Choosing Appropriate Clinical Endpoints

1
Choosing Appropriate Clinical Endpoints

• Neil Clendeninn
• Mark Krailo
• Manuel Hidalgo
• Col (Ret) James Williams

2
THE ULTIMATE ENDPOINT IN CLINICAL TRIALS

• Survival
• Sometimes, clinical benefit (Surrogate endpoints
  are usually used)

3
Concepts/Definitions proposed by the NIH Workshop
for Evaluation of Surrogate Endpoints

• Biomarker a characteristic that is objectively
  measured and evaluated as an indicator of normal
  biologic processes, pathogenic processes, or
  pharmacologic responses to a therapeutic
  intervention
• Clinical Endpoint a characteristic or variable
  that reflects how a patient feels or functions or
  how long a patient survives

4
Concepts/Definitions proposed by the NIH Workshop
for Evaluation of Surrogate Endpoints

• Surrogate Endpoint A biomarker intended to
  substitute for a clinical endpoint. The
  investigational team uses epidemiologic,
  therapeutic, pathophysiologic, or other
  scientific evidence to select a surrogate
  endpoint that is expected to predict clinical
  benefit, harm, or lack of benefit or harm.

5
Biomarkers

• Understanding the underlying mechanism of action
  to gain insight into the relationship between the
  biomarker, treatment, and clinical endpoint.
• Treatment may alter the biomarker but not affect
  clinical endpoint
• Means of assessing the biomarker must be valid
  and reproducible.
• Treatment may alter the biomarker but it may be
  an artifact of the means of specimen
  ascertainment, preparation, transportation, or
  measurement technique.
• intra-patient variability may be falsely
  attributed to treatment effect.

6
Surrogate Endpoints in Medical Practice

• Blood pressure
• Intraocular pressure (glaucoma)
• HgbA1c (diabetes)
• Psychometric testing
• Tumor shrinkage (cancer)
• ACR criteria (rheumatoid arthritis)
• Pain scales (pain)

Janet Woodcock, 11/4/04
7
Surrogate Endpoints in Clinical Trials

• New/Surrogate Endpoints
• Tumor markers
• Molecular endpoint assessment
• Surrogate marker or biomarker
• Functional Imaging

8
Rationale for Use of Surrogate Endpoints

• Decrease duration of clinical trial
• Decrease sample size for trial
• Integrity of the inference about the ultimate
  effects of treatment are maintained

9
Paradigm for Valid Surrogate Endpoint
Time
Intervention
Surrogate endpoint
True clinical endpoint
Disease
Fleming DeMets, 1996
10
Paradigm for Failure of Surrogate Endpoint(1.
Surrogate is not in causal pathway)
Time
Intervention
Surrogate endpoint
Disease
True clinical endpoint
Fleming DeMets, 1996
11
Paradigm for Failure of Surrogate Endpoint(2.
Intervention affects only one of several causal
pathways)
Time
Intervention
Surrogate endpoint
True clinical endpoint
Disease
Fleming DeMets, 1996
12
Paradigm for Failure of Surrogate Endpoint(3.
Surrogate is not in pathway of intervention, or
is insensitive to it)
Time
Intervention
Surrogate endpoint
True clinical endpoint
Disease
Fleming DeMets, 1996
13
Paradigm for Failure of Surrogate Endpoint(4.
Intervention has mechanisms independent of
disease process)
Time
Intervention
Surrogate endpoint
True clinical endpoint
Disease
Fleming DeMets, 1996
14
Trial Design Issues with Targeted Therapy
15
Establish reliability, validity, and feasibility
of Measurement Technique

• Reliability - propensity for the measurement
  procedure to provide the same value when repeated
  in the same subject/specimen under the same
  conditions
• Validity degree to which the observed value
  actually measures the desired quantity or
  characteristic

16
Specimen Collection Issues

• feasibility
• patient burden
• when assessing target inhibition in tumor tissue
  difficulties may arise in obtaining serial
  biopsies and intra-tumor variability
• use of peripheral blood as an alternative
  requires clinical or preclinical data supporting
  the contention that measurement of target
  inhibition in peripheral blood correlates with
  target inhibition in the tumor
• time commitment for monitoring schedule
• trial cost of monitoring endpoints

17
Preclinical developmentCytostatic Agents

• What is the expected effects once the target is
  inhibited?
• does target inhibition lead to cellular
  apoptosis, senescence, growth inhibition
• does target need to be inhibited continuously,
  intermittently, or only sporadically
• is the target the tumor or the surrounding
  stroma/microenvironment
• what is the function of the target in normal
  cells
• critical to establish that any observed
  preclinical activity in animal models is
  attributable to modulation of the target

18
Phase I trials

• Cytotoxic agents - administer the maximum amount
  of the agent in as intense a manner as possible
  to destroy as many tumor cells as possible while
  destroying a minimum amount of normal cells and
  inducing tolerable toxicity.

19
Phase I Trial Designs Cytotoxic Agents

• toxic/poisonous mechanism of action which
  generally results in a monotone S-shaped dose
  response relationship for tumor cells as well as
  a similar dose response relationship for normal
  cells
• maximum tolerated dose is based on the type and
  severity of normal tissue toxicities
• safety and feasibility

20
Phase I trials

• Cytostatic agents - administer the amount of the
  agent needed to maximally inhibit the relevant
  target while avoiding an assault on normal cells
  or intolerable toxicity

21
Biomarkers in Phase I Clinical Trials with
Targeted Agents

• Goal is to reach a dose that consistently
  inhibits the target
• Safety and pharmacology applies.
• Implement pharmacodynamic studies to assess
  target inhibition.
• The hypothesis is if you do not hit the target,
  it will not work.
• Provided you know what that target is, this may
  be correct.
• How do we do this?

22
What Are PD Studies?

• Assessment of drug effects on
• Target itself pErk in PD0325901 study.
• Proximal to the target pS6K in mTOR inhibitors.
• Distal to the target Ki67, blood flow.
• In early drug development, these studies are
  useful
• To prove the MOA.
• Dose and schedule finding tools.
• To validate normal tissues as potential
  surrogates.
• Studies are not useful to predict efficacy
• Too few patients.
• Even fewer responders.

23
Important Elements in Biomarker Development for
Phase I Trials

• Relevant preclinical model.
• Develop a method to assess target inhibition PD
  marker
• Needs to be well validated quantitative,
  reproducible, precise, accurate, etc
• Needs to be applicable to samples collected in
  clinical trials.
• Develop tissues that are appropriate to measure
  target inhibition.
• Tumor tissues.
• Surrogate tissue.

24
Surrogate Tissues in Phase I Trials

• Normal tissues used to measured PD endpoints
  skin, hair follicle, endothelial cells, PBMC
• Easier to collect in a serial manner.
• May be adequate for target inhibition assessment
  Pharmacological questions.
• Unlikely to be useful for oncological questions.
• Need to be validated.

25
Phase II clinical trial-cytostatic agents

• Disease stability-delay tumor growth
• demonstration of disease stability likely to
  translate to improvements in survival
• difficulty in assessing disease stability
• observation of tumor growth inhibition is linked
  to baseline tumor growth rate.
• slow growing tumor will appear stable for a short
  period of observation
• a rapidly growing tumor may grow within the same
  observation period but the rate of growth may
  have significantly decreased

26
Phase II clinical trial

• Disease stability as endpoint
• high degree of variability in human cancers and
  the tendency for patients with the most indolent
  disease to be over-represented in phase II trials
  leads to the issue of what level of activity is
  high enough to consider it significantly higher
  than what is expected with standard treatment.
• comparison to historical controls, concurrent
  controls receiving standard treatment, or
  patients serving as their own controls.

27
Validation of Surrogate Endpoints is Difficult

• Requires large sample size, evaluating both the
  surrogate endpoint and the true clinical outcome
• Requires understanding of disease process and
  causal pathways
• Depends on the specific intervention in question

28
AT THE END OF THE DAY, WHATS IN IT FOR ME (A
PATIENTS PERSPECTIVE) (1)

• Endpoints must be clearly defined in
  patient-friendly
• language i.e.
• Toxicity Can I tolerate the effects of agent?
• Tumor response How will I respond to treatment?
• Survival How long can I expect to live?
• Quality of life i.e.,
• What will it cost me to participate in study?
• Will I be able to work?
• How will the study affect my daily routine?
• How about pain management?
• How sick will I get?
• How frequently do I have to travel to
• trial site?
• What will occur during each visit?
• Provide a treatment/dateline plan

29
AT THE END OF THE DAY, WHATS IN IT FOR ME (A
PATIENTS PERSPECTIVE) (2)
(3) Will I receive any feedback/ results of
the trial? (4) When all else fails tell the
truth i.e., patients are, in many cases, more
resilient than you believe
Remember - the patient is also a member / partner
of the clinical trial team!
30
Getting your Biomarker Study Approved

• Spend sometime describing in in the protocol
  rationale, examples, references.
• IRB member may not be so acquainted with what
  you propose.
• Make it an objective in the study to test a
  hypothesis.
• Provide any preliminary data it may be useful for
  the science as it will not be for the patient.
• Provide safety assurance with the procedure.
• Write a clear ICF.

31
ExampleUsing DCE-MRI to Establish the Optimal
Therapeutic Dose for PTK/ZK
Morgan B, et al. J Clin Oncol 2003213955-3964
32
Example ADVL0015 Bortezomib in recurrent solid
tumors in Peds

• Use of composite endpoint-clinical tolerability
  biologic inhibition (20S proteosome)
• -determined by inhibition of the average 20S
  proteosome on day 1 hour 1 divided by average 20S
  protease prior to therapy and the result is 0.20
  or less

33
ADVL0015 Bortezomib

• Tolerability
• Non-hemotologic DLT
• Any Grade III or IV EXCEPT
• Grade III Nausea Vomiting
• Grade III transaminase, which returns to normal
  prior to next treatment
• Grade III Fever
• Hematologic
• Grade IV neutropenia/thrombocytopenia gt 7days
• Other
• Grade II non-hematologic causing gt 7 day
  interruption

34
ADVL0015 Escalation Scheme based on Tolerability
Inhibition
35
Practical Considerations

• Demonstrated relationship to MOA
• Properly validated surrogate endpoints are not
  common
• Most useful in Phase II setting as a basis for
  justifying further evaluation
• Problematic in Phase III trials, particularly
  when the interventions being compared are based
  on different modalities