Latent Diffusion Models with Image-Derived Annotations for Enhanced AI-Assisted Cancer Diagnosis in Histopathology.

Artificial Intelligence (AI)-based image analysis has immense potential to support diagnostic histopathology, including cancer diagnostics. However, developing supervised AI methods requires large-scale annotated datasets. A potentially powerful solution is to augment training data with synthetic data. Latent diffusion models, which can generate high-quality, diverse synthetic images, are promising. However, the most common implementations rely on detailed textual descriptions, which are not generally available in this domain. This work proposes a method that constructs structured textual prompts from automatically extracted image features. We experiment with the PCam dataset, composed of tissue patches only loosely annotated as healthy or cancerous. We show that including image-derived features in the prompt, as opposed to only healthy and cancerous labels, improves the Fréchet Inception Distance (FID) by 88.6. We also show that pathologists find it challenging to detect synthetic images, with a median sensitivity/specificity of 0.55/0.55. Finally, we show that synthetic data effectively train AI models.

Replacing concurrent controls with virtual control groups in rat toxicity studies.

Virtual control groups (VCGs) in nonclinical toxicity represent the concept of using appropriate historical control data for replacing concurrent control group animals. Historical control data collected from standardized studies can serve as base for constructing VCGs and legacy study reports can be used as a benchmark to evaluate the VCG performance. Replacing concurrent controls of legacy studies with VCGs should ideally reproduce the results of these studies. Based on three four-week rat oral toxicity legacy studies with varying degrees of toxicity findings we developed a concept to evaluate VCG performance on different levels: the ability of VCGs to (i) reproduce statistically significant deviations from the concurrent control, (ii) reproduce test substance-related effects, and (iii) reproduce the conclusion of the toxicity study in terms of threshold dose, target organs, toxicological biomarkers (clinical pathology) and reversibility. Although VCGs have shown a low to moderate ability to reproduce statistical results, the general study conclusions remained unchanged. Our results provide a first indication that carefully selected historical control data can be used to replace concurrent control without impairing the general study conclusion. Additionally, the developed procedures and workflows lay the foundation for the future validation of virtual controls for a use in regulatory toxicology.

The sGC Activator Runcaciguat Has Kidney Protective Effects and Prevents a Decline of Kidney Function in ZSF1 Rats.

Chronic kidney disease (CKD) progression is associated with persisting oxidative stress, which impairs the NO-sGC-cGMP signaling cascade through the formation of oxidized and heme-free apo-sGC that cannot be activated by NO. Runcaciguat (BAY 1101042) is a novel, potent, and selective sGC activator that binds and activates oxidized and heme-free sGC and thereby restores NO-sGC-cGMP signaling under oxidative stress. Therefore, runcaciguat might represent a very effective treatment option for CKD/DKD. The potential kidney-protective effects of runcaciguat were investigated in ZSF1 rats as a model of CKD/DKD, characterized by hypertension, hyperglycemia, obesity, and insulin resistance. ZSF1 rats were treated daily orally for up to 12 weeks with runcaciguat (1, 3, 10 mg/kg/bid) or placebo. The study endpoints were proteinuria, kidney histopathology, plasma, urinary biomarkers of kidney damage, and gene expression profiling to gain information about relevant pathways affected by runcaciguat. Furthermore, oxidative stress was compared in the ZSF1 rat kidney with kidney samples from DKD patients. Within the duration of the 12-week treatment study, kidney function was significantly decreased in obese ZSF1 rats, indicated by a 20-fold increase in proteinuria, compared to lean ZSF1 rats. Runcaciguat dose-dependently and significantly attenuated the development of proteinuria in ZSF1 rats with reduced uPCR at the end of the study by -19%, -54%, and -70% at 1, 3, and 10 mg/kg/bid, respectively, compared to placebo treatment. Additionally, average blood glucose levels measured as HbA1C, triglycerides, and cholesterol were increased by five times, twenty times, and four times, respectively, in obese ZSF1 compared to lean rats. In obese ZSF1 rats, runcaciguat reduced HbA1c levels by -8%, -34%, and -76%, triglycerides by -42%, -55%, and -71%, and cholesterol by -16%, -17%, and -34%, at 1, 3, and 10 mg/kg/bid, respectively, compared to placebo. Concomitantly, runcaciguat also reduced kidney weights, morphological kidney damage, and urinary and plasma biomarkers of kidney damage. Beneficial effects were accompanied by changes in gene expression that indicate reduced fibrosis and inflammation and suggest improved endothelial stabilization. In summary, the sGC activator runcaciguat significantly prevented a decline in kidney function in a DKD rat model that mimics common comorbidities and conditions of oxidative stress of CKD patients. Thus, runcaciguat represents a promising treatment option for CKD patients, which is in line with recent phase 2 clinical study data, where runcaciguat showed promising efficacy in CKD patients (NCT04507061).