Tanja S. Zabka

In vitro to in vivo concordance of a high throughput assay of bone marrow toxicity across a diverse set of drug candidates.

The development of predictive toxicology assays is necessary to optimize the drug candidate selection process. The colony forming assay (CFA) is used routinely to assess bone marrow toxicity and represents a viable tool for the discovery toxicologist, but the assay is not widely accepted as a standard screening tool due to technical challenges. A higher throughput and standardized version of the assay recently was developed such that the proliferative capacity of a cell lineage is measured indirectly via ATP levels, replacing the cumbersome identification and enumeration of specific colonies. In this study, a high-throughput assay of bone marrow toxicity prediction using the granulocyte, erythrocyte, monocyte, and macrophage (GEMM) progenitor cell lineage was evaluated using a training set of 56 structurally diverse compounds with known in vivo bone marrow effects. In general, compounds identified as toxic in vivo had lower IC(50) values, whereas those identified as non-toxic had higher IC(50) values. Concordance (i.e., predictive accuracy) to in vivo bone marrow toxicity results was 82% when an in vitro toxicity threshold of 20 microM was used. Additional experiments in other hematopoietic lineages were conducted to determine if predictivity of several false positive and negative compounds in the GEMM lineage could be improved; however an increase in sensitivity or specificity was not observed. The high-throughput GEMM assay has good concordance to in vivo bone marrow toxicity results and, with the high-throughput and standardized format, can be incorporated readily into the pharmaceutical toxicological screening paradigm, aiding in the early identification of compounds that eventually may fail due to bone marrow toxicity.

Recommendations for the evaluation of pathology data in nonclinical safety biomarker qualification studies

A set of best practices for the conduct of histopathology evaluation in nonclinical safety studies was endorsed by the Society of Toxicologic Pathology (STP) in 2004. These best practices indicate that the study pathologist should have knowledge of the treatment group and access to all available study-related data for the animal from which the tissue was obtained. A new set of best practices for the conduct of histopathology review for safety biomarker qualification for nonclinical studies has been endorsed by the STP and is summarized in this document. These best practices are generally similar to those for nonclinical safety studies, specifically that the pathologist be “unblinded” or have access to study data. Although histopathology evaluation in biomarker qualification studies must be performed without knowledge of novel biomarker data, the study pathologist(s) should be involved in the attendant meta-analyses of these data. Blinded evaluation is an experimental tool in biomarker qualification studies that is appropriate only when well-defined criteria for specific histopathologic findings are identified prior to blinded review. Additionally, this paper also considers the management of bias, the use of a tiered evaluation approach, the importance of using qualified pathologists and standard reporting, and the management of spontaneous findings.

Characterization of Xenobiotic-Induced Hepatocellular Enzyme Induction in Rats: Anticipated Thyroid Effects and Unique Pituitary Gland Findings

During routine safety evaluation of RO2910, a non-nucleoside reverse transcriptase inhibitor for HIV infection, histopathology findings concurrent with robust hepatocellular induction occurred in multiple organs, including a unique, albeit related, finding in the pituitary gland. For fourteen days, male and female rats were administered, by oral gavage vehicle, 100, 300, or 1000 mg/kg/day of RO2910. Treated groups had elevated serum thyroid-stimulating hormone and decreased total thyroxine, and hypertrophy in the liver, thyroid gland, and pituitary pars distalis. These were considered consequences of hepatocellular induction and often were dose dependent and more pronounced in males than in females. Hepatocellular centrilobular hypertrophy corresponded with increased expression of cytochrome P450s 2B1/2, 3A1, and 3A2 and UGT 2B1. Bilateral thyroid follicular cell hypertrophy occurred concurrent to increased mitotic activity and sometimes colloid depletion, which were attributed to changes in thyroid hormone levels. Males had hypertrophy of thyroid-stimulating hormone–producing cells (thyrotrophs) in the pituitary pars distalis. All findings were consistent with the well-established adaptive physiologic response of rodents to xenobiotic-induced hepatocellular microsomal enzyme induction. Although the effects on the pituitary gland following hepatic enzyme induction-mediated hypothyroidism have not been reported previously, other models of stress and thyroid depletion leading to pituitary stimulation support such a shared pathogenesis.

Spontaneous Cardiomyopathy in Cynomolgus Monkeys (Macaca Fascicularis)

A previously undescribed spontaneous cardiomyopathy was identified by routine light microscopic examination of the heart from four clinically healthy purpose-bred cynomolgus monkeys that ranged from four to nine years of age and included 2 males and 2 females. Special stains of Sirius red, Masson’s trichrome, and Mallory’s phosphotungstic acid hematoxylin (PTAH); and immunohistochemistry using anti-CD68, troponin-I, and desmin antibodies were used to facilitate lesion characterization and assess cardiomyocyte viability. Microscopically, the apical to mid-ventricular myocardium to subendocardium had foci of cardiomyocyte disarray with cytoplasmic pallor to stippling and karyomegaly, vacuolization of the perimyseal connective tissue, a meshwork of fibrous tissue that concentrated around medium-sized blood vessels and dissected between or less often replaced affected cardiomyocytes; and a minimal, predominantly macrophage infiltrate. The disrupted cardiomyocytes were immunoreactive to desmin and troponin-I antibodies and had a normal cross-striation pattern by PTAH, indicating the chronic cardiomyopathy was not associated with active cardiomyocyte damage. The consistent distribution and morphology of the cardiomyopathy suggested a common etiology and pathogenesis. The features were reminiscent of chronic catecholamine-induced experimental cardiomyopathy and stress cardiomyopathy in monkeys and humans, respectively. This report documents another spontaneous heart lesion in clinically healthy monkeys for consideration during interpretation of toxicology studies.

Quantitative Histological Assessment of Xenobiotic-Induced Liver Enzyme Induction and Pituitary-Thyroid Axis Stimulation in Rats Using Whole-Slide Automated Image Analysis

Preclinical evaluation of a new compound, RO2910, identified a hypertrophic response in liver, thyroid gland, and pituitary gland (pars distalis). We aimed to develop and validate automated image analysis methods to quantify and refine the interpretation of semi-quantitative histology. Wistar-Han rats were administered RO2910 for 14 days. Liver, thyroid, and pituitary gland tissues were processed for routine histology and immunolabeled with anti–thyroid stimulating hormone (TSH) antibody (pituitary) and anti–topoisomerase II antibody (thyroid). Glass slides were scanned, image analysis methods were developed and applied to whole-slide images, and numerical results were compared with histopathology, circulating hormone levels, and liver enzyme mRNA expression for validation. Quantitative analysis of slides had strong individual correlation with semi-quantitative histological evaluation of all tissues studied. Hepatocellular hypertrophy quantification also correlated strongly with liver enzyme mRNA expression. In the pars distalis, measurement of TSH weak-staining areas correlated with both hypertrophy scores and circulating TSH levels. Whole-slide image analysis enabled automated quantification of semi-quantitative histopathology findings and a more refined interpretation of these data. The analysis also enabled a direct correlation with non-histological parameters using straightforward statistical analysis to provide a more refined dose- and sex-response relationship and integration among affected parameters. These findings demonstrate the utility of our image analysis to support preclinical safety evaluations.

Modeling Bone Marrow Toxicity Using Kinase Structural Motifs and the Inhibition Profiles of Small Molecular Kinase Inhibitors

The cellular function of kinases combined with the difficulty of designing selective small molecule kinase inhibitors (SMKIs) poses a challenge for drug development. The late-stage attrition of SMKIs could be lessened by integrating safety information of kinases into the lead optimization stage of drug development. Herein, a mathematical model to predict bone marrow toxicity (BMT) is presented which enables the rational design of SMKIs away from this safety liability. A specific example highlights how this model identifies critical structural modifications to avoid BMT. The model was built using a novel algorithm, which selects 19 representative kinases from a panel of 277 based upon their ATP-binding pocket sequences and ability to predict BMT in vivo for 48 SMKIs. A support vector machine classifier was trained on the selected kinases and accurately predicts BMT with 74% accuracy. The model provides an efficient method for understanding SMKI-induced in vivo BMT earlier in drug discovery.

Topic of Histopathology Blinding in Nonclinical Safety Biomarker Qualification Studies

Various consortia and working groups, composed of professionals from industry, academia, and government institutions, have undertaken or are undertaking nonclinical work to qualify safety biomarkers of tissue injury and function. As this work has developed and voluntary data have been submitted to regulatory authorities, study practices have come under close scrutiny in an attempt to ensure that the best science is consistently being applied. One practice that has been discussed in a variety of venues is the process used to generate histopathology data as additional end points and/or correlates in these studies. Histopathologic evaluation plays a critical role in these biomarker studies, because microscopic demonstration of given cellular processes are commonly used as a reference standard to assess diagnostic performance of candidate new biomarkers using methods such as receiver operating characteristic (ROC) analyses. In this regard, it is acknowledged that pathologists should not perform the histopathologic evaluation with knowledge of the candidate biomarker data; however, it has been questioned whether pathologists should conduct the histopathology evaluation for these biomarker studies without knowledge of treatment or other study-related data (i.e., ‘‘blinded’’ evaluation). Research pathologists, and especially the Society of Toxicologic Pathology (STP), have closely examined regulatory study histopathology practices (Crissman et al. 2004; Wandall, Hansson and Ruden 2007). Recently, a set of Best Practices for the conduct of histopathology review within nonclinical safety studies was endorsed by the STP (Crissman et al. 2004). Key elements of these Best Practices indicate that the study pathologist should be informed as follows: (1) have knowledge of the treatment group from which the sample was obtained; and (2) have complete knowledge of all available study-related data that are associated with the animal from which the tissue was obtained. As these practices have been employed successfully for decades in the context of regulatory toxicology studies designed to assess human safety of new chemical entities with complete regulatory acceptance, the rationale for following a different process for biomarker validation and qualification is unclear to these authors. This informed type of analysis (described above) is often referred to as ‘‘unblinded,’’ because the study pathologist has knowledge of dose groups and other study-related data at the time that the review is being conducted. It is considered that ‘‘unblinded’’ evaluation is critical to discriminate treatment-related changes from background, especially when subtle treatment-related effects increase the incidence or severity of spontaneous background findings. As needed, the study pathologist may reexamine tissues using a procedure sometimes termed ‘‘targeted masked’’ evaluation. This procedure entails reexamination of selected or all treated dose groups, randomly combined with controls and without knowledge of animal or group identity, to determine whether a subtle or equivocal finding can be identified consistently from control tissues. This second evaluation is performed after the pathologist feels confident that each finding has been fully characterized, to ensure either that subtle findings are assessed for incidence and severity or that equivocal findings are assessed to discriminate a true change from spontaneous background (in an unbiased manner). Further to the nonclinical safety study Best Practices endorsed by STP, a peer-review process is generally used as a method for quality assurance, in which a second pathologist can corroborate the interpretations of the study pathologist. This second evaluation is typically conducted in an ‘‘unblinded’’ fashion, as this evaluation is not meant to generate new data but rather to provide a second party evaluation of the interpretations and conclusions of the primary pathologist. Taking into consideration all available historical data as well as current practices of histopathology data generation, the current authors were tasked by the Scientific and Regulatory Policy Committee of the STP to consider the development of either a Best Practices or a Points to Consider guideline for industry and regulators, with emphasis on histopathology practices related to biomarker qualification. Although the final document will not be available until later, we wished to present our preliminary opinion based on the aforementioned Best Practices (Crissman et al. 2004). The present authors recognize that histopathology is a special discipline in which considerations for best practices may differ from those of other disciplines used in biomarker evaluation. The present authors also acknowledge that some toxicities are either unique to humans or idiosyncratic in nature and thus undetectable in standard nonclinical studies. Therefore, the method of evaluating slides would not affect the sensitivity for detection of such changes. Conversely, the present authors agree that ‘‘unblinded’’ slide evaluation creates a high degree of fidelity and consistency in the ability to identify animal toxicities. We concur with

Myocardial Mononuclear Cell Infiltrates Are Not Associated with Increased Serum Cardiac Troponin I in Cynomolgus Monkeys

Myocardial mononuclear cell infiltrate is a spontaneous cardiac finding commonly identified in laboratory cynomolgus monkeys. The infiltrates are predominantly composed of macrophages with lesser lymphocytes and are not typically associated with histologically detectable cardiomyocyte degeneration. These infiltrates are of concern because they confound interpretation of test article–related histopathology findings in nonclinical safety toxicology studies. The interpretation of safety studies would be simplified by a biomarker that could identify myocardial infiltrates prior to animal placement on study. We hypothesized that monkeys with myocardial mononuclear cell infiltrates could be identified before necropsy using an ultrasensitive immunoassay for cardiac troponin I (cTnI). Serum cTnI concentrations in monkeys with myocardial infiltrates were not higher than those in monkeys without infiltrates at any of the sampling times before and on the day of necropsy. Increased serum cTnI levels are not suitable for screening monkeys with myocardial mononuclear cell infiltrates before placement in the study.

Authors' response to letter to the editor on "image cytometry protocols".

Dear Editor, This letter is in response to the Letter to the Editor by Van Noorden and Chieco that references our article in the Journal of Histochemistry and Cytochemistry from May 2013, among other publications. We would like to thank the authors of the letter for their comments on the importance of thorough description of the methodology used when generating data from acquired images. Although we agree with their point on having understanding of the significance of proper methodology use and documentation, we believe that the level of description of the methodology, as often referenced to our prior article (Zabka et al. 2011), is correct and up to par with descriptions of similar methods in recent scientific publications. Nevertheless, to ensure that our main focus on the numerical correlation of image data with other molecular and clinical results would not be misinterpreted as shortage of procedural documentation, we would like to review and/or clarify some of the relevant methodology details that were mentioned in the letter to the editor and were not extensively specified or included in our article. Procedures for immunohistochemical staining were reviewed in the Materials & Methods section, whereas specific details were explained in a prior publication (Zabka et al. 2011) to which the audience was referred. Immunohistochemical staining intensity of positive regions was not quantified but rather utilized to define a minimum positive intensity threshold, which was set to achieve equivalence with manual scoring by the pathologist on a number of high-power slide fields. Thresholding for “dark staining” and “weak staining” was based on a comparison to the “normal cells” in the control group and set to achieve concordance with interpretation by the pathologist. For hematoxylin and eosin stains, segmentation was based on differences in grayscale intensities across image layers, as well as size/shape of the different image/slide elements. Color deconvolution for separation of hematoxylin and DAB was performed using the software tools referenced in the original article (Definiens Developer XD; Definiens Inc., Munich, Germany), which are based on color separation algorithms originally published by van der Laak et al. (2000). Last, as stated in the original article, whole-slide images were acquired at 20× magnification using a Mirax slide scanner (Carl Zeiss Microimaging; Thornwood, NY) with the standard factory equipment-mounted optics, which includes a 3 CCD color camera with a pixel resolution of 0.23 µm and white illumination light. In summary, we would like to emphasize that our original aim was to quantify the pathologist evaluation and provide a continuous scale of values for the morphological/cellular changes observed in the digital slides, so that these could be correlated to other study parameters easily, achieving a more refined interpretation of the changes. Quantitation of microscopic images is becoming increasingly important; however, meaningful data must be extracted from the original histological images. In order to accomplish this, precise, consistent methodology is necessary, and we have confidence that such was used in our research in the cited publication. Thus, the authors of the letter were correct in not assuming that “Garrido et al. (2013) have used inadequate image cytometry technology.”