Résumé : Background and Purpose: Evidence regarding radiation-induced lymphopenia and its negative impact on oncological outcome is incrementing. Therefore, the aim of this study is to evaluate the feasibility of lymphocyte-rich organs at risk (LOAR) sparing in pelvic irradiation for localized prostate cancer and to estimate its impact on the effective dose to circulating immune cells (EDIC). Materials and Methods: Twenty patients with pelvic nodal and prostate or prostate bed irradiation were included. The following bone marrow (BM) structures were delineated as LOARs using semi-automatic segmentation: lumbosacral spine (Ls-BM), ilium (Il-BM), lower pelvis (Lp-BM), and the combined whole-pelvis (Wp-BM). Twenty new lymphocyte sparing treatment plans (LS plans) were calculated, optimizing doses to LOARs while maintaining strict coverage of the targets and respecting standard OARs dose constraints. Finally, we elaborated an EDIC calculation model for pelvic irradiation. Results: LS plans showed a statistically significant dose decrease for LOAR compared to standard of care plans without compromising target coverage nor classic OAR dose constraints: in prostate plans, the V40Gy for Ls-BM, Il-BM, and Lp-BM was decreased by 23 %, 36 %, 52 % respectively. For prostate bed plans, the V40Gy for Ls-BM, Il-BM, and Lp-BM was decreased by 25 %, 59 %, 56 %, respectively. For Wp-BM, the V10Gy, V20Gy, and Dmean have been decreased by 3 %, 14 %, 15 %, and by 5 %, 15 %, 17 %, respectively for prostate and prostate bed plans. A statistically significant decrease in EDIC was seen for LS plans in both groups. Conclusions: We successfully demonstrated the feasability of lympocyte-sparing treatment planning in pelvic irradiation, also proposing a model for EDIC calculation.