Summary Basis of Decision for Bavencio

 

Clinical Pharmacology

The programmed death ligand 1 (PD-L1) may be expressed on tumour cells and/or tumour-infiltrating immune cells and can contribute to the inhibition of the antitumour immune response in the tumour microenvironment. Binding of PD-L1 to the programmed death 1 (PD-1) and B7.1 receptors found on T cells and antigen-presenting cells suppresses cytotoxic T-cell activity, T-cell proliferation and cytokine production.

Avelumab, the medicinal ingredient in Bavencio, is a fully human immunoglobulin G1 (IgG1) monoclonal antibody directed against PD-L1. Avelumab binds PD-L1 and blocks the interaction between PD-L1 and the PD-1 and B7.1 receptors. This removes the suppressive effects of PD-L1 on cytotoxic CD8+ T cells, resulting in the restoration of antitumour T-cell responses. In syngeneic mouse tumour models, blocking PD-L1 activity resulted in decreased tumour growth.

As a fully human IgG1, avelumab retains Fcγ receptor binding and has been shown to induce natural killer (NK) cell-mediated direct tumour cell lysis via antibody-dependent cell-mediated cytotoxicity (ADCC) in vitro.

A two-compartment population pharmacokinetic model was developed to describe avelumab concentration as a function of time, based on data from studies EMR100070-001, EMR100070-002 and EMR100070-003, representing a total of 1,629 subjects. The structural model was simple as it featured only a linear elimination from the central compartment.

The estimation of the terminal half-life in the population pharmacokinetic model differed by a factor of 1.5 from that in the non-compartmental analysis. The population pharmacokinetic analysis, when considering the entire therapeutic range, resulted in an estimate of 6 days, whereas after a single dose, the non-compartmental analysis estimate was 4 days. The ratio between the two estimates (6/4 = 1.5) was within the range of the ratios (1.6-1.8) between the minimum observed drug concentration (C_min) values at the start and the end of the therapeutic cycle. Since the population pharmacokinetic analysis does not correct for accumulation of avelumab in C_min, the half-life from this analysis might represent an artifact of the dosing regimen rather than an independent estimate. Both half-life estimates were included in the Bavencio Product Monograph.

The covariate of body weight had a positive correlation with the total systemic clearance (CL) and central volume of distribution (V1) parameters in the population pharmacokinetic model.

For further details, please refer to the Bavencio Product Monograph, approved by Health Canada and available through the Drug Product Database.

Clinical Efficacy

The clinical efficacy of Bavencio was evaluated in Part A of an ongoing pivotal Phase II, open-label, single-arm, multicenter trial, Study EMR100070-003 (Study 003 Part A), conducted in 88 patients with histologically confirmed metastatic Merkel cell carcinoma whose disease had progressed after at least one chemotherapy treatment for distant metastatic disease. The study excluded patients with: autoimmune disease; medical conditions requiring systemic immunosuppression; prior organ or allogeneic stem cell transplantation; prior treatment with anti-PD-1, anti-PD-L1 or anti-CTLA-4 antibodies; central nervous system metastases; infection with human immunodeficiency virus (HIV), hepatitis B virus or hepatitis C virus; or an Eastern Cooperative Oncology Group (ECOG) performance score ≥2. Patients received Bavencio at a dose of 10 mg/kg via intravenous infusion every two weeks until disease progression or unacceptable toxicity.

Of the 88 patients, 73.9% were male, the median age was 72.5 years (range, 33 years to 88 years), 92% were Caucasian, and 55.7% and 44.3% had an ECOG performance status of 0 and 1, respectively. Thirty-five percent (35%) of patients had received two or more prior therapies, and 53% of patients had visceral metastases. All patients had tumour samples evaluated for PD-L1 expression. Of the evaluated tumour samples, 66% were PD-L1 positive (≥1% of tumour cells), 18% were PD-L1 negative, and 16% had non-evaluable results by an investigational immunohistochemistry assay. Archival tumour samples from 77 patients were evaluated for Merkel cell polyomavirus using an investigational assay. Evidence of Merkel cell polyomavirus was found in 52% of the evaluated samples.

The primary efficacy endpoint of the pivotal Study EMR100070-003 was the confirmed best overall response according to the Response Evaluation Criteria in Solid Tumours (RECIST) (version 1.1), as determined by an Independent Endpoint Review Committee. The key secondary efficacy endpoint was duration of response. Progression-free survival and overall survival were also included in the efficacy assessment. The minimal follow-up period for the efficacy analyses was 12 months.

Best overall response was confirmed in 29 patients (33.0%, 95% confidence interval [CI]: 23.3, 43.8), and consisted of 10 complete responses (11.4%) and 19 partial responses (21.6%). Among the 29 patients with objective responses, median duration of response was not reached (range, 2.8 to 23.3+ months) at the time of data cut-off. Durable response (defined as 6 months or longer) was achieved in 25 patients (28.4%). A total of 21 patients remained in response at their last tumor assessment. Among those, 13 patients had a duration of response of at least 12 months.

The response to Bavencio treatment was observed in patients with tumour PD-L1 positive (≥1% tumour cells) or negative expression status, or with tumour positive or negative status for Merkel cell polyomavirus. However, both the PD-L1 and Merkel cell polyomavirus status of the tumours could not be considered reliable biomarkers for predicting treatment response because of small sample sizes and a high response variety, in addition to the experimental nature of the assay used for evaluating PD-L1 expression. The response to Bavencio treatment was also observed in patients with visceral metastasis or large tumour burden.

Median progression-free survival was 2.7 months (range, 0.03 to 24.5 months), (95% CI: 1.4, 6.9) and median overall survival was 12.9 months (range, 0.4 to 24.7 months), (95% CI: 7.5, not estimable). The results obtained for progression-free survival and overall survival have to be interpreted with caution and are considered of limited value due to the small number of patients enrolled in a study with a single-arm design.

Data were also submitted from the supportive Study 100070 Obs001 (Study Obs001), which was a retrospective, chart review and registry-based study to collect the treatment outcomes in patients with Merkel cell carcinoma treated with chemotherapy. Study Obs001 was intended to serve as a historical reference for the pivotal Study 003 Part A. A total of 686 patients with Merkel cell carcinoma were identified over a 10-year period in a United States health records database; 120 of whom were suspected to have metastatic disease. Thirty-nine (39) of the 120 patients had second-line chemotherapy. Among those, 14 patients were identified as similar to the patient population in Study 003 Part A and were included in the analysis. The overall response rate in these 14 patients was 28.6% (95% CI: 8.4, 58.1), with a median duration of response of 1.7 months (95% CI: 0.5, 3.0).

The efficacy data submitted from an early interim analysis of 16 treatment-naïve patients with metastatic Merkel cell carcinoma from the ongoing Study EMR100070-003 Part B were considered premature for an assessment. Consequently, Health Canada could not consider an indication of Bavencio for previously untreated patients with metastatic Merkel cell carcinoma.

In order to verify the clinical benefit of Bavencio, further follow-up is required, and further evaluation will take place upon the submission of the final report of the ongoing confirmatory trial, Study EMR100070-003 Part B (see What follow-up measures will the company take?).

Indication

The New Drug Submission for Bavencio was filed by the sponsor with the following indication:

• Bavencio (avelumab) is indicated for the treatment of adult patients with metastatic Merkel Cell Carcinoma (MCC).

Health Canada recommended that the sponsor revise the indication to adequately reflect the characteristics of the patients enrolled in the pivotal study for this submission. A caveat statement was also added to the indication, providing details regarding the limitations of the currently available data. Accordingly, Health Canada approved the following indication:

• Bavencio (avelumab) is indicated for the treatment of metastatic Merkel Cell Carcinoma (MCC) in previously treated adults.
• Marketing authorization with conditions was based on tumour response and durability of response.
• An improvement in survival or disease-related symptoms has not yet been established.

For more information, refer to the Bavencio Product Monograph, approved by Health Canada and available through the Drug Product Database.

Clinical Safety

The clinical safety of Bavencio was characterized in 88 patients with metastatic Merkel cell carcinoma, who participated in Part A of the pivotal EMR100070-003 study (Study 003 Part A) (described in the Clinical Efficacy section). Safety data were also derived from 1,650 patients with other solid tumours in an ongoing Study EMR 100070-001 (Study 001). Therefore, the safety database included pooled data from a total of 1,738 patients treated with the recommended therapeutic dose of Bavencio 10 mg/kg as an intravenous infusion every two weeks. In this population, 24% of patients were exposed to Bavencio for 6 months or longer, and 7% for 12 months or longer. Additionally, data from a small Phase I study, EMR100070-002 (Study 002) conducted in 17 Japanese patients with solid tumours were submitted and included in the safety assessment.

The most common (≥15%) adverse events in patients treated with Bavencio were fatigue, diarrhea, nausea, constipation, and decreased appetite. The most common adverse events and the severity of the adverse events were found to be consistent with the studied patient population.

A total of 911 patients died among the 1,783 patients (52.4%). Fifty percent (50%) of patients died in Study 003 Part A and 52.5% in Study 001. The majority of deaths (744 or 42.8%) were due to disease progression. Sixty-three patients (3.6%) died due to adverse events.

The primary safety risks associated with Bavencio were immune-mediated adverse drug reactions and infusion reactions.

In Study 003 Part A and Study 001, immune-mediated adverse drug reactions were reported in 17% and 14% of patients, and ≥Grade 3 immune-mediated adverse drug reactions occurred in 3.4% and 2.2% of patients, respectively.

Infusion reactions occurred in 21.6% and 25.5% of patients, and ≥Grade 3 infusion reactions were reported in 0% and 0.7% of patients in Study 003 Part A and Study 001, respectively.

The most common (≥1%) immune-mediated adverse drug reactions were hypothyroidism, pruritus, rash, and maculo-papular rash. No fatal case or Grade 4 immune-mediated adverse drug reactions were reported in Study 003 Part A. In Study 001, there were six immune-mediated fatal cases (autoimmune hepatitis, hepatic failure, pneumonitis, myocarditis, acute respiratory distress/failure) and two Grade 4 immune-mediated adverse drug reactions (autoimmune disorders and pneumonitis). Two patients (2.2%) in Study 003 Part A and 25 patients (1.7%) in Study 001 experienced serious immune-mediated adverse drug reactions including pneumonitis, autoimmune hepatitis, hypothyroidism, adrenal insufficiency, and autoimmune disorder. No additional safety issues were identified in Study 002. Overall, the incidence, severity and types of immune-mediated adverse drug reactions in Study 003 Part A, Study 001 and the pooled analysis are consistent with the safety profiles of other monoclonal antibody immune checkpoint inhibitors. Most immune-mediated adverse drug reactions were reversible and managed with dose interruption, administration of corticosteroids, and supportive care.

As with all therapeutic proteins, Bavencio has the potential for immunogenicity. Of the 1,738 patients treated with Bavencio at the recommended dose, 1,558 patients were evaluable for treatment-emergent anti-drug antibodies and 64 (4.1%) tested positive. Neutralizing antibodies were not tested.

Appropriate warnings and precautions are in place in the approved Bavencio Product Monograph to address the identified safety concerns.

For more information, refer to the Bavencio Product Monograph, approved by Health Canada and available through the Drug Product Database.