Abstract
Background
The association between membranous nephropathy (MN) and malignant tumor has long been focused. However, existing studies mostly focused on patients diagnosed of malignant tumors within a limited timeframe (typically defined as 1 year) before or after the diagnosis of MN. Actually, this represents only a subgroup of MN patients complicated with malignant tumors, and those complicated with malignant tumors without a limited period of time haven’t received attention and research. In this study, we aimed to explore the clinicopathologic characteristics of MN patients complicated with malignant tumors, and establish an effective predictive model for identifying the risk of malignant tumors in patients with MN.
Methods
A total of 194 MN patients with malignant tumors and 604 idiopathic MN patients without malignant tumors were retrospectively recruited in this study. All of the patients were then randomly separated (3:1) into the training cohort (n = 599) and the validation cohort (n = 199). A predictive model was constructed based on regression analysis and the model performance, calibration ability and clinical utility were subsequently assessed via the area under the ROC curve (AUC), calibration curve and decision curve analysis (DCA).
Results
A predictive model basedd on age, hemoglobin, degree of arteriole injury, glomerular IgG1, IgG2, IgG3, IgG4, and PLA2R deposition were constructed. The predictive model exhibited a diagnostic power of 0.890 and 0.960 in the training and validation cohorts, respectively, and was validated to demonstrate strong calibration capability and clinical utility.
Conclusion
In this largest cohort with MN and malignant tumors up to date, we constructed a model based on clinical and pathological parameters, to effectively estimate the risk of malignant tumors in patients with MN. This tool aims to assist clinicians in their decision-making process and improve the prognosis for high-risk MN patients by facilitating tumors screening at the time of initial diagnosis.