Eurotox 2018 – Scientific Activities

symposium registration

Lunch industry symposium:
Advances in immunological safety evaluation of immunomodulatory drugs


Monday, September 3rd – 13.00 to 14.00


The SQUARE – Rue Mont des Arts – 1000 Brussels, Belgium
Room: “The Arc”


Immunomodulatory drugs are increasingly being developed for the treatment of inflammatory disorders, auto-immune diseases, cancer or organ transplant rejection.

These compounds alter signalling pathways involved in the immune response and thus can exert immunotoxic effects mainly by exaggerated pharmacodynamics.

Nowadays, immunomodulators tend to affect more specific and narrow pathways in the immune system, sometimes making it more difficult to predict any potential immunological safety risk in preclinical models.

During this symposium, the first presentation will show two case studies for the infection risk evaluation associated with treatment by immunosuppressant compounds, taking into account human genetic data as well as data from preclinical models.

The second presentation will show recent developments of the T-cell dependent antibody response (TDAR) model used to evaluate immunostimulation in non-human primates using sub-optimal doses of model antigens such as KLH or multiple antigens. The incorporation of multiple immunomonitoring end-points in non-human primate studies allows for a comprehensive monitoring of various aspects of the immune response involved, and can be used for compounds with various mechanisms of action.


“Evaluation of infection risk with immunomodulators.”
M. Kammueller – Novartis

“Use of model antigens to evaluate T-cell dependent immune stimulation in non-human primates”
P. Ancian – Citoxlab


Scientific posters

P02-10 • Effects of phenobarbital (PB) acute exposure on minipig liver gene expression

J. Silvano, P. Ancian, C. Bansard, J. Nielsen, L. Richert, S. Pramila and F. Roy

Presenting author: Jérémy Silvano

Phenobarbital is a widely used anti-seizure medication. Treatment of laboratory animals with PB results in hepatomegaly and induction of CYP450 xenobiotic metabolizing enzymes, and long term treatment of rats and mice results in hepatic tumors. While PB has been extensively studied in rodents, little is known about the impact of PB treatment on the gene expression profile in minipig liver. For that purpose, 3 male Göttingen minipigs aged 4-5 months were treated orally with vehicle or 15 mg/kg/day PB for 6 days. On day 7, animals were euthanized and an 80 mg section of liver was snap frozen in liquid nitrogen. Total RNA was extracted from liver samples using a combined Trizol/RNeas® method and showed adequate quality for Affymetrix target preparation using the 3’-IVT Plus kit. Labeled material was then hybridized to Affymetrix GeneChip® Porcine Arrays. The differentially expressed genes were filtered using an absolute fold-change threshold at 1.5 with a corrected p-value lower than 0.05. A total of 161 and 139 probesets were found to be significantly down and upregulated in minipig liver by PB treatment, respectively. CYP2A19, CYP2B22, CYP2C42, CYP3A39 and CYP3A46 transcripts were up-regulated with mean fold changes varying between 31.2 and 1.5. Among the top 20 most regulated genes, CYP2A19, CYP2B22, FADS2, tsukushin-like, FADS1, HMGCS2, and SULT1A1 were up-regulated and SDS, GARNL3, CLDN14, acetoacetyl-CoA synthetase-like were down regulated. Pathways and ontology classification using the DAVID software revealed that the oxidoreductase, retinol metabolism, chemical carcinogenesis and steroid hormone biosynthesis pathways were significantly enriched, with modulated genes such as CYPs, SULT1A1 (Sulfotransferase Family 1A Member 1), as well as fatty acid desaturases FADS1 and FADS2. These data will be correlated with additional analyses such as CYPs and UGT expression and ex-vivo enzyme activities in liver microsomes. In conclusion, this study aids understanding the mechanisms/modes of action of PB exposure on minipig liver and the identification of potential biomarkers of exposure and response.

Keywords: Phenobarbital, liver, minipig, Gene Expression profiling, CYP450


P16-35 • Hepatic effects of phenobarbital in the minipig

J. B. Nielsen, J. Decorde, C-A. Erratico, C. Parmentier, M. Untrau, C. Bansard, P. Ancian, M. Fonsi,
L. Richert, R. Forster and P. Singh

Presenting author: Jonas Boje Nilsen

Rodent models are often used to study biochemical mechanisms of toxicity and to establish a mode of action (MOA) for adverse, chemical-induced effects. Phenobarbital (PB) has been shown to induce differential gene expression and CYP2B, CYP3A and UGT enzymes in the rodent liver leading to increased thyroid hormone clearance and subsequent stimulation of cell proliferation in the thyroid that eventually produces thyroid tumors after long-term exposures. This pathway has not been completely investigated in non-rodents. In this study, male Gottingen minipigs were orally administered 15 mg/kg/day PB for 6 days followed by investigation of phase I and II liver enzymes (mRNA expression and enzyme activities), circulating thyroid hormone levels and differential gene expression in the liver, in addition to standard toxicological evaluations. PB-administered minipigs had 46% increased absolute and 42% increased relative liver weights with mild, diffuse hepatocellular hypertrophy versus the control group. PB reduced plasma concentrations of T3 (by about 27% and 47%) and T4 (by about 15% and 20%) versus predose values at 24 hours and after 6 days of administration, respectively. PB induced a 2-fold increase in CYP3A429 and CYP4A24, 5-fold increase in CYP1A2 and around 11-fold increase in CYP2B22 compared to the control group, with increases in mRNA expression of these genes. A 1.6-fold increase in T4-specific UGT phase II enzyme activity was observed in the PB-administered group, and CYP2B22, CYP2A19, CYP2C42, CYP3A39 and CYP3A46 were among the top 20 upregulated genes. This study indicates that several of the early biochemical key events in the PB-induced (rodent) MOA are present in minipigs. It is concluded that the minipig is an interesting model for further study of the effects of xenobiotics on the liver-thyroid axis.

Keywords: Mechanisms of Toxicity, Alternative Animal Models, Target Organ Toxicity

P02-12 • Effects of rapamycin on in vivo and ex vivo cell-mediated immune responses in cynomolgus monkeys

P. Ancian, E. Grosdidier, A. Nguyen, S. Sarlang, F. Gervais and J. Descotes

Presenting author: Philippe Ancian

Inconsistent delayed-type hypersensitivity (DTH) responses have been reported in non-human primates, hence the need to study alternative endpoints. For this purpose, 3 male and 3 female purpose-bred Cynomolgus monkeys aged 33-36 months received a human dose of tetanus vaccine containing aluminum hydroxide IM on days 1 and 14, and were treated orally with 1 mg/kg/day rapamycin (RPM) from days 36 to 66. DTH was induced by an intradermal challenge of tetanus toxoid (TTx) and aluminum hydroxide on days 28 and 63. Injection sites were observed before challenge and after 24, 48 and 72 hours. Then, skin biopsies were taken and stained with anti-CD3, CD4, CD8, CD68, Ki67, FoxP3 antibodies. Blood was collected on days 0, 28, 31 and 35 to measure anti-TTx IgG using ELISA; on day 0 and then weekly, to study standard hematology; finally, at pre-test and on days 28, 35, 59 and 63, to quantify IFN-g secreting cells by ELISPOT and cytokine levels using a multiplex assay. No clinical signs were noted. Hematological findings were consistent with known effects of RPM. RPM blood levels measured by LC-MS/MS were within the immunosuppression range. The expected anti-TTx humoral response was seen in all animals prior to RPM treatment, and a 3-8 fold drop in anti-TTX IgG levels was measured after 4 weeks of RPM treatment. DTH reactions in animals prior to RPM treatment were slight and inconsistent and no DTH inhibition could be observed after RPM treatment. CD4+ and CD8+ T-cells, and CD68+ macrophages infiltrates at the injection site were observed after the first DTH reaction, and markedly decreased after RPM treatment. Despite inter-animal variability, the number of anti-TTX IFN-g secreting cells surged after the first DTH response, and markedly declined at the end of RPM treatment. No clear trend in any cytokine level was noted after the first DTH response, but all levels declined after RPM treatment. This study confirms that DTH in Cynomolgus monkeys requires further study and that a better understanding of DTH related events may be gained using ex vivo endpoints (cytokine analysis, immunohistochemistry etc), and in particular IFN-g ELISPOT.

Keywords: Non-human primate, tetanus toxoid, delayed-type hypersensitivity, rapamycin, immunosuppression

P16-32 • A human relevance investigation of PPARα-mediated key events in the hepatocarcinogenic mode of action of propaquizafop in rats

P. Singh, W. H. Bomann, F. Spezia, F. Gervais, R. Forster, L. Richert and C. Strupp

Presenting author: Pramila Singh

Propaquizafop is an herbicide with a long history of use. Chronic dietary exposure of rats led to liver tumors without genotoxicity, subchronic studies demonstrated dose-dependent increases in liver weights with hepatocellular hypertrophy, and two weeks administration induced liver CYP4A and peroxisomal enzymes. A rodent-specific mode of action (MOA) via activation of the peroxisome proliferator-activated receptor α (PPARα), increased liver peroxisomal activity, hypertrophy, liver enlargement, subsequent cell proliferation and finally adenoma formation was postulated by others based on the alignment of these findings with an established MOA for fibrates. Experience with PPARα-inducing pharmaceuticals indicates a lack of human relevance. In this study, the results of a MOA investigation following two weeks of dietary feeding in wildtype (WT) and PPARα-knockout (KO) rats are presented to confirm the dependency of key events following propaquizafop administration on PPARα. In WT animals dose-dependent liver weight increases (65-84%), liver CYP4A (20 fold) and acyl-CoA oxidase (10-15 fold) induction, hepatocellular hypertrophy and proliferation were observed, while in KO rats only a marginal increase in liver weight (24%) was observed without any effect on the other parameters confirming PPARα-dependency and sequelae comparable to a well-understood hepatocarcinogenic MOA in rodents. Based on an evaluation of the data according to the MOA-Human Relevance Framework, we conclude that liver tumors observed in rodents after long-term dietary administration of propaquizafop do not pose a relevant health risk to humans.

Keywords: Propaquizafop, carcinogenicity, hepatocarcinogenic, PPARα, liver, peroxisome proliferators, mode of action, human relevance framework


P19-18 • Rotational thromboelastometry for ex vivo assessment of drug effects on platelet function

S. Authier, P. Chang, L. F. N. Silva, J. Bakke, J. Gahagen, K. Wong and C. St-Jean

Presenting author: Simon Authier

High sensitivity assays to identify pro or antithrombotic effects are critical in preclinical safety testing considering the life-threatening consequences of drug-induced thrombosis or hemorrhage. This study evaluated the sensitivity of a global coagulation assessment platform, to detect alterations in platelet response, clot formation and thrombus lysis ex vivo. Blood was obtained from Cynomolgus monkey and human normal volunteers. Rising concentrations or inhibitors, such as heparin and acetylsalicylic acid, and platelet activator, such as collagen, were added to the blood samples, ex vivo. Viscoelastic properties of the clot formation were assessed by thromboelastometry (ROTEM®, for 1.5 hours after recalcification of citrated whole blood with calcium chloride (Star-tem reagent). Platelet response in whole blood decreases abruptly after 2 hours post collection. In Cynomolgus monkeys, mean coagulation time (CT) increased by 17%, two hours post collection while clot formation time (CFT) increased by 51% and maximum clot firmness (MCF) decreased by 9% in the same interval. The results after 4 hours were similar to those at 2 hours. Because of the inter-individual variations observed, the results were presented as the % of change from an untreated baseline, which was the value obtained without any agonist or inhibitor. With this approach, inter-individual variations were significantly reduced. With rising collagen concentrations (0 µg/mL, 1 µg/mL, 2 µg/mL, 5 µg/mL), the CT was shortened by 23%, 31% and 36% with collagen concentrations of 1, 2 and 5 µg/mL, respectively. Signs of saturation were observed at higher collagen concentrations. Collagen had no significant impact on the CFT and the MCF. As expected, heparin at pharmacologically relevant concentrations completely inhibited clot formation at concentrations of 1 units/mL or above. However, at a lower concentration of 0.1 units/mL, CT was extended by 148%, CFT was prolonged by 454% while MCF was reduced by 32%. As previously reported, acetylsalicylic acid (0.1 mg/mL, 1 mg/mL, 10 mg/mL) did not significantly affect rotational thromboelastometry. This study reports ex vivo platelet response characterization using rotational thromboelastometry as a strategy for pharmacological evaluations of drug effects using non-human primates and human blood.

Keywords: thromboelastometry, drug effect, platelet function, ex vivo assessment, antothrombotic effect, coagulation, non-human primates


P12-43 • Evaluation of the dose-response and fate in the lung and pleura of chrysotile containing brake dust compared to chrysotile or crocidolite asbestos in a 28 day inhalation toxicology study

B. Tóth, D. M. Bernstein, R. A. Rogers, P. Kunzendorf and H. Ernst

Presenting author: Balázs Tóth

This study was designed to provide an understanding of the biokinetics and potential toxicology in the lung and pleura following inhalation of brake dust (from brakes manufactured with chrysotile) in a 28-day repeated multi-dose inhalation toxicology study (6 h/d, 5 d/wk. 4 wk) followed by 28-days of exposure free recovery and served as a range finding study for a subsequent ongoing 90-day repeated dose inhalation toxicology study with lifetime recovery. Comparative fiber control groups were included of a similar grade commercial chrysotile as used in the brakes and a commercial crocidolite asbestos sample. The aerosol fiber distribution of the chrysotile (chry) and crocidolite (croc) asbestos were similar (Fibers L>20 µm/cm3: Chry-LD 42, Chry-HD 62; Croc-LD 36, Croc-HD 55; WHO fibers/cm3 Chry-LD 192, Chry-HD 219; Croc-LD 211, Croc-HD 255). The total number of particles in the aerosol of the brake dust groups was similar to that in the chrysotile groups (Part/cm3: B-D 710 – 1065; Chry 532 – 1442). A macrophage dose-response to the brake dust groups was observed with no fiber related effects. In the fiber control groups, the study differentiated between similar exposures to chrysotile and crocidolite asbestos. The chrysotile exposure resulted in a dose-dependent particle laden macrophage response characterized as Wagner Grade 1 to 3. In contrast, following crocidolite exposure there was a dose-response accumulation of fiber laden macrophages and interstitial fibrosis during exposure (Wagner score Grades 3 to 4) which increased in incidence following exposure (Wagner score Grade 4). Confocal microscopy images (determined following deep freezing of the chest wall at sacrifice) revealed no difference between the air control, brake dust and chrysotile high-dose exposure groups and no difference in the visceral or parietal pleura thickness at 28 or 56 days. The crocidolite exposure resulted in more than double the thickness of the visceral pleura and parietal pleura, with associated extensive inflammatory response and collagen development observed including adhesions between the visceral and parietal surfaces. These results provided a basis for the design of the 90 day study. (Study funded by Honeywell International Inc).

Keywords: respiratory toxicology, inhalation toxicology, lung, pulmonary, respiratory system, Brake dust, Chrysotile, Crocidolite

P25-21 • J-Tp and Tp-e as biomarkers of proarrhythmic risk in nonclinical models: historical data evaluation by the HESI consortium

O. Foulon, S. Authier, M. Abernathy, E. Boulay, R. Chui, G. S. Friedrich, N. Gendron-Parra, Greiter-Wilke, J-M. Guillon, D. Leishman, J. Nichols, J. Pierson, M. Skinner, M. Pugsley, J-P. Valentin, H. Vargas and T. Wisialowsky

Presenting author: Olivier Foulon

A comprehensive analysis of T-wave morphology changes has recently emerged as a testing strategy to evaluate drug-induced ion channel blockade and proarrhythmic risk using ECG data from clinical trials. This retrospective study assessed the JTpeak and Tp-e intervals as novel non-clinical proarrhythmia biomarkers. Lead II ECG telemetry data was analyzed from Beagle dogs and monkey studies by members of the HESI consortium using pattern recognition methods with various software platforms. JTpeak and Tp-e duration was also quantified in groups given known drugs. Baseline JTpeak and Tp-e were HR dependent in both species and individual rate correction methods were evaluated as JTpeak corrected (JTpc) and Tp-e corrected (Tp-ec). Drugs such as sotalol, dofetilide, thioridazine, cisapride, quinine and moxifloxacin were generally associated with an increase in JTpc similar to QTc interval prolongation. Drugs such as captopril, verapamil, nifedipine, minodixil and milrinone were generally associated with minimal effects or shortening of JTpc. The anesthetic medetomidine decreased body temperature, increased JTpc in monkeys; however, shortened the Tp-ec interval. Atenolol and pimobendan showed no notable effect on these two QT subinterval biomarkers. Both JTp-c and Tp-ec demonstrate promise as complementary biomarkers to drug-induced QT interval changes and may play a role in proarrhythmia risk assessment. However, experimental methods and factors that influence variability of these ECG parameters in non-clinical species need to be evaluated. Evaluation of a broad range of control drugs to validate these measures is required to establish their translational and discriminative value in drug safety pharmacology studies.

Keywords: proarrhythmic risk, ECG, JTp and Tpe intervals, proarrhythmia biomarkers, QTc prolongation