Pembrolizumab/Radiation Improves T Cell Infiltration in Early Breast Cancer

Combining preoperative radiation therapy with pembrolizumab (Keytruda) significantly boosted T cell infiltration among patients with node-positive, higher-risk, hormone receptor (HR)–positive, HER2-negative, early-stage breast cancer, according to data from the phase 2 TBCRC-053 (P-RAD) trial (NCT04443348) highlighted in a press briefing at the 2025 San Antonio Breast Cancer Symposium (SABCS).¹

Only patients who received radiation at 24 Gy in combination with pembrolizumab were determined to have a statistically significant increase in T cell infiltration for the primary tumor (P = .023). Patients with high T cell infiltration also tended to have higher levels of PD-L1 expression compared with those who did not have high infiltration.

“When we look by gene expression analysis, we see that beyond just T cell infiltration, we are getting infiltration with B cells, dendritic cells, immunostimulatory macrophages, and natural killer [NK] cells,” lead study investigator Gaorav Gupta MD, PhD, an assistant professor in the Department of Radiation Oncology at Lineberger Comprehensive Cancer Center of the University of North Carolina at Chapel Hill, noted in his presentation of the data.

Using spatial transcriptomics, Gupta and coinvestigators discovered an increased abundance of immune structures called tertiary lymphoid structures (TLS) following treatment with pembrolizumab/radiation. He described how these structures have correlated with better responses to immunotherapy in other cancer types.

Regarding surgical outcomes, the rate of ypN0 disease or nodal pathologic complete response (pCR) at the time of definitive surgery was 23.5% (n = 4/17) with no radiation, 29.4% (n = 5/17) with 9 Gy of radiation, and 33.3% (n = 5/15) with 24 Gy of radiation. In each respective cohort, the pCR rates were 5.9% (n = 1/17), 29.4% (n = 5/17), and 20.0% (n = 3/15); the rates of residual cancer burden (RCB) 0 or 1 were 17.7% (n = 3/17), 29.4% (n = 5/17), and 33.3% (n = 5/15). Gupta noted that the trial was not powered to show clinically meaningful differences for these end points.

Exploratory analyses revealed that the rate of pCR was 6.25% (n = 1/16) for patients with Luminal A disease and 46.7% (n = 7/15) for those with non-Luminal A status (P = .016). Those with PD-L1 Q4 marker expression status experienced a pCR rate of 50% (n = 8/16) compared with 0% (n = 0/15) in those without the marker (P = .0024). Among patients with non-Luminal A disease and Q4 PD-L1 expression, the pCR rate was 70% (n = 7/10) vs 4.8% (n = 1/21) in those with Luminal A disease or non-Q4 PD-L1 expression (P = .0003).

“Despite [patients] having a high burden of disease, and including non-grade 3 tumors, we have observed encouraging rates of pCR in the radiation plus pembrolizumab arms. Exploratory analyses suggest the greatest benefit of this radiation/immunotherapy [combination] is in non-Luminal A tumors that induce PD-L1 expression at the 2-week [biopsy] timepoint,” Gupta said in the presentation. “Future trials examining preoperative radiation with pembrolizumab in HR-positive, HER2-negative breast cancer are needed and encouraged to clarify the disease control benefit that this may bring.”

According to Gupta, data from the phase 3 CheckMate 7FL trial (NCT04109066) and the phase 3 KEYNOTE-756 trial (NCT03725059) previously demonstrated the possibility of increasing pCRs by adding immune checkpoint inhibitors to neoadjuvant chemotherapy in patients with HR-positive, HER2-negative early-stage breast cancer.^2,3 However, these immunotherapy strategies only appeared to show a benefit in tumors that showed high baseline tumor-infiltrating lymphocytes (TILs) and PD-L1 expression.

“If we could perhaps improve the immune infiltration of these tumors before we start the immunochemotherapy, we may get better responses,” Gupta noted. “That was the motivation to ask whether localized radiation could have this effect on these tumors, perhaps leading to better immunotherapy responses.”

In the TBCRC-053 trial, patients with node-positive, HER2-negative breast cancer were randomly assigned to receive no, low-, or high-dose preoperative radiation plus pembrolizumab and chemotherapy. The trial included a cohort of patients with triple-negative disease (n = 48) as well as those with high-risk, HR-positive, HER2-negative disease (n = 48), which constituted the focus of Gupta’s presentation. Two weeks after initiating 0 Gy, 3 fractions of 3 Gy, or 3 fractions of 8 Gy of radiation, patients underwent biopsy for their primary tumor and metastatic lymph node before starting neoadjuvant chemotherapy, which consisted of weekly paclitaxel plus adriamycin/cyclophosphamide with concurrent pembrolizumab.

The trial’s primary end point was the breast tumor T cell infiltration at the 2-week biopsy, measured as a rank-based Immunoscore. This evaluation provided a percentile rank of peritumoral and intratumoral CD3/CD8 T cell to tumor ratio compared with an untreated cohort via multiplexed panCK/CD3/CD8 immunofluorescence. Investigators defined an upper quartile of 0.75 or higher T cell infiltration as a T cell responder.

The trial’s co-primary end point was the ypN0 rate, which was not powered for comparative analyses. Secondary end points included the composite pCR rate and RCB 0/1 rate, as well as PD-L1 expression. Exploratory end points included transcriptional signatures of response.

Patients with estrogen receptor expression of 10% or higher, progesterone receptor expression of 10% or higher, and HER2 non-amplified disease were eligible for inclusion in this cohort of the P-RAD trial. Other eligibility criteria included having cN1 to N3, cT1c to T4c, cM0 disease, as well as grade 3 disease with a Ki67 index of at least 20% or a high-risk Genomic assay score; those with grade 1/2 disease with cN2 to N3 status were permitted to enroll.

Across the overall population, the median age was 51 years (range, 23-78). Of note, 47% and 7.8% of patients, respectively, had T3 and T4 disease; 25% and 24% had N2 and N3 status. Moreover, 2.0%, 35.0%, and 63.0% had grade 1, grade 2, and grade 3 disease, respectively.

Disclosures: Gupta noted relevant financial relationships as an employee of the University of North Carolina at Chapel Hill and having received grant/research support from Merck, Breakpoint Therapeutics, NIH/NCI, the Department of Defense, the Breast Cancer Research Foundation, and the V Foundation. He also highlighted being a stockholder in Naveris, Inc. and having received IP/Royalty payments from Naveris, Inc.

References

1. Gupta GP, Ho AY, Gharibpoor F, et al. Primary results from the HR+/HER2- cohort of TBCRC-053 (P-RAD): a randomized trial of no, low, or high dose preoperative RADiation with pembrolizumab and chemotherapy in node-positive, HER2-negative breast cancer. Presented at the 2025 San Antonio Breast Cancer Symposium (SABCS); December 9-12, 2025; San Antonio, TX. Abstract GS2-05.
2. Loi S, Salgado R, Curigliano G, et al. Neoadjuvant nivolumab and chemotherapy in early estrogen receptor-positive breast cancer: a randomized phase 3 trial. Nat Med. 2025;31:433-441. doi:10.1038/s41591-024-03414-8
3. Cardoso F, O’Shaughnessy J, Liu Z, et al. Pembrolizumab and chemotherapy in high-risk, early-stage, ER+/HER2− breast cancer: a randomized phase 3 trial. Nat Med. 2025;31:442-448. doi:10.1038/s41591-024-03415-7