Vinícius Luiz Menezes Jesus, Gabriel Carvalho Dos Anjos, Gabriel Kledeglau Jahchan Alves, Edgar Alexandre Aviles Sandoval, Guilherme Garcia Barros, David Hamilton Cho, Francisco Valdez Santos Lima, Paulo Afonso de Carvalho, Fabio Pescarmona Gallucci, Leonardo Cardili, Jose Pontes, William Carlos Nahas, Mauricio Dener Cordeiro
We developed and internally validated two nomograms to predict recurrence after partial nephrectomy in patients with renal cell carcinoma. These models are based on readily available clinicopathological factors and provide individualized risk estimates. While external clinical validation is required, they may help guide postoperative counseling and surveillance.
INTRODUCTION: Partial nephrectomy is the gold standard treatment for localized renal tumors, yet recurrence remains a significant concern. Most existing prognostic tools have been developed using data from radical nephrectomy cohorts, often failing to adequately predict recurrence after partial nephrectomy. We aimed to identify independent predictors of recurrence and develop clinically applicable nomograms to forecast both overall and local recurrence.
MATERIALS AND METHODS: We retrospectively reviewed data from 694 patients who underwent partial nephrectomy for non-metastatic renal cell carcinoma at a tertiary hospital between 2001 and 2020. Demographic, clinical, and pathological variables were analyzed using univariable and multivariable Cox regression analyses. Independent predictors were incorporated into nomograms for overall and local recurrence rates. Discriminatory accuracy was evaluated using the concordance index (c-index) and calibration plots.
RESULTS: During a median follow-up of 61.6 months, 70 patients (10.1%) experienced recurrence (48 local and 44 distant), while 22 patients experienced both. Positive surgical margins, Fuhrman grade 3 to 4, tumor size ≥4.5 cm, and male sex were independent predictors of overall recurrence. For local recurrence, positive margins and Fuhrman grades 3 to 4 remained significant. Five-year recurrence-free survival was 73.8% for patients with positive margins versus 92.6% for those with negative margins (P < .0001). The nomograms demonstrated good discrimination, with c-index values of 0.78 for overall recurrence and 0.80 for local recurrence, and calibration showed close agreement between the predicted and observed probabilities.
CONCLUSION: We developed and internally validated two nomograms to predict recurrence after partial nephrectomy in patients with renal cell carcinoma. These models are based on readily available clinicopathological factors and provide individualized risk estimates. While external clinical validation is required, they may help guide postoperative counseling and surveillance.