Endpoint considerations in cancer clinical trials

1. Dr. Bhaswat S. Chakraborty
Senior Vice President and Chairman,
R&D Science Core Committee, Cadila Pharmaceuticals Ltd.
National Conference on Innovation in Pharmaceutical
Industry
L.J. Institute of Pharmacy 28
th
January 2012

2. Contents
Clinical Development of a New Drug
Endpoint considerations
Merits and demerits
Overall survival
Tumor assessment based endpoints
QoL, biomarkers and symptom trial
Trial designs
Concluding comments

3. Clinical Development of New Drugs
Development Plan
Define target disease population
Dose range and schedule at which the drug can be shown to be
simultaneously safe and effective, to the extent that the risk-benefit
relationship is acceptable
Satisfying these broad aims usually requires an ordered program of
clinical trials, each with its own specific objectives (ICH E8).
A marketing application should clearly describe the main content of
such plans, and the contribution made by each trial.
A statistical summary, overview, or meta-analysis may be informative when
medical questions are addressed in more than one trial.
Other major statistical issues (if any) that are expected to affect a number of
trials in a common plan should be addressed in that plan

4. Cancer Research Today
Research is conducted mainly on
New Drugs
New Combinations
Radiotherapy
Surgery
In the West, research is usually done by large co-operative groups, in
addition to those mentioned for India
In India
Large Pharmaceuticals
Co-operative Groups, e.g., ICON (Indian Co-operative Oncology Network)
Regional Cancer Centres & Govt. sponsored studies
Academia

5. What does FDA Look for in a Ca RCT?
FDA approves a drug application based on
Substantial evidence of efficacy & safety from “adequate
and well-controlled investigations”
A valid comparison to a control
Quantitative assessment of the drug’s effect

(21 CFR 314.126.)
The design of cancer trials intended to support drug
approval is very important
Oncology based NDAs are usually reviewed on fast track

6. Endpoints in Oncology Trials
Must show either direct evidence of clinical benefit or improvement
in an established surrogate for clinical benefit
Clinical benefit: survival improvement
Overall survival (OS)
Progress-free survival (PFS)
Improvement in a patient’s quality of life (QOL)
Other endpoints on which approval has been given are:
Objective response rate (ORR)
by RECIST or any radiological tests or physical examinations
Improvement in survival, improvement in a QOL, improved physical
functioning, or improved tumor-related symptoms do not always be
predicted by, or correlate with, ORR

7. So, the Endpoint Metrics are:
Time to event end points
Survival
Disease free survival
Progress -free survival
Objective response rates
Complete
Partial
Stable disease
Progressive disease
Symptom end points
Palliation
QoL

8. Relative Merits
Endpoint Evidence Assessment Some AdvantagesSome Disadvantages
Survival Clinical benefit· RCT needed
· Blinding not
essential

· Direct measure of
benefit
· Easily
measured
· Precisely
measured
· Requires larger and
longer studies
· Potentially affected by
crossover therapy
· Does not capture
symptom benefit
· Includes noncancer
deaths
Disease-Free
Survival
(DFS)
Surrogate for
accelerated
approval or
regular
approval*
· RCT needed
· Blinding
preferred
· Considered to
be clinical benefit
by some
· Needs fewer
patients and
shorter studies than
survival
· Not a validated
survival surrogate in most
settings
· Subject to assessment
bias
· Various definitions
exist

9. Relative Merits..
Endpoint Evidence Assessment Some AdvantagesSome Disadvantages
Objective
Response
Rate (ORR)
Surrogate for
accelerated
approval or
regular
approval*

· Single-arm or
randomized
studies can be
used
· Blinding
preferred in
comparative
studies
· Can be assessed
in single-arm
studies
· Not a direct measure of
benefit
· Usually reflects drug
activity in a minority of
patients
· Data are moderately
complex compared to
survival
Complete
Response
(CR)

Surrogate for
accelerated
approval or
regular
approval*
· Single-arm or
randomized
studies can be
used
· Blinding
preferred in
comparative
studies
· Durable CRs
represent obvious
benefit in some
settings (see text)
· Can be assessed
in single-arm
studies
· Few drugs produce high
rates of CR
· Data are moderately
complex compared to
survival

10. Overall Survival (OS)
OS: The time from randomization until death from any cause
Measured usually in the intent-to-treat (ITT) population
Most reliable cancer endpoint, and when studies can be
conducted to adequately assess survival, it is usually the
preferred endpoint
Precise and easy to measure
Bias is not a factor in endpoint measurement
Survival improvement should be analyzed as a risk-benefit
analysis to assess clinical benefit
OS should be evaluated in RCTs
Historical trials are seldom reliable for time-dependent endpoints
(e.g., OS, PFS).

12. Rosell et al., Annals of Oncology 19: 362–369, 2008

13. Endpoints Based on Tumor Assessments
Disease-free survival (DFS)
Objective response rate (ORR)
Time to tumor progression (TTP)
Progress-free survival (PFS)
Time-to-treatment failure (TTF)
They are all time-dependent endpoints
Collection and analysis of these endpoints are based on indirect
assessments, calculations, and estimates (e.g., tumor measurements)
Two critical judgments:
1.whether the endpoint will support either accelerated approval or regular
approval
2.endpoint should be evaluated for the potential of bias or uncertainty in
tumor endpoint assessments
 Drug applications using studies that rely on tumor measurement-
based endpoints as sole evidence of efficacy may need
confirmatory evidence from a second trial

14. Rosell et al., Annals of Oncology 19: 362–369, 2008

15. Cautions in Tumor Assessments
Accuracy in measuring tumors can differ among tumor settings
Imprecision can happen in locations where there is a lack of
demarcated margins (e.g., malignant mesothelioma, pancreatic
cancer, brain tumors).
When the primary study endpoint is based on tumor
measurements (e.g., PFS or ORR), tumor endpoint assessments
generally should be verified by central reviewers blinded
to study treatments
 This measure is especially important when the study is not blinded
 It may be appropriate for the FDA to audit a sample of the scans to
verify the central review process

16. Quality of Life (QoL) Endpoints
Global health-related quality of life (HRQL) have not served
as primary efficacy endpoints in oncology drug approvals
They are usually patient reported outcome measures
For example, the FACT-L is a 44-item self-report instrument which measures
multidimensional quality of life in Phase II and III lung cancer clinical trials
Reliability and validity of such multi-item instruments must be thoroughly
examined
For QOL to be used as primary endpoints to support cancer
drug approval, the FDA should be able to distinguish between
improvement in tumor symptoms and lack of drug toxicity
An apparent effectiveness advantage based on a global QoL
instrument can simply indicate less toxicity rather than
effectiveness

17. Biomarkers
Usually not a good idea for cancer drug approval
Other than paraprotein levels measured in blood and urine for
myeloma, biomarkers assayed from blood or body fluids have not
served as primary endpoints
Not considered good predictors of clinical benefit
The FDA has sometimes accepted tumor markers as elements of a
composite endpoint
e,g., clinical events such as significant decrease in performance status,
or bowel obstruction in conjunction with marked increases in CA-125
was considered progression in ovarian cancer patients
Biomarkers, however, can be useful in identifying prognostic
factors
 and in selection of patients and stratification factors to be
considered in study designs

18. Specific Symptom Endpoints
Time to progression of cancer symptoms, an endpoint similar to TTP, is a
direct measure of clinical benefit rather than a potential surrogate
Problems in measuring progression (e.g., missing assessments) also exist in
evaluating time to symptomatic progression
Because few cancer trials are blinded, assessments can be biased
delay between tumor progression and the onset of cancer symptoms can occur
alternative treatments are initiated before achieving the symptom endpoint,
confounding this analysis
patients may have minimal cancer symptoms
also, tumor symptoms can be difficult to differentiate from drug toxicity
Important
composite symptom endpoint should have components of similar clinical
importance and the results should not be exclusively attributed to one
component
missing data & infrequent treatment are also confounding factors

19. Trial Designs
Randomized Clinical Trials (RCTs)
Gold standard in Phase III
Single centre CT
Primary and secondary indications
Safety profile in patients
Pharmacological / toxicological characteristics
Multi-centre CT
Confirmation of the above
Effect size
Site, care and demographic differences
Epidemiological determination
Complexity
Far superior to meta-analyzed determination of effect

20. 0
D
i
f
f
e
r
e
n
c
e

i
n

C
l
i
n
i
c
a
l

E
f
f
i
c
a
c
y

(
Є
)
d = Meaningful Difference
Non-Inferiority
Equivalence
Inferiority
Superiority
-d
+d
Non-Superiority
Equality

21. Other Trial Designs..
Single arm studies
no available therapy and where major tumor regressions can be presumed to be
attributed to the tested drug
ORR and response duration measurements
Non-inferiority (NI) trials
should demonstrate that the new drug is not less effective than a standard
regimen (the active control) by a prespecified amount (noninferiority margin)
NI margin is some fraction of (e.g., 50 percent) of the control drug effect (assay
sensitivity)
the control should have a well-characterized survival benefit
If the new drug is inferior to the active control by more than the NI margin, it
will be presumed to be ineffective
Other approaches
No Treatment or Placebo Control Studies
Isolating Drug Effect in Combinations
Studies for Radio- and Chemotherapy Protectants

22. Study Design: Approaches
Randomised Controlled Trials (RCT) most preferred
approach
Demonstrating superiority of the new therapy
Other approaches
Single arm studies (e.g., Phase II)
e.g., when many complete responses were observed or when
toxicity was minimal or modest
Equivalence Trials
No Treatment or Placebo Control Studies
Isolating Drug Effect in Combinations
Studies for Radio- and Chemotherapy Protectants

23. Placebo Control Equality Trials
No anticancer drug treatment in the control arm is
unethical
Sometimes acceptable
E.g., in early stage cancer when standard practice is to give no
treatment
Add-on design (also for adjuvants)
all patients receive standard treatment plus either no additional
treatment or the experimental drug
Placebos preferred to no-treatment controls because they permit
blinding
Unless very low toxicity, blinding may not be feasible because of
a relatively high rate of recognizable toxicities

24. Drug or Therapy Combinations
Use the add-on design
Standard + Placebo
Standard + Drug X
Effects seen in early phases of development
Establish the contribution of a drug to a standard regimen
Particularly if the combination is more effective than any of
the individual components

25. Concluding Remarks
Clinical testing of new Oncology products is very sophisticated and complex
Cancer clinical data is very complex (censored, skewed, often fraught with
missing data point), therefore, proper hypothesization and statistical treatment
of data are required
There are many endpoints that are scientifically valid but OS as primary end
point is often preferred by regulatory agencies
Tumor assessment trials may need another confirmatory CT
Endpoints must be demonstrative (directly or indirectly) of clinical benefit
Missing data, infrequent treatment, increased type I error and other
confounding factors must be addressed
Prospective RCTs are usually the preferred approach for evaluation of new
therapies
Despite good knowledge in endpoints & trial design, meet & consult FDA
before initiating a pivotal trial.

26. Thank you Very Much