Until prospective randomized clinical data accrue and mature, this controversial issue will continue to suffer from overinterpretation of inadequate supporting data.
Oncology (Williston Park). 30(4):314, 316, 317.
Minesh P. Mehta, MBChB
Current management of patients with multiple (≥ 5) brain metastases (MBMs) is in flux and remains controversial, based on increasing emphasis on the use of stereotactic radiosurgery (SRS) for patients with three or fewer brain metastases,[1] and the desire to expand this indication further. Though there have been many retrospective studies showing feasibility and efficacy of SRS for MBMs, few practice-changing messages have emerged from them. Substantial selection biases in these studies exist; patients with poorer performance status and uncontrolled systemic disease are often more likely to receive whole-brain radiation therapy (WBRT), and subsequent interpretations of superior efficacy are therefore flawed. In light of these well-documented caveats, there are very few data to support SRS, especially SRS only, for MBMs at present (a situation that could change pending completion of appropriate randomized trials, some of which are ongoing), a primary reason for the National Comprehensive Cancer Network’s current recommendations for WBRT in patients with more than three brain metastases.[2]
One prospective study has offered the best available data to observationally evaluate overall survival (OS) for patients with 5 to 10 brain metastases treated with SRS only compared with patients with 1 or 2 to 4 brain metastases treated with SRS only.[3] This is clearly not a comparison of SRS with WBRT or SRS with SRS + WBRT; therefore, unbridled enthusiasm for over-interpretation of these data should be curtailed. Even in this trial, the OS for the single brain metastasis group was clearly the highest (13.9 months); there was no significant difference in OS for patients with 2 to 4 or 5 to 10 metastases (10.8 months each), indirectly underscoring the long-held observation that the addition of SRS to WBRT improves survival only in patients with more indolent biology, as represented by patients with a single metastasis, a high Graded Prognostic Assessment (GPA) score, etc. The fact that no survival discrimination in favor of SRS exists for patients with 2 to 10 lesions challenges SRS to become cost-effective, as well as quality-effective, if it is to replace WBRT.
SRS is clearly an effective focal therapy, but unlike WBRT, which categorically decreases failure elsewhere in the brain in patients with MBMs, SRS is simply unable to achieve this endpoint. It has often been promulgated that the consequences of failure elsewhere in the brain are trivial or irrelevant, but genuine, high-quality data supporting such claims are simply not available. In fact, at least two recent studies in the literature, one a re-analysis of a Japanese randomized trial,[4] and the second, a retrospective analysis,[5] suggest that SRS alone actually results in inferior survival. An apparent benefit of SRS is its ability to offer superior short-term neurocognition as compared with WBRT, although long-term data from randomized trials to support this benefit are not yet available in the peer-reviewed literature, and therefore some question and debate persists regarding this issue. Highly promising randomized clinical data using memantine[6] and hippocampal-sparing WBRT[7] could potentially lead to similar posttreatment cognitive outcomes, and ongoing randomized trials will answer these questions. A randomized study from the North American Gamma Knife Consortium for patients with MBMs is testing WBRT vs SRS for patients with up to 10 brain metastases (ClinicalTrials.gov identifier: NCT01731704), although it is unclear to what extent hippocampal-sparing and memantine will be utilized in this trial; in fact, it is likely that if only conventional WBRT is used, it could deliver results that almost instantaneously become less relevant.
Moreover, we posit that the questions and research directions of substantial clinical relevance are actually not nested in the “WBRT vs SRS for x or y number of brain metastases” concept, but rather in appropriate and selected use of both modalities, with a view toward integrating these with the rapid advances in multiple other therapeutic options. For example, rather than blithely accepting the almost 70% or higher likelihood of eventual failure in the brain following SRS alone, we should be considering judicious studies to identify patients at greatest risk of early and multiple lesion failure, and utilizing WBRT (with strategies to diminish neurocognitive sequelae, such as the use of memantine, hippocampal avoidance, the emerging field of striatal-enriched protein tyrosine phosphatase [STEP] inhibitors,[6-8] etc.) more extensively in these patients, while deferring it in others. This might require histology-specific molecular and clinical factor stratification. We should also consider incorporating blood–brain barrier (BBB)-penetrating therapies, specifically in target-enriched clinical scenarios in combination with SRS, to replace WBRT, and simultaneously eliminate the associated unacceptably high elsewhere-in-the-brain failure rates (eg, the SRS plus BBB-penetrating anti-HER2–directed systemic therapy approach based on the LANDSCAPE trial as an illustrative example[9]). Several other options, such as epidermal growth factor receptor–directed therapies for mutated non–small-cell lung cancer (NSCLC), second-generation anaplastic lymphoma kinase (ALK) inhibitors for NSCLC, and others, are now emerging. A third such opportunity would be either SRS or fractionated stereotactic radiotherapy approaches to one or a limited number of brain metastases in combination with immune checkpoint inhibitors to exploit the immunogenic potential of radiation; NRG Oncology is developing one such protocol for melanoma brain metastases. Lastly, it is important to test alternative therapeutic strategies, such as NovoTTF, in combination with SRS to achieve the microscopic disease control so readily attained by WBRT, a concept to be tested in the forthcoming METIS trial. Outside of the clinical trial context, individualized assessment of each patient is of primary importance to guide treatment decisions, such as whether expected patient survival is too short to warrant expensive technologies like SRS alone or MRI surveillance, or if the risk of intracranial relapse from SRS alone puts them at risk for loss of survival, as elegantly demonstrated by Aoyama et al[4] and others, therefore necessitating the incorporation of WBRT.
In summary, until prospective randomized clinical data accrue and mature, this controversial issue will continue to suffer from overinterpretation of inadequate supporting data. This absence of data will, at least for the time being, provide an insufficient evidence base for using SRS alone in patients with MBMs, which remains a problem from the health economics and insurance payment perspective.[10] The importance of careful patient selection for the use of SRS, with WBRT as appropriate, in patients with MBMs cannot be overemphasized. Doing so can offer the best prospects of effective and cost-conscious cancer care in this patient population, aimed at sparing neurocognition, improving survival, and achieving the best possible tumor local control.
Financial Disclosure:Dr. Mehta has served as a consultant for Abbott, Bristol-Myers Squibb, Cavion, Celldex, Elekta, Novartis, Novocure, and Roche; has research funding from Cellectar and Novocure; has served in a leadership capacity on the Pharmacyclics BOD (with stock options); and serves on the Data Monitoring and Safety Board of Monteris. Dr. Verma has no significant financial interest or other relationship with the manufacturers of any products or providers of any service mentioned in this article.
1. Mehta MP, Tsao MN, Whelan TJ, et al. The American Society for Therapeutic Radiology and Oncology (ASTRO) evidence-based review of the role of radiosurgery for brain metastases. Int J Radiat Oncol Biol Phys. 2005;63:37-46.
2. National Comprehensive Cancer Network. NCCN Guidelines for CNS Cancer. www.nccn.org/professionals/physician_gls/f_guidelines.asp#cns. Accessed March 10, 2016.
3. Yamamoto M, Serizawa T, Shuto T, et al. Stereotactic radiosurgery for patients with multiple brain metastases (JLGK0901): a multi-institutional prospective observational study. Lancet Oncol. 2014;15:387-95.
4. Aoyama H, Tago M, Shirato H, et al. Stereotactic radiosurgery with or without whole-brain radiotherapy for brain metastases: secondary analysis of the JROSG 99-1 randomized clinical trial. JAMA Oncol. 2015;1:457-64.
5. Wang TJ, Saad S, Qureshi YH, et al. Outcomes of gamma knife radiosurgery, bi-modality & tri-modality treatment regimens for patients with one or multiple brain metastases: the Columbia University Medical Center experience. J Neurooncol. 2015;122:399-408.
6. Brown PD, Pugh S, Laack NN, et al. Memantine for the prevention of cognitive dysfunction in patients receiving whole-brain radiotherapy: a randomized, double-blind, placebo-controlled trial. Neuro Oncol. 2013;15:1429-37.
7. Gondi V, Pugh SL, Tome WA, et al. Preservation of memory with conformal avoidance of the hippocampal neural stem-cell compartment during whole-brain radiotherapy for brain metastases (RTOG 0933): a phase II multi-institutional trial. J Clin Oncol. 2014;32:3810-6.
8. Goebel-Goody SM, Baum M, Paspalas CD, et al. Therapeutic implications for striatal-enriched protein tyrosine phosphatase (STEP) in neuropsychiatric disorders. Pharmacol Rev. 2012;64:65-87.
9. Bachelot T, Romieu G, Campone M, et al. Lapatinib plus capecitabine in patients with previously untreated brain metastases from HER2-positive metastatic breast cancer (LANDSCAPE): a single-group phase 2 study. Lancet Oncol. 2013;14:64-71.
10. Sullivan R, Peppercorn J, Sikora K, et al. Delivering affordable care in high-income countries. Lancet Oncol. 2011;12:933-80.