By Sunandana Chandra, MD, MS
- The most common malignancies that metastasize to the brain are melanoma and non–small cell lung cancer.
- CNS-directed therapies, including surgery, stereotactic radiosurgery, or whole-brain radiation therapy, have been the mainstay of local brain-directed therapy, but the outcomes from these treatments vary widely.
- The newest data of immunotherapy use in melanoma brain metastases may be practice changing and may obviate or delay stereotactic radiosurgery in the salvage setting.
Metastatic brain tumors are by far the most common intracranial malignancy; approximately 170,000 per year are identified, and they develop in approximately 40% of all patients with advanced cancer.1-3 Patients with advanced malignancies, especially melanoma, have been surviving longer with improved control of extracranial disease because of new targeted and immune therapies that lead to improved survival. This improved survival—along with improved diagnostics, including improved detection of small metastases by MRI—is one of the reasons that the actual incidence of brain metastases has been increasing.
Two of the most common malignancies that metastasize to the brain are melanoma and non–small cell lung cancer (NSCLC).4-7 The lifetime incidence of brain metastases is greater than 50% for patients with metastatic melanoma, and it is 40% to 50% in patients with lung cancer.8,9 The incidence of brain metastases in melanoma may even be higher: on autopsy, at least 70% of patients with melanoma have brain metastases.10-12 Progressive disease in the brain is the major cause of tumor-related death in patients with melanoma. Approximately 10% of patients with NSCLC have brain metastases at initial diagnosis, and an additional 30% of patients develop brain metastases during the course of their disease.13
The features that affect outcome of brain metastases include the primary histology; the number, size, and site of brain metastases; the status of extracranial disease; and the performance status of the patient. Prognostic factors may have the most impact on disease-specific outcomes in patients with brain metastases.14,15
A number of central nervous system (CNS)-directed therapies, including surgery, stereotactic radiosurgery (SRS), or whole-brain radiation therapy (WBRT), have been the mainstay of local brain-directed therapy. The use of WBRT has recently fallen out of favor in some malignancies. However, the optimal treatment of multiple brain metastases is not well defined and remains inadequate. The paucity of data on the management of CNS metastases is compounded by exclusion of patients with brain metastases—even when asymptomatic—from participation in clinical trials, which limits outcomes data even more in this patient population. Many patients who have controlled systemic disease frequently develop progressive or recurrent brain metastases.
No single standard-of-care option for treatment exists for patients with multiple brain metastases. WBRT has limited efficacy and harms quality of life. In a small population of 86 patients with melanoma, those treated with WBRT alone had decreased median survival rates compared with SRS or surgery. The outcomes of the patients with brain metastases varied widely. In those treated with WBRT alone, the median survival was 3 months; in those treated with SRS, the median survival was 7 months; and surgery was associated with a median survival of 11 to 12 months.15
One of the challenges of treating brain metastases is that the blood–brain barrier leads to poor permeability of many drugs. The blood–brain barrier is a highly selective permeability barrier composed of capillary endothelial cells, which are connected by tight junctions and astrocyte foot processes that limit diffusion of molecules and drugs into the CNS compartment.16 In addition, because brain metastases tend to occur later in the disease course, these tumors may be biologically and genomically distinct and more complex from the primary site or extracranial metastatic sites and therefore may be more resistant to particular therapies.17,18
Small-molecule inhibitors, such as BRAF and MEK, have shown good intracranial responses in patients with BRAF-mutant melanoma with brain metastases.19 The immune checkpoint inhibitors anti–CTLA-4 and anti–PD-1 are large monoclonal antibodies that were considered unlikely to cross an intact blood–brain barrier. However, a small meta-analysis (four studies) showed that, when melanoma brain metastases were treated with SRS alone versus SRS and ipilimumab, a survival benefit was seen with the combination.20
A retrospective review of data from a single institution in which patients treated with immunotherapy and SRS versus those treated with SRS alone were evaluated showed that those who were treated with both modalities experienced higher rates of radionecrosis (13.8%) and intracranial hemorrhage (15.5%).21 However, those who received immunotherapy and SRS had a lower risk of death, albeit with increased risk of brain toxicity, than did those who received SRS alone.
A study presented at the 2017 ASCO Annual Meeting revealed that radionecrosis was a significant toxicity in patients with melanoma and brain metastases who were treated with an anti–PD-1 antibody and radiotherapy (either SRS or WBRT).22 Of the 118 patients evaluated, 85% were treated with pembrolizumab; 10% with nivolumab; and 5% with combination ipilimumab and nivolumab. Overall, 69% of patients underwent SRS; 19% had WBRT; and 12% had both. The median progression-free survival and overall survival times were 24 months and 46 months, respectively. Eighteen percent of patients developed radionecrosis: 17% after SRS; 9% after WBRT; and 36% after both modalities. In this study, factors that may have led to a higher incidence of radionecrosis were elevated LDH, prior ipilimumab therapy, and more fractions of WBRT.
A previous study of patients with melanoma and brain metastases treated with high-dose ipilimumab for four doses followed by maintenance high-dose ipilimumab were evaluated in two parallel cohorts: (A) those who were neurologically asymptomatic, and (B) those who were symptomatic and on a stable dose of corticosteroids. Of 51 patients in cohort A, and 21 patients in cohort B, intracranial control was exhibited in 12 and two patients (24% and 10%), respectively.23 Extracranial disease control also was noted in 14 patients and one patient (27% and 5%) in the two cohorts, respectively.23
More recently, a phase II study showed that the anti–PD-1 antibody nivolumab alone versus nivolumab in combination with the anti–CTLA-4 antibody ipilimumab are both active in patients with melanoma and brain metastases; the intracranial response was 44% in asymptomatic patients with brain metastases and no prior local therapy treated with combination therapy versus 20% in patients treated with nivolumab alone.24 In patients whose disease failed to respond to local therapy, who were neurologically symptomatic, and/or who had leptomeningeal disease, the intracranial response was 6%. The 6-month progression-free survival and overall survival times were greater in those treated with combination immunotherapy than with an individual drug. Patients who experienced disease progression with BRAF inhibitor therapy did not respond as well to combination immunotherapy, and patients who had been treated with prior brain-directed therapy and those with leptomeningeal disease did not respond well to nivolumab alone.
Data from a separate phase II trial presented at the 2017 ASCO Annual Meeting showed that, when asymptomatic patients with melanoma and at least one brain metastasis were treated with combination nivolumab and ipilimumab (with SRS allowed if there was progression of a single brain metastasis), there was a 56% intracranial objective response rate for greater than 6 months.25 Approximately 19% had a complete response; furthermore, the intracranial and extracranial responses generally were correlated.25
The newest data of immunotherapy use in melanoma brain metastases may be practice changing and may obviate or delay SRS in the salvage setting given the rare but often significant morbidities, such as those seen with radiation necrosis and hemorrhage, associated with these local therapies.
About the Author: Dr. Chandra is an assistant professor of Hematology/Oncology at Northwestern University in Chicago.