Mechanisms underlying an enhanced navigation-based intervention to improve timely adjuvant radiotherapy in head and neck cancer: Secondary outcomes from the NDURE RCT.

95Background: Initiation of postoperative radiation therapy (PORT) ≤ 6 weeks of surgery is part of NCCN Guidelines for head and neck squamous cell carcinoma (HNSCC) and a Commission on Cancer quality metric, yet more than 50% of patients experience a delay. Our randomized clinical trial (RCT) demonstrated that NDURE, an enhanced navigation-based intervention, improved timely PORT for patients with HNSCC relative to usual care (UC) patient navigation (PN). This planned secondary analysis tests the hypothesis that NDURE improves initiation of PORT ≤ 6 weeks of surgery relative to UC by improving the delivery of underlying processes of care related to timely PORT. Methods: Adults with HNSCC planning to undergo surgery and PORT were randomly assigned 1:1 to NDURE, a multilevel navigation-based intervention to enhance key processes of care and overcome barriers to timely PORT, or UC, which consisted of standard PN. The primary endpoint was initiation of timely PORT (≤ 6 weeks postoperatively). Hypothesized process of care mediators included: 1) preoperative (pre-op) consultation with a radiation oncologist, 2) pre-op dental evaluation and extractions during the ablative surgery, 3) timely postoperative evaluation by the radiation oncologist ( < 21 days of surgery), 4) timely CT simulation ( < 7 days after the radiation oncology appointment), and 5) timely radiation planning (starting PORT < 14 days after CT simulation). Causal mediation analysis was performed to determine the effects of hypothesized mediating process of care variables on initiation of timely PORT. All binary endpoints were modeled using a generalized linear model approach in R Studio version 4.4.1 with mediation package 4.5.0. Results: Among 145 (NDURE, n = 67; UC, n = 78) patients evaluable for the primary endpoint, NDURE increased the probability of attending a pre-op consultation with a radiation oncologist (79.1% vs 43.6%; p < 0.001), pre-op dental evaluation and extractions during the ablative surgery (80.6% vs 55.1%; p < 0.001), timely postoperative evaluation by the radiation oncologist (56.7% vs 20.5%; p < 0.001), and timely radiation planning (65.7% vs 47.4%; p = 0.02) relative to UC. In a causal mediation analysis, the effect of NDURE on initiation of timely PORT relative to UC was most strongly mediated through improving timely postoperative evaluation by the radiation oncologist (proportion of total effect mediated = 0.24; 95% CI 0.08 to 0.65) and timely radiation planning (proportion of total effect mediated = 0.14; 95% CI -0.03 to 0.45). Conclusions: In this RCT, NDURE improved numerous processes of care relative to UC. The primary mechanisms by which NDURE improved timely PORT were enhancing timely postoperative evaluation by a radiation oncologist and timely radiation planning after simulation. These data may help optimize strategies to improve timely PORT. Clinical trial information: NCT04030130.

Duke Scholars

Author Russel Roy Kahmke Head & Neck Surgical Oncology