PD-L1 blockade breaks through triple-negative breast cancer




Metastatic triple-negative breast cancer appears to be the latest hard-to-treat cancer to yield to the juggernaut that is now anti-PD-L1 immunotherapy.

MPDL3280L, an investigational monoclonal antibody against programmed death ligand 1 (PD-L1), posted an overall response rate of 19% among 21 evaluable patients in a phase Ia trial (95% confidence interval, 5-42).

Dr. Leisha Emens © 2015 AACR/Todd Buchanan

Dr. Leisha Emens

This included two complete responses in patients with high PD-L1 expression and two partial responses. Three of the four responses are ongoing, Dr. Leisha Emens reported at the annual meeting of the American Association for Cancer Research.

“I think it very well could be the first targeted therapy that bears out in a larger trial,” she said during a press briefing. “These data are still early, and we need to enroll and treat a lot more patients with this agent, but I think it has great, great promise for this particular breast cancer subtype.”

There is great unmet need for new treatments in triple-negative breast cancer (TNBC) because it has a worse prognosis than other breast cancer subtypes do, and the only approved treatment option in the United States is chemotherapy.

TNBC is a good candidate for immunotherapy, particularly PD-L1 targeted therapies, because it has a higher mutation rate than do other breast cancer subtypes. This produces neoantigens that can be recognized as foreign by the immune system and be more effective targets for an immune response, Dr. Emens of the Johns Hopkins Sidney Kimmel Comprehensive Cancer Center, Baltimore, explained.

TNBC also has higher PD-L1 expression levels, which can inhibit T-cell antitumor responses, and more tumor-infiltrating lymphocytes, which can facilitate a immune response and are associated with improved outcomes when present in high numbers.

MPDL3280A is designed to inhibit the binding of PD-L1 to programmed death receptor 1 (PD-1) and B7.1, which can restore antitumor T-cell activity and enhance T-cell priming, she said.

The checkpoint inhibitor received breakthrough therapy designation for metastatic bladder cancer in 2014 and a second designation in non–small cell lung cancer in February.

The ongoing phase Ia trial enrolled 54 women with metastatic TNBC and an ECOG performance status of 0 or 1. This included 21 patients initially selected for high PD-L1 expression levels (at least 5%) on their immune cells and 33 all-comers. MPDL3280A intravenous infusions were given every 3 weeks at doses of 15 mg/kg, 20 mg/kg, or 1,200 mg. Efficacy was evaluated in the 21 patients and safety in all 54 patients.

At 24 weeks, progression-free survival was 27% (95% CI, 7-47), Dr. Emens said.The median duration of response (range, 18-56+ weeks) has not been reached.

Three patients with progressive disease experienced pseudoprogression, where the target lesion shrank, but new lesions developed. Pseudoprogression, a feature of checkpoint inhibition that also has been seen with ipilimumab (Yervoy), is new for many physicians to manage and requires the patient’s entire clinical picture be taken into account, Dr. Emens said.

“An important component of the phenomenon of pseudoprogression is that if you see evidence of new lesions on a scan and the patient’s doing clinically well, you continue to treat and then reevaluate subsequent to that,” she said. “If there’s progression at that point, then potentially you consider changing the therapy or just following the patient more closely. Another potential option to help sort through that is to try and obtain tissue from one of those lesions to get some idea of what is happening, if it’s a phenomenon of the inflammatory response or a response to the therapy.”

Dr. Emens detailed one such case in which three target lesions decreased in size from baseline on 9- and 20-month follow-up scans, but newly enlarged axillary nodes that appeared inflammatory or necrotic developed near the third target lesion at 9 months. The patient remained on therapy and is doing well today, with further shrinkage of the target lesion and regression of the axillary nodes at 20 months.

MPDL3280A was generally well tolerated, with fatigue, nausea, fever, decreased appetite, and asthenia being the most common adverse events, Dr. Emens said. In all, 11% of patients experienced grade 3 treatment-related events. Two deaths, assessed as drug related by the investigator, are under investigation.

Press briefing moderator Louis M. Winer, director of the Georgetown Lombardi Comprehensive Cancer Center, commented that it wasn’t that long ago that phase I investigators were pleased if they saw even a hint of activity that would justify moving forward to phase II. The activity signals with the checkpoint inhibitors, however, are “unequivocal” and the implications for the future treatment of people with triple negative breast cancer are “very, very exciting,” according to Dr. Winer.

Next Article: