Retrospective Analysis of Late Toxicity in Cervical Cancer Patients Treated With CT-Based High-Dose-Rate Brachytherapy

Abstract

Purpose: The purpose of this study was to (1) compare volumetric (D2cc) and ICRU point doses delivered to the bladder and rectum with CT based intracavitary brachytherapy (BT) and (2) evaluate late toxicity relative to total linear quadratic equivalent dose in 2Gy per fraction (LQED2) to bladder and rectum using volumetric and point based methods of dose reporting. Materials and Methods: This retrospective study evaluated 22 women treated for cervical cancer at our institution with external beam radiation (EBRT) and CT based high-dose-rate (HDR) intracavitary BT. All patients were followed for a minimum of two years, with the exception of those who developed toxicity but died within two years of treatment. Patient, treatment and toxicity data were extracted from paper and electronic charts. During the study period, CT images were acquired at the time of each BT insertion and BT plans were evaluated and approved using ICRU point dose tolerances. To allow evaluation of D2cc doses, organs at risk (bladder and rectum) were retrospectively contoured on the original CT images by one radiation oncologist (RO) and reviewed by a second RO. The position of the ICRU bladder and rectal points were reviewed by a single medical physicist. Dose to the bladder and rectum was evaluated for each fraction of BT using volumetric (D2cc) and point dose (ICRU) methods of dose reporting. Total LQED2 (EBRT + BT) to point A, bladder and rectum was calculated using an α/β of 3 for normal tissue and 10 for tumor. Late toxicity was graded as per the common terminology criteria for adverse events version 3.0 (CTCAE). Correlative analysis was performed to determine association between late toxicity and total LQED2 to organs at risk. Results: Twenty-two women received a total of 88 HDR intracavitary insertions. The median age was 50 years and the median followup was 48.8 months. All patients were treated with 45Gy in 25 fractions to the pelvis; nine patients (41%) received an external beam boost to involved nodes and/or parametria (range:5.4Gy/3 – 9Gy/5). BT was delivered with tandem and ovoid applicators and the most common fractionation for BT was 26Gy in 4 fractions (82%). The mean dose per fraction to point A was 6.39Gy and the mean total LQED2 to point A was 79.2Gy. The mean dose per fraction to the ICRU rectal point and D2cc rectum was 4.1Gy and 3.3Gy; corresponding values for the bladder were 4.6Gy and 5.8Gy. The mean total LQED2 to the ICRU rectal point and D2cc rectum were 66.7Gy and 60.4Gy respectively; corresponding values for the bladder were 72.5Gy and 85.0Gy. Ten women (45%) experienced a total of fourteen grade 1/2 late gastrointestinal (GI) toxicities. Eleven women (50%) experienced a total of sixteen grade1/2, and three grade 3/4 late genitourinary (GU) toxicities. The three grade 3/4 toxicities (dysuria, frequency and vesicovaginal fistula) occurred in a single patient (ICRU bladder 103.0Gy, D2cc bladder 88.3Gy). There was no correlation between toxicity and total LQED2 to D2cc bladder or rectum, nor was there correlation between toxicity and total LQED2 to ICRU bladder or rectal points (all r2 <0.07). Conclusions: Similar to previously published studies, the ratio of mean D2cc/ICRU dose to bladder and rectum in our series was 1.3 and 0.82, respectively. We found no correlation between late toxicity and total LQED2 to bladder and rectum using either point dose or volumetric evaluation of organs at risk. Large prospective studies are needed to evaluate late toxicity relative to total LQED2 to bladder and rectum in the definitive treatment of cervical cancer.