Outcome by treatment modality in sinonasal undifferentiated carcinoma (SNUC): A case-series, systematic review and meta-analysis

Objective Sinonasal undifferentiated carcinoma (SNUC) is an aggressive malignancy first described by Frierson et al. in 1986. As the tumor is very rare, current treatment recommendations are based on institutional case reports. We thus felt the need to perform a comprehensive systematic review and meta-analysis to investigate how treatment modalities are associated with survival. Design Case-series, systematic review and meta-analysis Methods We searched the OvidMedline, OvidEmbase, Web of Science, Biosis, Scopus and the Cochrane Library database libraries. We extracted aggregate and individual patient data for statistical analysis. To study the association between treatment modalities and survival, we used random-effects meta-regression for the aggregate- and cox mixed-effects models. Results 379 citations were found; 29 case series could be included in the final analysis, including a total number of 390 single patients (34.6% female). Median age at diagnosis was 52 years. 80.9% of patients presented with a T4 tumor and 16.0% with nodal metastasis at diagnosis. In individual patient data (IPD) meta-analysis, single modality (surgery alone or radiation alone) treatment was associated with reduced survival compared to double modality (surgery & radiation or chemoradiation) treatment (adjusted Hazard Ratio [aHR] 2.97, 95% ConfidenceInterval [1.41–6.27]) and compared to triple modality (surgery & radiation & chemotherapy) treatment (aHR 2.80 95%-CI 1.29–6.05 for triple vs. single modality). Triple modality treatment was not superior to double modality treatment. (aHR 1.06, 95%-CI 0.59–1.92). Conclusion Double and triple modality treatment are associated with improved survival over single modality but there is no evidence that triple modality is superior to double modality treatment. © 2017 Elsevier Ltd