ONCOLOGICAL OUTCOMES OF RADICAL NEPHRECTOMY WITH VENOUS THROMBECTOMY FOR RENAL CELL CARCINOMA AND DEVELOPMENT OF A RECURRENCE RISK CALCULATOR

Abstract

Introduction

The likelihood of recurrence after surgery for non-metastatic renal cell carcinoma (RCC) with venous tumor thrombus (VTT) remains considerable and previously developed predictive models remain underutilized in clinical practice. Adjuvant pembrolizumab was recently FDA approved and all RCC patients with VTT technically fall under the FDA approval guidance. However, we have previously conducted a cost-effectiveness analysis to demonstrate that the 3% survival benefit of adjuvant pembrolizumab outweighs its costs and risks when the 5-year risk of metastasis is at least 60%. As such, many patients with RCC and VTT may not benefit from adjuvant pembrolizumab treatment. The purpose of this study was to develop and internally validate an easy-to-use metastasis risk calculator after radical nephrectomy for non-metastatic RCC with VTT.

Methods

We performed a single-institution retrospective analysis of all adult patients who underwent radical nephrectomy with thrombectomy for non-metastatic RCC with VTT between 2000 and 2021. Demographic, clinicopathologic, and procedural characteristics were examined for association with the primary outcome of metastasis-free survival (MFS). A 70%-30% split was used to divide the cohort into a development/training and validation cohort, respectively. A least absolute shrinkage and selection operation (LASSO) Cox regression model was used to select variable combinations that best correlated with RCC metastasis. These variables were used to develop an MFS nomogram for which the area under the curve (AUC) was measured at 5 years. Decision curve analysis was performed to compare the net benefit of a nomogram-based strategy vs a treat-all strategy.

Results

Of the 532 M0 patients, 278 (52.3%), 66 (12.4%), 116 (21.8%), 35 (6.6%), and 37 (7.0%) had a level 0, I, II, III, and IV thrombus, respectively. Baseline characteristics are found in Table 1. The 5-year MFS for VTT level 0, I, II, III, IV was 51.2%, 34.7%, 28.5%, and 33.7%, respectively (p<0.01). Using LASSO feature selection, an MFS nomogram (Figure 1A) was built using four pathologic variables: thrombus level, necrosis, sarcomatoid, and positive nodes. The nomogram separated patients into low (36% of cohort), medium, and high-risk groups for metastasis (Figure 1B) with a 5-year risk of metastasis of approximately 30%, 60%, and 80%, respectively (p<0.001). The AUC at 5-years was 0.74 for both the development and validation cohorts (Figure 1C). Decision curve analysis found a significant net benefit favoring the nomogram over a treat-all strategy when adjuvant therapy treatment thresholds were over 30% metastasis risk (Figure 1D).

Conclusions

Identifying VTT patients who are at increased risk of recurrence is important in determining post-operative follow-up and potentially who might benefit from adjuvant therapy. Our study introduces a MFS nomogram that relies on just four pathologic variables to estimate postoperative recurrence in patients after radical nephrectomy for RCC with VTT. Further validation of this MFS nomogram and understanding more about the implications for post-operative management will be important in the future.