In the Journals

Pembrolizumab shows activity in brain metastases from NSCLC, melanoma

Show Citation

June 24, 2016

Pembrolizumab demonstrated therapeutic activity in brain metastases of patients with non–small cell lung cancer or melanoma, according to early results of an ongoing randomized phase 2 trial.

“In the United States, about 50,000 patients with metastatic melanoma or non–small cell lung cancer develop brain metastases every year. ... At diagnosis, 10% of patients with metastatic NSCLC have brain metastases and another 30% develop brain involvement,” Sarah B. Goldberg, MD, MPH, assistant professor in the department of medical oncology in the Smilow Cancer Hospital at Yale School of Medicine, and colleagues wrote. “Many effective drugs in development have not been well studied for central nervous system penetrations, and patients with untreated brain metastases are excluded from most clinical trials.”

Sarah Goldberg

Sarah B. Goldberg

Based on the demonstrated efficacy of PD-1 checkpoint inhibitors for various tumor types, Goldberg and colleagues evaluated the safety and activity of pembrolizumab (Keytruda, Merck) for 36 patients who have brain metastases from NSCLC (n = 18; median age, 59; range, 33-82) or melanoma (n = 18; median age, 65; range, 41-85).

Patients had untreated brain metastases between 5 mm and 20 mm and did exhibit neurologic symptoms. PD-L1 expression was required for patients with NSCLC but not for patients with melanoma.

All patients received 10 mg/kg IV pembrolizumab every 2 weeks until progression. Brain metastasis response served as the primary endpoint.

Median follow-up was 11.6 months for the melanoma cohort and 6.8 months for the NSCLC cohort.

Four (22%; 95% CI, 7-48) patients with melanoma and six (33%; 95% CI, 14-59) patients with NSCLC demonstrated a brain metastasis response. Nine of 10 responses were ongoing at the time of data analysis on June 30, 2015.

One patient in the melanoma group experienced grade 3 cognitive dysfunction and elevated aminotransferases. Grade 4 hyperkalemia; grade 3 colitis, pneumonitis and fatigue; and grade 2 acute kidney injury each occurred in one patient with NSCLC.

Grade 1 to 2 neurologic adverse events in the melanoma group included seizure (17%; n = 3) headache (17%; n = 3) and dizziness (6%; n = 1). In the NSCLC group, grade 1 to 2 neurologic adverse events included cognitive dysfunction (6%; n = 1), headache (22%; n = 4), dizziness (11%; n = 2) and stroke (6%; n = 1).

Six patients in the melanoma cohort and nine in the NSCLC cohort died. One patient died of an unknown cause 8 months after the last dose of study treatment. All other deaths resulted from disease progression.

Median OS was not reached for the melanoma group and was 7.7 months (95% CI, 3.5-not reached) for the NSCLC group.

Researchers plan to report additional endpoints, PFS, OS as well as biomarker information when study data are mature. They acknowledged the small sample size as a limitation of these results.

“Our findings clearly show that systemic PD-1 inhibition can be effective in treating nonirradiated or progressing brain metastases,” Goldberg and colleagues wrote. “The relatively low proportion of patients with melanoma who responded compared with other studies of prembrolizumab might underrepresent clinical benefit and might reflect complexities of clinical trials in this patient population.”

Because new targeted treatments have prolonged survival in patients, and because whole-brain radiotherapy can impair memory, research into novel treatment for patients with brain metastases are warranted, Rafael Rosell, MD, director of the Cancer Biology and Precision Medicine Program at Catalan Institute of Oncology of Hospital Germans Trias i Pujol in Barcelona, Spain, and Niki Karachaliou, MD, medical oncologist and member of the Translational Research Group in Pangaea Biotech SA in Barcelona, Spain, wrote in an invited commentary.

“Mounting evidence suggests that the therapeutic benefit of monoclonal antibodies against PD-1 and PD-L1 is achieved in patients with a preexisting T-cell response against their tumors, as established by baseline CD8 T-cell infiltration in the tumor microenvironment,” Rosell and Karachaliou write. “Treatment with anti–PD-1 antibodies is a promising new therapeutic approach that should be used in combination with targeted treatments to inhibit crucial oncogenic signaling pathways to prevent the absence of T-cell infiltrates in tumor sites.” by Nick Andrews

Disclos ure: Goldberg reports grants from AstraZeneca and Merck, as well as fees from Clovis. Rosell and Karachaliou report no relevant financial disclosures. Please see the full study for a list of all other researchers’ relevant financial disclosures.