Prognostic Nomograms to Predict Overall Survival and Cancer-specific Survival in Patients with Anaplastic Astrocytoma: A Surveillance, Epidemiology, and End Results (SEER) Analysis

Introduction: Anaplastic astrocytoma (AA) is a grade III glioma with limited life expectancy. Developing a nomogram to predict overall survival (OS) and cancer-specific survival (CSS) is critical for individualized treatment of AA patients.

Methods: Data of patients diagnosed with AA between 1975 and 2019 were extracted from SEER database using SEER*Stat software. We identified 541 patients who were randomized into a training (n=379) and validation group (n=162) at a ratio of 7:3. Multivariate Cox regression analysis was performed in the training group to identify prognostic factors used to construct the OS and CSS nomograms. C-indices, area under the curve (AUC), and calibration curves were used to evaluate the predictive value of both nomograms at 6-, 12-, 24- and 36-months duration.

Results: Overall, patients had a median age of 50 years, and 52.7% were male. The median OS and CSS were 34.0 months (95% CI: 25.1-42.9) and 37.0 months (95% CI: 28.6-45.4), respectively. Multivariate Cox regression identified age, chemotherapy, and surgery type as independent prognostic factors affecting OS and CSS. In the training cohort, the C-indices for the OS nomogram and the CSS nomogram were 0.808 (95% CI: 0.775-0.841) and 0.807 (95% CI: 0.772-0.842), respectively. In the validation cohort, the C-indices for the OS nomogram and CSS nomogram were 0.812 (95% CI: 0.774-0.851) and 0.790 (95% CI: 0.751-0.828), respectively. The AUC values at 6-, 12-, 24-, and 36-months were 0.887, 0.845, 0.819, and 0.828 for OS and 0.890, 0.847, 0.819, and 0.822 for CSS, respectively. The calibration curves showed excellent consistency between the predicted and actual survival probabilities for both OS and CSS.

Conclusion : Age, chemotherapy, and surgery type were identified as independent prognostic factors on which nomograms with high predictive value and consistency were constructed. These nomograms may be useful in predicting an individual’s prognosis to build an individualized treatment plan.