This article is more than 5 years old. Information may no longer be current.
Pembrolizumab shows promise in metastatic triple-negative breast cancer
CHICAGO — Monotherapy with pembrolizumab safely led to durable responses in patients with heavily pretreated triple-negative breast cancer, according to results from a cohort of the KEYNOTE-086 trial.
“Triple-negative breast cancer is an aggressive heterogenous breast cancer subtype, with high unmet need,” Sylvia Adams, MD, medical oncologist at NYU Langone Medical Center, said during her presentation. “Due to the lack of receptor expression, therapy options are limited to chemotherapy. Despite chemotherapy, median OS in the second line and above is usually less than 1 year, and for many patients [treatment is] associated with significant toxicity.”
Pembrolizumab (Keytruda, Merck) showed durable activity in patients with PD-L1–positive metastatic triple-negative breast cancer in KEYNOTE-12. In the current trial, researchers sought to determine pembrolizumab’s efficacy in this population without selection for PD-L1.
The analysis included 170 women (median age, 54 years) included in cohort A of the trial who had received at least one prior chemotherapy for metastatic disease and an ECOG performance score of 0 to 1. Patients received 200 mg pembrolizumab every 3 weeks for up to 24 months.
Objective response rate — in all patients and in PD-L1–positive patients — and safety served as the study’s primary endpoints. Secondary endpoint included duration of response, disease control rate, PFS and OS.
Sixty-two percent of women had PD-L1–positive tumors, 44% had received three or more prior lines of therapy, and 74% has visceral metastases.
After a median follow-up of 10.9 months, nine patients (5%) remained on treatment. Disease progression was the most common reason for treatment discontinuation.
ORR was 4.7%, regardless of PD-L1 expression. This included one complete response and seven partial responses. Thirty-five patients (21%) achieved stable disease, five of which were durable for at least 6 months.
ORR appeared lower in patients with poor prognostic factors, such as high lactate dehydrogenase (high vs. low, 2% vs. 7%), liver metastases (with vs. without, 0% vs. 7%) and visceral metastases (2% vs. 11%).
The disease control rate was 8% (95% CI, 4-13) and median duration of response was 6.3 months (range, 1.2+ to 10.3+).
Five of the responding patients (63%) had not experienced progressive disease at the time of the data cutoff.
Twenty-seven percent of patients demonstrated decrease in target lesion size from baseline, which were maintained over several assessments. Many patients with decreased tumor burden remained on therapy on time of data cutoff.
Median PFS was 2 months (95% CI, 1.9-2) and median OS was 8.9 months (95% CI, 7.2-11.2); PFS and OS did not differ by PD-L1 status.
Twelve percent of patients achieved 12-months PFS and 69% achieved 12-month OS.
Further, all responding patients were alive and only six of the 35 women with stable disease died at the time of the analysis. For these cohorts, median OS has not been reached, whereas it was 7.1 months for patients with progressive disease as best response.
Sixty percent of patients experienced any-grade adverse event, but only 12.4% experienced a grade 3 or grade 4 adverse event, none of which were fatal.
Adams also presented preliminary data from cohort B of the trial, which included 52 previously untreated patients. The ORR in this cohort was 23.1%, which was significantly higher than that observed in cohort A. Further, 50% of patients demonstrated decrease in target lesion size.
“Although it requires further investigation, the activity seemed greater in patients with less heavily pretreated disease,” Adams said. “PD-L1 was not a predictive factor in this cohort, therefore analysis of additional biomarkers such as tumor-infiltrating lymphocytes are ongoing.” – by Alexandra Todak
Reference:
Adams S, et al. Abstract 1008. Presented at: ASCO Annual Meeting; June 2-6, 2017; Chicago.
Disclosure: Adams reports no relevant financial disclosures. Please see the abstract for a list of all other researchers’ relevant financial disclosures.
Perspective
Back to Top
Although the overall data would not suggest that pembrolizumab (Keytruda, Merck) would be an appropriate therapy in this patient population, this was a heavily pretreated group, and the response rate appeared considerably higher (23%) in the fraction of patients who were previously untreated. This is very promising, and there are ongoing trials of this drug and other checkpoint inhibitors, both alone and in combination.
In the broader view, it is very impressive how immunotherapy is working in cancers we did not think were particularly immunogenic. For a long time, we thought there were only a few cancers that would respond to immunotherapy — like melanoma and renal cell carcinomas — but we are learning that is not necessarily true. That has opened many opportunities; however, prompting the immune system to react remains a challenge for certain cancers and certain subsets. For those cancers, we have to figure out how to better combine therapies or specifically target patients most likely to respond.
Additional data are showing that not all triple-negative breast cancers are the same. We tend to think of this as one population, but we are now seeing that there are different subtypes of triple-negative breast cancer. We also are learning more about those factors that impact the role PD-1 and PD-L1 play in the immune response, such as mutational burden, HLA status or other co-stimulatory molecules. This is why continued basic science research in this arena is so important. The key is to do a better job identifying those biomarkers that might select the subsets of patients with triple-negative breast cancer who will best respond to PD-1 inhibitors.
This research shows us that PD-L1 alone likely won't be enough to identify the patient populations most likely to benefit. Researchers are exploring how specific genetic mutations or the expression of other biomarkers may predict response to checkpoint inhibitors. There also are potential immunologic and physiologic markers that may help identify patients whose immune systems may be able recognize the cancer, but aren't potent enough to make a difference clinically. There may be other approaches to help these patients.
Read more about