Deep Learning-Enabled Diagnosis of Liver Adenocarcinoma.

BACKGROUND & AIMS: Diagnosis of adenocarcinoma in the liver is a frequent scenario in routine pathology and has a critical impact on clinical decision making. However, rendering a correct diagnosis can be challenging, and often requires the integration of clinical, radiologic, and immunohistochemical information. We present a deep learning model (HEPNET) to distinguish intrahepatic cholangiocarcinoma from colorectal liver metastasis, as the most frequent primary and secondary forms of liver adenocarcinoma, with clinical grade accuracy using H&E-stained whole-slide images. METHODS: HEPNET was trained on 714,589 image tiles from 456 patients who were randomly selected in a stratified manner from a pool of 571 patients who underwent surgical resection or biopsy at Heidelberg University Hospital. Model performance was evaluated on a hold-out internal test set comprising 115 patients and externally validated on 159 patients recruited at Mainz University Hospital. RESULTS: On the hold-out internal test set, HEPNET achieved an area under the receiver operating characteristic curve of 0.994 (95% CI, 0.989-1.000) and an accuracy of 96.522% (95% CI, 94.521%-98.694%) at the patient level. Validation on the external test set yielded an area under the receiver operating characteristic curve of 0.997 (95% CI, 0.995-1.000), corresponding to an accuracy of 98.113% (95% CI, 96.907%-100.000%). HEPNET surpassed the performance of 6 pathology experts with different levels of experience in a reader study of 50 patients (P = .0005), boosted the performance of resident pathologists to the level of senior pathologists, and reduced potential downstream analyses. CONCLUSIONS: We provided a ready-to-use tool with clinical grade performance that may facilitate routine pathology by rendering a definitive diagnosis and guiding ancillary testing. The incorporation of HEPNET into pathology laboratories may optimize the diagnostic workflow, complemented by test-related labor and cost savings.

[Current therapeutic options in Mycosis fungoides and Sézary syndrome]. / Aktuelle Therapieoptionen bei Mycosis fungoides und Sézary-Syndrom.

Driven by the approval of new targeted therapies, significant progress has been made in recent years in the clinical management of cutaneous T­cell lymphomas. Although there are no curative treatment options for cutaneous T­cell lymphomas, response rates are often encouraging, in particular when using combination therapies. The decision for the appropriate form of treatment depends on the specific diagnosis, disease stage, and the history of prior therapies. This article provides a comprehensive overview of current treatment options in mycosis fungoides and Sézary syndrome, based on the recently published, revised German S2k guidelines on cutaneous lymphomas (update 2021). In addition, we present promising, yet-to-be-approved therapies that at least in part can be already used off-label in clinical practice today.

Primary cutaneous diffuse large B-cell lymphoma, NOS and leg type: Clinical, morphologic and prognostic differences.

BACKGROUND AND OBJECTIVES: Primary cutaneous diffuse large B-cell lymphoma, NOS (PCLBCL/NOS) is a rare PCLBCL. Only few data are available for this tumor. The aim of this study was to identify clinical and/or immunohistochemical markers (in addition to Bcl-2) that characterize PCLBCL/NOS, assist in differentiating it from PCLBCL, leg type (PCLBCL/LT) and help to assess the clinical course/prognosis. PATIENTS AND METHODS: Bcl-2- PCLBCL/NOS) cases (n = 14 were compared with Bcl-2+ PCLBCL/LT cases (n = 29). RESULTS: PCLBCL/NOS patients were younger, predominantly male and had better survival rates than patients with PCLBCL/LT. Patients with PCLBCL/NOS presented more often with larger plaques limited to one or two contiguous body regions, whereas PCLBCL/LT cases often presented with disseminated lesions. Neoplastic cells had a higher proliferation rate (Ki67) in PCLBCL/LT patients. The tumor microenvironment of PCLBCL/NOS had a more prominent CD3+ infiltrate. Overall survival data for the whole cohort (n = 37) revealed that female gender and Bcl-2 expression correlated with a worse survival rate. Bcl-6 expression and centroblastic subtype correlated with better outcomes. None of the other markers studied (e.g. GCB/non-GCB subtype) correlated with survival rate. CONCLUSIONS: PCLBCL/NOS and PCLBCL/LT differ in their clinical behavior and outcomes. Bcl-2 still seems to be the best marker for discriminating between these two subgroups. Bcl-2, female gender and Bcl-6 represent prognostic markers for PCLBCL.

Cutaneous tumor cell load correlates with survival in patients with Sézary syndrome.

BACKGROUND: Sézary syndrome (SS) is defined by the triad of erythroderma, generalized lymphadenopathy and more than 1 000 circulating Sézary cells/µl in the peripheral blood. PATIENTS AND METHODS: We screened the cutaneous lymphoma registry of our department for SS patients to identify clinical features of SS besides the defining criteria and to correlate them with disease survival. RESULTS: 24 SS patients were analyzed retrospectively. The mean age was 65 years with 62 % male patients. The median follow-up time was 32.5 months with an estimated 5-year overall survival rate of 76 %. All patients complained about itching and presented with palmoplantar keratoderma. 62.5 % had nail involvement, 21 % alopecia, 12.5 % ectropion, 4 % prurigo nodularis, 8 % localized and 8 % generalized skin tumors, including leonine facies. In addition, 33 % had infections and also 33 % had venous thromboembolism. We identified cutaneous tumor cell load as a significant prognostic marker for SS. None of the other parameters were associated with disease specific survival. CONCLUSIONS: Clinically SS is characterized by various presentations beyond erythroderma. The cutaneous tumor cell load in SS is strongly associated with outcome and survival. We demonstrate a high risk for venous thromboembolism in SS patients who might benefit from anti-coagulation therapies.