RESUMO
Objective We examine the heterogeneity and distribution of the cohort populations in two publicly used radiological image cohorts, Cancer Genome Atlasâ¯Kidney Renal Clear Cell Carcinoma (TCIA TCGA KIRC) collection and 2019 MICCAI Kidney Tumor Segmentation Challenge (KiTS19), and deviations in real world population renal cancer data from National Cancer Database (NCDB) Participant User Data File (PUF) and tertiary center data. PUF data is used as an anchor for prevalence rate bias assessment. Specific gene expression and therefore biology of RCC differ by self-reported race especially between the African American and Caucasian population. AI algorithms learn from datasets, but if the dataset misrepresents the population, reinforcing bias may occur. Ignoring these demographic features may lead to inaccurate downstream effects, thereby limiting the translation of these analyses to clinical practice. Consciousness of model training biases is vital to patient care decisions when using models in clinical settings. Method Data evaluated included the gender, demographic and reported pathologic grading and cancer staging. American Urological Association risk levels were used. Poisson regression was used to estimate the population-based and sample specific estimation for prevalence rate and corresponding 95% confidence interval. SAS 9.4 was used for data analysis. Result Compared to PUF, KiTS19 and TCGA KIRC over sampled Caucasian by 9.5% (95% CI, -3.7% to 22.7%) and 15.1% (95% CI, 1.5% to 28.8%), under sampled African American by -6.7% (95% CI, -10% to -3.3%), -5.5% (95% CI, -9.3% to -1.8%). Tertiary also under sampled African American by -6.6% (95% CI, -8.7% to -4.6%). The tertiary cohort largely under sampled aggressive cancers by -14.7% (95% CI, -20.9% to -8.4%). No statistically significant difference was found among PUF, TCGA, and KiTS19 in aggressive rate, however heterogeneities in risk are notable. Conclusion Heterogeneities between cohorts need to be considered in future AI training and cross-validation for renal masses.
RESUMO
OBJECTIVE: To analyze predictors, extent and functional implications associated with renal parenchymal volume replacement (PVR) by renal cell carcinoma (RCC) prior to intervention. This phenomenon is well-recognized yet not adequately studied, and, if severe, can influence management. MATERIALS AND METHODS: A retrospective review was performed of partial nephrectomy (PN) and radical nephrectomy (RN) patients with available preoperative nuclear-renal-scan and imaging demonstrating solitary RCC with normal contralateral kidney. Normal renal parenchymal volume of each kidney was measured by free-hand scripting from preoperative axial images. Primary endpoint was percent PVR which was estimated assuming that the contralateral-kidney serves as a control: PVRâ¯=â¯(volume contralateral kidney - volume ipsilateral kidney) normalized by volume contralateral kidney. Multivariable linear-regression analysis assessed factors associated with preoperative PVR. Further analysis evaluated the functional effect of PVR prior to surgery. RESULTS: 146 PN and 136 RN patients with necessary studies were analyzed. For RN, the median PVR was 15% and a quarter of patients had PVR ≥27%. In contrast, PVR was negligible in PN patients for whom median preoperative parenchymal volumes were nearly identical in the ipsilateral/contralateral kidneys (179/180cc, respectively). PVR inversely correlated with preoperative renal function in the ipsilateral kidney (P <.01). Tumor-size (P <.01), stage (Pâ¯=â¯.03), and endophytic properties (Pâ¯=â¯.03) associated with PVR on multivariable-analysis. CONCLUSION: Our data suggest that substantial replacement of normal parenchyma by RCC occurs in many patients selected for RN and can contribute to preexisting renal-insufficiency. PVR prior to intervention is mainly driven by tumor characteristics in RN patients, but is negligible in most PN patients.
