AUTHOR=Lam Sai-Kit , Zhang Yuanpeng , Zhang Jiang , Li Bing , Sun Jia-Chen , Liu Carol Yee-Tung , Chou Pak-Hei , Teng Xinzhi , Ma Zong-Rui , Ni Rui-Yan , Zhou Ta , Peng Tao , Xiao Hao-Nan , Li Tian , Ren Ge , Cheung Andy Lai-Yin , Lee Francis Kar-Ho , Yip Celia Wai-Yi , Au Kwok-Hung , Lee Victor Ho-Fun , Chang Amy Tien-Yee , Chan Lawrence Wing-Chi , Cai Jing TITLE=Multi-Organ Omics-Based Prediction for Adaptive Radiation Therapy Eligibility in Nasopharyngeal Carcinoma Patients Undergoing Concurrent Chemoradiotherapy JOURNAL=Frontiers in Oncology VOLUME=11 YEAR=2022 URL=https://www.frontiersin.org/journals/oncology/articles/10.3389/fonc.2021.792024 DOI=10.3389/fonc.2021.792024 ISSN=2234-943X ABSTRACT=Purpose

To investigate the role of different multi-organ omics-based prediction models for pre-treatment prediction of Adaptive Radiotherapy (ART) eligibility in patients with nasopharyngeal carcinoma (NPC).

Methods and Materials

Pre-treatment contrast-enhanced computed tomographic and magnetic resonance images, radiotherapy dose and contour data of 135 NPC patients treated at Hong Kong Queen Elizabeth Hospital were retrospectively analyzed for extraction of multi-omics features, namely Radiomics (R), Morphology (M), Dosiomics (D), and Contouromics (C), from a total of eight organ structures. During model development, patient cohort was divided into a training set and a hold-out test set in a ratio of 7 to 3 via 20 iterations. Four single-omics models (R, M, D, C) and four multi-omics models (RD, RC, RM, RMDC) were developed on the training data using Ridge and Multi-Kernel Learning (MKL) algorithm, respectively, under 10-fold cross validation, and evaluated on hold-out test data using average area under the receiver-operator-characteristics curve (AUC). The best-performing single-omics model was first determined by comparing the AUC distribution across the 20 iterations among the four single-omics models using two-sided student t-test, which was then retrained using MKL algorithm for a fair comparison with the four multi-omics models.

Results

The R model significantly outperformed all other three single-omics models (all p-value<0.0001), achieving an average AUC of 0.942 (95%CI: 0.938-0.946) and 0.918 (95%CI: 0.903-0.933) in training and hold-out test set, respectively. When trained with MKL, the R model (R_MKL) yielded an increased AUC of 0.984 (95%CI: 0.981-0.988) and 0.927 (95%CI: 0.905-0.948) in training and hold-out test set respectively, while demonstrating no significant difference as compared to all studied multi-omics models in the hold-out test sets. Intriguingly, Radiomic features accounted for the majority of the final selected features, ranging from 64% to 94%, in all the studied multi-omics models.

Conclusions

Among all the studied models, the Radiomic model was found to play a dominant role for ART eligibility in NPC patients, and Radiomic features accounted for the largest proportion of features in all the multi-omics models.