RAPHIA: A deep learning pipeline for the registration of MRI and whole-mount histopathology images of the prostate.

Image registration can map the ground truth extent of prostate cancer from histopathology images onto MRI, facilitating the development of machine learning methods for early prostate cancer detection. Here, we present RAdiology PatHology Image Alignment (RAPHIA), an end-to-end pipeline for efficient and accurate registration of MRI and histopathology images. RAPHIA automates several time-consuming manual steps in existing approaches including prostate segmentation, estimation of the rotation angle and horizontal flipping in histopathology images, and estimation of MRI-histopathology slice correspondences. By utilizing deep learning registration networks, RAPHIA substantially reduces computational time. Furthermore, RAPHIA obviates the need for a multimodal image similarity metric by transferring histopathology image representations to MRI image representations and vice versa. With the assistance of RAPHIA, novice users achieved expert-level performance, and their mean error in estimating histopathology rotation angle was reduced by 51% (12 degrees vs 8 degrees), their mean accuracy of estimating histopathology flipping was increased by 5% (95.3% vs 100%), and their mean error in estimating MRI-histopathology slice correspondences was reduced by 45% (1.12 slices vs 0.62 slices). When compared to a recent conventional registration approach and a deep learning registration approach, RAPHIA achieved better mapping of histopathology cancer labels, with an improved mean Dice coefficient of cancer regions outlined on MRI and the deformed histopathology (0.44 vs 0.48 vs 0.50), and a reduced mean per-case processing time (51 vs 11 vs 4.5 min). The improved performance by RAPHIA allows efficient processing of large datasets for the development of machine learning models for prostate cancer detection on MRI. Our code is publicly available at: https://github.com/pimed/RAPHIA.

UCHL1 is a potential molecular indicator and therapeutic target for neuroendocrine carcinomas.

Neuroendocrine carcinomas, such as neuroendocrine prostate cancer and small-cell lung cancer, commonly have a poor prognosis and limited therapeutic options. We report that ubiquitin carboxy-terminal hydrolase L1 (UCHL1), a deubiquitinating enzyme, is elevated in tissues and plasma from patients with neuroendocrine carcinomas. Loss of UCHL1 decreases tumor growth and inhibits metastasis of these malignancies. UCHL1 maintains neuroendocrine differentiation and promotes cancer progression by regulating nucleoporin, POM121, and p53. UCHL1 binds, deubiquitinates, and stabilizes POM121 to regulate POM121-associated nuclear transport of E2F1 and c-MYC. Treatment with the UCHL1 inhibitor LDN-57444 slows tumor growth and metastasis across neuroendocrine carcinomas. The combination of UCHL1 inhibitors with cisplatin, the standard of care used for neuroendocrine carcinomas, significantly delays tumor growth in pre-clinical settings. Our study reveals mechanisms of UCHL1 function in regulating the progression of neuroendocrine carcinomas and identifies UCHL1 as a therapeutic target and potential molecular indicator for diagnosing and monitoring treatment responses in these malignancies.

High expression of Trop2 is associated with aggressive localized prostate cancer and is a candidate urinary biomarker.

Distinguishing indolent from clinically significant localized prostate cancer is a major clinical challenge and influences clinical decision-making between treatment and active surveillance. The development of novel predictive biomarkers will help with risk stratification, and clinical decision-making, leading to a decrease in over or under-treatment of patients with prostate cancer. Here, we report that Trop2 is a prognostic tissue biomarker for clinically significant prostate cancer by utilizing the Canary Prostate Cancer Tissue Microarray (CPCTA) cohort composed of over 1100 patients from a multi-institutional study. We demonstrate that elevated Trop2 expression is correlated with worse clinical features including Gleason score, age, and pre-operative PSA levels. More importantly, we demonstrate that elevated Trop2 expression at radical prostatectomy predicts worse overall survival in men undergoing radical prostatectomy. Additionally, we detect shed Trop2 in urine from men with clinically significant prostate cancer. Our study identifies Trop2 as a novel tissue prognostic biomarker and a candidate non-invasive marker for prostate cancer.

Diagnostic impact of RNA-based next-generation sequencing fusion panel for solid tumors: A single-institution experience.

OBJECTIVES: Gene rearrangements frequently act as oncogenic driver mutations and determine the onset and progression of cancer. RNA-based next-generation sequencing (NGS) is being used with increasing frequency for solid tumors. The purpose of our study is to investigate the feasibility and utility of an RNA-based NGS fusion panel for solid tumors. METHODS: We conducted a retrospective, single-institution review of fusion panels requested between May 2022 and March 2023. Demographic, clinical, pathologic, and molecular findings of the patients were reviewed. The utility of the RNA-based NGS fusion panel for the pathologic diagnosis of solid tumors was assessed. RESULTS: Our study included 345 cases, and a fusion event was identified in 24.3% (78/321) of cases. Among the 110 cases submitted for diagnostic purposes, a fusion event was detected in 42.7% (47/110) of cases. The results led to refinement or clarification of the initial diagnosis in 31.9% (15/47) of cases and agreement or support for the initial diagnosis in 59.6% (28/47) of cases. Furthermore, our study indicated that the overall cellularity (tumor and normal tissue) of the tested specimen influences the success of the testing process. CONCLUSIONS: In summary, this study demonstrated the feasibility and utility of an RNA-based NGS fusion panel for a wide variety of solid tumors in the appropriate clinicopathologic context. These findings warrant further validation in larger studies involving multiple institutional patient cohorts.

Spatial transcriptomics identifies candidate stromal drivers of benign prostatic hyperplasia.

Benign prostatic hyperplasia (BPH) is the nodular proliferation of the prostate transition zone in older men, leading to urinary storage and voiding problems that can be recalcitrant to therapy. Decades ago, John McNeal proposed that BPH originates with the "reawakening" of embryonic inductive activity by adult prostate stroma, which spurs new ductal proliferation and branching morphogenesis. Here, by laser microdissection and transcriptional profiling of the BPH stroma adjacent to hyperplastic branching ducts, we identified secreted factors likely mediating stromal induction of prostate glandular epithelium and coinciding processes. The top stromal factors were insulin-like growth factor 1 (IGF1) and CXC chemokine ligand 13 (CXCL13), which we verified by RNA in situ hybridization to be coexpressed in BPH fibroblasts, along with their cognate receptors (IGF1R and CXCR5) on adjacent epithelium. In contrast, IGF1 but not CXCL13 was expressed in human embryonic prostate stroma. Finally, we demonstrated that IGF1 is necessary for the generation of BPH-1 cell spheroids and patient-derived BPH cell organoids in 3D culture. Our findings partially support historic speculations on the etiology of BPH and provide what we believe to be new molecular targets for rational therapies directed against the underlying processes driving BPH.

Low-grade fibromyxoid sarcoma of the breast: genetic characterization and immunohistochemical comparison to morphologic mimics.

Spindle cell lesions of the breast elicit a specific, relatively limited differential diagnosis, and accurate classification often requires careful morphologic evaluation and immunohistochemical workup. Low-grade fibromyxoid sarcoma (LGFMS) is a rare malignant fibroblastic tumor with deceptively bland spindle cell morphology. Involvement of the breast is exceedingly rare. We examined the clinicopathologic and molecular characteristics of three cases of breast/axillary LGFMS. In addition, we interrogated the immunohistochemical expression of MUC4, a commonly used marker of LGFMS, in other breast spindle cell lesions. LGFMS presented in women at 23, 33, and 59 years of age. Tumor size ranged from 0.9 to 4.7 cm. Microscopically, they were circumscribed nodular masses composed of bland spindle cells with fibromyxoid stroma. Immunohistochemically, tumors were diffusely positive for MUC4 and negative for keratin, CD34, S100 protein, and nuclear beta-catenin. Fluorescence in-situ hybridization demonstrated FUS (n = 2) or EWSR1 (n = 1) rearrangements. Next-generation sequencing identified FUS::CREB3L2 and EWSR1::CREB3L1 fusions. MUC4 immunohistochemistry performed on 162 additional breast lesions demonstrated only weak and limited expression in a subset of cases of fibromatosis (10/20, ≤30% staining), scar (5/9, ≤10%), metaplastic carcinoma (4/23, ≤5%), and phyllodes tumor (3/74, ≤10%). MUC4 was entirely negative in cases of pseudoangiomatous stromal hyperplasia (n = 9), myofibroblastoma (n = 6), periductal stromal tumor (n = 3), and cellular/juvenile fibroadenoma (n = 21). LGFMS can rarely occur in the breast and should be considered in the differential diagnosis of breast spindle cell lesions. Strong and diffuse MUC4 expression is highly specific in this histologic context. Detection of an FUS or EWSR1 rearrangement can confirm the diagnosis.

A review of artificial intelligence in prostate cancer detection on imaging.

A multitude of studies have explored the role of artificial intelligence (AI) in providing diagnostic support to radiologists, pathologists, and urologists in prostate cancer detection, risk-stratification, and management. This review provides a comprehensive overview of relevant literature regarding the use of AI models in (1) detecting prostate cancer on radiology images (magnetic resonance and ultrasound imaging), (2) detecting prostate cancer on histopathology images of prostate biopsy tissue, and (3) assisting in supporting tasks for prostate cancer detection (prostate gland segmentation, MRI-histopathology registration, MRI-ultrasound registration). We discuss both the potential of these AI models to assist in the clinical workflow of prostate cancer diagnosis, as well as the current limitations including variability in training data sets, algorithms, and evaluation criteria. We also discuss ongoing challenges and what is needed to bridge the gap between academic research on AI for prostate cancer and commercial solutions that improve routine clinical care.

Computational Detection of Extraprostatic Extension of Prostate Cancer on Multiparametric MRI Using Deep Learning.

The localization of extraprostatic extension (EPE), i.e., local spread of prostate cancer beyond the prostate capsular boundary, is important for risk stratification and surgical planning. However, the sensitivity of EPE detection by radiologists on MRI is low (57% on average). In this paper, we propose a method for computational detection of EPE on multiparametric MRI using deep learning. Ground truth labels of cancers and EPE were obtained in 123 patients (38 with EPE) by registering pre-surgical MRI with whole-mount digital histopathology images from radical prostatectomy. Our approach has two stages. First, we trained deep learning models using the MRI as input to generate cancer probability maps both inside and outside the prostate. Second, we built an image post-processing pipeline that generates predictions for EPE location based on the cancer probability maps and clinical knowledge. We used five-fold cross-validation to train our approach using data from 74 patients and tested it using data from an independent set of 49 patients. We compared two deep learning models for cancer detection: (i) UNet and (ii) the Correlated Signature Network for Indolent and Aggressive prostate cancer detection (CorrSigNIA). The best end-to-end model for EPE detection, which we call EPENet, was based on the CorrSigNIA cancer detection model. EPENet was successful at detecting cancers with extraprostatic extension, achieving a mean area under the receiver operator characteristic curve of 0.72 at the patient-level. On the test set, EPENet had 80.0% sensitivity and 28.2% specificity at the patient-level compared to 50.0% sensitivity and 76.9% specificity for the radiologists. To account for spatial location of predictions during evaluation, we also computed results at the sextant-level, where the prostate was divided into sextants according to standard systematic 12-core biopsy procedure. At the sextant-level, EPENet achieved mean sensitivity 61.1% and mean specificity 58.3%. Our approach has the potential to provide the location of extraprostatic extension using MRI alone, thus serving as an independent diagnostic aid to radiologists and facilitating treatment planning.

GRM1 Immunohistochemistry Distinguishes Chondromyxoid Fibroma From its Histologic Mimics.

Chondromyxoid fibroma (CMF) is a rare benign bone neoplasm that manifests histologically as a lobular proliferation of stellate to spindle-shaped cells in a myxoid background, exhibiting morphologic overlap with other cartilaginous and myxoid tumors of bone. CMF is characterized by recurrent genetic rearrangements that place the glutamate receptor gene GRM1 under the regulatory control of a constitutively active promoter, leading to increased gene expression. Here, we explore the diagnostic utility of GRM1 immunohistochemistry as a surrogate marker for GRM1 rearrangement using a commercially available monoclonal antibody in a study of 230 tumors, including 30 CMF cases represented by 35 specimens. GRM1 was positive by immunohistochemistry in 97% of CMF specimens (34/35), exhibiting moderate to strong staining in more than 50% of neoplastic cells; staining was diffuse (>95% of cells) in 25 specimens (71%). Among the 9 CMF specimens with documented exposure to acid decalcification, 4 (44%) exhibited diffuse immunoreactivity (>95%) for GRM1, whereas all 15 CMF specimens (100%) with lack of exposure to decalcification reagents were diffusely immunoreactive ( P =0.003). High GRM1 expression at the RNA level was previously observed by quantitative reverse transcription polymerase chain reaction in 9 CMF cases that were also positive by immunohistochemistry; low GRM1 expression was observed by quantitative reverse transcription polymerase chain reaction in the single case of CMF that was negative by immunohistochemistry. GRM1 immunohistochemistry was negative (<5%) in histologic mimics of CMF, including conventional chondrosarcoma, enchondroma, chondroblastoma, clear cell chondrosarcoma, giant cell tumor of the bone, fibrous dysplasia, chondroblastic osteosarcoma, myoepithelial tumor, primary aneurysmal bone cyst, brown tumor, phosphaturic mesenchymal tumor, CMF-like osteosarcoma, and extraskeletal myxoid chondrosarcoma. These results indicate that GRM1 immunohistochemistry may have utility in distinguishing CMF from its histologic mimics.

Immunofluorescent and molecular characterization of effusion tumor cells reveal cancer site-of-origin and disease-driving mutations.

BACKGROUND: Effective cancer treatment relies on precision diagnostics. In cytology, an accurate diagnosis facilitates the determination of proper therapeutics for patients with cancer. Previously, the authors developed a multiplexed immunofluorescent panel to detect epithelial malignancies from pleural effusion specimens. Their assay reliably distinguished effusion tumor cells (ETCs) from nonmalignant cells; however, it lacked the capacity to reveal specific cancer origin information. Furthermore, DNA profiling of ETCs revealed some, but not all, cancer-driver mutations. METHODS: The authors developed a new multiplex immunofluorescent panel that detected both malignancy and pulmonary origin by incorporating the thyroid transcription factor-1 (TTF-1) biomarker. Evaluation for TTF-1-positive ETCs (T-ETCs) was performed on 12 patient samples. T-ETCs and parallel ETCs from selected patients were collected and subjected to DNA profiling to identify pathogenic mutations. All samples were obtained with Institutional Review Board approval. RESULTS: Malignancy was detected in all samples. T-ETCs were identified in 9 of 10 patients who had clinically reported TTF-1 positivity (90% sensitivity and 100% specificity). Furthermore, DNA profiling of as few as five T-ETCs identified pathogenic mutations with equal or greater sensitivity compared with profiling of ETCs, both of which showed high concordance with clinical findings. CONCLUSIONS: The findings suggest that the immunofluorescent and molecular characterization of tumor cells from pleural effusion specimens can provide reliable diagnostic information, even with very few cells. The integration of site-specific biomarkers like TTF-1 into ETC analysis may facilitate better refined diagnosis and improve patient care.

The genetic landscape of germline DDX41 variants predisposing to myeloid neoplasms.

Germline DDX41 variants are the most common mutations predisposing to acute myeloid leukemia (AML)/myelodysplastic syndrome (MDS) in adults, but the causal variant (CV) landscape and clinical spectrum of hematologic malignancies (HMs) remain unexplored. Here, we analyzed the genomic profiles of 176 patients with HM carrying 82 distinct presumably germline DDX41 variants among a group of 9821 unrelated patients. Using our proposed DDX41-specific variant classification, we identified features distinguishing 116 patients with HM with CV from 60 patients with HM with variant of uncertain significance (VUS): an older age (median 69 years), male predominance (74% in CV vs 60% in VUS, P = .03), frequent concurrent somatic DDX41 variants (79% in CV vs 5% in VUS, P < .0001), a lower somatic mutation burden (1.4 ± 0.1 in CV vs 2.9 ± 0.04 in VUS, P = .012), near exclusion of canonical recurrent genetic abnormalities including mutations in NPM1, CEBPA, and FLT3 in AML, and favorable overall survival (OS) in patients with AML/MDS. This superior OS was determined independent of blast count, abnormal karyotypes, and concurrent variants, including TP53 in patients with AML/MDS, regardless of patient's sex, age, or specific germline CV, suggesting that germline DDX41 variants define a distinct clinical entity. Furthermore, unrelated patients with myeloproliferative neoplasm and B-cell lymphoma were linked by DDX41 CV, thus expanding the known disease spectrum. This study outlines the CV landscape, expands the phenotypic spectrum in unrelated DDX41-mutated patients, and underscores the urgent need for gene-specific diagnostic and clinical management guidelines.

Bridging the gap between prostate radiology and pathology through machine learning.

BACKGROUND: Prostate cancer remains the second deadliest cancer for American men despite clinical advancements. Currently, magnetic resonance imaging (MRI) is considered the most sensitive non-invasive imaging modality that enables visualization, detection, and localization of prostate cancer, and is increasingly used to guide targeted biopsies for prostate cancer diagnosis. However, its utility remains limited due to high rates of false positives and false negatives as well as low inter-reader agreements. PURPOSE: Machine learning methods to detect and localize cancer on prostate MRI can help standardize radiologist interpretations. However, existing machine learning methods vary not only in model architecture, but also in the ground truth labeling strategies used for model training. We compare different labeling strategies and the effects they have on the performance of different machine learning models for prostate cancer detection on MRI. METHODS: Four different deep learning models (SPCNet, U-Net, branched U-Net, and DeepLabv3+) were trained to detect prostate cancer on MRI using 75 patients with radical prostatectomy, and evaluated using 40 patients with radical prostatectomy and 275 patients with targeted biopsy. Each deep learning model was trained with four different label types: pathology-confirmed radiologist labels, pathologist labels on whole-mount histopathology images, and lesion-level and pixel-level digital pathologist labels (previously validated deep learning algorithm on histopathology images to predict pixel-level Gleason patterns) on whole-mount histopathology images. The pathologist and digital pathologist labels (collectively referred to as pathology labels) were mapped onto pre-operative MRI using an automated MRI-histopathology registration platform. RESULTS: Radiologist labels missed cancers (ROC-AUC: 0.75-0.84), had lower lesion volumes (~68% of pathology lesions), and lower Dice overlaps (0.24-0.28) when compared with pathology labels. Consequently, machine learning models trained with radiologist labels also showed inferior performance compared to models trained with pathology labels. Digital pathologist labels showed high concordance with pathologist labels of cancer (lesion ROC-AUC: 0.97-1, lesion Dice: 0.75-0.93). Machine learning models trained with digital pathologist labels had the highest lesion detection rates in the radical prostatectomy cohort (aggressive lesion ROC-AUC: 0.91-0.94), and had generalizable and comparable performance to pathologist label-trained-models in the targeted biopsy cohort (aggressive lesion ROC-AUC: 0.87-0.88), irrespective of the deep learning architecture. Moreover, machine learning models trained with pixel-level digital pathologist labels were able to selectively identify aggressive and indolent cancer components in mixed lesions on MRI, which is not possible with any human-annotated label type. CONCLUSIONS: Machine learning models for prostate MRI interpretation that are trained with digital pathologist labels showed higher or comparable performance with pathologist label-trained models in both radical prostatectomy and targeted biopsy cohort. Digital pathologist labels can reduce challenges associated with human annotations, including labor, time, inter- and intra-reader variability, and can help bridge the gap between prostate radiology and pathology by enabling the training of reliable machine learning models to detect and localize prostate cancer on MRI.

Combinatorial Inactivation of Tumor Suppressors Efficiently Initiates Lung Adenocarcinoma with Therapeutic Vulnerabilities.

Lung cancer is the leading cause of cancer death worldwide, with lung adenocarcinoma being the most common subtype. Many oncogenes and tumor suppressor genes are altered in this cancer type, and the discovery of oncogene mutations has led to the development of targeted therapies that have improved clinical outcomes. However, a large fraction of lung adenocarcinomas lacks mutations in known oncogenes, and the genesis and treatment of these oncogene-negative tumors remain enigmatic. Here, we perform iterative in vivo functional screens using quantitative autochthonous mouse model systems to uncover the genetic and biochemical changes that enable efficient lung tumor initiation in the absence of oncogene alterations. Generation of hundreds of diverse combinations of tumor suppressor alterations demonstrates that inactivation of suppressors of the RAS and PI3K pathways drives the development of oncogene-negative lung adenocarcinoma. Human genomic data and histology identified RAS/MAPK and PI3K pathway activation as a common feature of an event in oncogene-negative human lung adenocarcinomas. These Onc-negativeRAS/PI3K tumors and related cell lines are vulnerable to pharmacologic inhibition of these signaling axes. These results transform our understanding of this prevalent yet understudied subtype of lung adenocarcinoma. SIGNIFICANCE: To address the large fraction of lung adenocarcinomas lacking mutations in proto-oncogenes for which targeted therapies are unavailable, this work uncovers driver pathways of oncogene-negative lung adenocarcinomas and demonstrates their therapeutic vulnerabilities.

Spitz nevus with EHBP1-ALK fusion and distinctive membranous localization of ALK.

ALK rearrangements define a histopathologically distinctive yet diverse subset of Spitz tumors characterized by fusiform to epithelioid melanocytes with frequent fascicular growth and ALK overexpression. Molecularly, these tumors are characterized by fusions between ALK and a variety of gene partners, most commonly TPM3 and DCTN1. We describe an unusual case of a Spitz nevus occurring in a 13-year-old female that manifested ALK immunopositivity with cell membrane localization. The proliferation was polypoid and composed of elongated nests of epithelioid melanocytes with enlarged nuclei, prominent nucleoli, and abundant cytoplasm without significant atypia and lacking mitotic figures. The nevus exhibited strong and diffuse expression of p16. Targeted next-generation RNA sequencing revealed an in-frame EHBP1-ALK fusion, which has been reported only once in the literature. EHBP1 encodes an adaptor protein with plasma membrane targeting potential. Together, these findings suggest that the 5' ALK fusion partner in Spitz tumors may dictate the subcellular localization of the ALK chimeric oncoprotein. In summary, this case highlights a rare ALK fusion associated with a distinct immunohistochemical staining pattern and further expands the spectrum of ALK-rearranged melanocytic tumors.

Well-differentiated lipomatous neoplasms with p53 alterations: a clinicopathological and molecular study of eight cases with features of atypical pleomorphic lipomatous tumour.

AIMS: Well-differentiated lipomatous neoplasms encompass a broad spectrum of benign and malignant tumours, many of which are characterised by recurrent genetic abnormalities. Although a key regulator of p53 signalling, MDM2, is characteristically amplified in well-differentiated liposarcoma, recurrent abnormalities of p53 itself have not been reported in well-differentiated adipocytic neoplasms. Here, we present a series of well-differentiated lipomatous tumours characterised by p53 alterations and histological features in keeping with atypical pleomorphic lipomatous tumour (APLT). METHODS AND RESULTS: We reviewed the morphological, immunohistochemical and molecular genetic features of eight lipomatous tumours with p53 alterations. Four tumours arose in the thigh, and one case each arose in the shoulder, calf, upper back, and subclavicular regions; six tumours were deep/subfascial and two were subcutaneous. Relevant clinical history included two patients with Li-Fraumeni syndrome. Morphologically, all cases showed well-differentiated adipocytes with prominent nuclear pleomorphism, limited mitotic activity, and no tumour cell necrosis. All cases were negative for MDM2 overexpression and amplification as determined with immunohistochemistry and fluorescence in-situ hybridisation, respectively. Immunohistochemically, p16 was diffusely overexpressed in all cases; seven tumours (88%) showed abnormal loss of Rb and p53. TP53 mutation or deletion was identified in four of six tumours evaluated with exon-targeted hybrid capture-based massively parallel sequencing; RB1 mutation or deletion was present in five of six cases. CONCLUSIONS: We present a series of eight well-differentiated lipomatous neoplasms characterised by p53 alterations in addition to Rb loss and histological features of APLT. These findings suggest that impaired p53 signalling may contribute to the pathogenesis of APLT in a subset of cases.

Identification and characterization of effusion tumor cells (ETCs) from remnant pleural effusion specimens.

BACKGROUND: Cancer is a leading cause of death worldwide, and patients may have advanced disease when diagnosed. Targeted therapies guided by molecular subtyping of cancer can benefit patients significantly. Pleural effusions are frequently observed in patients with metastatic cancer and are routinely removed for therapeutic purposes; however, effusion specimens have not been recognized as typical substrates for clinical molecular testing because of frequent low tumor cellularity. METHODS: Excess remnant pleural effusion samples (N = 25) from 21 patients with and without suspected malignancy were collected at Stanford Health Care between December 2019 and November 2020. Samples were processed into ThinPrep slides and underwent novel effusion tumor cell (ETC) analysis. The ETC results were compared with the original clinical diagnoses for accuracy. A subset of confirmed ETCs was further isolated and processed for molecular profiling to identify cancer driver mutations. All samples were obtained with Institutional Review Board approval. RESULTS: The authors established novel quantitative standards to identify ETCs and detected epithelial malignancy with 89.5% sensitivity and 100% specificity in the pleural effusion samples. Molecular profiling of confirmed ETCs (pools of 5 cells evaluated) revealed key pathogenic mutations consistent with clinical molecular findings. CONCLUSIONS: In this study, the authors developed a novel ETC-testing assay that detected epithelial malignancies in pleural effusions with high sensitivity and specificity. Molecular profiling of 5 ETCs showed promising concordance with the clinical molecular findings. To promote cancer subtyping and guide treatment, this ETC-testing assay will need to be validated in larger patient cohorts to facilitate integration into cytologic workflow.

Nodular fasciitis of the breast: clinicopathologic and molecular characterization with identification of novel USP6 fusion partners.

Nodular fasciitis is a benign, self-limited, pseudosarcomatous neoplasm that can mimic malignancy due to its rapid growth, cellularity, and mitotic activity. Involvement of the breast is rare and diagnosis on biopsy can be challenging. In this largest series to date, we examined the clinicopathologic and molecular characteristics of 12 cases of nodular fasciitis involving the breast/axilla. All patients were female, with a median age of 32 years (range 15-61). The lesions were 0.4 to 5.8 cm in size (median 0.8). All cases presented as palpable masses, and two patients had overlying skin retraction. Microscopically, lesions were relatively well-circumscribed nodular masses of bland myofibroblastic spindle cells within a variably myxoid stroma. Infiltrative growth into adipose tissue or breast epithelium was frequent. Mitotic figures were present in all cases, ranging from 1 to 12 per 10 high-power fields (median 3). Immunohistochemically, all cases expressed smooth muscle actin and were negative for pan-cytokeratin, p63, desmin, CD34, and nuclear beta-catenin. Targeted RNA sequencing performed on 11 cases identified USP6 gene fusions in eight; one additional case was positive by break-apart fluorescence in situ hybridization. The common MYH9-USP6 rearrangement was detected in four cases; another case had a rare alternative fusion with CTNNB1. Three cases harbored novel USP6 gene fusions involving NACA, SLFN11, or LDHA. All fusions juxtaposed the promoter region of the 5' partner gene with the full-length coding sequence of USP6. Outcome data were available for eight patients; none developed recurrence or metastasis. Five patients elected for observation without immediate excision, and self-resolution of the lesions was reported in three cases. Albeit uncommon, nodular fasciitis should be considered in the differential diagnosis of breast spindle cell lesions. A broad immunohistochemical panel to exclude histologic mimics, including metaplastic carcinoma, is important. Confirmatory detection of USP6 rearrangements can aid in classification, with potential therapeutic implications.

Automated detection of aggressive and indolent prostate cancer on magnetic resonance imaging.

PURPOSE: While multi-parametric magnetic resonance imaging (MRI) shows great promise in assisting with prostate cancer diagnosis and localization, subtle differences in appearance between cancer and normal tissue lead to many false positive and false negative interpretations by radiologists. We sought to automatically detect aggressive cancer (Gleason pattern ≥ 4) and indolent cancer (Gleason pattern 3) on a per-pixel basis on MRI to facilitate the targeting of aggressive cancer during biopsy. METHODS: We created the Stanford Prostate Cancer Network (SPCNet), a convolutional neural network model, trained to distinguish between aggressive cancer, indolent cancer, and normal tissue on MRI. Ground truth cancer labels were obtained by registering MRI with whole-mount digital histopathology images from patients who underwent radical prostatectomy. Before registration, these histopathology images were automatically annotated to show Gleason patterns on a per-pixel basis. The model was trained on data from 78 patients who underwent radical prostatectomy and 24 patients without prostate cancer. The model was evaluated on a pixel and lesion level in 322 patients, including six patients with normal MRI and no cancer, 23 patients who underwent radical prostatectomy, and 293 patients who underwent biopsy. Moreover, we assessed the ability of our model to detect clinically significant cancer (lesions with an aggressive component) and compared it to the performance of radiologists. RESULTS: Our model detected clinically significant lesions with an area under the receiver operator characteristics curve of 0.75 for radical prostatectomy patients and 0.80 for biopsy patients. Moreover, the model detected up to 18% of lesions missed by radiologists, and overall had a sensitivity and specificity that approached that of radiologists in detecting clinically significant cancer. CONCLUSIONS: Our SPCNet model accurately detected aggressive prostate cancer. Its performance approached that of radiologists, and it helped identify lesions otherwise missed by radiologists. Our model has the potential to assist physicians in specifically targeting the aggressive component of prostate cancers during biopsy or focal treatment.

In Vivo Imaging of Methionine Aminopeptidase II for Prostate Cancer Risk Stratification.

Prostate cancer is one of the most common malignancies worldwide, yet limited tools exist for prognostic risk stratification of the disease. Identification of new biomarkers representing intrinsic features of malignant transformation and development of prognostic imaging technologies are critical for improving treatment decisions and patient survival. In this study, we analyzed radical prostatectomy specimens from 422 patients with localized disease to define the expression pattern of methionine aminopeptidase II (MetAP2), a cytosolic metalloprotease that has been identified as a druggable target in cancer. MetAP2 was highly expressed in 54% of low-grade and 59% of high-grade cancers. Elevated levels of MetAP2 at diagnosis were associated with shorter time to recurrence. Controlled self-assembly of a synthetic small molecule enabled design of the first MetAP2-activated PET imaging tracer for monitoring MetAP2 activity in vivo. The nanoparticles assembled upon MetAP2 activation were imaged in single prostate cancer cells with post-click fluorescence labeling. The fluorine-18-labeled tracers successfully differentiated MetAP2 activity in both MetAP2-knockdown and inhibitor-treated human prostate cancer xenografts by micro-PET/CT scanning. This highly sensitive imaging technology may provide a new tool for noninvasive early-risk stratification of prostate cancer and monitoring the therapeutic effect of MetAP2 inhibitors as anticancer drugs. SIGNIFICANCE: This study defines MetAP2 as an early-risk stratifier for molecular imaging of aggressive prostate cancer and describes a MetAP2-activated self-assembly small-molecule PET tracer for imaging MetAP2 activity in vivo.

A Functional Taxonomy of Tumor Suppression in Oncogenic KRAS-Driven Lung Cancer.

Cancer genotyping has identified a large number of putative tumor suppressor genes. Carcinogenesis is a multistep process, but the importance and specific roles of many of these genes during tumor initiation, growth, and progression remain unknown. Here we use a multiplexed mouse model of oncogenic KRAS-driven lung cancer to quantify the impact of 48 known and putative tumor suppressor genes on diverse aspects of carcinogenesis at an unprecedented scale and resolution. We uncover many previously understudied functional tumor suppressors that constrain cancer in vivo. Inactivation of some genes substantially increased growth, whereas the inactivation of others increases tumor initiation and/or the emergence of exceptionally large tumors. These functional in vivo analyses revealed an unexpectedly complex landscape of tumor suppression that has implications for understanding cancer evolution, interpreting clinical cancer genome sequencing data, and directing approaches to limit tumor initiation and progression. SIGNIFICANCE: Our high-throughput and high-resolution analysis of tumor suppression uncovered novel genetic determinants of oncogenic KRAS-driven lung cancer initiation, overall growth, and exceptional growth. This taxonomy is consistent with changing constraints during the life history of cancer and highlights the value of quantitative in vivo genetic analyses in autochthonous cancer models.This article is highlighted in the In This Issue feature, p. 1601.