Clinical trials in melanoma patients with brain metastases.

Abstract

Dear Editor With the approval of inhibitors of CTLA-4, BRAF, MEK, and PD-1 for advanced melanoma, systemic therapy has dramatically improved. Due to the historically poor prognosis for patients with melanoma brain metastases (MBMs), coupled with concerns about blood–brain barrier penetration of larger molecules, the original trials with most of these agents excluded patients with brain involvement. Treatment of MBMs has consisted primarily of local therapy with surgery, whole brain radiation, or stereotactic radiosurgery. Each carries its own set of short- and long-term risks that may affect neurological and neurocognitive functions (Flanigan et al., 2013). Effects of newer systemic therapies on MBMs therefore need assessment. A substantial proportion of metastatic melanoma patients develop MBMs; the incidence at autopsy is 75% and ~40% develop clinically apparent disease (Flanigan et al., 2011). Among the first trials evaluating BRAF-targeting therapy for untreated MBMs was a trial with sorafenib and temozolomide (Table 1). Patients with asymptomatic MBMs had a median progression-free survival (PFS) of 3.5 months, compared to 4.2 months in patients without, demonstrating feasibility of systemic therapy for untreated MBMs (Amaravadi et al., 2009). The phase I trial of dabrafenib in unresectable BRAF-mutant melanoma included ten patients with small, untreated MBMs; tumor shrinkage was seen with no additional toxicities (Falchook et al., 2012). This led to a phase II trial of dabrafenib in 172 patients with lesions <40 mm diameter. The intracranial response rate (RR) among patients without and with prior surgery or radiation was 39 and 30.8% (Long et al., 2012). A pilot study of 19 patients with previously untreated MBMs using vemurafenib showed ≥30% shrinkage in MBMs in 37% (Dummer et al., 2014). Table 1 Completed melanoma brain metastases trials Immune therapies also have activity in MBMs. A retrospective study of patients on a phase II trial with ipilimumab identified 12 patients with MBMs at trial entry. Two had a partial response (PR) in the brain, three had stable disease (SD); three of these five were still alive at 4 yr (Weber et al., 2011). A subsequent prospective, multicentered phase II trial studied ipilimumab specifically in patients with MBM (Margolin et al., 2012). Patients with either asymptomatic MBMs or symptomatic MBMs on a stable steroid dose were enrolled. The 12-week cerebral disease control rate was 24% in patients off steroids, 10% in patients on steroids. The difference is likely due to steroid immunesuppression and patient characteristics. In a phase II study of ipilimumab plus fotemustine, 20 patients had asymptomatic brain metastases, 25% had SD or PR in the brain, and 25% had a CR. Interestingly, patients with MBMs had the same median overall and 3-yr survival as those without (di Giacomo et al., 2012). These limited studies suggest that both targeted and immune therapies have activity in the brain, and might reduce the need for radiation and surgery for local control. Case series for both targeted and immune therapies have confirmed activity of these drugs in patients with MBM (Gibney et al., 2015; Knisely et al., 2012). These studies have demonstrated that accrual of patients with MBM to trials is feasible and have resulted in additional studies for MBMs with newly approved drugs or drugs in late stages of development (Table 2). Table 2 Current clinical trials for patients with melanoma brain metastases involving systemic therapy With the increasing population of patients with MBM, revision of clinical research programs to include them in disease-specific systemic therapy trials is required. In recent years, trials tend to include patients with treated, stable brain metastases, and pharmaceutical companies are now supporting MBM-specific trials prior to drug approval. The trials in Table 1 show that brain and extracerebral responses are typically concordant, and studies in other malignancies confirm this, suggesting that drugs cross the impaired blood–brain barrier seen in brain metastases (Bachelot et al., 2013; Costa and Kobayashi, 2012). Additional studies are evaluating systemic therapies together with radiation. Systemic therapy studies use variable inclusion criteria specific to brain metastases. Allowable prior local therapy is inconsistent, as are concurrent steroid use, allowable lesion size, and response criteria (Table 3). Limiting the number and/or size of allowable metastases in immune therapy trials and use of prophylactic anti-epileptics might increase RRs and decrease steroid use. Separate cohorts for patients with leptomeningeal disease are warranted. Image interpretation in MBMs is challenging, and clinical trial end points and response criteria require revision. MBM-related adverse events and neurotoxicity similarly need refinement, including long-term neurotoxicity and effects of local and systemic intervention on adjacent brain parenchyma. Moreover, the interaction between systemic therapies and radiation and resultant neurological toxicities is unknown. These newer clinical trial designs for brain metastasis patients will contribute to a rapidly evolving field while providing hope to patients with MBM. Table 3 Inclusion criteria and response assessment for patients with melanoma brain metastases (MBMs)—completed studies