Predicting radiotherapy-induced xerostomia in head and neck cancer patients using day-to-day kinetics of radiomics features

Thomas Berger; David J Noble; Leila E A Shelley; Thomas McMullan; Amy Bates; Simon Thomas; Linda J Carruthers; George Beckett; Aileen Duffton; Claire Paterson; Raj Jena; Duncan B McLaren; Neil G Burnet; William H Nailon

doi:10.1016/j.phro.2022.10.004

Predicting radiotherapy-induced xerostomia in head and neck cancer patients using day-to-day kinetics of radiomics features

Phys Imaging Radiat Oncol. 2022 Nov 4:24:95-101. doi: 10.1016/j.phro.2022.10.004. eCollection 2022 Oct.

Authors

Thomas Berger¹, David J Noble^{2

3}, Leila E A Shelley¹, Thomas McMullan¹, Amy Bates², Simon Thomas⁴, Linda J Carruthers¹, George Beckett⁵, Aileen Duffton⁶, Claire Paterson⁶, Raj Jena², Duncan B McLaren³, Neil G Burnet⁷, William H Nailon^{1

8}

Affiliations

¹ Department of Oncology Physics, Edinburgh Cancer Centre, Western General Hospital, Crewe Road South, Edinburgh EH4 2XU, UK.
² The University of Cambridge, Department of Oncology, Cambridge Biomedical Campus, Hills Road, Cambridge CB2 0QQ, UK.
³ Department of Clinical Oncology, Edinburgh Cancer Centre, Western General Hospital, Crewe Road South, Edinburgh EH4 2XU, UK.
⁴ Department of Medical Physics and Clinical Engineering, Cambridge University Hospitals NHS Foundation Trust, Cambridge Biomedical Campus, Hills Road, Cambridge CB2 0QQ, UK.
⁵ Edinburgh Parallel Computing Centre, Bayes Centre, 47 Potterrow, Edinburgh EH8 9BT, UK.
⁶ Beatson West of Scotland Cancer Centre, Great Western Road, Glasgow G12 0YN, UK.
⁷ The Christie NHS Foundation Trust, Wilmslow Road, Manchester, M20 4BX, UK.
⁸ School of Engineering, the University of Edinburgh, the King's Buildings, Mayfield Road, Edinburgh EH9 3JL, UK.

Abstract

Background and purpose: The images acquired during radiotherapy for image-guidance purposes could be used to monitor patient-specific response to irradiation and improve treatment personalisation. We investigated whether the kinetics of radiomics features from daily mega-voltage CT image-guidance scans (MVCT) improve prediction of moderate-to-severe xerostomia compared to dose/volume parameters in radiotherapy of head-and-neck cancer (HNC).

Materials and methods: All included HNC patients (N = 117) received 30 or more fractions of radiotherapy with daily MVCTs. Radiomics features were calculated on the contra-lateral parotid glands of daily MVCTs. Their variations over time after each complete week of treatment were used to predict moderate-to-severe xerostomia (CTCAEv4.03 grade ≥ 2) at 6, 12 and 24 months post-radiotherapy. After dimensionality reduction, backward/forward selection was used to generate combinations of predictors.Three types of logistic regression model were generated for each follow-up time: 1) a pre-treatment reference model using dose/volume parameters, 2) a combination of dose/volume and radiomics-based predictors, and 3) radiomics-based predictors. The models were internally validated by cross-validation and bootstrapping and their performance evaluated using Area Under the Curve (AUC) on separate training and testing sets.

Results: Moderate-to-severe xerostomia was reported by 46 %, 33 % and 26 % of the patients at 6, 12 and 24 months respectively. The selected models using radiomics-based features extracted at or before mid-treatment outperformed the dose-based models with an AUC_train/AUC_test of 0.70/0.69, 0.76/0.74, 0.86/0.86 at 6, 12 and 24 months, respectively.

Conclusion: Our results suggest that radiomics features calculated on MVCTs from the first half of the radiotherapy course improve prediction of moderate-to-severe xerostomia in HNC patients compared to a dose-based pre-treatment model.

Keywords: Head and neck cancer; Mega-voltage CT; Parotid gland; Radiomics; Xerostomia.

Abstract

Grants and funding