3582 - Comparative Outcomes and Toxicity of Surgical and Nonsurgical Salvage Approaches for Locoregionally Recurrent Head and Neck Cancer

Purpose/Objective(s): Management of locoregionally recurrent head and neck cancer (LRR-HNC) remains challenging, particularly in patients eligible for salvage interventions. We aimed to evaluate oncologic outcomes among patients with LRR-HNC treated with different salvage strategies and to assess the association between salvage modality and treatment-related toxicity.

Materials/Methods:

We performed a retrospective single-institution analysis of 50 patients with locoregionally recurrent head and neck cancer treated with salvage therapy. Descriptive statistics are reported as mean (SD), median (IQR), or proportions with 95% exact (Clopper-Pearson) confidence intervals (CI). Fisher’s exact test was used to compare categorical outcomes. Overall survival (OS) was estimated using Kaplan–Meier methods. Statistical significance was defined as p<0.05. Analyses were performed using SAS v9.4. Primary endpoints included development of second recurrence and time to second recurrence. Secondary endpoints included OS and treatment-related toxicity graded by CTCAE.

Results: Among 50 patients, mean age was 71.4±12.9 years, 64% male, and predominantly squamous cell carcinoma (82%). Primary salvage modalities for first recurrence included surgery (24%), radiation (24%), chemotherapy/immunotherapy combinations (24%), immunotherapy alone (8%), and other systemic approaches. Second recurrence occurred in 16% of patients (8/50; 95% CI 7–29%). Among those with a second recurrence, median time to second recurrence following first recurrence was 9.5 months (IQR 6.5–26.0). For treatment of second recurrence, 37.5% received immunotherapy alone, 25% radiation alone, 12.5% chemo-immunotherapy-radiation, and 12.5% surgery-chemo-radiation. Any treatment-related toxicity was observed in 90% of patients. Maximum toxicity grades were: grade 1 (30%), grade 2 (34%), and grade 3 (26%) . No grade 4–5 toxicities were reported. There was no statistically significant association between primary salvage modality and presence of toxicity (Chi-square p=0.59; Fisher’s exact p=0.50) , nor between modality and maximum toxicity grade (Fisher’s exact p=0.58). At last follow-up, 50% of patients had died. Among evaluable patients (n=39), median OS was 46.0 months (95% CI 24.0–78).

Conclusion: In this retrospective cohort of LRR-HNC, second recurrence occurred in 16% of patients with a median time of 16 months. Median OS approached 4 years. Although treatment-related toxicity was common, severe (grade =3) toxicity occurred in approximately one-quarter of patients and did not significantly differ by salvage modality. These findings support individualized selection of salvage strategy based on clinical context, as comparable oncologic and toxicity outcomes were observed across modalities.