Tumor contour irregularity on preoperative imaging: a practical and useful prognostic parameter for papillary renal cell carcinoma

• Published in: European radiology 2021-06, Vol.31 (6), p.3745-3753
• Main Authors: Dai, Chenchen, Huang, Jiaqi, Li, Yaohui, Zhang, Sihong, Tan, Qinxuan, Hou, Jun, Tang, Qiying, Hu, Xiaoyi, Guo, Jianming, Zeng, Mengsu, Wang, Hang, Zhou, Jianjun
• Format: Article
• Language: English
• Subjects: Contours; Cross-sections; Diagnostic Radiology; Imaging; Impact analysis; Internal Medicine; Interventional Radiology; Irregularities; Kidney cancer; Mathematical models; Medical imaging; Medical prognosis; Medicine; Medicine & Public Health; Metastases; Neuroradiology; Papillary renal cell carcinoma; Parameters; Radiology; Risk; Risk groups; Survival; Tumors; Ultrasound; Urogenital
• ISSN: 0938-7994; 1432-1084
• DOI: 10.1007/s00330-020-07456-7

Objectives To illustrate tumor contour irregularity on preoperative imaging with a practical method and further determine its value in predicting disease-free survival (DFS) in patients with pRCC (papillary renal cell carcinoma). Methods We performed a retrospective single-institution review of 267 Chinese pRCC patients between March 2009 and May 2019. Contour irregularity on cross-section was classified into smooth but distorted margin, unsmooth and sharply nodular margin, and blurred margin. Then, the ratio of the cross-section numbers of irregularity and the total tumor was defined as the contour irregular degree (CID). Cox regression and Kaplan-Meier analysis were performed to analyze the impact of CID on DFS. Then, the prognostic performance of CID was compared with pRCC risk stratification published by Leibovich et al. Results The median follow-up was 45 months (IQR: 23–69), in which 27 (10%) patients had metastasis or recurrence. Observed DFS rates were 95%, 90%, and 88% at 1, 3, and 5 years. The CID was an independent prognostic factor of DFS (HR = 1.048, 95% CI = 1.029–1.068, p < 0.001). The Kaplan-Meier plot showed that high-risk patients (CID ≥ 50%) tended to have a significantly shorter DFS ( p < 0.001). The CID and Leibovich’s pRCC model for DFS prediction had a C-index of 0.934 (95% CI = 0.907–0.961) and 0.833 (95% CI = 0.739–0.927) respectively. Conclusions With our standard and practical method, the CID can be a reliable imaging marker for DFS prediction in patients with pRCC. Key Points • The updated contour irregularity was an independent parameter for predicting disease-free survival in patients with pRCC. • High-risk pRCC patients (contour irregular degree ≥ 50%) tended to have a shorter disease-free survival. • Tumor contour irregularity in pRCC risk stratification outperformed Leibovich’s model from our cohort.