Favorable Initial Outcomes of Abdominopelvic Reirradiation Using Dose-Escalated Magnetic Resonance Image-Guided Radiation Therapy

Abstract

Abdominopelvic re-irradiation (reRT) is associated with increased risk of potentially severe or fatal toxicity. To minimize the risk of such toxicity, prescribed reRT dose is routinely limited to 35-40 Gy in 2 Gy equivalent dose (EQD2), although higher doses would be preferred to improve local control (LC), if they could be delivered safely. Magnetic resonance image (MRI)-guided radiation therapy (MRgRT) may facilitate dose-escalated reRT through the use of reduced uncertainty margins, continuous intrafraction motion assessment, and daily on-table adaptive replanning. We retrospectively evaluated clinical outcomes of cancer patients from a single institution who received MRgRT on an MR Linac to a previously irradiated site. All patients were treated with continuous intrafraction monitoring. Tumors with significant respiratory motion were usually treated in mid-inspiration breath hold without an internal target volume (ITV). Toxicity was evaluated according to Common Terminology Criteria for Adverse Events, version 5.0. 12 consecutive patients who received reRT were evaluated. Median age was 65 years (range: 33-88) and ECOG performance status was 0-1 in nearly all (91.7%). No patient had >1 prior course of RT. ReRT was initiated after a median of 29.1 months (range: 7.6-153.9) from prior RT completion. Concurrent systemic therapy was delivered to 25%. Conventionally fractionated RT was used prior to reRT in 83.3% (median EQD2: 54.0 Gy) while stereotactic body radiation therapy (SBRT) was used in 16.7% (median EQD2: 61.7 Gy). The most commonly reirradiated sites were abdominopelvic nodes (33.3%) and pancreatic cancer (33.3%). ReRT was delivered with SBRT in 66.7% (median EQD2: 51.3 Gy) and hyperfractionation in 33.3% (median EQD2: 39.2 Gy). On-table adaptive replanning was performed for 3 patients, each receiving reRT in 5 fractions to a median 40 Gy (range: 33-50 Gy). With median follow-up of 7.4 months (range: 1.4-12.8 months) in-field LC was 83.3%; median EQD2 among the 2 patients with local progression was 31.8 Gy. The incidence of acute grade 2 toxicity was 8.3% (fatigue; n = 1). No patient had acute grade 3+ or late grade 2+ toxicity. These initial safety and efficacy outcomes of reRT delivered on an MR Linac, typically with the intent of dose escalation, are encouraging. Based on these data, a prospective study of dose-escalated MR-guided reRT is currently under development at our institution. The range of doses used will allow us to generate a tumor dose-response curve for reRT.