APPLYING PRE-CLINICAL DATA TO CLINICAL STUDIES-I

1
APPLYING PRE-CLINICAL DATA TO CLINICAL STUDIES-I

• Edward A. Sausville, M.D., Ph.D.
• Developmental Therapeutics Program
• National Cancer Institute
• October 17, 2002

2
GOALS OF PRECLINICAL DRUG STUDIES
Regulatory framework

• IND Investigational New Drug application
  approval by FDA to conduct human studies main
  criterion SAFETY AND LIKELY REVERSIBLE TOXICITY
  allows start of Phase I trials
• Pediatric Phase I Oncology Drugs Special Issues
• Few (thankfully) patients many agents
• Unmet medical need ethical concerns
• Prior Rx Concomitant meds
• Unique pediatric biology (tumor and host) vs.
  
• value of adult data in study design

3
PEDIATRIC PHASE I STUDIES CLASSICAL NCI
RECOMMENDATIONS

• Begin in pediatric patients with solid tumors
  and
• leukemias at 80 of max tol dose (MTD) in
  adults with solid tumors
• Solid tumor and leukemia pts at each level
• Escalate in 20 increments distinguish
  myelosuppressive tox (desired in leukemia) vs
  non myeloid tox
• In absence of non-myelo tox, escalate beyond
  
• solid tumor MTD in leukemia patients.

Marsoni et al. Cancer Treatment Reports 69, 1263,
1985
4
PEDIATRIC PHASE I STUDIES BASIS FOR
RECONSIDERATION OF CLASSICAL PRACTICE

• BIOLOGY Pediatric tumors may have targets
  intrinsically different from adults therefore
  adult data will never be available for drugs
  directd to those targets.
• PHARMACOLOGY Past practice weighted toward
  cytotoxics ? Relevance to targeted agents.
• TIMING Many new agents delay in completing
  adult studies before application in peds
  neoplasms therefore exacerbate unmet need.

5
COMPONENTS OF AN IND
The goal of the pre-clinical process

• Form 1571
• Table of Contents
• Intro Statement / Plan
• Investigator Brochure
• Clinical Protocol
• Chemistry, Manufacture, Control

• Pharmacology/ Toxicology
• Prior Human Experience
• Additional Info - Data monitoring, Quality
  Assurance

6
HOW ARE PHASE I DOSE AND SCHEDULE FIXED IN
ADULTS ?

• Animal (usually mouse) model studies define
  likely active
• schedules in animals bearing human-derived
  tumors
• Likelihood of human activity stochastic more
  models
• with activity, greater likelihood of human
  activity
• Limitations difference between animal /
  human metabolism
• Drug concentrations OR effect on target
  provide
• important ancillary info.
• Toxicology according to NCI guidelines / FDA
  requirements
• Starting dose a fraction of dose causing no or
  minimal
• reversible toxic effect escalate dose in
  steps likely to capture
• reversible toxic effect

7
PROBLEMS WITH MTD DRIVEN ENDPOINTS

• Drugs regulating pathways important in
  oncogenesis are effective by combining with high
  affinity binding sites therefore must
  distinguish targeted vs non-targeted toxicity
  related to these binding sites
• Whether dosing beyond effect on desired target
  buys therapeutic value not clear
• Therefore must define in pre-clinical studies
  BIOLOGICALLY EFFECTIVE DOSE and MAXIMUM
  TOLERATED DOSE
• Use BIOLOGIC rather than TOXIC endpoints in
  PhaseI?

8
REGULATORY CONSIDERATIONS FOR PRE-CLINICAL
DEVELOPMENT OF ANTICANCER DRUGS DeGeorge, et
al, CCP (1998) 41 173-185
The types of preclinical studies expected for
support of clinical trials and marketing of a new
drug depends on both the intended use of the drug
and the population of patients being studied and
treated. In situations where potential benefits
are greatest (Advanced, life-threatening
disease), greater risks of treatment toxicity can
be accepted and the required preclinical testing
can be minimal.
9
FDA PRECLINICAL PHARMACOLOGY TOXICOLOGY
REQUIREMENTS

• DRUGS
• Two Species - Rodent Non-rodent
• Clinical Route Schedule
• Follow NCI Guidelines
• Pharmacokinetics - Optional
• BIOLOGICALS
• Most Relevant Species
• Clinical Route Schedule

10
OBJECTIVES OF PRECLINICAL TOXICOLOGY STUDIES FOR
ANTI-NEOPLASTIC DRUGS

• DETERMINE IN APPROPRIATE ANIMAL MODELS
• The Maximum Tolerated Dose (MTD)
• Dose Limiting Toxicities ( DLT )
• Schedule-Dependent Toxicity
• Reversibility of Adverse Effects
• A Safe Clinical Starting Dose

11
NCI STANDARDIZED PRECLINICAL TOXICOLOGY PROTOCOLS
FOR ANTI-NEOPLASTIC AGENTS (1980 - 1988)

• Mouse Lethality Studies
• Dog Toxicity Studies
• Rodent (Rat) Toxicity Studies

• Determine LD10 on Dx1 Dx5 Schedules
• Assess safety of 1/10 LD10 Determine DLTs on
  Dx1 Dx5 Schedules.
• Assess safety of 1/10 LD10 Determine DLTs on
  Dx1 Dx5 Schedules.

12
CURRENT NCI TOX PHILOSOPHY

• Agent-Directed Studies
• Pharmacologically- Guided
• Integrate With Preclinical Efficacy Data the
  Proposed Clinical Protocol
• Rational Evaluation of Role of Schedule
  Dependence, Pharmacokinetics Metabolism in the
  Development of Toxicity
• Relate Plasma Drug LeveLs and/or AUC to Safety
  and to Occurrence Severity of Toxicity
• Extrapolate Toxic Effects Across Species

13
AGENT-DIRECTED versus STANDARD PROTOCOL DRUG
DEVELOPMENT

• Greater Scientific Basis for Development
• Permits Greater Flexibility
• Data Rich IND Submission to Support Phase I
• Preclinical Potential Less Expensive
• Permits PK-Guided Dose Escalation in Phase I
• Clinical Trial Potential Shorter, Less
  Expensive
• Optimal Schedule Greater Chance of Success?
• Patients Greater Chance of Effective Therapy?

14
(No Transcript)
15
(No Transcript)
16
(No Transcript)
17
BENZOYLPHENYLUREA PRECLINICAL MTD DLTs
Starting Dose 24 mg/m2
18
ADDITIONAL AGENT-DIRECTED TOXICOLOGY STUDY
REQUIREMENTS

• Attain Efficacious Drug Levels in Plasma In Vivo
• Correlate Drug Plasma Levels and/or AUC with
  Toxicity and Safety Across Species
• Ameliorate Toxicity by Change in Route/Schedule
• Compare Toxicity with Accepted Clinical Agents
  as Necessary

19
SCHEDULE and ROUTE versus TOXICITY

• Pyrazoloacridine Bolus iv Neurotoxicity
• 1 Hr civ Bone Marrow
• Penclomedine Bolus iv Neurotoxicity
• 1 Hr Dx5 BM ( Neuro)
• 5 Hr civ Neuro Death
• Oral BM
• O6 Benzylguanine Bolus CNS, HR ?
• Infusion Neutrophilia

20
HOW PREDICTIVE OF HUMAN EXPERIENCE ARE THESE
SAFETY-TESTING ALGORITHMS?

• Predictive power varies with endpoint desired
• - 97 safe starting dose if 2-3 species (rodent
  plus dog)
• - 83 mouse only
• (Smith, A.C. Proc AACR 35 459)
• BUT Human MTD accurately predicted by
• -mouse in 36
• -rat in 19
• -dog in 44
• NO in vitro or silico methodology has yet
  emerged to
• predict human toxicity or dosing scheme (?marrow)

NCI data
21
CONCLUSIONS FROMILSI HUMAN TOXICITY PROJECT DATA

• 71 (151/214) human toxicities(HTs) associated
  with
• toxicities in animals
• - 63 in non-rodents
• - 36 in rodent plus non-rodent combinations
• - 43 in rodents
• 29 of human toxicities not predicted for by
  animals
• (8HTs insufficient animal data)
• Two species best predictor
• for single species non-rodent (dog and/or
  primate)
• better than rodent (mainly rat)

SOT.BiomarkersSymposium Regul. Toxicol. 32 56,
2000
22
CONSIDERATIONS IN APPLYING DATA TO PEDS
POPULATIONS

• How closely do adult / peds MTDs correspond?
• cytotoxics?
• noncytotoxics
• Are classical MTDs still relevant?
• Age / maturity / nature of tox species
  relevant
• if adult human phase I data not to drive
  peds dosing?
• Importance of efficacy model PK / PD?
• Chronicity / reversibility / age-relatedness of
• target related toxs (e.g.s)
• bone growth and anti-VEGFRs?
• skin (anti-EGFR)

23
ACKNOWLEDGEMENTS
J. Covey M. Hollingshead A. Smith J.
Tomaszewski S. Decker