CADTH Reimbursement Review

Olaparib (Lynparza)

Sponsor: AstraZeneca Canada Inc.

Therapeutic area: gBRCAm, HER2-negative, high-risk early breast cancer

This multi-part report includes:

Clinical Review

Pharmacoeconomic Review

Stakeholder Input

Clinical Review

Executive Summary

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

Table 1: Submitted for Review

BRCA = breast cancer susceptibility gene; gBRCA = germline BRCA; HER2 = human epidermal growth factor receptor 2; NOC = Notice of Compliance; NOC/c = Notice of Compliance with conditions.

Introduction

Breast cancer is the most commonly diagnosed cancer among women in Canada and the second most common cancer in men and women combined.¹ In 2020, 27,700 women were diagnosed with breast cancer, representing about 25% of new cancer cases in Canada.¹ Breast cancer is the second leading cause of cancer deaths among women, accounting for 14% of all cancer deaths.¹ The 5-year net survival for breast cancer is more than 85% among women diagnosed before 85 years of age, after which it drops to about 73%.¹ In men, the incidence of breast cancer is less than 1% per year, with 260 new cases diagnosed in 2021 in Canada.¹ BRCA1 and BRCA2 are human breast cancer susceptibility genes (BRCAs) that produce proteins responsible for repairing damaged DNA and play an important role in maintaining the genetic stability of cells.^2,3 A mutation in 1 or both BRCA genes reduces gene expression, which can lead to uncontrolled cell growth and is associated with an increased risk of cancer, including breast cancer.^4-6

Hereditary, deleterious mutations account for 5% to 10% of all breast cancers, and 60% to 68% of these hereditary cancers occur in individuals with a germline BRCA (gBRCA) mutation (gBRCAm).^4,7 In women harbouring a BRCA1 gene mutation, the estimated lifetime risk of developing breast cancer by the age of 80 years is 65% to 80%, and the 10-year actuarial risk of developing contralateral breast cancer is 25% to 31%.^8,9 The estimated lifetime risk of developing breast cancer is approximately 76% among women with a BRCA2 mutation, while among men with BRCA1 or BRCA2 mutations, it ranges from 3% to 8%.^8-10 BRCA mutations occur in women with all subtypes of breast cancer, but more commonly in those with early onset or a family history.⁹ Approximately 75% of patients with breast cancer who have a mutation in the BRCA1 gene are classified as having triple-negative breast cancer (TNBC).^11-13 In contrast, patients with breast cancer carrying mutations in the BRCA2 gene are more likely to be positive for expression of a hormone receptor (HR-positive), and only approximately 20% have TNBC.¹³

Olaparib is a selective inhibitor of human poly-(ADP-ribose) polymerase (PARP) enzymes (PARP1, PARP2, and PARP3) involved in normal cellular functions, such as DNA transcription and DNA repair. Olaparib is approved by Health Canada for the adjuvant treatment of adult patients with a deleterious or suspected deleterious gBRCAm who have high-risk early breast cancer that is negative for human epidermal growth factor receptor 2 (HER2) and that has been treated with neoadjuvant or adjuvant chemotherapy.¹⁴ The recommended total daily dose of olaparib is 600 mg taken as 2 150 mg tablets twice daily for a total of 1 year or until disease recurrence or unacceptable toxicity, whichever occurs first. Olaparib is available as a 150 mg or 100 mg tablet.

Olaparib has been previously reviewed by CADTH for other indications, including as monotherapy for the maintenance treatment of adult patients with:

• newly diagnosed advanced BRCA-mutated high-grade epithelial ovarian, fallopian tube, or primary peritoneal cancer who are in response (complete response or partial response) to first-line platinum-based chemotherapy, until disease progression or up to 2 years if no evidence of disease

• platinum-sensitive relapsed BRCA-mutated epithelial ovarian, fallopian tube, or primary peritoneal cancer who are in response to platinum-based chemotherapy.^15,16

The objective of this report is to perform a systematic review of the beneficial and harmful effects of olaparib 300 mg twice daily for the adjuvant treatment of adult patients with deleterious or suspected deleterious gBRCA-mutated, HER2-negative, high-risk early breast cancer who have been treated with neoadjuvant or adjuvant chemotherapy.

Stakeholder Perspectives

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

Patient Input

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

Two patient groups, the Canadian Breast Cancer Network (CBCN) and Rethink Breast Cancer (Rethink) provided input for this review. CBCN is a national health charity aiming to ensure the best quality of care for all Canadians affected by breast cancer. The CBCN patient input was based on an online survey of 6 patients with gBRCA-mutated early breast cancer and a literature review of current studies and grey literature. Rethink is a Canadian charity with a focus on improving the experience and outcomes of patients with breast cancer. Rethink gathered information for this review from general observations and insights through various ongoing initiatives (including stories shared by patients, virtual support groups, working groups, and patient advisory boards), in-depth telephone interviews with 3 patients with a BRCA-mutated breast cancer who participated in the OlympiA study, as well as responses from people in the Rethink Instagram community with high-risk early breast cancer.

According to the patient input received, BRCA-mutated breast cancer is more likely to be detected in people at a young age. These young patients can face several age-specific issues such as fertility or family-planning challenges, diagnosis during pregnancy, childcare, impact on relationships, body image, dating and sexuality, feeling isolated from peers who do not have cancer, career hiatuses, and financial insecurity. The main factors influencing patients’ decisions about currently available treatment options included effectiveness of the treatment (i.e., how well the treatment could help stabilize disease and delay recurrence), prolonging life without sacrificing quality of life (i.e., how well the treatment could help the patient maintain a productive, active life with minimal disruption to daily routines), risk of side effects, and the cost and accessibility of treatments. The Rethink input revealed that patient respondents, especially those with stage III breast cancer, tended to endure side effects as well as their impacts on quality of life to achieve satisfactory effectiveness. In terms of experience with olaparib, none of the 3 patients from the Rethink group who had participated in the OlympiA study and did not experience a recurrence mentioned unendurable side effects.

Clinician Input

Input From Clinical Experts Consulted by CADTH

The clinical experts consulted by CADTH noted that current systemic treatment of patients with early breast cancer is based on the receptor status and pathological findings and does not consider the patient’s BRCA mutation status, as there is no treatment specific for mutation status. Therefore, there is a need for new therapeutic options to improve survival outcomes and increase the overall cure rate in this subgroup of patients. The clinical experts noted that patients who meet the inclusion criteria outlined in the OlympiA trial will be best suited for treatment with olaparib. According to the clinical experts, by improving survival and reducing disease recurrence, the patients who are cured will have a higher quality of life and longer life. The goal of treating BRCA-mutated early breast cancer is to eradicate disease and prevent metastatic spread, resulting in cure. It was further noted by the clinical experts that it remains unclear how to integrate olaparib within the current treatment paradigm with other drugs, such as capecitabine, pembrolizumab, abemaciclib, or a combination of drugs in the treatment of early breast cancer. The clinical experts consulted mentioned that companion diagnostic testing is perceived as a barrier, given that not all patients qualify for genetic testing based on provincial guidelines (i.e., patients who have a low likelihood of hereditary syndromes). The clinical experts indicated that current genetic testing guidelines vary by province, and BRCA mutations are underdiagnosed based on most provincial testing criteria. According to the clinical experts, toxicities and disease recurrence will be the main factors to consider when deciding to discontinue treatment with olaparib.

Clinician Group Input

The clinician group input was obtained from 2 clinician groups, including the Ontario Health (Cancer Care Ontario) (OH-CCO) Breast Cancer Drug Advisory Committee (1 clinician provided input) and a group of medical oncologists across Canada (4 clinicians provided input). Both clinician groups identified that the important goal of treatment for early breast cancer, including gBRCA-mutated early breast cancer, is to decrease the recurrence of cancer and improve survival. One potential barrier, which was mentioned by the OH-CCO Breast Cancer Drug Advisory Committee, is that the current guidelines for BRCA mutation testing are restrictive in terms of eligibility criteria; thus, many patients who carry a BRCA mutation may not receive the testing and thereby lose the opportunity to receive olaparib. Both clinician groups noted the reasons that may lead to the discontinuation of olaparib are recurrence or progression of disease, intolerant toxicity or severe side effects, and patient or physician preference.

Drug Program Input

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

• considerations for the initiation of therapy

• considerations for the discontinuation of therapy

• considerations for prescribing therapy

• care provision issues

• generalizability.

The clinical experts consulted by CADTH provided advice on the potential implementation issues raised by the drug programs.

Clinical Evidence

Pivotal Studies and Protocol Selected Studies

Description of Studies

The OlympiA trial is an ongoing, phase III, randomized, multicentre, double-blind, placebo-controlled trial. The primary objective of the trial was to assess the efficacy and safety of olaparib versus placebo for the adjuvant treatment of patients with deleterious or suspicious deleterious germline BRCA1 or BRCA2 gene mutations, high-risk, HER2-negative early-stage breast cancer who had completed definitive local treatment and neoadjuvant or adjuvant chemotherapy. A total of 1,836 patients with breast cancer and gBRCA mutations were enrolled across 546 sites in 23 countries in North America (34 patients from Canada), South America, Europe, Asia Pacific, and South Africa. The primary efficacy end point was invasive disease–free survival (IDFS), and the key secondary efficacy end points were overall survival (OS) and distant disease–free survival (DDFS). Patient-reported outcomes (PROs) were assessed using the Functional Assessment of Chronic Illness Therapy–Fatigue (FACIT-F) questionnaire and the European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire Core 30 (EORTC QLQ-C30). Treatment with olaparib was given for up to 12 months, or until disease recurrence or unacceptable toxicity, whichever occurred first.

Overall, baseline characteristics were well balanced between treatment groups in the OlympiA trial. The mean age of patients was 43.3 years (standard deviation [SD] = 9.97 years), and about 68.7% of patients were between 30 and 49 years of age. Most patients were female (99.7%), premenopausal (61.3%), with an Eastern Cooperative Oncology Group (ECOG) Performance Status of 0 (88.7%), white (66.7%), and nonhispanic or Latino (88.1%). A total of 26.4% of patients received prior platinum therapy, and half of patients (50.1%) received neoadjuvant treatment. A total of 82.3% of the patients had TNBC, while 17.7% had HR-positive, HER2-negative breast cancer. Germline BRCA1 deleterious or suspected deleterious mutations were identified in 72.2% of patients, germline BRCA2 mutations in 27.1% of patients, and both germline BRCA1 and BRCA2 mutations in 0.4% of patients. The majority of patients with TNBC (60.3%) had a mutation in BRCA1, while the majority of patients with HR-positive, HER2-negative breast cancer (51.4%) had a mutation in BRCA2. A total of 36.1% of patients had clinical American Joint Committee on Cancer (AJCC) stage IIA, 21.0% had AJCC stage IIB, and 13.0% had AJCC stage IIIA.

Efficacy Results

Table 2 and Table 3 present a summary of key results from the OlympiA trial.

Overall Survival

At interim analysis 1 (March 27, 2020), the OS data were 7.9% mature. Deaths were reported in 59 patients (6.4%) in the olaparib group and 86 patients (9.4%) in the placebo group. The median OS was not estimable in either treatment arm and the stratified hazard ratio (HR) was 0.68 (99% confidence interval [CI], 0.44 to 1.05; P = 0.0236). At interim analysis 2 (July 12, 2021), the OS data were 10.0% mature. In the full analysis set (FAS), deaths were reported in 75 patients (8.1%) in the olaparib group and 109 patients (11.9%) in the placebo group. The median OS was not estimable, and the stratified HR was 0.68 (98.5% CI, 0.47 to 0.97; P = 0.0091) in favour of the olaparib group. The proportion of patients who remained alive at 4 years was 89.8% (95% CI, 87.2% to 91.9%) in the olaparib group and 86.4% (95% CI, 83.6% to 88.7%) in the placebo group (difference = 3.4%; 95% CI, −0.1% to 6.8%).¹⁷ The results of prespecified sensitivity and subgroup analyses were consistent with the primary analysis.

Table 2: Summary of Key Results From Pivotal Study

AE = adverse event; CI = confidence interval; CTCAE = Common Terminology Criteria for Adverse Events; DDFS = distant disease–free survival; FACIT-F = Functional Assessment of Chronic Illness Therapy–Fatigue; FAS = full analysis set; HER2 = human epidermal growth factor receptor 2; HR = hazard ratio; HR-positive = positive for expression of a hormone receptor; IDFS = invasive disease–free survival; KM = Kaplan-Meier; LS = least squares; MMRM = mixed-model for repeated measures; OS = overall survival; PRO = patient-reported outcome; SAE = serious adverse event; TRAE = treatment-related adverse event.

^aData cut-off date: March 27, 2020.

^bEstimate of the treatment HR was based on the stratified Cox proportional hazards model. The stratification factors were the same as those used in the stratified log-rank test. The CI for the HR was estimated using the profile likelihood approach.

^cInferential, according to the alpha spending rules for the interim analysis.

^dP value from a stratified log-rank test. Stratification was by chemotherapy type (adjuvant vs. neoadjuvant), hormone receptor status (HR-positive and HER2-negative vs. triple-negative breast cancer), and prior platinum therapy (yes vs. no). Stratification factors were based on the categories used in the randomization system and were chosen by the pooling strategy. Once the pooling strategy was applied, only the hormone receptor status stratification factor was selected.

^eMedian clinical follow-up was calculated using the reverse censoring method.

^fPatients who have not had a recorded event at the time of the analysis will be censored at the date of their last disease evaluation.

^gData cut-off date: July 12, 2021.

^hOnly patients with an evaluable baseline form were included.

^IAdjusted LS mean changes, P values (2-sided), and 95% CIs were obtained from an MMRM analysis of the change from baseline. The model included treatment, time and treatment by time interaction, corresponding baseline score, and the baseline score by time interaction. The difference was the values for olaparib minus placebo.

^jP value was not adjusted for multiple comparisons.

Sources: Clinical Study Reports for OlympiA.^18,19

Table 3: Change From Baseline for EORTC QLQ-C30 Subscale Scores — PRO

CI = confidence interval; EORTC QLQ-30 = European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire Core 30; LS = least squares; MMRM = mixed-model for repeated measures; PRO = patient-reported outcome; QoL = quality of life; SD = standard deviation.

Note: Only patients with an evaluable baseline form were included.

Adjusted LS mean changes, P values (2-sided), and 95% CIs were obtained from an MMRM analysis of the change from baseline. The model included treatment, time and treatment by time interaction, corresponding baseline score, and the baseline score by time interaction. The difference was the values for olaparib minus placebo.

^aP value was not adjusted for multiplicity.

Source: Clinical Study Report for OlympiA.¹⁸

Invasive Disease–Free Survival

At interim analysis 1 (March 27, 2020), 106 patients (11.5%) in the olaparib group and 178 patients (19.5%) in the placebo group had an IDFS event. The median IDFS was not estimable in either treatment arm, and the stratified HR for invasive disease recurrence or death was 0.58 (99.5% CI, 0.41 to 0.82; P = 0.0000073) in favour of the olaparib group. At interim analysis 2 (July 12, 2021), the stratified HR for invasive disease recurrence or death was 0.63 (95% CI, 0.50 to 0.78). The proportion of patients who remained invasive disease–free at 4 years was 82.7% (95% CI, 79.6% to 85.4%) in the olaparib group and 75.4% (95% CI, 72.2% to 78.3%) in the placebo group (difference = 7.3%; 95% CI, 3.0% to 11.5%).¹⁷ The results of prespecified sensitivity and subgroup analyses were consistent with the primary analysis.

Distant Disease–Free Survival

At interim analysis 1 (March 27, 2020), 89 patients (9.7%) in the olaparib group and 152 patients (16.6%) in the placebo group had a DDFS event. The median DDFS was not estimable in either treatment arm, and the stratified HR for distant disease recurrence or death was 0.57 (99.5% CI, 0.39 to 0.83; P = 0.0000257) in favour of the olaparib group. At interim analysis 2 (July 12, 2021), the stratified HR for distant disease recurrence or death was 0.61 (95% CI, 0.48 to 0.77). The proportion of patients who remained distant disease–free at 4 years was 86.5% (95% CI, 83.8% to 88.8%) in the olaparib group and 79.1% (95% CI, 76.0% to 81.8%) in the placebo group (difference = 7.4%; 95% CI, 3.6% to 11.3%).¹⁷ The results of prespecified sensitivity and subgroup analyses were consistent with the primary analysis.

Health-Related Quality of Life

Health-related quality of life (HRQoL) data were assessed only in the PRO analysis set using the FACIT-F or EORTC QLQ-C30 questionnaires. No strong conclusions could be drawn about the effect of olaparib compared with placebo on HRQoL due to an increased risk of type I error and a high risk of attrition bias.

Functional Assessment of Chronic Illness Therapy–Fatigue

No clinically meaningful differences were found between treatment groups in mean change in FACIT-F score at follow-up (less than the minimal important difference [MID] of 3 points). For the subgroup of patients who had previously received neoadjuvant chemotherapy (N = 727), the adjusted least squares (LS) mean difference for olaparib versus placebo was −1.3 (95% CI, −2.4 to −0.2; P = 0.024) at 6 months, and −1.5 (–2.8 to −0.2; P = 0.025) at 12 months. For the subgroup of patients who had previously received adjuvant chemotherapy (N = 778), the adjusted LS mean difference for olaparib versus placebo was −1.3 (95% CI, –2.3 to −0.2; P = 0.017) at 6 months and −1.3 (95% CI, –2.4 to 0.1; P = 0.027) at 12 months.

EORTC QLQ-C30

The adjusted LS mean difference for olaparib versus placebo in the global health status score was █ ███ █ █ █ █ ██ ██ █ █ ██ ██████) at 6 months, █ █ ██ █ █ █ █ ██ ██ █ █ ██ ███████ at 12 months, − █ ██ █ █ █ █ ██ ██ █ █ ██ ███████ at 18 months, and █ █ ██ █ █ █ █ ██ ██ █ █ █ ██ ███████ at 24 months. The adjusted LS mean difference between treatment groups in the nausea and vomiting symptom scale was 6.0 (95% CI, 4.0 to 8.0; P █ █████) at 6 months, 6.3 (95% CI, 4.4 to 8.2; █ ██████) at 12 months, 0.4 (95% CI, –1.2 to 1.9; █ ██████) at 18 months, and 1.4 (95% CI, –0.4 to 3.3; █ ███████ at 24 months. The adjusted LS mean difference between treatment groups in the diarrhea symptom scale was 0.3 (95% CI, –2.0 to 2.7; █ ███████ at 6 months, 2.0 (95% CI, –1.0 to 4.9; █ ██████) at 12 months, 1.1 (95% CI, –2.0 to 4.3; █ ██████) at 18 months, and 1.8 (95% CI, –1.5 to 5.0; █ ██████) at 24 months.

Patients Who Had Previously Received Adjuvant Chemotherapy

The adjusted LS mean difference for olaparib versus placebo in the global health status score was █ █ ██ █ █ █ █ ██ ██ █ █ █ ██ ███████ at 6 months, █ █ ██ █ █ █ █ ██ ██ █ █ █ ██ ███████ at 12 months, − █ ██ █ █ █ █ ██ ██ █ █ ██ ██████) at 18 months, and █ █ ██ █ █ █ █ ██ ██ █ █ ██ ███████ at 24 months. The adjusted LS mean difference between treatment groups in the nausea and vomiting symptom scale was 5.3 (95% CI, 3.4 to 7.2; █ ██████) at 6 months, 4.5 (95% CI, 2.8 to 6.2 █ ██ ██████ at 12 months, −0.3 (95% CI, –1.9 to 1.3; █ ██████) at 18 months, and −0.6 (95% CI, –2.5 to 1.2; █ ██████) at 24 months. The adjusted LS mean difference between treatment groups in the diarrhea symptom scale was −1.7 (95% CI, –4.1 to 0.7; █ ██████) at 6 months, 0.1 (95% CI, –2.2 to 2.4; █ ██████) at 12 months, 0.4 (95% CI, –1.9 to 2.7 █ ██ █████) at 18 months, and −1.0 (95% CI, –3.4 to 1.4; █ ██████) at 24 months.

Harms Results

A total of 836 patients (91.8%) in the olaparib group and 758 patients (83.8%) in the placebo group experienced at least 1 adverse event (AE). Common Terminology Criteria for Adverse Events (CTCAE) grade 3 to 5 AEs occurred in 24.5% of patients in the olaparib group and 11.3% of patients in the placebo group. A total of 736 patients (80.8%) in the olaparib group and 480 patients (53.1%) in the placebo group experienced at least 1 treatment-related adverse event (TRAE). The most common TRAEs occurring in the olaparib or placebo groups were anemia (20.6% and 1.7%, respectively), diarrhea (12.0% and 7.5%, respectively), decreased neutrophil count (14.9% and 4.6%, respectively), and decreased white blood cell count (14.1% and 4.5%, respectively). A total of 33 patients (33.6%) in the olaparib group and 6 patients (0.7%) in the placebo group experienced at least 1 serious TRAE. The majority of TRAEs were manageable with supportive care and/or dose modifications and consistent with the known safety profile of olaparib. There were 2 fatal AEs in the placebo group and 1 fatal AE in the olaparib group during the treatment period or within the 30-day follow-up period, as well as 2 fatal AEs in the placebo group and 1 fatal AE in the olaparib group 30 days after discontinuation.

The frequency of notable harms identified in the protocol was comparable between the treatment groups. The most commonly reported notable AE was new primary cancer (2.3% and 4.0% in the olaparib and placebo groups, respectively), followed by pneumonitis (1.0% and 1.3% in the olaparib and placebo groups, respectively), and myelodysplastic syndrome or acute myeloid leukemia (0.2% and 0.3% in the olaparib and placebo groups, respectively). No new safety concerns have been identified compared with previous trials in patients with metastatic breast cancer.

Critical Appraisal

The OlympiA trial used accepted methods for blinding, allocation concealment, and randomization with stratification. The demographic and baseline patient characteristics were generally balanced between the treatment groups, so randomization was successful. A relatively high proportion of patients prematurely discontinued the trial medication (25.6% and 20.4% in the olaparib and placebo groups, respectively); however, the clinical experts noted that this is reflective of clinical practice. Since the OlympiA trial is ongoing, the longer-term efficacy of adjuvant olaparib for IDFS, DDFS, and OS is unknown. Further, since all results are based on interim analyses, there is the potential that the benefit of olaparib relative to placebo is overestimated; however, the presence and extent of any overestimation is uncertain.^20,21 All interim and subgroup analyses were prespecified in the statistical plan. Multiplicity adjustments for type I error were conducted for IDFS, DDFS, and OS according to a prespecified statistical hierarchy plan. The results were robust to a number of supportive and sensitivity analyses for the primary and key secondary outcomes. Subgroup analyses were prespecified in the OlympiA trial but may not have been powered to detect a treatment difference and there were no adjustments made for multiplicity. While improvement in quality of life was of primary importance to both patients and clinical experts, conclusions from the HRQoL assessment were limited, as no adjustments for multiplicity were made (so there is an increased risk of type I error). In addition, HRQoL was assessed using FACIT-F and EORTC QLQ-C30 questionnaires only in the PRO analysis set based on the evaluable baseline data; thus, there is a high risk of bias due to missing data, especially at later follow-up. There was a potential for unblinding of patients and investigators due to differences in the AE profile for olaparib relative to placebo. If unblinding were to occur, there would be a risk of performance and detection bias for self-reported quality of life and safety data; however, the direction and extent of any bias is uncertain.

The patient population in the OlympiA trial generally reflects patients in clinical practice in this setting. The majority of the study population was white and nonhispanic, and only 34 patients from Canada were recruited. However, the clinical experts consulted noted that although this may not be representative of the general breast cancer population, it is reflective of the population eligible for olaparib treatment, and the lack of representation of patients in Canada does not reduce the generalizability of results to Canadian clinical practice. To be enrolled in the OlympiA trial, patients were required to have completed at least 6 cycles of chemotherapy and all local therapies at least 2 weeks before randomization. The clinical experts consulted noted that olaparib would probably not be withheld if patients had previously received fewer than 6 cycles of chemotherapy for medical reasons. Patients with HR-positive, HER2-negative breast cancer were underrepresented in the OlympiA trial (17.7% with HR-positive, HER2-negative disease versus 82.3% with TNBC). The clinical experts consulted noted that these proportions are reflective of the group with hereditary breast cancer with BRCA mutations in clinical practice. Health Canada reviewers noted that due to the small number of patients with HR-positive, HER2-negative disease and lack of statistical power, the magnitude of the clinical benefit of olaparib in this subpopulation remains unclear.²² It was further indicated by the clinical experts that the criteria used in the OlympiA trial to determine a high risk of disease recurrence were reasonable, with the exception of a clinical and post-treatment pathologic stage and estrogen receptor (ER) status and histologic grade (CPS&EG) score of 3 or higher, which is not commonly used in clinical practice, although it is easily calculated. About 87.2% of patients in the OlympiA trial did not pass screening, most commonly because patients did not have a deleterious or suspected deleterious BRCA mutation in screening part 1. The clinician groups and clinical experts consulted agreed that companion diagnostic testing would be a challenge in Canada. They noted that current BRCA testing guidelines vary by province, and BRCA mutations are underdiagnosed based on most provincial testing criteria because current guidelines are restrictive in terms of eligibility criteria; thus, many patients who may carry a BRCA mutation may not receive the testing and thereby lose the opportunity to receive treatment with olaparib under current local or regional guidelines.

Indirect Comparisons

Description of Studies

To date, there have been no clinical trials directly comparing the efficacy of olaparib with other adjuvant treatments in patients diagnosed with HER2-negative, gBRCA-mutated, high-risk nonmetastatic breast cancer. The sponsor conducted a Bucher indirect treatment comparison (ITC) to address this gap.

The sponsor selected studies identified from a systematic literature review (SLR) to ensure that the population (or subpopulation), the control treatment, and the study design were aligned with those from the sponsor-conducted OlympiA trial.¹⁷ Relevant comparator interventions included adjuvant HER2-negative, high-risk breast cancer treatments publicly reimbursed in Canada. A feasibility assessment was then conducted to assess homogeneity between included studies and determine the appropriateness of inclusion in an ITC. The sponsor identified 1 randomized controlled trial (RCT), CIBOMA,¹⁹ that was feasible to be included in the ITC along with the sponsor-conducted OlympiA trial. The OlympiA trial was a phase III, double-blind RCT comparing olaparib with placebo in patients who were diagnosed with HER2-negative, gBRCA-mutated, high-risk early breast cancer and had received local treatment and neoadjuvant or adjuvant chemotherapy. The CIBOMA trial was a phase III, open-label RCT that compared capecitabine with observation in patients with TNBC who had been treated with neoadjuvant or adjuvant chemotherapy. Unlike the OlympiA trial, the CIBOMA trial did not require the participants to have confirmed gBRCA mutations. The median duration of follow-up was 2.5 years for the OlympiA trial and more than 7 years for the CIBOMA trial (interquartile range not reported).

The sponsor adopted the Bucher method to perform the ITC. The clinical end points included 3-year IDFS or disease-free survival (DFS) and OS. The risk of bias in the included studies was assessed independently by 2 reviewers using the checklist of the National Institute for Health and Care Excellence single technology appraisal user guide.²³

Efficacy Results

The Bucher ITC compared olaparib versus capecitabine via the common comparator — placebo or observation — and estimated the HRs for IDFS or DFS and OS in patients with TNBC from the OlympiA and CIBOMA trials. No conclusions could be drawn about the efficacy of olaparib compared with capecitabine due to imprecision in the effect estimates (i.e., wide 95% CIs, including HR = 1).

No analysis of harms was reported in the sponsor-submitted ITC report.

Critical Appraisal

The SLR used to identify relevant studies was methodologically sound in terms of the sponsor using a comprehensive literature search strategy as well as performing study selection, data extraction, and risk-of-bias assessment in duplicate. However, it was unclear in the ITC report whether the feasibility assessment was carried out by a single or multiple assessors. Moreover, although the risk of bias in individual studies was assessed in the SLR, the assessment results were not incorporated and discussed in the ITC report. The sponsor conducted the ITC, based on the Bucher method,²⁰ to estimate the relative treatment efficacy of olaparib against capecitabine. The Bucher method assumes that the trials included in the ITC should be sufficiently similar with respect to study population, study design, outcome measurements, and the distribution of treatment effect modifiers. The ITC has some limitations that reduce the CADTH team’s confidence in the effect estimates. There were notable differences across the 2 trials in the patient baseline demographics and disease characteristics (e.g., unknown BRCA mutation status in the CIBOMA trial) and trial design (e.g., double blind versus open label, outcome definitions) that might threaten the plausibility of the assumptions of the Bucher method. In addition, safety outcomes were not analyzed in the ITC report and no justification was provided, which precludes a balanced judgment of comparative benefits relative to comparative harms. Other outcomes that are important to patients, such as symptoms and HRQoL, were not investigated. Finally, the ITC was performed only in patients with TNBC, which only aligned with a part of the indication population specified in the sponsor’s application; therefore, the results may not be generalizable to all patients who meet the criteria in the reimbursement request.

Other Relevant Evidence

No other relevant evidence was submitted by the sponsor or identified from the literature.

Conclusions

Based on data from the OlympiA trial, olaparib demonstrated a clinically meaningful and statistically significant benefit compared with placebo in improving IDFS, DDFS, and OS in adult patients with HER2-negative, high-risk early breast cancer. The median IDFS, OS, and DDFS were not estimable in either treatment group because insufficient follow-up time had elapsed for these outcomes; thus, the longer-term efficacy of adjuvant olaparib is unknown. In addition, the estimates of the benefit of olaparib may be overestimated because the results are from interim analyses, although the presence and extent of any overestimation is uncertain. However, olaparib could help optimize adjuvant treatment in patients with BRCA-mutated early breast cancer to improve outcomes in terms of disease recurrence and survival, given its acceptable and manageable safety profile. The safety profile of olaparib was consistent with the known adverse effects profile of olaparib, and no new safety signals were identified. Strong conclusions could not be drawn related to the effect of olaparib on HRQoL due to the high risk of attrition bias and increased risk of type I error in the analyses of these outcomes. The evidence of olaparib was limited to 1 placebo-controlled pivotal trial, and no direct evidence of olaparib compared against other comparators was available for this review, most likely because the current systemic treatment of early breast cancer does not consider the patient’s BRCA mutation status. Uncertainties remain regarding the availability of BRCA mutation testing in Canada for clinical implementation in determining patient eligibility for olaparib treatment. No conclusions could be drawn from the ITC about the effect of olaparib relative to capecitabine on IDFS, DFS, or OS due to methodological limitations and imprecision in the effect estimates (wide 95% CIs, including HR = 1).

Introduction

Disease Background

Breast cancer is the most commonly diagnosed cancer among women in Canada, and the second most common cancer in men and women combined.¹ In 2020, 27,700 women were diagnosed with breast cancer, representing about 25% of new cancer cases in Canada.¹ Breast cancer is the second leading cause of cancer deaths among women, accounting for 14% of all cancer deaths.¹ It is estimated that about 1 in 8 women in Canada (12%) will develop breast cancer during their lifetime, and about 38% of cases will be diagnosed in females aged 30 to 59 years.¹ The 5-year net survival for breast cancer is more than 85% among women diagnosed before 85 years of age, after which it drops to about 73%.¹ In men, the incidence of breast cancer is less than 1% per year, with 260 new cases diagnosed in 2021 in Canada.¹ Breast cancer risk is influenced by several factors including age, family history, reproductive status (e.g., late menopause), birth control use, hormone exposures (e.g., estrogen), menopausal hormone therapy, inherited gene changes, race, and lifestyle factors, such as alcohol intake and physical inactivity.^1,24 More than 90% of patients with breast cancer are diagnosed with early-stage disease, which is defined as not having spread beyond the breast tissue or that has spread to the axillary lymph nodes only. Early breast cancer includes ductal carcinoma in situ (stage 0) and stages I to IIIA but may also include invasive cancers within stages I to IIIC, excluding stage 0.^1,25,26

Breast cancer susceptibility genes (BRCA1 and BRCA2) are human genes that produce proteins responsible for repairing damaged DNA through homologous recombination of DNA replication forks and double strand breaks, and play an important role in maintaining the genetic stability of cells.^2,3 Germline mutations in 1 or both BRCA genes reduce gene expression, which can lead to uncontrolled cell growth and is associated with an increased risk of cancer, including breast cancer.^4-6 Hereditary deleterious mutations contribute to 5% to 10% of all breast cancers, and 60% to 68% of these hereditary cancers occur in individuals with gBRCA mutations.^4,7 The BRCA mutations occur in women with all subtypes of breast cancer, but more commonly in those with early onset or a family history of breast cancer.⁹ Approximately 2.5% of Ashkenazi Jewish women carry 1 of 3 founding mutations in the BRCA1 and BRCA2 genes, which are associated with a high lifetime risk of invasive breast cancer.²⁷

Patients with breast cancer who inherit a harmful variant in the BRCA1 or BRCA2 genes have worse survival outcomes than those without BRCA mutations.²⁸ In women harbouring a BRCA1 mutation, the estimated lifetime risk of developing breast cancer by the age of 80 years is 65% to 80%, and the 10-year actuarial risk of developing contralateral breast cancer is 25% to 31%.^8,9 The estimated lifetime risk of developing breast cancer among women with a BRCA2 mutation is approximately 76%, while among men with BRCA1 or BRCA2 mutations, it ranges from 3% to 8%.^8-10 Approximately 75% of patients with breast cancer who have a mutation in the BRCA1 gene are classified as having TNBC, which is distinguished by the absence of a hormone receptor (HR-negative), and no expression of HER2 (HER2-negative).^11-13 More than 50% of patients with early-stage TNBC are bound to experience recurrence, of which 37% will die in the first 5 years after surgery.²⁹ In contrast, patients with breast cancer carrying mutations in the BRCA2 gene are more likely to be positive for expression of the hormone receptor (HR-positive), and only approximately 20% have TNBC.¹³ Although many patients with HR-positive, HER2-negative disease will not experience recurrence or have distant recurrence with standard therapies alone, around 7% to 11% of patients with early breast cancer will experience a local recurrence during the first 5 years after treatment.³⁰

The diagnosis of breast cancer is based on clinical examination in combination with imaging and is confirmed by pathological evaluation.⁹ Clinical examination includes bimanual palpation of the breasts and regional lymph nodes and assessment for distant metastases, while imaging includes bilateral mammography or ultrasound of the breast and regional lymph nodes. Pathological diagnosis should be based on a core needle biopsy. It also includes presence or absence of ductal carcinoma in situ, the histological type and grade, HER2 expression or HER2 gene amplification, and immunohistochemistry evaluation of progesterone receptor (PgR) or ER status.⁹ Genetic counselling and testing for a BRCA mutation are offered to patients with breast cancer in high-risk groups, including those with a strong family history of cancer, a diagnosis of breast cancer before the age of 50, and a recent or past history of ovarian or second breast cancer.⁹ Testing for gBRCA mutations involves blood or saliva tests.⁴ In addition, a small percentage of BRCA-related cancers contain purely somatic mutations that can be detected through direct analysis of the tumour tissue or circulating cell-free DNA.¹⁰

Standards of Therapy

The following section is based on input from the clinical experts and clinician groups consulted by CADTH for this review. The clinical experts and clinician groups consulted indicated that olaparib would be prescribed for the adjuvant treatment of adult patients with gBRCA-mutated, HER2-negative, high-risk early-stage breast cancer for 1 year after completion of adjuvant or neoadjuvant chemotherapy, and local therapy. Both the clinical experts and clinician groups agreed that the current systemic treatment for breast cancer does not consider the patient’s BRCA gene mutation status, as there is no treatment specific to mutation status. The clinical experts consulted indicated that, depending on the stage of the disease, most patients with high-risk early breast cancer start treatment with neoadjuvant therapy, such as chemotherapy (doxorubicin-cyclophosphamide plus paclitaxel [AC-T] or 5-fluorouracil-epirubicin-cyclophosphamide plus docetaxel [FEC-D]), followed by adjuvant therapy, depending on residual disease burden and tumour receptors (ER- or PgR-positive, or triple-negative). Both the clinical experts and clinician groups noted that clinicians currently use pembrolizumab therapy in combination with chemotherapy in the neoadjuvant setting followed by adjuvant pembrolizumab therapy for the treatment of patients with TNBC. They also noted that sometimes capecitabine can be used to treat patients with TNBC who have residual disease. The clinical experts consulted indicated that for the treatment of patients with HR-positive, HER2-negative breast cancer, clinicians currently use adjuvant endocrine therapy with or without abemaciclib, depending on whether patients meet monarchE trial criteria. The clinical experts highlighted that it remains unclear how adjuvant olaparib will be integrated into the treatment of HR-positive, HER2-negative breast cancer along with abemaciclib, as well as into the treatment of early TNBC along with pembrolizumab and capecitabine, as there are no clinical data to directly compare the efficacy and safety of these therapies for treating patients with a BRCA mutation. The clinical experts indicated the goal of therapy is to eradicate disease and prevent metastatic spread, resulting in cure.

Preferred treatment options outlined in the National Comprehensive Cancer Network,³¹ the St. Gallen International Consensus Guidelines,⁷ and the American Society of Clinical Oncology Hereditary Breast Cancer Guideline³² for patients with HER2-negative, early-stage (II or III) breast cancer with a high risk of recurrence and BRCA1 or BRCA2 mutations include olaparib or talazoparib therapy, irrespective of hormone receptor status.

Drug

Olaparib is a selective inhibitor of human PARP enzymes (PARP1, PARP2, and PARP3) involved in normal cellular functions, such as DNA transcription and DNA repair. It has been shown that olaparib blocks base excision repair by trapping PARP at the site of DNA damage, leading to the collapse of DNA replication forks and the accumulation of DNA double-stranded breaks. Thus, PARP inhibition has been identified as a targeted therapy that may exploit intrinsic defects in numerous cancer cells and has been reported to have selective cytotoxicity in breast cancer with mutations in the BRCA1 and/or BRCA2 genes.¹⁴

Olaparib underwent a priority review at Health Canada and obtained a Notice of Compliance with conditions on July 27, 2022, for the adjuvant treatment of adult patients with deleterious or suspected deleterious gBRCA-mutated, HER2-negative high-risk early breast cancer who have been treated with neoadjuvant or adjuvant chemotherapy. Patients must have confirmation of a gBRCAm before olaparib treatment is initiated.¹⁴ The sponsor’s requested reimbursement criteria for olaparib are aligned with the Health Canada–approved indication.³³ Olaparib is the first and only PARP inhibitor approved for the adjuvant treatment of gBRCA-mutated high-risk stage breast cancer in Canada. Olaparib is available as a 100 mg or 150 mg tablet. The Health Canada–recommended total daily dose of olaparib is 600 mg, taken as 2 150 mg tablets twice daily for a total of 1 year or until disease recurrence or unacceptable toxicity, whichever occurs first.¹⁴ The recommended reduced total daily dose of olaparib for the management of AEs is 500 mg. If a further dose reduction is required, the recommended reduced total daily dose of olaparib is 400 mg.¹⁴ Patients with HR-positive breast cancer should continue concurrent treatment with endocrine therapy as per current clinical practice guidelines.¹⁴

Olaparib has been previously approved by Health Canada and reviewed by CADTH for use as monotherapy for the maintenance treatment of adult patients with newly diagnosed advanced BRCA-mutated high-grade epithelial ovarian, fallopian tube, or primary peritoneal cancer who are in response (complete response or partial response) to first-line platinum-based chemotherapy until disease progression or up to 2 years if there is no evidence of disease.¹⁶ Olaparib has been reviewed for use as monotherapy maintenance treatment of adult patients with platinum-sensitive relapsed BRCA-mutated epithelial ovarian, fallopian tube, or primary peritoneal cancer who are in response to platinum-based chemotherapy.³⁴ Olaparib has also been reviewed for use as monotherapy for the treatment of adult patients with metastatic castration-resistant prostate cancer and deleterious or suspected deleterious germline and/or somatic mutations in the homologous recombination repair genes (BRCA1, BRCA2, and ATM) who have progressed following prior treatment with a new hormonal drug.³⁵

Key characteristics of commonly used medical treatments for early breast cancer are presented in Table 4.

Table 4: Key Characteristics of Pharmacotherapies for Early-Stage Breast Cancer

AML = acute myeloid leukemia; ATP = adenosine triphosphate; BRCA = breast cancer susceptibility gene; CDK = cyclin-dependent kinase; ET = endocrine therapy; gBRCAm = germline BRCA mutation; HER2 = human epidermal growth factor receptor 2; HR-positive = positive for expression of a hormone receptor; MDS = myelodysplastic syndrome; PARP = poly-(ADP-ribose) polymerase; PD-1 = programmed cell death 1 protein; PD-L1 = programmed cell death 1 ligand 1; PD-L2 = programmed cell death 1 ligand 2; Rb = retinoblastoma; RNA = ribonucleic acid.

^aCapecitabine may be used off-label for patients with triple-negative breast cancer.

^bHealth Canada–approved indication.

Source: Product monographs for Lynparza¹⁴ Verzenio,³⁶ Keytruda,³⁷ and Xeloda.³⁸

Stakeholder Perspectives

Patient Group Input

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

Two patient groups, the CBCN and Rethink Breast Cancer (Rethink) provided input for this review. CBCN is a national health charity aiming to ensure the best quality of care for all Canadians affected by breast cancer. The CBCN patient input was based on an online survey (the CBCN’s 2017 Lived Experience Breast Cancer Patient Survey) of 6 patients with gBRCA-mutated early breast cancer, and a literature review of current studies and grey literature. Rethink is a Canadian charity with a focus on improving the experience and outcomes of patients with breast cancer. Rethink gathered information for this review from general observations and insights through various ongoing initiatives (including stories shared by patients, virtual support groups, working groups, and patient advisory boards), in-depth telephone interviews with 3 patients with a BRCA-mutated breast cancer who participated in the OlympiA study, as well as responses from people in the Rethink Instagram community with high-risk early breast cancer.

According to the patient input received from Rethink, BRCA-mutated breast cancer is more likely to be detected in people at a young age. These young patients can face several age-specific issues such as fertility or family-planning challenges, diagnosis during pregnancy, childcare, impact on relationships, body image, dating and sexuality, feeling isolated from peers who do not have cancer, career hiatuses, and financial insecurity. Moreover, they may suffer a physical and emotional toll, as being diagnosed with a BRCA-mutated breast cancer at a young age may require making treatment-related decisions such as double mastectomy, oophorectomy, hysterectomy, egg or embryo preservation, and even decisions such as possible genetic screening of embryos.

The survey conducted by CBCN identified several key factors influencing patient decision-making about currently available treatment options. These factors included effectiveness of the treatment (i.e., how well the treatment could help stabilize disease and delay recurrence), prolonging life without sacrificing quality of life (i.e., how well the treatment could help the patient maintain a productive, active life with minimal disruption to daily routines), risk of side effects, and the cost and accessibility of treatments. The Rethink input revealed that patient respondents, especially those with stage III breast cancer, tended to endure side effects as well as impacts on quality of life to achieve satisfactory effectiveness. In terms of experience with olaparib, none of the 3 patients from the Rethink group who had participated in the OlympiA study who had not had a recurrence mentioned unendurable side effects. Four of the 6 (66%) CBCN survey respondents with inherited BRCA mutations indicated they received chemotherapy and 2 (33%) received radiation therapy as part of their overall breast cancer treatment. None of the CBCN survey respondents reported having experience with olaparib treatment. Both patient groups expressed high interest in olaparib, as they had high hopes that it would bring benefits, from extension of progression-free disease to avoidance of metastasis.

Clinician Input

Input From the Clinical Experts Consulted by CADTH

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

Unmet Needs

The clinical experts consulted by CADTH indicated that most patients with BRCA mutations have TNBC and, accordingly, different mechanisms of disease development due to their mutation status, which is not targeted by current treatment options. They also noted that not all patients respond to current treatments; therefore, there is a need for new therapeutic options to improve survival outcomes and increase the overall cure rate in patients with BRCA-mutated breast cancer.

Place in Therapy

The clinical experts agreed that current systemic treatment of breast cancer is based on the receptor status and pathological findings and considers the patient’s BRCA mutation status. According to the clinical experts, as the mechanism of action of olaparib is complementary, patients should still pursue appropriate neoadjuvant and adjuvant therapies as per their pathology and stage; however, olaparib would be an additional drug that would be recommended. The clinical experts indicated that olaparib would be used as adjuvant therapy for 1 year in patients with BRCA mutations who meet the OlympiA trial inclusion criteria, regardless of tolerance to prior therapies. The clinical experts agreed that it remains unclear how different drugs would be prioritized if tolerance became an issue to multiple drugs; it is likely that the drug most likely to cause toxicity will be stopped (i.e., in the case of severe hot flashes or vaginal dryness, endocrine therapy would be modified). They also highlighted that it remains unclear how to integrate olaparib into the current treatment paradigm with other drugs, such as capecitabine, pembrolizumab, abemaciclib, or a combination of drugs in the treatment of patients with breast cancer who have a BRCA mutation. The clinical experts consulted indicated that this is a change in practice for patients who carry the BRCA mutation and that a shift in the treatment paradigm would occur, as this is the new standard-of-care therapy in these patients.

Patient Population

The clinical experts consulted by CADTH agreed that patients who meet the inclusion criteria outlined in the OlympiA trial will be best suited for treatment with olaparib (i.e., patients who have a germline BRCA1 or BRCA2 pathogenic or likely pathogenic variant; have high-risk, HER2-negative primary breast cancer who have completed at least 6 cycles of neoadjuvant or adjuvant chemotherapy containing anthracyclines, taxanes, or both drugs; and completed all local therapy, including radiotherapy at least 2 weeks and not more than 12 weeks before trial entry). The clinical experts noted that adjuvant bisphosphonates in postmenopausal patients and endocrine therapy for HR-positive breast cancer are allowed; however, no chemotherapy is allowed after surgery in patients who received neoadjuvant chemotherapy. The clinical experts consulted mentioned that companion diagnostic testing would be a challenge, especially as clinicians would request BRCA mutation testing for patients who do not qualify for genetic testing based on provincial guidelines (i.e., patients who have a low likelihood of hereditary syndromes) but who meet the criteria for adjuvant olaparib in the presence of a BRCA mutation. The clinical experts indicated that current genetic testing guidelines vary by province, and BRCA mutations are underdiagnosed based on most provincial testing criteria. In addition, there are concerns that BRCA mutation testing may not happen fast enough; therefore, increased access to genetic testing and reduced turnaround times will likely be necessary.

Assessing Response to Treatment

The clinical experts consulted noted that since the current indication is for the adjuvant setting, treatment response cannot be assessed. They agreed that improved long-term survival is the important outcome at a population level. It was further noted by the clinical experts that by improving survival and reducing disease recurrence, patients who are cured will have a higher quality and quantity of life. The goal of therapy is to eradicate disease and prevent metastatic spread, resulting in cure.

Discontinuing Treatment

According to the clinical experts consulted, toxicity from treatment with olaparib and disease recurrence would be the main factors to consider when deciding to discontinue treatment with olaparib.

Prescribing Conditions

The clinical experts consulted agreed that olaparib should be prescribed in a standard outpatient medical oncology clinic (i.e., in a community hospital or tertiary setting) by medical oncologists and general practitioners in oncology who have medical oncology training. The clinical experts also noted that ideally, pharmacists with expertise in oncology drugs should be utilized; however, outpatient retail pharmacies will also be able to access and dispense this drug in certain jurisdictions, as it is an oral drug.

Clinician Group Input

This section was prepared by CADTH staff based on the input provided by clinician groups. The full original clinician group input received by CADTH has been included in the stakeholder section at the end of this report.

Clinician group input was obtained from 2 clinician groups, including the OH-CCO Breast Cancer Drug Advisory Committee (1 clinician provided input) and a group of medical oncologists across Canada (4 clinicians provided input). The medical oncologist group stated that no therapy other than surgical treatment has been adopted specifically for patients with BRCA mutation–associated tumours, which are often more aggressive. Both clinician groups identified that the important goal of treatment for early breast cancer, including gBRCA-mutated early breast cancer, is to decrease the recurrence of cancer and improve survival.

Both clinician groups agreed that patients must have confirmation of a gBRCA mutation before receiving olaparib. The medical oncologist group highlighted the increasing demand and workload for germline mutation testing. One potential barrier, which was mentioned by the OH-CCO Breast Cancer Drug Advisory Committee, is that the current guidelines for BRCA mutation testing are restrictive, so patients who may carry a BRCA mutation may not receive the testing and thereby lose the opportunity to receive olaparib. It was further noted by the clinician groups consulted that timely BRCA testing will be needed to ensure patients and clinicians have access to the results to initiate treatment with olaparib.

To assess treatment response, the OH-CCO Breast Cancer Drug Advisory Committee indicated that standard cancer care follow-up without diagnostic imaging will suffice unless patients show signs or symptoms suggestive of recurrent or progressive disease; in such cases, radiographic imaging or biopsy may be conducted. Both clinician groups pointed out several reasons that may lead to the discontinuation of olaparib, including recurrence or progression of disease, intolerant toxicity or severe side effects, and patient or physician preference.

Drug Program Input

The drug programs provide input on each drug reviewed through CADTH’s reimbursement review processes by identifying issues that may impact their ability to implement a recommendation. The implementation questions and corresponding responses from the clinical experts consulted by CADTH are summarized in Table 5.

Table 5: Summary of Drug Plan Input and Clinical Expert Response

BRCA = breast cancer susceptibility gene; CPS&EG = clinical and post-treatment pathologic stage and estrogen receptor status and histologic grade; HER2 = human epidermal growth factor receptor 2; HR-positive = positive for expression of a hormone receptor; pCR = pathological complete response; pERC = CADTH pan-Canadian Oncology Drug Review Expert Review Committee; TNBC = triple-negative breast cancer.

Clinical Evidence

The clinical evidence included in the review of olaparib is presented in 3 sections. The first section, the systematic review, includes pivotal studies provided in the sponsor’s submission to CADTH and Health Canada, as well as those studies that were selected according to an a priori protocol. The second section includes indirect evidence from the sponsor and indirect evidence selected from the literature that met the selection criteria specified in the review. The third section includes sponsor-submitted long-term extension studies and additional relevant studies that were considered to address important gaps in the evidence included in the systematic review.

Systematic Review (Pivotal and Protocol Selected Studies)

Objectives

To perform a systematic review of the beneficial and harmful effects of olaparib, 300 mg (2 × 150 mg tablets) twice daily for the adjuvant treatment of adult patients with deleterious or suspected deleterious gBRCA-mutated, HER2-negative, high-risk early breast cancer who have been treated with neoadjuvant or adjuvant chemotherapy.

Methods

Studies selected for inclusion in the systematic review include pivotal studies provided in the sponsor’s submission to CADTH and Health Canada as well as those meeting the selection criteria presented in Table 6. The outcomes included in the CADTH review protocol reflect outcomes considered to be important to patients, clinicians, and drug plans.

Table 6: Inclusion Criteria for the Systematic Review

AML = acute myeloid leukemia; CPS&EG = clinical and post-treatment pathologic stage and estrogen receptor status and histologic grade; ET = endocrine treatment; gBRCA = germline breast cancer susceptibility gene; gBRCAm = germline BRCA mutation; HER2 = human epidermal growth factor receptor 2; HR-positive = positive for expression of a hormone receptor; HRQoL = health-related quality of life; MDS = myelodysplastic syndrome; RCT = randomized controlled trial; TNBC = triple-negative breast cancer.

^aMay be used off-label for the treatment of patients with triple-negative breast cancer.

^bNot funded in Canada; however, considered relevant by clinicians.

The literature search for clinical studies was performed by an information specialist using a peer-reviewed search strategy according to the PRESS Peer Review of Electronic Search Strategies checklist.³⁹

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

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

CADTH-developed search filters were applied to limit retrieval to RCTs or controlled clinical trials. Retrieval was not limited by publication date or by language. Conference abstracts were excluded from the search results. See Appendix 1 for the detailed search strategies.

The initial search was completed on August 31, 2022. Regular alerts updated the search until the meeting of the CADTH pan-Canadian Oncology Drug Review Expert Committee (pERC) on January 11, 2023.

Grey literature (literature that is not commercially published) was identified by searching relevant websites from the Grey Matters: A Practical Tool For Searching Health-Related Grey Literature checklist.³⁹ Included in this search were the websites of regulatory agencies (FDA and European Medicines Agency). Google was used to search for additional internet-based materials. See Appendix 1 for more information on the grey literature search strategy. A focused literature search for network meta-analyses dealing with olaparib or breast cancer was run in MEDLINE All (1946–) on August 26, 2022. No search limits were applied.

These searches were supplemented by reviewing bibliographies of key papers. In addition, the sponsor of the drug was contacted for information regarding unpublished studies.

Findings From the Literature

A total of 4 reports^17-19,40 of 1 study were identified from the literature for inclusion in the systematic review (Figure 1). The included studies are summarized in Table 7. A list of excluded studies is presented in Appendix 2.

Figure 1: Flow Diagram for Inclusion and Exclusion of Studies

Table 7: Details of the OlympiA Study

AE = adverse event; BGI = Beijing Genomics Institute; BRCA = breast cancer susceptibility gene; CPS&EG = clinical and post-treatment pathologic stage and estrogen receptor status and histologic grade; CTCAE = Common Terminology Criteria for Adverse Events; DCIS = ductal carcinoma in situ; DDFS = distant disease–free survival; ECG = electrocardiogram; ECOG = Eastern Cooperative Oncology Group; ER = estrogen receptor; EORTC QLQ-C30 = European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire Core 30; FACIT-F = Functional Assessment of Chronic Illness Therapy–Fatigue; gBRCA = germline breast cancer susceptibility gene; gBRCA1 = germline BRCA1 gene; gBRCA2 = germline BRCA2 gene; gBRCAm = germline BRCA mutation; GI = gastrointestinal; HER2 = human epidermal growth factor receptor 2; IDFS = invasive disease–free survival; IHC = immunohistochemistry; ISH = in situ hybridization; OS = overall survival; PARP = poly-(ADP-ribose) polymerase; pCR = pathological complete response; PgR = progesterone receptor; pN0 = axillary node–negative; pN1 = axillary node–positive; pT2 = pathological tumour size > 2 cm; QoL = quality of life; RCT = randomized controlled trial; SAE = serious adverse event; TNBC = triple-negative breast cancer; ULN = upper limit of normal.

^aDefined as IHC 0 or 1+ without ISH, or IHC 2+ and ISH nonamplified with ratio < 2.0 and, if reported, average HER2 copy number < 4 signals per cell, or ISH nonamplified with ratio < 2.0 and if reported, average HER2 copy number < 4 signals per cell (without IHC).

^bIHC 0 or 1+ without ISH, or IHC 2+ and ISH nonamplified with ratio < 2.0 and, if reported, average HER2 copy number < 4 signals per cell, or ISH nonamplified with ratio < 2.0 and, if reported, average HER2 copy number < 4 signals per cell (without IHC).

^cKnown or predicted to be detrimental or lead to a loss of function. Local gBRCA testing results, if available, were used for establishing eligibility.

^dFor example, “variants of uncertain clinical significance” or “variant of unknown significance” or “variant, favour polymorphism,” or “benign polymorphism.”

^eThere were 2 exceptions. The first was adequately treated nonmelanoma skin cancer, curatively treated in situ cancer of the cervix, DCIS, or stage I grade 1 endometrial carcinoma. The second was other solid tumours and lymphomas (without bone marrow involvement) diagnosed > 5 years before randomization and treated with no evidence of disease recurrence and for which no more than 1 line of chemotherapy was applied.

^fThe required washout period before starting study treatment was 2 weeks.

^gThe required washout period before starting study treatment was 5 weeks for enzalutamide or phenobarbital and 3 weeks for other drugs.

^hOnly patients requiring a gBRCA status Myriad or BGI test before randomization.

^IPatients with confirmed gBRCA status by Myriad or BGI, or if BRCA status was known before study entry.

^jSafety follow-ups were implemented during the 1-year study treatment period and 30 days after its discontinuation.

^kEfficacy follow-ups were implemented every 3 months during the first 2 years following end of treatment, and every 6 months for years 3, 4, and 5 and annually thereafter for approximately 10 years.

^lUntil 10 years after the last patient was randomized.

^mDefined as the time from randomization to date of first recurrence, where recurrence was defined as invasive locoregional, distant recurrence, contralateral invasive breast cancer, second primary nonbreast invasive malignancy, or death from any cause.

ⁿDefined as the time from the date of randomization until death due to any cause.

^oDefined as the time from randomization until documented evidence of the first distant recurrence of breast cancer or death from any cause.

Sources: Clinical Study Reports for OlympiA.^18,19

Description of the OlympiA Trial

The OlympiA trial is an ongoing, phase III, randomized, multicentre, double-blind, placebo-controlled trial. The primary objective of the trial was to assess the efficacy and safety of olaparib versus placebo as adjuvant treatment in patients with deleterious or suspicious deleterious germline BRCA1 or BRCA2 mutations and high-risk HER2-negative early-stage breast cancer who had completed definitive local treatment and neoadjuvant or adjuvant chemotherapy. This was a collaborative study that was coordinated worldwide by Breast International Group in partnership with Frontier Science, NRG Oncology, and AstraZeneca. A total of 1,836 patients with breast cancer and gBRCA mutations were enrolled across 546 sites in 23 countries in North America (34 patients from Canada), South America, Europe, Asia Pacific, and South Africa. The primary efficacy end point was IDFS and the key secondary efficacy end points were OS and DDFS. PROs were assessed using the FACIT-F questionnaire and European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire Core 30 (EORTC QLQ-C30). Treatment with olaparib was given for up to 12 months, or until disease recurrence or unacceptable toxicity, whichever occurred first.

In the OlympiA trial, the first superiority interim analysis was performed when 50% of the events required for the primary IDFS analysis (165 events) had been reached in the first 50% of recruited patients (900 patients). The second interim analysis was performed when 330 IDFS events had been reported. The first interim database lock was executed on March 27, 2020, and the first clinical database lock was executed on December 15, 2020. The second interim database lock was executed on July 12, 2021, and the second clinical database lock was executed on December 17, 2021. Both interim analyses were performed by the independent data-monitoring committee.

Randomization, Treatment Allocation, and Blinding

Interactive response technology was used to randomly assign patients using permuted blocks in a 1:1 ratio to either olaparib at a dose of 300 mg twice daily (N = 921) or matching placebo (N = 915). The last patient was randomized to study treatment on May 28, 2019. Randomization was stratified by:

• hormone receptor status (ER or PgR), HER2-negative versus TNBC

• prior neoadjuvant versus adjuvant chemotherapy

• prior platinum use for breast cancer (yes versus no).

Ideally, patients should have been randomized in the OlympiA trial within a maximum of 8 weeks after the completion of their last treatment, including surgery, chemotherapy, or radiation therapy, but in no case longer than 12 weeks. Both patients and investigators were blinded to the study treatments administered during the trial. A schematic of the OlympiA trial is presented in Figure 2. An independent data-monitoring committee was responsible for reviewing the unblinded safety and efficacy analyses.

Study Phases

Screening Phase: In the OlympiA trial, there were 2 screening parts. Screening part 1 was performed at any time before, during, or after neoadjuvant or adjuvant chemotherapy. It was applied only to patients with unknown gBRCA status before enrolment in the trial and was conducted to determine if the patient was considered eligible for local or central Myriad testing for BRCA status. Once this phase was successfully completed and patients had the BRCA test results, they moved on to screening part 2. Screening part 2 was applied to patients with known gBRCA status, confirmed by Myriad or local testing, who had a deleterious or suspected deleterious mutation. These patients underwent screening assessments, including evaluation of eligibility criteria, baseline characteristics, and concomitant medications. Patients with known BRCA status were required to consent to participate in the OlympiA trial within 28 days before randomization.

Treatment Phase: The treatment phase started with the first dose of treatment following randomization (olaparib and placebo in the intervention and control groups, respectively). During the treatment phase, all patients had assessments every 2 weeks during the first month, every 4 weeks for the following 5 months, and every 3 months for the remaining 6 months of study treatment. The following assessments were performed during the onsite visits: physical examination, vital signs, hematology, radiological tests, and concomitant medications.

Safety Follow-Up: A follow-up visit was conducted 30 days after the last dose of olaparib or placebo. Patients who discontinued study treatment before completing week 52 were required to attend the study treatment discontinuation visit followed by the 30-day follow-up visit. The date of discontinuation, the reasons, and details of the specific cancer therapies were recorded.

Efficacy Follow-Up: Efficacy assessments were performed every 3 months for the first 2 years following end of treatment to assess disease recurrence and new cancers. During years 3, 4, and 5, patients were assessed every 6 months and then annually for approximately 10 years. Assessments for disease recurrence, new cancers, and OS were performed based on signs and symptoms, clinical assessment, mammograms, and/or MRI and vital signs.

Survival Follow-Up: After completion of approximately 10 years of efficacy follow-up assessment, patients will enter the annual survival follow-up phase, which will continue until 10 years after the last patient is randomized. If the patient met the primary end point (IDFS) due to a breast cancer–related distant recurrence, the patient entered the survival follow-up phase of the trial with annual assessments, which continued until 10 years after the last patient was randomized. If the IDFS end point was reached due to events other than distant recurrence, patients continued efficacy assessments until breast cancer–related distant recurrence or approximately 10 years after their randomization into the study. During this phase, no onsite clinical visits were required.

Protocol Amendments

Several modifications were made to the study protocol and after the start of patient recruitment that were documented in the Clinical Study Report. The amendments dated October 21, 2015, included the:

• inclusion of patients with ER-positive and/or PgR-positive high-risk HER2-negative breast cancer

• addition of hormone receptor status to the stratification factors

• extension of the randomization window

• inclusion of specific requirements and definitions for patients with HR-positive and HER2-negative cancer

• collection of the family cancer detection

• increase in the number of centres and number of randomized patients from 1,500 to 1,800.

The amendments dated July 27, 2018, included:

• extension of study recruitment to 5 years

• confirmation that the interim analysis would be triggered by events in the first 900 patients

• inclusion of pneumonitis in the events occurring after 30-day follow-up period

• addition of serious adverse events (SAEs) reports for all new primary cancers.

There were several modifications to the statistical analysis plan for the study that were documented in the statistical analysis plan amendments. The amendments dated April 9, 2015, included: addition of prior platinum therapy and HR status to the subgroup analyses; update time for interim analysis to a minimum of 165 IDFS events observed from the first 750 patients recruited; addition of formal statistical analyses for multiple end points for IDFS, OS, and DDFS; and addition of a description for the imputation for missing data. The amendments dated May 18, 2018, included: sample size was increased to 1,800 with 90% power, time for the primary analysis was changed to 330 IDFS events, time for the interim analysis was updated to a minimum of 165 IDFS events observed for the first 900 patients, censoring rules were added, a hierarchy of IDFS events was added, the baseline and demographic analyses were updated, the incidence of new cancers was updated, several sensitivity analyses were added, the method of subgroup analysis was expanded, and more detail was added to global interaction. Additional changes not specified in the statistical analysis plan amendment included the addition of a list of all protocol deviations related to COVID-19 and the exclusion of 2 patients who were stratified as patients with TNBC but whose disease was subsequently found to be HER2-positive or of unknown status.

Figure 2: Study Schema for the OlympiA Trial

BRCA = breast cancer susceptibility gene; CPS&EG = clinical and post-treatment pathologic stage and estrogen receptor status and histologic grade; ER = estrogen receptor; gBRCA = germline BRCA; HER2 = human epidermal growth factor receptor 2; ICF = informed consent form; path CR = pathological complete resistance; PgR = progesterone receptor; pN0 = axillary node–negative; pN1 = axillary node–positive; pT2 = tumour > 2 cm; t = tumour; TNBC = triple-negative breast cancer.

Sources: Clinical Study Reports for OlympiA.^18,19

Populations

Inclusion and Exclusion Criteria

The key inclusion and exclusion criteria applied to the OlympiA trial are summarized in Table 6. Briefly, patients eligible for enrolment in the OlympiA trial had a documented germline BRCA1 or BRCA2 pathogenic or likely pathogenic variant, and high-risk early-stage HER2-negative primary breast cancer after definitive local treatment and neoadjuvant or adjuvant chemotherapy. To be enrolled in the trial, patients had to have known BRCA status, which could be determined using local genetic testing or could be assessed as part of the study-entry procedure through central Myriad testing. Patients must have completed all local therapy and at least 6 cycles of neoadjuvant or adjuvant chemotherapy containing anthracyclines, taxanes, or both drugs. All patients had either TNBC or ER- and/or PgR-positive disease, HER2-negative breast cancer. Patients with TNBC who were treated with adjuvant chemotherapy were required to have axillary node–positive disease or axillary node–negative disease with an invasive primary tumour measuring at least 2 cm on pathological analysis. Those who were treated with neoadjuvant chemotherapy were required to have residual invasive breast cancer in the breast and/or resected lymph nodes. Patients who were treated with adjuvant chemotherapy for HR-positive, HER2-negative breast cancer were required to have at least 4 pathologically confirmed positive lymph nodes. Those who were treated with neoadjuvant chemotherapy were required to have residual invasive cancer in the breast and/or resected lymph nodes (i.e., did not achieve a pathological complete response) and a CPS&EG score of 3 or higher. Patients were excluded from the OlympiA trial if they had metastatic breast cancer, any previous treatment with a PARP inhibitor, secondary primary malignancy, or were concomitantly using strong or moderate CYP3A inhibitors or CYP3A inducers, as the required washout period before starting study treatment was 2 to 3 weeks. Patients who received systemic chemotherapy within 3 weeks before randomization or those who received adjuvant radiotherapy within 2 weeks before randomization were also excluded.

Baseline Characteristics

A summary of baseline characteristics is presented in Table 8 and Table 9. Baseline characteristics were well balanced between the treatment groups. The mean age of all randomized patients in the OlympiA trial was 43.3 years (SD = 9.97 years), and 68.7% of patients were between 30 and 49 years of age. Most patients were female (99.7%) and premenopausal (61.3%), had an ECOG Performance Status of 0 (88.7%), and were white (66.7%) or nonhispanic or Latino (88.1%). A total of 26.4% of patients received prior platinum therapy and half of patients (50.1%) received neoadjuvant treatment. Most patients underwent nonconservative surgery (74.7%) and received a regimen that included both anthracycline and taxane therapy (93.7%) as prior adjuvant or neoadjuvant chemotherapy.

A total of 82.3% of the patients had TNBC (HR-negative and HER2-negative), while 17.7% had HR-positive, HER2-negative breast cancer. Germline BRCA1 deleterious or suspected deleterious mutations were identified in 72.2% of patients, gBRCA2 mutations in 27.1% of patients, and both gBRCA1 and gBRCA2 mutations in 0.4% of patients. The majority of patients with TNBC (60.3%) had a mutation in BRCA1, while the majority of patients with HR-positive, HER2-negative breast cancer (51.4%) had a mutation in BRCA2. For patients for whom both local and central Myriad BRCA testing was available, the test results were concordant. Of 1,836 patients randomized in the OlympiA trial, 83.8% were confirmed to have a gBRCA mutation using the central Myriad test. A total of 36.1% of patients had clinical AJCC stage IIA, 21.0% had AJCC stage IIB, and 13.0% had AJCC stage IIIA breast cancer.

Table 8: Summary of Baseline Characteristics — FAS

BC = breast cancer; BRCA = breast cancer susceptibility gene; BRCA1 = breast cancer susceptibility gene 1; BRCA2 = breast cancer susceptibility gene 2; DCIS = ductal carcinoma in situ; ECOG = Eastern Cooperative Oncology Group; eCRF = electronic case report form; ER = estrogen factor; FAS = full analysis set; gBRCAm = germline BRCA mutation; HER2 = human epidermal growth factor 2; NOS = not otherwise specified; PgR = progesterone factor; pN1 = axillary node–positive; Q1 = 25th percentile; Q3 = 75th percentile; SD = standard deviation; TNBC = triple-negative breast cancer.

^aNot Ashkenazi Jewish could mean that the patient is Jewish but not Ashkenazi Jewish, not Jewish, or descent recorded as unknown.

^bPost randomization, 2 patients (included as TNBC) were found not to have confirmed negative HER2 status. These patients were captured as important protocol deviations in the “No histologically confirmed nonmetastatic primary invasive adenocarcinoma of the breast” category.

^cLocal BRCA results were only available for patients who knew their gBRCA mutation status before entry into the study. All randomized patients were required to have a central test. Myriad testing was not done in China. Beijing Genomics Institute (China) results were considered local results.

^dPatient E106578 (placebo group) was reported to have had a right mastectomy (DCIS) in 2001 in the concomitant procedure in the eCRF and the patient received axillary lymph node dissection (pN1) before entering the OlympiA study. Patient E112449 (placebo group) was reported to have occult breast carcinoma on the histology report and no visible tumour in the breast per MRI; the patient received axillary lymph node dissection (pN1) before entering OlympiA study.

Sources: Clinical Study Reports for OlympiA.^18,19

Table 9: Pathological Characteristics of Primary Breast Cancer — FAS

AJCC = American Joint Committee on Cancer; BC = breast cancer; BRCA = breast cancer susceptibility gene; FAS = full analysis set; gBRCAm = germline BRCA mutation; ITT = intention to treat; NOS = not otherwise specified.

^aMay be more than 1 type per patient.

Sources: Clinical Study Reports for OlympiA.^18,19

Interventions

In the OlympiA trial, all eligible patients were randomized in a 1:1 ratio to receive 1 of 2 interventions: olaparib at a dose of 300 mg twice daily or matching placebo in a double-blind manner. The drugs were administered at the same time each morning and evening with an interval of about 12 hours with light food. The olaparib and placebo tablets were identical in packaging and labelling. Dose reduction was available with 100 mg tablets. If AEs occurred, such as moderate renal impairment, the initial daily dose of olaparib (600 mg) could be reduced to 500 mg or 400 mg. The daily dose of olaparib was reduced to 300 mg or 200 mg during concomitant administration of moderate or strong CYP3A inhibitors, respectively. After a dose reduction of olaparib, an increase to a higher dose was not permitted. Patients were not prescribed olaparib after the discontinuation of the study treatment.

Concomitant Medications

All concomitant medications were documented at each visit. Concomitant medications that were considered necessary for the patient’s welfare and that did not interfere with the study medication were allowed at the discretion of the investigator. Patients received adjuvant endocrine therapy per local policy or international guidelines. Bisphosphonates or denosumab were allowed during the study treatment and follow-up phases. Other antineoplastic medications were prohibited during study treatment and IDFS follow-up. Subsequent anticancer treatment was expected to be initiated after breast cancer recurrence. If it was necessary to initiate during the 1-year treatment phase, it was important that the study treatment be discontinued. If a patient developed nausea, vomiting, or diarrhea, these symptoms were recorded as AEs and appropriately treated. The use of natural or herbal products or other traditional remedies was discouraged.

Medications prohibited during the trial included: strong CYP3A inhibitors (e.g., itraconazole, clarithromycin), moderate CYP3A inhibitors (e.g., ciprofloxacin, erythromycin), strong CYP3A inducers (e.g., phenobarbital, rifampicin), moderate CYP3A inducers (bosentan, efavirenz, and modafinil), CYP3A4 substrates, CYP2B6 substrates (bupropion, efavirenz), OATP1B1 substrates, OCT2 substrates, OAT3 substrates (furosemide, methotrexate), and anticoagulant therapy.

Study Treatment Discontinuation

Patients had to be discontinued from the study treatment in the following circumstances:

• patient decision: the patient was free at any time to discontinue study treatment, without prejudice to further clinical care

• AE

• completion of 1-year treatment period

• the patient has a confirmed pregnancy during treatment

• severe noncompliance with the clinical study protocol

• locoregional breast cancer recurrence (ipsilateral invasive breast cancer recurrence, regional invasive breast cancer recurrence)

• distant breast cancer recurrence

• contralateral invasive breast cancer or new primary nonbreast invasive cancer

• death

• bone marrow findings consistent with myelodysplastic syndrome or acute myeloid leukemia

• diagnosis of breast cancer recurrence or diagnosis of a secondary primary malignancy.

Patients were to be seen within at least 30 days of discontinuation of the study drug to collect and/or complete information on AEs.

Study Withdrawal

Patients withdrew from the study in the following circumstances:

• voluntary withdrawal by the patient who was at any time free to discontinue their participation in the study without prejudice to further treatment

• patient was enrolled incorrectly, for instance, the patient did not meet the required inclusion or exclusion criteria for the study; this option was applicable only to patients not randomized into the study (e.g., patients who did not pass screening and who were identified before randomization)

• patient lost to follow-up

• death.

Outcomes

A list of efficacy end points identified in the CADTH review protocol that were assessed in the clinical trials included in this review is provided in Table 10. These end points are further summarized subsequently. A detailed discussion and critical appraisal of the outcome measures is provided in Appendix 4.

The primary end point for the OlympiA trial was IDFS in the intention-to-treat (ITT) population. IDFS was defined per the Standardized Definitions for Efficacy End Points (STEEP) criteria⁴¹ as the time from randomization to date of first recurrence, where recurrence is defined as follows:

• ipsilateral invasive breast tumour recurrence: invasive breast cancer involving the same breast parenchyma as the original primary

• regional invasive breast cancer recurrence: invasive breast cancer in the axilla, regional lymph nodes, chest wall, and skin of the ipsilateral breast

• distant recurrence, i.e., metastatic breast cancer that had either been biopsy-confirmed or radiologically diagnosed as recurrent invasive breast cancer

• death attributable to any cause, including breast cancer, nonbreast cancer, or unknown cause

• contralateral invasive breast cancer

• new primary nonbreast invasive cancers (i.e., excluding new in situ carcinomas of any site). New primary nonbreast invasive cancers include hematologic cancers and myelodysplastic syndrome. Squamous or basal cell skin cancers were not counted as primary end point events.

Table 10: Summary of Outcomes of Interest Identified in the CADTH Review Protocol

EORTC QLQ-C30 = European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire Core 30; FACIT-F = Functional Assessment of Chronic Illness Therapy–Fatigue; GI = gastrointestinal; QoL = quality of life.

Sources: Clinical Study Reports for OlympiA.^18,19

Locoregional recurrence of the disease (ipsilateral or regional invasive breast cancer) had to be confirmed by cytological or histological examination and appropriate imaging. Distant recurrence had to be diagnosed by radiological examination and/or histopathological assessment when the metastatic lesion is easily accessible for biopsy. Invasive contralateral breast cancer or invasive nonbreast secondary primary cancer was required to be confirmed by histopathological report.

The key secondary end points for the OlympiA trial were OS and DDFS in the ITT population. OS was defined as the time from the date of randomization to death due to any cause. Any patient not known to have died at the time of analysis was censored based on the last recorded date the patient was known to be alive.

DDFS was defined as the time from randomization to documented evidence of the first distant recurrence of breast cancer, including:

• distant recurrence: metastatic disease breast cancer that was either biopsy-confirmed or radiologically diagnosed as recurrent invasive breast cancer

• death attributable to any cause, including breast cancer, nonbreast cancer, or unknown cause

• new primary nonbreast invasive cancer.

Evidence of distant recurrence had to be either biopsy-confirmed or radiologically diagnosed. New primary nonbreast invasive cancer had to be confirmed by histopathological report. The first site of distant recurrence was recorded and used in the analysis.

The incidence of contralateral invasive breast cancer, contralateral noninvasive breast cancer, new primary ovarian cancer, new primary fallopian tube cancer, and new primary peritoneal cancer were each considered separately in the ITT population. Evidence of new contralateral cancer and new ovarian, fallopian, or peritoneal cancers was confirmed by radiologic examination, with positive histology or cytology. Time to new contralateral cancer and new ovarian, fallopian, or peritoneal cancer was defined as the date of randomization to the date of the first occurrence of the event.

Patient-Reported Outcomes

The primary PRO was the composite fatigue score, which was assessed in the PRO dataset using the FACIT-F questionnaire and measured at 6 and 12 months after randomization. The secondary PRO outcomes included:

• patient-reported gastrointestinal symptoms (nausea, vomiting, and diarrhea) at 6, 12, 18, and 24 months as measured by the EORTC QLQ-C30 scale

• composite fatigue score (FACIT-F questionnaire) measured at 6, 12, 18, and 24 months after randomization

• 2-item global quality of life score of the EORTC QLQ-C30 scale at 6, 12, 18, and 24 months

• different functional subscale scores as measured by the EORTC QLQ-C30 with particular interest in the emotional and physical subscales.

FACIT-F Questionnaire

The FACIT-F scale was originally developed for use in patients with cancer. It is a 13-item patient-reported scale of fatigue symptoms.⁴² Patients are asked to indicate to what extent each of the 13 items applies to them over the course of the previous 7 days. The response for each item is scored on a 5-point Likert scale from 0 (not at all) to 4 (very much). The total score of the FACIT-F scale is calculated by summing the scores of all 13 measurement items and ranges from 0 to 52, with a lower score indicating more fatigue.^43,44

The FACIT-F is a generally valid and reliable scale for cancers, including breast cancer. Yellen et al.⁴⁴ reported convergent–divergent validity (n = 50, including 12 patients with breast cancer with no staging information) through a strong correlation with the FACIT-F scale with the Profile of Mood States (POMS) fatigue scale (r = –0.83), the Piper Fatigue Scale (r = 0.61), and the POMS vigour scale (r = –0.77), indicating that the FACIT-F scale was significantly related to fatigue or anemia-relevant concepts (convergent validity) and was unrelated to concepts that were not assumed to be associated with fatigue or anemia (divergent validity) (if r > 0.35, or < –0.35, then significant at a significance level of 0.05). The FACIT-F scale was found to be strongly correlated with patient-rated ECOG Performance Status (r = –0.55).⁴² The FACIT-F scale was also able to discriminate patient membership classified by ECOG Performance Status (i.e., performance status of 0 versus 1 versus 2 or 3; P < 0.0001) under the assumption that a better performance status would be associated with higher quality of life scores.⁴⁴ In terms of reliability, the FACIT-F scale demonstrated strong internal consistency (Cronbach alpha = 0.93 to 0.96), and a good test–retest reliability over a 3- to 7-day window (r = 0.90) in patients with cancer.^42,44 No evidence of responsiveness to change was identified for the FACIT-F scale in patients with breast cancer.

Two studies estimated the MIDs for change in the FACIT-F scale in patients with cancer, including those with breast cancer.^45,46 However, neither study provided details on the breast cancer stage. The estimates of the MID were reported as 3.0 points (no less than 2.7 points) using the anchor-based and distribution-based approaches, meaning 1 category change (better or worse) on 3 of the 13 items (23%) on the FACIT-F scale with no change in the other 10 items.⁴⁵

EORTC QLQ C-30

The EORTC QLQ C-30 version 3.0 is 1 of the most-used PRO measures in oncology clinical trials. It is a multidimensional, cancer-specific, self-administered measure of HRQoL.⁴⁷ Version 3.0 of the questionnaire is the most recent version and has been in use since December 1997.⁴⁸ It is intended for use in the adult population only. The EORTC QLQ-C30 is composed of both multi-item scales and single-item measures. These include 5 functional scales (physical, role, cognitive, emotional, and social), 3 symptom scales (fatigue, pain, and nausea and vomiting), a global health status (HRQoL) scale, and 6 single items assessing additional symptoms commonly reported by patients with cancer (dyspnea, loss of appetite, insomnia, constipation, and diarrhea) as well as the perceived financial impact of the disease.⁴⁷ The EORTC QLQ-C30 uses a 1-week recall period to assess functional status and symptoms. All scales and single-item measures are scored from 0 to 100. Most questions have 4 response options (“not at all,” “a little,” “quite a bit,” “very much”), with scores on these items ranging from 1 to 4. For the 2 items that form the global HRQoL scale, the response format is a 7-point Likert scale with anchors at 1 = “very poor” and 7 = “excellent.” Raw scores for each scale are computed as the average of the items that contribute to a particular scale. Each raw scale score is converted to a standardized score that ranges from 0 to 100 using a linear transformation, with a higher score reflecting better function on the function scales and better HRQoL, while a higher score in the symptom scale means a higher burden of symptoms and therefore a worse health state.^49,50

The content validity was demonstrated by mapping the EORTC QLQ-C30 questionnaire with the WHO’s International Classification of Functioning framework based on the opinions of 21 experts.⁵¹ The discriminative validity of the psychosocial subscales of EORTC QLQ-C30 in patients with breast cancer (n = 150) was demonstrated through correlation with the ECOG Performance Status (Spearman rank correlation coefficient > 0.2).⁵² In terms of the interinstrument associations that occurred between the EORTC QLQ-C30 and POMS subscales (also known as convergent validity), the emotional function of the EORTC QLQ-C30 correlated very strongly with tension (Spearman rank correlation coefficient = 0.76), depression and dejection (Spearman rank correlation coefficient = 0.74) and the POMS total mood disturbance subscales (Spearman rank correlation coefficient = 0.74).⁵² A study investigated the inter-rater reliability among patients with metastatic breast cancer (n = 46). The median kappa coefficient for agreement across the 30 items in the EORTC QLQ-C30 was 0.86 (range, 0.48 to 1.00), and there was almost perfect agreement (kappa coefficient 0.8 to 1.00) for 19 items and “substantial agreement” (kappa coefficient, 0.61 to 0.80) for 6 items.

The estimates of the MID were evaluated in patients with advanced breast cancer (n = 723) using the anchor-based and distribution-based approaches and ranged from 5 to 14 points for within-group improvements and from −14 to −4 points for within-group deterioration, and from 4 to 11 points for between-group improvements and −18 to −4 points for between-group deterioration.⁵³ Another study estimated the MIDs, using the anchor-based approach for any scale of the EORTC QLQ-C30 questionnaire, to be 10 points in patients with breast cancer (n = 246) and small-cell lung cancer (n = 111).⁵⁴ Kawahara et al.⁵⁵ indicated that the estimated MIDs in patients with metastatic breast cancer using the anchor-based approach ranged from 7 to 15 points for within-group improvements, −17 to −7 points for within-group deterioration, 5 to 11 points for between-group improvements, and –8 to −5 points for between-group deterioration.

Statistical Analysis

Sample Size Calculations

In the OlympiA trial, at least 330 IDFS events in the ITT population were required to achieve a 90% power at a 2-sided 5% significance level. Data on patients with high-risk breast cancer with gBRCA mutations are limited; therefore, some assumptions were made about the expected hazard rates in the study population to allow the estimation of the sample size. At the primary analysis, a true HR of 0.7 between the olaparib and placebo arms was chosen with a critical HR value of 0.805. It has been estimated that the 3-year incidence of IDFS is approximately 60% for patients without a complete pathological response who have previously received neoadjuvant therapy, and 77% for patients who have previously received adjuvant therapy. Therefore, assuming approximately 50% of patients will receive prior neoadjuvant therapy, the estimated overall 3-year IDFS rate for the total population is approximately 68%. Thus, the sponsor estimated that at least 1,800 patients needed to be enrolled, assuming a dropout rate of 8% for the first 12 months, and approximately 5 years to complete patient recruitment.

Analysis of Outcomes

Primary Outcome

IDFS was planned to be analyzed at the interim and primary IDFS analyses. The interim analysis was planned to be analyzed in both the ITT population with an estimated 254 events and in the supportive mature cohort population, which included the first 900 patients with 165 events anticipated. The inclusion of a mature cohort population at the interim analysis was intended to provide confidence that any observed treatment effect from the interim analysis in the ITT population is maintained with longer-term follow-up. Statistical significance was not required for the mature cohort population, and this analysis was considered supportive. Evidence of the superiority of IDFS could only be determined in the interim analysis if the results of both analyzed populations (ITT and mature cohort) show a sustained clinically relevant treatment effect. An independent data-monitoring committee evaluated both efficacy and safety data, and recommendations for the study were made based on the body of the evidence.

To set boundaries, a 1-sided significance level (i.e., 2.5%) was to be split between the final and interim analyses using a bespoke spending function, where a fixed significance level was assigned at the interim analysis, and the remaining significance level at the final analysis, taking into account the correlation.⁵⁶ A 2-sided significance level of 0.005 was assigned to the interim analysis, so a statistically significant result would need to be observed to declare superiority in the IDFS. A hierarchical testing strategy was utilized to control the overall type I error for multiple end points.⁵⁷ If the primary IDFS end point was statistically significant (using the full 2.5% 1-sided alpha), then key secondary outcomes (DDFS and OS) were tested based on the weighted proportion of the significance level (1-sided 2% for DDFS and 1-sided 0.5% for OS). This testing procedure continued through each of the key secondary end points until the end point failed to reach statistical significance, after which subsequent key secondary end points were considered exploratory (Figure 3).

IDFS was analyzed using a log-rank test stratified by the stratification factors at randomization, and the corresponding P value was reported as the primary analysis result (Table 11). In addition, IDFS was analyzed using a stratified Cox proportional hazard model using the Efron approach for handling ties. The stratification factors were hormone receptor status (ER- and/or PgR-positive, HER2- negative versus TNBC), prior chemotherapy (neoadjuvant versus adjuvant chemotherapy), and prior platinum use for breast cancer (yes versus no). The difference between IDFS rates for each arm was reported with 95% CIs estimated by the profile likelihood approach. A Kaplan-Meier plot for IDFS was presented by treatment arms. The proportional hazards assumption was assessed by inspecting plots of complementary log-log (time) versus log (time), and formally testing using the Grambsch-Therneau test based on scaled Schoenfeld residuals from a Cox model including treatment group as a factor.

Figure 3: Multiplicity Strategy for the OlympiA Trial

DDFS = distant disease–free survival; IDFS = invasive disease–free survival; OS = overall survival.

Source: Sponsor-submitted statistical analysis plan.⁵⁸

Table 11: Statistical Analysis of Efficacy End Points

EORTC QLQ-C30 = European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire Core 30; FACIT-F = Functional Assessment of Chronic Illness Therapy–Fatigue; gBRCAm = germline BRCA mutation; MMRM = mixed-model for repeated measures; NA = not applicable; QoL = quality of life.

Sources: Clinical Study Reports for OlympiA^18,19 and sponsor-submitted statistical analysis plan.⁵⁸

IDFS was defined as the time from randomization to date of first recurrence. For confirmed local recurrence, the earliest date of diagnosis based on objective findings was used to determine the time to event for the analysis. For confirmed distant recurrence, the date of the earliest radiological or imaging examination or cytological or histological assessment was used to determine the time to event for the analysis. If 2 recurrence events (local and distant) were reported within 2 months of each other, then this was referred to as a simultaneous event and was considered as a single event, but the date of recurrence was the earliest date of the 2 events. If 2 recurrence events were reported more than 2 months apart, then these were treated as 2 separate recurrences and the earliest event and corresponding event date were used to derive the IDFS. If a patient died during the study with no documented disease recurrence or invasive contralateral breast cancer, or invasive nonbreast secondary primary cancer, then the date of death was used to derive the IDFS. Patients who did not have an IDFS event at the time of analysis were censored at the date of their last disease assessment. Patients were censored at 0.5 days after randomization under the following 3 circumstances:

• patients who had an event before randomization

• patients who were identified as inadvertent randomizations

• patients who withdrew consent, received no treatment, and did not provide any follow-up data.

Sensitivity Analyses

The following sensitivity (exploratory) analyses were conducted for the primary end point:

• A stratified Cox regression analysis, based on all randomized patients confirmed by a central Myriad test to have a gBRCAm, to compare against the primary IDFS analysis result if the analysis population differs from the ITT population.

• Any patient mis-stratified in the randomization system was included in the primary analysis as stratified.

• If the results of ER and PgR status from the local and central laboratories differed in more than 5% of randomized patients, then the same model as described previously was performed using the central laboratory results.

• A “deviation bias” sensitivity analysis may be performed after excluding patients with important protocol deviations that may affect the efficacy of the trial therapy.

• An unadjusted analysis was performed and compared with the primary results.

Subgroup Analyses

Subgroup analyses of IDFS in the ITT population were performed for the prespecified subgroups. Of the subgroups listed in the CADTH review protocol, the following were prespecified in the OlympiA trial:

• BRCA mutation status: BRCA1, BRCA2, or both.

• prior chemotherapy: neoadjuvant or adjuvant.

• prior platinum therapy: yes or no.

• hormone receptor status: ER- and/or PgR-positive and HER2-negative, or TNBC.

• axillary nodal status: node-negative, or node-positive.

• CPS&EG score: 2, 3, or 4 versus 5 or 6 (for the postneoadjuvant group only).

There was no multiplicity control for the subgroup analyses. As such, all subgroup analyses were exploratory in nature. Forest plots were created including the HRs and 95% CI and interaction P values for treatment-by-subgroup interactions.

Secondary Outcomes

Both key secondary outcomes, DDFS and OS, were analyzed in the ITT population. OS and DDFS were analyzed at the time of the IDFS analysis using the same methodology and model as for the IDFS analysis. Similar to IDFS, a bespoke spending function was applied to the OS and DDFS, where a fixed significance level was assigned to each analysis time point and to the remaining significance level assigned to the final analysis time point controlling the overall type I error. If the null hypothesis is rejected for DDFS and not for OS, then the following significance levels were recycled to OS from DDFS: IDFS, 99.5%; DDFS, 99.375%; and OS, 99%. If the null hypothesis is rejected for OS and not for DDFS, then the significant levels recycled to DDFS from OS were as follows: IDFS, 99.5%; DDFS, 99.375%; and OS, 99.0%.

For the analysis of OS, any patient not known to have died at the time of analysis was censored based on the last recorded date on which the patient was known to be alive. The final OS analysis will be conducted once the trial follow-up is complete.

For the analysis of DDFS, the date of the earliest radiological, cytological, or histological assessment was used to determine the time to event for the analysis. Patients without documented evidence of DDFS at the time of the analysis were censored at the date of their last clinical examination.

The incidence of contralateral invasive breast cancer, contralateral noninvasive breast cancer, new primary ovarian cancer, new primary fallopian tube cancer, and new primary peritoneal cancer was compared between groups using a Fine-Gray competing risk analysis, adjusted for stratification factors. Death was considered a competing risk for each end point. In addition, bilateral mastectomy after randomization was a competing risk for the contralateral breast cancer analyses, bilateral oophorectomy after randomization was considered a competing risk for the new primary ovarian cancers analysis, and removal of the fallopian tubes was considered a competing risk for the new primary fallopian tube cancer analysis.

Patient-Reported Outcomes

PROs were tested without adjustments for multiplicity and analyzed only in the PRO analysis set, which included patients who started treatment and provided evaluable data at baseline. The composite fatigue score (FACIT-F) measured at 6 and 12 months after randomization was compared between the 2 arms using a mixed-model for repeated measures (MMRM) analysis adjusted for time, treatment by time interaction, corresponding baseline score, and the baseline score by time interaction. Separate analyses were performed for patients who received neoadjuvant and adjuvant chemotherapy. To further assess the effect of olaparib on fatigue score over time, the composite fatigue score (FACIT-F questionnaire) measured at 6, 12, 18, and 24 months after randomization was analyzed using an MMRM model of all the postbaseline scores. Gastrointestinal symptoms, the 2-item global health status (quality of life) score measured at 3, 12, 18, and 24 months and the scores from the 5 multi-item EORTC QLQ-C30 functional scales were analyzed using the same MMRM model.

Missing PRO Data

If less than 50% of the subscale items were missing, the subscale score based on the nonmissing items was divided by the number of nonmissing items and multiplied by the total number of items on the subscale. If at least 50% of the items were missing, that subscale was treated as missing.

Harms

In the OlympiA trial, AEs were graded according to CTCAE version 4.03. Safety analyses were based on the safety population and included patients who had received at least 1 dose of the trial medication. All safety end points were reported using descriptive statistics. Separate summaries were provided for most notable safety end points, including myelodysplastic syndrome, acute myeloid leukemia, new primary cancers, and pneumonitis.

Analysis Populations

The FAS, also known as the ITT population (N = 1,836), consisted of all randomized patients (Table 12). The study was powered based on this population. Patients were analyzed according to their randomized group. Unless otherwise specified, all clinical efficacy end points were summarized and analyzed using the FAS.

The mature cohort (N = 900) included the first 900 randomized patients only. The analysis based on the mature cohort was regarded as supportive.

The PRO analysis set (N = 1,751) consisted of patients who consented to participate in the PRO assessment and who started treatment and provided evaluable FACIT-F or EORTC QLQ-C30 data at baseline. Baseline was defined as the last result on or before the first day of study treatment.

The safety analysis set or safety population (N = 1,815) consisted of all patients who received at least 1 dose of the randomized study drug.

Table 12: Analysis Sets

Note: The mature cohort consists of the first 900 randomized patients.

Sources: Clinical Study Reports for OlympiA.^18,19

Results

Patient Disposition

In the OlympiA trial, the first patient was randomized to study treatment on June 5, 2014, and the last patient was randomized on May 28, 2019. A total of 14,387 participants were screened; of these, 12,551 (87.2%) did not pass screening. The main reason for screening failures was not meeting eligibility criteria, predominantly because the patient did not have a deleterious or suspected deleterious BRCA mutation (95.0%) in screening part 1.

At interim analysis 1 (March 27, 2020), a total of 1,836 patients were successfully randomized (34 patients from Canada) in a 1:1 ratio to receive either olaparib (n = 921) or matching placebo (n = 915). A total of 294 patients (16.0%) discontinued from the study, 1,815 patients (98.9%) received study treatment, and 1,353 patients (73.7%) completed study treatment. A total of 423 patients (23.0%) discontinued study treatment, including 25.6% in the olaparib group and 20.4% in the placebo group (Table 36).

At interim analysis 2 (July 12, 2021), a total of 350 patients (19.1%) discontinued from the study, mainly due to death (18.0% and 20.1% in the olaparib and placebo groups, respectively), and patient decision (7.6% and 5.8% in the olaparib and placebo groups, respectively). A total of 1,815 patients received study medication, and 1,389 patients (75.7%) completed study treatment (Table 13). A total of 237 patients (25.7%) in the olaparib group and 189 patients (20.7%) in the placebo group discontinued study treatment. The most frequently reported reasons for discontinuation were AEs (10.6% and 4.7% in the olaparib and placebo groups, respectively), recurrence of disease (4.3% and 8.7% in the olaparib and placebo groups, respectively), and patient decision to stop study drug (6.4% and 3.3% in the olaparib and placebo groups, respectively).

Data on gBRCAm status for this study were obtained locally from medical records and/or a central Myriad laboratory. In terms of patient disposition, the subset of patients with confirmed Myriad gBRCAm status was similar to the overall study population (Table 38).

Table 13: Patient Disposition at Interim Analysis 2 — FAS

FAS = full analysis set.

^aJuly 12, 2021, date cut-off.

^bDoes not include patients who did not receive treatment.

Source: Clinical Study Reports for OlympiA.¹⁹

Patients Affected by the COVID-19 Pandemic

About █ ██ ███████ patients, including █████ in the olaparib group and █████ in the placebo group, had at least 1 visit affected by the COVID-19 pandemic. According to the sponsor, none of the protocol deviations related to the COVID-19 pandemic were considered significant and raised no concerns about the conduct of the study.

Exposure to Study Treatments

At interim analysis 1 (March 27, 2020), the median actual exposure to treatment was 350.0 days (range, 1 to 420) in the olaparib group, and 358.0 days (range, 2 to 404) in the placebo group (Table 38). The median treatment duration was 338.0 days (range, 1 to 420) in the olaparib group and 358.0 days (range, 2 to 404) in the placebo group.

At interim analysis 2 (July 12, 2021), the median intended duration of treatment exposure was 364.0 days (range, 1 to 492 days) in the olaparib group and 365.0 days (range, 2 to 414 days) in the placebo group (Table 14). The median actual exposure to treatment was similar for the olaparib and placebo groups (350.0 days (range, 1 to 420 days) and 359.0 days (range, 2 to 404 days), respectively) in the safety study population. The median actual treatment duration was 341.0 days (range, 1 to 420 days) in the olaparib group and 358.0 days (range, 2 to 404 days) in the placebo group.

A total of 228 patients (25.0%) in the olaparib group required a dose reduction compared with 47 patients (5.2%) in the placebo group, with AEs being the main reason for the dose reduction (24.4% and 3.9% in the olaparib and placebo groups, respectively) (Table 14). Dose interruptions lasting at least 3 days occurred in 405 patients (44.5%) in the olaparib group and 279 patients (30.9%) in the placebo group. The most frequently reported reasons for dose interruptions were AEs (36.2% and 14.2% in the olaparib and placebo groups, respectively), and surgery (24.8% and 26.3% in the olaparib and placebo groups, respectively). Treatment adherence was assessed based on the amount of study drug returned at the end of the treatment period.

Prior Therapy for Primary Breast Cancer

A total of 926 patients (50.4%) received antineoplastic therapy, including taxanes (49.6% and 49.1% in the olaparib and placebo groups, respectively), anthracyclines (47.7% and 46.2% in the olaparib and placebo groups, respectively), and nitrogen mustard analogues (48.0% and 47.2% in the olaparib and placebo groups, respectively). A total of 291 patients (89.5%) received prior and/or concurrent hormone therapy, including 86.9% of patients in the olaparib group and 92.4% in the placebo group. Most patients underwent a nonconservative breast surgery before randomization, including 755 patients (42.2%) who had previously received neoadjuvant chemotherapy, and 667 patients (36.3%) who had previously received adjuvant chemotherapy.

Table 14: Treatment Exposure at Interim Analysis 2 — Safety Population

SD = standard deviation.

Note: July 12, 2021, data cut-off date.

Patients with partial treatment end dates were excluded.

Dose interruptions and reductions are based on investigator-initiated decisions; interruptions and reductions due to patient noncompliance are omitted.

^aTotal intended exposure (days) = last dose date minus first dose date plus 1.

^bActual treatment exposure (days) = intended exposure minus total duration of dose interruptions, where intended exposure was calculated as per previous footnote.

^cNumber of days on 300 mg olaparib or placebo twice daily (actual exposure for the assigned starting dose).

Sources: Clinical Study Reports for OlympiA.^18,19

Concomitant Therapy

Reported concomitant medication use was generally balanced between treatment arms. A total of 788 patients (85.6%) in the olaparib group and 762 patients (83.3%) in the placebo group received at least 1 concomitant medication initiated on or after the randomization date. The most commonly used classes of drugs were: propionic acid derivatives (18.3% and 18.6% in the olaparib and placebo groups, respectively), proton pump inhibitors (16.4% and 14.8% in the olaparib and placebo groups, respectively), propulsives (16.2% and 5.5% in the olaparib and placebo groups, respectively), antibacterials for systemic use (28.9% and 29.2% in the olaparib and placebo groups, respectively), paracetamol (20.8% and 20.5% in the olaparib and placebo groups, respectively), and ibuprofen (11.5% in both olaparib and placebo groups). A total of 5 patients (0.5%) in the olaparib group and 4 patients (0.4%) in the placebo group received disallowed concomitant medications, including antineoplastic drugs (0.4% and 0.2% in the olaparib and placebo groups, respectively), and radiotherapy (0.1% and 0.2% in the olaparib and placebo groups, respectively).

Subsequent Post-Event Therapy

In total, 258 patients (14.1%) received post-event treatments for breast cancer recurrence, new primary, contralateral invasive breast cancer malignancy, or nonprimary end point malignancies, including systemic therapy (10.9% and 17.3% in the olaparib and placebo groups, respectively), post-event radiotherapy (4.2% and 7.7% in the olaparib and placebo groups, respectively), and post-event surgery (5.8% and 8.6% in the olaparib and placebo groups, respectively). The most common subsequent systemic therapies received included platinum-containing regimens (4.7% and 8.4% in the olaparib and placebo groups, respectively), pyrimidine analogues (4.1% and 6.6% in the olaparib and placebo groups, respectively), taxanes (4.6% and 4.8% in the olaparib and placebo groups, respectively), and PARP inhibitors (4.7% and 5.2% in the olaparib and placebo groups, respectively).

Protocol Deviations

In total, 251 patients (13.7%) had at least 1 important protocol deviation, with similar frequencies across the treatment groups: 129 patients (14.0%) in the olaparib arm, and 122 patients (13.3%) in the placebo arm (Table 15). The most common protocol deviations included not fulfilling criteria for high-risk disease (2.7% and 1.5% in the olaparib and placebo arms, respectively), and no staging or insufficient staging as patients with metastatic breast cancer should have been excluded from the trial according to the exclusion criteria (7.3% in both olaparib and placebo arms). The important protocol deviations were identified and classified before database lock in a blinded manner.

Efficacy

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

Overall Survival

At interim analysis 1 (March 27, 2020), the OS data were 7.9% mature. The median follow-up was 2.3 years in the olaparib group and 2.5 years in the placebo group (Table 16). In the FAS, median OS was not estimable, and the stratified HR was 0.68 (99% CI, 0.44 to 1.05; P = 0.0236). In the FAS, deaths were reported in 59 patients (6.4%) in the olaparib group and 86 patients (9.4%) in the placebo group. The proportion of patients who were alive at 3 years was 92.0% (95% CI, 89.6% to 93.9%) in the olaparib group and 88.3% (95% CI, 85.4% to 90.7%) in the placebo group. The Kaplan-Meier plot of OS for the FAS is presented in Figure 4.

Table 15: Protocol Deviations — FAS

FAS = full analysis set; GCP = good clinical practice.

^aThe same patient may have had more than 1 important protocol deviation.

Sources: Clinical Study Reports for OlympiA.^18,19

At interim analysis 2 (July 12, 2021), the OS data were 10.0% mature. The median follow-up was 3.5 years in the olaparib group, and 3.6 years in the placebo group (Table 16). In the FAS, deaths were reported in 75 patients (8.1%) in the olaparib group and 109 patients (11.9%) in the placebo group. The median OS was not estimable, and the stratified HR was 0.68 (98.5% CI, 0.47 to 0.97; P = 0.0091) in favour of the olaparib group. The proportion of patients who were alive at 4 years was 89.8% (95% CI, 87.2% to 91.9%) in the olaparib group and 86.4% (95% CI, 83.6% to 88.7%) in the placebo group (difference = 3.4%; 95% CI, –0.1% to 6.8%).¹⁷ The Kaplan-Meier plot of IDFS for the FAS is presented in Figure 5. The sensitivity and subgroup analyses were consistent with the primary OS analysis (Table 45 and Table 46).

Table 16: Overall Survival — FAS

CI = confidence interval; ER = estrogen receptor; FAS = full analysis set; HER2 = human epidermal growth factor receptor 2; HR = hazard ratio; KM = Kaplan-Meier; NA = not applicable; OS = overall survival; PgR = progesterone receptor; TNBC = triple-negative breast cancer.

^aData cut-off date: March 27, 2020.

^bData cut-off date: July 12, 2021.

^cEstimate of the treatment HR was based on the stratified Cox proportional hazards model. Stratification factors were the same as those used in the stratified log-rank test.

^dInferential, according to the alpha spending rules for the interim analysis. The CI for the HR was estimated using the profile likelihood approach.

^eExploratory, not inferential.

^fP value from a stratified log-rank test. Stratification was by chemotherapy type (adjuvant vs. neoadjuvant), hormone receptor status (ER- and/or PgR-positive and HER2-negative vs. TNBC), and prior platinum therapy (yes vs. no). Stratification factors were based on the categories used in the randomization system and were chosen by the pooling strategy. Once the pooling strategy was applied, only the hormone receptor status stratification factor was selected. A 2-sided significance level of 0.01 was assigned to interim analysis 1. A 2-sided significance level of 0.015 was assigned to interim analysis 2.

^gAny patient not known to have died at the time of analysis will be censored based on the last recorded date on which the patient was known to be alive.

^hThese randomized patients will be counted in the FAS; however, they will be treated as being censored for the OS event just after randomization. Censoring these patients at day 0.5 will not affect the log-rank test; however, they will be counted in the total number on day 1 and in the KM plot. The reason for censoring at day 0.5 is to avoid ties with other patients censored on day 1.

^IPercentage of patients (95% CI) alive from the KM estimates; the 95% CIs were calculated using the Greenwood formula.

Sources: Clinical Study Reports for OlympiA.^18,19

Figure 4: Kaplan-Meier Estimates of OS at Interim Analysis 1 — FAS

Data cut-off date: March 27, 2020.

OS = overall survival; FAS = full analysis set.

Source: Clinical Study Report for OlympiA.¹⁸

Figure 5: Kaplan-Meier Estimates of OS at Interim Analysis 2 — FAS

Note: July 12, 2021, data cut-off date.

OS = overall survival; FAS = full analysis set.

Source: Clinical Study Report for OlympiA.¹⁹

Invasive Disease–Free Survival

At interim analysis 1 (March 27, 2020), a total of 284 patients had an IDFS event (the data reached 15.5% maturity), including 106 patients (11.5%) in the olaparib group, and 178 patients (19.5%) in the placebo group. In the mature cohort that included the first 900 randomized patients, 65 patients (14.5%) in the olaparib group and 104 patients (23.1%) in the placebo group had an IDFS event (Table 40).

At interim analysis 2 (July 12, 2021), a total of 341 patients had an IDFS event (the data reached 18.6% maturity), including 134 patients (14.5%) in the olaparib group, and 207 patients (22.6%) in the placebo group. The most common type of event was a distant recurrence (Table 17).

Table 17: Summary of Type of First IDFS Event — FAS

CNS = central nervous system; FAS = full analysis et; IDFS = invasive disease–free survival.

^aData cut-off date: March 27, 2020.

^bData cut-off date: July 12, 2021.

Sources: Clinical Study Reports for OlympiA.^18,19

Table 18: Invasive Disease–Free Survival — FAS

CI = confidence interval; ER = estrogen receptor; FAS = full analysis set; HER2 = human epidermal growth factor receptor 2; HR = hazard ratio; IDFS = invasive disease–free survival; KM = Kaplan-Meier; NA = not applicable, PgR = progesterone receptor; TNBC = triple-negative breast cancer.

^aData cut-off date: March 27, 2020.

^bData cut-off date: July 12, 2021.

^cEstimate of the treatment HR was based on the stratified Cox proportional hazards model. Stratification factors were the same as those used in the stratified log-rank test. The CI for the HR was estimated using the profile likelihood approach.

^dInferential, according to the alpha spending rules for the interim analysis.

^eExploratory, not inferential.

^fP value from a stratified log-rank test. Stratification was by chemotherapy type (adjuvant vs. neoadjuvant), hormone receptor status (ER- and/or PgR-positive and HER2-negative vs. TNBC), and prior platinum therapy (yes vs. no). Stratification factors were based on the categories used in the randomization system and were chosen by the pooling strategy. Once the pooling strategy was applied, only the hormone receptor status stratification factor was selected. A 2-sided significance level of 0.005 was assigned to IDFS at interim analysis 1.

^gMedian clinical follow-up was calculated using the reverse censoring method.

^hPatients who have not had a recorded IDFS event at the time of the analysis will be censored at the date of their last disease evaluation. Disease evaluation includes mammogram and/or breast MRI (MRI preferred for patients younger than 50 years), other radiological or other imaging or clinical examination (e.g., physical exam).

^IThese randomized patients will be counted in the FAS; however, they will be treated as being censored for the IDFS event just after randomization. Censoring these patients at day 0.5 will not affect the log-rank test; however, they will be counted in the total number at risk and in the KM plot. The reason for censoring at day 0.5 is to avoid ties with other patients censored on day 1.

^jPercentage (95% CI) of patients who were free from invasive disease from the KM estimates and the 95% CIs were calculated using the Greenwood formula.

The mature cohort includes the first 900 patients randomized.

Sources: Clinical Study Reports for OlympiA.^18,19

At interim analysis 1 (March 27, 2020), the median follow-up was 2.3 years in the olaparib group and 2.5 years in the placebo group (Table 18), and the median IDFS was not estimable in either treatment arm. The stratified HR for invasive disease recurrence or death was 0.58 (99.5% CI, 0.41 to 0.82; P = 0.0000073) in favour of the olaparib group. The proportion of patients who remained invasive disease–free at 3 years was 85.9% (95% CI, 82.8% to 88.4%) in the olaparib group and 77.1% (95% CI, 73.7% to 800.12%) in the placebo group. The IDFS results in the FAS were consistent with the IDFS analysis performed using a mature cohort with a median follow-up of 3.5 years in both treatment groups (stratified HR = 0.61; 99.5% CI, 0.39 to 0.95) (Table 40). After reviewing the data for interim analysis 1, an independent data-monitoring committee concluded that the prespecified statistical threshold for the superiority of olaparib over placebo for IDFS was reached at a maturity of 15.5%; thus, this analysis constitutes the primary IDFS analysis for the OlympiA trial. The Kaplan-Meier plot of IDFS for the FAS is presented in Figure 6.

Subgroup Analyses

Primary end point subgroup analyses at interim analysis 1 are presented in Table 19. The treatment effects for all prespecified subgroups in the CADTH protocol were consistent with the main effect.

Figure 6: Kaplan-Meier Estimates of IDFS at Interim Analysis 1 — FAS

FAS = full analysis set; IDFS = invasive disease–free survival.

Note: March 27, 2020, data cut-off date.

Source: Clinical Study Reports for OlympiA.¹⁸

Table 19: Subgroup Analyses of IDFS at Interim Analysis 1 — FAS

BRCA = breast cancer susceptibility gene; CI = confidence interval; CPS&EG = clinical and post-treatment pathologic stage and estrogen receptor status and histologic grade; ER = estrogen receptor; FAS = full analysis set; HER2 = human epidermal growth factor receptor 2; HR = hazard ratio; HR-positive = positive for expression of a hormone receptor; IDFS = invasive disease–free survival; NA = not applicable; PgR = progesterone receptor; TNBC = triple-negative breast cancer.

^aData cut-off date: March 27, 2020.

^bThe Cox model included factors for treatment group, subgroup factor, and the treatment-by-subgroup interaction. All patients with nonmissing subgroup data were included in the model. The CI was calculated using a profile likelihood approach. These analyses are not inferential.

^cHR-positive is defined as ER-positive and/or PgR-positive.

^dTwo patients are excluded from the summary of the TNBC subset because they do not have confirmed HER2-negative status.

Source: Clinical Study Reports for OlympiA.¹⁸

Sensitivity Analyses

The sensitivity analyses for the primary IDFS end point at interim analysis 1 (Table 42), including patients with central pathology review data, unadjusted analysis, and interval censoring, as well as supportive analysis in patients with a Myriad-confirmed gBRCAm (Table 41), were consistent with the primary IDFS analysis.

At interim analysis 2 (July 12, 2021), the median follow-up was 3.5 years in the olaparib group and 3.6 years in the placebo group (Table 18). This was an exploratory analysis, as no alpha was reserved for IDFS since the superiority of olaparib over placebo for IDFS was demonstrated at interim analysis 1. The median IDFS was not estimable, and the stratified HR for invasive disease recurrence or death was 0.63 (95% CI, 0.50 to 0.78) in favour of the olaparib group. The proportion of patients who remained invasive disease–free at 4 years was 82.7% (95% CI, 79.6% to 85.4%) in the olaparib group and 75.4% (95% CI, 72.2% to 78.3%) in the placebo group (difference = 7.3%; 95% CI, 3.0% to 11.5%).¹⁷ The Kaplan-Meier plot of IDFS for the FAS is presented in Figure 7. The subgroup analyses across prespecified subgroups and sensitivity analyses were consistent with the primary IDFS analysis for interim analysis 2.

Figure 7: Kaplan-Meier Estimates of IDFS at Interim Analysis 2 — FAS

Note: July 12, 2021, data cut-off date.

IDFS = invasive disease–free survival; FAS = full analysis set.

Source: Clinical Study Reports for OlympiA.¹⁹

Distant Disease–Free Survival

At interim analysis 1 (March 27, 2020), the median follow-up was 2.3 years in the olaparib group and 2.5 years in the placebo group (Table 20). In the FAS, 89 patients (9.7%) in the olaparib group and 152 patients (16.6%) in the placebo group had a DDFS event. The median DDFS was not estimable in either arm, and the stratified HR for distant disease recurrence or death was 0.57 (99.5% CI, 0.39 to 0.83; P = 0.0000257) in favour of the olaparib group. The proportion of patients who remained distant disease–free at 3 years was 87.5% (95% CI, 84.6% to 89.9%) in the olaparib group and 80.4% (95% CI, 77.2% to 83.3%) in the placebo group. The Kaplan-Meier plot of DDFS for the FAS is presented in Figure 8.

Table 20: Distant Disease–Free Survival — FAS

CI = confidence interval; DDFS = distant disease–free survival; ER = estrogen receptor; FAS = full analysis set; HER2 = human epidermal growth factor receptor 2; HR = hazard ratio; IDFS = invasive disease–free survival; KM = Kaplan-Meier; NA = not applicable; NR = not reported; PgR = progesterone receptor; TNBC = triple-negative breast cancer.

^aData cut-off date: March 27, 2020.

^bData cut-off date: July 12, 2021.

^cInferential, according to the alpha spending rules for the interim analysis. The CI for the HR was estimated using the profile likelihood approach.

^dEstimate of the treatment HR was based on the stratified Cox proportional hazards model. Stratification factors were the same as those used in the stratified log-rank test.

^eExploratory, not inferential.

^fP value from a stratified log-rank test. Stratification was by chemotherapy type (adjuvant vs. neoadjuvant), hormone receptor status (ER- and/or PgR-positive and HER2-negative vs. TNBC), and prior platinum therapy (yes vs. no). Stratification factors were based on the categories used in the randomization system and were chosen by the pooling strategy. Once the pooling strategy was applied, only the hormone receptor status stratification factor was selected. A 2-sided significance level of 0.005 was assigned to the interim analysis of DDFS.

^gPatients who have not had a recorded DDFS event at the time of the analysis will be censored at the date of their last clinical evaluation (i.e., last physical exam or radiological evaluation).

^hThese randomized patients will be counted in the FAS; however, they will be treated as being censored for the DDFS event just after randomization. Censoring these patients at day 0.5 will not affect the log-rank test, however, they will be counted in the total number at risk and in the KM plot.

^IPercentage (95% CI) of patients who were distant disease–free from the KM estimates and the 95% CIs were calculated using the Greenwood formula.

Sources: Clinical Study Reports for OlympiA.^18,19

Figure 8: Kaplan-Meier Estimates of DDFS at Interim Analysis 1 — FAS

DDFS = distant disease–free survival; FAS = full analysis set.

Note: March 27, 2020, data cut-off date.

Source: Clinical Study Reports for OlympiA.¹⁸

Subgroup Analyses

Subgroup analyses for the DDFS end point for interim analysis 1 in the OlympiA trial are presented in Table 45. The treatment effects for all prespecified subgroups in the CADTH protocol were consistent with the main effect.

Sensitivity Analyses

The sensitivity analyses for the DDFS end point at interim analysis 1 (Table 43), including patients with central pathology review data, unadjusted analysis, and interval censoring, as well as the supportive analysis in patients with a Myriad-confirmed gBRCAm (Table 42), were consistent with the primary analysis of DDFS.

At interim analysis 2 (July 12, 2021), the median follow-up was 3.5 years in the olaparib group and 3.6 years in the placebo group (Table 20). This was an exploratory analysis, as no alpha was reserved for DDFS since the superiority of olaparib over placebo for DDFS had been demonstrated at interim analysis 1. In the FAS, 107 (11.6%) patients in the olaparib group and 172 patients (18.8%) in the placebo group had a DDFS event. The stratified HR for distant disease recurrence or death was 0.61 (95% CI, 0.48 to 0.77). The proportion of patients who remained distant disease–free at 4 years was 86.5% (95% CI, 83.8% to 88.8%) in the olaparib group and 79.1% (95% CI, 76.0% to 81.8%) in the placebo group (difference = 7.4%; 95% CI, 3.6% to 11.3%).¹⁷ The Kaplan-Meier plot of IDFS for the FAS is presented in Figure 9.

Figure 9: Kaplan-Meier Estimates of DDFS at Interim Analysis 2 — FAS

DDFS = distant disease–free survival; FAS = full analysis set.

Note: July 12, 2021, data cut-off date.

Source: Clinical Study Report for OlympiA.¹⁹

Health-Related Quality of Life

In the OlympiA trial, HRQoL data were assessed only in the PRO analysis set using the FACIT-F or EORTC QLQ-C30 questionnaires. The PRO set comprised patients who consented to participate in HRQoL assessment and started treatment, and the analyses were based only on a PRO subset with evaluable baseline data for FACIT-F or EORTC QLQ-C30. For both the FACIT-F and EORTC QLQ-C30 questionnaires, compliance rates were similar between treatment groups at baseline and follow-up in the placebo group compared with the olaparib group, and compliance diminished over time in both groups (to less than 70% after 24 months) (Table 21). Two separate analyses were conducted to compare differences between treatment groups at follow-up in FACIT-F and EORTC QLQ-C30 scores among patients who had previously received neoadjuvant versus those who had received postsurgical adjuvant chemotherapy, as the timing of chemotherapy can have a potential impact on the HRQoL outcome.

Table 21: Compliance Rates for HRQoL Questionnaires — PRO

EORTC QLQ-30 = European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire Core 30; FACIT-F = Functional Assessment of Chronic Illness Therapy–Fatigue questionnaire; HRQoL = health-related quality of life; PRO = patient-reported outcome.

^aNumber of evaluable forms divided by the number of expected forms.

Source: Clinical Study Report for OlympiA.¹⁸

Functional Assessment of Chronic Illness Therapy–Fatigue

Using an MMRM analysis, no clinically meaningful differences were found between treatment groups in mean change in FACIT-F score at 6 and 12 months (less than the MID of 3 points⁴⁵). For the subgroup of patients who had previously received neoadjuvant chemotherapy (N = 727), the adjusted LS mean difference for olaparib versus placebo was –1.3 (–2.4 to –0.2; P = 0.024) at 6 months and –1.5 (–2.8 to –0.2; P = 0.025) at 12 months. For the subgroup of patients who had previously received adjuvant chemotherapy (N = 780), the adjusted LS mean difference for olaparib versus placebo was –1.3 (–2.3 to –0.2; P = 0.017) at 6 months and –1.3 (–2.4 to 0.1; P = 0.027) at 12 months (Table 22, Figure 10, and Figure 11).

Table 22: Change in FACIT-F Score From Baseline at 6 and 12 Months — PRO

CI = confidence interval; FACIT-F = Functional Assessment of Chronic Illness Therapy–Fatigue; LS = least squares; MMRM = mixed-model for repeated measures; PRO = patient-reported outcome.

^aOnly patients with an evaluable baseline form were included.

^bThe adjusted LS mean changes, P values (2-sided), and 95% CIs were obtained from an MMRM analysis of the change from baseline. The model included treatment, time and treatment by time interaction, corresponding baseline score, and the baseline score by time interaction. The difference was the values for olaparib minus placebo.

^CThe P value was not adjusted for multiplicity.

Source: Clinical Study Report for OlympiA.¹⁸

Figure 10: Change From Neoadjuvant Baseline for FACIT-F Score — PRO

CI = confidence interval; FACIT-F = Functional Assessment of Chronic Illness Therapy–Fatigue questionnaire; LS = least squares; MMRM = mixed-model for repeated measures; PRO = patient-reported outcome.

Note: Adjusted LS mean changes and 95% CI are obtained from an MMRM analysis of the change from baseline. The model includes treatment, time and treatment by time interaction, corresponding baseline score, and the baseline score by time interaction.

Source: Clinical Study Report for OlympiA.¹⁸

Figure 11: Change From Adjuvant Baseline for FACIT-F Score — PRO

CI = confidence interval; FACIT-F = Functional Assessment of Chronic Illness Therapy–Fatigue questionnaire; LS = least squares; MMRM = mixed-model for repeated measures; PRO = patient-reported outcome.

Note: Adjusted LS mean changes and 95% CIs were obtained from an MMRM analysis of the change from baseline. The model includes treatment, time and treatment by time interaction, corresponding baseline score, and the baseline score by time interaction.

Source: Clinical Study Report for OlympiA.¹⁸

EORTC QLQ-C30

Patients Who Had Previously Received Neoadjuvant Chemotherapy: The adjusted LS mean difference for olaparib versus placebo in the global health status score was █ █ ██ █ █ █ █ ██ ██ █ █ ██ ███████ at 6 months, █ █ ██ █ █ █ █ ██ ██ █ █ ██ ██████) at 12 months, █ █ ██ █ █ █ █ ██ ██ █ █ ██ ██████) at 18 months, and █ █ ██ █ █ █ █ ██ ██ █ █ █ ██ ██████) at 24 months (Table 23). The adjusted LS mean difference between treatment groups in the nausea and vomiting symptom scale was 6.0 (95% CI, 4.0 to 8.0; █ ██ ██████ at 6 months, 6.3 (95% CI, 4.4 to 8.2 █ ██ █████) at 12 months, 0.4 (95% CI, –1.2 to 1.9; █ ██████) at 18 months, and 1.4 (95% CI, –0.4 to 3.3; █ ██████) at 24 months. The adjusted LS mean difference between treatment groups in the diarrhea symptom scale was 0.3 (95% CI, –2.0 to 2.7; █ ██████) at 6 months, 2.0 (95% CI, –1.0 to 4.9; █ ██████) at 12 months, 1.1 (95% CI, –2.0 to 4.3 █ ██ █████) at 18 months, and 1.8 (95% CI, –1.5 to 5.0; █ ██████) at 24 months.

Patients Who Had Previously Received Adjuvant Chemotherapy: The adjusted LS mean difference for olaparib versus placebo in the global health status score was █ █ ██ █ █ █ █ ██ ██ █ █ █ ██ ██████) at 6 months █ █ █ ██ █ █ █ █ ██ ██ █ █ █ ██ ███████ at 12 months, █ █ ██ █ █ █ █ ██ ██ █ █ ██ ███████ at 18 months, and █ █ ██ █ █ █ █ ██ ██ █ █ ██ ███████ at 24 months (Table 23). The adjusted LS mean difference between treatment groups in the nausea and vomiting symptom scale was 5.3 (95% CI, 3.4 to 7.2; █ ██ █████) at 6 months, 4.5 (95% CI, 2.8 to 6.2; █ ██ █████) at 12 months, –0.3 (95% CI, –1.9 to 1.3; █ ██████) at 18 months, and –0.6 (95% CI, –2.5 to 1.2; █ ███████ at 24 months. The adjusted LS mean difference between treatment groups in the diarrhea symptom scale was –1.7 (95% CI, –4.1 to 0.7; █ ███████ at 6 months, 0.1 (95% CI, –2.2 to 2.4; █ ███████ at 12 months, 0.4 (95% CI, –1.9 to 2.7; █ ██████) at 18 months, and –1.0 (95% CI, –3.4 to 1.4; █ ██████) at 24 months.

Incidence of Contralateral Breast Cancer and New Primary Ovarian, Fallopian Tube, and Peritoneal Cancer

The incidences of contralateral breast cancers (invasive and noninvasive), new primary ovarian cancer, and new primary invasive nonbreast nonovarian cancers in the olaparib and placebo groups are shown in Table 24. The competing risks analysis showed that the stratified HR for the incidence of contralateral invasive breast cancer was 0.97 (95% CI, 0.52 to 1.79; P = 0.900) (Table 24).

Table 23: Change From Baseline for EORTC QLQ-C30 Subscale Scores — PRO

CI = confidence interval; EORTC QLQ-30 = European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire Core 30; LS = least squares; MMRM = mixed-model for repeated measures; PRO = patient-reported outcome; QoL = quality of life; SD = standard deviation.

Note: Only patients with an evaluable baseline form were included.

Adjusted LS mean changes, P values (2-sided), and 95% CIs were obtained from an MMRM analysis of the change from baseline. The model included treatment, time and treatment by time interaction, corresponding baseline score, and the baseline score by time interaction. The difference was the values for olaparib minus placebo.

^aP value was not adjusted for multiplicity.

Source: Clinical Study Report for OlympiA.¹⁸

Table 24: Summary of Cancers Occurring Post Randomization at Interim Analysis 2 — FAS

FAS = full analysis set.

Note: Summary of cancers without considering competing risks.

^aData cut-off date: July 12, 2021.

^bIncludes new primary ovarian, fallopian, and peritoneal cancers, without considering competing risks.

^cOne instance of ovarian cancer recurrence (patient E104618) was captured in the database.

Source: Clinical Study Report for OlympiA.¹⁹

Progression-Free Survival

Progression-free survival was not measured or reported in the OlympiA trial.

Time to Progression

Time to progression was not measured or reported in the OlympiA trial.

Breast Cancer Symptoms

Breast cancer symptoms were not measured or reported in the OlympiA trial.

Need for Surgery

Need for surgery was not measured or reported in the OlympiA trial.

Table 25: Contralateral Invasive Breast Cancer at Interim Analysis 2 — FAS

CI = confidence interval; ER = estrogen receptor; FAS = full analysis set; HER2 = human epidermal growth factor receptor 2; HR = hazard ratio; PgR = progesterone receptor; TNBC = triple-negative breast cancer.

^aData cut-off date: July 12, 2021.

^bEstimate of the treatment HR based on the cause-specific hazards from a stratified Cox proportional hazards model (accounting for competing risks); < 1 indicates a lower risk with olaparib arm compared with placebo arm. Stratification is by chemotherapy type (adjuvant vs. neoadjuvant), hormone receptor status (ER- and/or PgR-positive and HER2-negative vs. TNBC), and prior platinum therapy (yes vs. no). Stratifications was based on the categories used in the randomization system and are chosen by the pooling strategy.

^cP value calculated from the Fine-Gray model adjusted for treatment group and stratification factors. The P value was not adjusted for multiplicity.

^dThese randomized patients will be counted in the FAS; however, they will be treated as being censored for the contralateral invasive breast cancer event just after randomization. Censoring these patients at day 0.5 will not affect the log-rank test; however, they will be counted in the total number at risk and in the Kaplan-Meier plot. The reason for censoring at day 0.5 is to avoid ties with other patients censored on day 1.

Source: Clinical Study Report for OlympiA.¹⁹

Harms

Only those harms identified in the review protocol are reported subsequently.

Adverse Events

At interim analysis 2 (July 12, 2021), 836 patients (91.8%) in the olaparib group and 758 patients (83.8%) in the placebo group experienced at least 1 AE (Table 26). Common AEs, AEs with a CTCAE grade of 3 or higher, and the preferred term for the AE are summarized in Table 27. A total of 736 patients (80.8%) in the olaparib group and 480 patients (53.1%) in the placebo group experienced at least 1 TRAE (Table 28). The most common TRAEs occurring in the olaparib and placebo groups were anemia (20.6% and 1.7%, respectively), diarrhea (12.0% and 7.5%, respectively), decreased neutrophil count (14.9% and 4.6%, respectively), and decreased white blood cell count (14.1% and 4.5%, respectively). The majority of TRAEs were manageable with supportive care and/or dose modifications and consistent with the known safety profile for olaparib.

Table 26: Summary of Adverse Events — Safety Population

AE = adverse event; CTCAE = Common Terminology Criteria for Adverse Events; SAE = serious adverse event; TRAE = treatment-related adverse event.

^aPatients with multiple events in the same category were counted only once in that category; patients with events in more than 1 category were counted once in each of the categories.

^bAs assessed by the investigator.

^cIncludes AEs that led to a dose interruption or reduction and did not lead to the permanent discontinuation of the study treatment.

Source: Clinical Study Report for OlympiA.¹⁸

Table 27: Adverse Events Reported in at Least 5% of Patients in Either Treatment Arm — Safety Population