Prediction of post‑nephroureterectomy renal function in non‑dialysis upper tract urothelial carcinoma using machine learning with PyCaret and SHAP explanations

Lee,Hsiang Ying; Li,Wei-Ming; Yeh,Hsin-Chih; Wu,Wen-Jeng; Li,Ching-Chia; Ke,Hung-Lung; Wang,Yen-Chun

doi:10.3892/ol.2025.14946

Prediction of post‑nephroureterectomy renal function in non‑dialysis upper tract urothelial carcinoma using machine learning with PyCaret and SHAP explanations

Authors:
- Hsiang Ying Lee
- Wei-Ming Li
- Hsin-Chih Yeh
- Wen-Jeng Wu
- Ching-Chia Li
- Hung-Lung Ke
- Yen-Chun Wang
View Affiliations

Affiliations: Department of Urology, Kaohsiung Medical University Hospital, Kaohsiung 807, Taiwan, R.O.C.
Published online on: February 25, 2025 https://doi.org/10.3892/ol.2025.14946
Article Number: 200
Copyright: © Lee et al. This is an open access article distributed under the terms of Creative Commons Attribution License.

Metrics: Total Views: 0 (Spandidos Publications: | PMC Statistics: )

Metrics: Total PDF Downloads: 0 (Spandidos Publications: | PMC Statistics: )

Cited By (CrossRef): 0 citations Loading Articles...

Abstract

Renal function after radical nephroureterectomy (RNU) in patients with advanced upper tract urothelial carcinoma (UTUC) is an indicator of eligibility for postoperative chemotherapy. The present study aimed to utilize machine learning for predicting renal function status after RNU and to investigate the contribution of several features to this prediction. The present retrospective study included 764 medical records of patients with non‑metastatic UTUC who received RNU from 2008 to 2022. The records were divided into training (n=534) and testing (n=230) datasets. Several demographic and clinicopathological parameters were collected for analysis. PyCaret was utilized for data processing, model establishment and comparison, while SHapley Additive exPlanations values were adopted to assess the contribution of each variable to the predictive performance of the model. The results demonstrated that the Random Forest Regressor model had improved accuracy in predicting postoperative renal function compared with other algorithms. In both the estimated glomerular filtration rate (eGFR) and chronic kidney disease outcome, preoperative renal function had the most marked contribution to postoperative renal function. Additionally, the Charlson Comorbidity Index (CCI), body mass index (BMI) and tumor size were highly associated with renal function after RNU, and hydronephrosis was another main factor for predicting postoperative eGFR. The retrospective design was a limitation of the present study. The results demonstrated that important predictors, including preoperative eGFR, tumor size, BMI, CCI and hydronephrosis, were associated with postoperative renal function. To conclude, using a machine learning prediction model may be used in the future to determine appropriate therapeutic strategies and inform the timing of perioperative systemic chemotherapy, in particular perioperative systemic chemotherapy, for advanced UTUC.

View Figures

View References

1	Han J, Xian Z, Zhang Y, Liu J and Liang A: Systematic overview of aristolochic acids: Nephrotoxicity, carcinogenicity, and underlying mechanisms. Front Pharmacol. 10:6482019. View Article : Google Scholar : PubMed/NCBI
2	Chang YH, Hsu WL, Lee YK, Chiang CJ, Yang YW, You SL, Chen YC and Lai TS: Trends and sex-specific incidence of upper urinary tract cancer in Taiwan: A birth cohort study. Cancer Med. 12:15350–15357. 2023. View Article : Google Scholar : PubMed/NCBI
3	Yeh HC, Chang CH, Fang JK, Chen IHA, Lin JT, Hong JH, Huang CY, Wang SS, Chen CS, Lo CW, et al: The value of preoperative local symptoms in prognosis of upper tract urothelial carcinoma after radical nephroureterectomy: A retrospective, multicenter cohort study. Front Oncol. 12:8728492022. View Article : Google Scholar : PubMed/NCBI
4	Chen IA, Chang CH, Huang CP, Wu WJ, Li CC, Chen CH, Huang CY, Lo CW, Yu CC, Tsai CY, et al: Factors predicting oncological outcomes of radical nephroureterectomy for upper tract urothelial carcinoma in Taiwan. Front Oncol. 11:7665762022. View Article : Google Scholar : PubMed/NCBI
5	Margulis V, Shariat SF, Matin SF, Kamat AM, Zigeuner R, Kikuchi E, Lotan Y, Weizer A, Raman JD and Wood CG; Upper Tract Urothelial Carcinoma CollaborationThe Upper Tract Urothelial Carcinoma Collaboration, : Outcomes of radical nephroureterectomy: A series from the upper tract urothelial carcinoma collaboration. Cancer. 115:1224–1233. 2009. View Article : Google Scholar : PubMed/NCBI
6	Chang LW, Hung SC, Chen CS, Li JR, Chiu KY, Wang SS, Yang CK, Lu K, Chen CC, Wang SC, et al: Geriatric nutritional risk index as a prognostic marker for patients with upper tract urothelial carcinoma receiving radical nephroureterectomy. Sci Rep. 13:45542023. View Article : Google Scholar : PubMed/NCBI
7	Abouassaly R, Alibhai SMH, Shah N, Timilshina N, Fleshner N and Finelli A: Troubling outcomes from population-level analysis of surgery for upper tract urothelial carcinoma. Urology. 76:895–901. 2010. View Article : Google Scholar : PubMed/NCBI
8	Birtle A, Johnson M, Chester J, Jones R, Dolling D, Bryan RT, Harris C, Winterbottom A, Blacker A, Catto JWF, et al: Adjuvant chemotherapy in upper tract urothelial carcinoma (the POUT trial): A phase 3, open-label, randomised controlled trial. Lancet. 395:1268–1277. 2020. View Article : Google Scholar : PubMed/NCBI
9	Leow JJ, Chong YL, Chang SL, Valderrama BP, Powles T and Bellmunt J: Neoadjuvant and adjuvant chemotherapy for upper tract urothelial carcinoma: A 2020 systematic review and meta-analysis, and future perspectives on systemic therapy. Eur Urol. 79:635–654. 2021. View Article : Google Scholar : PubMed/NCBI
10	Petros FG, Qiao W, Singla N, Clinton TN, Robyak H, Raman JD, Margulis V and Matin SF: Preoperative multiplex nomogram for prediction of high-risk nonorgan-confined upper-tract urothelial carcinoma. Urol Oncol. 37:292.e1–292.e9. 2019. View Article : Google Scholar : PubMed/NCBI
11	Levey AS, Stevens LA, Schmid CH, Zhang YL, Castro AF III, Feldman HI, Kusek JW, Eggers P, Van Lente F, Greene T, et al: A new equation to estimate glomerular filtration rate. Ann Intern Med. 150:604–612. 2009. View Article : Google Scholar : PubMed/NCBI
12	Webster AC, Nagler EV, Morton RL and Masson P: Chronic kidney disease. Lancet. 389:1238–1252. 2017. View Article : Google Scholar : PubMed/NCBI
13	Kidney Disease: Improving Global Outcomes (KDIGO) CKD-MBD Work Group: KDIGO clinical practice guideline for the diagnosis, evaluation, prevention, and treatment of chronic kidney disease-mineral and bone disorder (CKD-MBD). Kidney Int Suppl. 76 (Suppl 113):S1–S130. 2009.
14	Ray S: ‘A quick review of machine learning algorithms’. 2019 International Conference on Machine Learning, Big Data, Cloud and Parallel Computing (COMITCon); Faridabad, India: pp. 35–39. 2019
15	Favaretto RL, Shariat SF, Savage C, Godoy G, Chade DC, Kaag M, Bochner BH, Coleman J and Dalbagni G: Combining imaging and ureteroscopy variables in a preoperative multivariable model for prediction of muscle-invasive and non-organ confined disease in patients with upper tract urothelial carcinoma. BJU Int. 109:77–82. 2012. View Article : Google Scholar : PubMed/NCBI
16	Qi Y: Random forest for bioinformatics. Zhang C and Ma Y: Ensemble Machine Learning. Springer; New York, NY: pp. 307–323. 2012
17	Smith PF, Ganesh S and Liu P: A comparison of random forest regression and multiple linear regression for prediction in neuroscience. J Neurosci Methods. 220:85–91. 2013. View Article : Google Scholar : PubMed/NCBI
18	Tafuri A, Marchioni M, Cerrato C, Mari A, Tellini R, Odorizzi K, Veccia A, Amparore D, Shakir A, Carbonara U, et al: Changes in renal function after nephroureterectomy for upper urinary tract carcinoma: Analysis of a large multicenter cohort (radical nephroureterectomy outcomes (RaNeO) research consortium). World J Urol. 40:2771–2779. 2022. View Article : Google Scholar : PubMed/NCBI
19	Kaag M, Trost L, Thompson RH, Favaretto R, Elliott V, Shariat SF, Maschino A, Vertosick E, Raman JD and Dalbagni G: Preoperative predictors of renal function decline after radical nephroureterectomy for upper tract urothelial carcinoma. BJU Int. 114:674–679. 2014. View Article : Google Scholar : PubMed/NCBI
20	Rodríguez Faba O, Palou J, Breda A, Maroto P, Fernández Gómez JM, Wong A and Villavicencio H: Predictive factors for impaired renal function following nephroureterectomy in upper urinary tract urothelial cell carcinoma. Urol Int. 92:169–173. 2014. View Article : Google Scholar : PubMed/NCBI
21	Song W, Sung HH, Han DH, Jeong BC, Seo SI, Jeon SS, Lee HM, Choi HY and Jeon HG: The effect of contralateral kidney volume on renal function after radical nephroureterectomy: Implications for eligibility for neoadjuvant chemotherapy for upper tract urothelial cancer. Urol Oncol. 35:114.e1–114.e7. 2017. View Article : Google Scholar : PubMed/NCBI
22	Yun HR, Kim H, Park JT, Chang TI, Yoo TH, Kang SW, Choi KH, Sung S, Kim SW, Lee J, et al: Obesity, metabolic abnormality, and progression of CKD. Am J Kidney Dis. 72:400–410. 2018. View Article : Google Scholar : PubMed/NCBI
23	Madero M, Katz R, Murphy R, Newman A, Patel K, Ix J, Peralta C, Satterfield S, Fried L, Shlipak M and Sarnak M: Comparison between different measures of body fat with kidney function decline and incident CKD. Clin J Am Soc Nephrol. 12:893–903. 2017. View Article : Google Scholar : PubMed/NCBI
24	Kanno T, Kobori G, Saito R, Ito K, Nakagawa H, Takahashi T, Koterazawa S, Takaoka N, Somiya S, Haitani T, et al: Hydronephrosis severity as a predictor of postoperative renal function decline following laparoscopic radical nephroureterectomy. Int J Clin Oncol. 29:464–472. 2024. View Article : Google Scholar : PubMed/NCBI
25	Singla N, Hutchinson R, Haddad A, Sagalowsky A, Lotan Y and Margulis V: Preoperative hydronephrosis is associated with less decline in renal function after radical nephroureterectomy for upper tract urothelial carcinoma. Can J Urol. 23:8334–8341. 2016.PubMed/NCBI
26	Fang D, Zhang Q, Li X, Qian C, Xiong G, Zhang L, Chen X, Zhang X, Yu W, He Z and Zhou L: Nomogram predicting renal insufficiency after nephroureterectomy for upper tract urothelial carcinoma in the Chinese population: Exclusion of ineligible candidates for adjuvant chemotherapy. Biomed Res Int. 2014:5291862014. View Article : Google Scholar : PubMed/NCBI
27	Hashimoto T, Ohno Y, Nakashima J, Gondo T, Nakagami Y, Namiki K, Horiguchi Y, Yoshioka K, Ohori M and Tachibana M: Prediction of renal function after nephroureterectomy in patients with upper tract urothelial carcinoma. Jpn J Clin Oncol. 45:1064–1068. 2015. View Article : Google Scholar : PubMed/NCBI
28	Hensley PJ, Labbate C, Zganjar A, Howard J, Huelster H, Durdin T, Pham J, Xiao L, Pallauf M, Lombardo K, et al: Development and validation of a multivariable nomogram predictive of post-nephroureterectomy renal function. Eur Urol Oncol. 7:1313–1319. 2024. View Article : Google Scholar : PubMed/NCBI
29	Liu J, Wu P, Lai S, Wang J, Hou H and Zhang Y: Prognostic models for upper urinary tract urothelial carcinoma patients after radical nephroureterectomy based on a novel systemic immune-inflammation score with machine learning. BMC Cancer. 23:5742023. View Article : Google Scholar : PubMed/NCBI