Abstract: PD20-06
Sources of Funding: none

Introduction

Partial Nephrectomy (PN) is a well established treatment for small localized disease – pT1a tumors. Better renal function compared to radical nephrectomy (RN), with similar oncologic control, expanded the PN indication for larger and more complex tumors in some reference institution, including pT3a. The role of PN in this type of tumor still unknown and the potential risk of positive margins and negative outcomes are uncertain. We performed this study to assess the impact of PN on outcomes among patients with pT3a tumors

Methods

692 patients underwent PN or RN from 2005 to 2016 at MSKCC and expressed a pathological stage of pT3a. 18 patients with metastasis at diagnosis, 1 patient with a single kidney, 1 patient with hereditary cancer syndrome and 68 patients with histology other than Clear Cell, Chromophobe, or Papillary were excluded. Group comparisons were made using Wilcoxon rank-sum and Fisher’s exact test for continuous and categorical variables, respectively. Outcomes of interest included estimated blood loss, warm ischemia time, estimated glomerular filtration rate at 6 months, length of stay, margin status, Fuhrman grade, tumor size, pathological histology, and symptoms index at presentation

Results

Among our cohort, 376 (62%) and 228 (38%) patients with were scheduled for RN and PN, respectively. Of the 228 patients originally scheduled for PN, 12% were converted to RN intraoperatively. A smaller proportion of patients scheduled to undergo PN had clear cell/conventional histology (77% vs 88%; p=0.001) on pathology compared to patients scheduled for RN. Among patients with clear cell or papillary histology, a larger proportion of patients scheduled for PN had lower Fuhrman grade (24% vs 10.3% had FG 1 or 2; p<0.0001) on pathology than patients scheduled for RN. Of our 604 patients, 111 patients died, 33 from kidney disease. The median follow up time for survivors was 2.0 years from surgery. On multivariable analysis, scheduled PN was non-significantly associated with better OS (HR 0.62; 95% C.I. 0.37, 1.03; p = 0.064), better CSS (HR 0.51; 95% C.I. 0.18, 1.49; p = 0.2), and better RFS (HR 0.56; 95% C.I. 0.29, 1.07; p=0.081). From the estimates of the hazard ratio, we suspect that the bias related to surgeons choosing PN or RN based on low or high risk disease is not appropriately adjusted for in our model

Conclusions

We found no evidence to suggest that PN has poorer outcomes than RN in patients with pT3a tumors. The inherent benefits of PN on renal function preservation make this approach very attractive even in larger and complex tumors

Funding

none