Patients with previously untreated, locally recurrent, inoperable, or metastatic triple-negative breast cancer derived a statistically significant survival benefit following treatment with pembrolizumab and chemotherapy.
Treatment with pembrolizumab (Keytruda) and chemotherapy yielded a statistically significant and clinically meaningful survival benefit compared with chemotherapy alone in a population of patients with previously untreated, locally recurrent, inoperable, or metastatic triple-negative breast cancer (TNBC) with a PD-L1 combined positive score (CPS) of 10 or more, according to updated findings from the phase 3 KEYNOTE-355 trial (NCT02819518).1
Furthermore, a PD-L1 CPS of at least 10 was found to be a reasonable cut-off to define the population of patients expected to derive the most benefit from this regimen.
Results from the study, which were presented at the 2021 San Antonio Breast Cancer Symposium, showed that the combination of pembrolizumab and chemotherapy yielded a median overall survival (OS) of 23.0 months vs 16.1 months with chemotherapy alone in patients with a CPS of 10 or higher (HR, 0.73; 95% CI, 0.55-0.95; P = .0093). The 18-month OS rates were 58.3% with pembrolizumab vs 44.7% with placebo/chemotherapy.
Moreover, the median progression-free survival (PFS) elicited by the combination was 9.7 months vs 5.6 months in patients with a CPS of at least 10 (HR, 0.66; 95% CI, 0.50-0.88). The 12-month PFS rate was 39.1% with pembrolizumab and 23.0% with placebo/chemotherapy.
“These results provide further support for pembrolizumab in combination with chemotherapy as the new standard-of-care treatment regimen for patients with locally recurrent, unresectable or metastatic TNBC whose tumors express a PD-L1 CPS of 10 or more,” lead study author Javier Cortés, MD, PhD, the head of breast cancer and gynecological cancers at Hospital Universitario Ramón y Cajal, in Madrid, Spain, said in a presentation of the data.
Pembrolizumab is currently approved by the FDA for use in combination with chemotherapy for the treatment of patients with locally recurrent unresectable or metastatic TNBC whose tumors express PD-L1 (CPS ≥10) as determined by an FDA-approved test. The November 2020 approval was based on earlier findings of the KEYNOTE-355 trial.2
In the KEYNOTE-355 trial, the addition of pembrolizumab to chemotherapy resulted in a statistically significant and clinically meaningful improvement in both PFS and OS vs chemotherapy alone as a first-line treatment for patients with metastatic TNBC with a PD-L1 CPS of 10 or more. However, no statistically significant benefit in PFS or OS was previously observed in a subgroup of patients with PD-L1 CPS of 1 or higher, and the benefit of the regimen was not tested in the intent-to-treat (ITT) population due to the prespecified testing strategy.
In the trial, investigators enrolled patients 18 years or older with central determination of TNBC and PD-L1 expression. Moreover, patients had to have previously untreated, locally recurrent, inoperable, or metastatic disease; de novo metastasis or completion of treatment with curative intent within 6 months of first disease recurrence; and an ECOG performance status of 0 or 1. Patients’ life expectancy had to exceed 12 weeks, and adequate organ function was required.
Those with active central nervous system metastasis, active autoimmune disease, or who had been previously treated with systemic steroids were not eligible for enrollment.
Patients were randomized 2:1 to receive either intravenous (IV) pembrolizumab plus chemotherapy or placebo plus chemotherapy. Pembrolizumab was administered at a dose of 200 mg every 3 weeks. Chemotherapy regimens included:
Furthermore, patients were stratified by chemotherapy regimen (taxane vs gemcitabine/carboplatin), PD-L1 tumor expression (CPS ≥1 vs CPS <1), and prior treatment with the same class of chemotherapy in the neoadjuvant or adjuvant setting (yes vs no).
Primary end points for the study included PFS and OS in the PD-L1 CPS of 10 or higher, PD-L1 CPS of 1 or higher, and in the ITT populations. Secondary end points included objective response rate, duration of response, disease control rate, and safety in all treated patients.
In this analysis, investigators sought to assess outcomes in subgroups of patients by additional CPS cut-offs.
Of the 847 total patients enrolled on the study, 566 were randomized to the pembrolizumab group compared with 281 who were randomized to the placebo group.
A total 219 patients treated with pembrolizumab had a PD-L1 CPS of 10 or higher; 421 were treated and had a PD-L1 CPS of 1 or higher; and 562 were treated in the ITT population. Twenty-seven patients with a PD-L1 CPS of 10 or higher completed treatment while 189 discontinued, 35 patients with a PD-L1 CPS of 1 or higher completed treatment and 379 discontinued, and 39 patients in the ITT population completed treatment and 514 discontinued.
Among the 281 patients randomized to placebo, 103 were treated with a PD-L1 CPS of 10 or higher, 211 were treated with a PD-L1 CPS of 1 or higher, and 281 were treated in the ITT population. Five patients with a PD-L1 CPS of 10 or higher completed treatment and 95 discontinued, 8 patients with a PD-L1 CPS of 1 or higher completed treatment and 200 discontinued, and 12 patients in the ITT population completed treatment and 264 discontinued.
Baseline characteristics between the 2 treatment arms were well balanced. The median age was 53 years (range, 22-85), and 41.0% and 38.4% of patients on pembrolizumab and placebo, respectively, had an ECOG performance status of 1. Additionally, 75.1% of patients in each arm had a PD-L1 CPS of 1 or more; 38.9% and 36.7% of pembrolizumab- and placebo-treated patients, respectively, had a CPS of 10 or higher.
Furthermore, more than half of patients in each arm received gemcitabine/carboplatin at 54.9% vs 54.8%, respectively.
Additional data showed that, at a median follow-up of 44.0 months with pembrolizumab, the median OS among the PD-L1 CPS of 1 or higher group was 17.6 months with pembrolizumab vs 16.0 months with placebo (HR, 0.86; 95% CI, 0.72-1.04; P = .0563), and 17.2 months vs 15.5 months, respectively, in the ITT population (HR, 0.89; 95% CI, 0.75-1.05). The 18-month PFS rates with pembrolizumab in the PD-L1 CPS of 1 or higher and ITT groups were 48.4% and 47.8%, respectively; in the placebo groups, these rates were 41.4% and 41.8%, respectively.
Further analyses showed that the median OS for pembrolizumab vs placebo was 16.2 months vs 14.7 months, respectively, in those with a PD-L1 CPS of less than 1 (HR, 0.97; 95% CI, 0.72-1.32); 13.9 months vs 15.5 months, respectively, in those with a CPS of 1 to 9 (HR, 1.09; 95% CI, 0.85-1.40); 20.3 months vs 17.6 months, respectively, in those with a CPS of 10 to 19 (HR, 0.71; 95% CI, 0.46-1.09); and 24.0 months vs 15.6 months, respectively in those with a CPS of 20 or higher (HR, 0.72; 95% CI, 0.51-1.01).
At a data cutoff date of June 15, 2021, the median PFS among those with a PD-L1 CPS of 1 or higher was 7.6 months in the pembrolizumab group vs 5.6 months in the placebo arm (HR, 0.75; 95% CI, 0.62-0.91); the 1-year PFS rates were 31.7% and 19.4%, respectively. In the ITT population, the median PFS was 7.5 months vs 5.6 months, respectively (HR, 0.82; 95% CI, 0.70-0.98), and the 1-year PFS rates were 29.3% and 20.8%, respectively.
When broken down further, the median PFS for pembrolizumab vs placebo was 6.3 months vs 6.2 months, respectively, in those with a PD-L1 CPS of less than 1 (HR, 1.09; 95% CI, 0.78-1.52); 5.7 months vs 5.6 months, respectively, in those with a CPS of 1 to 9 (HR, 0.85; 95% CI, 0.65-1.11); 9.9 months vs 7.6 months, respectively, in patients with a CPS between 10 and 19 (HR, 0.70; 95% CI, 0.44-1.09); and 9.2 months vs 5.4 months, respectively in those with a CPS of 20 or higher (HR, 0.62; 95% CI, 0.44-0.88).
In terms of safety, 96.3% of patients on pembrolizumab vs 95.0% of patients on placebo reported treatment-related adverse effects (TRAEs) of any grade; 68.1% vs 66.9%, respectively, had TRAEs reported as grade 3 to 5 in severity. Serious TRAEs occurred in 17.8% of patients on pembrolizumab vs 12.1% of those on placebo, and 18.3% vs 11.0% of patients, respectively, experienced a TRAE that led to discontinuation of either study drug. Moreover, 0.4% of patients on the pembrolizumab arm experienced a TRAE that led to death.
The most common TRAEs reported on the pembrolizumab and placebo arms were anemia (49.1% vs 45.9%, respectively), neutropenia (41.1% vs 38.1%), and nausea (39.3% vs 41.3%).
Immune-mediated AEs (imAEs) were reported in 26.5% of patients in the pembrolizumab arm vs 6.4% of those on placebo. Additionally, 5.3% of patients on pembrolizumab experienced a imAE that was grade 3 to 5 in severity, 3.4% reported serious imAEs, and 2.8% had an imAE that led to treatment discontinuation; these rates were 0% on the placebo arm.