DEVELOPMENT AND VALIDATION OF A NOVEL NOMOGRAM TO PREDICT LYMPH NODE INVASION IN UPPER TRACT UROTHELIAL CARCINOMA
Recommended Citation
Saitta C, Afari JA, Hakimi K, Nguyen MV, Meagher MF, Wang L, Derweesh IH, Autorino R, Pandolfo S, Chiarelli G, Abdollah F, Davis M, Stephens A, Bell SH, Simone G, Lughezzani G, Buffi NM, Tuderti G, Ferro M, Yong C, Sundaram CP, Tozzi M, Taylor J, Margulis V, Checcucci E, Porpiglia F, Wood E, Ghoreifi A, Djaladat H, Wang L, Eilender BM, Mehrazin R, Gonzalgo ML. DEVELOPMENT AND VALIDATION OF A NOVEL NOMOGRAM TO PREDICT LYMPH NODE INVASION IN UPPER TRACT UROTHELIAL CARCINOMA. Urol Oncol 2024; 42:S103-S103.
Document Type
Conference Proceeding
Publication Date
3-1-2024
Publication Title
Urol Oncol
Abstract
Introduction: The role of lymphadenectomy in upper tract urothelial carcinoma (UTUC) remains controversial. We sought to develop a preoperative nomogram to predict nodal tropism (NT) defined as nodes invasion at the histological report (NI) or presence of nodes metastasis (NM) at follow up. Methods: We conducted a retrospective analysis of the ROBUUST database of UTUC patients who underwent robotic nephroureterectomy. NI was defined as presence of positive nodes at final histological report, while NM was defined as the emergence of newly detected retroperitoneal lymphadenopathy (>10 mm) during the follow-up period. Patients who underwent neoadjuvant or adjuvant chemotherapy were excluded from analysis. Primary objective was to develop a predictive model for NT. The model was developed through a stepwise multivariable logistic regression (MLR).;;Secondary outcomes pertain to internal validation though cross validation analysis. Accuracy of the model was tested with receiver operator characteristic/area under the curve (AUC), and calibration plot. Results: 1,117 were analyzed [755(64.1%) male and 422(35.8%) female]. On MVR cN+[Odds ratio(OR) 8.19, p<0.001];;cT4 vs cTa (OR 10.38, p=0.012);;history of bladder cancer (OR 6.66, p<0.001) high-grade cytology (OR 2.90, p<0.001), platelets lymphocyte ratio≥130 (OR 1.77, p=0.021), diabetes mellitus (OR 1.90, p=0.023), and symptoms at diagnosis (OR 2.16, p=0.008) were independent predictors for NT. A nomogram was developed based on the MVL (Figure 1). AUC of the model was 0.83 (Figure 2). AUC after internal validation was 0.81 (95% confidence interval 0.76-0.87;;Figure 2). A 7% threshold probability demonstrated 80.2% sensitivity, 75.4% specificity and 97.5% negative predictive value. Conclusions: By integrating patient characteristics and serum biomarkers, we've engineered a predictive model for node tropism. This model has the potential to improve clinical choices, with respect to the performance of lymphadenectomy at the time of surgery, post-surgery monitoring, and may spur considerations;for supplementary treatment.;Further investigation is requisite.
Volume
42
First Page
S103
Last Page
S103
