乌鸦传媒

SAN ANTONIO -- Adding an immune checkpoint inhibitor (ICI) to chemotherapy significantly increased the rate of pathologic complete response (pCR) and reduced residual cancer burden (RCB) in high-risk early-stage hormone receptor (HR)-positive/HER2-negative breast cancer, two randomized trials showed.

The pCR rate (ypT0/Tis ypN0) increased from 15.6% with chemotherapy plus placebo to 24.3% with the addition of pembrolizumab (Keytruda). The difference favoring the ICI arm persisted across a wide range of subgroups, including geographic location, disease stage, lymph node status, and different levels of PD-L1 expression.

Patients with node-positive disease, higher PD-L1 expression, and low estrogen receptor (ER) expression had larger increases in pCR. Patients who received less than the planned chemotherapy doses had lower pCR rates, although pembrolizumab improved pCR rates regardless of the completeness of chemotherapy exposure, reported Joyce O'Shaughnessy, MD, of Baylor University Medical Center in Dallas, at the San Antonio Breast Cancer Symposium (SABCS).

"Immune-mediated adverse event rates were consistent with the known toxicity profile of pembrolizumab plus neoadjuvant chemotherapy and no new safety concerns were observed," she said. "The study is powered to evaluate event-free survival (EFS) as the dual primary endpoint. EFS results are immature and continue to be evaluated."

Similar results emerged from a prespecified biomarker analysis of a trial evaluating the addition of nivolumab (Opdivo) to neoadjuvant chemotherapy. An intention-to-treat (ITT) analysis showed improved pCR in the ICI arm, and the biomarker analysis showed higher pCR and low RCB rates in patients with higher PD-L1 expression, lower HR expression, and a higher stromal tumor infiltrating lymphocyte (sTIL) score, reported Sherene Loi, MD, of Peter MacCallum Cancer Center in Melbourne, Australia.

Collectively, the two studies will help improve understanding about the use of immunotherapy in early HR-positive/HER2-negative breast cancer, but the as-yet unreported event-free survival (EFS) data will be even more informative, said SABCS invited discussant Laura A. Huppert, MD, of the University of California San Francisco.

"It's particularly critical to see the EFS data before we consider introducing IO [immuno-oncology] in these patients, as there is less data for the use of pCR as an intermediate biomarker in hormone receptor-positive, HER2-negative early-stage breast cancer," said Huppert. "How these data correlate with EFS is a critical question."

KEYNOTE-756

O'Shaughnessy reported findings from a prespecified biomarker analysis of the trial, which compared neoadjuvant chemotherapy with or without pembrolizumab, followed by adjuvant endocrine therapy plus pembrolizumab or placebo. The from the study showed that pembrolizumab significantly improved pCR in the ITT population, irrespective of tumor PD-L1 expression status.

Data analysis included 1,278 patients with previously untreated T1c-T2 cN1-2 or T3-4 cN0-2 disease. The trial has dual primary endpoints of pCR and EFS. Overall survival is a secondary endpoint, and the proportion of patients achieving RCB 0-1 is an exploratory endpoint.

The patients had a median age of 49, and about 90% of the study population had ECOG performance status 0. Three-fourths of the patients had a PD-L1 combined positive score (CPS) ≥1, including 40% with a CPS ≥10. About 90% of the patients had node-positive disease.

The primary analysis showed an 8.5% absolute difference in pCR (P=0.00005). Results were similar for alternative pCR definitions (ypT0 ypN0 and ypT0/Tis) and disease stage (II vs III). The pCR rate in node-positive disease was similar to the overall results but lower in the node-negative group (16.9% vs 13.1%).

The pCR rate with pembrolizumab and the difference favoring the ICI are increased with PD-L1 CPS, including 42.3% vs 29.0% for CPS ≥10 and 53.6% vs 36.4% for CPS ≥20. Among patients with CPS ≥1, those with ER expression <10% had substantially higher pCR rates with pembrolizumab (57.6% vs 33.3%) as compared with those who had ER expression ≥10% (27.6% vs 18.4%).

Patients allocated with pembrolizumab had a shift toward lower RCB after definitive surgery: RCB 3 - 20.5% vs 28.9%; RCB 2 - 40.8% vs 45.3%; and RCB 0-1 - 35% vs 23.6%.

As expected, immune-mediated AEs in the neoadjuvant phase occurred more often in the pembrolizumab arm 32.8% vs 7.0%, including grade 3-5 (7.1% vs 1.2%) and serious AEs (6.2% vs 1.7%). Rates of discontinuation associated with any drug were 7.7% in the pembrolizumab arm and 1.6% with chemotherapy plus placebo.

CheckMate 7FL

Loi reported findings from the phase III trial, involving 510 patients with high-risk HR-positive/HER2-negative breast cancer, including 349 with complete biomarker data. Patients were randomized to neoadjuvant chemotherapy with or without nivolumab, followed by adjuvant endocrine therapy. An showed that adding nivolumab to chemotherapy significantly increased the pCR rate in the ITT population (24.5% vs 13.8%, P=0.0021). Patients with higher PD-L1 expression (≥1% immune cells, ICs) had higher pCR rates (44.3% vs 20.2%).

The new analysis examined pCR and RCB 0-1 rates by PD-L1 expression (IC % positivity), various CPS cutoffs in baseline core biopsies, sTIL score, estrogen and/or progesterone receptor (ER/PR) expression, and Ki67 proliferation index.

The pCR rate increased with PD-L1 expression. Investigators identified PD-L1 CPS ≥3 as the optimal cutoff for predicting pCR benefit from the addition of nivolumab. At that level, the pCR rates were 53.0% with nivolumab versus 25.8% with chemotherapy alone.

The proportion of patients achieving RCB 0-1 with nivolumab also increased with PD-L1 expression. The overall rate was 30.7% with nivolumab versus 21.3% without. The rate increased from 46.8% at CPS ≥1 (vs 29.8% without nivolumab) to 78.9% at CPS ≥20 (vs 26.7%).

Additional analyses showed that nivolumab-associated pCR and RCB 0-1 rates increased with sTIL values and were higher in association with lower levels of ER and PR expression. Investigators found no associations between nivolumab benefit and Ki67 index.

"Additional exploratory and correlative analyses are ongoing to further refine the patient subpopulation with primary ER-positive/HER2-negative breast cancer who could benefit from the addition of nivolumab to neoadjuvant chemotherapy," said Loi.