A Novel Machine Learning Algorithm Combined With Multivariate Analysis for the Prognosis of Renal Collecting Duct Carcinoma

ObjectivesTo investigate the clinical and non-clinical characteristics that may affect the prognosis of patients with renal collecting duct carcinoma (CDC) and to develop an accurate prognostic model for this disease.MethodsThe characteristics of 215 CDC patients were obtained from the U.S. National Cancer Institute’s surveillance, epidemiology and end results database from 2004 to 2016. Univariate Cox proportional hazard model and Kaplan-Meier analysis were used to compare the impact of different factors on overall survival (OS). 10 variables were included to establish a machine learning (ML) model. Model performance was evaluated by the receiver operating characteristic curves (ROC) and calibration plots for predictive accuracy and decision curve analysis (DCA) were obtained to estimate its clinical benefits.ResultsThe median follow-up and survival time was 16 months during which 164 (76.3%) patients died. 4.2, 32.1, 50.7 and 13.0% of patients were histological grade I, II, III, and IV, respectively. At diagnosis up to 61.9% of patients presented with a pT3 stage or higher tumor, and 36.7% of CDC patients had metastatic disease. 10 most clinical and non-clinical factors including M stage, tumor size, T stage, histological grade, N stage, radiotherapy, chemotherapy, age at diagnosis, surgery and the geographical region where the care delivered was either purchased or referred and these were allocated 95, 82, 78, 72, 49, 38, 36, 35, 28 and 21 points, respectively. The points were calculated by the XGBoost according to their importance. The XGBoost models showed the best predictive performance compared with other algorithms. DCA showed our models could be used to support clinical decisions in 1-3-year OS models.ConclusionsOur ML models had the highest predictive accuracy and net benefits, which may potentially help clinicians to make clinical decisions and follow-up strategies for patients with CDC. Larger studies are needed to better understand this aggressive tumor.

Protein kinase A catalytic-α and catalytic-β proteins have nonredundant regulatory functions

Vasopressin regulates osmotic water transport in the renal collecting duct by protein kinase A (PKA)-mediated control of the water channel aquaporin-2 (AQP2). Collecting duct principal cells express two seemingly redundant PKA catalytic subunits, PKA catalytic α (PKA-Cα) and PKA catalytic β (PKA-Cβ). To identify the roles of these two protein kinases, we carried out deep phosphoproteomic analysis in cultured mpkCCD cells in which either PKA-Cα or PKA-Cβ was deleted using CRISPR-Cas9-based genome editing. Controls were cells carried through the genome editing procedure but without deletion of PKA. TMT mass tagging was used for protein mass spectrometric quantification. Of the 4,635 phosphopeptides that were quantified, 67 phosphopeptides were significantly altered in abundance with PKA-Cα deletion, whereas 21 phosphopeptides were significantly altered in abundance with PKA-Cβ deletion. However, only four sites were changed in both. The target proteins identified in PKA-Cα-null cells were largely associated with cell membranes and membrane vesicles, whereas target proteins in PKA-Cβ-null cells were largely associated with the actin cytoskeleton and cell junctions. In contrast, in vitro incubation of mpkCCD proteins with recombinant PKA-Cα and PKA-Cβ resulted in virtually identical phosphorylation changes. In addition, analysis of total protein abundances in in vivo samples showed that PKA-Cα deletion resulted in a near disappearance of AQP2 protein, whereas PKA-Cβ deletion did not decrease AQP2 abundance. We conclude that PKA-Cα and PKA-Cβ serve substantially different regulatory functions in renal collecting duct cells and that differences in phosphorylation targets may be due to differences in protein interactions, e.g., mediated by A-kinase anchor proteins, C-kinase anchoring proteins, or PDZ binding.

Protein kinase A catalytic-α and catalytic-β proteins have non-redundant functions

ABSTRACTVasopressin regulates osmotic water transport in the renal collecting duct by PKA-mediated control of the water channel aquaporin-2 (AQP2). Collecting duct principal cells express two seemingly redundant PKA catalytic subunits, PKA catalytic α (PKA-Cα) and PKA catalytic β (PKA-Cβ). To identify the roles of these two protein kinases, we carried out deep phosphoproteomic analysis in cultured mpkCCD cells in which either PKA-Cα or PKA-Cβ was deleted using CRISPR-Cas9-based genome editing. Controls were cells carried through the genome editing procedure, but without deletion of PKA. TMT mass tagging was used for protein mass spectrometric quantification. Of the 4635 phosphopeptides that were quantified 67 were significantly altered in abundance with PKA-Cα deletion, while 21 were significantly altered in abundance with PKA-Cβ deletion. However, only four sites were changed in both. The target proteins identified in PKA-Cα-null cells were largely associated with cell membranes and membrane vesicles, while target proteins in the PKA-Cβ-null cells were largely associated with the actin cytoskeleton and cell junctions. In contrast, in vitro incubation of mpkCCD proteins with recombinant PKA-Cα and PKA-Cβ resulted in virtually identical phosphorylation changes. In addition, analysis of total protein abundances in the in vivo samples showed that PKA-Cα deletion resulted in a near disappearance of AQP2 protein, while PKA-Cβ deletion did not decrease AQP2 abundance. We conclude that PKA-Cα and PKA-Cβ serve substantially different functions in renal collecting duct cells and that differences in phosphorylation targets may be due to differences in protein interactions, e.g. mediated by AKAP, C-KAP or PDZ binding.