First-Line Rucaparib Maintenance Improves PFS in Ovarian Cancer, Regardless of HRD Status

Article

The ATHENA-MONO trial showed a significant improvement in progression-free survival when patients with platinum-sensitive ovarian cancer were treated with maintenance rucaparib vs placebo in the first-line setting, regardless of HRD status.

Rucaparib (Rubraca) significantly improved progression-free survival (PFS) when used as a first-line maintenance for patients with ovarian cancer who responded to platinum-based chemotherapy, regardless of homologous recombination deficiency (HRD) status, according to results of the ATHENA-MONO trial (NCT03522246), presented at the 2022 American Society of Clinical Oncology (ASCO) Annual Meeting.

Results in the intent-to-treat (ITT) population showed a median PFS of 20.2 months (95% CI, 15.2-24.7) in the rucaparib arm and 9.2 months (95% CI, 8.3-12.2) in the placebo arm (HR, 0.52; 95% CI, 0.40-0.68; log-rank P <.0001). The secondary end point of PFS by blinded-independent central radiology review (BICR) showed a median that was 25.9 months (95% CI, 16.8- NR) in the rucaparib arm and 9.1 months (95% CI, 6.4-9.7) in the placebo arm (HR, 0.47; 95% CI, 0.36-0.63; log-rank P <.0001).

“Patients with measurable disease at baseline have further tumor reduction with rucaparib. [Additionally], rucaparib safety profile is consistent with prior studies,” Bradley J. Monk, MD, FACS, FACOG, professor in the Division of Gynecologic Oncology at the University of Arizona College of Medicine, medical director of the Gynecologic Program at US Oncology Research Network, and lead investigator of this study, said during the presentation.

Overall, the ATHENA trial was divided into 2 study analyses after randomization based on treatment regimens. Patients who had newly diagnosed stage III to IV high-grade epithelial ovarian, fallopian tube, or primary peritoneal cancer and who complete frontline platinum-doublet chemotherapy were randomized 4:4:1:1 to 1 of 4 treatment arms. Patients must have achieved investigator-assessed complete response (CR) or partial response (PR), received cytoreductive surgery, had an ECOG performance status of 0 or 1, and had no prior treatments for ovarian cancer.

In ATHENA, patients were randomized to either arm A with oral rucaparib at 600 mg twice daily plus intravenous (IV) nivolumab (Opdivo) at 480 mg; arm B with IV placebo and rucaparib; arm C with oral placebo and nivolumab; or arm D comprised of an intravenous and oral placebo regimen. Patients were treated for 24 months or until radiographic progression, unacceptable toxicity, or other reasons for discontinuation. ATHENA-MONO comprised arms B (n = 400) and D (n = 100), and were reported at 2022 ASCO. ATHENA-COMBO will include arms A (n = 400) and B (n = 400). The primary end point for the overall trial is investigator-assessed PFS.

“Arm C which, is nivolumab only, will not be analyzed. In ATHENA-MONO, rucaparib is the experimental arm, and in ATHENA-COMBO, the combination of rucaparib plus nivolumab is the experimental arm, and rucaparib plus placebo is the control arm,” Monk said.

A hierarchical step-down design was developed for the primary end point, which first evaluated PFS by HRD status in those with BRCA mutant and loss of heterozygosity (LOH)–high, BRCA wild-type disease. Given the regimen met statistical significance in this group, then the study would “step down” to an ITT analysis. At a 90% power at a 2-sided significance level of 0.025, intended HR values were 0.45 for those with HRD and 0.60 for those in the ITT population.

In the HRD group, investigator-assessed median PFS was 28.7 months (95% CI, 23.0-not reached) vs 11.3 months (95% CI, 9.1-22.1; HR, 0.47; 95% CI, 0.31-0.72; P = .0004). Median PFS by BICR was NR (95% CI, 28.7-NR) vs 9.9 months (95% CI, 6.5-NR; HR, 0.44; 95% CI, 0.28-0.70; P = .0004).

In the investigator-assessed PFS exploratory subgroup analysis, rucaparib continued to demonstrate increased benefit compared with the placebo, regardless of the presence of a BRCA mutation (HR, 0.40; 95% CI, 0.21-0.75), LOH-high disease (HR, 0.58; 95% CI, 0.33-1.01), or HRD negativity (HR, 0.65; 95% CI, 0.45-0.95). The data were similar in subgroups evaluated by BICR.

The investigator-assessed objective response rate (ORR) in the HRD population was 58.8% (95% CI, 32.9% vs 81.6%) in the rucaparib arm vs 20.0% (95% CI, 0.5%-71.6%) in the placebo arm, both comprised exclusively of PRs, with median duration of response (DOR) at 16.7 months vs 5.5 months, respectively. In the ITT population, the ORR in the rucaparib arm was 48.8% (95% CI, 32.9%- 64.9%) vs 9.1% (95% CI, 0.2%-41.3%) in the placebo arm, with 1 CR in the rucaparib group and the rest of responses being PRs. Corresponding median DOR between groups was 22.1 months and 5.5 months.

At least 1 grade 3 or higher adverse effect (AE) was reported for 60.5% of patients in the rucaparib group vs 22.7% in the placebo group, with treatment interruptions or dose reductions from treatment-emergent AEs (TEAEs) occurring in 63.8% and 21.8%, respectively. The most common grade 3 or higher AEs occurring with rucaparib vs placebo were anemia or increased hemoglobin (28.7% vs 0%, respectively), neutropenia or neutrophil count decrease (14.6% vs 0.9%), and increased alanine aminotransferase or aspartate aminotransferase (10.6% vs 0.9%).

Patients did not experience significant changes in bilirubin or increased levels of drug-induced liver toxicity. Additionally, through month 12, 70% of patients continued to receive at least 500 mg of rucaparib twice daily, which was more than 80% of the starting dose.

Of the 425 patients treated in the rucaparib arm vs 110 in the placebo, 12.4% vs 9.9%, respectively, were still on treatment at the data cutoff of March 23, 2022, and 63.5% vs 80.2% discontinued treatment before 2 years because of disease progression (41.0% vs 64.9%), AEs (12.6% vs 5.4%), withdrawal of consent (4.9% vs 2.7%), clinical progression (3.3% vs 5.4%), ot other reasons (1.6% vs 1.8%). The median treatment duration was 14.7 months (95% CI, 0.1-32.7) in the rucaparib group and 9.9 months (95% CI, 0.9-25.9) in the placebo, with a median duration of follow-up of 26.1 months (95% CI, 25.8-26.9) vs 26.2 (95% CI, 24.0-27.7).

In the HRD population, patient characteristics included a median age of 57 years in the rucaparib arm vs 59 years in the placebo group, White race in 74.1% and 71.4%, ECOG performance status of 0 in 71.4% and 79.6%, and stage III disease in 73.5% and 63.3%, respectively. Types of cancer included epithelial ovarian (82.7% vs 79.6%), fallopian tube (11.4% vs 10.2%), and primary peritoneal (5.9% vs 10.2%). Primary surgery was performed in 56.2% of patients in the rucaparib arm and 55.1% in the placebo arm, 74.1% vs 71.4% had no residual disease, 50.8% vs 51.0% were BRCA wild-type/LOH-high, and 9.2% vs 10.2% had measurable disease at baseline.

In the ITT populations, the median age in the rucaparib and placebo arms was 61 years, White race in 76.8% vs 78.4%, an ECOG performance score of 0 in 69.1% vs 68.5%, and stage III disease in 75.6% vs 70.3%, respectively. Types of cancer included epithelial ovarian (78.7% vs 76.6%), fallopian tube (11.7% vs 16.2%), and primary peritoneal (9.6% vs 7.2%). Primary surgery was performed in 48.9% of patients in the rucaparib arm vs 48.6% in the placebo and no residual disease was found in 75.4% vs 73.9%. Patients in the rucaparib vs placebo arms were had HRD-positive, BRCA wild-type/LOH-high status at rates of 22.0% vs 22.5%, HRD negativity in 44.3% vs 44.1%, with HRD unknown in 12.4% vs 11.7%. Additionally, measurable disease at baseline was found in 9.6% of patients in the rucaparib arm and 9.9% in the placebo arm.

Reference

Monk BJ, Parkinson C, Lim MC, et al. ATHENA–MONO (GOG-3020/ENGOT-ov45): A randomized, double-blind, phase 3 trial evaluating rucaparib monotherapy versus placebo as maintenance treatment following response to first-line platinum-based chemotherapy in ovarian cancer. J Clin Oncol. 2022;40(suppl 17):LBA5500. doi: 10.1200/JCO.2022.40.17_suppl.LBA5500

Recent Videos
Medical use of AI increases every day, and in the future, will be exponentially greater and many forms of treatment will be improved, according to Russell C. Langan, MD, FACS, FSSO.
Shubham Pant, MD, MBBS, highlights an “exciting time” in the treatment of patients with RAS-mutated pancreatic cancer.
Greater direct access to academic oncologists may help address challenges associated with a lack of CAR T education in the community setting.
A computational linguistics model designed to locate pancreatic cysts that started to locate pancreatic cancer has the potential to lead to more efficient treatment.
Brett L. Ecker, MD, discusses the importance of multidisciplinary collaboration in improving patient outcomes in neuroendocrine tumors.
A computational linguistics model has mitigated disparities in surveillance of the pancreas that primarily affected racial and ethnic minorities.
Brett L. Ecker, MD, focused on the use of de-escalation therapy, which is gaining momentum in neuroendocrine tumors.
Certain bridging therapies and abundant steroid use may complicate the T-cell collection process during CAR T therapy.
Related Content