NEW YORK – Data from two Phase III trials presented at the European Society for Medical Oncology Congress have suggested that adding immunotherapy to chemotherapy first in the neoadjuvant setting and then to endocrine therapy in the adjuvant setting may lead to better outcomes in hormone receptor-positive, HER2-negative early breast cancer.
In the CheckMate 7FL and KEYNOTE-756 trials, researchers tested the activity of Bristol Myers Squibb's Opdivo (nivolumab) or Merck's Keytruda (pembrolizumab), respectively, in combination with chemotherapy in the neoadjuvant setting, and then in combination with endocrine therapy in the adjuvant setting in patients with high-risk estrogen receptor (ER)-positive, HER2-negative early breast cancer. Patients receiving immunotherapy-containing regimens were compared to patients who received neoadjuvant chemo followed by endocrine therapy in the adjuvant setting. In both CheckMate 7FL and KEYNOTE-756, patients on the immunotherapy-containing regimens had better pathologic complete responses, as compared to the control group.
In discussing the results from the two studies at the meeting, Stephen Johnston, head of medical oncology at the Royal Marsden NHS Foundation Trust, said, "This is really important data showing for the first time that [pathologic complete response] can be improved" with immunotherapy for HR-positive breast cancer patients in early-stage treatment. He further called the two studies "potentially practice-changing."
In general, breast cancer isn't very immunogenic, and patients tend not to respond well to immune checkpoint inhibitors. Nonetheless, researchers and drugmakers have been trying to work out which subsets of breast cancer patients may derive some benefit from checkpoint inhibitors but have had mixed success.
The KEYNOTE-355 study, for example, showed that Keytruda added to chemo could improve overall survival compared to just chemo in metastatic, PD-L1-positive triple-negative breast cancer. The IMpassion130 had also suggested initially that Roche's checkpoint inhibitor Tecentriq (atezolizumab) with chemo benefited PD-L1-positive advanced TNBC patients, but in 2021, the company withdrew this indication, which had received accelerated approval from the US Food and Drug Administration, after the IMpassion131 trial failed to confirm the progression-free survival benefit seen in the earlier trial.
Keytruda plus chemo for neoadjuvant treatment followed by single-agent Keytruda as adjuvant treatment is also approved in the US for high-risk early-stage TNBC patients, though this indication doesn't require PD-L1 testing. Roche subsidiary Genentech was similarly exploring Tecentriq as neoadjuvant and adjuvant treatment but stopped the studies after lackluster results.
Meanwhile, the I-SPY2 platform trial suggested the potential benefit of neoadjuvant Keytruda plus chemo in not just early-stage TNBC but also in HR-positive, HER2-negative breast cancer. Given that most of the benefit seen with immunotherapy has been in TNBC, some experts have interpreted this data as suggesting that some ER-positive breast cancers may be more TNBC-like.
These efforts set the stage for the present CheckMate 7FL and KEYNOTE-756 studies.
In CheckMate 7FL, more than 500 patients with high-risk, early-stage, ER-positive, HER2-negative breast cancer received Opdivo plus neoadjuvant chemotherapy followed by Opdivo plus adjuvant endocrine therapy, or placebo plus neoadjuvant chemotherapy followed by placebo plus adjuvant endocrine therapy.
Sherene Loi, lead investigator of the trial and a medical oncologist at Peter MacCallum Cancer Center in Victoria, Australia, said that the trial was affected by several events while it was being conducted from November 2019 to April 2022. In that time, Eli Lilly's CDK4/6 inhibitor Verzenio (abemaciclib) became an adjuvant treatment option for HR-positive early breast cancer patients, for example, requiring extensive protocol modification in CheckMate 7FL, which affected enrollment numbers.
"The treatment landscape changed in October 2021 with the approval of adjuvant [Verzenio] in the same population as [CheckMate 7FL]," Loi said. "This was expected to result in a high rate of withdrawals due to safety concerns when combining a CDK4/6 inhibitor with an anti-PD-1 [therapy]. This put the co-primary endpoint of [progression-free survival] at risk."
As a result, the researchers amended the protocol for the trial to focus solely on pathologic complete response as the primary endpoint. And instead of evaluating event-free survival as a co-primary endpoint, researchers made this into an exploratory endpoint and stopped enrolling patients at 521 patients instead of the planned 1,000. The enrollment target was then further reduced to 510 patients for the primary efficacy analysis when trial sites in Russia shuttered due to the Russia-Ukraine war. Follow-up was reduced to one year post-surgery and the adjuvant portion of the treatment became open label.
"And, of note, this was during the COVID-19 pandemic," Loi added.
Even with the reduction in sample size and other headwinds faced by researchers, Loi said, they "still had adequate power to evaluate our hypothesized difference in [pathologic complete response] rates in this population from 12 to 22 percent."
About one-third of the patients were PD-L1 positive, and 80 percent were clinically node-positive at diagnosis. Among those who had Opdivo added to their neoadjuvant and adjuvant treatments, the pathologic complete response rate was 24.5 percent compared to 13.8 percent for those who had placebo plus standard neoadjuvant and adjuvant therapy. In PD-L1 positive patients, the pathologic complete response rate increased from 20.2 percent to 44.3 percent with the addition of Opdivo. The pathologic complete response rate, by comparison, increased only by 3.6 percent when Opdivo was added in PD-L1 negative patients.
The benefit seen with Opdivo was consistent between subgroups including nodal status and clinical stages and across age and ethnic backgrounds. Researchers will present detailed biomarker analysis at a future scientific conference.
In the similarly designed KEYNOTE-756 study, researchers tested neoadjuvant Keytruda or placebo plus chemotherapy followed by adjuvant Keytruda or placebo plus endocrine therapy in 1,278 patients with early-stage, high-risk, ER-positive, HER2-negative invasive ductal breast cancer. Pathologic complete response and event-free survival were dual primary endpoints, with secondary endpoints including overall survival and other measures of pathologic complete response.
"This subtype is very heterogeneous. It has different clinical behaviors and outcomes with standard treatment," said Fatima Cardoso, lead investigator of the study and director of the breast unit at Champalimaud Clinical Center in Lisbon, Portugal. She added that although these subtypes generally have a better prognosis than other breast cancer subtypes, there is a high-risk subpopulation that is typically treated with chemotherapy. In that group, after neoadjuvant chemotherapy, published pathologic complete response rates range from zero to 18 percent.
In the first interim analysis from KEYNOTE-756, as of a May 2023 data cutoff, the pathologic complete response rate for patients who had Keytruda added to their standard neoadjuvant and adjuvant treatment regimens was 24.3 percent compared to 15.6 percent for those in the control arm. Among PD-L1-positive patients, the pathologic complete response rate was 29.7 percent in the Keytruda arm versus 19.5 in the placebo arm, and in PD-L1-negative patients, 7.2 percent and 2.6 percent, respectively. Event-free survival results were not mature at data cutoff. (About 75 percent of patients in both arms of the trial were PD-L1 positive.)
As with CheckMate 7FL, results were consistent across subgroups, with a significant exception. The subgroup of patients who had tumors with low ER-positivity, around 5 percent of the study population, "seems to derive a bigger benefit [with Keytruda], although with a larger confidence interval," Cardoso said. "This is a piece of data that is very important for those of us who believe that these tumors behave more as triple-negative than as ER-positive."