Development and validation of a vascularity-based architectural classification for clear cell renal cell carcinoma: correlation with conventional pathological prognostic factors, gene expression patterns, and clinical outcomes

Abstract

The prognostic significance of an architectural grading system for clear cell renal cell carcinoma (ccRCC) has recently been demonstrated. The present study aimed to establish a vascularity-based architectural classification using the cohort of 436 patients with localized ccRCC who underwent extirpative surgery and correlated the findings with conventional pathologic factors, gene expression, and prognosis. First, we assessed architectural patterns in the highest-grade area on hematoxylin and eosin-stained slides, then separately evaluated our surrogate score for vascularity. We grouped nine architectural patterns into three categories based on the vascular network score. "Vascularity-based architectural classification" was defined: category 1: characterized by enrichment of the vascular network, including compact/small nested, macrocyst/microcystic, and tubular/acinar patterns; category 2: characterized by a widely spaced-out vascular network, including alveolar/large nested, thick trabecular/insular, papillary/pseudopapillary patterns; category 3: characterized by scattered vascularity without a vascular network, including solid sheets, rhabdoid and sarcomatoid patterns. Adverse pathological prognostic factors such as TNM stage, WHO/ISUP grade, and necrosis were significantly associated with category 3, followed by category 2 (all p < 0.001). We successfully validated the classification using The Cancer Genome Atlas (TCGA) cohort (n = 162), and RNA-sequencing data available from TCGA showed that the angiogenesis gene signature was significantly enriched in category 1 compared to categories 2 and 3, whereas the immune gene signature was significantly enriched in category 3 compared to categories 1 and 2. In univariate analysis, vascularity-based architectural classification showed the best accuracy in pathological prognostic factors for predicting recurrence-free survival (c-index = 0.786). The predictive accuracy of our model which integrated WHO/ISUP grade, necrosis, TNM stage, and vascularity-based architectural classification was greater than conventional risk models (c-index = 0.871 vs. 0.755-0.843). Our findings suggest that the vascularity-based architectural classification is prognostically useful and may help stratify patients appropriately for management based on their likelihood of post-surgical recurrence.