Predictors of Hematologic Toxicity and Chemotherapy Dose Intensity in Patients Undergoing Chemoradiation for Pancreatic Cancer.

Objectives: Intensity-modulated radiation therapy (IMRT) has been shown to decrease abdominal toxicity in patients undergoing chemoradiation (CRT) for pancreatic cancer. We evaluated whether IMRT impacts the rates of hematologic toxicity and chemotherapy dose intensity in patients undergoing CRT. Methods: We retrospectively reviewed patients with borderline resectable or locally advanced pancreatic cancer undergoing CRT between 2006 and 2012. Exclusion criteria included receipt of non-gemcitabine therapy, chemotherapy before CRT, or abnormal baseline hematologic indices. Endpoints included total gemcitabine dose received, dose intensity, unplanned dose reductions, and hematologic toxicity (WBC, ANC, platelet, and hemoglobin). Patient/treatment factors were evaluated for their relationship to the above endpoints during CRT and within the first 3 months post-CRT. Statistical analysis was performed using the Fisher exact test and regression models. Because of the multiple comparisons in the presented analysis, a false discovery rate adjustment was performed at the 5% false discovery rate level. Results: Eighty-five patients met the inclusion criteria. Fifty-eight (68.2%) patients received treatment with IMRT, and 27 (31.8%) patients were treated with 3D-conformal radiation. During CRT, there was no relationship between radiation technique and gemcitabine dose received, dose intensity, or hematologic grade 3 + toxicity. Post-CRT, there was no relationship between radiation technique and total gemcitabine dose received, dose intensity, or dose reduction. Patients receiving IMRT were more likely to have ANC grade 3 + toxicity (P = 0.007) post-CRT, although this was no longer statistically significant after correction. There were no other relationships between treatment technique and hematologic toxicity. Conclusions: IMRT technique may be associated with higher hematologic toxicity in patients undergoing CRT for pancreatic cancer. Given the expanding use of CRT, additional study is needed to identify the impact of IMRT on myelosuppression in these patients.