Deepfake Histologic Images for Enhancing Digital Pathology.

A pathologist's optical microscopic examination of thinly cut, stained tissue on glass slides prepared from a formalin-fixed paraffin-embedded tissue blocks is the gold standard for tissue diagnostics. In addition, the diagnostic abilities and expertise of pathologists is dependent on their direct experience with common and rarer variant morphologies. Recently, deep learning approaches have been used to successfully show a high level of accuracy for such tasks. However, obtaining expert-level annotated images is an expensive and time-consuming task, and artificially synthesized histologic images can prove greatly beneficial. In this study, we present an approach to not only generate histologic images that reproduce the diagnostic morphologic features of common disease but also provide a user ability to generate new and rare morphologies. Our approach involves developing a generative adversarial network model that synthesizes pathology images constrained by class labels. We investigated the ability of this framework in synthesizing realistic prostate and colon tissue images and assessed the utility of these images in augmenting the diagnostic ability of machine learning methods and their usability by a panel of experienced anatomic pathologists. Synthetic data generated by our framework performed similar to real data when training a deep learning model for diagnosis. Pathologists were not able to distinguish between real and synthetic images, and their analyses showed a similar level of interobserver agreement for prostate cancer grading. We extended the approach to significantly more complex images from colon biopsies and showed that the morphology of the complex microenvironment in such tissues can be reproduced. Finally, we present the ability for a user to generate deepfake histologic images using a simple markup of sematic labels.

Biphasic squamoid alveolar renal cell carcinoma: Cytologic features of a rare entity.

Biphasic squamoid alveolar renal cell carcinoma is a newly described rare morphologic variant of papillary renal cell carcinoma. Its characteristic histomorphology and immunophenotype have been well described in the literature. Namely, BSARCC is composed of a dual-cell population with nests of larger squamoid cells surrounded by a single layer of cuboidal cells in alveolar arrangements. Invariably, the squamoid component expresses cyclin D1. More recently, MET alterations have been identified within a subset of BSARCC, raising the possibility for targeted MET inhibitor therapy. To the best of our knowledge the cytomorphologic features of BSARCC have yet to be described. Herein we correlate the cytologic features (percutaneous image-guided fine needle aspiration) of BSARCC to its corresponding histomorphology and immunophenotype (core needle biopsy).

NTRK3-Rearranged Prostatic Acinar Adenocarcinoma: Report of a Patient and Review of the Literature.

Molecular investigations have led to increased therapeutic options for prostatic adenocarcinoma. A single case report of a PRPSAP1::NTRK3 gene fusion occurring in prostate cancer was previously reported. A review of the literature revealed that NTRK gene rearrangements are exceedingly rare molecular events in prostate cancer. NTRK gene fusions can be oncogenic drivers or develop as resistance mechanisms. The tumor-agnostic approvals of TRK inhibitors by the FDA provide additional rationale for molecular investigations of aggressive prostatic adenocarcinomas. This may prove to be an additional therapeutic option for patients with aggressive prostatic carcinomas refractory to initial therapy. We report a case of an aggressive castrate-resistant prostatic adenocarcinoma with a BMP6::NTRK3 gene fusion.

Genomic attributes of prostate cancer across primary and metastatic noncastrate and castrate resistant disease states: a next generation sequencing study of 183 patients.

Primary prostatic adenocarcinoma (pPC) undergoes genomic evolution secondary to therapy-related selection pressures as it transitions to metastatic noncastrate (mNC-PC) and castrate resistant (mCR-PC) disease. Next generation sequencing results were evaluated for pPC (n = 97), locally advanced disease (involving urinary bladder/rectum, n = 12), mNC-PC (n = 21), and mCR-PC (n = 54). We identified enrichment of TP53 alterations in high-grade pPC, TP53/RB1 alterations in HGNE disease, and AR alterations in metastatic and castrate resistant disease. Actionable alterations (MSI-H phenotype and HRR genes) were identified in approximately a fifth of all cases. These results help elucidate the landscape of genomic alterations across the clinical spectrum of prostate cancer.

Accelerating Cancer Histopathology Workflows with Chemical Imaging and Machine Learning.

Histopathology has remained a cornerstone for biomedical tissue assessment for over a century, with a resource-intensive workflow involving biopsy or excision, gross examination, sampling, tissue processing to snap frozen or formalin-fixed paraffin-embedded blocks, sectioning, staining, optical imaging, and microscopic assessment. Emerging chemical imaging approaches, including stimulated Raman scattering (SRS) microscopy, can directly measure inherent molecular composition in tissue (thereby dispensing with the need for tissue processing, sectioning, and using dyes) and can use artificial intelligence (AI) algorithms to provide high-quality images. Here we show the integration of SRS microscopy in a pathology workflow to rapidly record chemical information from minimally processed fresh-frozen prostate tissue. Instead of using thin sections, we record data from intact thick tissues and use optical sectioning to generate images from multiple planes. We use a deep learning­based processing pipeline to generate virtual hematoxylin and eosin images. Next, we extend the computational method to generate archival-quality images in minutes, which are equivalent to those obtained from hours/days-long formalin-fixed, paraffin-embedded processing. We assessed the quality of images from the perspective of enabling pathologists to make decisions, demonstrating that the virtual stained image quality was diagnostically useful and the interpathologist agreement on prostate cancer grade was not impacted. Finally, because this method does not wash away lipids and small molecules, we assessed the utility of lipid chemical composition in determining grade. Together, the combination of chemical imaging and AI provides novel capabilities for rapid assessments in pathology by reducing the complexity and burden of current workflows. SIGNIFICANCE: Archival-quality (formalin-fixed paraffin-embedded), thin-section diagnostic images are obtained from thick-cut, fresh-frozen prostate tissues without dyes or stains to expedite cancer histopathology by combining SRS microscopy and machine learning.

Molecular and Immunophenotypic Correlates of Metastatic Epithelioid Angiomyolipoma Include Alterations of TP53, RB1, and ATRX.

CONTEXT.­: Epithelioid angiomyolipomas (eAMLs) are rare tumors of the kidney that occur in patients with tuberous sclerosis complex or in a sporadic setting; a subset of these tumors exhibit metastatic behavior. OBJECTIVE.­: To analyze molecular profiling data to identify pathogenic alterations in rare cases of metastatic eAML, and to identify immunohistochemistry (IHC)-based surrogate markers. DESIGN.­: Molecular profiling data from the American Association for Cancer Research GENIE registry was accessed for 23 patients with angiomyolipomas, and 9 of 16 patients with eAMLs in our institutional registry were evaluated with next-generation sequencing. IHC was performed to screen for alterations of P53, RB, and ATRX for all 16 institutional cases. RESULTS.­: Combined alterations of 5 tumor-suppressor genes (TP53, ATRX, RB1, APC, and NF1) were identified using next-generation sequencing in 7 of 8 (88%) patients with metastatic disease compared to a single patient with nonmetastatic disease (RB1 variant of uncertain significance; 1 of 24, 4%). No cases with abnormal IHC results were identified in 11 patients with nonmetastatic disease compared to 3 of 5 patients with metastatic disease. CONCLUSIONS.­: Our results show that the majority of metastatic eAMLs have mutations of 5 tumor-suppressor genes (TP53, ATRX, RB1, APC, and NF1), while these are rare in patients with nonmetastatic disease. Furthermore, IHC for P53, RB, and ATRX may serve as a screen for a subset of these alterations in resource-limited settings. These findings, if validated in larger data sets, have the potential to predict metastatic behavior in eAMLs.

Renal Neoplasia in Polycystic Kidney Disease: An Assessment of Tuberous Sclerosis Complex-associated Renal Neoplasia and PKD1/TSC2 Contiguous Gene Deletion Syndrome.

To determine the incidence of renal neoplasia among patients undergoing nephrectomy for polycystic kidney disease (PKD), we queried our institutional nephrectomy registry (years 2000-2020). Approximately 4% (231 of 5757) of patients who underwent nephrectomy had PKD, and 26 of these 231 patients (11.3%) had renal neoplasia. Tumors from an additional two patients with PKD were also evaluated. Patients with PKD who had tuberous sclerosis complex (TSC)-associated renal neoplasia were screened for PKD1/TSC2 contiguous gene deletion syndrome (CGS) using single nucleotide polymorphism arrays. The median age of patients with PKD and renal neoplasia at nephrectomy was 54 yr. The median tumor size was 2.0 cm and the tumors were predominantly of low grade and stage. The tumors consisted of 23 renal cell carcinomas (RCCs), one epithelioid angiomyolipoma, and four angiomyolipomas. The median follow-up was 59.5 mo (n = 26) and only one patient with clear cell RCC developed metastases. Two patients with angiomyolipomas had PKD1/TSC2 CGS. Our results support screening of patients with PKD and TSC-associated renal neoplasia as well as TSC patients with cystic renal disease for CGS, as identification of patients with CGS can better define the manifestation and prognosis of CGS and guide counseling regarding patterns of inheritance. PATIENT SUMMARY: We identified patients with abnormal kidney cell growth (called renal neoplasia) among those undergoing removal of kidney tissue for polycystic kidney disease (PKD) and patients with a syndrome involving deletions in two genes, called PKD1/TSC2 contiguous gene deletion syndrome (CGS) at our institution. Of 231 PKD patients with removal of kidney tissue, 11.3% had renal neoplasia, and two patients with angiomyolipoma tumors had PKD1/TSC2 CGS. Detection of renal neoplasia associated with a condition called tuberous sclerosis complex in PKD may increase the identification of patients with PKD1/TSC2 CGS and guide patient counseling regarding outcomes and patterns of inheritance.

Incidentally Detected Celiac Disease with Splenomegaly on 18F FDG PET/CT: A Potential Lymphoma Mimic.

Celiac disease is an immune-mediated disorder triggered by hypersensitivity to gluten occurring in genetically susceptible individuals. A high-index of suspicion is needed for diagnosis as patients can be asymptomatic or present with atypical symptoms or extra-intestinal manifestations. Typical 18F-Fluorodeoxyglucose (FDG) Positron Emission Tomography (PET)/Computed Tomography (CT) gastrointestinal manifestations of celiac disease include increased multifocal or diffuse jejunal and ileal uptake; focal duodenal uptake is less common. Splenomegaly with increased splenic FDG uptake is also uncommon in celiac disease in the absence of portal hypertension; small-sized spleen and functional hyposplenism are more typical. We report a case of celiac disease diagnosed after PET/CT showed FDG uptake in the duodenum and enlarged spleen. Follow-up after gluten-free diet showed complete metabolic resolution and regression of splenomegaly. The combination of focal bowel and splenic uptake is unusual in celiac disease and may be mistaken for a lymphoproliferative disorder. Awareness of this entity may avoid misdiagnosis and guide appropriate management.