ABSTRACT Background This study aimed to develop predictive models for local (LR), regional (RR), and distant recurrence (DR) after skull base surgery for primary malignant tumors. Methods We analyzed 2179 patients (1995–2015) from an international multicenter database, split into training ( n = 1297) and validation ( n = 882) cohorts. Cox regression identified predictors of LR, RR, and DR, and nomograms were developed and validated using concordance indexes (C‐indexes). Results Median age was 56 years; 64% were male. The most common sites included the ethmoid sinus (38%), nasal cavity (25%), and maxillary sinus (20%). Squamous cell carcinoma was the most frequent histology (25%). LR predictors included brain invasion, higher‐risk histology, pT3/T4, pN+, positive margins, and adjuvant therapy (protective). RR predictors included high‐risk histology, pT3/T4, and pN+. DR predictors mirrored LR except for adjuvant therapy. C‐indexes were around 0.7 for all endpoints. Conclusions These validated nomograms provide useful tools for recurrence risk stratification and personalized management.
Valero et al. (Thu,) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: