Implementation of a Special Medical Physics Consult Process for Re-Irradiation in Patients with Brain Tumors: Planning and Toxicity Outcomes

Abstract

<h3>Purpose/Objective(s)</h3> We hypothesized that implementation of a systematic approach to re-irradiation (ReRT) using a special medical physics consult (SMPC) process would lead to acceptably low rates of high-grade toxicity associated with planning parameters impacting clinical management. <h3>Materials/Methods</h3> Consecutive patients treated using the ReRT-SMPC process during the first 18 months of its implementation were included. Accuracy of data set registration in the region of treatment was assessed by physics and verified by physician to generate cumulative physical and biologically corrected dose maps (equivalent dose in 2 Gy fractions EQD₂, alpha/beta 2.5 for normal tissue). Standardized dose discounts based on time interval from prior treatment and consistent guidelines for acceptable EQD₂ dose-limits per organ-at-risk (OAR) were used in all cases to provide allowable dose constraints prior to planning. Post-planning physics evaluation highlighted exceeded limits to assist clinical management. Comparison of planning parameters among patients with and without high-grade toxicity (grade 3 or higher) per CTCAE v5.0 was performed using 2-sided t-test and significance level 0.05. <h3>Results</h3> From April 2017 to September 2018, 261 consecutive patients underwent ReRT using this workflow, of whom 55 (21%) had brain tumors. ReRT with conventional fractionation (CFRT) was given to 47% primarily for gliomas (median EQD₂ 47 Gy (IQR, 41-56)), and with single or multi-fraction stereotactic radiosurgery in 53% primarily for brain metastases (median EQD₂ 100 Gy (IQR 57-120)). Median follow-up after ReRT among patients alive at time of analysis was 6.9 months (IQR, 5.3-10.8). No high-grade toxicity was observed in any delineated OAR including brainstem or optic structures. No specific normal brain constraint was employed in the SMPC process, and 5 (9%) patients experienced high-grade toxicity both early (cerebral edema, n=3) and late (necrosis, n=2). Cumulative EQD₂ was lower among patients with vs without toxicity (D0.03cc 87 vs 155 Gy, p=0.047), as 80% of patients with toxicity underwent ReRT with CFRT to larger volumes (39 vs 14 cc). While no constraints were applied for normal brain, recovery factors based on interval from index treatment were applied for all patients. However, no association with time interval was observed in patients with vs without toxicity (p=0.4). In two instances, the ReRT-SMPC analysis resulted in physician decision not to treat due to potential toxicity concerns. <h3>Conclusion</h3> Re-irradiation of primary and metastatic brain tumors using a standardized SMPC process yields low rates of high-grade CNS toxicity and meaningfully impacts physician decision-making. Future planning parameters including normal brain dose will be refined using this systematic process to improve the therapeutic ratio among patients undergoing re-irradiation.