Combination Regimen Shows Activity In HER2-Positive Breast Cancer
Jason M. Broderick
The combination of Keytruda (pembrolizumab) and Herceptin (trastuzumab) reached an objective response rate (ORR) of 15.2 percent in patients with Herceptin-resistant, PD-L1–positive, HER2-positive breast cancer, according to findings presented at 2017 San Antonio Breast Cancer Symposium (SABCS).
Results from the phase 1b/2 PANACEA trial also showed that Keytruda/Herceptin achieved a disease control rate of 24 percent in PD-L1–positive patients. Further, stromal tumor infiltrating lymphocytes (sTILs) were identified as a potential predictive marker. Among patients with sTIL levels of 5 percent or more, ORR was 39 percent versus 5 percent in patients with sTIL levels less than 5 percent.
“The PANACEA study of Keytruda with Herceptin in Herceptin-resistant metastatic HER2-positive patients met its primary endpoint in the PD-L1–positive cohort,” lead study author Sherene Loi, MD, PhD, associate professor at Peter MacCallum Cancer Centre in Melbourne, Australia, working with the International Breast Cancer Study Group, said at a press conference at SABCS.
PANACEA (SG 45-13/BIG 4-13/KEYNOTE-014) accrued patients with centrally confirmed HER2-positive breast cancer, ECOG performance status of 0 or 1, and measurable disease per RECIST 1.1. There was no limit on the number of prior systemic therapies, and patients had to have progressed on Herceptin or Kadcyla (T-DM1; ado-trastuzumab emtansine).
In the phase 1b part of the trial, PD-L1–positive patients received IV Keytruda at 2 mg/kg or 10 mg/kg plus Herceptin every three weeks. In phase 2, PD-L1–positive and –negative patients received 200 mg of IV Keytruda plus Herceptin every three weeks.
Treatment was administered until progression, unacceptable toxicity, patient withdrawal, investigator decision, or a maximum of two years.
The primary endpoint of the phase 2 part of the study was ORR. Secondary outcome measures included progression-free survival (PFS), disease control rate, duration of response, duration of disease control, and overall survival (OS).
Overall, 146 patients were screened between February 2015 and April 2017 at 11 sites across five countries. Sixty-eight (53.5 percent) patients were PD-L1 positive. It was determined that 10.6 percent of patients were ineligible for the study because they were HER2 negative. Investigators enrolled 58 (39.7 percent) of the screened patients: six PD-L1-positive patients in phase 1b, 40 PD-L1-positive patients in phase 2, and 12 PD-L1–negative patients in phase 2.
At a median follow-up of 13.6 months, three patients remained on treatment and 46 patients had discontinued treatment. Some of the reasons for discontinuation included side effects (n = 6), withdrawal of consent (n = 1), patient deterioration (n = 1), and death due to progressive disease (n = 1).
Across the cohorts, the median age was 50.5 years, 56.9 percent of patients were ER-negative, and 43.1 percent of patients were ER-positive. All patients had received chemotherapy and a Herceptin-containing regimen.
Loi noted that, “In the PD-L1-positive cohort, the patients were numerically younger and there was a higher frequency of ER-negative disease.”
Over 80 percent of each cohort had previously received an additional anti-HER2 therapy. In the phase 1b, phase 2 PD-L1-positive, and phase 2 PD-L1–negative cohorts, four, 29, and nine patients, respectively, had prior T-DM1. Three, 10, and four patients, respectively, had prior Perjeta (pertuzumab).
The ORR was 17 percent in the phase I PD-L1-positive group, comprising one complete response (CR). In the phase 2 PD-L1-positive group, the ORR was 15 percent, comprising one CR and five partial responses (PRs). Seven patients in this cohort had stable disease (SD), 25 had progressive disease, and two were not evaluable.
There were no responses in the PD-L1–negative group. Two patients had SD, nine had progressive disease, and one was not evaluable.
Among all PD-L1–positive patients, the median duration of disease control was 11.1 months and the median duration of response was 3.5 months. At the time of the data cutoff, five patients continued without progression. Loi noted that three of these five patients have completed two years of Keytruda.
The median PFS among all PD-L1-positive patients was 2.7 months versus 2.5 months in PD-L1–negative patients. The 12-month PFS rates were 13 percent versus 0, respectively.
Across all PD-L1-positive patients, the median OS was 16.1 months versus seven months in the PD-L1–negative group. The 12-month OS rates were 65 percent versus 12 percent, respectively.
Higher levels of sTILs were associated with improved response and disease control. Among all PD-L1-positive patients, 41 percent had sTILs of 5 percent or more. The disease control rate was 47 percent in patients with sTILs of 5 percent or more versus 5 percent in patients with sTILs less than 5 percent.
“An [sTIL level] of 5 percent provides high sensitivity, high negative predictive value, and potentially high reproducibility amongst pathologists,” said Loi.
Safety was evaluated across all 58 patients. The most common side effects included fatigue, diarrhea, arthralgia, headache, nausea, dyspnea and myalgia.
Commenting on the next steps in the HER2-positive setting, Loi said, “Future directions for immunotherapy in metastatic HER2+ breast cancer should focus on combinations with effective anti-HER2 therapy, especially in low TIL patients.”