Addition of pembrolizumab to chemotherapy demonstrates promise in triple-negative breast cancer

SAN ANTONIO — The addition of pembrolizumab to neoadjuvant chemotherapy demonstrated promising antitumor activity and exhibited a manageable toxicity profile in a cohort of patients with early-stage triple-negative breast cancer, according to results of the KEYNOTE-173 trial presented at San Antonio Breast Cancer Symposium.

The findings support the ongoing phase 3 KEYNOTE-522 trial, which is comparing the combination with placebo in this patient population.

“One of the key messages is it is safe to combine immune therapy with chemotherapy in this setting. Some of the toxicities of chemotherapy are not insignificant, but our impression is that the addition of immune therapy does not substantially alter the toxicity profile,” Peter Schmid, MD, PhD, FRCP, clinical director of St. Bartholomew Breast Cancer Centre in London and lead of the Centre of Experimental Cancer Medicine at Barts Cancer Institute, told HemOnc Today. “These combinations also are very active, especially combinations with platinum-based therapy.”

Limited treatment options are available for triple-negative breast cancer, and patient outcomes typically are poor.

The anti-PD-1 therapy pembrolizumab (Keytruda, Merck) has demonstrated antitumor activity and an acceptable safety profile for patients with metastatic triple-negative breast cancer.

In the six-cohort KEYNOTE-173 study, Schmid and colleagues evaluated different doses and schedules of taxanes and platinum in combination with pembrolizumab as neoadjuvant therapy for patients with locally advanced triple-negative breast cancer.

Researchers enrolled 60 women (median age, 48.5 years; range, 26-71) with newly diagnosed, locally advanced, previously untreated disease. Most women had ductal histology (83.3%), T2/T3 disease (88.3%) and nodal involvement (66.7%).

All patients received a single dose of pembrolizumab 200 mg on day 1 of the first cycle.

In the second through fifth cycles, patients received pembrolizumab 200 mg plus one of six chemotherapy regimens: nab-paclitaxel 125 mg/m² once weekly (cohort A); nab-paclitaxel 100 mg/m² once weekly plus carboplatin area under the curve 6 every 3 weeks (cohort B); nab-paclitaxel 125 mg/m² once weekly plus carboplatin area under the curve 5 every 3 weeks (cohort C); nab-paclitaxel 125 mg/m² once weekly plus carboplatin area under the curve 2 once weekly (cohort D); paclitaxel 80 mg/m² once weekly plus carboplatin area under the curve 5 every 3 weeks (cohort E); and paclitaxel 80 mg/m² once weekly plus carboplatin area under the curve 2 once weekly.

In the sixth through ninth cycles, all patients received doxorubicin 60 mg/m² every 3 weeks and cyclophosphamide 600 mg/m² every 3 weeks plus pembrolizumab 200 mg every 3 weeks.

Patients underwent breast MRI at screening, as well as after the fifth and ninth treatment cycles.

Safety and recommended phase 2 dose served as the primary endpoints.

Key efficacy endpoints included pathologic complete response — defined as complete absence of disease in the breast and lymph nodes — objective response rate, EFS and OS.

Researchers assessed dose-limiting toxicities during the first three cycles, as well as in the sixth and seventh cycles.

Median follow-up was 19.6 months.

Twenty-two (36.6%) experienced dose-limiting toxicities (cohort A, n = 2; cohort B, n = 4; cohort C, n = 6; cohort D, n = 6; cohort E, n = 0; cohort F, n = 4).

The most common grade 3 or higher treatment-related adverse events were neutropenia (73%), febrile neutropenia (22%), anemia (20%) and thrombocytopenia (8%).

Eighteen patients (30%)experienced immune-related adverse events, the most common of which were grade 2 hypothyroidism (n = 4), grade 1 hyperthyroidism (n = 3), grade 3 colitis (n = 2) and grade 3 rash (n = 2).

Eleven patients (18.3%) discontinued pembrolizumab due to treatment-related adverse events (one each in cohorts A, B and E, and four each in cohorts D and F).

Researchers reported ORRs of 100% (90% CI, 74-100) in cohorts B and C, 90% (90% CI, 61-100) in cohorts D and F, 80% (90% CI, 49-96) in cohort A, and 70% (90% CI, 39-91) in group E.

Sixty percent (90% CI, 30-85) of all patients achieved pathologic complete response.

“That is a figure we normally don’t achieve with chemotherapy-free combinations for triple-negative breast cancer,” Schmid said.

Patients for whom surgery was not possible, as well as those who declined an operation, were classified as not achieving pathologic complete response, Schmid said.

One such patient had a complete resolution of cancer in her breast but did not undergo axillary surgery because the axilla appeared clear, Schmid said.

“In a study as small as this, you can imagine it biases against the combination, so what we are reporting likely is a minimum rate of activity,” he said.

Investigators reported 12-month EFS rates of 100% in cohorts B, C, E and F; 90% (90% Ci, 58-98) in cohort D; and 80% (90% CI, 49-93) in cohort A.

Twelve-month EFS rates were 100% among patients who achieved pathologic complete response and 88% (90% CI, 71-95) among those who did not.

“There is a clear correlation between pathologic complete response with chemotherapy and long-term outcomes in triple-negative breast cancer,” Schmid told HemOnc Today. “What we don’t know yet is how clear the correlation is if we add in a biological or immune therapy.”

Three patients developed recurrence. None of those three received platinum-based chemotherapy.

“These are very early data and very small numbers,” Schmid said. “We can’t make final conclusions, but the outcomes are very encouraging, particularly considering tripe-negative breast cancer is a very difficult-to-treat subtype.” – by Mark Leiser

Reference:

Schmid P, et al. Abstract PD5-01. Presented at: San Antonio Breast Cancer Symposium; Dec. 4-8, 2018; San Antonio.

Disclosures: Please see the abstract for all other authors’ relevant financial disclosures.