Exploratory Analysis Identifies Biomarkers of Response in TNBC

New biomarker analyses from the phase 3 KEYNOTE-522 trial provide insights into the biological factors that may influence response to pembrolizumab plus chemotherapy in early-stage triple-negative breast cancer (TNBC).

The new research, presented by Joyce O’Shaughnessy, MD, at San Antonio Breast Cancer Symposium (SABCS) 2024, also reinforces the broad benefit of this immunotherapy-chemotherapy combination across patient subgroups.

The authors examined multiple potential biomarkers including tumor mutational burden (TMB), T-cell–inflamed gene expression profile (GEP), and various non-T-cell–inflamed consensus signatures in relation to pathological complete response (pCR) and event-free survival (EFS).

The KEYNOTE-522 trial previously demonstrated that the addition of pembrolizumab to neoadjuvant chemotherapy, followed by adjuvant pembrolizumab, significantly improved pCR, EFS, and overall survival compared with chemotherapy alone in patients with high-risk, early-stage TNBC.

“This prespecified exploratory analysis evaluated associations between potential biomarkers for pathological response and EFS with pembrolizumab, according to a prespecified statistical analysis plan,” O’Shaughnessy explained.

T-Cell Inflammation and TMB are Associated With pCR and EFS

The analysis included 946 patients with whole exome sequencing data and 904 patients with ribonucleic acid sequencing data from pretreatment tissue samples. The researchers found that T-cell–inflamed GEP was positively associated with both pCR and EFS, regardless of treatment group (P

Priyanka Sharma, MD, Frank B. Tyler Professor in Cancer Research at the University of Kansas Medical Center, Kansas City, Kansas, who was not involved in the study, noted that the finding of T-cell-inflamed signature as prognostic was not surprising, as “many immune gene signatures have been shown to be prognostic in TNBC in settings of both chemotherapy and chemoimmunotherapy.”

Although TMB was significantly associated with pCR in both groups (P P P = .422). However, when asked about the proportion of patients with high TMB, O’Shaughnessy noted that it was

In an interview, Sharma cautioned about overinterpreting the association between TMB and improved EFS in the pembrolizumab plus chemotherapy group. “Very few patients had high TMB, and although this exploratory analysis is intriguing, it does not inform patient practice due to such small numbers,” she said. “The criterion for high TMB used in this analysis is not something we have available in commercial assays, which report TMB on a different scale.”

Association Between Molecular Subtypes and Patient Outcomes

Addressing molecular subtypes, O’Shaughnessy highlighted findings regarding the luminal androgen receptor (LAR) subtype: “The luminal androgen receptor subtype had the lowest pCR rate with chemotherapy alone and the largest improvement in pCR rate with the addition of pembrolizumab. All subtypes had improved EFS with the addition of pembrolizumab, with the largest absolute benefit in the LAR subgroup.”

The analysis also revealed an inverse correlation between human epidermal growth factor receptor 2 (HER2) gene expression and T-cell–inflamed GEP (correlation factor, −0.32), although HER2 levels and molecular subtypes did not show significant associations with pCR or EFS in the pembrolizumab group after adjusting for T-cell–inflamed GEP. The analysis also revealed that the PTEN loss signature was positively associated with pCR rates in both groups and with EFS in the pembrolizumab group.

In an interview, Sharma explained that “PTEN loss has also been previously associated with poor prognosis both in chemotherapy and chemo-plus immunotherapy in TNBC,” adding that “these findings confirm what we have known about TNBC biology.”

The study also examined homologous recombination deficiency (HRD) status, which was determined through whole exome sequencing. O’Shaughnessy explained in response to an audience question.

“High HRD is more than 16% [of the 552 genomic regions showing loss of heterozygosity], and low HRD is less than 16%. That has been validated with the Myriad myRisk test,” she said.

Although HRD positivity correlated with higher pCR rates in both treatment groups, both HRD-positive and HRD-negative patients showed improved EFS with the addition of pembrolizumab. This finding suggests that HRD status alone should not be used to select patients for pembrolizumab therapy.

Implications and Future Directions

Despite the broad benefits of pembrolizumab across biomarker subgroups, Sharma said she believes that there is still value in pursuing biomarker-driven treatment selection. “Immunotherapy, while effective, is quite toxic, and it is clear that not all patients need it,” she said. “Analysis of other biomarkers like tumor-infiltrating lymphocytes, MHC-II [major histocompatibility complex class II], Determa-IO signature, and TNBC-Dx would be very valuable from this and other phase 3 trials.”

O’Shaughnessy concluded that while several biomarkers showed prognostic values for pCR and EFS, most were not predictive of pembrolizumab benefit. She added that the combination of pembrolizumab plus chemotherapy demonstrated superior efficacy compared with chemotherapy alone across various biomarker-defined subgroups.

Although this analysis provides valuable insights into the biological factors influencing treatment outcomes in early-stage TNBC, Sharma emphasized that “none of these biomarkers are clinically available and [that the analysis] did not identify patients who would not benefit from addition of pembrolizumab to neoadjuvant chemotherapy” and that “these findings do not impact current patient decisions.”

O’Shaughnessy said the findings of this study lay the groundwork for future research into optimizing immunotherapy-based treatment strategies for patients with early-stage TNBC.

Sharma argued that “at this point, this exploratory analysis may not inform future trial designs.” She further explained that “biomarkers that have been shown to be predictive of immunotherapy benefit in previous retrospective analyses were not reported.”

This study was funded by Merck Sharp & Dohme LLC, a subsidiary of Merck & Co, Rahway, New Jersey, USA (MSD). O’Shaughnessy reported receiving honoraria from Agendia, AstraZeneca, Bristol-Myers Squibb, Eisai, Genentech, Lilly, MSD, Novartis, Pfizer, Puma, Roche, Gilead, Stemline, Scorpion, and Pierre Fabre.Sharma reported receiving consulting fees and honoraria from Merck, Gilead, Genzyme Corporation (Sanofi), Novartis, AstraZeneca, GSK, Pfizer, and Menarini.

Christos Evangelou, PhD, is a freelance medical writer and science communications consultant.