Hyperfractionated abdominal reirradiation for gastrointestinal malignancies

Background: We sought to determine the role of abdominal reirradiation for patients presenting with recurrent or new primary gastrointestinal (GI) malignancies. At our institution, we have established a hyperfractionated, accelerated reirradiation regimen consisting of 39 Gray (Gy) in 26 twice-daily fractions. Although this regimen is used frequently in the pelvis, we sought to determine its toxicity and efficacy for abdominal tumors. Methods: Twenty-four patients who received abdominal reirradiation with a hyperfractionated, accelerated approach from 2000 to 2017 were identified. Overall survival (OS) and local progression-free survival (LPFS) were calculated using the Kaplan-Meier method. Several patient, tumor and treatment characteristics were evaluated on univariate analyses for association with OS and LPFS using a Cox proportional hazards model. Results: Of the twenty-four patients identified, the majority (n=11, 46%) had pancreatic adenocarcinoma as their primary disease but also included upper GI adenocarcinoma (n=4), colon adenocarcinoma (n=3), hepatobiliary cancers (n=4) and other malignancies (n=2). The majority of patients received 45-50.4Gy in 1.8Gy fractions as their initial abdominal radiation course. The median reirradiation dose was 39Gy in 26 twice-daily fractions with a minimum six hour interval. The median [interquartile range (IQR)] interval between the courses of radiotherapy was 28 [18.6-38.9] months. Only palliative reirradiation intent was associated with decreased OS. While colon adenocarcinoma primary was significantly associated with increased LPFS, the sample size was small (n=3). The 1-yr rate of LPFS was 38%. The median [IQR] duration of freedom from local progression was 8 [3.8-19.2] months. The 1-year OS was 50% and the median (IQR) OS was 14 [6.3-19.6] months. Thirteen patients (54%) had acute side effects with one patient experiencing G3 nausea and one experiencing a G4 bleed; the remaining patients experienced G1-G2 symptoms. Conclusion: Hyperfractionated, accelerated reirradiation to the abdomen was relatively well-tolerated but provided limited local control to recurrent or second primary abdominal malignancies. Reirradiation could play a role in treating these patients with palliative or curative intent, but alternative strategies for delivering increased biologically effective dose should be further explored.