Martha M. Moore, Ph.D. Director, Division of Genetic and Reproductive Toxicology

1. Evaluation of a potential mutagenic MOA based on analysis of the weight of evidence and using the modified Hill criteria Martha M. Moore, Ph.D. Director, Division of Genetic and Reproductive Toxicology National Center for Toxicological Research Food and Drug Administration Jefferson, Arkansas Disclaimer: Not FDA Policy

2. Topics • Mutagenic mode of action (How high a burden of proof??) • Overview of genetic toxicology assays—how to weigh the evidence • Strategy for using in vivo mutation for MOA assessment (modified Hill Criteria) • Hazard identification vs. mode of action

3. Three Steps (Questions) To Determine if Mutagenic MOA for Tumor Induction • Potential mutagen? Hazard Identification (Evaluated using standard mutation assays) • In vivo rodent/human mutagen? (Mutagenic in the target tissue?) • Is mutation a “key” event in the development of the tumor?

4. Weigh the Evidence (MOA)How High a Burden of Proof? Mutagenic MOA Nonmutagenic MOA Mixed MOA

5. How to weigh the evidence as to whether a chemical causes specific tumors by a mutagenic mode of action (Mutation is THE key event) (Listed in decreasing order of relevance/importance) Cancer relevant oncogene/tumor suppressor gene mutations can be detected in the target tissue following chemical exposure Surrogate gene mutations can be detected in the target tissue following chemical exposure 3. DNA adducts (known to be mutagenic adducts) can be detected in the target tissue following chemical exposure Primary DNA damage can be detected in the target tissue following chemical exposure Gene mutations and/or DNA adducts or other measures of primary DNA damage can be detected in vivo. 6. Evidence that the chemical can induce mutations, cytogenetic damage, DNA adducts and/or primary DNA damage in vitro.

6. Types of Mutagenic Damage Point mutations Deletions Translocations Recombination Gene Conversion Amplification Aneuploidy

7. Definitions: Genotoxicity: Damage to the DNA or chromosome. Interactions with DNA (DNA adducts). Mutagenicity: Damage to the DNA or chromosome structure (or number) that is passed to daughter cells. Generally, the key question is whether the test agent is a mutagen

8. Assays that detect genotoxicity but not mutagenicity (primary DNA damage) DNA adduct analysis DNA strand breakage Unscheduled DNA synthesis (UDS) Sister chromatid exchange These assays indicate whether the test agent has the capability to reach and interact with DNA or chromosomes Positive in vivo data more significant than in vitro.

9. Assays that detect genotoxicity but not mutagenicity • Chromosome aberration analysis • Micronucleus analysis Positive in vivo data more significant than in vitro--however a negative in vivo does not automatically out weigh a positive in vitro

10. Some general considerations about chromosomal aberration assays • Damaged chromosomes must replicate and cell division must progress to • metaphase to generate a visible chromosomal aberration (CA) • cell division to generate a micronucleated erythrocyte • Breakage events leading to CAs and MN are generally lethal, and cells that contain them are rapidly eliminated from dividing cell populations • Most visible CAs and MN are not mutations, but the events that generate them also generate mutations

11. Assays that detect mutagenicity • Ames Test • Mouse Lymphoma Assay (MLA) • HPRT mutation in various cell lines or in vivo lymphocytes • Transgenic rodents (Big Blue, GPT/Delta) Positive in vivo data more significant than in vitro--however a negative in vivo does not automatically out weigh a positive in vitro

12. Standard Genetic Toxicology Battery Ames Test or other bacterial assay Mouse Lymphoma Tk assay or in vitro cytogenetics In vivo cytogenetics

13. Genotoxicity Assays MeasureDifferent Genetic Events(One should not expect to obtain the same results with all assays) Mutation Ames MN Chrom Abs MLA Point Mutation yes no noyes Insertion/Deletion yes no no yes Allele Loss no no no yes Multi-locus Mutation no no no yes Small Chrom. Damage no yes no yes Large Chrom. Damage no yes yesyes Aneuploidy no yes yesyes

14. Weight-of-evidence assessment • Need to assess each genotoxicity experiment for quality/consistency of data (just because it is published does not mean that the data is valid). • Need to integrate the data from the various assays based upon an understanding of the assays (what types of events they do and do not detect).

15. Mutagen = Mutagenic Carcinogen • Cancer is a multi-stage, multiple event disease • Mutation is involved in the etiology of cancer • Cancer is a disease of mutation and cell division • Increased mutations can occur by the induction of new mutations or by increasing the “spontaneous” mutation rate

16. Genotoxicity data informs MOA for cancer risk assessment Key Question: Does the agent mediate the induction of tumors via a mutational mode of action?

17. MOA Evaluation Should involve • Assessment of mutation in the target tissue • Time to mutation (temporality) • Dose response concordance (Mutation and tumor induction)

18. Multi-Stage Tumor Induction Requires the Accumulation of Events over Time (Many/most rodent carcinogens require long chronic exposure) Mutagenic Carcinogen Multiple events Initiating Mutation Tumor Nonmutagenic Carcinogen Initiating Mutation Multiple events Toxicity/ Cell Proliferation Tumor

19. Temporality—time-to-mutation vs time-to-tumor Predictions Mutagenic carcinogens would be expected to show a positive mutation response after relatively short treatment periods Mutant Frequency Time in Weeks Nonmutagenic carcinogens would be expected to be negative after long chronic treatment, or show a positive response only after long chronic treatment

20. Dose Response Concordance: Predictions If mutation is THE key event: The mutation dose response will lead the tumor dose response If the tumor dose response leads the mutation dose response or the mutation response is negative after long chronic exposures Consistent with mutation is not the key event

21. How do you evaluate dose response concordance?(A work in progress) • Visual inspection of the curves • Quantitative modeling to compare mutation and tumor response • Benchmark dose (BMD) (single dose assessment) • Compare the probability of an “adverse” MF response with the probability of tumor response (dose response curve assessment)

22. Unfortunately most of the available in vivo mutation studies were conducted for hazard identification—not optimally designed for MOA evaluation We present 2 case studies that can be used to demonstrate the general approach • Riddelliine (consistent with mutagenic MOA) • Dichloroacetic Acid (DCA) (consistent with nonmutagenic MOA)

23. Case study 1: Visual inspection and BMD analysis for riddelline: (Consistent with mutagenic MOA) BMD10(M)=0.16 BMD10(C)=0.39

24. Case study 2: Visual inspection and BMD analysis for DCA (60-week exposure for MF) (Consistent with nonmutagenic MOA) BMD10(C)=0.22 BMD10(M)=0.97

26. General Features of the Hazard Identification Design • Acute exposure (single or small # of doses) • IWGT recommended design is 28 days treatment and 3 days for mutant manifestation • Generally a negative control and 2 doses • Generally the high dose is the MTD and the low dose is approximately ½ the top dose

27. MOA Basic Design • Same species/strain as cancer study • Same exposure route as cancer study • Multiple doses (6 or 7 or more) based on tumor data--Enough doses to adequately “define” the dose response curve • Chronic treatment ( up to 12 months) modeled on the tumor bioassay • Interim analysis of MF (to define time-to-mutation) • Detection of mutation in the target tissue(s) • Evaluate the dose response concordance

28. Bottom-line Questions for MOA Assessment • What happens? • When does it happen? • At what dose does it happen? • This information can then be used to develop a timeline for the various events and also to understand the dose response curves for the various events

29. Conclusions • A weight of the evidence assessment of all of the gene-tox data is useful for MOA • The most important question is whether the chemical induces mutation in the target tissue • Future research to assess mutation as a key event should use a MOA design rather than hazard ID design • Further work needed to assess dose response concordance

30. Acknowledgements • Robert Heflich, NCTR • Mugamine Manjanatha, NCTR • Lynne Haber, TERA