Prognostic Relevance of Treatment Failure Patterns in Pediatric High Grade Glioma; Is there a Role for a Revised Failure Classification System?

• Published in: International journal of radiation oncology, biology, physics Vol. 99; no. 2; pp. 450 - 458
• Main Authors: Lucas, John T., MS, MD, Cooper, David A., MD, Hwang, Scott, MD, Tinkle, Christopher, MD, PhD, Li, Xingyu, PhD, Li, Yimei, PhD, Orr, Brent, MD, PhD, Merchant, Thomas E., DO, PhD, Broniscer, Alberto, MD
• Format: Journal Article
• Language: English
• Published: United States Elsevier Inc 01-10-2017
• Subjects: Adolescent; Antineoplastic Agents - therapeutic use; Brain Neoplasms - mortality; Brain Neoplasms - pathology; Brain Neoplasms - surgery; Brain Neoplasms - therapy; Child; Disease Progression; Erlotinib Hydrochloride - therapeutic use; Glioma - mortality; Glioma - pathology; Glioma - surgery; Glioma - therapy; GLIOMAS; GY RANGE; Hematology, Oncology and Palliative Medicine; Humans; Magnetic Resonance Imaging; Neuroimaging - methods; PATIENTS; PEDIATRICS; Prognosis; Prospective Studies; RADIATION DOSES; Radiology; RADIOLOGY AND NUCLEAR MEDICINE; Radiotherapy Dosage; Regression Analysis; Treatment Failure
• ISSN: 0360-3016; 1879-355X
• DOI: 10.1016/j.ijrobp.2017.04.039

Abstract Purpose To investigate the pattern of treatment failure with respect to anatomic extent, radiation dose, and criteria for failure according to the Response Assessment in Neuro Oncology (RANOc). We evaluate the corresponding prognostic significance of these factors in patients with pHGG. Methods and Materials Fifty-six patients with pHGG were enrolled on an institutional Phase I-II prospective trial that included maximal safe resection and radiation with concurrent and adjuvant erlotinib. Radiation was administered to 54-59.4Gy at 1.8Gy per day. Tumor progression was defined according to clinical symptoms and imaging features and classified in relationship to original extent of tumor, radiation prescription target volume coverage and RANOc. Results With a median follow-up of 90 months (range 70-124 months), progression occurred in 48 patients (85.7%) while 8 (14.3%) were without progression. Central failure represented 42.6% of the total cohort, while infield, marginal, and distant failures occurred in 7.4%, 9.3%, and 22.2%, respectively. Patients with biopsy or subtotal resection had increased rates of central failure and represented 14.81% and 16.67% of the total cohort, respectively. Tumor progression was classified as local, local + distant, and distant. Among patients with local failure as a component of failure, 5 were considered marginal. Patients with frontal, temporal, and parietal disease had the highest rates of multifocal failure. A comparison between responses defined by RANOc demonstrated varied TTD from progression. Conclusions pHGG was shown to have high rates of central failure, particularly in those with limited resection. Patients with central failure had a trend towards more prolonged TTD from failure relative to other failure patterns. The low marginal failure rates seen in this group suggest that less conservative radiation target margins may be possible. TTD from failure varied according to RANO types, suggesting that adult RANOc require modification before being applied to pHGG.