The original trial data in more than 1,100 patients with early-stage TNBC indicated that adding pembrolizumab to chemotherapy prior to surgery and giving the drug for a year afterward improves event-free survival (EFS) over placebo by 37%.
Now, the researchers conducted a series of prespecified sensitivity and subgroup analyses, finding remarkably consistent EFS outcomes whether considering the addition of adjuvant chemotherapy, positive surgical margins, or disease characteristics such as nodal status and disease stage.
The analyses showed that the benefit with pembrolizumab over placebo was “robust,” said study presenter Peter Schmid, MD, PhD, Centre for Experimental Cancer Medicine, Barts Cancer Institute, Queen Mary University of London.
“These results further support pembrolizumab plus platinum-containing neoadjuvant chemotherapy followed by adjuvant pembrolizumab after surgery as a new standard of care treatment regimen for patients with high-risk, early-stage TNBC,” he said.
The research was presented at the San Antonio Breast Cancer Symposium on Dec. 7.
Hope S. Rugo, MD, who was invited to comment on the findings, noted that, while the sensitivity analyses showed the benefit with pembrolizumab was seen across the board, the numbers in each group of interest were “very small, making any impact unlikely.”
She continued that there also remain a number of unanswered questions, chief among them being: “Does everybody need a checkpoint inhibitor? Perhaps studies ... could help us understand which patients might do well with chemotherapy alone.”
Dr. Rugo, who is professor of medicine in the division of hematology and oncology at the Helen Diller Family Comprehensive Cancer Center at the University of California, San Francisco, , added that “we need to understand the balance of risk and toxicity” asking whether there are patients whose risk of an immunotoxicity is “so high that we should not give them a checkpoint inhibitor.”
It is not clear what constitutes the optimal chemotherapy backbone. “Does everybody need carboplatin? Does everyone need a year of pembrolizumab, even with a pathologic complete response given the intriguing data from GeparNUEVO and previously the I-SPY trial?” she asked.
“Of course, we don’t know the answers to those questions,” she said, but it is nevertheless possible to draw a roadmap for the treatment of early TNBC, although the choice of adjuvant therapy following surgery is less clear.
Dr. Rugo conducted a Twitter poll to canvas opinion on what to give to patients following surgery, depending on whether or not they have a pathological complete response.
At 73%, most of almost 200 respondents said patients with a pathological complete response should continue pembrolizumab for 1 year, while 72% said that patients without a pathological complete response should receive combination therapy of pembrolizumab and either capecitabine or olaparib, depending on mutational status.
Dr. Schmid began his presentation by noting that KEYNOTE-522 was the first prospective, randomized, phase 3 trial of pembrolizumab in early TNBC in the neoadjuvant and adjuvant setting.
Previously presented results showed that adding neoadjuvant pembrolizumab to chemotherapy was associated with a clinically meaningful increase in pathological complete response, while continuing with adjuvant chemotherapy after surgery led to a clinically meaningful improvement in EFS.
Consequently, the Food and Drug Administration approved pembrolizumab in this setting for patients with high-risk early-stage TNBC.
He reminded the audience that the trial included 1,174 patients randomized 2:1 to pembrolizumab or placebo every 3 weeks alongside eight cycles of chemotherapy, followed by pembrolizumab over placebo alone for up to nine cycles after undergoing definitive surgery.
After a median follow-up of 39.1 months, 15.7% of patients treated with pembrolizumab experienced an event versus 23.9% of those in the placebo group, at a hazard ratio of 0.63 (P = .00031). At 36 months, the EFS rate was 84.5% with pembrolizumab and 76.8% in patients treated with placebo.
Dr. Schmid said that they then performed five prespecified sensitivity analyses, which revealed that the results were “consistent with the primary EFS in all five sensitivity analyses, showing the robustnesses of the event-free survival benefit in the pembrolizumab arm.”
The first analysis, he continued, is of “particular interest as it considered the impact of postsurgery new anticancer therapy. For example, the use of adjuvant capecitabine.”
Censoring 31 patients from the pembrolizumab arm who received the drug and 13 of those given placebo, the team found that the hazard ratio for EFS for pembrolizumab versus placebo was 0.64.
Removing “positive margin at last surgery” as part of the definition of EFS also did not change the results substantially, with the HR for EFS for pembrolizumab versus placebo at 0.65.
Subgroup analysis revealed “consistent EFS results,” Dr. Schmid said, irrespective of whether stratifying the patients by nodal status, overall disease stage, menopausal status, HER2 status, or lactate dehydrogenase levels.
While patients in both treatment arms who had nodal involvement had worse outcomes than those without, those in the pembrolizumab arm “still had improved outcomes, compared with placebo, suggesting that it provides benefit regardless of nodal status.”
“Similarly, the EFS benefit with pembrolizumab was irrespective of disease stage,” Dr. Schmid said. Although the EFS improvement was greater in patients with stage II rather than III disease, at a HR of 0.60 versus 0.68, it highlights “the importance of early intervention.”
He said that the “rate of adverse events with pembrolizumab was low, especially in the adjuvant setting.”
Following his presentation, Dr. Schmid was asked whether he would consider retrying immunotherapy in patients after progression on pembrolizumab.
He replied that this is currently a “data-free zone.”
However, he said: “If a patient responded immunotherapy initially, had a disease-free interval and then has recurrence, then I would consider, if the patient is PD-L1 [programmed death–ligand 1] positive, at that time to add immunotherapy. We can’t say whether those patients will derive the same benefit” as that seen in randomized controlled trials in later stage TNBC, he added, “but there is, in my opinion, little to lose, especially if we have already established the patient tolerates immunotherapy well in that setting.”
Dr. Schmid continued that he “personally found it reassuring” that, in the current study, even patients without a complete pathological response “still showed a substantially better event-free survival compared to patients without immunotherapy, so I personally would consider immunotherapy for those patients when they relapse but we can discuss what the optimal disease-free interval is.”
The study was funded by Merck Sharp and Dohme. Both Dr. Rugo and Dr. Schmid reported relationships numerous pharmaceutical companies.
.