3503 - Neo-Adjuvant Radiotherapy Induced Pelvic Bone Marrow FDG PET Uptake Difference to Predict Chemotherapy Related Hematologic Toxicity in Rectal Cancer Patients

Purpose/Objective(s): Pelvic radiotherapy (RT) decreases hematopoietic activity in pelvic bone marrow (PBM). Some patients receiving post-RT chemotherapy (C) may have increased hematologic toxicity (HT) due to affected PBM. PBM sparing RT may help to decrease HT in these patients. The aim of this study was to predict pelvic RT induced HT during C comparing the pre-RT PET (PET1) and post-RT PET (PET2) PBM FDG uptake in rectal cancer patients treated with neo-adjuvant RT.

Materials/Methods: 24 rectal cancer patients treated with neo-adjuvant C-RT, between June 2016 and March 2022, followed by surgery and Capecitabine-Oxaliplatin C, and had PET1 and PET2 images were included. Median age was 55.3 (18.4-72). Female/male ratio was 10/14. Tumor stage was cT2 in 1, cT3 in 22 and cT4 in 1 patient. Twenty-two patients had cN(+) disease. Median tumor and LN doses were 56 (50.4-56) and 50.4 Gy, respectively. All patients were treated with VMAT and received concurrent Capecitabine. Except one, all patients received 4-8 cycles adjuvant C. PBM (whole bone) was contoured on planning CT images as Lumbo-Sacral (LS), Upper Pelvis (UP), Lower Pelvis (LP) and Total Pelvis (P), and copied to the registered PET1 and PET2 images for each patient. SUV_mean was measured for each structure on both PET images. Volumes of each PBM structure receiving 10 to 50 Gy in 5 Gy increment were recorded. Pre-RT, post-RT, pre-C and Nadir (during C) Hgb, Neu, PLT, and WBC values were obtained. Multiple linear regression models were used to analyze the difference between PET1 and PET2 SUVs and change in Post-RT, Pre-C and Nadir Hgb, Neu, PLT, and WBC, compared to pre-RT values. Also, effect of PBM doses on Nadir values was analyzed and PBM doses were compared with the difference between PET1 and PET2 SUV. The model’s results were evaluated based on p-values and R² (coefficient of determination).

Results: Median P PET1 and PET2 SUVs were 0.9 (0.6-2.9) and 0.7 (0.3-1), respectively. There was no significant correlation between PBM SUV differences and V10-50 Gy volumes. Median Pre-RT/Nadir Hgb, Neu, PLT and WBC were 12.6/11.5, 5/2.3, 268/125 and 8/3.7, respectively. RTOG Grade 2 and 3 HT were seen in 7 (29%) patients during C. Nadir Neu (p<0.05, R²=0.42) and PLT (p=0.01, R²=0.46) changes were significantly correlated with P V50 volume only. SUV difference in LP (p=0.02) and P (p=0.03) bone marrow was significantly correlated with nadir WBC change. The correlation between PBM SUV1/SUV2 difference and pre-RT/nadir Hgb, Neu, PLT, and WBC changes (p values and R²) are seen in the Table.

Conclusion: RT related PBM metabolic activity difference may better predict WBC toxicity during adjuvant C compared to PBM doses. Although there was no significant correlation with other hematologic parameters, PBM sparing RT may decrease WBC toxicity during adjuvant C in rectal cancer patients treated with neo-adjuvant RT.