Clinical Management of Multiple Melanoma Brain Metastases: A Systematic Review

Importance  The treatment of multiple brain metastases (MBM) from melanoma is controversial and includes surgical resection, stereotactic radiosurgery (SRS), and whole-brain radiation therapy (WBRT). Several new classes of agents have revolutionized the treatment of metastatic melanoma, allowing some subsets of patients to have long-term survival. Given this, management of MBM from melanoma is continually evolving.

Objective  To review the current evidence regarding the treatment of MBM from melanoma.

Evidence Review  The PubMed database was searched using combinations of search terms and synonyms for melanoma, brain metastases, radiation, chemotherapy, immunotherapy, and targeted therapy published between January 1, 1995, and January 1, 2015. Articles were selected for inclusion on the basis of targeted keyword searches, manual review of bibliographies, and whether the article was a clinical trial, large observational study, or retrospective study focusing on melanoma brain metastases. Of 2243 articles initially identified, 110 were selected for full review. Of these, the most pertinent 73 articles were included.

Findings  Patients with newly diagnosed MBM can be treated with various modalities, either alone or in combination. Level 1 evidence supports the use of SRS alone, WBRT, and SRS with WBRT. Although the addition of WBRT to SRS improves the overall brain relapse rate, WBRT has no significant impact on overall survival and has detrimental neurocognitive outcomes. Cytotoxic chemotherapy has largely been ineffective; targeted therapies and immunotherapies have been reported to have high response rates and deserve further attention in larger clinical trials. Further studies are needed to fully evaluate the efficacy of these novel regimens in combination with radiation therapy.

Conclusions and Relevance  At this time, the standard management for patients with MBM from melanoma includes SRS, WBRT, or a combination of both. Emerging data exist to support the notion that SRS in combination with targeted therapies or immune therapy may obviate the need for WBRT; prospective studies are required to fully evaluate the efficacy of these novel regimens in combination with radiation therapy.

There are 170 000 to 200 000 new cases of brain metastases diagnosed each year, and 20% to 40% of patients with cancer will develop brain metastases.^1,2Quiz Ref ID Brain metastases are especially important in the context of malignant melanoma because 44% of patients with metastatic melanoma will develop symptomatic brain metastases, and intracranial disease accounts for 20% to 54% of deaths in patients with melanoma.^3-5 Prognosis is poor, and expected survival ranges from 3 to 13 months despite radiation therapy (RT).⁶ Whereas the clinical management of single metastases with craniotomy and/or stereotactic RT is well established, the treatment of patients with multiple brain metastases (MBM) is not well defined. Patients with newly diagnosed MBM can be treated with various modalities, either alone or in combination. Level 1 evidence supports the use of stereotactic radiosurgery (SRS) alone, whole-brain radiation therapy (WBRT) alone, and SRS with WBRT.^7-12 The rationale for WBRT is that patients with undetectable metastases after SRS may later develop additional brain lesions if left untreated.

Since 2011, the US Food and Drug Administration (FDA) has approved 7 new drugs for the systemic therapy of metastatic melanoma. Targeting mutations in the BRAF-MEK-ERK mitogen-activated protein kinase (MAPK) pathway has demonstrated significant improvements in progression-free and overall survival in randomized clinical trials (RCTs).^13-18 In addition, advances in tumor immunotherapy largely focused on cytokines and T-cell checkpoint inhibitors have shown durable therapeutic responses and emerging confirmation of overall survival benefits in RCTs.^19-21 These novel therapies have prolonged survival in a disease that previously had a dismal outcome. As patients are living longer as a result of more effective systemic therapy, surveillance and management of intracranial disease is of increasing importance. Emerging data support the activity of some targeted and immunotherapy agents in MBM, which suggests that control of MBM may be possible with aggressive multimodality management. Reports of an abscopal effect when RT is used in combination with immunotherapy also suggest that combination approaches may be especially interesting therapeutic strategies, and MBM represents a clinical scenario that is highly appropriate for testing these combinations.²²

Herein, we describe the contemporary peer-reviewed literature with high-quality data related to the clinical management of multiple melanoma brain metastases. On the basis of the review of the data, emerging recommendations for patient management are discussed and future areas of interest for clinical investigation are proposed. Given the advances in systemic therapy of melanoma, it is critical that oncologists treating these patients be aware of new treatment paradigms to optimize the outcomes for all patients with metastatic melanoma.

Box Section Ref ID

A literature search of PubMed was conducted by one of us (S.G.) using combinations of search terms and synonyms for melanoma, brain metastases, radiation, chemotherapy, immunotherapy, and targeted therapy in articles published between January 1, 1995, and January 1, 2015. Studies in languages other than English or involving animals and children were excluded. Articles were identified (n = 2243) from which titles and abstracts were examined to eliminate studies without evidence-based data, such as case reports, dosimetry studies, reviews, and studies of other cancers. All remaining articles were screened carefully; clinical trials, large observational studies, and studies focusing on melanoma brain metastases received priority in the selection process. Bibliographies of these studies were searched for other relevant studies. Initially, 110 articles were identified; duplicate studies or studies that did not meet these criteria on full review were then excluded (n = 37). Of these, the most pertinent 73 articles were selected for inclusion.

The results were reviewed by a multidisciplinary team composed of medical, surgical, and radiation oncologists. Critical issues were identified and key findings from the current literature are summarized in this report.

In an attempt to identify prognostic factors in patients with brain metastases, the Radiation Therapy Oncology Group (RTOG) conducted a recursive partitioning analysis that stratified patients with brain metastases from any histologic subtype into 3 groups using age, Karnofsky performance status (KPS) greater than 70, and status of systemic metastases. The most favorable group had a median survival of 7.1 months, and the least favorable group had a median survival of 2.3 months.²³Quiz Ref ID Sperduto et al⁶ developed a Graded Prognostic Assessment providing a disease-specific classification of outcomes in patients with brain metastases. In this classification, the only factors that affected outcomes in patients with melanoma brain metastases were KPS and number of metastases. In 86 patients with melanoma brain metastases receiving WBRT alone, the median survival was 3 months. In contrast, SRS was associated with a median survival of 7 months whereas surgery was correlated with a median survival of 11 to 12 months.⁶ Although these data were not based on RCTs and may be subject to selection bias, they do provide some insight into important prognostic variables in predicting outcome. In addition, it is notable that the melanoma Graded Prognostic Assessment was not influenced by the presence of extracranial disease or age, implying that aggressive management of intracranial disease is reasonable even in older patients with active systemic disease.

Advances in neurosurgical technique such as awake craniotomy, functional monitoring, and intraoperative magnetic resonance imaging (MRI) have revolutionized modern neurosurgery in an effort to improve gross total resection and reduce surgical morbidity. In patients with MBM, surgical resection is necessary for those patients who need a histologic diagnosis, whose neurological symptoms do not resolve with supportive care, or who have a dominant lesion deemed too large to be treated with SRS. Outside this context, the role of surgical intervention in patients with MBM remains controversial given factors involved in operative decision making such as the performance status and comorbidities of the patient, the size and accessibility of the lesion, and the proximity of the lesion to eloquent areas. To our knowledge, there have been no randomized studies of surgery in patients with MBM and as such, RT has been standard treatment in the management of MBM. One RCT currently open to enrollment is randomizing patients with 1 to 4 brain metastases to SRS or WBRT after resection of at least 1 of the lesions (NCT01372774).

Traditionally, the treatment of MBM has included WBRT and corticosteroids.^11,24 Meta-analyses of WBRT clinical reports have demonstrated that median survival is not altered with varying dosage, fractionation scheme, or overall treatment time.^25,26 The most commonly used regimens are 35 Gy (to convert to rad, multiply by 100) delivered in 2.5-Gy fractions during 14 treatment days or 30 Gy in 3-Gy fractions during 10 treatment days. Quiz Ref IDThe potential neurocognitive toxic effects related to WBRT include deterioration of verbal fluency, fine motor skills, immediate recall, and delayed recall. Sun et al²⁷ provided evidence that radiotherapy itself causes neurocognitive decline, independent of tumor burden as scored by the Hopkins Verbal Learning Test (HVLT) for immediate and delayed recall as early as 3 months in patients with lung cancer without evidence of brain metastases who were treated with prophylactic cranial irradiation.

Methods to reduce the potential for neurocognitive decline have been a focus of contemporary WBRT trials.^28-30 The RTOG 0933 study was a single-arm, phase 2 study investigating hippocampal-sparing WBRT in patients with brain metastases.²⁹ Irradiation of the subgranular zone of the hippocampal dentate gyrus, which contains neural stem cells, is thought to suppress the formation of new memories and impair recall.²⁹ With a primary end point of the HVLT–Revised Delayed Recall at 4 months, patients receiving hippocampal-sparing WBRT experienced a mean relative decline of 7.0%, which was significantly less than the 30% decline observed in historical controls (P < .001). Another large RCT, RTOG 0614, investigated the addition of memantine, an N-methyl-D-aspartate receptor antagonist, to WBRT.²⁸ Although the primary end point of HVLT–Revised Delayed Recall was not statistically significant, many of the secondary end points showed that memantine therapy delayed time to cognitive decline and reduced the rate of decline in memory, executive function, and processing speed compared with placebo.

Quiz Ref IDThere have been 2 RCTs that have shown that the addition of SRS to WBRT in patients with 1 to 2 or 3 to 4 lesions reduced the risk of central nervous system (CNS) relapse but did not portend a survival advantage compared with patients receiving WBRT alone (Table 1).^7,12 The largest study, RTOG 9508, randomized 333 patients with 1 to 3 brain metastases to WBRT with or without SRS, including 14 patients with melanoma.⁷ In the entire cohort, they found a 43% reduction in local failure with the addition of SRS to WBRT; this failed to translate into an overall survival benefit. These findings were confirmed in a meta-analysis of these studies.²⁶ Interestingly, patients receiving WBRT+SRS noted an improvement in KPS and decreased steroid use. It is important to note that neither study reported neurocognitive outcomes after WBRT or WBRT+SRS.

Although WBRT has been considered standard of care with MBM, neurologic toxicity and limited sustained local efficacy have called into question its use as initial therapy for MBM. Stereotactic radiosurgery has several advantages over WBRT, and changes in treatment guidelines have mirrored its increasing use.^34,35 Stereotactic radiosurgery is an especially attractive modality in the context of melanoma because systemic therapy is not delayed or disrupted for single-session SRS. Because of limited tumor control after WBRT, intracranial melanoma metastases were traditionally considered resistant to fractionated therapy. However, this limitation is not applicable to treatment with SRS. Local tumor control rates range from 70% to 80% after SRS, suggesting that SRS has the potential to overcome several limitations of fractionated RT.^8-10

There have been 2 RCTs that have shown that in patients with single or a limited number (3-4) of lesions SRS plus WBRT reduced the risk of CNS relapse without an improvement in overall survival.^8,10 A third RCT by Chang et al⁹ found an overall survival benefit to SRS alone compared with SRS plus WBRT, a secondary end point; the primary end point of this study was HVLT–Revised Total Recall. A meta-analysis of these 3 trials confirmed the reduced risk of CNS relapse, both local and distant, in addition to the lack of a survival advantage in patients receiving SRS plus WBRT compared with SRS alone.²⁶ A second meta-analysis using patient-level data noted a survival improvement in patients younger than 50 years who received SRS alone.³⁶ These data indicate that CNS relapses after SRS alone can be salvaged with either WBRT or SRS effectively without a detriment in survival.

Furthermore, the RCT by Chang et al⁹ revealed that the omission of WBRT improved neurocognitive function, and the other 2 RCTs^8,10 found significantly improved health-related quality-of-life measures and functional independence when omitting WBRT. These results suggest that cognitive function in patients with brain metastases declines faster because of adverse effects of WBRT rather than disease progression; however, this topic remains controversial. Given this, advocates of SRS believe strongly that WBRT should be omitted given the associated neurocognitive toxic effects seen in patients, particularly in those younger than 50 years old.³⁶ It should also be noted that patients with melanoma only made up 5% to 10% of the enrolled patients in all 3 RCTs of SRS with or without WBRT. Thus, we extrapolate the local control, survival, and neurocognitive data when applying it in the context of MBM from melanoma.

There are no RCTs investigating the role of SRS or WBRT in patients with 4 or more metastatic lesions in the brain. A recently published large prospective observational cohort of 1194 patients with MBM from various tumor types reported that the overall survival of patients with 2 to 4 brain metastases was noninferior to patients with 5 to 10 metastases (median survival, 10.8 months for each group).³⁷ These results suggest that SRS alone may be a viable treatment option even in patients with up to 10 metastases. Another study by the same group sought to determine whether outcomes from SRS alone for patients with 10 or more tumors differed from those for patients with 2 to 9 lesions (n = 1814) and found that the median survival time, neurological death–free survival times, neurological deterioration, and SRS-related complications were not statistically different between the 2 cohorts.³⁸ Neither study was focused solely on melanoma metastases, making histology-specific conclusions problematic. Nevertheless, improvements in technology have made tumor localization and measurement more accurate; this, coupled with the improved technical skills of neurosurgeons and radiation oncologists engaged in SRS, may be improving tumor control.

Melanoma has long been considered a radioresistant tumor, and more recent preclinical data suggest that melanoma cell lines show high levels of DNA damage repair at conventional fraction sizes and increased cell death with larger doses per fraction.³⁹ Many oncologists advocate for the omission of WBRT given the relative radio-insensitivity to standard fraction sizes. To date, there have been no reported RCTs of WBRT compared with SRS for patients with MBM from melanoma, although one study is currently enrolling patients.⁴⁰ There are, however, many RCTs of patients with brain metastases from mixed histologic subtypes that help provide insight into the role of RT for MBM.

There are instances in which an MBM will recur inside the field of treatment following maximum radiation exposure. In patients whose systemic disease is under relatively good control, the treatment of the in-field recurrence becomes the priority. A long-standing dilemma is the need to differentiate the recurrent brain metastasis from progressive radiation necrosis. Patients may present with similar radiographic imaging features and progressive symptoms; even advanced imaging modalities often fail to differentiate the 2 entities.^41,42 Surgical resection is typically indicated for those situations in which the diagnosis is unclear or for those patients whose disease is refractory to corticosteroid therapy.⁴³ Bevacizumab, a VEGF inhibitor, has also been shown to reverse the effects of radiation necrosis.⁴⁴ The use of MRI-guided laser therapy has reemerged as a potentially useful tool for in-field recurrences for which surgery is not possible, particularly because of advances in real-time magnetic resonance thermometry. Rao et al⁴⁵ demonstrated a greater than 90% control rate in lesions that exhibited progressive growth on 2 MRI scans; in all cases, a differentiation between a recurrent brain metastasis and radiation necrosis was not made. If a treatment modality exists that treats both recurrent brain metastasis and progressive radiation necrosis effectively, then the differentiation between the two may not be critical for clinical management.

The role of chemotherapy and radiation sensitizers in the management of patients with MBM is limited. Presumably, the failure of cytotoxic therapy is due to the low activity of the drugs and the limited penetration of the drugs across the blood-brain barrier (BBB). The BBB is a protective layer of tightly joined endothelial cells reinforced by the astrocyte foot processes called the glia limitans. There have been several RCTs in patients with MBM from mixed tumor types investigating various cytotoxic and radiosensitizing agents in combination with WBRT with negative results^46-50 (Table 2). Use of available cytotoxic treatments that can cross the BBB, such as temozolomide and fotemustine, has not resulted in significant improvements in either intracranial disease control or overall survival in patients with melanoma MBM (Table 2).^48,52-59 For example, temozolomide alone or in combination with WBRT resulted in a progression-free survival of only 1 to 2 months, which is comparable to the expected prognosis with supportive care only.

BRAF kinase is an intracellular enzyme that leads to increased signaling of the MAPK pathway. BRAF mutations occur in 50% of melanoma cases, providing a useful target for treatment in these patients. Vemurafenib is an oral inhibitor of mutated BRAF kinase that specifically targets mutations resulting from substitution of glutamic acid for valine at codon 600, known as BRAF V600E. Vemurafenib therapy was associated with improvements in overall survival for patients with MBM in an RCT.¹³ In a small pilot study for brain metastases, vemurafenib therapy resulted in a partial response rate of 16%, and the median survival was only 5.3 months.⁶⁰ In another study of patients who received SRS and concurrent vemurafenib, 58% of patients developed new brain metastases outside the irradiated volume.⁶¹ Furthermore, in patients without known brain metastases who are treated with vemurafenib, the brain is a common (20%-25%) site of treatment failure.⁶² Taken together, these data suggest that vemurafenib alone is not adequate for treatment of brain metastases.

Dabrafenib is another inhibitor of the mutated BRAF kinase, and a large phase 2 study of 172 patients demonstrated its efficacy in patients with melanoma metastatic to the brain, which is the largest prospective trial in patients with melanoma brain metastases to date.⁶³ In patients with previously untreated brain metastases who carried BRAF V600E mutations, 39% (95% CI, 28%-51%) achieved an overall intracranial response and the median overall survival was 7.7 months. Dabrafenib has also demonstrated enhanced clinical activity when combined with trametinib, a selective MEK inhibitor.¹⁴ Currently, NCT02039947 is under way to test the combination of a BRAF and MEK inhibitor specifically in patients with BRAF mutation–positive melanoma brain metastases.

Unlike cytotoxic or targeted agents, which must cross the BBB to be effective, it is not necessary for immunotherapy agents to gain direct access to the brain parenchyma to have therapeutic effect. Insights from recent experiences in autoimmune disease support this hypothesis.^64,65 Interleukin 2 (IL-2) and checkpoint inhibitors induce activation of T cells in the extracranial compartment, and the activated T cells can then enter the brain, a process that is mediated by P-selectin and α4 integrin.⁶⁵

Interleukin 2 is a cytokine produced primarily by CD4⁺ T cells that is necessary for T-cell and natural killer cell growth and differentiation. High-dose IL-2 has been used therapeutically in metastatic melanoma and is associated with a consistent objective response rate of approximately 16%, and many of the responses are durable for decades.⁶⁶ Administration of IL-2 requires hospitalization in specialized centers with experience in IL-2 management because of substantial toxic effects, including capillary leak syndrome.⁶⁷ Capillary leak syndrome results in generalized edema, which is a safety concern in patients with brain metastases because of the possibility of increased intracranial pressure. There are no RCTs or prospective studies investigating the role of IL-2 therapy in patients with MBM. There are limited retrospective data regarding the administration of high-dose IL-2 in melanoma patients with brain metastases.^67,68 In general practice, IL-2 is typically considered after control of MBM through RT has been achieved and only when intracerebral edema is minimal and patients do not require corticosteroid administration.

Cytotoxic T-lymphocyte antigen 4 (CTLA-4) is expressed on T cells, and it normally impedes the activation of T cells by antigen-presenting cells, acting as a checkpoint in the immune response. Ipilimumab is a humanized monoclonal antibody that blocks CTLA-4; therefore, inhibition of CTLA-4 signaling leads to continued T-cell activation and proliferation. Ipilimumab therapy has demonstrated an improvement in overall survival in an RCT and was approved by the FDA for the treatment of advanced and metastatic melanoma in 2011.²⁰ As a monoclonal antibody, Quiz Ref IDipilimumab does not appreciably cross the BBB. However, T cells that have been activated in the periphery are able to migrate into the CNS,⁶⁵ which may also occur in patients treated with ipilimumab.

There are no RCTs investigating the role of ipilimumab therapy in patients with melanoma MBM. Two prospective phase 2 studies found that ipilimumab has activity among patients with brain metastatic melanoma (Table 3).^19,69 The first study evaluated the use of ipilimumab and fotemustine in combination and found that the complete CNS response rate among patients with previously untreated brain metastases was 38%.¹⁹ The second study evaluating the efficacy of ipilimumab enrolled 72 patients, including 51 patients who were neurologically asymptomatic.⁶⁹ In this group, the CNS response rate was 16% and the median overall survival was 7.0 months. These data suggest that ipilimumab may have clinically relevant disease activity in the CNS, and additional studies are needed to clarify its role in these patients.

Similar to CTLA-4, the programmed cell death-1 (PD-1) is a T cell checkpoint receptor expressed on recently activated T cells, B cells, and myeloid cells and mediates inhibition of immune cell effector functions on binding to its natural ligands PD-L1 and PD-L2. Tumors coopt the PD-1 pathway by expressing the ligands PD-L1 or PD-L2, enabling them to evade immunosurveillance by inhibition of PD-1–expressing T cells within the tumor microenvironment. On the basis of preliminary data showing a 24% systemic objective response rate in an advanced patient population, pembrolizumab, a humanized monoclonal antibody against PD-1, achieved accelerated, conditional approval by the FDA for the treatment of metastatic melanoma in patients who have been previously treated with ipilimumab (and if the tumor is BRAF mutated, a BRAF inhibitor).²¹ Nivolumab, also known as BMS-936558, is the second fully humanized IgG4 monoclonal antibody against PD-1 to be conditionally approved by the FDA for second- or third-line treatment of metastatic melanoma. Although there are no completed studies of PD-1–blocking antibodies performed specifically in a brain metastases population, there are multiple studies under way (NCT02320058, NCT02374242).

Radiotherapy has the potential for synergizing with immunotherapy by increasing the permeability of the BBB, stimulating cytokine release, and increasing antigen presentation.⁷³ Combination treatment with SRS and immunotherapy is supported by multiple clinical studies.^22,74 Three retrospective single-institution studies suggest that ipilimumab in combination with RT may be more effective than RT alone in patients with brain metastases (Table 3).^70-72 In all 3 studies, approximately 35% of patients received ipilimumab prior to RT whereas 65% received the drug after RT. Notably, responses to ipilimumab therapy in the brain, as in extracranial disease, have been reported to be durable beyond 4 years.⁷⁵ The term abscopal effect (from the Latin “ab” and Greek “scopus”) denotes a phenomenon of tumor regression at sites that are remote from an irradiated target.²² The abscopal effect is uncommon, but it has been described with RT to the body and the brain.

In summary, concurrent treatment with ipilimumab and SRS seems to be safe and may increase efficacy of SRS; this practice is supported by consensus in the National Comprehensive Cancer Network guidelines.⁷⁶ At present, mature studies are needed to determine whether immunotherapies in conjunction with SRS have the potential to supplant WBRT to control microscopic intracranial disease. At least 3 ongoing studies will combine SRS with ipilimumab for treatment of brain metastases in patients with advanced melanoma (NCT01950195, NCT01703507, and NCT02097732).

At this time, the standard management for patients with MBM from melanoma includes SRS, WBRT, or a combination of both. Stereotactic radiosurgery is taking hold as the preferred treatment for metastases in the brain because it avoids the neurologic decline experienced with WBRT. Cytotoxic chemotherapy regimens have largely been ineffective without evidence for survival benefit. Emerging data exist to support the notion that SRS in combination with immune checkpoint inhibitors may be an effective treatment, and prospective studies are required to fully evaluate the efficacy of these novel combination regimens and whether they can supplant the use of WBRT. The improvements in survival for patients with melanoma seem to be affecting those with MBM, and patients should be evaluated by a multidisciplinary team to individualize the therapeutic approach and maximize clinical benefit. Further clinical investigation will help provide definitive evidence-based data for optimizing the treatment of patients with melanoma with MBM.

Accepted for Publication: March 30, 2015.

Corresponding Author: Sharad Goyal, MD, Department of Radiation Oncology, Rutgers Cancer Institute of New Jersey, 195 Little Albany St, New Brunswick, NJ 08903 (goyalsh@rutgers.edu).

Published Online: May 21, 2015. doi:10.1001/jamaoncol.2015.1206.

Author Contributions: Drs Goyal and Silk had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

Study concept and design: Goyal, Tian, Kaufman.

Acquisition, analysis, or interpretation of data: Goyal, Silk, Mehnert, Danish, Ranjan, Kaufman.

Drafting of the manuscript: All authors.

Critical revision of the manuscript for important intellectual content: Goyal, Silk, Danish, Kaufman.

Administrative, technical, or material support: Goyal, Kaufman.

Study supervision: Goyal, Kaufman.

Conflict of Interest Disclosures: Dr Kaufman serves as a consultant for Alkermes, Amgen, EMD Serono, Prometheus, and Sanofi, has received research funding from Bristol-Myers Squibb, and serves on the speaker’s bureau for Merck. No other disclosures are reported.