Pathologist-trained machine learning classifiers developed to quantitate celiac disease features differentiate endoscopic biopsies according to modified marsh score and dietary intervention response.

BACKGROUND: Histologic evaluation of the mucosal changes associated with celiac disease is important for establishing an accurate diagnosis and monitoring the impact of investigational therapies. While the Marsh-Oberhuber classification has been used to categorize the histologic findings into discrete stages (i.e., Type 0-3c), significant variability has been documented between observers using this ordinal scoring system. Therefore, we evaluated whether pathologist-trained machine learning classifiers can be developed to objectively quantitate the pathological changes of villus blunting, intraepithelial lymphocytosis, and crypt hyperplasia in small intestine endoscopic biopsies. METHODS: A convolutional neural network (CNN) was trained and combined with a secondary algorithm to quantitate intraepithelial lymphocytes (IEL) with 5 classes on CD3 immunohistochemistry whole slide images (WSI) and used to correlate feature outputs with ground truth modified Marsh scores in a total of 116 small intestine biopsies. RESULTS: Across all samples, median %CD3 counts (positive cells/enterocytes) from villous epithelium (VE) increased with higher Marsh scores (Type 0%CD3 VE = 13.4; Type 1-3%CD3 VE = 41.9, p < 0.0001). Indicators of villus blunting and crypt hyperplasia were also observed (Type 0-2 villous epithelium/lamina propria area ratio = 0.81; Type 3a-3c villous epithelium/lamina propria area ratio = 0.29, p < 0.0001), and Type 0-1 crypt/villous epithelial area ratio = 0.59; Type 2-3 crypt/villous epithelial area ratio = 1.64, p < 0.0001). Using these individual features, a combined feature machine learning score (MLS) was created to evaluate a set of 28 matched pre- and post-intervention biopsies captured before and after dietary gluten restriction. The disposition of the continuous MLS paired biopsy result aligned with the Marsh score in 96.4% (27/28) of the cohort. CONCLUSIONS: Machine learning classifiers can be developed to objectively quantify histologic features and capture additional data not achievable with manual scoring. Such approaches should be further investigated to improve biopsy evaluation, especially for clinical trials.

Effects of 0.5% and 2.0% Sodium Lauryl Sulfate in Male CD-1 Mice From a 3-Month Oral Gavage Toxicity Study.

The tolerability of single daily gavage doses of 0.5% or 2.0% (wt/vol) sodium lauryl sulfate (SLS) in 11- to 12-week-old male CD-1 mice was evaluated in a study of 3 months in duration. Live-phase, gross necropsy, and histopathologic parameters were evaluated. Mortality of 14% occurred in mice administered formulations containing SLS. Clinical observations in mice administered SLS included abnormal respiration (audible, irregular, and/or labored), swollen abdomen, rough haircoat, hunched appearance, and hypoactivity. Necropsy findings in mice administered SLS consisted of enlarged intestines containing abnormal contents with gas. There were no instances of mechanical gavage-related injury. Histologic evaluation of the respiratory tract revealed injury to the nasal passages and nasopharynx, including, but not limited to, inflammation, exudate, apoptosis/necrosis of epithelium, and atrophy of epithelium or olfactory nerves. Collectively, the data indicated that under the experimental conditions of our 3-month study in male CD-1 mice, once-daily gavage administration of vehicle formulations containing SLS at 0.5% or 2.0% resulted in nasal injury and 14% mortality supportive of gastroesophageal reflux. Sponsors utilizing formulations containing SLS in toxicity studies in CD-1 mice should exclude gastroesophageal reflux as a confounding factor in studies with morbidity or mortality associated with respiratory distress or evidence of aerophagia.

Myostatin Neutralization Results in Preservation of Muscle Mass and Strength in Preclinical Models of Tumor-Induced Muscle Wasting.

Skeletal muscle wasting occurs in a great majority of cancer patients with advanced disease and is associated with a poor prognosis and decreased survival. Myostatin functions as a negative regulator of skeletal muscle mass and has recently become a therapeutic target for reducing the loss of skeletal muscle and strength associated with clinical myopathies. We generated neutralizing antibodies to myostatin to test their potential use as therapeutic agents to attenuate the skeletal muscle wasting due to cancer. We show that our neutralizing antimyostatin antibodies significantly increase body weight, skeletal muscle mass, and strength in non-tumor-bearing mice with a concomitant increase in mean myofiber area. The administration of these neutralizing antibodies in two preclinical models of cancer-induced muscle wasting (C26 colon adenocarcinoma and PC3 prostate carcinoma) resulted in a significant attenuation of the loss of muscle mass and strength with no effect on tumor growth. We also show that the skeletal muscle mass- and strength-preserving effect of the antibodies is not affected by the coadministration of gemcitabine, a common chemotherapeutic agent, in both non-tumor-bearing mice and mice bearing C26 tumors. In addition, we show that myostatin neutralization with these antibodies results in the preservation of skeletal muscle mass following reduced caloric intake, a common comorbidity associated with advanced cancer. Our findings support the use of neutralizing antimyostatin antibodies as potential therapeutics for cancer-induced muscle wasting.

LY2875358, a neutralizing and internalizing anti-MET bivalent antibody, inhibits HGF-dependent and HGF-independent MET activation and tumor growth.

PURPOSE: MET, the receptor for hepatocyte growth factor (HGF), has been implicated in driving tumor proliferation and metastasis. High MET expression is correlated with poor prognosis in multiple cancers. Activation of MET can be induced either by HGF-independent mechanisms such as gene amplification, specific genetic mutations, and transcriptional upregulation or by HGF-dependent autocrine or paracrine mechanisms. EXPERIMENTAL DESIGN/RESULTS: Here, we report on LY2875358, a novel humanized bivalent anti-MET antibody that has high neutralization and internalization activities, resulting in inhibition of both HGF-dependent and HGF-independent MET pathway activation and tumor growth. In contrast to other bivalent MET antibodies, LY2875358 exhibits no functional agonist activity and does not stimulate biologic activities such as cell proliferation, scattering, invasion, tubulogenesis, or apoptosis protection in various HGF-responsive cells and no evidence of inducing proliferation in vivo in a monkey toxicity study. LY2875358 blocks HGF binding to MET and HGF-induced MET phosphorylation and cell proliferation. In contrast to the humanized one-armed 5D5 anti-MET antibody, LY2875358 induces internalization and degradation of MET that inhibits cell proliferation and tumor growth in models where MET is constitutively activated. Moreover, LY2875358 has potent antitumor activity in both HGF-dependent and HGF-independent (MET-amplified) xenograft tumor models. Together, these findings indicate that the mechanism of action of LY2875358 is different from that of the one-armed MET antibody. CONCLUSIONS: LY2875358 may provide a promising therapeutic strategy for patients whose tumors are driven by both HGF-dependent and HGF-independent MET activation. LY2875358 is currently being investigated in multiple clinical studies.

Tasisulam sodium, an antitumor agent that inhibits mitotic progression and induces vascular normalization.

LY573636-sodium (tasisulam) is a small molecule antitumor agent with a novel mechanism of action currently being investigated in a variety of human cancers. In vitro, tasisulam induced apoptosis via the intrinsic pathway, resulting in cytochrome c release and caspase-dependent cell death. Using high content cellular imaging and subpopulation analysis of a wide range of in vitro and in vivo cancer models, tasisulam increased the proportion of cells with 4N DNA content and phospho-histone H3 expression, leading to G(2)-M accumulation and subsequent apoptosis. Tasisulam also blocked VEGF, epidermal growth factor, and fibroblast growth factor-induced endothelial cell cord formation but did not block acute growth factor receptor signaling (unlike sunitinib, which blocks VEGF-driven angiogenesis at the receptor kinase level) or induce apoptosis in primary endothelial cells. Importantly, in vivo phenocopying of in vitro effects were observed in multiple human tumor xenografts. Tasisulam was as effective as sunitinib at inhibiting neovascularization in a Matrigel plug angiogenesis assay in vivo and also caused reversible, non G(2)-M-dependent growth arrest in primary endothelial cells. Tasisulam also induced vascular normalization in vivo. Interestingly, the combination of tasisulam and sunitinib significantly delayed growth of the Caki-1 renal cell carcinoma model, whereas neither agent was active alone. These data show that tasisulam has a unique, dual-faceted mechanism of action involving mitotic catastrophe and antiangiogenesis, a phenotype distinct from conventional chemotherapies and published anticancer agents.

LY317615 decreases plasma VEGF levels in human tumor xenograft-bearing mice.

Angiogenesis plays an important role in tumor growth. Angiogenic growth factors may be useful as biomarkers of antiangiogenic activity since their plasma concentrations correlate with the efficacy of treatments directed toward angiogenic targets. SW2 small-cell lung carcinoma (SCLC), Caki-1 renal cell carcinoma and HCT-116 colon carcinoma tumors produce measurable plasma VEGF, bFGF and TGFbeta in nude mice. Mice bearing these human tumor xenografts were treated orally twice daily with the PKCbeta inhibitor, LY317615 (days 14-30 for SW2 and HCT116, and days 21-39 for Caki-1). Plasma was collected every 3 days from control and treated mice. LY317615 significantly decreased plasma VEGF levels in mice bearing SW2 SCLC and Caki-1 renal cell carcinoma compared to control plasma concentrations beginning 5-7 days after initiating therapy. VEGF plasma levels remained suppressed after termination of LY317615 treatment and for the duration of the study (an additional 2 to 3 weeks). Plasma VEGF levels in mice bearing HCT116 xenografts were not altered by LY317615 treatment and plasma bFGF and TGF-beta were not altered by LY317615 in any of the animals. As shown by CD31 immunohistochemical staining, LY317615 decreased intratumoral vessel density by nearly 40% in all three tumors. Only the Caki-1 tumor responded to single-agent LY317615 therapy with a measurable tumor growth delay. Thus, unexpectedly inhibition of PKCbeta in vivo led to decreased VEGF production that persisted after therapy as well as to decreased intratumoral vessels. Plasma VEGF was a weak marker of response to LY317615, and plasma bFGF and TGFbeta were not markers of LY317615 activity.