The reporting of this OPAL-C study protocol (version 2.0; Supplement C, Version 1.0, dated 13 September 2021) conforms to the Standard Protocol Items: Recommendations for Interventional Trials (SPIRIT) guidelines [11]. The order of the SPIRIT checklist items has been modified to group similar items together.

Trial design

OPAL (NCT03574779) is a phase 1B/2 multicohort proof-of-concept umbrella study designed to evaluate the safety and efficacy of novel treatments and/or combinations of treatments in patients with ovarian cancer. Objectives are cohort-specific, and the design of each cohort is tailored to the patient population and treatment of interest. At the time of this publication, OPAL has 2 ongoing cohorts, cohort A and cohort C. Cohort A (OPAL-A) focuses on evaluating the efficacy and safety of the dostarlimab, bevacizumab, and niraparib combination in patients with recurrent platinum-resistant ovarian cancer who have received 1 to 2 prior lines of therapy. Notably, patients in cohort A must not have received prior treatment with PARP inhibitors. Cohort B was canceled before activation.

The global, multicenter, randomized, open-label, phase 2 OPAL-C cohort is focused on patients with newly diagnosed, advanced ovarian cancer (high-grade nonmucinous epithelial ovarian, fallopian tube, or peritoneal cancer, collectively referred to as ovarian cancer) with homologous recombination–deficient tumors who are eligible for neoadjuvant platinum-taxane doublet chemotherapy followed by IDS. Figure 1 and Table 1 show the OPAL-C schematic and study schedule, respectively. The estimated study duration for each patient will be up to 48 months.

Fig. 1

Study schematic. The study duration for each patient will be up to 48 months. aAn estimated sample size of 40 patients in each of the 2 arms (80 patients total) is needed for the half-width of a 2-sided 80% CI for the pre-IDS unconfirmed overall response rate difference of 12%. This estimate assumes that the pre-IDS unconfirmed overall response rate in the neoadjuvant niraparib arm is 65% and that the pre-IDS unconfirmed overall response rate difference between arms is 20%. bPatients with a complete or partial response after neoadjuvant therapy per RECIST v1.1 go on to receive IDS. Patients with stable disease or disease progression may proceed to an alternative therapy at the investigator’s discretion. cDefined as the percentage of patients with unconfirmed complete or partial response on study treatment before IDS as assessed by the investigator using the RECIST v1.1 criteria. dThird cycle optional. eUp to 36 months in the absence of progressive disease, death, unacceptable toxicity, or patient/physician decision to withdraw from the study. CI, confidence interval; IDS, interval debulking surgery; RECIST v1.1, Response Evaluation Criteria for Solid Tumors version 1.1

Table 1 Schedule for enrolment, interventions, and assessments (SPIRIT figure)Participants

Table 2 shows the key eligibility criteria. Adult patients with stage III/IV ovarian cancer according to International Federation of Gynecology and Obstetrics staging guidelines and an Eastern Cooperative Oncology Group performance status ≤ 2 may enroll. All patients must have homologous recombination–deficient tumors as determined by central testing using the myChoice CDx PLUS assay (Myriad Genetics, Salt Lake City, UT, USA). Eligible patients with germline BRCA1/2 deleterious or presumed deleterious mutations by an authorized test (e.g., BRACAnalysis CDx; Myriad Genetics) will need central homologous recombination deficiency testing using the myChoice CDx PLUS assay to confirm eligibility. Before study entry, patients must have completed 1 run-in cycle of carboplatin-paclitaxel. In addition to the exclusion criteria given in Table 2, patients who are immunocompromised or who have an autoimmune disease that has required systemic treatment in the last 2 years are ineligible.

Table 2 Key patient eligibility criteriaScreening

All patients will undergo baseline assessments and central homologous recombination deficiency tumor tissue testing using the myChoice CDx PLUS assay during a 5-week prescreening period. All patients will receive one run-in cycle of carboplatin-paclitaxel per local standard of care because of the urgency of initiating therapy and the time required to receive central homologous recombination deficiency testing results. Patients with confirmed homologous recombination–deficient tumors who meet all eligibility criteria and have signed informed consent forms will be enrolled in the study after recovery from the run-in cycle of carboplatin-paclitaxel.

Randomization and blinding

Patients will be randomly assigned in a 1:1 ratio to receive three 21-day cycles of neoadjuvant platinum-taxane doublet chemotherapy (carboplatin-paclitaxel, arm 1) or neoadjuvant niraparib (arm 2) approximately 3 weeks after the initial run-in cycle of carboplatin-paclitaxel treatment. Randomization with a block size of four will be performed using an interactive voice/web response technology system. Randomization will be stratified based on BRCA mutational status. As an open-label trial, patients will not be blinded to study treatment. To mitigate the risk of introducing bias into the assessment of treatment effect, clinical and statistical team members involved in the study will remain blinded to treatment-sensitive data until database freeze. Before the database freeze, no data aggregation (efficacy, safety, or pharmacokinetics) by treatment arm will be performed, except for prespecified interim analyses. The study will be monitored by the Data Review Committee, and only the Data Review Committee will review the aggregate data from the prespecified interim analysis.

Neoadjuvant therapy

Patients randomized to neoadjuvant niraparib treatment will receive an individualized starting dose based on body weight and platelet count at the time of screening. Patients with a body weight < 77 kg or screening platelet count < 150,000/μL will receive a niraparib dose of 200 mg; patients with a body weight ≥ 77 kg and screening platelet count ≥ 150,000/μL will receive a niraparib dose of 300 mg. Niraparib will be taken orally once daily throughout each 21-day treatment cycle.

Patients randomized to neoadjuvant platinum-taxane doublet chemotherapy will receive carboplatin-paclitaxel without bevacizumab per local standard of care. On day 1 of each cycle, carboplatin will be administered at the investigator’s prescribed dose, intravenously over 60 min, followed by paclitaxel at a dose of 175 mg/m2, administered intravenously over 180 min. Before each carboplatin-paclitaxel administration, patients must have an absolute neutrophil count ≥ 1500/μL (≥ 1000/μL when granulocyte colony–stimulating factor will be administered), platelet count ≥ 100,000/μL, and hemoglobin levels ≥ 9 g/dL. If a patient is unable to tolerate carboplatin or paclitaxel, they will receive cisplatin or docetaxel, respectively, and remain in the treatment arm for analysis.

Neoadjuvant treatment may be discontinued because of toxicity, disease progression per Response Evaluation Criteria in Solid Tumors version 1.1 (RECIST v1.1), or investigator/medical monitor decision. Clinical progression based on signs, symptoms, physical examination results, or increased cancer antigen 125 (CA-125) (> 15% from lowest recorded value) will not necessitate discontinuation of neoadjuvant treatment but may trigger an additional scan per RECIST v1.1. If patients do not meet stable disease or progression criteria per RECIST v1.1, but the investigator believes it is in the patient’s best interest to discontinue neoadjuvant treatment, neoadjuvant therapy may be discontinued after consultation with the study’s medical monitor; in such an event, the additional scan performed for this evaluation will be used for the primary efficacy analysis. Patients who receive additional cycles of platinum-taxane doublet chemotherapy before IDS will be considered as having discontinued treatment and will undergo end-of-treatment procedures and provide scans for assessment per RECIST v1.1 after the 3 cycles of neoadjuvant therapy.

IDS and pre-IDS assessment

IDS will be performed 3 to 6 weeks after the final cycle of neoadjuvant therapy. Before surgery, patients will undergo a pre-IDS assessment. Figure 2 shows the criteria used for selecting patients for IDS. Patients with an unconfirmed complete or partial response according to RECIST v1.1 will undergo IDS by a qualified gynecologic/surgical oncologist aimed at achieving no visible residual disease. Patients with stable disease per RECIST v1.1 may proceed to IDS or an alternative therapy at the investigator’s discretion. Patients with disease progression will exit the study treatment and proceed to an alternative therapy at the investigator’s discretion. Surgical and postsurgical outcomes will be collected. Postsurgical safety will be assessed according to local standard of care for 30 days (± 7 days) after IDS and before adjuvant chemotherapy.

Fig. 2

Criteria for selecting patients for IDS. IDS, interval debulking surgery; RECIST v1.1, Response Evaluation Criteria for Solid Tumors version 1.1

Adjuvant therapy

After a post-IDS recovery period of up to 6 weeks, all patients will receive up to three 21-day cycles of platinum-taxane doublet chemotherapy (and optional bevacizumab or biosimilar for patients deemed high risk by the investigator) regardless of neoadjuvant treatment arm assignment. The third treatment cycle will be optional. Platinum-taxane doublet chemotherapy will be administered identically to treatment arm 1 during neoadjuvant therapy. For patients deemed high risk by the investigator, 7.5 or 15 mg/kg of optional bevacizumab or biosimilar will be administered intravenously over 90 min. After adjuvant therapy and within 14 days before the first day of the first cycle of the maintenance treatment period, diagnostic imaging assessments per RECIST v1.1 will be performed.

Maintenance therapy

Patients without disease progression per RECIST v1.1 will start niraparib (and optional bevacizumab or biosimilar for patients deemed high risk by the investigator) maintenance treatment in 3-week cycles for up to 36 months. To enable patients to recover adequately from hematologic and nonhematologic toxicity, maintenance treatment can be delayed 6 to 9 weeks after the final adjuvant chemotherapy cycle.

Maintenance imaging assessments will be performed on the first day or within 14 days before the first day of the first maintenance cycle, then every 3 months (± 7 days) for 1 year after the first day of the first maintenance cycle, every 6 months (± 7 days) during the second year, and yearly thereafter until disease progression per RECIST v1.1, death, unacceptable toxicity, patient/physician withdrawal, or study completion. If a patient has increasing CA-125 levels or concerning symptoms, unscheduled computerized tomography or magnetic resonance imaging scans may be conducted per RECIST v1.1 criteria.

Treatment adjustment and discontinuation

Niraparib treatment can be interrupted because of adverse events for up to 14 and 21 days during neoadjuvant and maintenance treatment, respectively. Treatment will be interrupted for niraparib-related, nonhematologic, Common Terminology Criteria for Adverse Events grade ≥ 3 adverse events. Relation to study treatment will be determined by investigator assessment. Treatment interruption for hematologic adverse events will be determined based on prespecified blood counts. Additional file 1: Supplementary Table 1 shows the niraparib dose interruption and adjustment criteria for hematologic parameters.

If nonhematologic toxicity resolves to baseline or Common Terminology Criteria for Adverse Events grade ≤ 1 within 2 weeks for neoadjuvant therapy or within 3 weeks for maintenance treatment, patients may restart niraparib at a reduced dose if prophylaxis is not feasible. If the nonhematologic toxicity adverse event recurs at a similar or worse grade, treatment will be interrupted and a dose reduction made on resolution. Patients will discontinue niraparib treatment permanently if the toxicity requiring dose interruption does not resolve to baseline or Common Terminology Criteria for Adverse Events grade ≤ 1 or specified hematologic levels within the allowable period or if the patient has already had the maximum allowed 2 dose reductions.

Dose adjustments and supportive care for platinum-taxane doublet chemotherapy, bevacizumab, or biosimilar will follow guidelines specified in the prescribing information for each medication or local standard of care.

Concomitant medications

Any medication the patient takes other than the study treatment, including herbal and other nontraditional remedies, will be considered concomitant medications. For each patient in the study, a record will be kept of all concomitant medication that they are using or used, at screening and at each subsequent study visit. Permitted concomitant medications include contraceptives (only those prespecified in the protocol) and rescue medications for adverse events (as deemed necessary by the treating investigator according to local institutional practice and/or guidance in the appropriate prescribing information). During the treatment phase of the study, patients will be prohibited from receiving systemic anticancer or biological therapy, immunotherapy, chemotherapy (except for the chemotherapy treatments specified and approved for use in the study), hormonal therapy, radiation therapy, surgery (except for the prespecified IDS), live vaccines, and investigational agents other than niraparib and prophylactic cytokines (prohibited only in the first cycle of the study but allowed in subsequent cycles as per American Society of Clinical Oncology guidelines).

Outcomes

The primary endpoint is pre-IDS unconfirmed overall response rate, defined as the investigator-assessed percentage of patients with unconfirmed complete or partial response on study treatment before IDS per RECIST v1.1. Secondary endpoints were the incidence of patients with CA-125 progression per Gynecological Cancer InterGroup CA-125 response criteria, progression-free survival (time from randomization to earliest date of disease progression per RECIST v1.1 by investigator assessment or death by any cause), and progression-free survival rate at 12, 18, and 24 months (proportion of participants without documented disease progression per RECIST v1.1 per investigator assessment or death within 12, 18, or 24 months after randomization); overall survival (time from randomization to the date of death by any cause); and time to first subsequent treatment (date of treatment randomization to the date of first subsequent anticancer therapy or death). Patient-reported outcomes will also be assessed using the Patient Reported Outcomes version of the Common Terminology Criteria for Adverse Events (PRO-CTCAE), Functional Assessment of Cancer Therapy- General Population (FACT-GP5), European Organization for Research and Treatment of Cancer (EORTC) IL136, and the EORTC IL137. For the PRO-CTCAE, selected items (taste changes, nausea, vomiting, abdominal pain, bloating, constipation, diarrhea, concentration, memory, muscle pain, joint pain, fatigue, anxiety, neuropathy, and sadness) will be assessed weekly during neoadjuvant chemotherapy and at end of treatment. FACT-GP5, EORTC IL136, and EORTC IL137 assessments will be collected weekly during the neoadjuvant treatment period, every 3 cycles for the first 13 to 14 months, and then every 3 months thereafter during niraparib maintenance treatment. Exploratory endpoints include pathological complete response rate, identification of potential disease or treatment-related biomarkers, and evaluation of surgical and postsurgical outcomes. A neoadjuvant treatment modality response score may be investigated post hoc if needed.

Safety

Known risks associated with niraparib treatment include hematological adverse events, such as thrombocytopenia, and the development of myelodysplastic syndrome or acute myeloid leukemia. Both niraparib and bevacizumab have also been associated with an increased risk of hypertension. Within OPAL-C, safety will be evaluated based on the incidence and severity of treatment-emergent adverse events, serious adverse events, treatment discontinuations or dose delays or reductions due to adverse events, predefined adverse events of special interest (i.e., myelodysplastic syndrome, acute myeloid leukemia, and secondary cancer), changes in Eastern Cooperative Oncology Group performance status, changes in clinical laboratory results (hematology and chemistry), vital sign measurements, observations during symptom-directed physical examination, and use of concomitant medications. Events may be reported spontaneously or solicited through patient questioning and/or review of clinical data. All adverse events, regardless of causality, will be collected and recorded for each participant; events will be coded using the current version of the Medical Dictionary for Regulatory Activities coding system.

Statistical methodsSample size

A total of 84 patients are planned for randomization, such that an estimated sample size of 40 patients per treatment arm is evaluable for pre-IDS unconfirmed overall response rate. The sample size estimation was calculated using EAST software, version 6.5 (Cytel, Waltham, MA, USA). With 40 patients in each treatment arm, the half width of a 2-sided 80% confidence interval (CI) for the pre-IDS unconfirmed overall response rate difference was 12%. This estimate assumes that the pre-IDS unconfirmed overall response rate in the platinum-taxane neoadjuvant treatment arm is 65% and that the difference between treatment arms is 20%. Estimates for expected effect size and overall response rate were based on findings from the niraparib literature as well as internal GSK data.

Interim analysis

A single planned interim analysis for futility using the pre-IDS unconfirmed overall response rate will be performed when at least 30 patients (approximately 15 patients per treatment arm) have pre-IDS radiographic data to determine their responses. Per the prespecified boundaries, the study will be allowed to continue unchanged if there is a greater than 30% predictive probability that the lower limit of the 80% CI of the pre-IDS overall response rate difference is greater than 0 at the study’s end. The interim analysis will be performed by a contract research organization, with results reviewed by the independent Data Review Committee. Enrollment may be paused during the interim analysis.

Analysis populations

This study will involve 6 analysis populations. The intent-to-treat population will include all randomized patients, while the safety population will consist of all patients who received at least one dose of the study treatment. The efficacy population will include all patients in the safety population with measurable disease at baseline tumor assessment, defined as the presence of at least one target lesion. The response-evaluable population will include all patients in the efficacy population with at least one evaluable postbaseline tumor assessment. The biomarker population will comprise all patients with at least one follow-up tumor assessment and a tumor or blood sample. Lastly, the pharmacokinetic population will include all patients who received niraparib and have at least one measurable niraparib concentration.

Outcome assessment

The primary endpoint of pre-IDS unconfirmed overall response rate will be analyzed in the response-evaluable population. The 2-sided 80% CI for the pre-IDS unconfirmed overall response rate difference between the 2 treatment arms will be evaluated. For the secondary endpoint of CA-125 progression per Gynecological Cancer InterGroup CA-125 response criteria, incidences for each study period (i.e., neoadjuvant treatment, adjuvant treatment, and maintenance treatment) will be compared between treatment arms with the 95% CI calculated using the stratified Miettinen and Nurminen’s method (stratification factor, BRCA mutation status). Kaplan–Meier plots of the survival distribution function will be generated and the median, 25th, and 75th percentiles and corresponding 95% CI will be estimated for each treatment arm. Progression-free survival rates at 12, 18, and 24 will be compared across treatment arms. Progression-free survival, overall survival, and time to first subsequent treatment, the distribution for each treatment arm will be estimated using the Kaplan–Meier method and will be compared between the 2 treatment arms using log-rank test stratified by the stratification factor used for randomization (i.e., BRCA mutational status). Hazard ratios and corresponding 90% CI will be estimated from the Cox proportional hazard model.

PRO-CTCAE scores for each item attribute (frequency, severity, and/or interference) will be summarized by visit. The frequency and proportions of participants in each response category for the FACT-GP5 at each time point will be evaluated. For the EORTC IL136 and EORTC IL137, summary statistics for observed values and changes from baseline will be calculated for scales and items by treatment group at each time point.

Demographic and baseline clinical characteristics will be summarized: number and percentage for categorical variables and the number of patients, mean, standard deviation, median, minimum, and maximum for continuous variables.