Abstract / Summary
ABSTRACT Background: Cellular therapies—chimeric antigen receptor T-cell (CAR-T), tumor-infiltrating lymphocyte (TIL), T-cell receptor (TCR)-engineered, gamma-delta (γδ) T-cell, and high-dose interleukin-2—have transformed the management of refractory malignancies. Over 6,000 interventional cell therapy trials are registered globally, yet patient selection criteria remain fragmented across trial protocols, regulatory labels, and institutional pathways, risking inappropriate exclusion and erroneous enrollment. Objective: The primary objective was to develop a unified, evidence-based patient selection algorithm for cellular therapy spanning 19 malignancies and 5 therapy modalities, with explicit decision-making logic for dual-pathway diseases, neurological performance overrides, and infection classification. The secondary objective was to implement the algorithm as a freely accessible, interactive R Shiny decision-support web application and to tabulate published efficacy and safety outcomes for each disease–antigen–therapy combination. Materials and Methods: Eligibility criteria were synthesized from published phase I–III trial protocols, regulatory submissions, and National Comprehensive Cancer Network guideline recommendations. The algorithm incorporates branching logic for dual-pathway diseases, neurological performance status overrides for patients with glioma, differentiation between controlled and uncontrolled infection, and age-appropriate pediatric inclusion. Results: The algorithm covers hematological malignancies, central nervous system tumors, thoracic, gastrointestinal, genitourinary, gynecological, musculoskeletal, and pediatric cancers. Each disease module specifies indications, tumor fitness criteria, standardized organ-function thresholds, antigen requirements, and published survival outcomes. Conclusion: This framework provides a reproducible, transparent tool for cellular therapy eligibility screening, promoting consistent multidisciplinary decision-making. The algorithm has been deployed as a freely accessible web application. Prospective multi-institutional validation is warranted.