Graboyes, E.M., Hill, E.G., Sterba, K.R., Chera, B.S., DeMass, R., Kistner-Griffin, E., McCay, J., Newman, J.G., Albergotti, W., Kejner, A., Skoner, J., Harper, J.L., Kaczmar, J., Zimmerman, S., Warren, G.W., Alberg, A.J., Calhoun, E., Nussenbaum, B., and Halbert, C. Hughes more...
Delays in initiating guideline-adherent postoperative radiation therapy (PORT) affect 50% of patients with head and neck squamous cell carcinoma (HNSCC), disproportionately burden racial minorities, are associated with worse survival, and contribute to racial disparities in mortality. This study tests the hypothesis that NDURE (Navigation for Disparities and Untimely Radiation thErapy), a patient navigation-based intervention, will decrease delays in starting guideline-adherent PORT relative to Usual Care (UC). This randomized clinical trial (RCT) included adults with locally-advanced HNSCC (i.e., oral cavity, oropharynx, hypopharynx, larynx or paranasal sinuses) planning to undergo curative-intent surgery and PORT. Preoperatively, patients were randomized to NDURE, a navigation-based intervention that addresses barriers to timely PORT at the patient-, healthcare team-, and organizational-levels, or UC. The primary endpoint was delay in initiating guideline-adherent PORT, defined as starting PORT > 6 weeks after surgery. The secondary endpoint was time-to-PORT (TTP). The difference in PORT delay between arms was evaluated using a binary regression generalized linear model with randomization stratification factors (race, expected PORT facility) as covariates. The difference in TTP between arms was evaluated by estimating hazard ratios (HRs) from Cox proportional hazards models, adjusting for stratification factors. Interactions between race and treatment arm were subsequently added to the binary regression and Cox models to compare racial disparities between arms. To detect a 20% reduction in the primary endpoint of PORT delay (45% vs 25%) assuming a two-sided α = 0.1 and power of 83%, we planned to accrue n=75 patients/arm evaluable for the primary endpoint, up to a total of n=180. Among 177 eligible patients randomized to NDURE (n=88) or UC (n=89), 146 patients underwent surgery and had a pathologic indication for PORT. NDURE decreased delays in initiating guideline-adherent PORT relative to UC (model-based PORT delay, 26% vs 61%; risk difference = -35%; 90% CI -48% to -23%; p < 0.001). Median TTP was 39 and 47 days in NDURE and UC, respectively. NDURE improved TTP relative to UC (HR = 1.92; 90% CI 1.43 to 2.58; p < 0.001). The difference in delays in initiating guideline-adherent PORT between Black and White patients was 12% in NDURE vs 24% in UC (p = 0.51). The difference in median TTP between Black and White patients was 1 day in NDURE vs 10 days in UC (NDURE Black/White HR = 0.89; 90% CI 0.48 - 1.53; UC Black/White HR = 0.65; 90% CI = 0.41 - 1.02; p = 0.53). In this RCT of patients with HNSCC undergoing surgery and PORT, NDURE decreased delays in starting guideline-adherent PORT and improved TTP. These data support conducting a large efficacy trial to evaluate patient navigation-based approaches to improving the timeliness and equity of PORT for patients with HNSCC and their effect on oncologic outcomes. [ABSTRACT FROM AUTHOR] more...