Extra domain B splice variant of fibronectin (EDB+FN) is an extracellular matrix protein (ECM) deposited by tumor-associated fibroblasts, and is associated with tumor growth, angiogenesis, and invasion. We hypothesized that EDB+FN is a safe and abundant target for therapeutic intervention with an antibody–drug conjugate (ADC). We describe the generation, pharmacology, mechanism of action, and safety profile of an ADC specific for EDB+FN (EDB-ADC). EDB+FN is broadly expressed in the stroma of pancreatic, non–small cell lung (NSCLC), breast, ovarian, head and neck cancers, whereas restricted in normal tissues. In patient-derived xenograft (PDX), cell-line xenograft (CLX), and mouse syngeneic tumor models, EDB-ADC, conjugated to auristatin Aur0101 through site-specific technology, demonstrated potent antitumor growth inhibition. Increased phospho-histone H3, a pharmacodynamic biomarker of response, was observed in tumor cells distal to the target site of tumor ECM after EDB-ADC treatment. EDB-ADC potentiated infiltration of immune cells, including CD3+ T lymphocytes into the tumor, providing rationale for the combination of EDB-ADC with immune checkpoint therapy. EDB-ADC and anti-PD-L1 combination in a syngeneic breast tumor model led to enhanced antitumor activity with sustained tumor regressions. In nonclinical safety studies in nonhuman primates, EDB-ADC had a well-tolerated safety profile without signs of either on-target toxicity or the off-target effects typically observed with ADCs that are conjugated through conventional conjugation methods. These data highlight the potential for EDB-ADC to specifically target the tumor microenvironment, provide robust therapeutic benefits against multiple tumor types, and enhance activity antitumor in combination with checkpoint blockade.

Whole slide imaging enables the use of a wide array of digital image analysis tools that are revolutionizing pathology. Recent advances in digital pathology and deep convolutional neural networks have created an enormous opportunity to improve workflow efficiency, provide more quantitative, objective, and consistent assessments of pathology datasets, and develop decision support systems. Such innovations are already making their way into clinical practice. However, the progress of machine learning - in particular, deep learning (DL) - has been rather slower in nonclinical toxicology studies. Histopathology data from toxicology studies are critical during the drug development process that is required by regulatory bodies to assess drug-related toxicity in laboratory animals and its impact on human safety in clinical trials. Due to the high volume of slides routinely evaluated, low-throughput, or narrowly performing DL methods that may work well in small-scale diagnostic studies or for the identification of a single abnormality are tedious and impractical for toxicologic pathology. Furthermore, regulatory requirements around good laboratory practice are a major hurdle for the adoption of DL in toxicologic pathology. This paper reviews the major DL concepts, emerging applications, and examples of DL in toxicologic pathology image analysis. We end with a discussion of specific challenges and directions for future research.

In the last decade, numerous initiatives have emerged worldwide to reduce the use of animals in drug development, including more recently the introduction of Virtual Control Groups (VCGs) concept for nonclinical toxicity studies. Although replacement of concurrent controls (CCs) by virtual controls (VCs) represents an exciting opportunity, there are associated challenges that will be discussed in this paper with a more specific focus on anatomic pathology. Coordinated efforts will be needed from toxicologists, clinical and anatomic pathologists, and regulators to support approaches that will facilitate a staggered implementation of VCGs in nonclinical toxicity studies. Notably, the authors believe that a validated database for VC animals will need to include histopathology (digital) slides for microscopic assessment. Ultimately, the most important step lies in the validation of the concept by performing VCG and the full control group in parallel for studies of varying duration over a reasonable timespan to confirm there are no differences in outcomes (dual study design). The authors also discuss a hybrid approach, whereby control groups comprised both concurrent and VCs to demonstrate proof-of-concept. Once confidence is established by sponsors and regulators, VCs have the potential to replace some or all CC animals.

Sinusoidal obstruction syndrome (SOS) is a unique form of liver injury that occurs after exposure to chemotherapeutic drugs and toxins. The diagnosis of SOS in humans remains a challenge as the clinical criteria have low specificity and there are no reliable noninvasive biomarkers. The mechanism of injury is believed to be damage to liver endothelial cells, primarily sinusoidal endothelial cells (SECs), which leads to sinusoidal dilation, central venous fibrosis, and/or nodular regeneration. Nonclinical data suggest that this uncommon liver toxicity can be recapitulated in cynomolgus monkeys, and it is critical that pathologists are familiar with its characteristic clinicopathologic features. Elevations in liver enzymes, in particular aspartate aminotransferase, associated with isolated thrombocytopenia, should raise the suspicion of SEC injury for specific drug classes. Characterization of liver microscopic findings in monkeys benefits from the use of appropriate stains, such as reticulin stain, and VEGFR2 and CD34 immunohistochemical (IHC) stains. CD41 IHC demonstrates platelet accumulation in injured sinusoids, the likely cause of thrombocytopenia commonly reported in SOS. In conclusion, this report provides a comprehensive characterization of the pathology findings of drug-induced SOS in monkeys with the objectives of ensuring appropriate nonclinical recognition of the liability and informing clinical development strategy and monitoring.

Enhanced histopathology of the immune system uses a precise, compartment-specific, and semi-quantitative evaluation of lymphoid organs in toxicology studies. The assessment of lymphocyte populations in tissues is subject to sampling variability and limited distinctive cytologic features of lymphocyte subpopulations as seen with hematoxylin and eosin (H&E) staining. Although immunohistochemistry is necessary for definitive characterization of T- and B-cell compartments, routine toxicologic assessments are based solely on H&E slides. Here, a deep learning (DL) model was developed using normal rats to quantify relevant compartments of the spleen, including periarteriolar lymphoid sheaths, follicles, germinal centers, and marginal zones from H&E slides. Slides were scanned, destained, dual labeled with CD3 and CD79a chromogenic immunohistochemistry, and rescanned to generate exact co-registered images that served as the ground truth for training and validation. The DL model identified individual splenic compartments with high accuracy (97.8% Dice similarity coefficient) directly from H&E-stained tissue. The DL model was utilized to study the normal range of lymphoid compartment area and cellularity. Future implementation of our DL model and expanding this approach to other lymphoid tissues have the potential to improve accuracy and precision in enhanced histopathology evaluation of the immune system with concurrent gains in time efficiency for the pathologist.

Pharmaceutical and biotechnology companies rarely disclose their use of translational emerging safety biomarkers (ESBs) during drug development, and the impact of ESB use on the speed of drug development remains unclear. A cross-industry survey of 20 companies of varying size was conducted to understand current trends in ESB use and future use prospects. The objectives were to: (1) determine current ESB use in nonclinical and clinical drug development and impact on asset advancement; (2) identify opportunities, gaps, and challenges to greater ESB implementation; and (3) benchmark perspectives on regulatory acceptance. Although ESBs were employed in only 5-50% of studies/programs, most companies used ESBs to some extent, with larger companies demonstrating greater nonclinical use. Inclusion of ESBs in investigational new drug applications (INDs) was similar across all companies; however, differences in clinical trial usage could vary among the prevailing health authority (HA). Broader implementation of ESBs requires resource support, cross-industry partnerships, and collaboration with HAs. This includes generating sufficient foundational data, demonstrating nonclinical to clinical translatability and practical utility, and clearly written criteria by HAs to enable qualification. If achieved, ESBs will play a critical role in the development of next-generation, translationally-tailored standard laboratory tests for drug development.