Association of aggressive resection with survival and progression-free survival in adult low-grade glioma: A systematic review and meta-analysis with numbers needed to treat.
Central Nervous System Tumors
Central Nervous System Tumors
2017 ASCO Annual Meeting
Poster Board Number:
Poster Session (Board #267)
J Clin Oncol 35, 2017 (suppl; abstr 2025)
Author(s): Timothy J Brown, Daniela Annenelie Bota, Elizabeth A. Maher, Dawit Gebremichael Aregawi, Linda M. Liau, Paul D. Brown, Jan C. Buckner, Michael Weller, Martin J. Van Den Bent, Mitchel S. Berger, Michael J. Glantz; The University of Texas Southwestern Medical Center, Dallas, TX; University of California Irvine Medical Center, Orange, CA; Penn State Milton S. Hershey Medical Center, Harrisburg, PA; University of California Los Angeles School of Medicine, Los Angeles, CA; Mayo Clinic, Rochester, MN; Laboratory of Molecular Neuro-Oncology, Department of Neurology, and Neuroscience Center Zurich, University Hospital and University of Zurich, Zurich, Switzerland; Erasmus MC Cancer Institute, Rotterdam, Netherlands; University of California, San Francisco, San Francisco, CA; Penn State Milton S. Hershey Medical Center, Hinsdale, MA
Background: Low-grade gliomas (LGG) account for 17-22% of all primary brain tumors. Optimal surgical management consists of optimum safe resection with the goal of complete resection. We performed a systematic review and meta-analysis to quantify the association of extent of resection with likelihood of survival, expressing our results in numbers needed to treat (NNT). Methods: A systematic review and study-level meta-analysis to determine the association of resection with overall survival and progression-free survival in newly diagnosed, supratentorial LGG in adults was performed by querying PubMed. Data were extracted to compare gross total resection (GTR) to subtotal resection (STR) and STR to biopsy (Bx) to determine relative risks (RR) of death and progression at 2, 5, and 10 years. Data were analyzed using a random effects model. NNT were calculated from significant comparisons and rounded up to the nearest whole number. Quality of evidence was determined by American Academy of Neurology criteria. Results: The systematic review resulted in 283 potential studies. Ultimately 29 studies were included in at least one comparison. There were no high quality (class I and II) or prospective studies discovered in the review. Comparing GTR to STR, RR with 95% confidence intervals (CI) of death at 2, 5, and 10 years, and NNT to avoid one death at 2, 5, and 10 years (GTR vs. STR) were 0.29 [0.17-0.52, p < 0.0001, NNT 17], 0.39 [0.29-0.51, p < 0.00001, NNT 6], and 0.50 [0.35-0.70, p < 0.0001 NNT 4]. RR and NNT for progression (GTR vs. STR) at 2, 5, and 10 years were 0.37 [0.24-0.57, p < 0.0001 NNT 7], 0.50 [0.39-0.64, p < 0.0001 NNT 4], and 0.67 [0.53-0.84, p = 0.0005 NNT 4]. Comparing STR to Bx, RR of death at 2, 5, and 10 years were 0.55 [0.34-0.88, p = 0.01 NNT 10], 0.9 [0.61-1.34], and 0.95 [0.73-1.23]. Conclusions: Increasing resection thresholds appear to be associated with improved overall and progression free survival, but the body of literature consists of low quality studies. Prospective studies are required to explore whether extent of resection matters or whether resectable tumors share a favorable biology associated with better outcome.
Categories: Brain Tumor