
Editor's Note: At the 2026 ESMO Congress, Professor Hope S. Rugo of City of Hope Comprehensive Cancer Center, California, USA, presented the interim analysis (IA) and final analysis (FA) results of the KEYNOTE-B49 study. The findings showed that pembrolizumab plus chemotherapy significantly improved progression-free survival (PFS) in patients with PD-L1 CPS ≥1, HR-positive/HER2-negative advanced breast cancer (ABC). The PFS benefit was more pronounced in the subgroup with PD-L1 CPS ≥10. In addition, the immunotherapy combination more than tripled the 24-month PFS rate, with an overall manageable safety profile. These findings suggest that some patients with endocrine-resistant and more aggressive disease may benefit from immunotherapy. Oncology Frontier interviewed Professor Hope S. Rugo at the ESMO Congress to discuss the key findings from KEYNOTE-B49, the clinical benefits observed in specific subgroups, and the role of PD-L1 testing in guiding clinical practice.
Oncology Frontier: You presented the results of the KEYNOTE-B49 study at this year’s ESMO Congress. Could you introduce the background of the study and its key findings?
Professor Hope S. Rugo: I am a breast medical oncologist at City of Hope Comprehensive Cancer Center in California, USA, and also serve as Director of the Breast Medical Oncology Division. At this year’s ESMO Congress, I presented data from the KEYNOTE-B49 study. Initiated several years ago, this study was designed to evaluate pembrolizumab (pembro) plus chemotherapy (CT) versus placebo (pbo) plus CT in patients with PD-L1-positive, HR-positive/HER2-negative advanced breast cancer (ABC).
The study enrolled 342 patients with centrally confirmed PD-L1 CPS ≥1 who had previously received endocrine therapy (ET) and were candidates for first-line chemotherapy. Eligible patients had locally recurrent unresectable or metastatic HR-positive/HER2-negative advanced breast cancer.
In this study, PD-L1 positivity was defined as CPS ≥1. Due to slow enrollment and a relatively low rate of PD-L1 positivity, the planned sample size was revised from the original 800 patients to approximately 340. The primary endpoint was PFS assessed by blinded independent central review (BICR) in patients with PD-L1 CPS ≥1. The key secondary endpoint was overall survival (OS) in the CPS ≥1 population. Other secondary endpoints included PFS and OS in patients with CPS ≥10, objective response rate (ORR), disease control rate (DCR), duration of response (DOR), and safety.
At this year’s ESMO Congress, I presented the results of both the interim analysis (IA) and final analysis (FA), with median follow-up periods of 29.1 months (data cutoff: September 22, 2025) and 34.9 months (data cutoff: March 16, 2026), respectively. We found that most enrolled patients had a PD-L1 CPS of 1–9 (78.9%), while 21.1% had a CPS ≥10. The two treatment arms were well balanced in terms of PD-L1 CPS distribution.
Among patients with PD-L1 CPS ≥1, pembrolizumab plus chemotherapy significantly improved PFS compared with placebo plus chemotherapy. The hazard ratio (HR) was 0.74 in both the interim and final analyses, with median PFS of 6.7 versus 6.0 months, respectively. However, the absolute difference in median PFS was modest, at just 0.7 months.
A greater difference was observed in the PD-L1 CPS ≥10 subgroup. In the interim analysis, median PFS was 7.5 versus 4.4 months, with an HR of 0.59 (95% CI, 0.35–0.98), favoring pembrolizumab plus chemotherapy. However, this subgroup was relatively small, accounting for only 21% of the overall study population, with 36 patients in each arm.
When we examined PFS in the PD-L1 CPS ≥1 population, it became clear that some patients derived substantially greater benefit than the median values alone would suggest. The 24-month PFS rate more than tripled in patients with PD-L1 CPS ≥1 (14.1% vs. 3.5%), which is truly remarkable. The proportion of patients in this subgroup who remained progression-free at 24 months increased from less than 5% to approximately 14%. We rarely see results like this. These findings suggest that a subset of patients with HR-positive, endocrine-refractory metastatic disease who are PD-L1-positive—and particularly those with PD-L1 CPS ≥10—can benefit from the addition of pembrolizumab to standard chemotherapy.
We also evaluated OS. However, no significant OS benefit was observed with pembrolizumab plus chemotherapy in either the PD-L1 CPS ≥1 or CPS ≥10 subgroup. Nevertheless, after adjusting for subsequent salvage treatments received by patients, the HR decreased to below 1 (HR, 0.94). Interestingly, patients who received pembrolizumab were less likely to receive subsequent chemotherapy or antibody-drug conjugates (ADCs). After adjusting for this variable, the HR improved slightly, although the difference remained statistically nonsignificant because of the very small patient numbers.
Regarding adverse events, thyroid disorders were the most common. We did not observe any other adverse reactions, including immune-related adverse events, beyond what has already been reported in previous studies.
Overall, I believe the key message from this study is that there is indeed a subset of patients who can benefit, although it is a relatively small subgroup. These patients must be PD-L1-positive, and the benefit appears more pronounced in those with strongly positive PD-L1 expression, defined as CPS ≥10. We do not yet know whether patients with ER expression below 10% would respond differently, as very few such patients were included in this trial. However, I believe it is important to assess PD-L1 status in patients with ER-positive, endocrine-refractory metastatic breast cancer who appear to have more aggressive disease and are less sensitive to endocrine therapy, because these patients may benefit from immunotherapy.