Deep learning has proven capable of automating key aspects of histopathologic analysis. However, its context-specific nature and continued reliance on large expert-annotated training datasets hinders the development of a critical mass of applications to garner widespread adoption in clinical/research workflows. Here, we present an online collaborative platform that streamlines tissue image annotation to promote the development and sharing of custom computer vision models for PHenotyping And Regional Analysis Of Histology (PHARAOH; https://www.pathologyreports.ai/ ). Specifically, PHARAOH uses a weakly supervised, human-in-the-loop learning framework whereby patch-level image features are leveraged to organize large swaths of tissue into morphologically-uniform clusters for batched annotation by human experts. By providing cluster-level labels on only a handful of cases, we show how custom PHARAOH models can be developed efficiently and used to guide the quantification of cellular features that correlate with molecular, pathologic and patient outcome data. Moreover, by using our PHARAOH pipeline, we showcase how correlation of cohort-level cytoarchitectural features with accompanying biological and outcome data can help systematically devise interpretable morphometric models of disease. Both the custom model design and feature extraction pipelines are amenable to crowdsourcing, positioning PHARAOH to become a fully scalable, systems-level solution for the expansion, generalization and cataloging of computational pathology applications. Faust, Chen, and colleagues present PHARAOH, a collaborative computational pathology platform that allows histologists to quickly develop custom labelled image datasets to train and catalogue a variety of machine learning models for histopathological analysis.

Papillary renal neoplasm with reverse polarity (PRNRP) is a recently described indolent entity with distinct features and its recognition from other oncocytic/eosinophilic papillary renal cell carcinoma (ePRCC) has important prognostic implications. ABCC2, a renal drug transporter, is overexpressed in aggressive PRCCs. In this study, we compared the clinicopathological parameters and the biological ABCC2 expression between PRNRP and ePRCC. PRNRP (n = 8) and ePRCC (n = 21) cases were selected from resection specimens and corresponding clinicopathological data were collected. ABCC2 immunohistochemical (IHC) staining was performed and ABCC2 staining patterns were classified as negative, cytoplasmic, and brush-border. RNA in-situ hybridization (ISH) was used to assess ABCC2 transcript levels. All eight PRNRP cases had weak cytoplasmic ABCC2 IHC reactivity; however, they showed no detectable ABCC2 transcripts on RNA ISH. In comparison, 76% (16/21) of ePRCCs showed ABCC2 IHC brush-border expression and significantly higher ABCC2 RNA ISH transcript levels (p < 0.001). Additionally, the ePRCC group showed a significantly larger tumor size (p = 0.004), higher WHO/ISUP grade (p < 0.001), and stage (p = 0.044). None of the PRNRP cases showed disease progression, while 9.5% (2/21) ePRCCs had disease progression. PRNRP is clinically and biologically distinct from ePRCC. Hence, it is crucial to differentiate between these two entities, particularly in needle core biopsies.

Historically, papillary renal cell carcinoma (PRCC) was divided into two types, type 1 and type 2, based solely on morphology. However, it is apparent that PRCC is far more complex and represents a histological, clinical, and molecular spectrum. There has been a significant evolution in our understanding of PRCC, highlighted by the recognition of new and molecularly defined entities that were previously included in PRCC type 2. This contemporary review addresses the evolving concepts regarding the PRCC, including why it is no longer needed to subtype PRCC, the current molecular landscape, prognostic parameters, and PRCC variants, including biphasic PRCC, papillary renal neoplasm with reverse polarity, and Warthin-like PRCC. Pathologists should also be aware of the potential mimickers of both low-grade and high-grade PRCCs as well as some new and emerging entities that may show papillary growth that should be excluded in the diagnostic workup. The evolving knowledge of PRCC biomarkers, morphologic patterns, and PRCC variants could also have important implications for clinical management. Lastly, the heterogeneity within the PRCC spectrum needs to be further studied, aiming to better stratify PRCC for appropriate clinical management and systemic therapy.

AIMS:Papillary renal cell carcinoma (PRCC) histologic subtyping is no longer recommended in the 2022 WHO classification. Currently, WHO/ISUP nucleolar grade is the only accepted prognostic histologic parameter for PRCC. ABCC2, a renal drug transporter, has been shown to significantly predict outcomes in PRCC. In this study we evaluated the prognostic significance of ABCC2 IHC staining patterns in a large, multi-institutional PRCC cohort and assessed the association of these patterns with ABCC2 mRNA expression. METHODS AND RESULTS:We assessed 254 PRCCs for ABCC2 IHC reactivity patterns that were stratified into negative, cytoplasmic, brush-border <50%, and brush-border ≥50%. RNA in situ hybridization (ISH) was used to determine the transcript level of each group. Survival analysis was performed with SPSS and GraphPad software. RNA-ISH showed that the ABCC2 group with any brush-border staining was associated with a significant increase in the transcript level, when compared to the negative/cytoplasmic group (P = 0.034). Both ABCC2 groups with brush-border <50% (P = 0.024) and brush-border ≥50% (P < 0.001) were also associated with worse disease-free survival (DFS) in univariate analysis. Multivariate analysis showed that only ABCC2 IHC brush-border (<50% and ≥50%) reactivity groups (P = 0.037 and P = 0.003, respectively), and high-stage disease (P < 0.001) had a DFS of prognostic significance. In addition, ABCC2 brush-border showed significantly worse DFS in pT1a (P = 0.014), pT1 (P = 0.013), ≤4 cm tumour (P = 0.041) and high stage (P = 0.014) groups, while a similar analysis with high WHO/ISUP grade in these groups was not significant. CONCLUSION:ABCC2 IHC brush-border expression in PRCC correlates with significantly higher gene expression and also independently predicts survival outcomes.