The evolving role of systemic therapy and local, brain-directed treatment in patients with melanoma and brain metastases

See article by Tawbi et al. pp 1961–1973. Patients with advanced non-uveal melanoma carry significant risk for developing brain metastases. Population-based data has indicated that, among allcomers with de novo metastatic melanoma, 28% will harbor intracranial oncologic disease.1 In addition, a significant percentage of patients with advanced melanoma without intracranial disease at presentation will develop brain metastases over time. The prognosis of patients with melanoma and brain metastases has historically been dismal and often limited by neurologic disease progression, but advances in immunotherapeutic and, for the 40%-50% of patients harboring a targetable BRAF mutation, targeted approaches, have markedly improved outcomes in this population.2–5 The systemic management of brain metastases secondary to melanoma is dependent on multiple factors, including intracranial/extracranial disease burden, the presence vs absence of targetable mutations in genes such as BRAF, NRAS, and KIT, age, performance status, comorbidities, symptomatology, steroid requirements, prior treatment, and other factors. Among patients with bulky/progressive systemic or intracranial oncologic disease lacking a viable local option, the use of BRAF/MEK inhibition, if viable, has appeal given the potential for rapid and profound response. However, among patients with melanoma and smaller, asymptomatic brain metastases not requiring steroids, who lack an overwhelming burden of extracranial disease and who can tolerate dual-agent immunotherapy, the preferred systemic approach likely entails ipilimumab and nivolumab; targeted agents may be less-preferred in such patients given concerns regarding durability of response and rebound of disease post-discontinuation of targeted treatment.5,6 Supporting data for dual-agent immunotherapy in the management of such patients stem from CheckMate 204 (a single-arm phase II study of ipilimumab/nivolumab in patients with active/unirradiated brain metastases secondary to melanoma without (Cohort A) or with (Cohort B) symptoms/steroid requirements) and the ABC study (a randomized phase II study of patients with melanoma and asymptomatic/unirradiated brain metastases comparing ipilimumab/nivolumab (Cohort A) to nivolumab (Cohort B) with a single-arm cohort (Cohort C) consisting of patients managed with nivolumab monotherapy for progressive disease after local therapy, neurologic symptoms, or leptomeningeal disease).2–4 Among patients with asymptomatic brain metastases devoid of steroid requirements, Cohort A from both the CheckMate 204 and ABC studies displayed an intracranial response/clinical benefit rate of approximately 50%-60% and more impressively a strong correlation between extracranial and intracranial response, suggesting a limited impact of the blood-brain barrier,7 was noted, as were prolonged responses to systemic therapy and sustained long-term overall survival. It should be noted that the ABC study permitted prednisone at ≤10 mg/day, or equivalent.3,4 Among patients with smaller, asymptomatic brain metastases secondary to melanoma who are not on steroids, the role of radiation remains unclear. The historical rationale for radiation stems from concern regarding the intracranial efficacy of systemic therapy.7 Both CheckMate 204 and the ABC trial challenged this assertion. In addition, concerns regarding necrosis risk when stereotactic radiation is utilized in patients receiving immunotherapy, particularly dual-agent immunotherapy in melanoma, are significant.8,9 However, the use of concomitant stereotactic radiation and immunotherapy has also been demonstrated to yield marked and durable responses.10 Ultimately, randomized data are required to better inform the role of stereotactic radiation in the management of such patients. In this regard, a randomized study of ipilimumab plus nivolumab with or without brain-directed stereotactic radiation among patients with melanoma and asymptomatic brain metastases without a need for prednisone (or equivalent) >10 mg/day is underway (NCT03340129). Among patients with brain metastases secondary to melanoma who are neurologically symptomatic/steroid-dependent, CheckMate 204 indicated a lower intracranial response rate with ipilimumab and nivolumab (22%),3,4 potentially related to steroid-mediated inhibition of immunotherapy-based efficacy, albeit with a very small sample size (N = 18).3,4 Consequently, systemic therapy alone is likely insufficient for management, especially among patients with little anatomic/physiologic reserve to tolerate further disease progression. CheckMate 204 stipulated a maximum dose of dexamethasone of 4 mg per day (or equivalent) in Cohort B. Many patients with symptomatic brain metastases are managed with >4 mg/day of dexamethasone; in such patients, the efficacy of immunotherapy may be even more limited. In addition, CheckMate 204 required stable neurologic signs/symptoms and generally included patients with limited intracranial disease, warranting concern that patients harboring progressive symptomatology related to more advanced intracranial disease may perform even more poorly. This notion is consistent with the observation that 3 of the 4 patients in Cohort B who responded to dual-agent immunotherapy had a limited intracranial disease burden (median of 2 brain metastases, none >8 mm in diameter). Consequently, among patients with neurologic symptoms and/or steroid requirements in the real world, local therapy may prove necessary. Given the importance of minimizing steroid utilization in patients receiving immunotherapy, neurosurgical resection may prove especially useful in removing bulky intracranial lesions, relieving edema, and facilitating a more prolonged attempt at immunotherapy. For multifocal symptomatic brain metastases, where neurosurgical resection may be less viable, integration of radiation therapy (ideally stereotactically) with immune-based therapy has potential to achieve sustained intracranial disease control.10 In patients with targetable BRAF mutations, combination studies of targeted and immunotherapeutic approaches are underway (NCT04511013), and have particular appeal given the potential for targeted therapy to yield rapid and profound, although typically not durable, responses. Ultimately, however, more data are required to optimally understand the role, sequencing, and integration of local brain-directed therapy with systemic therapy in patients with melanoma and active brain metastases, particularly those who are symptomatic- or steroid-dependent. The investigators from CheckMate 204 and the ABC study should be commended on prospectively evaluating the intracranial efficacy of systemic therapy in patients with active brain metastases. These studies represent a transition from excluding patients with brain metastases from systemic therapy-based trials to including such patients only if brain metastases are small, asymptomatic, stable, and/or radiated, to truly measuring the potential for systemic therapy to control active intracranial disease. CheckMate 204 and the ABC study led to a paradigm shift in the management of patients with melanoma and brain metastases despite a limited number of patients, speaking to the thoughtful design and implementation of both studies. The hope is that more studies of systemic agents with potential to control intracranial disease will follow a similar mold. The text is the sole product of the authors and no third party had input or gave support to its writing. Conflict of interest statement.