Survival Win for Pembrolizumab in Early Triple-Negative Breast Cancer

PD-1 inhibition before and after surgery for high-risk early triple-negative breast cancer (TNBC) led to significant improvement in overall survival (OS), according to updated results from a large randomized trial.

After a median follow-up of 75 months, the 5-year OS was 86.6% in patients who received pembrolizumab (Keytruda) before and after surgery compared with 81.7% in those who received only neoadjuvant chemotherapy. The advantage for PD-1 inhibition persisted across most prespecified subgroups, including PD-L1 expression status, nodal status, and tumor size.

Notably, patients benefited from pembrolizumab regardless of whether they achieved a pathologic complete response (pCR), reported Peter Schmid, MD, PhD, of Barts Cancer Institute and Queen Mary University in London, at the European Society for Medical Oncology congress in Barcelona.

“Neoadjuvant pembrolizumab with chemotherapy followed by pembrolizumab continues to show a clinically meaningful improvement in survival, the data plateauing over the last few years with neoadjuvant chemotherapy alone, after a median follow-up of over 6 years,” said Schmid. “With this longer follow-up, the toxicity profile remained consistent with the previous analyses and the established safety profiles of pembrolizumab and chemotherapy, with no new safety concerns.

“These results provide further support for neoadjuvant platinum-containing chemotherapy with pembrolizumab and pembrolizumab after surgery as a standard of care for patients with high-risk stage II or stage III triple-negative breast cancer.”

The finding of an OS benefit caps a progression of positive reports from the KEYNOTE-522 trial, which previously demonstrated significant improvements in pCR and event-free survival (EFS) with pembrolizumab.

The survival data from the trial show that “pembrolizumab contributes to the cure of triple-negative breast cancer,” said invited discussant Marleen Kok, MD, PhD, of the Netherlands Cancer Institute in Amsterdam. Data from the trial also showed that use of pembrolizumab leads to a higher pCR rate and improves survival in patients who do not achieve pCR.

Moving forward, the challenge is to build on these practice-changing results. Kok cited three key questions that need to be addressed in future investigations and analyses:

• Which patients are cured with neoadjuvant pembrolizumab?
• Do all patients need a full year of treatment with the PD-1 inhibitor?
• What is the long-term impact on quality of life in the young patient population that TNBC comprises?

“Neoadjuvant pembrolizumab added to chemotherapy significantly improves overall survival in triple-negative breast cancer, and this is, or should be, the new standard of care for stage II or III triple-negative breast cancer,” said Kok. “It is the shared responsibility of industry, academia, guideline committees, and regulatory bodies to ensure that patients get the most comprehensive information to guide decisions and minimize over- and undertreatment.

“We need biomarkers from registration trials. We have to limit trials where we have no clue on the contribution of the components [of treatment]. We need special attention for long-term immunotherapy toxicity, now that we are curing young breast cancer patients with immunotherapy.”

KEYNOTE-522 included 1,174 patients with newly diagnosed stage II or III TNBC. All patients received neoadjuvant chemotherapy with carboplatin-paclitaxel followed by doxorubicin or epirubicin plus cyclophosphamide. They were randomized 2:1 to neoadjuvant pembrolizumab or placebo, followed by pembrolizumab or placebo after surgery.

The trial met the dual primary endpoints of pCR (64.8% vs 51.2% in favor of pembrolizumab) and EFS (84.5% vs 76.8% at 3 years). Schmid reported findings from a prespecified analysis of OS, a principal secondary endpoint.

The updated analysis showed a continuation of the EFS advantage with pembrolizumab, 81.2% versus 72.2% at 5 years, representing a 35% reduction in the hazard ratio (95% CI 0.51-0.83). The 5-year results differed little from the 3-year results (84.6% vs 76.4%) in favor of pembrolizumab.

“The hazard ratio has remained stable throughout, as most of the recurrence events in triple-negative breast cancer tend to occur relatively early,” said Schmid. “We see probably about 70% of recurrences in the first 2 to 3 years, about 90% of recurrences in the first 5 years. Fortunately, we saw relatively few events between year 3 and year 5.”

The 4.9% absolute difference in OS at 5 years represented a 34% reduction in the survival hazard in favor of pembrolizumab (95% CI 0.50-0.87, P=0.00150). As expected, patients who achieved a pCR had better OS (95.1% with pembrolizumab, 94.4% with placebo) than those who did not. Nonetheless, treatment with pembrolizumab conferred a survival benefit among patients who did not achieve a pCR (71.8% vs 65.7%, HR 0.76, 95% CI 0.56-1.05).

With regard to safety, KEYNOTE-522 investigators “haven’t seen any new signals,” said Schmid. “If you look at the chemotherapy phase of treatment, the toxicity is clearly dominated by chemotherapy.”

Immune-mediated side effects were as expected, he continued. The most common immune side effect was hypothyroidism (15.1%), followed by skin reactions (5.7%) and hyperthyroidism (5.2%).