Scott Isom

1
An Overview of Phase I and II Clinical Trial
Designs for Cancer Research

• Scott Isom
• Biostatistics Core
• Comprehensive Cancer Center
• April 2, 2009

2
What is a clinical trial?

• A definitive tool for evaluation of the
  applicability of medical research
• A key research activity with the potential to
  improve the quality of health care and control
  costs though careful comparison of alternative
  treatments.
• ClinicalTrials.gov
• Biomedical or health-related research studies in
  human beings that follow a pre-defined protocol.
• Types treatment, prevention, diagnostic,
  screening, QOL

3
Cancer Center Trial Phases

• Phase I
• Dose Finding/Toxicity
• Phase II
• Safety/Efficacy
• Phase III
• Comparative Treatment Efficacy

4
Phase I - Definition

• ClinicalTrials.gov
• Initial studies to determine the metabolism and
  pharmacologic actions of drugs in humans, the
  side effects associated with increasing doses,
  and to gain early evidence of effectiveness may
  include healthy participants and/or patients.
• Often first study done in humans and have small
  sample sizes

5
Phase I - Purpose

• Determine the Maximum Tolerated Dose (MTD)
• Assess pharmacology of drug/treatment
• Evaluate toxicities
• Evidence of antitumor activity (exploratory)

6
Phase I Patients

• Have a particular disease of interest
• Fairly advanced disease (often have disease that
  is unresponsive to standard treatments)
• Very sick usually excluded

7
Phase I Sample Design

• 33 dose escalation
• Evaluate 3 subjects at a dose
• A1) If 0 of 3 have DLT, then increase to next
  dose step 1
• B1) If 1 of 3 have DLT, then go to step 2
• C1) If gt1 of 3 have DLT then go to step 3
• Evaluate an additional 3 patients at same dose
• A2) If 1 of 6 have DLT, then increase to next
  dose step 1
• B2) If gt1 of 6 have DLT, then go to step 3
• Discontinue dose escalation
• When trial is stopped, then the dose level below
  that at which excessive DLT was observed is the
  MTD.
• MTD is then used as the dose in the Phase II trial

8
Phase I Determining Initial Dose

• New drug/treatment
• 1/10 the LD10 (lethal dose 10 dose causing
  toxicity in 10 of animals but not death if dose
  is doubled)
• 1/3 TDL (toxic dose low dose causing toxicity
  in animals but not death if dose is doubled)
• Drug used previously
• Conservative estimate based on other studies

9
Phase I Dose escalation
Fibonacci Numbers - 1, 1, 2, 3, 5, 8, 13, 21, etc
Fibonacci Scheme 100 200
300 500 800 1300
Modified Fibonacci 100
200 330 500
700 900
Standard Increments 100
150 200 250
300 325
10
Phase I Limitations/Concerns

• Small sample size
• With 33 design, response at a dose would have,
  at most, 6 subjects.
• Difficult to conduct in multicenter setting
• The MTD chosen is what will be used in the Phase
  II trial

11
Phase II - Definition

• ClinicalTrials.gov
• Controlled clinical studies conducted to
  evaluate the effectiveness of the drug for a
  particular indication or indications in patients
  with the disease or condition under study and to
  determine the common short-term side effects and
  risks.

12
Phase II - Purpose

• Evaluate biologic activity
• Short term endpoints (tumor response)
• Most designs have a binary response variable
• example gt50 decrease in tumor burden
• Estimate the rate of adverse events
• Other possible endpoints
• Survival
• Time to progression
• Time to recurrence

13
Phase II - Patients

• Patients with a particular disease
• Use explicit eligibility criteria
• Patients with few prior treatments, good
  performance status, measurable disease
• Typically between 14-50 patients are enrolled to
  see how many have a response

14
Phase II - Design

• Fixed Design decision after fixed number of
  patients accrued (e.g., rlt30 no further testing
  vs rgt50 further testing. Then accrue 38
  patients if 15 or fewer respond (39) then no
  further testing. If response is really 30, the
  probability of no further testing is 92 if it
  is really 50, the probability of further testing
  is 87)
• -Simplest
• -Not the most efficient or the most ethical

15
Phase II - Design

• Two-Stage Design decisions made after groups of
  patients accrued (Previous example accrue 22
  patients, if 7 or fewer respond then no further
  testing if gt7 respond, accrue additional 24
  patients. If 17/46 (37) or fewer respond then
  no further testing, otherwise go on to phase III
  trial.)
• -More efficient than fixed design
• --Most Common
• Simon two-stage design, Min-Max design

16
Phase II - Design

• Randomized designs
• - used to decide which of several new regimens
  should be taken to the next phase of testing.
• - not designed for a statistical comparison
  of the regimens, but rather to determine which
  regimen seems the best for further study

17
Phase II - Limitations

• Dependent upon the MTD from the Phase I trial
• If does was too low, it will seem ineffective
• Too high and too many adverse events may occur
• Usually single arm studies
• Must be compared to historical controls

18

• QUESTIONS?