# Prognostic value of the GRANT score and development of a nomogram in papillary renal cell carcinoma: a SEER-based study with external validation in a Chinese cohort

## TL;DR

This study validates the GRANT score and creates a new nomogram to predict survival in papillary renal cell carcinoma patients, using data from the SEER database and a Chinese cohort.

## Contribution

A novel nomogram was developed and externally validated for predicting survival in papillary renal cell carcinoma patients.

## Key findings

- The GRANT score is an independent prognostic factor for overall and cancer-specific survival in pRCC patients.
- The nomogram achieved high C-indices for predicting cancer-specific survival in training, internal validation, and external validation cohorts.
- Risk stratification using the nomogram effectively distinguished low-, intermediate-, and high-risk patient groups with different survival outcomes.

## Abstract

Papillary renal cell carcinoma (pRCC) exhibits significant heterogeneity, and robust prognostic tools specifically validated for this subtype are lacking. The GRANT score, incorporating grade, age, nodes, and tumor stage, shows promise but requires extensive validation in pRCC-specific cohorts. This study aimed to evaluate the prognostic value of the GRANT score and develop a novel nomogram for predicting survival in pRCC patients.

A multi-center retrospective study was conducted. Patients undergoing surgery for pRCC were identified from the SEER database (2004-2015) and formed the training (n=4,001) and internal validation (n=1,689) cohorts. An external validation cohort (n=151) was sourced from a Chinese institution. Overall survival (OS) and cancer-specific survival (CSS) were primary endpoints. The GRANT score was calculated for all patients. Univariate and multivariate Cox analyses identified independent prognostic factors, which were incorporated into nomograms for predicting 1-, 3-, and 5-year OS and CSS. Model performance was assessed using the concordance index (C-index), time-dependent receiver operating characteristic curves, and calibration plots.

Multivariate analysis confirmed the GRANT score as an independent prognostic factor for both OS and CSS. The prognostic nomograms integrated key variables, including surgical approach, marital status, TNM stage, tumor size, Fuhrman grade, and the GRANT score. For OS prediction, the nomogram achieved C-indices of 0.711 (training), 0.720 (internal validation), and 0.740 (external validation). For CSS prediction, the model demonstrated superior performance, with C-indices of 0.860 (training), 0.873 (internal validation), and 0.826 (external validation). Calibration curves showed excellent agreement between predicted and observed outcomes. Risk stratification based on nomogram scores effectively distinguished low-, intermediate-, and high-risk patient groups with significantly different survival.

This study validates the GRANT score as an independent prognostic factor in a large pRCC cohort. The developed and externally validated nomogram provides a clinically useful tool with robust performance, particularly for predicting CSS, facilitating personalized risk assessment and postoperative management for pRCC patients.