Commentary: The Effects of Postoperative Neurological Deficits on Survival in Patients With Single Brain Metastasis

In Medikonda et al,¹ “The effects of postoperative neurological deficits on survival in patients with single brain metastasis”, the authors describe significantly worse outcomes in patients with new postoperative neurological deficits, relative to those without, after surgery for a single brain metastasis. They report a hazard ratio of 2.45 and a decreased median survival of 2.4 vs 12.6 mo, respectively. With a rising number of cancer patients developing brain metastases, this work is powerful and important.² It follows recent studies by Rahman et al³ and McGirt et al⁴ who similarly described potential setbacks in overall survival among glioblastoma patients experiencing functional deficits after surgery. In the glioblastoma experience, any postoperative deficit led to decreased survival (9.2 vs 14.7 mo).³ More specifically, patients with postoperative motor deficits saw a lower two-year survival, relative to patients with postoperative language deficits and patients with no deficits (0%, 8%, and 23%).⁴ This paper by Medikonda et al¹ is the first to describe similar results for brain metastases patients, which reminds us to weigh the risks and benefits of brain metastases surgery in the current age of targeted therapies and stereotactic radiosurgery (SRS).

Resection of single brain metastases followed by whole brain radiation (WBRT) has long been standard of care after the landmark manuscript by Patchell in 1990.⁵ Since then, medical and radiosurgery (SRS) treatments have increasingly been a part of metastatic central nervous system (CNS) disease. Better systemic therapies have led to increased overall survival in most cancer types, and advances in SRS have replaced WBRT.^6,7

The success of immunotherapy in many cancer types and targeted therapies in other cancer types has changed the landscape of current brain metastases practice. Checkpoint inhibitor treatment of melanoma metastasis to the brain has led to cohort survivals as long as 2 plus years, and durable disease control in about one-third of the patients.^8-10 Likewise, EGFR and ALK/ROS1 inhibitor treatment of nonsmall cell lung cancer with CNS spread have been successful. Early alectinib trial data for the ALK-mutated patients have shown median progression free survival of about 34 mo, and an overall CNS response rate of 79%.^11-13 Further gains can be expected as combination therapies and sequential regimens are optimized. However, toxicities such as immune related adverse events are common, with a quarter of patients experiencing severe grade events that merit hospitalization.¹⁴

SRS outcomes and safety have been extensively evaluated since the publication of the Patchell et al study.⁵ Randomized studies have demonstrated equivalent survival but improved cognitive outcomes in patients with up to 4 brain metastases treated with upfront SRS instead of WBRT.¹⁵ Many groups are looking at the benefit of treating patients with up to 10 to 15 brain metastases with upfront radiosurgery.

Even though surgery will clearly continue to be indicated for removal of many symptomatic lesions, this manuscript reminds us that a postoperative neurological deficit has survival consequences. In addition to postoperative morbidity, leptomeningeal disease (LMD) seeding has been associated with upfront surgical management for metastatic disease.^16-18 Various mechanisms have been proposed, including spread from piecemeal dissection or spread due to proximity to CSF flow.^19,20 In a recent prospective study of 320 patients, Nguyen et al¹⁸ described a 1- and 2-yr LMD rate of 20% and 24% vs 6% and 10% for hypofractionated SRS to surgical cavities, compared to intact metastases. Such findings have led to considerations for neo-adjuvant SRS to “sterilize” the anticipated surgical.¹⁶ Some primary pathologies are at greater, cited risks for LMD spread such as breast cancer or tumors in the posterior fossa while targeted or immunotherapies have been seen as protective, or even a potential treatment option for LMD.^17,18,21-24

To produce future iterations of Medikonda et al¹ and McGirt et al,⁴ complications reporting will require more transparency about the procedure and people studied. First, as seen with the featured article, informative data needs to pair individual postoperative deficits with outcomes. Many prior clinical studies have commonly tallied postoperative complications, but subsequently reported clinical outcomes only for the aggregated cohort. The consequence has been losses in reporting-granularity and exclusion from future, potentially revealing meta-analyses. With better systemic therapies now available, physicians will need to assess quality of life outcomes after surgical, medical, and radiation interventions.

These early deficit-associated outcome studies also highlight how a common classification scheme is needed for early validation. The current variations in reporting practices make it difficult to resolve incongruencies between conclusions. Using hemiparesis as an example, we see Medikonda et al¹ defined permanent motor weakness as a deficit sustained after four weeks. Conversely, permanent weakness was assigned by McGirt et al⁴ for glioblastoma patients after 6 months. Additional knowledge of the location of these lesions would clarify if attributing permanency is appropriate. For instance, transient weakness may be expected for up to 6 mo after operating around the supplementary motor cortex.²⁵ Additionally, the authors defined hemiparesis as a major, postoperative deficit, and excluded aphasia. Conversely, McGirt et al⁴ demonstrated for glioblastoma language deficits had a more significant impact on survival than motor deficits.

The featured article is an exciting reminder of the evolving landscape for metastatic brain tumors and surgery's contribution. Survival after brain metastasis has exhibited a substantial increase over the past decade, in an increasingly growing field of promising treatment options. The extent to which surgical resection plays a role in the future of metastatic disease management greatly depends on how we measure its safety and value. The heterogeneous clinical history of brain metastasis needs to be systematically captured in the data. Deficits need to be linked with individuals’ outcomes. Goals of care need to be acknowledged. Only thereafter will our cohorts be conducive towards generalizing findings, measuring effect sizes, and comparing between studies. Most importantly we will then have accurately accounted for survival quality and enable this generation of neurosurgery to redefine metastatic disease management.

Funding

This study did not receive any funding or financial support.

Disclosures

The authors have no personal, financial, or institutional interest in any of the drugs, materials, or devices described in this article. Dr Li receives grant support from NIH, Bristol Myers Squibb, and Novocure, has a consulting agreement with Medtronic, and participates on the speaker's bureau for Johnson & Johnson.