Prognostic Subgroups for Disease-Free Survival With Cutaneous Squamous Cell Carcinoma of the Head and Neck: A Secondary Analysis of a Randomized Clinical Trial.

Importance: Cutaneous squamous cell carcinoma is a common cancer, and approximately 5% of cases present with high-risk features, such as regional nodal metastases. Postoperative radiotherapy (PORT) improves outcomes in this group, but the TROG 05.01 trial found no added benefit of concurrent chemotherapy. Identification of patients with poor disease-free survival (DFS) is critical to optimize future adjuvant treatment strategies. Objective: To identify a subgroup of patients with poor DFS following PORT with or without concurrent chemotherapy who may be appropriate candidates for future adjuvant trials. Design, Setting, and Participants: This was a secondary analysis of the prospective, randomized phase 3 TROG 05.01 trial, which enrolled 321 patients with high-risk cutaneous squamous cell carcinoma of the head and neck between April 2005 and July 2014 at multiple centers in Australia. Data were analyzed in January 2025. A recursive partitioning analysis using classification and regression trees was applied to identify prognostic subgroups. Disease-free and overall survival (OS) were measured from randomization until relapse or death, with appropriate censoring. Main Outcomes and Measures: The primary outcome was DFS; secondary outcomes included OS and identification of prognostic subgroups. Results: Of 321 patients randomized, 310 (23 female individuals [7%] and 287 male individuals [93%]; median [range] age, 64 [32-85] years) initiated the allocated treatment. The recursive partitioning analysis identified 6 nodes and grouped patients into 3 DFS risk categories. A high-risk group (n = 88) was defined by the presence of extranodal extension and a nodal size of 22 mm or greater. This group had 5-year DFS and OS rates of 56% (95% CI, 45%-66%) and 59% (95% CI, 46%-69%), respectively, compared with 75% (95% CI, 68%-81%) and 85% (95% CI, 79%-90%) in the non-high-risk group. The hazard ratios for high vs non-high risk were 2.0 for DFS (95% CI, 1.3-3.1; P = .001) and 3.2 for OS (95% CI, 1.9-5.3; P < .001). Conclusions and Relevance: The results of this secondary analysis suggest that patients with extranodal extension and a nodal size of 22 mm or greater represent a high-risk subgroup with inferior DFS following PORT with or without concurrent chemotherapy. These patients may benefit from future trials that evaluate adjuvant treatment intensification. Trial Registration: ClinicalTrials.gov Identifier: NCT00193895.