Quantifying the Dosimetric Impact of Proton Range Uncertainties on RBE-Weighted Dose Distributions in Intensity-Modulated Proton Therapy for Bilateral Head and Neck Cancer.

BACKGROUND: In current clinical practice, intensity-modulated proton therapy (IMPT) head and neck cancer (HNC) plans are generated using a constant relative biological effectiveness (cRBE) of 1.1. The primary goal of this study was to explore the dosimetric impact of proton range uncertainties on RBE-weighted dose (RWD) distributions using a variable RBE (vRBE) model in the context of bilateral HNC IMPT plans. METHODS: The current study included the computed tomography (CT) datasets of ten bilateral HNC patients who had undergone photon therapy. Each patient's plan was generated using three IMPT beams to deliver doses to the CTV_High and CTV_Low for doses of 70 Gy(RBE) and 54 Gy(RBE), respectively, in 35 fractions through a simultaneous integrated boost (SIB) technique. Each nominal plan calculated with a cRBE of 1.1 was subjected to the range uncertainties of ±3%. The McNamara vRBE model was used for RWD calculations. For each patient, the differences in dosimetric metrices between the RWD and nominal dose distributions were compared. RESULTS: The constrictor muscles, oral cavity, parotids, larynx, thyroid, and esophagus showed average differences in mean dose (Dmean) values up to 6.91 Gy(RBE), indicating the impact of proton range uncertainties on RWD distributions. Similarly, the brachial plexus, brain, brainstem, spinal cord, and mandible showed varying degrees of the average differences in maximum dose (Dmax) values (2.78-10.75 Gy(RBE)). The Dmean and Dmax to the CTV from RWD distributions were within ±2% of the dosimetric results in nominal plans. CONCLUSION: The consistent trend of higher mean and maximum doses to the OARs with the McNamara vRBE model compared to cRBE model highlighted the need for consideration of proton range uncertainties while evaluating OAR doses in bilateral HNC IMPT plans.

shinyDeepDR: A user-friendly R Shiny app for predicting anti-cancer drug response using deep learning.

Advancing precision oncology requires accurate prediction of treatment response and accessible prediction models. To this end, we present shinyDeepDR, a user-friendly implementation of our innovative deep learning model, DeepDR, for predicting anti-cancer drug sensitivity. The web tool makes DeepDR more accessible to researchers without extensive programming experience. Using shinyDeepDR, users can upload mutation and/or gene expression data from a cancer sample (cell line or tumor) and perform two main functions: "Find Drug," which predicts the sample's response to 265 approved and investigational anti-cancer compounds, and "Find Sample," which searches for cell lines in the Cancer Cell Line Encyclopedia (CCLE) and tumors in The Cancer Genome Atlas (TCGA) with genomics profiles similar to those of the query sample to study potential effective treatments. shinyDeepDR provides an interactive interface to interpret prediction results and to investigate individual compounds. In conclusion, shinyDeepDR is an intuitive and free-to-use web tool for in silico anti-cancer drug screening.