Patterns in ano-rectal dose maps and the risk of late toxicity after prostate IMRT

Abstract

Purpose: The aim of this work was to determine how the spatial pattern of dose in the ano-rectal wall is related to late gastro-intestinal toxicity for prostate cancer patients treated with mainly IMRT.

Patients and methods: Patients from the DUE-01 multicentre study with patient-reported (prospective) follow-up and available dosimetric data were included. Conventionally fractionated patients received 74–80 Gy and hypofractionated patients received 65–75.2 Gy. A large majority of the patients were treated with intensity-modulated radiotherapy (IMRT). Dose-surface maps (DSMs) for the anal canal and rectum as a single structure, and for the anal canal and the rectum separately, were co-registered rigidly in two dimensions and, for the patients with and without toxicity, respectively, the mean value of the dose in each pixel was calculated. A pixel-wise t-test was used to highlight the anatomical areas where there was a significant difference between the ‘mean dose maps’ of each group. Univariate models were also fitted to a range of spatial parameters. The endpoints considered were a mean grade ≥1 late fecal incontinence and a maximum grade ≥2 late rectal bleeding.

Results: Twenty-six out of 213 patients had fecal incontinence, while 21/225 patients had rectal bleeding. Incontinence was associated with a higher dose in the caudal region of the anal canal; the most relevant spatial parameter was the lateral extent of the low and medium isodoses (5–49 Gy in EQD2). Bleeding was associated with high isodoses reaching the posterior rectal wall. The spatial dose parameters with the highest AUC value (.69) were the lateral extent of the 60–70 Gy isodoses.

Conclusions: To avoid fecal incontinence it is important to limit the portion of the anal canal irradiated. Our analysis confirms that rectal bleeding is a function of similar spatial dose parameters for patients treated with IMRT, compared to previous studies on patients treated with three-dimensional conformal radiotherapy.

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Funding

This work was supported by Fondazione Italo Monzino and Associazione Italiana per la Ricerca sul Cancro (AIRC, grant IG16087).