Defining Intraprostatic Gross Tumor Volume and Microboosting through Consensus: A Contour-Based Modified Delphi...

Advances in imaging have enabled identification of intraprostatic gross tumor volumes (GTVs), creating opportunities for microboosting in prostate radiation therapy. However, variability exists in patient selection, target delineation, and treatment planning. This study aimed to characterize expert contouring variability and establish consensus guidance for GTV microboosting using a contour-based modified Delphi consensus.

International radiation oncologists with expertise in GTV microboosting participated in a 2-part study comprising a contouring exercise and a 2-round modified Delphi consensus. Participants contoured GTVs in 2 prostate cancer cases with multiparametric magnetic resonance imaging (MRI) and prostate-specific membrane antigen positron emission tomography imaging, including 1 complex case with imaging discordance. Contouring agreement was assessed using Fleiss κ, intraclass correlation coefficients, and simultaneous truth and performance level estimation analysis. Participants subsequently completed iterative Delphi surveys addressing patient selection, imaging requirements, contouring practices, and treatment planning. Consensus and strong consensus were predefined as ≥75% and ≥90% agreement, respectively.

Fifteen of 20 invited experts completed the study. Contouring agreement was high in the simpler case but substantially lower in the complex case. Prostate-specific membrane antigen positron emission tomography-based GTVs were consistently larger than multiparametric MRI-based GTVs. Overall, consensus or strong consensus was achieved for 43 of 131 statements (33%). Agreement was reached on key aspects of GTV delineation, including the use of combined T2-weighted and diffusion weighted images/Apparent Diffusion Coefficient sequences on multiparametric MRI, eligibility of Prostate Imaging Reporting and Data System 4 to 5 and PROMISE (Prostate Cancer Molecular Imaging Standardize Evaluation) 4 to 5 lesions for boosting, preferred use of prostate-specific membrane antigen-targeted tracers, and avoidance of clinical target volume margins when positron emission tomography and MRI information are used in combination. No consensus was achieved on stereotactic body radiation therapy microboosting outside clinical trials, margin expansion using single-modality imaging, or optimal dose prescriptions to the GTV or uninvolved prostate.

This contour-based modified Delphi study demonstrates variability in GTV delineation for complex cases and identifies areas of expert consensus that can inform standardized implementation of prostate GTV microboosting.