What’s a GPA?
The Study: Never far behind, Merck funded this single institution phase 2 trial in which patients with untreated or even progressing brain mets 5-20 mm received pembrolizumab without local therapy. Importantly, here the population largely did not consist of de novo brain mets with roughly 80% previously treated with local therapy. Patients with synchronous (non-evaluable) brain mets >20 mm or in eloquent areas received local therapy to those lesions prior to pembro. Among a grand total of 23 enrollees, 6 (26%) responded—outnumbered by the 8 (35%) deemed “unevaluable” prior to follow-up scan due to clinical progression and/or need for radiation. One potential reason for this? The upper limit of the range of brain mets in this population not receiving whole brain radiation was...81. We can only hope this was a typo. In the only defense for Merck we can find, the few observed responses were durable and ongoing at 24 months. Another noteworthy finding was the 30% necrosis rate in lesions previously treated with radiosurgery. The obvious problem here: the design seems largely ignorant of decades of brain met trials championed by radiation oncology and neurosurgery and seems to prioritize elimination of focal therapy over improvement in patient outcomes.
Bottom Line: The intracranial response rate for pembro monotherapy in a small cohort of patients with a wide variety of melanoma brain mets is roughly one-quarter, outweighed by the rate of rapid clinical progression. | Kluger, J Clin Oncol 2018