Comparing Healthcare Utilization, Hospitalization, and Costs with Neoadjuvant Intensity-Modulated Radiation Therapy (IMRT) versus Proton Beam Therapy (PBT) for Esophageal Cancer

Abstract

In the setting of tri-modality therapy for esophageal cancer (EC), prior studies demonstrate that neoadjuvant proton therapy (PBT) compared with intensity-modulated radiation therapy (IMRT) yields lower toxicity burden and fewer postoperative complications. However, it is not known whether there are also benefits of neoadjuvant PBT for healthcare utilization outcomes in EC patients undergoing surgery. We hypothesized that patients treated neoadjuvantly with PBT vs. IMRT plus esophagectomy would demonstrate healthcare utilization outcomes benefits, specifically, lower postoperative costs and shorter hospitalization length of stay (LOS). We tested differences in healthcare utilization outcomes in 287 patients treated with neoadjuvant PBT vs. IMRT, first in a retrospective test cohort (N = 237), and then in a prospective validation cohort of patients enrolled in a randomized trial of PBT vs. IMRT [NCT 01512589] (N = 50). These healthcare utilization outcomes were a pre-specified secondary outcome in this trial. In patients receiving neoadjuvant PBT (N = 81 test, 21 validation) vs. IMRT (N = 156 test, 29 validation) chemoradiation followed by esophagectomy, we compared costs (healthcare charges), length of stay (LOS) for esophagectomy, and postoperative complications (POC). Absolute (Δ = $PBT-$IMRT) and relative costs as in cost ratios (CR = $PBT/$IMRT) were calculated, compared using generalized linear models with gamma distribution. LOS was compared using negative binomial regression. There were no significant differences in patient characteristics by treatment group in both test and validation cohorts, including age, gender, race, histology, stage, location in the esophagus, radiation dose, and comorbidities. Neoadjuvant chemoradiation costs were higher for PBT vs. IMRT in both the test and validation cohorts: Test CR = 1.80 (95% CI 1.72-1.90, P<0.001); Δ = +$57,319; Validation CR = 1.50 (95% CI 1.40-1.61, P<0.001); Δ = +$55,958. As expected, surgery costs for esophagectomy did not significantly differ in PBT vs. IMRT in either cohort. Mean LOS during hospitalization for esophagectomy were lower for PBT vs. IMRT in both cohorts (Test LOS 8.4 vs. 10.4 days; P<0.001; Validation LOS 9.1 vs. 13 days; P = 0.02). Shorter LOS (P<0.001) and fewer POCs (P<0.001) were associated with lower postoperative costs in test cohort. Hospitalization costs for esophagectomy were lower for PBT vs. IMRT in both the test and validation cohorts: Test CR = 0.77 (95% CI 0.65-0.92, P = 0.004); Δ = - $15,251; Validation CR = 0.47 (95% CI 0.30-0.74; P = 0.002); Δ = - $46,923. After adjusting for covariates, costs remained higher for PBT in the chemoradiation period but lower in the postoperative period. Despite up-front chemoradiation costs, substantial healthcare costs were recovered after PBT during the period at highest risk for POCs, during hospitalization after esophagectomy.