The induction of apoptosis is a direct way to eliminate tumor cells and improve cancer therapy. Apoptosis is tightly controlled by the balance of pro- and antiapoptotic Bcl-2 proteins. BH3 mimetics neutralize the antiapoptotic function of Bcl-2 proteins and are highly promising compounds inducing apoptosis in several cancer entities including pediatric malignancies. However, the clinical application of BH3 mimetics in solid tumors is impeded by the frequent resistance to single BH3 mimetics and the anticipated toxicity of high concentrations or combination treatments. One potential avenue to increase the potency of BH3 mimetics is the development of immune cell-based therapies to counteract the intrinsic apoptosis resistance of tumor cells and sensitize them to immune attack. Here, we describe spheroid cultures of pediatric cancer cells that can serve as models for drug testing. In these 3D models, we were able to demonstrate that activated allogeneic Natural Killer (NK) cells migrated into tumor spheroids and displayed cytotoxicity against a wide range of pediatric cancer spheroids, highlighting their potential as anti-tumor effector cells. Next, we investigated whether treatment of tumor spheroids with subtoxic concentrations of BH3 mimetics can increase the cytotoxicity of NK cells. Notably, the cytotoxic effects of NK cells were enhanced by the addition of BH3 mimetics. Treatment with either the Bcl-XL inhibitor A1331852 or the Mcl-1 inhibitor S63845 increased the cytotoxicity of NK cells and reduced spheroid size, while the Bcl-2 inhibitor ABT-199 had no effect on NK cell-mediated killing. Taken together, this is the first study to describe the combination of BH3 mimetics targeting Bcl-XL or Mcl-1 with NK cell-based immunotherapy, highlighting the potential of BH3 mimetics in immunotherapy.

In gastric cancer (GC), there are four molecular subclasses that indicate whether patients respond to chemotherapy or immunotherapy, according to the TCGA. In clinical practice, however, not every patient undergoes molecular testing. Many laboratories have used well-implemented in situ techniques (IHC and EBER-ISH) to determine the subclasses in their cohorts. Although multiple stains are used, we show that a staining approach is unable to correctly discriminate all subclasses. As an alternative, we trained an ensemble convolutional neuronal network using bagging that can predict the molecular subclass directly from hematoxylin-eosin histology. We also identified patients with predicted intra-tumoral heterogeneity or with features from multiple subclasses, which challenges the postulated TCGA-based decision tree for GC subtyping. In the future, deep learning may enable targeted testing for molecular subtypes and targeted therapy for a broader group of GC patients.

No abstract

No abstract

Fluorescence confocal microscopy (FCM) is a rapidly evolving tool that provides real-time virtual HE images of native tissue. Data about the potential of FCM as an alternative to frozen sections for the evaluation of donor liver specimens are lacking so far. The aim of the current study was to determine the value of FCM in liver specimens according to the criteria of the German Society for Organ Procurement. In this prospective study, conventional histology and FCM scans of 50 liver specimens (60% liver biopsies, 26% surgical specimens, and 14% donor samples) were evaluated according to the German Society for Organ Procurement. A comparison of FCM scans and conventional frozen sections revealed almost perfect levels of agreement for cholangitis (κ = 0.877), fibrosis (κ = 0.843), and malignancy (κ = 0.815). Substantial levels of agreement could be obtained for macrovesicular steatosis (κ = 0.775), inflammation (κ = 0.763), necrosis (κ = 0.643), and steatohepatitis (κ = 0.643). Levels of agreement were moderate for microvesicular steatosis (κ = 0.563). The strength of agreement between frozen sections and FCM was superior to the comparison of conventional HE and FCM imaging. We introduce FCM as a potential alternative to the frozen section that may represent a novel approach to liver transplant pathology where timely feedback is crucial and the deployment of human resources is becoming increasingly difficult.

MUC16/CA125 is associated with cancer proliferation in several tumor entities. The data on MUC16 expression in cholangiocarcinoma (CCA) tissue are very limited. The aim of this study was to assess the MUC16 status and its impact on survival in CCA patients. All the patients with surgically resected CCA that were diagnosed between August 2005 and December 2021 at the University Hospital Frankfurt were retrospectively analyzed. A 7-Mucin biomarker panel was assessed by immunohistochemistry. For overall survival (OS), Kaplan–Meier curves and Cox-regression analyses were performed. Randomly selected intrahepatic cholangiocarcinoma (iCCA) were further processed for differential expression profiling. A total of 168 patients with CCA were classified as MUC16 (−) (66%, n = 111) and MUC16 (+) (34%, n = 57). Subgroup analyses revealed a median OS of 56.1 months (95% CI = 42.4–69.9 months) and 27.4 months (95% CI = 15.8–39.1 months) for MUC16 (−) and MUC16 (+), respectively (p =< 0.001). In multivariate analysis, MUC16 (+) (HR = 1.6, 95% CI = 1–2.6, p = 0.032) was an independent risk factor for poor prognosis. Prominently deregulated pathways have been identified following MUC16 expression, overrepresented in cell cycle and immune system exhaustion processes. These findings suggest including MUC16 in clinical routine diagnostics as well as studying its molecular pathways to identify further mechanistic key players.

There is a lot of recent interest in the field of computational pathology, as many algorithms are introduced to detect, for example, cancer lesions or molecular features. However, there is a large gap between artificial intelligence (AI) technology and practice, since only a small fraction of the applications is used in routine diagnostics. The main problems are the transferability of convolutional neural network (CNN) models to data from other sources and the identification of uncertain predictions. The role of tissue quality itself is also largely unknown. Here, we demonstrated that samples of the TCGA ovarian cancer (TCGA-OV) dataset from different tissue sources have different quality characteristics and that CNN performance is linked to this property. CNNs performed best on high-quality data. Quality control tools were partially able to identify low-quality tiles, but their use did not increase the performance of the trained CNNs. Furthermore, we trained NoisyEnsembles by introducing label noise during training. These NoisyEnsembles could improve CNN performance for low-quality, unknown datasets. Moreover, the performance increases as the ensemble become more consistent, suggesting that incorrect predictions could be discarded efficiently to avoid wrong diagnostic decisions.

Background: Hirschsprung disease (HD) is an aganglionosis of variable length starting at the rectosigmoid colon with surgery as sole therapeutic option. The length of the resected bowel segment is a crucial information for the treating surgeons and influences the prognosis of the patient. It is often artificially altered due to post operative tissue shrinkage. The objective of this study is to quantify the extent tissue shrinkage of HD specimens. Material and Methods: Colorectal HD specimens were measured at the time of surgery and at the time of cut-up, either fresh or after formalin fixation and statistically analyzed. Results: Sixteen colorectal specimens were included. Following formalin fixation the specimen length decreased by 22.7% ( P < .001). Without formalin fixation the specimens shrank by an average of 24.9% ( P = .05). There was no significant difference in the extent of tissue shrinkage with or without formalin fixation ( P = .76). Conclusion: This study showed that there is significant tissue shrinkage in HD specimens. The 2 different cohorts revealed that tissue shrinkage is mostly caused by tissue retraction/alteration after organ removal but also to a lesser extent by fixation with formalin. Surgeons and (neuro-)pathologists should be aware of the sizeable shrinking artifact to avoid unnecessary confusion.