment, Development and Evaluation framework for each reimbursement review. As such, the framework was not used for the current review.

Disease Background

Contents within this section were informed by materials submitted by the sponsor and clinical expert input. The following summary was validated by the review team.

Hodgkin lymphoma is a B-cell lymphoid malignancy that originates in the lymphocytes.³ Classical HL accounts for 95% of all HL cases and is characterized by the presence of Hodgkin Reed-Sternberg cells, which express the surface antigen CD30.^17,18 While the median age of diagnosis for HL is 39 years, the incidence of HL is age-related and bimodal, predominantly affecting people in their 20s, early 30s, and those older than 55 years.^19,20 A common early sign of HL is painless enlargement of 1 or more lymph nodes.⁹ Other signs and symptoms of HL include fatigue, shortness of breath, itchiness on the trunk of the body, unusual back or abdominal pain, and abdominal swelling.²¹ A diagnosis of HL is done by a lymph node biopsy.^8,9,22

In 2022, an estimated 1,050 new cases of HL occurred in Canada overall, for an incidence rate of 2.6 cases per 100,000.⁴ Based on the 2018 Canadian Cancer Statistics, approximately 23.3% of Canadian patients presenting with HL have stage III disease and 22.7% have stage IV disease.⁵ Childhood HL represents 6% of all cancers and has an incidence rate of 12 cases per million (1.2 cases per 100,000) per year among children aged 0 to 14 years.⁶ In 2019, a total of 25 children in Canada in this age group were diagnosed with HL.⁷

The clinical experts consulted by the review team noted that advanced-stage HL in adult patients refers to Ann Arbor stage III and IV HL (Appendix 1). The clinical experts noted that pediatric patients with HL are usually classified into low-, intermediate-, and high-risk groups in clinical practice, and the high-risk group is generally considered equivalent to the advanced-stage classical HL group in adults. The clinical experts reported that pediatric patients with high-risk or advanced-stage HL are generally defined as stage II with bulk tumour, stage III with B symptoms (stage IIIB), and stage IV with or without B symptoms (stage IVA and stage IVB). However, the clinical experts consulted by the review team also indicated that the definition of advanced-stage HL in pediatric patients is evolving and may vary by centre, with some centres defining any stage III or IV as advanced-stage disease in pediatric patients.

HL is considered a curable disease.²³ In 2021, the age-standardized mortality rate was 0.2 per 100,000 with 5- and 10-year net survival rates of 85% and 81%, respectfully.²⁴

Standards of Therapy

Contents within this section were informed by materials submitted by the sponsor and clinical expert input. The following summary was validated by the review team.

The goal of therapy in patients with advanced-stage (stage III and IV) HL is curative.^8,9 The clinical experts consulted by the review team noted that the treatments for patients in the advanced-stage HL group (stage III and IV) are similar. The clinical experts also emphasized that treatments may differ between classical HL and nodular lymphocyte-predominant HL, the latter of which does not have classical Reed-Sternberg cells and may behave more like indolent non-HL.

The clinical experts consulted by the review team noted that current treatment regimens for adult patients in Canada with advanced-stage HL mainly include: an approach based on ABVD for 6 cycles with PET2-directed de-escalation for PET2-negative patients to AVD and, in some jurisdictions, PET2-directed escalation to escalated BEACOPP in PET2-positive patients; an approach based on upfront escalated BEACOPP with PET2-driven treatment adaptation; and BV + AVD for patients with stage IV HL. For pediatric patients, the clinical experts consulted by the review team noted that most clinical centres in Canada have adopted ABVE-PC for 5 cycles with radiation therapy determined by PET2 response, while relatively few centres use vincristine-etoposide-prednisone-doxorubicin–cyclophosphamide-vincristine-prednisone-dacarbazine.

The clinical experts consulted by the review team noted that current drugs for HL that are essentially cytotoxic and/or DNA-damaging do not target the underlying pathogenetic mechanisms, except for programmed cell-death protein 1 inhibitors, which target the underlying immune escape essential for classical HL proliferation. The clinical experts noted that no drugs are currently accessed through special programs to treat advanced-stage HL in the front-line setting, and radiotherapy is usually reserved for earlier-stage (I and II) disease or may be given at the end of therapy for patients with advanced-stage HL.

Drug Under Review

Brentuximab vedotin is an antibody-drug conjugate directed against CD30.² Currently, BV has an indication approved by Health Canada² for the treatment of previously untreated patients with stage IV HL, in combination with AVD; BV was reviewed and recommended for reimbursement for that indication in 2020.²⁵

The recommended dose of BV for previously untreated advanced-stage HL is 1.2 mg/kg up to a maximum of 120 mg in combination with AVD administered every 2 weeks for a maximum of 12 doses or until disease progression or unacceptable toxicity occurs. BV is administered as an IV infusion over 30 minutes.

Key characteristics of BV are summarized in Table 3, along with other treatments available for advanced-stage HL.

Table 3: Key Characteristics of BV and Bleomycin

ADC = antibody-drug conjugate; AVD = doxorubicin-vinblastine-dacarbazine; BV = brentuximab vedotin; MMAE = monomethyl auristatin E; NA = not applicable; PML = progressive multifocal leukoencephalopathy.

^aHealth Canada–approved indication.

Sources: Sponsor’s Clinical Summary Evidence,¹ BV product monograph,^2, and bleomycin for injection product monograph.²⁶

Stakeholder Perspectives

Patient-Group Input

This section was prepared by the review team based on the input provided by patient groups. The full original patient input(s) received by the review team are included in the stakeholder section at the end of this report.

Patient Input

One patient group, Lymphoma Canada, provided input for BV for the treatment of previously untreated patients with advanced-stage HL, in combination with AVD. Patient input was gathered from an online anonymous patient survey from March 14 to May 2, 2023. A total of 26 responses were gathered, and 3 of these patients reported receiving BV + AVD treatment.

Patients were asked questions regarding the physical and psychosocial symptoms experience at the time of diagnosis, current quality of life, and how these symptoms affected their daily activities. At the time of their lymphoma diagnosis, most of the patients reported fatigue (79%) as the symptom with the greatest impact (5 out of 5), followed by enlarged lymph nodes (58%), shortness of breath (63%), and weight loss (47%). In addition, 74% of patients reported experiencing anxiety or worry, 68% stressing about their diagnosis, 63% difficulty in sleeping, and 58% fearing progression of their lymphoma. Regarding physical symptoms that currently affect their quality of life, out of 7 responses, fatigue (29%) and headaches (14%) were identified as the most significant factors imposing a negative impact on quality of life. The most consequential psychosocial factors affected patients’ current quality of life were the stress of having cancer (71%), fear of progression (71%), anxiety or worry (71%), difficulty sleeping (43%), problems concentrating (43%), and inability to attend work or school (43%).

Among the surveyed patients, 3 reported receiving ABVD in the front-line setting, 2 were treated with other forms of chemotherapy, and 1 was treated with cyclophosphamide-doxorubicin-prednisone-rituximab-vincristine. While evaluating the importance of outcomes of new treatments, patients from the survey emphasized the need for a novel lymphoma therapy to control disease symptoms, lengthen disease remission, extend life spans, and improve quality of life.

While describing the experience with the treatment under review, 1 of the 3 patients who had received BV + AVD indicated they had been in remission for between 6 months and a year, another had been in remission for longer than a year, and the third patient was in post-treatment (and unsure about their remission status). Side effects from the BV + AVD treatment reported by patients were fatigue (n = 3), neutropenia (n = 2), constipation (n = 2), joint or muscle pain (n = 2), low platelet count (n = 1), low blood pressure (n = 1), and decreased appetite (n = 1). Two patients reported experiencing financial setbacks — 1 due to absence from work and 1 due to cost of other medications. One of these patients mentioned having a poor experience with BV, and the other 2 rated their experience as very good.

Clinician Input

Input From Clinical Experts Consulted by the Review Ream

All our review teams include at least 1 clinical specialist with expertise in the diagnosis and management of the condition for which the drug is indicated. Clinical experts are a critical part of the review team and are involved in all phases of the review process (e.g., providing guidance on the development of the review protocol, assisting in the critical appraisal of clinical evidence, interpreting the clinical relevance of the results, and providing guidance on the potential place in therapy). The following input was provided by 3 clinical specialists with expertise in the diagnosis and management of previously untreated patients with advanced-stage HL.

Unmet Needs

The clinical experts consulted by the review team noted that advanced-stage HL in adult patients refers to Ann Arbor stage III and IV HL and agreed that the treatment goals as well as unmet needs are the same between adult patients with stage III HL and those with stage IV HL. The clinical experts noted that, in clinical practice, pediatric patients with HL are usually classified into low-, intermediate-, and high-risk groups, and the high-risk group is generally considered equivalent to advanced-stage classical HL in adults. The clinical experts reported that pediatric patients with high-risk or advanced-stage HL are generally defined as having stage II with bulk tumour, stage III with B symptoms (stage IIIB), and stage IV with or without B symptoms (stage IVA and stage IVB), and these patients are treated the same in clinical practice — as advanced-stage HL. However, the clinical experts consulted by the review team indicated that the definition of advanced-stage HL in pediatric patients is evolving and may vary by centre, with some centres defining any stage III or IV as advanced-stage disease in pediatric patients.

The clinical experts noted that cure of the disease is the main goal of treatment. Moreover, the cure should be achieved with an acceptable level of short- and long-term toxicity. In younger patients with advanced-stage HL, the goal of therapy is also to restore normal life expectancy and productivity while minimizing toxicity by increasing the cure rate, lowering the risk of relapse, and reducing the need for more cytotoxic and/or DNA-damaging therapies.

According to the clinical experts, an improved cure is an important unmet need for patients with advanced-stage HL to reduce treatment failure, prevent disease progression or relapse, and avoid late side effects (e.g., secondary malignancies, cardiac and pulmonary late effects) and further therapies that are toxic (e.g., ASCT), particularly for younger patients diagnosed with advanced-stage HL and older patients who have poor tolerance to treatment. In addition, 1 clinical expert noted that a combination therapy without bleomycin (i.e., part of the standard AVD) is needed for older patients with advanced-stage HL due the potentially fatal lung toxicity of bleomycin.

Place in Therapy

The clinical experts noted that BV + AVD is considered a front-line therapy and should not be used for patients who are intolerant to other therapies. The purpose of using BV + AVD is for curing the disease, not for symptomatic control of disease. The clinical experts noted that, at the time of the review, BV + AVD had been approved only for the treatment of previously untreated patients with stage IV HL, and the use of BV + AVD in patients with stage III classical HL would cause a shift in the current treatment paradigm for those patients. The clinical experts noted that BV + AVD would not be added to other treatments, except that radiotherapy may be given at the end of therapy. The clinical experts indicated that, in pediatric patients, BV would be used in combination with a different chemotherapy backbone that has been investigated in a phase III randomized controlled trial¹¹ (i.e., BV + AVEPC) instead of AVD, as trials of BV + AVD in pediatric patients had not been completed at the time of this review.¹²

Patient Population

The clinical experts noted that any previously untreated adult patients with stage III or IV classical HL who meet the eligibility criteria of the ECHELON-1 trial are best suited for the use of BV + AVD. The clinical experts indicated that patients with nodular lymphocyte-predominant HL are not suitable for BV + AVD. The clinical experts emphasized that BV + AVD should be used with caution in older patients (i.e., those aged ≥ 60 years) due to potential toxicity. The clinical experts also noted that factors such as older age, stage IV, International Prognostic Score (IPS), and ECOG PS are accepted risk factors and may be associated with adverse outcomes. The clinical experts indicated that liver function tests, ECOG PS, and blood counts are important when considering giving BV + AVD. The clinical experts also noted that any pediatric patient with advanced-stage classical HL could also be eligible for BV in combination with chemotherapy, and eligibility for therapy should be determined by the treating physician.

Assessing the Response Treatment

The clinical experts noted that PET scans, typically PET2 and a PET scan after all 6 cycles (end of treatment), is used for response assessment. The desired outcome is a complete metabolic response on the end-of-treatment PET scan. In follow-ups of patients who are found to have achieved a CR, visits take place every 3 to 4 months for 2 years, then typically every 6 months for 3 more years, and there is no serial imaging to monitor disease. In the absence of a CR, patients with a partial response may undergo radiation therapy, and patients with refractory disease would undergo further investigations (e.g., biopsy) and treatment with a second-line regimen.

Whether patients who receive BV + AVD and remain PET-positive on PET2 should be escalated to a different treatment remained an unresolved question for 1 clinical expert, while a second expert suggested continuing BV + AVD. The clinical experts pointed out that, although PET2 positivity is important for most regimens, its role in the BV + AVD regimen is not yet clearly established.

The clinical expert specializing in pediatrics noted that outcomes used in clinical practice to assess treatment response are generally aligned with outcomes typically used in adult therapeutics. All 3 clinical experts noted that OS is the most clinically meaningful outcome to assess the efficacy of BV in combination with chemotherapy in patients with advanced-stage HL, and PFS is an important outcome. In pediatric patients, EFS is also important. However, the importance of mPFS remains controversial. The clinical experts noted that end points such as duration of response and duration of CR are not typically used in a disease such as HL.

Discontinuing Treatment

The clinical experts agreed that, overall, discontinuation of BV + AVD is uncommon because toxicity or refractory HL is not common. The clinical experts identified several situations in which BV + AVD can be discontinued: if treatment is complete; there is clear progression of disease (although a Deauville 4 or 5 PET2 without new disease would not warrant discontinuing therapy); or there is an AE such as an allergic reaction to BV, severe neuropathy, pneumonitis, elevated liver enzymes, or neutropenia. For elevated liver enzymes and neutropenia, dose delays or dose adjustments are required before discontinuation.

Prescribing Considerations

The clinical experts noted that diagnosis of the disease in adult patients must be made by an experienced pathologist. Selection of patients for BV + AVD should be made by a hemato-oncologist experienced in treating HL. However, the treatment can be delivered in specialty clinics of nonacademic centres.

The clinical expert specializing in pediatrics noted that all pediatric oncology patients are cared for by pediatric oncology teams at tertiary care centres. Some aspects of care may be provided at satellite centres after diagnostic and management decisions are made.

Clinician Group Input

This section was prepared by the review team based on the input provided by clinician groups. The full original clinician group inputs received by the review team are included in the stakeholder section at the end of this report.

Clinician group input on the review of BV was received from the OH-CCO Hematology Cancer Drug Advisory Committee and POGO. Six clinicians provided input on behalf of OH-CCO. POGO is a collaboration of Ontario’s 5 specialized childhood cancer centres. The input collected from POGO was prepared in a consultative manner, with 1 clinician discussing the indication with other members of the submission panel and seeking input from POGO’s Therapeutic and Technology Advisory Committee, and the final submission was based on the results of this process.

The OH-CCO input highlighted the need to improve outcomes with first-line therapy to avoid the need for second-line therapy. The group noted that patients with stage III and IV disease would be best suited for current treatment. The group indicated that typical lymphoma response measures, including PET scans, are used in clinical practice to assess patients’ response to treatment. They indicated they would discontinue treatment with BV + AVD in cases of significant toxicities or disease progression.

The POGO input noted that, while a variety of chemotherapy and radiation approaches are available as options for the standard of care, these vary by region and between pediatric- and adult-focused practitioners in Canada. POGO reported that, historically, the ABVD chemotherapy backbone used with BV in adult patients has not been used by pediatric oncologists to treat pediatric patients due to concerns about higher anthracycline (doxorubicin) and bleomycin exposure, and known dose-dependent cardiac and pulmonary toxicities. POGO noted that BV has been studied and used in combination with another chemotherapy regimen (AVEPC) in patients aged 2 to 21 years with previously untreated high-risk HL.¹¹ POGO indicated that this alternative chemotherapy backbone is more commonly used in the pediatric setting, and BV + AVEPC has become standard care for high-risk pediatric patients in Ontario. Regarding treatment goals for HL in pediatric patients with, POGO emphasized the need to avoid disease recurrence to minimize potential late effects from subsequent chemotherapies and ASCT received at relapse, and the associated impact on HRQoL. While describing the outcomes used to determine whether a pediatric patient is responding to treatment for HL, the POGO group emphasized the importance of OS and EFS, considering the higher chance of experiencing late effects of therapy after treatment among the younger patient population. Like OH-COO input, POGO suggested treatment be discontinued at disease progression.

Drug Program Input

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

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

ABVD = doxorubicin-bleomycin-vinblastine-dacarbazine; ABVE-PC = adriamycin-bleomycin-vincristine-etoposide-prednisone-cyclophosphamide; ADV = doxorubicin-vinblastine-dacarbazine; AVPEC = doxorubicin-vincristine-etoposide-prednisone-cyclophosphamide; BEACOPP = bleomycin-etoposide-doxorubicin-cyclophosphamide, vincristine, procarbazine, and prednisone; BV = brentuximab vedotin; CNS = central nervous system; CPG = clinical practice guidelines; CVPP = cyclophosphamide-vinblastine-procarbazine-prednisone; ECOG PS = Eastern Cooperative Oncology Group Performance Status; G-CSF = granulocyte colony-stimulating factor; HL = Hodgkin lymphoma; OPEA-COPDAC = vincristine-etoposide-prednisone-doxorubicin–cyclophosphamide-vincristine-prednisone-dacarbazine; OS = overall survival; PAG = Provincial Advisory Group; pERC = pan-Canadian Oncology Review Expert Review Committee; PML = progressive multifocal leukoencephalopathy.

Clinical Evidence

The objective of the Clinical Review Report is to review and critically appraise the clinical evidence submitted by the sponsor on the beneficial and harmful effects of BV (50 mg per vial, lyophilized powder for reconstitution, IV infusion), in combination with AVD, in the treatment of previously untreated patients with advanced-stage HL. The focus will be placed on comparing BV + AVD to relevant comparators and identifying gaps in the current evidence.

A summary of the clinical evidence included by the sponsor in the review of BV + AVD is presented in 2 sections, and a critical appraisal of the evidence is included after this section. The first section, the systematic review, includes pivotal studies and RCTs selected according to the sponsor’s systematic review protocol. No long-term extension studies or indirect evidence was submitted by the sponsor. The next section includes an additional study that was considered to address important gaps in the pivotal RCT evidence.

Included Studies

Clinical evidence from 1 pivotal phase III, multicentre, open-label RCT (ECHELON-1) and 1 additional phase III, multicentre, open-label, RCT addressing gaps in the pivotal RCT evidence (AHOD1331) is included in the review and appraised in this document:

Pivotal Study and RCT Evidence

Contents within this section were informed by materials submitted by the sponsor. The following summary was validated by the review team.

Description of Studies

One study (ECHELON-1),^13,14 which was conducted by the sponsor, met the inclusion criteria for the sponsor-submitted systematic review. Characteristics of the included studies are summarized in Table 5.

The ECHELON-1 trial is a phase III, open-label, randomized, active-controlled, superiority trial comparing BV + AVD with ABVD in previously untreated adult patients with advanced-stage classical HL, consisting of stage III and stage IV patients as determined by the Ann Arbor classification system.^9,27,28 This study is being conducted in 1,334 participants from 21 countries, and 60 patients in Canada have been enrolled. Participants were randomized 1:1 and stratified by geographical region (i.e., Americas versus Asia versus Europe) and number of IPFP risk factors (i.e., 0 to1 versus 2 to 3 versus 4 to 7), with 664 patients randomized to the BV + AVD arm and 670 patients to the ABVD arm. The primary objective of the ECHELON-1 trial was to determine the efficacy of BV + AVD relative to ABVD as measured by mPFS. The key secondary objective was to compare OS between BV + AVD and ABVD.

The ECHELON-1 trial is ongoing. Data gathered at the cut-off dates of April 20, 2017, and June 1, 2021, were assessed for this review. New data from a descriptive analysis of OS conducted in response to a request for supplementary information from the EMA with a data cut-off date of March 11, 2023 was also included in this report.¹⁵

Table 5: Details of Pivotal Studies and RCT Evidence Identified by the Sponsor

ABVD = doxorubicin-bleomycin-vinblastine-dacarbazine; ADV = doxorubicin-vinblastine-dacarbazine; ALT = alanine transaminase; AST = aspartate transaminase; BV = brentuximab vedotin; ECOG PS = Eastern Cooperative Oncology Group Performance Status; EMA = European Medicines Agency; EORTC QLQ-C30 = European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire Core 30; FACT/GOG-Ntx Abbreviated = abbreviated Functional Assessment of Cancer Therapy/Gynecologic Oncology Group – Neurotoxicity; FACIT-Dyspnea 10 = Functional Assessment of Chronic Illness Therapy – Dyspnea 10-item short-form questionnaire; HBV = hepatitis B virus; HCV = hepatitis C virus; HL = Hodgkin lymphoma; HRQoL = health-related quality of life; NYHA = New York Heart Association; SAE = serious adverse event; TEAE = treatment-emergent adverse event; ULN = upper limit of normal.

^aWHO classification includes nodular-sclerosis, mixed-cellularity, lymphocyte-rich, lymphocyte-depleted, or classical HL not otherwise specified.

^bSpecified clinical laboratory values include: absolute neutrophil count of 1,500/μL or higher and platelet count of 75,000/μL or greater unless known HL marrow involvement, total bilirubin less than 1.5 times the ULN unless due to Gilbert syndrome, ALT or AST less than 3 times the ULN range (AST and ALT may be elevated up to 5 times the ULN if reasonably ascribed to presence of HL in liver), serum creatinine less than 2.0 mg/dL and/or creatinine clearance or calculated creatinine clearance 40 mL/minute or greater, hemoglobin 8 g/dL or greater.

^cNonmelanoma skin cancer or carcinoma in situ of any type are not excluded if they had been completely resected.

Sources: ECHELON-1 original Clinical Study Report (data cut-off: April 20, 2017)¹³ and Addendum 1 (data cut-off: June 01, 2021)¹⁴ and the sponsor’s Summary of Clinical Evidence.¹

Populations

Inclusion and Exclusion Criteria

The ECHELON-1 trial included previously untreated adult patients with histologically confirmed stage III or stage IV classical HL and with an ECOG PS of 2 or lower (Table 5). Patients with nodular lymphocyte-predominant HL and those with sensory or motor peripheral neuropathy were excluded.

Interventions

In the ECHELON-1 trial, patients in both groups received combination therapy (i.e., intervention group: BV + AVD; control group: ABVD), with each drug administered as an IV infusion on days 1 and 15 of a 28-day cycle. Treatment was continued up to a maximum of 6 cycles, which were completed before the final mPFS analysis (data cut-off: April 20, 2017). No new study treatments were administered during the post-treatment follow-up period leading up to Addendum 1 (data cut-off date: June 1, 2021).

The BV + AVD regimen consisted of the following drugs administered in sequential order: doxorubicin 25 mg/m² IV infusion, vinblastine 6 mg/m² IV infusion, and dacarbazine 375 mg/m² IV infusion. Then, approximately 1 hour after the end of the dacarbazine infusion, BV (i.e., 1.2 mg/kg) was administered via IV infusion over approximately 30 minutes. The dose of BV was calculated based on actual weight, except for patients weighing more than 100 kg. In such cases, the dose was calculated based on a weight of 100 kg. During the trial, 9% of patients (n = 58) randomized to BV + AVD received the 100 mg capped dose of BV. Dose adjustments were performed in patients who experienced a change in weight of 10% or more from baseline.

The ABVD regimen consisted of the following agents administered in sequential order: doxorubicin 25 mg/m² IV infusion, bleomycin 10 units/m² IV infusion, vinblastine 6 mg/m² IV infusion, and dacarbazine 375 mg/m² IV infusion.

Dose modification was permitted, based on treatment-associated toxicity. Administration of growth factors (i.e., G-CSF or granulocyte-macrophage colony-stimulating factor) were permitted to treat neutropenia, as were platelets and/or red blood cell–supportive growth factors or transfusions when needed. Due to the higher incidence of neutropenia in the BV + AVD group, the independent data and safety monitoring committee of the ECHELON-1 trial recommended primary prophylaxis with G-CSF for patients newly randomized to BV + AVD treatment. This change was made after 75% of enrolment was complete; primary prophylaxis was defined as use of G-CSF by day 5 of the treatment cycle.

Prophylactic antiemetics and antidiarrheals were not included in the study protocol but were permitted at the physician’s discretion. Corticosteroids were permitted as part of a chemotherapy premedication regimen or for treatment of HL according to an institution’s standards. If a PET scan performed on day 25 resulted in a Deauville score of 5, physicians had the option of switching the patient to an alternative treatment regimen. Any switch in therapy made before completion of front-line treatment was not considered an mPFS event. At the end of front-line treatment, patients in partial remission with persistent PET-positive disease were permitted to receive radiation therapy at the discretion of the investigator. However, such patients were counted has having an mPFS event only if they were deemed to have a non-CR (i.e., a Deauville score ≥ 3) confirmed by the IRF.

Outcomes

A list of end points assessed in this clinical review report is provided in Table 6 and further summarized in the following section. Summarized end points are based on those included in the sponsor’s Summary of Clinical Evidence as well as any identified as important to this review according to stakeholders (the clinical experts, clinician groups, or patient groups). In addition to the end points listed in the table, the review team noted that post-progression survival (PPS), which was defined as time from progression to death and calculated for patients who experienced disease progression by subtracting PFS from OS obtained from the ECHELON-1 trial, was submitted by the sponsor to support pharmacoeconomic analyses (Appendix 2). Of note, PPS was not prespecified in the ECHELON-1 trial protocol.

Table 6: Outcomes Summarized From the ECHELON-1 Trial

EORTC QLQ-C30 = European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire Core 30; IRF = independent review facility; mPFS = modified progression-free survival; PFS = progression-free survival.

^aPFS according to IRF (data cut-off date: April 20, 2017) was also reported by the ECHELON-1 trial. Results can be found in Appendix 4.

Sources: ECHELON-1 original Clinical Study Report (data cut-off: April 20, 2017)¹³ and Addendum 1 (data cut-off: June 01, 2021),¹⁴ and the sponsor’s Summary of Clinical Evidence.¹

Efficacy Outcomes

Several definitions of key efficacy outcomes were provided in the ECHELON-1 trial.^37,38

Overall Survival

In the ECHELON-1 trial, OS was defined as the time from the date of randomization to the date of death.

Percentage of Patients Alive Without HL

The percentage of patients alive without HL was defined as the proportion of patients who were alive without classical HL at 3 years or 5 years after the patient’s randomization date.

Progression-Free Survival According to Investigator

Progression-free survival was defined as the time from the date of randomization to the date of the first of documentation of progressive disease or death due to any cause.

Modified Progression-Free Survival According to IRF

The mPFS^37,38 was assessed by the IRF according to the Revised Response Criteria for Malignant Lymphoma and defined as the time from the date of randomization to the date of the first of documentation of progressive disease, or death due to any cause, or modified progression, which was defined as receipt of anticancer therapy or radiotherapy for HL after completion of front-line therapy for patients who were confirmed noncomplete responders according to IRF. Completion of front-line therapy was defined upon receipt of the planned study drug regimen with no more than 2 missed doses of BV + AVD or ABVD in patients who did not switch therapy, or upon conclusion of 1 alternative anticancer regiment for HL after BV + AVD or ABVD discontinuation in patients who switched therapy before completion of BV + AVD or ABVD. The mPFS event date of these patients was the date of the first PET scan after completion of front-line therapy demonstrating the absence of a complete remission, defined as a Deauville score of 3 or higher.

CR Rate According to IRF

A CR according to IRF was defined as the proportion of patients who achieved a CR at the end of front-line therapy as determined by an IRF.

Health-Related Quality of Life

The summary of EORTC QLQ-C30 and EQ-5D-3L is shown in Table 7.

Table 7: Summary of HRQoL Outcome Measures and Their Measurement Properties

ECOG PS = Eastern Cooperative Oncology Group Performance Status; EORTC QLQ-C30 = European Organisation for the Research and Treatment of Cancer Quality of Life Questionnaire Core 30; EQ VAS = visual analogue scale; HL = Hodgkin lymphoma; HRQoL = health-related quality of life; MID = minimal important difference.

Harms Outcomes

The harms outcomes assessed in the ECHELON-1 trial included TEAEs, TESAEs, deaths, withdrawals due to AEs, and notable harms (treatment-emergent peripheral neuropathy, secondary malignancy, neutropenia, febrile neutropenia, and pulmonary-related toxicity). AEs were coded using the Medical Dictionary for Regulatory Activities (MedDRA). TEAEs were defined as any AE that occurs after administration of the first dose of study drug and through 30 days after the last dose of front-line therapy. Pulmonary-related toxicity included preferred terms of “dyspnea” and “hypoxia,” and all preferred terms in interstitial lung disease standardized MedDRA query, and preferred terms with the high-level term “respiratory and pulmonary function diagnostic procedures.”

Statistical Analysis

Sample Size and Power Calculation

Sample size was calculated based on the primary end point of mPFS according to IRF, with a planned enrolment of approximately 1,240 patients. To detect an HR of 0.67 in mPFS (assuming an emergent plateau in the mPFS rate after 2 years), which would indicate a 73% improvement in 2-year mPFS in the ABVD arm compared to the 81% in the BV + AVD arm, approximately 260 mPFS events were needed to achieve 90% power at a 1-sided significance level of 0.025 using a log-rank test. With 1,240 participants, there was a 95% likelihood of obtaining 260 mPFS over 60 months (assuming 36-month accrual, 24 months mPFS follow-up after the last patient recruitment, and a 5% annual dropout rate).

Statistical Testing

Details of the statistical analysis of selected efficacy end points are summarized in Table 8.

Both OS and mPFS were compared between treatment arms using the stratified log-rank test, with the stratification factors being the number of IPFP risk factors at baseline and region. A stratified Cox regression model was used to estimate HRs and 95% CIs. The Kaplan-Meier approach was used to estimate distribution of OS or mPFS end points for each treatment group. The model included covariates such as age, race, baseline ECOG PS score, baseline cancer stage, baseline B symptoms, and PET2 results, along with treatment.

Two formal analyses for OS, including an OS interim analysis at the time of the final mPFS analysis (data cut-off date of April 20, 2017) and a final analysis when 112 deaths had occurred were planned in the original protocol. An additional interim analysis of OS, which was submitted by the sponsor for the present reimbursement request, was added to protocol Amendment 9 and performed after the accrual of 103 deaths. The final OS analysis will be carried out after accrual of 112 deaths or 10 years from the randomization date of the last patient, whichever occurs first. When there was no documented death at the time of analysis, patients were censored at the date last known to be alive.

Modified PFS was tested at a 1-sided significance level of 0.025. OS was tested at a 1-sided significance level of 0.025 given that the test of mPFS reached statistical significance in the final mPFS analysis, which was prespecified in the protocol. The overall type I error was controlled using the O’Brien-Fleming boundary with a Lan-DeMets alpha spending function. The alpha level of OS at the final analysis will be adjusted for the final observed death count to maintain the overall type I error. Statistical testing adjusted for multiplicities were only conducted for OS and mPFS; P values were used for descriptive purposes only for other secondary and exploratory end points.

Sensitivity and subgroup analyses are shown in Table 8. Two additional exploratory analyses were performed, including 1 for mPFS restricted to patients who did not switch therapy and 1 for PFS according to investigator. The statistical methods of PFS according to investigator were like those of mPFS. PFS was censored at the last known alive date for those who did not have events. Patients with PFS events according to investigator after more than 1 missed visit were censored at the date of the last adequate assessment.

The percentage of patients alive without HL and the CR rate according to IRF between the 2 treatment arms were compared using a stratified Cochran-Mantel-Haenszel test. Descriptive summaries of observed EORTC QLQ-C30 and EQ-5D-3L scores were generated at each scheduled assessment time point by treatment.

All patients who received at least 1 dose of the study drug were included in the safety analysis set and were analyzed according to the actual treatment received. AEs were categorized according to the Medical Dictionary for Regulatory Activities and intensity was measured using National Cancer Institute Common Terminology Criteria for Adverse Events version 4.03. TEAEs, TESAEs, mortality, and withdrawals due to AEs are presented in this report as harms outcomes. Peripheral neuropathy, secondary malignancy, neutropenia, febrile neutropenia, pneumonitis, and pulmonary toxicity were reported as notable harms.

Table 8: Statistical Analysis of Efficacy End Points in the ECHELON-1 Trial

CI = confidence interval; CR = complete response; 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; HL = Hodgkin lymphoma; IPFP = International Prognostic Factor Project; IRF = intendent review facility; ITT = intention to treat; mPFS = modified progression-free survival; OS = overall survival; PET2 = PET response after 2 cycles of chemotherapy; PFS = progression-free survival; SD = standard deviation; VAS = visual analogue scale; vs. = versus.

Sources: Drug Reimbursement Review sponsor submission¹⁶ and the sponsor’s Summary of Clinical Evidence.¹

Analysis Populations

Analysis populations of the ECHELON-1 trial are summarized in Table 9.

Table 9: Analysis Populations of the ECHELON-1 Trial

ITT = intention-to-treat; mPFS = modified progression-free survival.

Sources: ECHELON-1 original Clinical Study Report (data cut-off: April 20, 2017)¹³ and the sponsor’s Summary of Clinical Evidence.¹

Protocol Amendments and Deviations

Nine amendments were made to the protocol. The original protocol and the protocol with the ninth amendment were issued on March 29, 2012, and September 24, 2021, respectively. A summary of key protocol amendments implemented in the ECHELON-1 trial is presented in Appendix 3. Patients were not enrolled into the study until Amendment 4. The sample size was increased in Amendment 7. In Amendment 9, an additional interim analysis of OS, which was submitted by the sponsor for the present reimbursement request, was performed.

Major protocol deviations occurred in 16 patients who were enrolled despite not satisfying eligibility criteria, and in 10 patients who received an incorrect treatment or dose of the study drug. No deviations were identified in patients who received excluded medications or were not discontinued from the study despite study withdrawal criteria not being met.

Results

Patient Disposition

A summary of patient disposition in the ECHELON-1 trial is presented in Table 10. Overall, patient dispositions were balanced between the BV + AVD and the ABVD arms. For example, similar proportions of patients completed front-line therapy between the BV + AVD and ABVD groups (92% [608 of 664] versus 93% [622 of 670]). Other than the completed maximum number of cycles per protocol (89% [593 of 664] in the BV + AVD group versus 91% [607 of 670] in the ABVD group), the second-most common reason for treatment discontinuation was AEs (4% [28 of 664] in the BV + AVD group versus 3% [22 of 670] in the ABVD group). About 12% (12 of 664) and 14% (95 of 670) were lost to follow-up.

Table 10: Summary of Patient Disposition From Pivotal Studies and RCT Evidence Submitted by the Sponsor

ABVD = doxorubicin-bleomycin-vinblastine-dacarbazine; AVD = doxorubicin-vinblastine-dacarbazine; BV = brentuximab vedotin; HL = Hodgkin lymphoma; ITT = intention-to-treat; OS = overall survival; PFS = progression-free survival.

^aThe ITT population was defined as patients who were randomized to treatment.

^bThe safety population was defined as all enrolled patients who received at least 1 dose of study medication.

^cPatients were considered to have completed study treatment per protocol if they completed front-line treatment or experienced progressive disease according to investigator or died before completion of front-line treatment.

^dCompletion of front-line treatment was defined as: upon receipt of planned study drug regimen with no more than 2 missed doses of BV + AVD or ABVD, or upon conclusion of 1 alternative anticancer regiment for HL after discontinuation of BV + AVD or ABVD.

^eOn-study deaths were defined as deaths that occurred within 30 days of the last dose of front-line therapy.

^fFollow-up deaths were defined as deaths that occurred after 30 days of the last dose of front-line therapy.

Sources: ECHELON-1 Clinical Study Report Addendum 1 (data cut-off: June 1, 2021)¹⁴ and the sponsor’s Summary of Clinical Evidence.¹

Baseline Characteristics

A summary of baseline patient demographics and disease characteristics in the ECHELON-1 trial is shown in Table 11.

A total of 1,334 patients were randomly assigned to receive BV + AVD (n = 664) or ABVD (n = 670). Baseline demographics and disease characteristics were generally well balanced between the 2 treatment groups. Overall, the median age of enrolled patients was 36 years (range = 18 to 83); most (66%, 874 of 1,334) were younger than 45 years, and 14% (186 of 1,334) were aged 60 years or older. Of the 1,334 patients enrolled, 58% (776) were male and 84% (1,114) were white. A majority (64%) of patients had stage IV disease (846), 2 or 3 (53%, or 705) IPFP risk factors, an ECOG PS of 0 (57%, or 754), extranodal involvement at diagnosis (62%, or 827), and B symptoms (59%, or 781) at baseline.

Table 11: Summary of Baseline Characteristics of Pivotal Studies and RCT Evidence Submitted by the Sponsor

ABVD = doxorubicin-bleomycin-vinblastine- dacarbazine; AVD = doxorubicin-vinblastine-dacarbazine; BV = brentuximab vedotin; ECOG PS = Eastern Cooperative Oncology Group Performance Status; HL = Hodgkin lymphoma; IPFP = International Prognostic Factor Project; ITT = intention-to-treat; SD = standard deviation.

Note: Percentages are based on nonmissing values in the ITT population in each column.

^aAge on date of informed consent.

^bTime since initial diagnosis = (first dose date of study drug – date of initial diagnosis) / 30.4375.

^cClassical HL included patients who were diagnosed with classical HL not otherwise specified.

^dPatients whose initial diagnosis was HL, then subsequently found to have been misdiagnosed.

^eStage II disease was captured as a protocol violation.

^fPatients who presented with a B symptom for ≥ 1 visit before the start of study drug administration.

Sources: ECHELON-1 Clinical Study Report Addendum 1 (data cut-off: June 1, 2021)¹⁴ and the sponsor’s Summary of Clinical Evidence.¹

Exposure to Study Treatments

Overall, patients in the BV + AVD group and those in the ABVD group received a similar median number of treatment cycles, which were administered over a similar median duration of treatment at similar median relative dose intensities. Details on the extent of exposure to study treatments in the ECHELON-1 trial are summarized in Table 12. The assigned treatments had been completed by the data cut-off date of April 20, 2017, and no new study treatments were administered as of the data cut-off on June 1, 2021.

Before the completion of the front-line therapy, of the physician’s choice, 15 patients (2%) in the BV + AVD arm and 9 (1%) in the ABVD arm switched to an alternative front-line treatment; all were treated with another form of chemotherapy only. ABVD was the most frequently reported alternative front-line treatment for the BV + AVD group and BEACOPP was the most frequently reported alternative front-line treatment for the ABVD group. In addition, in the BV + AVD arm, the switch was made mainly due to AEs (12 patients), followed by other reasons (2 patients), and the Deauville score after the PET2 assessment (1 patient). In the ABVD arm, 4 patients switched to another form of chemotherapy due to a Deauville score and other reasons, while only 1 patient switched due to AEs.

Concomitant Medications

As of April 20, 2017, at least 1 concomitant medication was reported for 659 patients (100%) treated with BV + AVD and for 653 patients (99%) treated with ABVD. Details about commonly reported concomitant medications (used in ≥ 20% of patients in either group) are shown in Table 13.

Table 12: Summary of Patient Exposure From Pivotal Studies and RCT Evidence Submitted by the Sponsor

ABVD = doxorubicin-bleomycin-vinblastine-dacarbazine; AVD = doxorubicin-vinblastine-dacarbazine; BV = brentuximab vedotin; RCT = randomized controlled trial; SD = standard deviation.

^aDuration of treatment was defined as (last dose date – first dose date + 14) / 7.

^bA treated cycle was defined as a 28-day period, during which the patient received any amount of individual drugs within the BV + AVD regimen or the ABVD regimen.

^cRelative dose intensity (%) was defined as: 100 × (total dose received) / (total dose intended). Total dose intended was the summation of the intended doses in all treatment cycles before patients switch to alternative front-line medication. The intended dose in each cycle was determined by the dose at the time the patient was randomized. For BV, total dose intended = sum of intended dose at each dosing visit over treated dosing visit, where intended dose = prescribed dose at cycle 1, day 1 × actual weight; For ABVD, total dose intended = sum of intended dose at each cycle over treated cycle, where intended dose = (prescribed dose level at cycle 1, day 1 × body surface area (m²). The unit for bleomycin is IU/m².

Sources: ECHELON-1 original Clinical Study Report (data cut-off: April 20, 2017)¹³ and the sponsor’s Summary of Clinical Evidence.¹

Table 13: Concomitant Medications for 20% or Greater of Patients in the ECHELON-1 Trial

ABVD = doxorubicin-bleomycin-vinblastine-dacarbazine; AVD = doxorubicin-vinblastine-dacarbazine; BV = brentuximab vedotin.

Sources: ECHELON-1 original Clinical Study Report (data cut-off: April 20, 2017)¹³ and the sponsor’s Summary of Clinical Evidence.¹

Subsequent Treatment

As of the data cut-off date of June 1, 2021, 20% (135 of 662) of the patients in the BV + AVD group and 24% (157 of 659) of the patients in the ABVD group received at least 1 subsequent anticancer therapy after completion of the front-line therapy (Table 14).

The most common subsequent therapy was chemotherapy (12% [79 of 662] in the BV + AVD group; 17% [111 of 659] in the ABVD group). The most common chemotherapy for the BV + AVD group was cisplatin plus cytarabine plus dexamethasone (3% [17 of 662]). In the BV + AVD arm, 3 patients (< 1%) received ABVD as subsequent treatment. In the ABVD arm, 48 (7%) patients received BV as subsequent anticancer treatment.

Other subsequent treatments included radiation therapy (8% [54 of 662] in the BV + AVD group; 8% [54 of 659] in the ABVD group), high-dose chemotherapy in combination with transplant (7% [45 of 662] in the BV + AVD group; 9% [62 of 659] in the ABVD group), immunotherapy (3% [18 of 662] in the BV + AVD group; 4% [28 of 659] in the ABVD group), hormonal therapy (0 in the BV + AVD group; < 1% [2 of 659] in the ABVD group), and other investigational drug (0 in the BV + AVD group; < 1% [2 of 659] in the ABVD group).

Table 14: Summary of Subsequent Treatment From Pivotal Studies and RCT Evidence Submitted by the Sponsor

ABVD = doxorubicin-bleomycin-vinblastine-dacarbazine; AVD = doxorubicin-vinblastine-dacarbazine; BV = brentuximab vedotin; RCT = randomized controlled trial.

Sources: ECHELON-1 Clinical Study Report Addendum 1 (data cut-off: June 01, 2021),¹⁴ Drug Reimbursement Review sponsor submission,^16, and the sponsor’s Summary of Clinical Evidence.¹

Efficacy

Key efficacy results in the ITT population of the ECHELON-1 trial are presented in Table 15. The data cut-off dates for OS were June 1, 2021 (i.e., the cut-off date in the sponsor’s Clinical Study Report) and March 11, 2023 (i.e., the cut-off date in the sponsor’s response to a request for supplementary information from the EMA). Data for PFS according to investigator, percentage of patients alive without HL, EOQTC QLQ-C30, and EQ-5D-3L were obtained from the data with a cut-off date of June 1, 2021, while data for mPFS according to IRF and CR rate according to IRF were from the data with a cut-off date of April 20, 2017. Results of subgroup analyses by disease stage (stage III only versus stage IV only) are shown in Table 16. Subgroup analyses by other factors are presented in Appendix 4.

Table 15: Summary of Key Efficacy Results From the ECHELON-1 Trial (ITT Population)

ABVD = doxorubicin-bleomycin-vinblastine-dacarbazine; AVD = doxorubicin-vinblastine-dacarbazine; BV = brentuximab vedotin; CI = confidence interval; CR = complete response; EORTC QLQ-C30 = European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire Core 30; HL = Hodgkin lymphoma; IPFP = International Prognostic Factor Project; IRF = intendent review facility; ITT = intention-to-treat; LS = least squares; mPFS = modified progression-free survival; NE = not estimable; OS = overall survival; PFS = progression-free survival; QoL = quality of life; SD = standard deviation; SE = standard error.

^aHR and 95% CI were based on a stratified Cox’s proportional hazard regression model with stratification factors region and number of IPFP risk factors at baseline with treatment as the explanatory variable in the model. A HR of less than 1 favours the BV + AVD group.

^bMedian OS follow-up was calculated from the Kaplan-Meier method switching the OS event and censored status, i.e., OS event as censored and censored as OS event.

^cMedian PFS follow-up was calculated from the Kaplan-Meier method switching the PFS according to investigator event and censored status, i.e., PFS according to investigator event as censored and censored as PFS according to investigator event.

^dMedian mPFS follow-up was calculated from the Kaplan-Meier method switching the mPFS event and censored status, i.e., mPFS event as censored and censored as mPFS event.

^eThe EORTC QLQ-C30 score range is 0 to 100. A high summary score represents a high quality of life.

^fThe range of EQ-5D VAS scores is 0 to 100. A higher score indicates a more preferred health status.

Sources: ECHELON-1 original Clinical Study Report (data cut-off: April 20, 2017),¹³ ECHELON-1 Clinical Study Report Addendum 1 (data cut-off: June 01, 2021),¹⁴ Drug Reimbursement Review sponsor submission,^16, and the sponsor’s Summary of Clinical Evidence.¹

Table 16: Summary of Key Efficacy Results From the ECHELON-1 Trial (Stage III Only Versus Stage IV Only)

ABVD = doxorubicin-bleomycin-vinblastine-dacarbazine; AVD = doxorubicin-vinblastine-dacarbazine; BV = brentuximab vedotin; CI = confidence interval; IRF = intendent review facility; ITT = intention-to-treat; mPFS = modified progression-free survival; NE = not estimable; OS = overall survival; PFS = progression-free survival; vs. = versus.

Note: OS, PFS according to investigator, the percentage of patients alive without HL, European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire Core 30, and EQ-5D-3L were obtained from the data with a cut-off date of June 1, 2021, while mPFS according to IRF, CR rate according to IRF were from the data with a cut-off date of April 20, 2017. Subgroup analyses for OS and PFS according to investigator were obtained from the data with a cut-off date of June 1, 2021, while subgroup analyses for mPFS according to IRF was from the data with a cut-off date of April 20, 2017.

^aThe HR and 95% CI were based on an unstratified Cox’s proportional hazard regression model. A hazard ratio of less than 1 favours the BV + AVD group.

^bMedian OS follow-up was calculated from the Kaplan-Meier method switching the OS event and censored status, i.e., OS event as censored and censored as OS event.

^cMedian PFS follow-up was calculated from the Kaplan-Meier method switching the PFS according to investigator event and censored status, i.e., PFS according to investigator event as censored and censored as PFS according to investigator event.

^dThe median mPFS follow-up was calculated from the Kaplan-Meier method switching the mPFS event and censored status, i.e., mPFS event as censored and censored as mPFS event.

Source: ECHELON-1 original Clinical Study Report (data cut-off: April 20, 2017),¹³ ECHELON-1 Clinical Study Report Addendum 1 (data cut-off: June 01, 2021),¹⁴ Drug Reimbursement Review sponsor submission,^16, and the sponsor’s Summary of Clinical Evidence.¹

Overall Survival

As of the data cut-off date of June 1, 2021, the median follow-up times were 73.3 months (95% CI, 72.61 to 74.05) in the BV + AVD group and 72.4 months (95% CI, 71.10 to 73.63) in the ABVD group (Table 15). In the ITT population, the HR for OS was 0.59 (95% CI, 0.396 to 0.879; P = 0.009), indicating that patients treated with BV + AVD had a 41% lower risk of death than those treated with BV + AVD. The median OS was not reached for patients with advanced-stage classical HL for either the BV + AVD group or the ABVD group. The absolute difference in number of events between the BV + AVD and ABVD arms was 4%, favouring BV + AVD (6% versus 10%). The proportion of patients who were alive on the date of last contact was 63% (415 of 664) in the BV + AVD group and 56% (372 of 670%) in the ABVD group, respectively. Of note, 10% (65 of 664) and 19% (126 of 664) of patients in the BV + AVD group were censored due to loss to follow-up and withdrawal by participants, respectively, while these proportions were 11% (76 of 670) and 21% (141 of 670), respectively, for patients receiving ABVD.

The HRs for OS were 0.863 (95% CI, 0.452 to 1.648; P = 0.654) for patients with stage III classical HL and 0.478 (95% CI, 0.286 to 0.799; P = 0.004) for patients with stage IV classical HL (Table 16). Kaplan-Meier plots of OS for the ITT population, the subpopulation with stage III classical HL, and the subpopulation with stage IV classical HL are presented in Figure 1, Figure 2, and Figure 3, respectively.

Figure 1: Kaplan-Meier Plot of OS for Patients With Advanced-Stage Classical HL (ITT Population, Data Cut-Off Date: June 1, 2021)

ABVD (denoted as A + AVD in the sponsor’s Clinical Study Report) = doxorubicin-bleomycin-vinblastine-dacarbazine; AVD = doxorubicin-vinblastine-dacarbazine; BV = brentuximab vedotin; CI = confidence interval; HL = Hodgkin lymphoma; HR = hazard ratio; IPFP = International Prognostic Factor Project; ITT = intention-to-treat; OS = overall survival.

Note: HR ([BV + AVD] / ABVD) and 95% CI were based on a stratified Cox’s proportional hazard regression model with the stratification factors, region, and number of IPFP risk factors at baseline with treatment as the explanatory variable in the model. An HR of less than 1 favours the BV + ABVD arm.

Sources: ECHELON-1 Clinical Study Report Addendum 1 (data cut-off: June 01, 2021)¹⁴ and the sponsor’s Summary of Clinical Evidence.¹

Figure 2: Kaplan-Meier Plot of OS for Patients With Stage III Classical HL (Data Cut-Off Date: June 1, 2021)

ABVD (denoted as A + AVD in the sponsor’s Clinical Study Report) = doxorubicin-bleomycin-vinblastine-dacarbazine; AVD = doxorubicin-vinblastine-dacarbazine; BV = brentuximab vedotin; CI = confidence interval; HL = Hodgkin lymphoma; HR = hazard ratio; ITT = intention-to-treat; OS = overall survival.

Note: HR ([BV + AVD] / ABVD) and 95% CI were based on unstratified Cox’s proportional hazard regression model. An HR of less than 1 favours the BV + ABVD arm.

Sources: ECHELON-1 Clinical Study Report Addendum 1 (data cut-off: June 01, 2021)¹⁴ and the sponsor’s Summary of Clinical Evidence.¹

Figure 3: Kaplan-Meier Plot of OS for Patients With Stage IV Classical HL (Data Cut-Off Date: June 1, 2021)

ABVD (denoted as A + AVD in the sponsor’s Clinical Study Report) = doxorubicin-bleomycin-vinblastine-dacarbazine; AVD = doxorubicin-vinblastine-dacarbazine; BV = brentuximab vedotin; CI = confidence interval; HL = Hodgkin lymphoma; HR = hazard ratio; ITT = intention-to-treat; OS = overall survival.

Note: HR ([BV + AVD] / ABVD) and 95% CI were based on unstratified Cox’s proportional hazard regression model. An HR of less than 1 favours the BV + ABVD arm.

Sources: ECHELON-1 Clinical Study Report Addendum 1 (data cut-off: June 01, 2021)¹⁴ and the sponsor’s Summary of Clinical Evidence.¹

New OS Data — 2023 Descriptive Analysis¹⁵

As of the data cut-off date of March 11, 2023, the descriptive analysis had a median follow-up of approximately 88 months for the ITT population (Table 17). The median follow-up durations were 89.7 months (95% CI, 86.57 to 90.55) for the BV + AVD group and 86.3 months (95% CI, 84.53 to 89.33) for the ABVD group. This analysis included 111 OS events (deaths): 44 deaths (7%) occurred in the BV + AVD group and 67 (10%) in the ABVD group. The median OS was not reached for either group. The HR for OS was 0.61 (95% CI, 0.414 to 0.892).

In the stage III subgroup (Table 18), the median OS was not reached for either treatment arm, and the HR for OS was 1.004 (95% CI, 0.540 to1.866) for BV + AVD patients compared with ABVD patients. In the stage IV subgroup, the median OS was not reached for either treatment arm, and the HR for OS was 0.48 (95% CI, 0.291 to 0.784) for BV + AVD patients compared with ABVD patients. Kaplan-Meier plots of OS for the ITT population, the subpopulation with stage III classical HL, and the subpopulation with stage IV classical HL are presented in Figure 4, Figure 5, and Figure 6, respectively.

Table 17: Summary of OS Results From the ECHELON-1 Trial (ITT Population) — 2023 Descriptive Analysis (Data Cut-Off Date: March 11, 2023)

ABVD = doxorubicin-bleomycin-vinblastine-dacarbazine; AVD = doxorubicin-vinblastine-dacarbazine; BV = brentuximab vedotin; CI = confidence interval; ITT = intention-to-treat; NE = not estimable; OS = overall survival.

^aThe HR ([BV + AVD] / ABVD) and 95% CI were based on a stratified Cox’s proportional hazard regression model. A hazard ratio of less than 1 favours the BV + AVD group.

^bThe median OS follow-up was calculated from the Kaplan-Meier method switching the OS event and censored status (i.e., OS event as censored and censored as OS event).

Source: EMA Assessment Report: Adcetris.¹⁵

Table 18: Summary of OS Results From the ECHELON-1 Trial in Stage III Versus Stage IV Disease Subgroups — 2023 Descriptive Analysis (Data Cut-Off Date: March 11, 2023)

ABVD = doxorubicin-bleomycin-vinblastine-dacarbazine; AVD = doxorubicin-vinblastine-dacarbazine; BV = brentuximab vedotin; CI = confidence interval; NE = not estimable; OS = overall survival.

^aThe HR ([BV + AVD] / ABVD) and 95% CI were based on an unstratified Cox’s proportional hazard regression model. A HR of less than 1 favours the BV + AVD group.

^bThe median OS follow-up was calculated from the Kaplan-Meier method switching the OS event and censored status (i.e., OS event as censored and censored as OS event).

Source: EMA Assessment Report: Adcetris.¹⁵

Figure 4: Kaplan-Meier Plot of OS for Patients With Advanced-Stage Classical HL (ITT Population) — 2023 Descriptive Analysis (Data Cut-Off Date: March 11, 2023)

ABVD (denoted as A + AVD) = doxorubicin-bleomycin-vinblastine-dacarbazine; AVD = doxorubicin-vinblastine-dacarbazine; BV = brentuximab vedotin; CI = confidence interval; HL = Hodgkin lymphoma; ITT = intention-to-treat; OS = overall survival.

Note: The hazard ratio ([BV + AVD] / ABVD) and 95% CI were based on a stratified Cox’s proportional hazard regression model. A HR of less than 1 favours the BV + ABVD arm.

Source: EMA Assessment Report: Adcetris.¹⁵

Figure 5: Kaplan-Meier Plot of OS for Patients With Stage III Classical HL — 2023 Descriptive Analysis (Data Cut-Off Date: March 11, 2023)

ABVD (denoted as A + AVD) = doxorubicin-bleomycin-vinblastine-dacarbazine; AVD = doxorubicin-vinblastine-dacarbazine; BV = brentuximab vedotin; CI = confidence interval; HL = Hodgkin lymphoma; OS = overall survival.

Note: The HR ([BV + AVD] / ABVD) and 95% CI were based on a stratified Cox’s proportional hazard regression model. A HR of less than 1 favours the BV + ABVD arm.

Source: EMA Assessment Report: Adcetris.¹⁵

Figure 6: Kaplan-Meier Plot of OS for Patients With Stage IV Classical HL — 2023 Descriptive Analysis (Data Cut-Off Date: March 11, 2023)

ABVD (denoted as A + AVD) = doxorubicin-bleomycin-vinblastine-dacarbazine; AVD = doxorubicin-vinblastine-dacarbazine; BV = brentuximab vedotin; CI = confidence interval; HL = Hodgkin lymphoma; OS = overall survival.

Note: The HR ([BV + AVD] / ABVD) and 95% CI were based on a stratified Cox’s proportional hazard regression model. A HR of less than 1 favours the BV + ABVD arm.

Source: EMA Assessment Report: Adcetris.¹⁵

Alive Without HL

In the ITT population as of the data cut-off date June 1, 2021, the 3-year rates of being alive without HL were about 96% (546 of 567) in the BV + AVD group and 93% (503 of 540) in the ABVD group (Table 15). The 5-year rates of being alive without HL were about 94% (450 of 480) in the BV + AVD group and 92% (408 of 443) in the ABVD group. No subgroup analyses by disease stage were reported for this efficacy end point.

PFS According to Investigator

As of the data cut-off date June 1, 2021, the median follow-ups were 73.2 months (95% CI, 72.48 to 74.05) in the BV + AVD group and 71.6 months (95% CI, 70.37 to 72.87) in the ABVD group (Table 15). In the ITT population, the HR of PFS according to investigator was 0.678 (95% CI, 0.532 to 0.863; P = 0.002), indicating that patients treated with BV + AVD had a 33.2% lower risk of experiencing PFS events compared with those treated with BV +AVD. There was a 7% absolute difference in number of PFS events between the BV + AVD group and the ABVD group, favouring BV + AVD (17% versus 24%, respectively). The median PFS according to investigator was not reached for patients with advanced-stage classical HL for either the BV + AVD group or the ABVD group. The most common reason for PFS event was progressive disease, and the most common reason for being censored in both treatment groups was the absence of a documented PFS event. Of note, 11% (73 of 664) and 15% (98 of 664) of patients in the BV + AVD group were censored due to loss to follow-up and withdrawal by participants, respectively, while these proportions were 11% (72 of 670) and 15% (101 of 670), respectively, for patients receiving ABVD.

The HRs for PFS according to investigator were 0.603 (95% CI, 0.391 to 0.930; P = 0.021) for patients with stage III classical HL and 0.715 (95% CI, 0.534 to 0.959; P = 0.024) for patients with stage IV classical HL (Table 16). Kaplan-Meier plots of PFS according to investigator for the ITT population, the subpopulation with stage III classical HL, and the subpopulation with stage IV classical HL are presented in Figure 7, Figure 8, and Figure 9, respectively.

Figure 7: Kaplan-Meier Plot of PFS According to Investigator for Patients With Advanced-Stage Classical HL (ITT Population)

ABVD (denoted as A + AVD in the sponsor’s Clinical Study Report) = doxorubicin-bleomycin-vinblastine-dacarbazine; AVD = doxorubicin-vinblastine-dacarbazine; BV = brentuximab vedotin; CI = confidence interval; HL = Hodgkin lymphoma; HR = hazard ratio; ITT = intention-to-treat; PFS = progression-free survival.

Note: The HR ([BV + AVD] / ABVD) and 95% CI were based on a stratified Cox’s proportional hazard regression model with the stratification factors, region, and number of IPFP risk factors at baseline with treatment as the explanatory variable in the model. An HR of less than 1 favours the BV + ABVD arm.

Sources: ECHELON-1 Clinical Study Report Addendum 1 (data cut-off: June 01, 2021)¹⁴ and the sponsor’s Summary of Clinical Evidence.¹

Figure 8: Kaplan-Meier Plot of PFS According to Investigator for Patients With Stage III Classical HL

ABVD (denoted as A + AVD in the sponsor’s Clinical Study Report) = doxorubicin-bleomycin-vinblastine-dacarbazine; BV + AVD = brentuximab vedotin plus doxorubicin, vinblastine, dacarbazine; HL = Hodgkin lymphoma; HR = hazard ratio; PFS = progression-free survival.

Note: The HR ([BV + AVD] / ABVD) and 95% CI were based on unstratified Cox’s proportional hazard regression model. A HR of less than 1 favours the BV + ABVD arm.

Sources: ECHELON-1 Clinical Study Report Addendum 1 (data cut-off: June 01, 2021)¹⁴ and the sponsor’s Summary of Clinical Evidence.¹

Figure 9: Kaplan-Meier Plot of PFS According to Investigator for Patients With Stage IV Classical HL

ABVD (denoted as A + AVD in the sponsor’s Clinical Study Report) = doxorubicin-bleomycin-vinblastine-dacarbazine; AVD = doxorubicin-vinblastine-dacarbazine; BV = brentuximab vedotin; CI = confidence interval; HL = Hodgkin lymphoma; HR = hazard ratio; PFS = progression-free survival.

Note: The HR ([BV + AVD] / ABVD) and 95% CI were based on unstratified Cox’s proportional hazard regression model. An HR of less than 1 favours the BV + ABVD arm.

Sources: ECHELON-1 Clinical Study Report Addendum 1 (data cut-off: June 01, 2021)¹⁴ and the sponsor’s Summary of Clinical Evidence.¹

Modified PFS According to IRF

As of the data cut-off date April 20, 2017, the median mPFS was not reached in either the BV + AVD group or the ABVD group (Table 15). In the ITT population, the HR for mPFS according to IRF was 0.770 (95% CI, 0.603 to 0.982; P = 0.035). The absolute difference in number of mPFS events between the BV + AVD arm and the ABVD arm was 4%, favouring BV + AVD (18% versus 22%). The most common reason for being censored in both treatment groups was no documented mPFS event.

The HR for mPFS according to IRF was 0.923 (95% CI, 0.600 to 1.420; P = 0.716) for patients with stage III classical HL and 0.712 (95% CI, 0.530 to 0.957; P = 0.024) for patients with stage IV classical HL (Table 16). Only the Kaplan-Meier plot of mPFS according to IRF for the ITT population was available (Figure 10).

Figure 10: Kaplan-Meier Plot of mPFS According to IRF for Patients With Advanced-Stage Classical HL (ITT Population)

ABVD (denoted as A + AVD in the sponsor’s Clinical Study Report) = doxorubicin-bleomycin-vinblastine-dacarbazine; AVD = doxorubicin-vinblastine-dacarbazine; BV = brentuximab vedotin; CI = confidence interval; HL = Hodgkin lymphoma; HR = hazard ratio; IRF = independent review facility; ITT = intention-to-treat; mPFS = modified progression-free survival.

Note: The HR ([BV + AVD] / ABVD) and 95% CI were based on a stratified Cox’s proportional hazard regression model with the stratification factors, region, and number of IPFP risk factors at baseline with treatment as the explanatory variable in the model. An HR of less than 1 favours the BV + ABVD arm.

Sources: ECHELON-1 original Clinical Study Report (data cut-off: April 20, 2017)¹³ and the sponsor’s Summary of Clinical Evidence.¹

CR Rate According to IRF

As of April 20, 2017, in the ITT population, the CR rates after the end of front-line therapy were 73% in the BV + AVD group and 71% in the ABVD group, with a corresponding relative risk of 1.038 (95% CI, 0.97 to 1.11) (Table 15). No subgroup analyses by disease stage were reported for CR rate according to IRF.

EORTC QLQ-C30

Summary and subscale scores for the EORTC QLQ-C30 were reported in the ECHELON-1 trial. Table 15 summarizes the summary scores as well as the EORTC QLQ-C30 Global Health Status/quality-of-life subscale scores at baseline, end of treatment, and 36 months after end of treatment. The mean EORTC QLQ-C30 summary scores and the Global Health Status/quality-of-life subscale scores were similar between the BV + AVD group and the ABVD group between 3 months and 36 months after end of treatment. No subgroup analyses by disease stage were reported for this efficacy end point.

EQ-5D-3L

Visual analogue scale (VAS) scores for the EQ-5D-3L were reported in the ECHELON-1 trial. Table 15 summarizes the VAS scores at baseline, end of treatment, and 36 months after end of treatment. No subgroup analyses by disease stage were reported for this efficacy end point. The scores were similar between the BV + AVD group and the ABVD group between 3 months and 36 months after end of treatment. No subgroup analyses by disease stage were reported for this efficacy end point.

Harms

A summary of harms in the ECHELON-1 trial is provided in Table 19. Deaths during the post-treatment follow-up period, resolution or improvement in peripheral neuropathy events, and secondary malignancies were from the data cut-off date of June 1, 2021, while the remaining data were from the data cut-off of April 20, 2017.

Adverse Events

The proportions of patients experiencing TEAEs up to 30 days after the last front-line dose were similar between patients treated with BV + AVD (99%) and those treated with ABVD (98%) (Table 19). However, compared to patients treated with ABVD, higher proportions of patients treated with BV + AVD were found in several TEAEs, such as neutropenia (58% versus 45%, respectively) and peripheral sensory neuropathy (29% versus 17%, respectively).

Serious Adverse Events

Higher percentages of patients in the BV + AVD group experienced TESAEs up to 30 days after the last front-line dose, compared to the percentages of patients in the ABVD group (43% versus 27%, respectively) (Table 19). Notably, 17% of patients treated with BV + AVD and 7% of patients treated with ABVD experienced febrile neutropenia.

Mortality

Deaths were reported in 6% of the patients in the BV + AVD arm and 10% of the patients in the ABVD arm (Table 19). Among them, deaths occurred in 1% of the patients treated with BV + AVD and 2% of the patients treated with ABVD within 30 days of the last dose of front-line therapy, as well as in 5% of the patients in the BV + AVD group and 8% of the patients in the ABVD group during the post-treatment follow-up period (> 30 days after the last dose of front-line therapy).

Withdrawal Due to AEs

Treatment discontinuation due to AEs occurred in 13% of the patients in the BV + AVD arm and 16% of the patients in the ABVD arm (Table 19). The most common reason for withdrawal in the BV + AVD arm was peripheral sensory neuropathy (3%), followed by neuropathy peripheral (2%) and peripheral motor neuropathy (2%).

Notable Harms

Details on notable harms are shown in Table 19. In the safety population, 67% of the patients in the BV + AVD group and 43% in the ABVD group experienced at least 1 peripheral neuropathy event. The percentages of patients who experienced treatment-emergent peripheral neuropathy of grade 3 or higher were 10% in the BV + AVD group and 2% in the ABVD group. Resolution or improvement was reported for 379 patients (86%) treated with BV + AVD and 249 patients (87%) treated with ABVD at the last follow-up.

Approximately 3% of the patients in the BV + ABVD group and 5% of the patients in the ABVD group developed secondary malignancies. The proportions of patients who experienced neutropenia as TEAEs or TEAEs of grade 3 or higher were higher in the BV + ABVD group compared with those in the ABVD group (TEAEs: 58% versus 45%; TEAEs grade 3 or higher: 54% versus 39%). Similarly, the proportions of patients who experienced febrile neutropenia as TEAEs or TEAEs of grade 3 or higher were also higher in the BV + ABVD group than in the ABVD group (TEAEs: 19% versus 8%; TEAEs grade 3 or higher: 19% versus 8%). Primary prophylaxis with G-CSF was used in 83 patients in the BV + AVD arm and 43 patients in the ABVD arm to prevent neutropenia or febrile neutropenia (Table 20). Implementation of G-CSF prophylaxis in patients receiving BV + AVD resulted in relatively lower rates of neutropenia (any events and grade 3 or higher events) and febrile neutropenia compared to those who had not received primary prophylaxis, although these rates remained higher than the rates in patients in the ABVD group who received G-CSF prophylaxis.

The ECHELON-1 trial defined pulmonary-related toxicity to include the preferred terms of “dyspnea” and “hypoxia,” all preferred terms in an interstitial lung disease standardized MedDRA query, and preferred terms with the high-level term “respiratory and pulmonary function diagnostic procedures.” Overall, fewer patients in the BV + AVD arm experienced AEs of pulmonary-related toxicity compared with those in the ABVD arm (13% versus 25%). The most common AE of pulmonary-related toxicity for either group was dyspnea (12% versus 24%).

Table 19: Summary of Harms — Pivotal and RCT Evidence

ABVD = doxorubicin-bleomycin-vinblastine-dacarbazine; AVD = doxorubicin-vinblastine-dacarbazine; BV = brentuximab vedotin.

Note: Deaths during post-treatment follow-up period, resolution or improvement in peripheral neuropathy events, and secondary malignancies were from data cut-off date June 1, 2021, while the remaining were from data as of April 20, 2017.

^aPreferred terms of neutropenia and neutrophil count decreased are counted as neutropenia.

^bFor patients without end-of-treatment date, the date was imputed as the last dose of front-line therapy plus 30 days.

^cResolution was defined as event outcome of “resolved” or “resolved with sequelae.” Patients with resolution of all peripheral neuropathy events at end of treatment (or last follow-up) were patients with all peripheral neuropathy resolved and all the resolution dates were on or before end of treatment or last follow-up date.

^dResolution implies improvement. In addition, for events that were not resolved, improvement was defined as decrease by at least 1 grade from the worst grade with no higher grade thereafter. Patients with improvement in any event at end of treatment or at last follow-up were those with at least 1 improved event and with a date of improvement before the end-of-treatment date or last follow-up date. Patients with all events resolved were excluded.

^eAn ongoing event at end of treatment was defined as event end date after the end-of-treatment date if the event end date was not missing or the last follow-up date was on or after the end-of-treatment date if the event end date is missing. Maximum severity is the maximum among grades that were occurring at end of treatment.

^fAn ongoing event at last follow-up was defined as event outcome of “unresolved” or “unknown” or events without an end date. Maximum severity is the maximum among grades that were occurring at last follow-up.

^gPulmonary-related toxicity include preferred terms of “dyspnea” and “hypoxia,” and all preferred terms in an interstitial lung disease standardized Medical Dictionary for Regulatory Activities query, and preferred terms with the high-level term “respiratory and pulmonary function diagnostic procedures.”

Sources: ECHELON-1 original Clinical Study Report (data cut-off: April 20, 2017),¹³ ECHELON-1 Clinical Study Report Addendum 1 (data cut-off: June 01, 2021),¹⁴ Drug Reimbursement Review sponsor submission,^16, and the sponsor’s Summary of Clinical Evidence.¹

Table 20: Treatment-Emergent Neutropenia and Febrile Neutropenia by G-CSF Primary Prophylaxis (ECHELON-1 Safety Population)

ABVD = doxorubicin-bleomycin-vinblastine-dacarbazine; AVD = doxorubicin-vinblastine-dacarbazine; BV = brentuximab vedotin; G-CSF = granulocyte colony-stimulating factor.

^aPreferred term of neutropenia and decreased neutrophil count are counted as neutropenia.

Sources: ECHELON-1 original Clinical Study Report (data cut-off: April 20, 2017),¹³ Drug Reimbursement Review sponsor submission,^16, and the sponsor’s Summary of Clinical Evidence.¹

Critical Appraisal

Internal Validity

The ECHELON-1 trial was an international, multicentre, phase III, open-label, randomized, active-controlled superiority trial. A stratified randomization procedure based on region (i.e., Americas versus Europe versus Asia) and number of IPFP risk factors (i.e., IPS of 0 to 1 versus 2 to 3 versus 4 to 7) was used to minimize potential imbalances between the study groups that might bias results. Baseline characteristics were similar overall between the treatment groups, although patients in the ABVD group were aged about 2 years older on average, and a larger percentage of those randomized to BV + ABVD had nodular-sclerosis classical HL (64% versus 58%, respectively); however, these differences were determined not to be clinically important. The sampled population had a generally good performance status at baseline, and approximately two-thirds of patients had stage IV disease. Neither of these factors were likely to have influenced the results in a meaningful way. The ECHELON-1 trial therefore did not appear to have major concerns related to selection bias.

As an open-label trial, the investigators, patients, and sponsor in the ECHELON-1 trial were aware of the patients’ treatment allocation. The primary outcome of mPFS and secondary outcome of CR were assessed by an IRF that was blinded, which would help limit detection bias related to the open-label design.⁵⁴ The level of concordance between mPFS according to IRF and mPFS according to investigators was reported as 91%, suggesting good agreement between the methods of assessing the primary outcome (the Internal Validity section of this review provides further discussion of mPFS). However, other important outcomes, such as PFS and HRQoL, had no blinded assessment and are therefore more prone to detection bias and performance bias. The sponsor indicated that, once the primary analysis of mPFS according to IRF was met, the trial protocol no longer required scans and disease-progression information to be sent to the IRF. As a result, PFS according to IRF was discontinued. PFS was assessed by the investigators instead in the long-term follow-up period. There was no clear evidence that the knowledge of treatment assignment among investigators influenced the assessment of PFS and the results were consistent with those from the mPFS analyses. The sponsor confirmed that no sensitivity analyses were conducted to explore the robustness of the PFS findings, including the potential impact of investigator assessment.¹⁶ Similarly, there were no obvious differences between the treatment groups in the HRQoL measurements, indicating that, if performance bias was present, it likely affected both groups.

Generally, no serious concerns were identified in the conduct of the ECHELON-1 trial, including protocol amendments and protocol deviations. The use of concomitant medications was considered by the clinical experts consulted by the review team to reflect practice settings, and the types of medications used were not expected to modify treatment effects. It was noted that, before the completion of the front-line treatment to which patients were randomized, patients were allowed to switch to an alternative front-line therapy of the physician’s choice. Overall, 15 patients (2%) in the BV + AVD arm and 9 (1%) in the ABVD arm switched to another form of chemotherapy. ABVD was the most frequently reported alternative front-line treatment for the patients originally randomized to the BV + AVD group, and BEACOPP was the most common alternative for patients randomized to the ABVD group. One concern regarding the switching was that study outcomes in those who switched might not adequately reflect the effects of the randomized study therapies. However, given that the switching rates were small (1% to 2%) and ITT analyses were applied, the risk of bias was considered low.

After completion of the front-line treatment to which patients were randomized, patients were permitted to receive subsequent treatment for HL. Receipt of subsequent therapy was considered an mPFS event. As of June 1, 2021, overall, 20% (135 of 662) of patients in the BV + AVD group and 24% (157 of 659) in the ABVD group received at least 1 subsequent anticancer therapy. The difference between the groups was primarily related to subsequent use of chemotherapy in the ABVD group (17% versus 12% in the BV + AVD group). The frequencies of use of other anticancer therapies were similar between the groups. Among the subsequent chemotherapies, 7% (48 of 659) of the patients randomized to the ABVD group received BV and 3% of patients received BV as maintenance therapy after transplant. Numerous other subsequent chemotherapy regimens were reported, with no single regimen standing out as a preferred alternative (generally used in < 2% of patients). Receiving subsequent treatment might confound the assessment of PFS and OS by delaying progression and prolonging survival beyond what would have occurred with front-line treatment alone. The overall difference between the groups in subsequent therapy use was 5%, which may or may not be important. However, as described, the only apparent difference between the groups was in the subsequent use of BV in the ABVD group. Any bias would therefore likely be in the direction of the null. The sponsor confirmed that no sensitivity analyses were conducted on PFS or OS outcomes to assess the impact of subsequent therapies. The review team concluded that the limited potential bias from the use of subsequent therapies was unlikely to have a large influence on the PFS and OS results.

The validity of the primary outcome of mPFS is a key consideration in evaluating the evidence for BV + AVD. The outcome of mPFS was adopted in the ECHELON-1 trial to capture all events that reflect a failure of front-line chemotherapy by counting a response that was less than complete at the end of the front-line therapy as an event. The ECHELON-1 trial defined a response that was less than complete as “receipt of anticancer therapy or radiotherapy for HL after completion of front-line therapy for patients who were confirmed noncomplete responders.” However, the clinical experts consulted by the review team noted that this definition is not consistent with practice in defining disease progression or first-line treatment failure in advanced HL. Furthermore, receipt of radiotherapy does not necessarily indicate disease progression in clinical practice because certain patients (e.g., those with bulky disease) may receive radiotherapy as part of their initial treatment. Despite the end-of-treatment PET scans conducted by IRF, there is a concern that the results would be biased given the administration of new anticancer therapy was at the discretion of the treating physician. Health Canada also raised this concern in its original review of BV.⁵⁵ While sensitivity analyses, including additional analyses requested by the EMA, showed the mPFS results were robust to various censoring rules. However, 2 sensitivity analyses related to how subsequent therapies were treated in the primary analysis suggested sensitivity to the definitions. In sensitivity analysis 6 of mPFS according to IRF, “subsequent therapy after completion of front-line therapy without confirmed non-CR was considered as event.”¹³ The HR for the analysis was reported to be 0.837 (95% CI, 0.679 to 1.033). In sensitivity analysis 12a of mPFS according to IRF, “for the purpose of determining a modified event, the definition of subsequent therapy was restricted to anticancer chemotherapy,” thereby excluding radiotherapy from the definition. The HR for the analysis was reported to be 0.791 (95% CI 0.618 to 1.013). In both sensitivity analyses the upper confidence limit crosses 1 and in analysis 6 the HR increases. The treatment of the “modified” component of mPFS in the analysis therefore appears to affect the results. The clinical experts noted that OS and PFS are more clinically relevant to assessing patient benefits from treatment in adults with advanced HL, while no evidence was included in the submission to the review team empirically validating mPFS as an outcome measure or that established the correlation with OS.

The proportional hazards assumption for mPFS, as well as PFS and OS, was evaluated and visual inspection of the curves indicated that the assumption was met. Multiplicities were only adjusted for the mPFS and OS, and P values for other end points, including patient-reported outcomes, were provided for descriptive purposes only. Data for OS (as of March 11, 2023) and PFS (as of June 1, 2022) were not mature (i.e., the median was not reached). A median follow-up time of approximately 88 months for OS or 70 months for PFS was considered by the clinical experts consulted by the review team to be sufficient to capture early events because most events are expected to occur within about 24 months of initial treatment in clinical practice. The clinical experts consulted by the review team also noted that a longer follow-up (≥ 10 years) for patients with HL may be needed to show a survival signal. High percentages in loss to follow-up and withdrawal by patients were noted in both OS and PFS analyses. Although the percentages of loss to follow-up and withdrawal by patients appeared to be balanced between treatment arms, reasons for loss to follow-up and withdrawal could be different between groups, which may lead to biased estimates of treatment effects. Because each patient’s outcome status is unknown and any ITT assumptions for such patients are unverifiable, ITT analysis alone is not appropriate for minimizing bias introduced by patients who are lost to follow-up. Moreover, relevant sensitivity analyses were not available for OS or PFS, and the potential impact of high loss to follow-up and withdrawal could not be determined further. Subgroup analyses by HL stage signalled that there might be a difference in treatment effects between patients with stage III classical HL and patients with stage IV classical HL for mPFS and OS. The review team’s ability to make a definitive conclusion as to whether the difference is true was limited by several concerns. First, sample sizes differed between the stages, with approximately two-thirds of patients classified with stage IV disease. Patient randomization was not stratified according to disease stage; the IPS was a stratification factor, and it includes disease stage as a component. However, the balance of known and unknown factors between treatment groups achieved by randomization may not have been preserved in stage III or stage IV subgroups. Additionally, although subgroup analyses by stage were prespecified, the study was not specifically designed to test statistical inferences between BV + AVD and ABVD in these subgroups. It is therefore impossible to rule out chance as a factor in subgroup effects between treatment arms given that no formal statistical tests for interaction between subgroups were done. It is biologically plausible that people with stage IV disease would show more benefit from treatment with BV + AVD than those with stage III disease because they may have a poorer prognosis. Because stage III disease is a lower risk, and progression or death occurs less frequently compared with stage IV HL, the treatment effect may not be observed in a trial.

External Validity

All participants in the pivotal ECHELON-1 trial were required to be aged 18 years or older. An evidence gap therefore remains with respect to the clinical efficacy and safety of BV + AVD in the pediatric population. In addition, the ECHELON-1 trial only included patients with classical HL. As a result, the pivotal trial did not reflect results for patients with nodular lymphocyte-predominant HL.

The inclusion and exclusion criteria of the ECHELON-1 trial in general were aligned with selection criteria in the Canadian settings when identifying suitable candidates for BV + AVD, according to the clinical experts consulted by the review team. However, the clinical experts noted that, in clinical practice, a small percentage of patients who were excluded from the ECHELON-1 trial may be eligible to receive BV + AVD. For example, BV + AVD can be given to patients with HIV if the disease is well managed, and to patients with a borderline left ventricle ejection fraction, after consultation with a cardiologist. The clinical experts also noted that patients with a higher ECOG PS (> 2) could be considered for treatment with BV + AVD on a case-by-case basis.

The clinical experts consulted by the review team noted that the doses of BV + AVD and ABVD used in the ECHELON-1 trial generally reflected the standard dose schedules used for adults in Canada. The clinical experts also confirmed that ABVD up to 6 cycles that is not adapted based on PET response is a relevant therapy used in the current standard of care and therefore a reasonable comparator for the adult population. However, the clinical experts also noted that ABVD up to 6 cycles not adapted based on PET response is not the only standard of care in Canada. In addition, the percentages of patients who received a transplant as subsequent treatment were lower in both groups than the experts would expect to see in clinical practice.

According to the clinical experts consulted by the review team, the study population characteristics generally reflected those of patients who would be eligible for BV + AVD in Canadian practice, and are representative of HL in Canada. However, the clinical experts noted that the percentage of patients with stage IV HL in the trial population and the percentage of white participants were higher than what would be seen in clinical practice.

Long-Term Extension Studies

No long-term extension studies were submitted by the sponsor.

Indirect Evidence

Contents within this section were informed by materials submitted by the sponsor. The following summary was validated by the review team.

No ITCs were submitted by the sponsor. The sponsor provided a feasibility assessment that determined it would be infeasible to conduct ITCs of BV + AVD versus other front-line therapies examined in clinical studies for advanced HL.

Studies Addressing Gaps in the Pivotal and RCT Evidence

In consultation with stakeholders, the review team identified a gap in the pivotal RCT evidence regarding the use of BV in pediatric patients (those aged < 18 years). The review team received input from POGO that front-line BV in combination with a chemotherapy backbone would be considered a treatment option in pediatric patients with HL based on the ECHELON-1 trial results and evidence from a published RCT that studied the addition of BV to a classic pediatric chemotherapy backbone in pediatric patients (the AHOD1331 trial¹¹). Furthermore, the sponsor identified the AHOD1331 trial as a study that could address a gap in the evidence regarding the use of BV in combination with chemotherapy in pediatric patients with high-risk HL, as the ECHELON-1 trial only enrolled adult patients (those aged ≥ 18 years). The review team confirmed with Health Canada and the sponsor that sponsor-supported data from pediatric patients with HL was not submitted to Health Canada for the supplement to a new drug submission. The review team, in consultation with the drug programs and a clinical expert specialized in treating pediatric patients with HL, identified children and adolescents with advanced or high-risk HL as having a need for new treatments, and that their exclusion resulted in an important gap in the evidence from the ECHELON-1 trial. With agreement from the sponsor, the review team therefore summarized and critically appraised the AHOD1331 trial.

Description of the AHOD1331 Trial

The AHOD1331 trial,¹¹ published in The New England Journal of Medicine, is a phase III, multicentre, open-label, randomized active-controlled trial comparing BV + AVEPC to ABVE-PC in previously untreated patients aged 2 to 21 years with high-risk classical HL, defined as patients with Ann Arbor stage IIB with bulk, stage IIIB, stage IVA and stage IVB HL. This trial was conducted in 587 participants across 153 Children’s Oncology Group institutions in North America. The number of trial sites in Canada was not reported. Patients were randomly assigned 1:1 to the BV + AVEPC or ABVE-PC group, stratified by the Ann Arbor stage (i.e., stage IIB with bulk tumour versus stage IIIB versus stage IVA versus stage IVB). The primary objective of the AHOD1331 trial was to determine the efficacy of BV + AVEPC relative to ABVE-PC as measured by EFS. Characteristics of the AHOD1331 trial are summarized in Table 21.

The AHOD1331 trial is ongoing. The final analysis of EFS was based on the database lock date of December 31, 2021.

Table 21: Details of the AHOD1331 Trial

ABVE-PC = doxorubicin-bleomycin-etoposide-prednisone-cyclophosphamide-vincristine; ALT = alanine transaminase; AST = aspartate aminotransferase; AVEPC = doxorubicin-vincristine-etoposide-prednisone-cyclophosphamide; BV = brentuximab vedotin; FEV₁ = forced expiratory volume in 1 second; FVC = forced vital capacity; HL = Hodgkin lymphoma; HRQoL = health-related quality of life; SRL = slow-responding lesion or nodal site; ULN = upper limit of normal.

^aBulk was defined as a contiguous extramediastinal nodal aggregate that measures greater than 6 cm across the longest transverse diameter (transaxial measurement) or craniocaudal dimension (measured on reformatted CT), or a mediastinal mass where the tumour diameter is greater than one-third the maximal thoracic diameter on an upright posteroanterior chest radiograph. If study eligibility by staging is uncertain, consultation with Imaging and Radiation Oncology Core Group Rhode Island may be obtained before study enrolment.

^bA formal list of end points was not available. End points were mainly identified from the objective section in the trial protocol.

^cOverall survival was not explicitly listed an exploratory end point.

Source: Castellino et al. (2022).¹¹

Populations

Inclusion and Exclusion Criteria

Previously untreated patients aged 2 to 21 years with classic HL of Ann Arbor stage IIB with bulk tumour, stage IIIB, stage IVA, or stage IVB were eligible (Table 21). Patients with nodular lymphocyte-predominant HL were excluded.

Interventions

In the AHOD1331 trial, patients in both groups received combination therapy (i.e., intervention group: BV + AVEPC, 5 cycles with 21 days for each cycle; control group: ABVE-PC, 5 cycles with 21 days for each cycle).

The BV + AVEPC regimen consisted of BV (1.8 mg/kg/dose IV) administered on day 1 before other chemotherapy, doxorubicin 25 mg/m²/dose IV or intermittent infusion on days 1 and 2, etoposide 125 mg/m²/dose IV from day 1 to day 3, prednisone 20 mg/m²/dose twice a day oral from day 1 to day 7, cyclophosphamide 600 mg/m²/dose IV on days 1 and 2, vincristine 1.4 mg/m²/dose IV or infusion (maximum dose: 2.8 mg) on day 8 only.

The ABVE-PC regimen (1 cycle) consisted of doxorubicin 25 mg/m²/dose IV or intermittent infusion on days 1 and 2, bleomycin with the dose varying on each day of administration (IV or subcutaneous), vincristine 1.4 mg/m²/dose IV or infusion (maximum dose: 2.8 mg) on days 1 and 8, etoposide 125 mg/m²/dose IV from day 1 to day 3, prednisone 20 mg/m²/dose twice a day oral from day 1 to day 7, and cyclophosphamide 600 mg/m²/dose IV on days 1 and 2.

Dose modification was permitted based on treatment-associated toxicity. G-CSF (i.e., filgrastim or pegfilgrastim) was given to all participants in the AHOD1331 trial.

Following the completion of cycle 2, patients were assessed for response to treatment. All participants received 3 additional cycles of chemotherapy unless they met criteria for removal from protocol therapy, such as progressive disease, physician’s determination, or development of a second malignancy. Patients with progressive or recurrent disease could receive subsequent anticancer therapy at the physician's discretion.

Patients without initial large mediastinal adenopathy and without any slow-responding lesion or nodal site of disease completed therapy after receiving 3 additional cycles of chemotherapy. Patients with initial large mediastinal adenopathy or any slow-responding lesion or nodal site continued to receive radiation therapy after chemotherapy. The percentages of patients who received radiation therapy were 53.4% (95% CI, 47.7 to 59.0) after receipt of BV + AVEPC and 56.8% (95% CI, 51.0 to 62.5) after the receipt of ABVE-PC.

Outcomes

A list of end points assessed in this clinical review report are provided in Table 22 and summarized further in the following section. Summarized end points are based on those included in the sponsor’s Summary of Clinical Evidence as well as any identified as important to this review according to stakeholders, for example the clinical expert, clinician groups, or patient groups. OS was not a protocol-specified end point but was reported in the published results of the AHOD1331 trial.

Table 22: Outcomes Summarized From the AHOD1331 Trial

^aOnly grade 3 or higher peripheral neuropathy was explicitly protocol prespecified. All other harms outcomes reported were not explicitly prespecified in the protocol.

Source: Castellino et al. (2022).¹¹

Overall Survival

Overall survival was defined as the time from randomization to death from any cause.

Event-Free Survival

EFS was defined as the time from randomization until disease progression or relapse, second malignancy, or death.

Harms Outcomes

Adverse events, which were considered clinically significant by the AHOD1331 trial investigators, were defined as AEs of grade 3 or higher graded with the National Cancer Institute Common Toxicity Criteria versions 4.0 and 5.0. The grade of peripheral neuropathy was determined using the Balis Pediatric Scale.⁵⁷

Statistical Analysis

Sample Size and Power Calculation

Limited information regarding the calculation of sample size was reported. The trial protocol stated that 580 eligible patients would be accrued to a maximum of 600 patients. The accrual rate for eligible patients was estimated to be about 146 patients per year, and the accrual duration was estimated to be about 4 years. An estimate of 77 observed events was considered necessary to have an 86% power to detect a difference of 8% in EFS in the BV + AVEPC group compared with the predicted EFS of 82% in the ABVE-PC group.

Statistical Testing

Details of statistical analysis of selected efficacy end points are summarized in Table 23.

For the primary end point of EFS in the ITT population, patients without an EFS event were censored at last contact. In the statistical analysis plan of the AHOD1331 trial, EFS analysis was claimed to be based on a 2-sided log-rank test with a significance level of 0.10 between the 2 randomized arms per ITT principle, and 90% CIs were planned. In the published results, EFS results were reported using a 2-sided log-rank test at a significance level of 0.05, and 95% CIs were used.

Interim and final EFS analyses were conducted. The planned final EFS analysis was scheduled to be performed after 77 events or 3 years after the last patient enrolled, whichever comes first. However, the actual final analysis of EFS was based on 74 EFS events due to the slow pace of events. The overall type I error of the EFS analyses was controlled using the Lan-DeMets alpha method.

Statistical analysis details about OS were not provided. No subgroup factors or related hypotheses were prespecified in the AHOD1331 trial protocol; post hoc subgroup analyses were conducted. No information about sensitivity analyses were identified from the AHOD1331 trial.

Table 23: Statistical Analysis of Efficacy End Points in the AHOD1331 Trial

CI = confidence interval; EFS = event-free survival; NR = not reported; OS = overall survival; vs. = versus.

Source: Castellino et al. (2022).¹¹

Analysis Populations

In the AHOD1331 trial, the ITT population, including all the patients who had undergone randomization, was used for both efficacy and harms outcomes summarized in the report.

Protocol Amendments and Deviations

The original and final protocols were issued on December 9, 2014, and June 22, 2020, respectively. The AHOD1331 trial initially enrolled patients younger than 18 years of age, and was later expanded to patients aged less than 22 years. Details on protocol deviations were not reported.

Results

Patient Disposition

Out of 600 participants screened, 13 were excluded due to a stage not consistent with eligibility at central review (n = 9), tests for eligibility out of window or out of range (n = 3), or incorrect consent process (n = 1).

Baseline Characteristics

A summary of baseline patient demographics and disease characteristics in the AHOD1331trial is shown in Figure 11.

A total of 587 patients were randomly assigned to receive BV plus chemotherapy (i.e., BV + AVEPC; n = 298) or standard care (ABVE-PC; n = 289). Baseline demographics and disease characteristics were generally well balanced between the 2 treatment groups. The median age of participants was 15.6 years (range = 3.4 to 21.99); most (84.7%) were between 12 and 22 years. Of the total number of patients enrolled, 47% were female and 57.6% were non-Hispanic white. The proportions of patients by disease stage were 20.6% for stage IIB with bulk tumour, 19.3% for stage IIIB, 28.4% for stage IVA, and 31.7% for stage IVB. Most of the patients had nodular-sclerosis classical HL (76.5%).

Figure 11: Summary of Baseline Characteristics of the AHOD1331 Trial

From: NEJM, Castellino et al,¹¹ Brentuximab vedotin with chemotherapy in pediatric high-risk Hodgkin’s lymphoma, Vol 387, p 1649 to 1660. Copyright © 2022 Massachusetts Medical Society. Reprinted with permission from Massachusetts Medical Society.

Exposure to Study Treatments, Concomitant Medications, Subsequent Treatment

Dose modification on any drug occurred in 28.2% of patients in the BV + AVEPC arm and 26.0% in the ABVE-PC arm. Dose modification on BV occurred in 8.1% of patients in the BV + AVEPC group. No concomitant cancer chemotherapy or immunomodulating agents were allowed. Corticosteroid therapy was only allowed for anaphylactic reactions, adrenal insufficiency, and severe asthma or as a breakthrough antiemetic. Antibiotics, blood products, antiemetics, fluids, electrolytes, and general supportive care were used.

Efficacy

The median follow-up time was 42.1 months (range = 0.1 to 80.9).

Three-Year Overall Survival

In terms of the 3-year OS in the ITT population, 99.3% (95% CI, 97.3 to 99.8) of patients were censored in the BV + AVEPC group and 98.5% (95% CI, 96.0 to 99.4) were censored in the ABVE-PC group. The HR for 3-year OS was not provided.

Three-Year EFS

In terms of the 3-year EFS in the ITT population, 92.1% (95% CI, 88.4 to 94.7) were censored in the BV + AVEPC group and 82.5% (95% CI, 77.4 to 86.5) were censored in the ABVE-PC group. The HR for 3-year EFS was 0.41 (95% CI, 0.25 to 0.67; P < 0.001) favouring the BV + AVEPC arm. First events occurred in 23 out of 298 patients and 51 out of 289 patients in the BV + AVEPC and ABVE-PC groups, respectively. Each group included 1 patient who developed secondary cancer (BV + AVEPC: acute myeloid leukemia; ABVE-PC: papillary thyroid carcinoma). One patient in the ABVE-PC group died due to a motor vehicle accident.

Harms

As shown in Figure 12, the occurrences of any AEs of grade 3 or higher were 73.5% in patients treated with BV + AVEPC and 68.2% in patients treated with ABVE-PC. Peripheral neuropathies of grade 3 or higher occurred in 6.7% of the patients in the BV + AVEPC arm and 5.5% % of the patients in the ABVE-PC arm. Febrile neutropenia occurred in 30.9% and 32.5% of patients in the BV + AVEPC and ABVE-PC arms, respectively. None of the patients in the BV + AVEPC group experienced pneumonitis, while only 1 patient in the ABVE-PC group did.

Figure 12: Summary of Harms — AHOD1331 Trial

From: NEJM, Castellino et al,¹¹ Brentuximab vedotin with chemotherapy in pediatric high-risk Hodgkin’s lymphoma, Vol 387, p 1649 to 1660. Copyright © 2022 Massachusetts Medical Society. Reprinted with permission from Massachusetts Medical Society.

Critical Appraisal

Internal Validity

Although details about the randomization process and allocation concealment were not reported in the research protocol or the main article, the risk of bias is anticipated to be low given that baseline characteristics between the treatment arms were generally similar for clinically important factors. The AHOD1331 trial was open-label but had blinded outcome assessors, the definition of EFS was aligned with accepted definitions from regulators, and treatment response was assessed via a centralized review, helping reduce the risk of detection bias related to the open-label design. Even though patients were aware of the treatment allocation, which could result in performance bias, the risk is considered low because the 3-year PFS in the ABVE-PC group (82.5%) and the types of AEs were generally in line with what the clinical experts consulted by the review team expected.

Those patients who remained PET-positive after 2 cycles of chemotherapy received response-adapted involved-site radiation therapy guided by blinded central assessment of PET scans. This could bias EFS results if the radiation therapy improved responses, reduced the likelihood of relapse, and/or increased the risk of secondary malignancy. However, the risk of this potential bias was mitigated by the requirement that radiation therapy could not be administered until directed to do so by the blinded assessment. Also, the percentages of patients who received involved-site radiation therapy were similar between the BV + AVEPC group and the ABVE-PC group (53.4% versus 56.8%, respectively).

Concomitant anticancer medications were not allowed. Antibiotics and supportive medications (e.g., antiemetics) were permitted as needed. Patients also received G-CSF support. None of the permitted medications would likely influence the results for either treatment group. However, after progression events, the treating physician could treat the patient at their discretion, which may affect the longer-term OS results.

External Validity

The AHOD1331 trial was appraised in this section to address an important gap with respect the unmet needs of using BV + chemotherapy in pediatric patients with high-risk HL. However, several notable issues need to be considered when generalizing results from the AHOD1331 trial. First, although the chemotherapy backbone used in the AHOD1331 trial (AVEPC) is a preferred backbone for pediatric patients according to POGO and the clinical expert consulted by the review team, it is different from the backbone used in adults (ABVD). Regarding the regulatory status of the pediatric regimen, the review team confirmed that BV is not approved for use in combination with the pediatric regimen and the sponsor confirmed that it is not planning to file for Health Canada approval of BV + AVEPC. Second, the clinical experts consulted by the review team noted the definition of high-risk or advanced-stage HL in pediatric patients varies. While the AHOD1331 trial adopted the definition of advanced-stage HL in pediatric patients as stage II with bulk tumour, stage IIIB, stage IVA, and stage IVB, some medical centres define any stage III or IV disease as advanced-stage disease in pediatric patients. Finally, the AHOD1331 trial involved both nonadults and young adults (up to the age of 22 years), while the pivotal ECHELON-1 trial enrolled patients aged 18 years and older. This created an overlap in patient ages between the pivotal ECHELON-1 trial and the AHOD1331 trial. The clinical experts consulted by the review team noted that the chemotherapy backbone AVEPC used in the AHOD1331 trial would not typically be used in patients aged 18 years or older in Canada and the chemotherapy backbone ABVD investigated in the pivotal ECHELON-1 trial may be used in adolescents aged close to 18 years with HL.

Discussion

Summary of Available Evidence

One ongoing phase III, open-label, randomized, active-controlled, superiority trial (ECHELON-1, N = 1,334, with 60 patients in Canada enrolled) was included. The primary objective of the ECHELON-1 trial was to determine the efficacy of BV + AVD relative to ABVD as measured by mPFS, while the key secondary objective was to compare OS. Other efficacy and safety end points were also examined, including PFS, percent alive without HL, CR rate, EORTC QLQ-30, EQ-5D-3L, TEAEs, TESAEs, deaths, withdrawals due to AEs, and notable harms (i.e., treatment-emergent peripheral neuropathy, secondary malignancy, neutropenia, febrile neutropenia, and pulmonary-related toxicity). The ECHELON-1 trial enrolled previously untreated adult patients with histologically confirmed stage III or stage IV classical HL with an ECOG PS of 2 or lower. The median age of enrolled participants was 36 years (range = 18 to 83), with 66% (874 of 1,334) aged younger than 45 years and 14% (186 of 1,334) aged 60 years or older. Of the total number of participants, 58% (n = 776) were male and 84% (n = 1,114) were white. Most patients had stage IV disease (64%, n = 846), 2 or 3 (53%, n = 705) IPFP risk factors, an ECOG PS of 0 (57%, n = 754), extranodal involvement at diagnosis (62%, n = 827), and B symptoms (59%, n = 781) at baseline.

No ITCs or long-term extension studies were submitted by the sponsor. The sponsor provided a feasibility assessment that determined it would be infeasible to conduct ITCs of BV + AVD versus other front-line therapies examined in clinical studies for advanced HL.

A phase III, multicentre, open-label, randomized active-controlled trial (AHOD1331, N = 587), published in The New England Journal of Medicine, was included to address the evidence gap regarding the use of BV + chemotherapy in pediatric patients with advanced-stage classical HL. The primary objective of the AHOD1331 trial was to assess the efficacy of BV + AVEPC versus ABVE-PC as measured by EFS. OS and safety outcomes were also reported in the published results. The AHOD1331 trial involved patients aged 2 to 21 years who were previously untreated for high-risk classical HL (i.e., Ann Arbor stage IIB with bulk tumour, stage IIIB, stage IVA, stage IVB). The median age of enrolled patients was 15.6 years (range = 3.4 to 21.99); most (84.7%, 497 of 587) were between 12 and 21 years. Of the 587 patients enrolled, 47% (276) were female and 57.6% (338) were non-Hispanic white. The proportions of patients by disease stage were 20.6% (121 of 587) for stage IIB with bulk tumour, 19.3% (113 of 587) for stage IIIB, 28.4% (167 of 587) for stage IVA, and 31.7% (186 of 587) for stage IVB. Most of the patients (76.5%, 449 of 587) had nodular-sclerosis classical HL.

Interpretation of Results

Efficacy

Overall, efficacy evidence from the ECHELON-1 trial indicated that BV + AVD is more efficacious than ABVD in patients with advanced-stage HL. This conclusion was made after considering the efficacy outcomes (i.e., OS, mPFS according to IRF, PFS according to investigator, percentage of patients who were alive without HL, and CR rate according to IRF) examined in the report and selected with input from the consulted clinical experts, and by input to the review team from patient and clinician groups. These efficacy outcomes aligned with patients’ expectation of important outcomes, including controlling disease symptoms, achieving longer disease remission, and prolonging life.

In general, the OS results (in the ITT population as well as the stage III and IV subgroups) were consistent between the updated descriptive analysis as of March 11, 2023, and the analysis as of June 1, 2021. OS was identified as the most clinically important outcome in our previous²⁵ and current reviews of BV + AVD for the treatment of advanced classical HL. An important consideration in our previous review was the relatively short duration of the follow-up in the OS analyses (median < 30 months), which was determined to be immature at that time. Given the natural history of HL and use of standard-of-care chemotherapy as a comparator (which includes ABVD), a much longer follow-up period would be required to assess the effects of BV + AVD on OS. The updated OS data (cut-off date of March 11, 2023) had a median follow-up of approximately 88 months for the ITT population. The HR for OS (data cut-off date of March 11, 2023) was 0.607 (95% CI, 0.414 to 0.892; P = 0.010). The absolute reduction in number of OS events between the BV + AVD group and the ABVD group was 3%, favouring the BV + AVD arm over the follow-up period, which the clinical experts considered to be meaningful. The 4-year (extent of separation: 2.8 months), 5-year (3.6 months), 6-year (4.4 months), and 7-year (4.7 months) Kaplan-Meier survival estimates suggest an increasing separation of the survival curves with longer follow-up; however, the extent of separation was slightly decreased at year 8 (4.3 months). Given that later estimates are obviously based on decreasing sample sizes and associated with less precision, as well as the fact that the median OS has not yet been reached in either group, the longer-term differences in OS remain unclear. As classical HL is generally considered a curable disease, the clinical experts consulted by the review team noted that the overall 3% absolute difference was, in their opinions, clinically meaningful. However, no empirically validated minimal important difference exists for OS in patients with advanced classical HL, making it difficult to interpret the size of the between-group difference in OS and to gauge the added clinical value of BV + AVD versus ABVD beyond statistical significance. Because patient-group input to the review team stated that treatments that allow for longer life are important to them, the observed OS difference with BV + AVD may be meaningful to patients. Patient-group input also highlighted the importance of improved HRQoL. The HRQoL outcome measures, the EORTC QLQ-C30, and EQ-5D-3L VAS, showed improvements in both treatment groups and similar scores between the BV + AVD and ABVD groups. However, interpretation of these results is complicated by the analyses’ exploratory nature and the absence of formal between-group comparisons for the EQ-5D-3L VAS. Combining the OS and HRQoL results indicates a favourable effect on OS with BV + AVD versus ABVD and no apparent meaningful difference in HRQoL. However, the percentage of patients who experienced SAEs and notable harms with BV + AVD versus ABVD was nearly double (as discussed in the Harms section).

Although uncertainty remains as to the magnitude of the OS benefit, the results are supported by evidence that BV + AVD reduces disease progression to a greater degree than does ABVD. Benefit with BV + AVD was also found in mPFS according to IRF (data cut-off date of April 20, 2017; HR for mPFS: 0.770; 95% CI, 0.603 to 0.982; absolute difference in number of mPFS events: 4%, favouring BV + AVD) and PFS according to investigator (data cut-off date of June 1, 2021; HR for PFS: 0.678; 95% CI, 0.532 to 0.863; absolute difference in number of PFS events: 7%, favouring BV + AVD). Despite the failure to reach the median OS, median mPFS according to IRF, and median PFS according to investigator, the follow-up time (i.e., median of about 88 months for OS, > 70 months for PFS, and 24.9 months for mPFS) was considered sufficient to capture early events, as in clinical practice most events are expected to occur within 24 months after treatment, according to the clinical experts consulted by the review team. Again, the clinical experts considered the results of the PFS assessments to be clinically meaningful, although there is no empirically derived threshold for clinical significance to determine this objectively. Further uncertainty comes from the validity of mPFS as an outcome in classical HL (as mentioned in the Critical Appraisal section). However, patient-group input identified delayed disease progression as an important outcome.

Other sources of uncertainty include the relatively high percentages in loss to follow-up and withdrawal by patients in both analyses for OS and PFS according to investigator. For example, in the interim OS data analysis (data cut-off date of March 11, 2023), 13% of the patients in the BV + AVD group and 13% of patients in the ABVD group were lost to follow-up; overall withdrawals occurred in 21% of the patients in the BV + AVD group and 22% in the ABVD group. The percentages were balanced between the 2 treatment arms. However, the reasons for the loss to follow-up and withdrawal were not reported and could differ between groups, which could lead to biased estimates. ITT analysis alone cannot minimize bias introduced by patients who are lost to follow-up because each patient’s outcome status is unknown and any ITT assumptions for such patients is unverifiable. Moreover, with no relevant sensitivity analyses available, as confirmed by the sponsor,¹⁶ the impact of higher loss to follow-up and withdrawal could not be further determined. It is acknowledged that larger losses to follow-up occur in longer-term studies such as the ECHELON-1 trial.

Another key source of uncertainty is the subgroup effect regarding the stage of advanced classical HL (stage III versus IV) at baseline. Potentially small treatment effects (and uncertain benefit-to-risk ratios) were identified in 2019 by Health Canada, which restricted the indication to stage IV disease.⁵⁵ Although PFS improved at both stage III (HR = 0.603; 95% CI, 0.391 to 0.930; data cut-off of June 1, 2021) and stage IV (HR = 0.715; 95% CI, 0.534 to 0.959; data cut-off date of June 1, 2021) subgroups, the hazards for OS and mPFS were lower in patients with stage IV classical HL (the HR for OS as of March 11, 2023 was 0.478; 95% CI, 0.286 to 0.799 and the HR for mPFS as of April 20, 2017 was 0.712; 95% CI, 0.530 to 0.957), compared to patients with stage III disease (HR for OS as of March 11, 2023 = 1.004, 95% CI, 0.540 to 1.866; HR for mPFS as of April 20, 2017 = 0.923; 95% CI, 0.600 to 1.420). However, as described in the Critical Appraisal section, there are numerous reasons the review team was not able to make definitive conclusions about whether there is a true difference in efficacy between patients with stage IV HL and those with stage III HL. This includes, among others, the fact that the majority (> 60%) of enrolled patients had stage IV disease, which caused an imbalance in sample size between the stage subgroups. This complicates interpretation of the subgroup analyses by stage, but it also means that the results of ECHELON-1 are easier to generalize to those with stage IV disease. The review team acknowledged that the inclusion of more patients with stage IV disease is practical for the trial, as it would allow for detection of a difference between treatments over a shorter duration of follow-up given the poorer prognosis in stage IV than stage III. However, determining whether the magnitude of benefit with BV + AVD is greater in those with stage IV disease than in those with stage III disease is challenging. While there is a signal of a subgroup difference, the design and analysis of the study precludes the review team from drawing any certain conclusion. The clinical experts consulted by the review team emphasized that the approach to treatment, particularly with for those with experience using ABVD, is to treat stage III and stage IV classical HL the same. It was therefore their opinion that BV + AVD would be used in the same manner as ABVD for advanced classical HL, regardless of stage.

An important gap in the evidence from the ECHELON-1 trial is the lack of data in people with advanced HL who are aged younger than 18 years. The review team received input from a clinical expert who specializes in treating pediatric patients with HL and a clinician group (POGO) indicating that BV in combination with chemotherapy would be used as front-line treatment for pediatric patients with advanced or high-risk HL. To address this gap, the review team reviewed evidence regarding the use of BV in combination with a pediatric chemotherapy backbone in patients aged 2 to 21 years with high-risk or advanced-stage classical HL from the AHOD1331 trial. Regarding the regulatory status of the pediatric regimen, the review team confirmed that BV is not approved for use in combination with the pediatric chemotherapy regimen and the sponsor confirmed that it is not planning to file for Health Canada approval for BV + AVEPC. Evidence from the AHOD1331 trial indicated that BV + AVEPC is more effective than ABVE-PC for EFS in patients aged 2 to 21 years with high-risk classical HL defined as Ann Arbor stage IIB with bulk tumour, stage IIIB, stage IVA, and stage IVB HL. According to the clinical experts consulted by the review team, the EFS in the ABVE-PC group (82.5% of patients were censored) reflected the effectiveness of ABVE-PC seen in clinical practice. The 6.6% absolute difference in the number of EFS events between the BV + AVEPC group and the ABVE-PC group favouring the former was clinically meaningful in the opinion of the clinical experts. The median follow-up time for EFS was considered adequate given that most events in clinical practice are expected to occur within 24 months after treatment. However, in the AHOD1331 trial, no difference between treatment groups was found in OS. The clinical expert consulted by the review team stated this would be expected, as 42 months of follow-up would be too short for OS events to occur in pediatric and young adult patients because HL is curable in these patients. The experts therefore indicated that EFS was a clinically relevant outcome.

Harms

In the pivotal ECHELON-1 trial, the observed AEs were consistent with the safety profiles of BV + AVD and ABVD according to the clinical experts consulted by the review team. Fewer patients treated with BV + AVD in the ECHELON-1 trial discontinued treatment due to AEs or experienced secondary malignancy and pulmonary-related toxicity compared with those receiving ABVD. More patients in the BV + AVD group experienced TEAEs and TESAEs, as well as notable harms such as grade 3 or higher treatment-emergent peripheral neuropathy, neutropenia, and febrile neutropenia, compared to those in the ABVD group. The clinical experts noted that the AEs were consistent with what they typically manage in clinical practice and were encouraged that key AEs with BV, such as peripheral neuropathy, resolved or improved in most patients. In the AHOD1331 trial, no difference was identified overall across harms outcomes between the BV + AVEPC group and the ABVE-PC group.

Conclusion

Overall, evidence from the phase III, open-label, randomized, ECHELON-1 trial suggested that BV + AVD is an effective front-line treatment for previously untreated adult patients diagnosed with advanced-stage classical HL. The clinically relevant efficacy end points examined in the report (OS, PFS, and mPFS) were consistently in favour of BV + AVD compared to ABVD in the ITT population. However, the clinical significance of the magnitude of the treatment differences is uncertain and concerns remain about the validity of the primary outcome (mPFS), as well as the high percentages in loss to follow-up and withdrawal by patients in both OS and PFS analyses. In addition, although subgroup analyses of the OS and mPFS results signalled that BV + AVD may be more effective in patients with stage IV classical HL than in those with stage III HL, conclusions regarding subgroup differences between stage III versus stage IV patients are uncertain because the study was not designed to detect differences between these subgroups and patients were not stratified at randomization according to disease stage. The safety profile of BV + AVD is consistent with the known AEs for the individual components of the regimen, but more patients treated with BV + AVD experienced SAEs compared with those in the ABVD group. An evidence gap remains with respect to the clinical efficacy and safety of BV + AVD in the pediatric population as all participants in the pivotal ECHELON-1 trial were required to be aged 18 years or older. To address the unmet needs in the pediatric patient population, the AHOD1331 trial was examined. Efficacy results from the AHOD1331 trial indicated that BV + AVEPC provides a clinically meaningful benefit in EFS compared to ABVE-PC in patients aged 2 to 21 years with high-risk classical HL defined as Ann Arbor stage IIB with bulk tumour, stage IIIB, stage IVA, and stage IVB HL.

References

1.Manufacturer Summary of Clinical Evidence [internal manufacturer's report]. In: Drug Reimbursement Review manufacturer submission: Brentuximab vedotin for injection (Adcetris), 50 mg/vial, lyophilized powder for reconstitution, intravenous infusion. Mississauga (ON): Seagen Canada Inc.; 2023 Feb.

2.Adcetris (brentuximab vedotin): Lyophilized powder for reconstitution with 10.5 mL of sterile water for injection, USP 50 mg [product monograph]. Bothell (WA): Seagen Inc.; 2021 Jun 11: https://pdf.hres.ca/dpd_pm/00061748.PDF. Accessed 2024 May 27.

3.Canadian Cancer Society. What is Hodgkin Lymphoma? [manufacturer-provided reference]. [2023]; https://cancer.ca/en/cancer-information/cancer-types/hodgkin-lymphoma/what-is-hodgkin-lymphoma. Accessed 2023 Jan 13.

4.Brenner DR, Poirier A, Woods RR, et al. Projected estimates of cancer in Canada in 2022. CMAJ. 2022;194(17):E601-E607. PubMed

5.Canadian Cancer Statistics: A 2018 special report on cancer incidence by stage [manufacturer-provided reference]. Toronto (ON): Canadian Cancer Society; 2018: https://cdn.cancer.ca/-/media/files/research/cancer-statistics/2018-statistics/canadian-cancer-statistics-2018-en.pdf.

6.Nagpal P, Akl MR, Ayoub NM, et al. Pediatric Hodgkin lymphoma: biomarkers, drugs, and clinical trials for translational science and medicine. Oncotarget. 2016;7(41):67551-67573. PubMed

7.Canadian Cancer Society. Childhood Hodgkin lymphoma statistics. 2024; https://cancer.ca/en/cancer-information/cancer-types/hodgkin-lymphoma-childhood/statistics. Accessed 2024 Mar 19.

8.Ansell SM. Hodgkin lymphoma: A 2020 update on diagnosis, risk-stratification, and management. Am J Hematol. 2020;95(8):978-989. PubMed

9.NCCN Clinical Practice Guidelines in Oncology: Hodgkin Lymphoma Version 1.2023 [manufacturer-provided reference]. Plymouth Meeting (PA): National Comprehensive Cancer Network (NCCN); 2022: https://www.nccn.org/.

10.CADTH Drug Reimbursement Expert Review Committee final recommendation: Brentuximab Vedotin (Adcetris - Seattle Genetics, Inc.). Ottawa (ON): CADTH; 2020 Dec 3: https://www.cda-amc.ca/sites/default/files/pcodr/Reviews2020/10214BrentuximabVedotinHLAVD_FnRec_approvedbyChair_REDACT_Post03Dec2020_final.pdf. Accessed 2023 Apr 26.

11.Castellino SM, Pei Q, Parsons SK, et al. Brentuximab Vedotin with Chemotherapy in Pediatric High-Risk Hodgkin’s Lymphoma. N Engl J Med. 2022;387(18):1649-1660. PubMed

12.Herrera AF, LeBlanc ML, Castellino SM, et al. SWOG S1826, a randomized study of nivolumab(N)-AVD versus brentuximab vedotin(BV)-AVD in advanced stage (AS) classic Hodgkin lymphoma (HL). J Clin Oncol. 2023;41(17_suppl):LBA4-LBA4.

13.Clinical Study Report: C25003. A randomized, open-label, phase 3 trial of A+AVD versus ABVD as front-line therapy in patients with advanced classical Hodgkin lymphoma [internal manufacturer's report]. Cambridge (MA): Millennium Pharmaceuticals, Inc.; 2017 Aug 28.

14.Clinical Study Report: C25003. A randomized, open-label, phase 3 trial of A+AVD versus ABVD as front-line therapy in patients with advanced classical Hodgkin lymphoma addendum 1 [internal manufacturer's report]. Lexington (MA): Takeda Development Center Americas, Inc.; 2022 Jun 16.

15.Committee for Medicinal Products for Human Use. Assessment Report: Adcetris. (European public assessment report). Amsterdam (NL): European Medicines Agency; 2023: https://www.ema.europa.eu/en/documents/variation-report/adcetris-h-c-002455-ii-0107-epar-assessment-report-variation_en.pdf. Accessed 2024 Mar 14.

16.Drug Reimbursement Review manufacturer submission: Brentuximab vedotin for injection (Adcetris), 50 mg/vial, lyophilized powder for reconstitution, intravenous infusion [internal manufacturer's package]. Mississauga (ON): Seagen Canada Inc.; 2023 Apr 12.

17.Kuppers R. The biology of Hodgkin's lymphoma. Nat Rev Cancer. 2009;9(1):15-27. PubMed

18.Ansell SM. Hodgkin lymphoma: 2018 update on diagnosis, risk-stratification, and management. Am J Hematol. 2018;93(5):704-715. PubMed

19.Leukemia and Lymphoma Society of Canada. Facts and Statistics About Blood Cancers [manufacturer-provided reference]. 2015; https://www.llscanada.org/disease-information/facts-and-statistics.

20.Shanbhag S, Ambinder RF. Hodgkin lymphoma: A review and update on recent progress. CA Cancer J Clin. 2018;68(2):116-132. PubMed

21.BC Cancer. Cancer Mangement Manual: Hodgkins Lymphoma [manufacturer-provided reference]. 2013; http://www.bccancer.bc.ca/health-professionals/clinical-resources/cancer-management-manual/lymphoma-chronic-leukemia-myeloma/hodgkin-lymphoma.

22.Lymphoma Clinical Practice Guidelines [manufacturer-provided reference]. Edmonton (AB): Cancer Care Alberta; 2019: https://www.albertahealthservices.ca/assets/info/hp/cancer/if-hp-cancer-guide-lyhe002-lymphoma.pdf.

23.Brice P, de Kerviler E, Friedberg JW. Classical Hodgkin lymphoma. Lancet. 2021;398(10310):1518-1527. PubMed

24.Canadian Cancer Statistics 2021 [manufacturer-provided reference]. Toronto (ON): Canadian Cancer Society; 2021: https://cdn.cancer.ca/-/media/files/research/cancer-statistics/2021-statistics/2021-pdf-en-final.pdf.

25.CADTH PCODR Final Clinical Guidance Report: Brentuximab Vedotin (Adcetris) [manufacturer-provided reference]. Ottawa (ON): CADTH; 2020: https://www.cadth.ca/sites/default/files/pcodr/Reviews2020/10214BrentuximabVedotinHLAVD_fnCGR_REDACT_Post03Dec2020_final.pdf.

26.Bleomycin for Injection [product monograph; manufacturer-provided reference]. Fresenius Kabi Canada, Ltd; 2016.

27.Cheson BD, Fisher RI, Barrington SF, et al. Recommendations for initial evaluation, staging, and response assessment of Hodgkin and non-Hodgkin lymphoma: the Lugano classification. J Clin Oncol. 2014;32(27):3059-3068. PubMed

28.Kaseb H, Babiker HM. Hodgkin Lymphoma [manufacturer-provided reference]. StatPearls. Treasure Island (FL): StatPearls; 2022.

29.Ansell SM, Radford J, Connors JM, et al. Overall Survival with Brentuximab Vedotin in Stage III or IV Hodgkin's Lymphoma. N Engl J Med. 2022;387(4):310-320. PubMed

30.Connors JM, Jurczak W, Straus DJ, et al. Brentuximab Vedotin with Chemotherapy for Stage III or IV Hodgkin's Lymphoma. N Engl J Med. 2018;378(4):331-344. PubMed

31.Evens AM, Connors JM, Younes A, et al. Older patients (aged ≥60 years) with previously untreated advanced-stage classical Hodgkin lymphoma: a detailed analysis from the phase III ECHELON-1 study. Haematologica. 2022;107(5):1086-1094. PubMed

32.Hutchings M, Radford J, Ansell SM, et al. Brentuximab vedotin plus doxorubicin, vinblastine, and dacarbazine in patients with advanced-stage, classical Hodgkin lymphoma: A prespecified subgroup analysis of high-risk patients from the ECHELON-1 study. Hematol Oncol. 2021;39(2):185-195. PubMed

33.Ramchandren R, Advani RH, Ansell SM, et al. Brentuximab Vedotin plus Chemotherapy in North American Subjects with Newly Diagnosed Stage III or IV Hodgkin Lymphoma. Clin Cancer Res. 2019;25(6):1718-1726p. PubMed

34.Straus DJ, Dlugosz-Danecka M, Connors JM, et al. Brentuximab vedotin with chemotherapy for stage III or IV classical Hodgkin lymphoma (ECHELON-1): 5-year update of an international, open-label, randomised, phase 3 trial. Lancet Haematol. 2021;8(6):e410-e421. PubMed

35.Straus DJ, lstrokugosz-Danecka M, Alekseev S, et al. Brentuximab vedotin with chemotherapy for stage III/IV classical Hodgkin lymphoma: 3-year update of the ECHELON-1 study. Blood. 2020;135(10):735-742. PubMed

36.Takeda. NCT01712490: A Randomized, Open-label, Phase 3 Trial of A+AVD Versus ABVD as Frontline Therapy in Patients With Advanced Classical Hodgkin Lymphoma [manufacturer-provided reference]. ClinicalTrials.gov. Bethesda (MD): U.S. National Library of Medicine; 2018: https://clinicaltrials.gov/ct2/show/NCT01712490.

37.C25003: Statistical Analysis Plan Amendment 1 [internal manufacturer's report]. Cambridge (MA): Millennium Pharmaceuticals, Inc.; 2021 Dec 7.

38.C25003: Clinical Study Protocol Incorporating Amendment 9 [internal manufacturer's report]. Lexington (MA): Takeda Development Center Americas, Inc.; 2021 Sept 24.

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

40.EORTC Quality of Life. EORTC Quality of life: FAQs. [2023]; https://qol.eortc.org/faq/. Accessed 2023 Jun 19.

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

42.EORTC QLC-C30 scoring manual. Brussels (BE): EORTC Data Center; 2001: https://www.eortc.org/app/uploads/sites/2/2018/02/SCmanual.pdf. Accessed 2023 Apr 26.

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

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

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

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

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

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

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

50.EuroQol Research Foundation. EQ-5D-3L | About [manufacturer-provided reference]. 2022; https://euroqol.org/eq-5d-instruments/eq-5d-3l-about/.

51.Sinnott PL, Joyce VR, Barnett PG. Guidebook: preference measurement in economic analysis. Menlo Park (CA): Health Economics Research Center; 2007.

52.Teckle P, Peacock S, McTaggart-Cowan H, et al. The ability of cancer-specific and generic preference-based instruments to discriminate across clinical and self-reported measures of cancer severities. Health Qual Life Outcomes. 2011;9:106. PubMed

53.Pickard AS, Neary MP, Cella D. Estimation of minimally important differences in EQ-5D utility and VAS scores in cancer. Health Qual Life Outcomes. 2007;5(1):70. PubMed

54.Clinical trial end points for the approval of cancer drugs and biologics - guidance for industry. Silver Spring (MD): U.S. Food and Drug Administration; 2018: https://www.fda.gov/media/71195/download. Accessed 2023 Apr 26.

55.Health Canada reviewer's report: Adcetris (brentuximab vedotin) [internal manufacturer' report]. In: Drug Reimbursement Review manufacturer submission: Brentuximab vedotin for injection (Adcetris), 50 mg/vial, lyophilized powder for reconstitution, intravenous infusion. Bothell (WA): Seattle Genetics, Inc.; 2019.

56.Schwartz GJ, Gauthier B. A simple estimate of glomerular filtration rate in adolescent boys. J Pediatr. 1985;106(3):522-526. PubMed

57.Lavoie Smith EM, Li L, Hutchinson RJ, et al. Measuring vincristine-induced peripheral neuropathy in children with acute lymphoblastic leukemia. Cancer Nurs. 2013;36(5):E49-60. PubMed

58.Parker C, Woods B, Eaton J, et al. Brentuximab vedotin in relapsed/refractory Hodgkin lymphoma post-autologous stem cell transplant: a cost-effectiveness analysis in Scotland. J Med Econ. 2017;20(1):8-18. PubMed

59.Large S, Hettle R, Balakumaran A, Wu E, Borse RH. Cost-effectiveness of pembrolizumab versus brentuximab vedotin for patients with relapsed or refractory classical Hodgkin's lymphoma: a United States payer perspective. J Med Econ. 2018:1-10. PubMed

60.Petrelli F, Barni S. Correlation of progression-free and post-progression survival with overall survival in advanced colorectal cancer. Ann Oncol. 2013;24(1):186-192. PubMed

61.Hotta K, Kiura K, Fujiwara Y, et al. Role of survival post-progression in phase III trials of systemic chemotherapy in advanced non-small-cell lung cancer: a systematic review. PLoS One. 2011;6(11):e26646. PubMed

62.Sundar S, Wu J, Hillaby K, Yap J, Lilford R. A systematic review evaluating the relationship between progression free survival and post progression survival in advanced ovarian cancer. Gynecol Oncol. 2012;125(2):493-499. PubMed

Appendix 1: Ann Arbor Staging Classification for HL

Please note that this appendix has not been copy-edited.

Following the initial diagnosis the disease is then staged using imaging and laboratory blood tests and categorized according to the Ann Arbor stage I through IV of disease severity where higher stages indicate more widespread disease (Table 24).^9,27,28

Table 24: Ann Arbor Definitions for HL

E = extranodal contiguous extension; HL = Hodgkin lymphoma; S = involvement of the spleen.

Source: NCCN (2022),⁹ Cheson et al. (2014),²⁷ A Kaseb et al. (2022).^9,27,28

Each Ann Arbor stage is further subdivided into A and B categories. Category A denotes that no systemic symptoms are present, and category B is assigned to patients who have unexplained fevers, drenching night sweats or unexplained weight loss. In addition, patients with classical HL are usually divided in 3 groups:

• early-stage favourable (stage I to II with no unfavourable factors)

• early-stage unfavourable (stage I to II with unfavourable factors such as large mediastinal adenopathy; > 3 nodal sites of disease; B symptoms; extranodal involvement; or significantly elevated erythrocyte sedimentation rate)

• advanced-stage disease (stage III to IV).⁹

Appendix 2: Appraisal of Post-Progression Survival

Please note this appendix has not been copy-edited.

PPS, which was defined as time from progression to death and calculated for participants who experienced disease progression by subtracting PFS from OS, was used by the sponsor in the pharmacoeconomic analysis of BV + AVD as front-line therapy for advance stage classical HL submitted to the review team.¹⁶ PPS was not a prespecified end point in the ECHELON-1 trial, and data for PPS analyses were obtained via post hoc analysis based on the ECHELON-1 trial data. PPS was estimated based on parametric survival distributions fit to individual participant data on PPS from the ECHELON-1 trial. The PPS analyses were conducted for cohorts of patients from the ECHELON-1 trial who had evidence of receipt of ASCT (N = 99) and those who did not (N = 126). Within these cohorts, analyses were stratified by front-line treatment arm (i.e., BV + AVD versus ABVD). Receipt of ASCT was determined using data on subsequent therapies received from the ECHELON-1 trial data.

PPS has been adopted in several cost-effectiveness analyses in oncology, including HL.^58,59 However, no studies with respect to the relationship between PPS and OS or PFS in the patient population with hematological malignancy (e.g., HL) were identified by the review team from the body of literature, while a few studies have been identified in patients with solid tumours. For instance, between PPS and OS, a systematic review of 34 RCTs indicated that PPS might be an appropriate surrogate for OS in patients with advanced colorectal cancer: Improvements in PPS were found strongly associated with improvements in OS.⁶⁰ Similarly, in patients with advanced non–small cell lung cancer,⁶¹ a systematic review of 70 phase III trials published between 1988 and 2007 found that PPS had become increasingly associated with OS over years. Between PPS and PFS, findings from an examination of 37 RCTs published between 1990 and 2010 suggested that percentage gains in PFS among patients with advanced ovarian cancer were related to no percentage gains or slight percentage gains or losses in PPS.⁶² The referenced systematic reviews pooled data from heterogeneous published studies that had wide ranging follow-up durations, differences in progression definitions, and limited information regarding post-progression disease management to mention some of the limitations. The authors of some of the reviews identify their analyses as not definitive of a correlation between PPS and OS. Given these limitations, lacking relevant evidence in hematological cancers, and the apparent differences between hematological and solid tumours, the association between PPS and OS/PFS in patients with classical HL remains unclear.

Appendix 3: Key Protocol Amendments in the ECHELON-1 Trial

Please note this appendix has not been copy-edited.

Table 25: Summary of Key Protocol Amendments in the ECHELON-1 Trial

ABVD = doxorubicin, bleomycin, vinblastine, dacarbazine; BSA = body surface area; BV + AVD = brentuximab vedotin plus doxorubicin, vinblastine, dacarbazine; CR = complete remission; EOT = end of treatment; HL = Hodgkin lymphoma; IRF = independent review facility; mPFS = modified progression-free survival; OS = overall survival; PRO = patient-reported outcome; RT = radiation therapy; SAE = serious adverse event.

Source: Drug Reimbursement Review manufacturer submission.¹⁶

Appendix 4: Additional Data From the ECHELON-1 Trial

Please note this appendix has not been copy-edited.

Figure 13: Forest Plot of HR in OS for Subgroup Analyses (ITT Population) in the ECHELON-1 Trial

ABVD (denoted as A + AVD in the manufacturer’s Clinical Study Report) = doxorubicin-bleomycin-vinblastine-dacarbazine; BV + AVD = brentuximab vedotin plus doxorubicin, vinblastine, dacarbazine; ECOG = Eastern Cooperative Oncology Group; IPFP = International Prognostic Factor Project; ITT = intention to treat; OS = overall survival.

Sources: ECHELON-1 Clinical Study Report Addendum 1 (data cut-off: June 01, 2021)¹⁴ and the sponsor’s Summary of Clinical Evidence.¹

Figure 14: Forest Plot of HR in mPFS According to IRF for Subgroup Analyses (ITT Population) in the ECHELON-1 Trial

ABVD (denoted as A + AVD in the manufacturer’s Clinical Study Report) = doxorubicin-bleomycin-vinblastine-dacarbazine; BV + AVD = brentuximab vedotin plus doxorubicin, vinblastine, dacarbazine; ECOG = Eastern Cooperative Oncology Group; IPFP = International Prognostic Factor Project; ITT = intention to treat; mPFS = modified progression-free survival.

Source: ECHELON-1 original Clinical Study Report (data cut-off: April 20, 2017)¹³ and the sponsor’s Summary of Clinical Evidence¹.

Figure 15: Forest Plot of HR in PFS According to Investigator for Subgroup Analyses (ITT Population) in the ECHELON-1 Trial

ABVD (denoted as A + AVD in the manufacturer’s Clinical Study Report) = doxorubicin-bleomycin-vinblastine-dacarbazine; BV + AVD = brentuximab vedotin plus doxorubicin, vinblastine, dacarbazine; ECOG = Eastern Cooperative Oncology Group; IPFP = International Prognostic Factor Project; ITT = intention to treat; PFS = progression-free survival.

Source: ECHELON-1 Clinical Study Report Addendum 1 (data cut-off: June 01, 2021)¹⁴ and the sponsor’s Summary of Clinical Evidence¹.

Table 26: PFS According to IRF (ITT Population) in the ECHELON-1 Trial

ABVD = doxorubicin-bleomycin-vinblastine-dacarbazine; BV + AVD = brentuximab vedotin in combination with doxorubicin, vinblastine, dacarbazine; 95% CI = 95% confidence interval; HL = Hodgkin lymphoma; IPFP = International Prognostic Factor Project; IRF = intendent review facility; ITT = intention to treat; NE = not estimable; PFS = progression-free survival.

Note; Data cut-off date for PFS according to IRF was April 20, 2017.

^aHR and 95% CI were based on a stratified Cox’s proportional hazard regression model with stratification factors region and number of IPFP risk factors at baseline with treatment as the explanatory variable in the model. A HR of less than 1 favours the BV + AVD group.

^bMedian PFS follow-up was calculated from the Kaplan-Meier method switching the PFS according to IRF event/censored status, i.e., PFS event as censored and censored as PFS event.

Source: ECHELON-1 original Clinical Study Report (data cut-off: April 20, 2017).¹³

Pharmacoeconomic Review

Abbreviations

Executive Summary

The executive summary comprises 2 tables (Table 1 and Table 2) and a conclusion.

Table 1: Submitted for Review

AVD = doxorubicin-vinblastine-dacarbazine; HL = Hodgkin lymphoma; NA = not applicable; NOC = Notice of Compliance; ASCT = autologous stem-cell transplant.

Table 2: Summary of Economic Evaluation

ABVD = doxorubicin-bleomycin-vinblastine-dacarbazine; AE = adverse event; ASCT = autologous stem-cell transplant; AVD = doxorubicin- vinblastine-dacarbazine; AVEPC = doxorubicin-vincristine-etoposide–prednisone-cyclophosphamide; BEACOPP = bleomycin-etoposide-doxorubicin-cyclophosphamide-vincristine-procarbazine; BV = brentuximab vedotin; ICER = incremental cost-effectiveness ratio; HL = Hodgkin lymphoma; LY = life-year; PFS = progression-free survival; QALY = quality-adjusted life-year; RDI = relative dose intensity.

Conclusions

Based on the CDA-AMC Clinical Review of the ECHELON-1 trial, a phase III, open-label, randomized controlled trial, overall survival (OS), progression-free survival (PFS), and modified progression-free survival (mPFS) were consistently in favour of brentuximab vedotin (BV) in combination with doxorubicin-vinblastine-dacarbazine (AVD) compared to doxorubicin-bleomycin-vinblastine-dacarbazine (ABVD) in the intention-to-treat (ITT) population, which comprised previously untreated adult patients with advanced-stage classical Hodgkin lymphoma (HL). However, the clinical significance of the magnitude of the treatment differences is uncertain and concerns remain about the high percentages in loss to follow-up and withdrawal of patients in both OS and PFS analyses. In addition, although subgroup analyses of the OS and mPFS results signalled that BV plus AVD may be more effective in patients with stage IV classical HL than in those with stage III HL, conclusions regarding subgroup differences between stage III versus stage IV patients are uncertain because the study was not designed to detect differences between these subgroups, and patients were not stratified according to disease stage at randomization. More patients treated with BV plus AVD experienced severe adverse events (AEs) compared with those in the ABVD group. Given that all patients in the pivotal ECHELON-1 trial were required to be aged 18 years or older, the ECHELON-1 trial did not reflect results for pediatric patients. The AHOD1331 trial indicated that BV plus doxorubicin-vincristine-etoposide-prednisone-cyclophosphamide (AVEPC) provides a clinically meaningful benefit in event-free survival (EFS) compared to doxorubicin-bleomycin-vincristine-etoposide–prednisone-cyclophosphamide (ABVE-PC) in patients aged 2 years to 21 years with high-risk classical HL.

CDA-AMC undertook reanalyses to address limitations in the economic evaluation submitted by the manufacturer by removing treatment-specific utility values, adding AE disutilities, and eliminating relative dose intensity (RDI). CDA-AMC could not resolve all outstanding limitations in the analysis and relied on scenario analyses to explore the impact of some of these limitations.

In the CDA-AMC reanalysis, the incremental cost-effectiveness ratio (ICER) for BV plus AVD compared to ABVD was $115,865 per quality-adjusted life-year (QALY) gained (incremental costs: $105,110; incremental QALYs: 0.91). As the manufacturer’s model used ECHELON-1 trial data, which excluded pediatric patients, results represent the cost-effectiveness of BV plus AVD for the treatment of previously untreated adult patients with advanced-stage HL. CDA-AMC’s base case was in line with the manufacturer’s results, which estimated an ICER of $118,104 per QALY gained when comparing BV plus AVD with ABVD. The incremental benefit of BV plus AVD was driven mainly by an improvement in PFS (incremental life-years [LYs] in front-line progression-free: 1.94). A reduction in the price of BV of at least 55% would be necessary for BV plus AVD to be considered cost-effective at a willingness-to-pay (WTP) threshold of $50,000 per QALY gained. A 55% price reduction would reduce the unit price of a 50 mg vial of BV from $4,840 to $2,178, which would reduce 28-day cycle costs for BV from $19,360 to $8,712. No information was provided for the population aged less than 18 years.

A number of limitations could not be addressed given the availability of clinical information and the nature of the manufacturer’s economic model. CDA-AMC was unable to address the lack of comparative clinical information for BV plus AVD versus other relevant comparators, including bleomycin-etoposide-doxorubicin-cyclophosphamide-vincristine-procarbazine (BEACOPP) and fluorodeoxyglucose PET-adapted BEACOPP and PET-adapted ABVD; as such, the cost-effectiveness of BV plus AVD versus these comparators is unknown. CDA-AMC was also unable to estimate the cost-effectiveness of BV plus AVEPC compared with ABVE-PC. The manufacturer’s submitted model structure did not allow for the exploration of the treatment effect of BV plus AVD by disease stage. As such, the cost-effectiveness of BV plus AVD by disease stage is unknown. Finally, as patients aged less than 18 years were excluded from the ECHELON-1 trial, which was used to populate the economic model, the clinical effectiveness, and therefore the cost-effectiveness, of BV plus AVD compared with most relevant comparators for those aged less than 18 years is unknown.

Additional limitations that were explored in scenario analyses included the potential bias around investigator-assessed PFS given the open-label nature of the ECHELON-1 trial, which could have biased results in favour of BV plus AVD. If the actual treatment effect of BV plus AVD on PFS was aligned more closely with the mPFS outcome (which was assessed by an independent review facility), the ICER for BV plus AVD is expected to be higher than estimated in the CDA-AMC base case. Additionally, increasing the proportion of patients in the model undergoing an autologous stem-cell transplant (ASCT), which was deemed to be lower than what the clinical experts consulted by CDA-AMC for this review expected to see clinical practice in Canada, resulted in a slightly higher ICER. These scenario analyses indicate that the CDA-AMC base case may overestimate the benefit associated with BV plus AVD and therefore likely represents a lower limit on what the ICER may be in actual practice.

Input Relevant to the Economic Review

This section is a summary of the feedback received from the patient groups, registered clinicians, and drug plans that participated in the CDA-AMC review process.

One patient group, Lymphoma Canada, provided input for this review. Patient input was received from an online survey focused on patients with HL. Twenty-six responses were received, with 5 responders being from Canada. Among the patients who answered the survey, 3 indicated they were treated with ABVD, 2 with other forms of chemotherapy, and 1 with R-CHOP regimen, which consists of cyclophosphamide, doxorubicin hydrochloride (hydroxydaunomycin), and vincristine sulphate. These patients reported being satisfied with the treatment provided to them. Patient feedback emphasized that controlling disease symptoms, maintaining disease remission, increasing length of life, and improving quality of life were the most important factors to be considered in a novel treatment. Three patients had experience with BV plus AVD. Two of these patients reported to be still in remission, and 1 did not report remission status. Patients reported the most common AEs during their experience with BV to be fatigue, neutropenia, constipation, and joint or muscle pain, and 2 patients referred to the AEs as manageable. Two patients rated their experience with BV plus AVD as very good, while 1 patient reported having had a poor experience.

Two clinician groups provided input for this review: the Ontario Health (Cancer Care Ontario) Hematology Cancer Drug Advisory Committee and Pediatric Oncology Group of Ontario. Clinician input included ABVD or BEACOPP as current first-line treatments for patients with stage III and IV HL. In the pediatric setting, ABVE-PC and vincristine-etoposide-prednisone-doxorubicin, followed by cyclophosphamide-vincristine-prednisone-dacarbazine (OEPA-COPDAC) regimens are the 2 treatments used in practice. In the pediatric setting, the chemotherapy regimen AVEPC is the current backbone used in practice with BV. Clinician input emphasized that treatment goals are to improve outcomes with first-line therapy to avoid the need for second-line therapy. For the pediatric setting, clinician input indicated that the main treatment goal is to avoid recurrence and consequently to minimize potential late effects which impact quality of life, add to health care utilization costs in survivors. Additionally, patients experiencing treatment failure or relapse may require additional cytotoxic therapy and ASCT.

Drug plan input indicated concerns with missing relevant comparators such as the BEACOPP regimen and PET-adapted regimens; according to the current provisional funding algorithm, patients who relapse would be eligible for BV re-treatment if a relapse occurs more than 12 months after completion of prior BV therapy with at least 6 months of response. The drug plans asked whether BV would be funded for patients who are aged less than 18 years, and those with an Eastern Cooperative Oncology Group Performance Status greater than 2, with nodular lymphocyte-predominant HL (no CD30 expression), with stage IIB disease at high risk, or with central nervous system involvement and symptoms of progressive multifocal leukoencephalopathy. Drug plan input indicated that primary prophylaxis with granulocyte colony-stimulating factor (G-CSF) is typically prescribed with BV plus AVD, and is associated with an additional cost.

Several of these concerns were addressed in the manufacturer’s model:

• The manufacturer’s submitted model accounted for quality of life and length of life by using QALYs as the primary outcome.

• The cost of G-CSF prophylaxis was included.

CDA-AMC was unable to address the following concerns raised from stakeholder input:

• BEACOPP was not included as a relevant comparator in the analysis.

• The pediatric population was not included in the ECHELON-1 study, and relevant comparators were not included in the analysis.

Economic Review

The current review is for BV (Adcetris) for previously untreated patients with advanced-stage HL and considers 2 reimbursement requests: BV in combination with ABVD for the treatment of previously untreated patients with advanced-stage HL, and BV in combination with AVEPC in previously untreated high-risk HL in the pediatric population.¹ This is a tumour group submission sponsored by the BC Cancer Agency and Pediatric Oncology Group of Ontario. However, the review is based on a previous manufacturer-initiated submission (Project Number PC0311 to 000) that was withdrawn. As such, the cost-effectiveness evidence is based on the evidence submitted by the manufacturer for Project Number PC0311 to 000.

Economic Evaluation

Summary of Manufacturer’s Economic Evaluation

Overview

The manufacturer submitted a cost-utility analysis assessing BV plus AVD compared with ABVD in patients with previously untreated patients with advanced-stage HL.¹ The modelled population was aligned with the patient population in the ECHELON-1 trial, which included adults with untreated advanced-stage HL.²

The recommended dose of BV is 1.2 mg/kg of body weight, given intravenously, on days 1 and 15 of each 28-day cycle for up to 6 cycles, according to the product monograph.³ BV is indicated in combination with 25 mg/m² of doxorubicin, 6 mg/m² of vinblastine, and 375 mg/m² of dacarbazine. The manufacturer-submitted price for BV is $4,840 per 50 mg vial. Assuming wastage, as well as a mean body weight (75 kg) and body surface area (1.88 m²) based on the ECHELON-1 trial, the cost of BV in the adult population is estimated to be $19,360 over a 28-day cycle and $116,160 over 6 cycles in the manufacturer’s model. The BV plus AVD regimen is estimated to cost $21,584 over a 28-day cycle and $129,506 over 6 cycles. The comparator regimen, ABVD, consisted of 25 mg/m² of doxorubicin, 10 units/m² of bleomycin, 6 mg/m² of vinblastine, and 375 mg/m² of dacarbazine on days 1 and 15 of each 28-day cycle for up to 6 cycles. The ABVD regimen is estimated to cost $3,902 over a 28-day cycle and $23,412 over 6 cycles.

The clinical outcome of interest was QALYs and LYs. The economic analysis was undertaken using a lifetime time horizon (60 years) from the perspective of a public health care payer. Discounting of 1.5% per year was applied to both costs and outcomes.

Model Structure

A semi-Markov model was developed in Microsoft Excel. The model consisted of 4 health states defined as front-line progression-free, post-progression and not receiving an ASCT, post-progression and receipt of an ASCT, and death (Figure 1). All patients entered the model in the front-line (i.e., receipt of their first-ever treatment) progression-free health state and received either BV plus AVD or ABVD. Patients in the front-line progression-free health state were at risk of disease progression or death. Upon progression, patients could go on to receive ASCT or no ASCT. Patients who failed front-line therapy and did not receive ASCT remained in a post-progression, no-ASCT state until death and incurred costs of subsequent therapies, including programmed cell death protein 1 (PD-1) inhibitors, such as nivolumab and pembrolizumab. Patients receiving ASCT after front-line failure were at risk of ASCT failure. Patients in whom ASCT failed less than 9 months after completion of front-line treatment were eligible to receive post-ASCT consolidation with BV, and incurred BV treatment costs. The proportion of patients who received BV consolidation was dependent on front-line treatment. In addition, a proportion of patients in whom ASCT failed incurred costs of subsequent therapies, which included nivolumab and pembrolizumab, and/or BV salvage therapy. However, patients who had received BV consolidation were assumed to be ineligible for BV salvage after ASCT failure. A proportion of patients progressing after ASCT were also assumed to undergo repeat ASCT, or an allogenic stem-cell transplant.

Model Inputs

Baseline characteristics (proportion of males: 58%; mean age: 39.5 years) and the probability of progression following front-line treatment were derived from the ECHELON-1 trial, an open-label, multicentre, randomized phase III trial.² Investigator-assessed PFS and post-progression survival (PPS) of the ITT population obtained from the ECHELON-1 trial were used to inform the base-case analysis. Patient progression from the front-line progression-free health state to the post-progression states or death was based on extrapolation curves fit to PFS data from the ECHELON-1 trial. Due to an observation that PFS curves in the ECHELON-1 trial plateaued, the manufacturer used a mixture cure model. The probability that the progression event, defined according to PFS, was death, was based on the ECHELON-1 trial and was treatment-specific. Based on the best-fitting mixture cure model, different front-line therapy cure proportions were observed for BV plus AVD and ABVD (proportion cured from front-line therapy at 8 years: 82.2% versus 74.7%, respectively). All extrapolated survival curves used in the model were selected based on the best fit according to the Akaike information criterion, Bayesian information criterion, visual inspection, and clinical plausibility. After 8 years, patients who remained in the front-line progression-free health state were assumed to be cured and no longer transitioned to post-progression health states.¹ At this time, patients remaining in the front-line progression-free health state could progress to death based on age- and gender-specific general population mortality, weighted to the gender distribution of patients in ECHELON-1 trial,⁴ and adjusted for non-HL–related excess mortality obtained from the literature.⁵ Transitions to death from the post-progression health states were based on PPS estimates derived from the ITT population of the ECHELON-1 trial who experienced disease progression and stratified by patients who had or did not have evidence of receipt of ASCT post-progression, as well as the general population mortality from Canadian life tables,⁴ adjusted for non-HL–related excess mortality.⁵ Survival distributions were fit separately for PPS with or without ASCT, independent of front-line therapy received. For the PPS of patients who had evidence of receipt of ASCT, the Gompertz distribution was chosen as it implied a cure fraction of approximately 83%. For the PPS of patients without evidence of receipt of ASCT, the log-normal distribution was chosen based on statistical fit and clinical plausibility.

The proportion of patients eligible for ASCT (44%) was based on data from the ECHELON-1 trial.² Among patients who failed therapy with ABVD, 68% were assumed to received BV consolidation therapy based on a previous CDA-AMC submission.⁶ The proportion of patients receiving BV consolidation after ASCT for who failed front-line therapy with BV plus AVD (33%) was based on clinician input. The manufacturer assumed that all patients who did not receive BV consolidation therapy and who progressed after ASCT would receive BV post–transplant failure. In addition, 75% of these patients were assumed to receive a PD-1 inhibitor (pembrolizumab or nivolumab) after BV post–transplant failure, based on a pivotal phase II trial.⁷ It was assumed that all patients who progress after receiving ASCT with BV consolidation would receive a PD-1 inhibitor as salvage therapy. The proportion of patients receiving G-CSF prophylaxis was obtained from clinician input and a previous CDA-AMC submission.⁶ Several other probabilities related to the likelihood of receiving an allogenic stem-cell transplant after ASCT failure, and receiving a repeat ASCT after initial ASCT failure were identified from the published literature and clinician input. Among patients who were not eligible for ASCT, the manufacturer assumed 50% would receive nivolumab and 50% pembrolizumab as subsequent therapies.

Probabilities of AEs were based on frequencies observed in the ECHELON-1 trial.² AEs considered in the model included serious neutropenic events, pulmonary toxicity, peripheral neuropathy, and anemia.

Health-state utility values for patients in the front-line PFS health state were derived from the ECHELON-1 trial using the EQ-5D-3L instrument and values sets from the UK.¹ These values were specific to the treatment regimen and stratified as to whether the patient was on or off treatment. The manufacturer also accounted for declining utilities with age, by setting utilities to the minimum of the state–specific or age- and sex-matched general population norm utility values obtained from literature using the Health Utilities Index Mark 3 questionnaire.⁸ Treatment-related AE disutilities were assumed to be included in these measures and were not explicitly modelled in the base case. Health-state utility values related to the post-progression ASCT state, which included separate values for the first 3 months and subsequent months following ASCT, were identified from the literature.^9,10 The utility for post-progression and not receiving ASCT was obtained from literature.¹⁰