Radiomics Analysis of 3D Dose Distributions to Predict Toxicity of Radiotherapy for Cervical Cancer.

François Lucia,Julien Bert,Olivier Pradier,Dominique Gouders,Vincent Bourbonne,Mathieu Hatt,Gurvan Dissaux,Vincent Jaouen,Florent Tixier,Dimitris Visvikis,Dorothy M Gujral,Ulrike Schick,Omar Miranda

doi:10.3390/jpm11050398

Abstract

Standard treatment for locally advanced cervical cancer (LACC) is chemoradiotherapy followed by brachytherapy. Despite radiation therapy advances, the toxicity rate remains significant. In this study, we compared the prediction of toxicity events after radiotherapy for locally advanced cervical cancer (LACC), based on either dose-volume histogram (DVH) parameters or the use of a radiomics approach applied to dose maps at the voxel level. Toxicity scores using the Common Terminology Criteria for Adverse Events (CTCAE v4), spatial dose distributions, and usual clinical predictors for the toxicity of 102 patients treated with chemoradiotherapy followed by brachytherapy for LACC were used in this study. In addition to usual DVH parameters, 91 radiomic features were extracted from rectum, bladder and vaginal 3D dose distributions, after discretization into a fixed bin width of 1 Gy. They were evaluated for predictive modelling of rectal, genitourinary (GU) and vaginal toxicities (grade ≥ 2). Logistic Normal Tissue Complication Probability (NTCP) models were derived using clinical parameters only or combinations of clinical, DVH and radiomics. For rectal acute/late toxicities, the area under the curve (AUC) using clinical parameters was 0.53/0.65, which increased to 0.66/0.63, and 0.76/0.87, with the addition of DVH or radiomics parameters, respectively. For GU acute/late toxicities, the AUC increased from 0.55/0.56 (clinical only) to 0.84/0.90 (+DVH) and 0.83/0.96 (clinical + DVH + radiomics). For vaginal acute/late toxicities, the AUC increased from 0.51/0.57 (clinical only) to 0.58/0.72 (+DVH) and 0.82/0.89 (clinical + DVH + radiomics). The predictive performance of NTCP models based on radiomics features was higher than the commonly used clinical and DVH parameters. Dosimetric radiomics analysis is a promising tool for NTCP modelling in radiotherapy.

Highlights

Cervical cancer is the fourth most frequent cancer in women, with an estimated incidence of 70,000 new cases in 2018 [1]
To explore the possible value of dosimetric radiomics in predictive modeling, we focused on the incidence of GU, rectal and vaginal toxicities, using the Common Terminology
For locally advanced cervical cancer (LACC) we compared the predictive performance of Normal Tissue Complication Probability (NTCP) models based on clinical parameters to commonly used dose-volume histogram (DVH) parameters of external beam radiation therapy (EBRT) and brachytherapy, such as Vx and Dmean, and to radiomics features derived from 3D dose distributions, alone or in combination

Summary

Introduction

Cervical cancer is the fourth most frequent cancer in women, with an estimated incidence of 70,000 new cases in 2018 [1]. Compared with tridimensionnal external beam radiation therapy (EBRT), intensity-modulated RT (IMRT) can deliver an adequate dose to the target structures whilst limiting the dose delivered to pelvic and abdominal organs-at-risk (OARs), including bowel, rectum and bladder. This results in a lower incidence of early and late gastrointestinal and genitourinary (GU) toxicity [2,3]. OARs, resulting in lower toxicity and improved local control [4,5] Despite these advances, the toxicity rate remains significant [6] and these side effects can have an impact on longterm quality of life for these young patients [7,8]

Objectives

Methods

Results

Discussion

Conclusion