A Simplified Prediction Tool for Metastasis After Nephrectomy With Venous Tumor Thrombectomy for RCC.

OBJECTIVE

To develop and comprehensively validate a simplified and easy-to-use prediction tool for the risk of metastatic recurrence after surgery for non-metastatic renal cell carcinoma with an associated venous tumor thrombus.

SUBJECTS/PATIENTS AND

METHODS

We evaluated 532 patients who underwent nephrectomy and venous thrombectomy for M0 RCC from 2000 to 2021. Clinicopathological and surgical data were analyzed. A 70-30 split (test and train cohort, respectively) and the least absolute shrinkage and selection operation regression (LASSO) model were used to select variables independently associated with recurrence. We developed a nomogram and validated it both internally and externally with 2 separate institutional datasets. Additionally, we performed decision curve analysis for the use of adjuvant Pembrolizumab.

RESULTS

We included 278 (52.3%), 66 (12.4%), 116 (21.8%), 35 (6.6%), and 37 (7.0%) patients with level zero through 4 tumor thrombi, respectively. Overall, estimated 5-year metastasis-free survival (MFS) was 41.3%. The resulting nomogram was composed of thrombus level, tumor necrosis, sarcomatoid features, and pN stage. AUC at 5 years was 0.74 in both our internal test and train cohorts, and 0.71 and 0.68 in 2 external cohorts. On decision curve analytics, our nomogram adds net benefit at a threshold probability between 0.33 and 0.80.

CONCLUSION

We provide an internally and externally validated risk calculator for 5-year MFS following radical nephrectomy with tumor thrombectomy. This prognostic tool may be used to guide selection for adjuvant therapies and to assist in clinical trial design.