• 027020

    A Preoperative Risk Classifier Predicts Tumor Progression in Patients with Cranial Base Chondrosarcoma


    Eric Wang, Michael McDowell, Ezequiel Goldschmidt, Paul Gardner, Carl Snyderman, Andrew Venteicher


    Introduction: Cranial base chondrosarcomas frequently involve critical neurovascular structures that make complete resection challenging. Predictors of tumor progression in the skull base as unknown.

    Objective: To identify clinical features of cranial base chondrosarcoma that will predict tumor progression.

    Methods: We performed a retrospective analysis at the University of Pittsburgh from 2004 to 2018. We recorded tumor locations and radiographic features. Our primary outcome was progression-free survival (PFS). Univariate and multivariate regression models were created. Of 41 patients treated for skull base chondrosarcoma, the mean age was 52 years, 60.5% were female, and mean follow-up was 55 months. For the entire cohort, the median PFS was 123 months.

    Results: Older age was correlated with shorter PFS (p < 0.017). We performed univariate regression analysis of involved tumor locations and of radiographic parameters, and those parameters with a p-value of < 0.1 were fed into a Cox proportional hazards model. This revealed an independent association of both age (p < 0.018) and increasing number of arteries encased (p < 0.045) with PFS.

    We next generated a grading scale based on both age and number of encased arteries (Figure 1a).   This allowed us to risk stratify patients into a high risk (10% patients; median PFS 22 months), intermediate risk (66% patients; median PFS 97 months), and low risk groups (24% patients; none progressed) (Figure 4b).

    Conclusions: In patients with skull base chondrosarcomas, age and artery encasement are two preoperative variables that predict tumor progression, which may help guide post-operative follow-up and adjuvant therapy.

We use cookies to improve the performance of our site, to analyze the traffic to our site, and to personalize your experience of the site. You can control cookies through your browser settings. Please find more information on the cookies used on our site here. Privacy Policy