Implementation and efficacy of a large-scale radiation oncology case-based peer-review quality program across a multinational cancer network.

1 Background: With rapid community expansion of academic cancer centers, ensuring high-quality delivery of care across all affiliated network sites is critical. Here we report the results of a radiation oncology peer-review system implemented across a large multinational cancer network. Methods: Weekly radiation oncology peer-review conferences were held between network centers and the main campus of a major cancer system; results of standardized peer-review for each case were recorded. Peer-review resulted in each case being scored as concordant or nonconcordant on initial review; nonconcordance was based on institutional guidelines, national standards, and/or expert opinion. Results: Between 2014 and 2018, 28,730 patient radiation treatment plans underwent peer-review at 10 network centers. The peer-review case volume increased over this study period, from 1,420 cases in 2014 to 9,112 in 2018, concomitant with network expansion. Examining cases reviewed in 2018 (N = 9,112), the most-commonly reviewed cases by disease site were breast (28.9%), head and neck (HN; 13.9%), and lung (12.6%). Of all cases in 2018, 452 (5.0%) were deemed nonconcordant. Higher nonconcordance rates were noted for HN cases (14.0%), and lower rates for lung cases (2.3%; p < 0.001). Of nonconcordant HN cases, the majority (69.5%) were deemed nonconcordant based on target volume delineation. Of nonconcordant breast cases, most (67.1%) were nonconcordant based on radiation field design. For centers added to the network during the study period, we observed a significant decrease in the nonconcordance rate over time after joining the network (average annual decrease of 5.4% in nonconcordant cases; p < 0.001). Conclusions: These data demonstrate the feasibility and efficacy of a large-scale multinational cancer network radiation oncology weekly peer-review program. Nonconcordance rates were highest for HN cases, primarily due to target volume delineation. With improved nonconcordance rates for newly-added network centers, these results offer the promise of improving the quality of radiotherapy delivery across an extensive cancer network with a major academic center as the nucleus.