Maintenance Pembrolizumab/Olaparib Falls Short in Locally Advanced, Metastatic TNBC

Maintenance Pembrolizumab/Olaparib Falls Short in Locally Advanced, Metastatic TNBC

Adding olaparib (Lynparza) to pembrolizumab (Keytruda) did not lead to improvements in progression-free or overall survival (OS) compared with pembrolizumab plus chemotherapy in patients with locally recurrent inoperable or metastatic triple-negative breast cancer (TNBC) who previously received induction pembrolizumab plus chemotherapy.

Findings from the phase 2 KEYLYNK-009 study (NCT04191135) were presented during the 2023 San Antonio Breast Cancer Symposium.¹

In the intention-to-treat (ITT) population, results showed that the progression-free survival (PFS) per RECIST v1.1 criteria by blinded independent central review (BICR) was 5.5 months (95% CI, 4.2-8.3) with pembrolizumab/olaparib (n = 135) compared with 5.6 months (95% CI, 4.3-6.9) with pembrolizumab/chemotherapy (n = 136; HR, 0.98; 95% CI, 0.72-1.33; P = .4556).

The 6-month PFS rates were 47.8% (95% CI, 38.5%-56.5%) with pembrolizumab plus olaparib and 45.8% (95% CI, 36.8%-54.4%) with pembrolizumab plus chemotherapy; the 1-year PFS rates were 33.3% (95% CI, 24.5%-42.3%) and 29.3% (95% CI, 21.2%-37.8%), respectively.

The median estimated OS in the ITT population with pembrolizumab/olaparib was 25.1 months (95% CI, 18.3-not reached [NR]) vs 23.4 months (95% CI, 15.8-NR) with pembrolizumab plus chemotherapy (HR, 0.95; 95% CI, 0.64-1.40). The 1-year and 18-month OS rates with pembrolizumab plus olaparib were 76.4% (95% CI, 68.2%-82.8%) and 62.0% (95% CI, 51.9%-70.6%); with pembrolizumab plus chemotherapy, these rates were 72.6% (95% CI, 64.0%-79.4%) and 55.7% (45.5%-64.7%), respectively.

However, there was a lower incidence of treatment-related adverse effects (TRAEs) with pembrolizumab plus olaparib vs pembrolizumab plus chemotherapy.

Lead author Hope S. Rugo, MD, FASCO, noted that the median time of induction chemotherapy was 4.2 months, which is not reflected in the PFS and OS data “because the timing started at randomization.”

“[For] patients who are receiving chemotherapy and pembrolizumab in the metastatic setting who have responding or stable disease, continuing with pembrolizumab alone after the best response to chemotherapy might be an effective strategy with less toxicity for our patients,” said Rugo, who is a professor of medicine in the Division of Hematology and Oncology, director of Breast Oncology, and director of Clinical Trials Education at the University of California San Francisco Helen Diller Family Comprehensive Cancer Center, in an oral presentation during the meeting.

Previously, in the phase 3 KEYNOTE-355 study (NCT02819518), pembrolizumab plus chemotherapy showed statistically significant improvements in PFS and OS in patients with previously untreated, locally recurrent inoperable or metastatic TNBC and PD-L1 combined positive score (CPS) of 10 or higher.

However, effective regimens are needed in the setting of locally recurrent inoperable or metastatic disease. Rugo mentioned that preclinical2,3 and phase 14 data suggest that maintenance combination therapy with a PD-L1 inhibitor and a PARP inhibitor could clinically benefit patients in the locally advanced/metastatic setting.

In the phase 2 KEYLYNK-009 trial, investigators tested the efficacy and safety of maintenance therapy with pembrolizumab and olaparib vs pembrolizumab with chemotherapy in patients with locally recurrent inoperable or metastatic TNBC who had clinical benefit from induction frontline pembrolizumab/platinum-based chemotherapy.

To be eligible for enrollment, patients had to have locally recurrent inoperable or metastatic TNBC that was not previously treated in the metastatic setting, measurable disease per RECIST v1.1 criteria by local radiology review, who had a 6-month or longer interval between treatment with curative intent and recurrence and confirmed PD-L1 status.

“But PD-L1 positivity was not required for entrance into the study,” Rugo, who is also a 2021 Giant of Cancer Care® inductee in education, said.

A total 460 patients started induction treatment with carboplatin at area under the curve (AUC) 2 on days 1 and 8 and gemcitabine at 1000 mg/m² on days 1 and 8 of each 21-day cycle, plus pembrolizumab at 200 mg every 3 weeks for 4 to 6 cycles. In the ITT population, 271 patients were then randomized 1:1 to receive post-induction olaparib at 300 mg twice daily plus pembrolizumab at 200 mg every 3 weeks for up to 35 cycles including induction or carboplatin at AUC 2 on days 1 and 8 and gemcitabine at 1000 mg/m² on days 1 and 8 of each 21-day cycle, plus pembrolizumab at 200 mg every 3 weeks for up to 35 cycles including induction.

A total of 189 patients were not randomized due to progressive disease (n = 143), AEs (n = 25), and other reasons (n = 21).

Stratification factors included induction response (complete response [CR] or partial response [PR] vs stable disease), PD-L1 status (CPS ≥1 vs <1), and genomic tumor status (BRCA mutant vs BRCA wild-type).

The primary end points were PFS per RECIST v1.1 criteria via BICR and OS, both in the ITT population. Secondary outcome measures were PFS and OS in the population of patients with a PD-L1 CPS of 10 or higher and the BRCA-mutant subgroup. Safety was also evaluated.

The median follow-up was 17.2 months in both the pembrolizumab/olaparib (range, 8.8-29.9) and pembrolizumab/chemotherapy (range, 9.1-29.5) arms. One and 2 patients, respectively, completed all 35 cycles of the pembrolizumab in each arm; 35 and 29 patients, respectively, were still receiving treatment.

Regarding baseline characteristics, the median age was 53 years (range, 25-82), 34.4% of patients had an ECOG performance status of 1, and 48.0% of patients had a PD-L1 CPS of 10 or greater. A total 21.8% of patients had a tumor BRCA mutation, and 59.1% of patients had a homologous recombination deficiency score of 33 or higher. Most patients had recurrent metastatic disease (67.5%), and 70.5% of patients had either a CR or PR at the time of randomization.

Additional data showed that when evaluated in the PD-L1 CPS of 10 or greater population, the median PFS was similar to that observed in the ITT population. The median PFS was 5.7 months (95% CI, 2.9-13.9) with pembrolizumab/olaparib and 5.7 months (95% CI, 3.8-7.6) with pembrolizumab/chemotherapy (HR, 0.92; 95% CI, 0.59-1.43).

However, patients in the tumor BRCA-mutant subgroup showed a trend toward improved PFS with pembrolizumab/olaparib vs pembrolizumab/chemotherapy, at 12.4 months (95% CI, 8.3-NR) and 8.4 months (95% CI, 5.4-NR), respectively (HR, 0.70; 95% CI, 0.33-1.48).

“It’s important to keep in mind that there were only 59 patients in this subgroup, but if you look at the landmark analyses, you also see a numerically greater percent of patients who are free from progression at 6 and 12 months,” Rugo added. The 6-month PFS rates were 84.4% (95% CI, 63.6%-93.9%) and 61.1% (95% CI, 40.8%-76.2%) with pembrolizumab/olaparib and pembrolizumab/chemotherapy, respectively; the 12-month PFS rates were 52.2% (95% CI, 30.0%-70.4%) and 45.1% (95% CI, 25.9%-62.6%), respectively.

The median OS was NR with both regimens in the PD-L1 CPS of 10 or greater population, but the hazard ratio was 0.97 (95% CI, 0.53-1.76). In the tumor BRCA-mutant populations, the median OS was NR (95% CI, 17.1-NR) with pembrolizumab/olaparib compared with 23.4 months (95% CI, 17.3-NR) with pembrolizumab/chemotherapy (HR, 0.81; 95% CI, 0.28-2.37). One-year and 18-month OS rates with pembrolizumab/olaparib were 96.6% (95% CI, 77.9%-99.5%) and 73.3% (95% CI, 45.9%-88.4%), respectively; these rates were 82.9% (95% CI, 63.7%-92.5%) and 70.4% (95% CI, 45.5%-85.5%), respectively, in the pembrolizumab/chemotherapy arms.

Regarding safety in the as-treated population, any-grade TRAEs occurred in 84.4% and 96.2% of patients on pembrolizumab/olaparib and pembrolizumab/chemotherapy, respectively; grade 3 to 5 TRAEs were reported in 32.6% and 68.4% of patients, respectively. A total of 8.9% and 19.5% of patients, respectively, experienced TRAEs that led to treatment discontinuation.

Additionally, any-grade and grade 3/4 immune-mediated AEs and infusion reactions occurred in 19.3% and 4.4% of patients on pembrolizumab plus olaparib vs 23.3% and 4.5% of those on pembrolizumab/chemotherapy. No patients on the pembrolizumab/olaparib arm had immune-mediated AEs or infusion reactions that led to treatment discontinuation compared with 4 on the pembrolizumab/olaparib arm.

Major differences in AE profiles between the olaparib and chemotherapy arms were more hematologic-related toxicities and infusion reactions with chemotherapy, Rugo concluded.

Editor’s Note: Dr Rugo disclosed that Institutional research support was received by AstraZeneca, Daiichi Sankyo, F. Hoffmann-La Roche AG/Genentech, Gilead Sciences, GlaxoSmithKline, Lilly, Merck & Co., Inc., Novartis Pharmaceuticals, OBI Pharma, Pfizer, Pionyr Immunotherapeutics, and Sermonix Pharmaceuticals. Dr Rugo also disclosed serving in a consultancy or advisory role for Puma, NAPO, Blueprint, and Daiichi Sankyo. Research funding was provided by Merck Sharp & Dohme LLC.

References

1. Rugo HS, Robson M, Im S-A, et al. Pembrolizumab + olaparib vs pembrolizumab + chemotherapy after induction with pembrolizumab + chemotherapy for locally recurrent inoperable or metastatic TNBC: randomized open-label phase 2 KEYLYNK-009 study. Presented at: 2023 San Antonio Breast Cancer Symposium; December 5-9, 2023; San Antonio, TX. Abstract GS01-05.
2. Jiao S, Xia W, Yamaguchi H, et al. PARP inhibitor upregulates PD-L1 expression and enhances cancer-associated immunosuppression. Clin Cancer Res. 2017;23(14):3711-3720. doi:10.1158/1078-0432.CCR-16-3215
3. Wang Z, Sun K, Xiao Y, et al. Niraparib activates interferon signaling and potentiates anti-PD-1 antibody efficacy in tumor models. Sci Rep. 2019;9(1):1853. doi:10.1038/s41598-019-38534-6
4. Domchek SM, Postel-Vinay S, Im S-A, et al. Olaparib and durvalumab in patients with germline BRCA-mutated metastatic breast cancer (MEDIOLA): an open-label, multicentre, phase 1/2, basket study. Lancet Oncol. 2020;21(9):1155-1164. doi:10.1016/S1470-2045(20)30324-7