3583 - Immune-Sparing Treatment Planning in Locally Advanced NSCLC using Transit Blood Dose Modelling

Purpose/Objective(s): Radiation-induced lymphopenia (RIL) is associated with inferior survival in locally advanced non–small cell lung cancer (LA-NSCLC). Conventional treatment planning does not account for effective dose to circulating immune cells (EDIC) nor the maximum single-cell transit dose exposure (Dmax), a proposed mechanistic driver of lymphocyte apoptosis. We performed a feasibility and mechanistic proof-of-concept study to determine whether transit blood dose modeling could identify immune-sparing planning strategies while maintaining target coverage and standard organ-at-risk constraints.

Materials/Methods: Eight photon plans (one clinical reference and seven re-optimizations) were generated for a patient treated to 60 Gy in 30 fractions. Re-optimization strategies prioritized cardiac substructure sparing, reduction of integral dose, and alternative beam geometries.

EDIC was calculated using mean heart (MHD), lung (MLD), and body (MBD) doses. ICE3(Immune Cell Circulatory Exposure Engine) V2.0, a Monte Carlo circulation model (2C3H) simulating 20,000 lymphocyte trajectories, was used to compute maximum single-cell transit dose exposure (Dmax).

Beam-on time (BOT) was extracted from RTPLAN control point data to account for gantry speed and dose-rate variability. Plans were evaluated against clinical constraints including PTV D95 =57 Gy, Lung V20 =35%, MHD =40 Gy, and esophageal limits.

Results: Transit dose modeling revealed substantial variation in predicted immune exposure despite comparable target coverage.

All re-optimized plans reduced EDIC relative to the clinical reference (-0.8% to -11.2%). Dmax decreased from 0.643 Gy (reference) to 0.590 Gy in cardiac-sparing configurations.

Beam-on time analysis demonstrated that gantry speed limits (maximum 6°/s) negated nominal dose rate differences between modalities, indicating that immune sparing was primarily driven by spatial dose redistribution

A cardiac substructure–sparing VMAT plan ranked favorably by both EDIC and Dmax while maintaining all clinical constraints. Lung V20 =35% was the binding constraint across re-optimized plans, reflecting a trade-off between pulmonary and cardiac immune exposure.

All plans exceeded a previously proposed high-risk Dmax threshold, indicating that further refinement may be necessary.

Conclusion: Transit-aware blood dose modeling reveals clinically meaningful differences in circulating immune exposure not captured by conventional planning metrics. Incorporation of mechanistic transit dose modeling into plan evaluation enables identification of immune-sparing configurations without loss of target coverage.

These findings support development of biologically informed treatment planning frameworks and provide a foundation for future studies correlating transit-based immune dose metrics with lymphocyte nadir and clinical outcomes.