The research team, led by Yosuke Aoyama and Takayuki Ueno, reviewed the records of 95 consecutive patients with early-stage triple-negative breast cancer or the closely related estrogen receptor-low, HER2-negative subtype who received neoadjuvant pembrolizumab plus chemotherapy at their institution between October 2022 and September 2024. These patients were not trial volunteers screened for perfect compliance and optimal organ function; they were ordinary women treated in a busy cancer center, with the comorbidities, tolerability issues, and treatment variations that characterize real-world oncology. The median age of the cohort was 52 years, and the median follow-up reached 24.2 months, a window long enough to capture most early recurrence events in this notoriously aggressive disease.

The centerpiece of the analysis was event-free survival, a measure that oncologists consider among the most clinically meaningful endpoints in early breast cancer. In this study, an event was defined as disease progression during treatment, local or distant recurrence, a second primary cancer, or death from any cause, with the clock starting at the initiation of neoadjuvant therapy. Using Kaplan-Meier estimation and log-rank comparisons, the investigators calculated that the estimated event-free survival rates stood at 96.8 percent at 12 months and 88.0 percent at 24 months across the entire cohort. For a disease that, before the immunotherapy era, carried a substantial risk of early relapse despite chemotherapy, these numbers represent a striking confirmation that the trial results translate into practice.

Perhaps the most important mechanistic insight came from stratifying patients by their pathological response to the pre-surgical treatment. A pathological complete response, meaning no residual invasive cancer detectable in the breast or lymph nodes at the time of surgery, remains the single most powerful predictor of long-term outcome in triple-negative breast cancer. In this real-world cohort, 62.1 percent of patients achieved a pathological complete response, a figure closely mirroring the rates observed in KEYNOTE-522. Among these responders, the estimated event-free survival was an extraordinary 100 percent at 12 months and 97.3 percent at 24 months. The difference between responders and non-responders was statistically significant, with the non-pCR group showing estimated rates of 91.7 percent and 75.3 percent at the same time points.

The timing of relapse events carried its own message. Every event-free survival event observed in the non-responder cohort occurred within 21 months of starting therapy, suggesting that the risk window for early recurrence in patients who do not achieve a pathological complete response is concentrated in the first two years after treatment initiation. This pattern has practical implications for surveillance strategies and for the design of future clinical trials aimed at improving outcomes for the residual-disease population, which remains the major unmet need in this field. For patients who clear the two-year mark without an event, the data suggest a favorable trajectory, though longer follow-up will be needed to confirm durable cure rates.

Beyond confirming efficacy, the study ventured into territory that has fascinated tumor immunologists for years: the relationship between immune-related adverse events and treatment benefit. Immune checkpoint inhibitors work by releasing the molecular brakes on T cells, and the same mechanism that unleashes an attack on tumor cells can also trigger inflammation in healthy organs, producing toxicities ranging from mild rash and thyroid dysfunction to severe pneumonitis, colitis, hepatitis, and endocrinopathies. A provocative hypothesis in the field holds that patients who develop these immune-related adverse events may have more biologically active immune systems and may therefore derive greater benefit from immunotherapy, although the evidence across tumor types has been inconsistent.

In this cohort, grade 3 or 4 immune-related adverse events, the severe toxicities that typically require immunosuppressive treatment and sometimes permanent discontinuation of pembrolizumab, occurred in 14 of the 95 patients. Remarkably, none of these 14 patients experienced any event-free survival event during the follow-up period. The authors are careful, and appropriately so, to frame this as a descriptive, hypothesis-generating observation rather than proof of a causal relationship. The number of patients with severe toxicities is small, the follow-up is limited to roughly two years, and retrospective single-institution analyses cannot control for the many confounders that influence both toxicity risk and relapse. Still, the complete absence of relapse in the severe-toxicity group adds a tantalizing data point to an ongoing scientific conversation about immune activation as a double-edged sword.

The study also carries significance for a subgroup that often falls between diagnostic categories: patients with estrogen receptor-low, HER2-negative disease. These tumors express barely detectable levels of the estrogen receptor, below the 10 percent threshold that defines conventional hormone receptor positivity, and they have historically been lumped with triple-negative cancers for treatment purposes despite limited evidence that the biology is identical. By including these patients and reporting outcomes for the combined population, the Japanese team provides real-world evidence that the pembrolizumab strategy performs well across this expanded indication, informing clinicians who face exactly these borderline receptor-low cases in daily practice.

Several technical caveats deserve honest acknowledgment. The retrospective design means that treatment decisions, including the specific chemotherapy backbones used alongside pembrolizumab, were made at the discretion of treating physicians rather than by protocol, introducing heterogeneity that a randomized trial would eliminate. The single-center setting, while ensuring consistent pathology review and data collection, limits generalizability to health systems with different patient demographics and resource levels. The relatively short median follow-up of 24.2 months means that late recurrences, which triple-negative breast cancer can certainly produce, have not yet had time to emerge, and the true event-free survival curves will only sharpen with time. The authors also note that the study received no external funding, and they disclose lecture honoraria and advisory relationships with several pharmaceutical companies, including MSD, the manufacturer of pembrolizumab, a common and transparent arrangement in Japanese academic oncology.

Nevertheless, the bottom line is difficult to escape: the combination of neoadjuvant pembrolizumab plus chemotherapy followed by adjuvant pembrolizumab, which entered clinical guidelines on the strength of KEYNOTE-522, is delivering on its promise outside the trial environment. For the roughly 10 to 15 percent of breast cancer patients whose tumors are triple-negative, a subtype that disproportionately affects younger women and carries a more aggressive clinical course than hormone receptor-positive disease, this confirmation matters enormously. It validates the treatment decisions being made today in cancer centers worldwide, it reinforces pathological complete response as the pivotal checkpoint guiding prognosis and follow-up intensity, and it flags the non-responder population, with all of its relapses clustering within 21 months, as the clear priority for the next generation of clinical trials. As immune checkpoint inhibitors continue their march through early-stage disease, studies like this one serve as the essential reality check that tells the field whether laboratory triumphs and trial statistics are truly changing the lives of the patients who need them most.

Subject of Research: Real-world event-free survival with neoadjuvant and adjuvant pembrolizumab in early-stage triple-negative breast cancer

Article Title: Real-world event-free survival with neoadjuvant pembrolizumab plus chemotherapy and adjuvant pembrolizumab in early-stage triple-negative and estrogen receptor-low, HER2-negative breast cancer

Article References: Aoyama, Y., Ozaki, Y., Kuno, M., Masuda, J., Nishimura, M., Kurata, M., Inagaki, L., Hosonaga, M., Fukada, I., Taniguchi, E., Kimura, Y., Maeda, T., Yoshida, K., Yamashita, N., Kobayashi, T., Takano, T., & Ueno, T. (2026). Real-world event-free survival with neoadjuvant pembrolizumab plus chemotherapy and adjuvant pembrolizumab in early-stage triple-negative and estrogen receptor-low, HER2-negative breast cancer. BMC Cancer. https://doi.org/10.1186/s12885-026-17065-0

Image Credits: AI Generated

DOI: 10.1186/s12885-026-17065-0

Keywords: triple-negative breast cancer, pembrolizumab, immunotherapy, neoadjuvant chemotherapy, event-free survival, pathological complete response, immune-related adverse events, KEYNOTE-522, real-world data, HER2-negative breast cancer, adjuvant therapy, breast cancer recurrence