Counterbalancing risks and gains from extended resections in malignant glioma surgery: a supplemental analysis from the randomized 5-aminolevulinic acid glioma resection study. Clinical article

Object Accumulating data suggest more aggressive surgery in patients with malignant glioma to improve outcome. However, extended surgery may increase morbidity. The randomized Phase III 5-aminolevulinic acid (ALA) study investigated 5-ALA–induced fluorescence as a tool for improving resections. An interim analysis demonstrated more frequent complete resections with longer progression-free survival (PFS). However, marginal differences were found regarding neurological deterioration and the frequency of additional therapies. Presently, the authors focus on the latter aspects in the final study population, and attempt to determine how safety might be affected by cytoreductive surgery. Methods Patients with malignant gliomas were randomized for fluorescence-guided (ALA group) or conventional white light (WL) (WL group) microsurgery. The final intent-to-treat population consisted of 176 patients in the ALA and 173 in the WL group. Primary efficacy variables were contrast-enhancing tumor on early MR imaging and 6-month PFS. Among secondary outcome measures, the National Institutes of Health Stroke Scale (NIH-SS) score and the Karnofsky Performance Scale (KPS) score were used for assessing neurological function. Results More frequent complete resections and improved PFS were confirmed, with higher median residual tumor volumes in the WL group (0.5 vs 0 cm3, p = 0.001). Patients in the ALA group had more frequent deterioration on the NIH-SS at 48 hours. Patients at risk were those with deficits unresponsive to steroids. No differences were found in the KPS score. Regarding outcome, a combined end point of risks and neurological deficits was attempted, which demonstrated results in patients in the ALA group to be superior to those in participants in the WL group. Interestingly, the cumulative incidence of repeat surgery was significantly reduced in ALA patients. When stratified by completeness of resection, patients with incomplete resections were quicker to deteriorate neurologically (p = 0.0036). Conclusions Extended resections performed using a tool such as 5-ALA–derived tumor fluorescence, carries the risk of temporary impairment of neurological function. However, risks are higher in patients with deficits unresponsive to steroids.