Comparison of surveillance modalities in the surveillance of sinonasal squamous cell carcinoma recurrence: A multi‐institutional study

Abstract Background Recurrence of sinonasal squamous cell carcinoma (SNSCC) follows an aggressive course, and early detection is paramount. This study identifies the parameters of different surveillance modalities. Methods We conducted a retrospective study of 105 SNSCC patients at three academic institutions from November 2009 to July 2024. Patient records were reviewed for demographics, tumor characteristics, endoscopy, CT, PET/CT, and MRI findings. Multivariable analyses were performed in RStudio. Results Mean time to recurrence was 12.1 months (SD 13.9 months). Patients with higher Charlson Comorbidity Index ( p = 0.041), endoscopic surgical approach ( p = 0.015), and suspicious surveillance findings ( p = 0.029) had higher rates of recurrence. Endoscopy showed a sensitivity of 18.5% and specificity of 99.2%, with a positive predictive value (PPV) of 45.5% and negative predictive value (NPV) of 97.0%. CT had a sensitivity of 75.0% and specificity of 100.0%, with a PPV of 100.0% and NPV of 97.6%. PET/CT demonstrated a sensitivity of 95.2% and specificity of 90.8%, with a PPV of 64.5% and NPV of 97.6%. MRI showed a sensitivity of 72.4% and specificity of 97.1%, with a PPV of 65.6% and NPV of 97.9%. The median time from the last normal surveillance to recurrence was 2.07 months for endoscopy, 8.61 months for CT, 8.15 months for PET/CT, and 6.49 months for MRI. Conclusions The high specificity and NPV of endoscopy, alongside the high sensitivity of PET/CT, support a multimodal approach for surveillance. Given the mean onset of SNSCC recurrence at 12.1 months, surveillance beyond the National Comprehensive Cancer Network's asymptomatic 6‐month guideline is warranted, and follow‐up should be tailored to patient‐specific risk factors.