Imaging in Clinical Development of Cancer Drugs

Download Report

Transcript Imaging in Clinical Development of Cancer Drugs

Molecular Imaging in Clinical Drug
Development:
Challenges of multi-site clinical trials
Andrea Pirzkall, MD
Genentech Research Early
Development (gRED)
CTN workshop 2/1/10
1
Disclosures
• I am an employee of Genentech, Inc
• I will discuss investigational use of:
– 18F-fluorothymidine (FLT)
2
Outline
• Value of novel imaging agents in clinical
drug development
• Importance of multi-center trials
• Challenges for drug developers doing
multi-center trials with investigational
imaging agents
• A specific example—FLT-PET
– How we did it
– What we would like to see in the future
3
The Drug Development Process:
Low success rates at every stage
Pre-Clinical
Drug Discovery Development
Clinical Development
Success rate (Ph1 to Approval):
Approx 10-20%
Kola & Landis Nat Rev Drug Disc 2004
DiMasi & Grabowski J Clin Onc 2007
Drug
Basic
Target
Candidate
Research
ID
Selection
IND
Phase 1 Phase 2 Phase 3 Approval
studies
IND=Investigation New Drug
Application
4
Oncology drug development: Low success
rates at every stage of clinical development
Data for 1991-2000 for 10 largest pharmaceutical companies
Kola & Landis, Nature Reviews Drug Discovery 2004
5
New tools for clinical drug
development needed
• High failure rate for drug candidates at
every stage of clinical development
• New approaches to clinical drug
development are needed
– FDA’s Critical Path Initiative
• Imaging approaches particularly promising
– Potential new tools to improve clinical drug
development
6
Potential uses of imaging in clinical
development of (oncology) drugs
1. Imaging presence of target on tumor
•
Identification of appropriate patient population for treatment
2. Imaging biodistribution of drug
–
How much drug reaches tumor compared to other
tissues/organs?
3. Imaging pharmacodynamic changes
–
Imaging biological effect of drug on tumor (or other
tissues/organs)
•
•
•
Is the drug binding to target?
Is the drug inhibiting the target?
Is the drug inducing the expected downstream biochemical
changes?
4. Imaging surrogate efficacy endpoints
–
Are changes occurring in tumor that are associated with
improved outcome (e.g. progression free or overall survival)?
7
Imaging presence of target on tumor:
111In-labeled
trastuzumab and her2+ tumors
• Single-photon emission computed tomography
(SPECT) to image labeled anti-her2 antibody
• Fused CT and 111In-DTPA-trastuzumab SPECT
image (96 hours after tracer injection)
8
Copyright ® American Society of Clinical Oncology
Perik, P. J. et al. J Clin Oncol; 24:2276-2282 2006
Imatinib in GIST: Early changes in FDGPET predict subsequent tumor shrinkage
FDG-PET
Pretreatment
Day 8
CT scans
Pretreatment
Week 4
Week 24
Stroobants et al Eur J Cancer 2003
GIST=Gastrointestinal stromal tumor
9
FDG-PET: Sunitinib effect in imatinibresistant GIST
• Single arm phase 1/2 study:
– 50 mg daily on different
schedules
– RR (RECIST) 9.1% (5/55)
– Qualitative PET response rate at
7 days 62% (33/53)
• Randomized, placebocontrolled Phase 3 (n=312):
Dileo et al GI ASCO 2005
Van den Abeele et al. ASCO 2005
Demetri et al ASCO 2005
Sunitinib package insert
– 50mg daily 4 wks on, 2 wks off
– Sunitinib arm: RR (RECIST) 6.8
%
– HR for TTP = 0.33 p<0.0001 (vs
placebo)
10
How to prioritize efforts to use imaging
for clinical drug development?
• High failure rate of molecules at every stage of
clinical development
– Imaging could potentially improve development at
every phase
• Late failure (e.g. failure in pivotal Phase 3
studies) is much more costly than early failures
• High priority goal: shift failures to earlier in
process
• Biggest impact on drug development:
Reduce Pivotal (Phase 3) failures
– Improve Go/No Go to Phase 3 decisions
11
Current basis for Go/No Go
decision to Phase 3 in Oncology
1. Small, single arm Phase 2
studies:
– Tumor shrinkage (RECIST)
used to decide Go/No Go to
Phase 3
– Inadequate for many new
oncology molecules
2. Large, randomized phase 2
studies
•
– Typically with time to
progression endpoints
– Long duration, large numbers
of patients
– Not sustainable:
Data for 1991-2000 for 10 largest
pharmaceutical companies
•
Increases cost of Phase 2
drug development
Kola & Landis, Nature Reviews Drug Discovery 2004
12
Using new technologies (imaging) to
improve Go/No Go decision to Phase 3
• Required characteristics of new
technology:
– Yield useful information in relatively small
Phase 2 studies:
• Single arm, short duration
– Assess drug activity in absence of tumor
shrinkage
• Improve upon current RECIST criteria
13
General performance requirements
for (imaging) test
• To guide individual patient decisions:
– Need excellent positive and negative
predictive value
– If test has high error rates won’t be used
• To guide development of a novel drug:
– Relatively low bar to improve upon current
decision making
– Relatively high error rates would still be an
improvement
14
Are available imaging technologies
sufficient?
• Available imaging technologies may be well suited
to the task:
– FDG-PET: Measure changes in tumor metabolic rate
– FLT-PET: Measure changes in tumor proliferative rate
– DCE-MRI: measure blood flow/vascular permeability
• These measure biological changes likely
associated with effective anti-cancer drugs
– could improve clinical drug development in the near future
• Other newer technologies may ultimately prove
superior
– But, establishing their place in drug development will take longer
15
FDG-PET imaging is promising for
clinical oncology drug development
• Wide clinical availability
• Numerous publications on clinical use
• Commonly used in management of many
patients with cancer
• In cancer drug development:
– Some dramatic examples
– However, more work needed to inform Go/No
Go to Phase 3 decisions
16
Need to do multi-center trials
• Many imaging agents have entered the clinic
• Few have been evaluated in multi-center trials
– significantly limiting impact
• Even FDG-PET: relatively few multi-center
results reported
• E.g., only now, are multi-center studies
underway to confirm the association between
FDG-PET response and clinical outcome from
standard therapy in common cancers:
– Non-small cell lung cancer: ACRIN 6678
– Non-Hodgkin Lymphoma: CALGB 580603
– Coordinated by the Foundation for NIH
17
Need for multi-center studies
• Increase confidence if similar results
obtained at different clinical sites
• Facilitates broader availability
• Adequate numbers of patients in an
acceptable time frame
• Fast way to impact clinical practice and
use in drug development
18
Challenges for therapeutics
developers doing multi-center trials
with investigational imaging agents
•
•
•
•
Regulatory
Quality/reliability of imaging agent
Quality/consistency of image acquisition
Quality/consistency of image interpretation
19
A Specific Example
• A study of FDG- and FLT-PET in patients
with non-small cell lung cancer receiving
erlotinib
• Purpose of study:
– Determine FDG- and FLT-PET response
rates and association with clinical outcome
– Determine feasibility of multi-center study with
FDG- and FLT-PET
20
Example: Erlotinib
• Small molecule, orally bioavailable inhibitor of
epidermal growth factor receptor (EGFR)
• Approved for treatment of patients with
advanced or metastatic non-small cell lung
cancer (NSCLC) after failure of at least one prior
chemotherapy regimen
• Randomized clinical study (BR.21) of erlotinib vs
placebo in NSCLC showed
– RECIST response rate of 8.9% with erlotinib (0.9%
with placebo)
– Median overall survival with erlotinib 6.7 months (4.7
months with placebo)
21
Evaluating FDG- and FLT-PET with
an established targeted therapy
• Purpose of study:
– Determine FDG- and FLT-PET response
rates and association with clinical outcome
– Determine feasibility of multi-center study with
FDG- and FLT-PET
• Use an established targeted therapy
(erlotinib in non-small cell lung cancer)
• Study is not intended to evaluate erlotinib
22
Study design
Diagnostic CT
FDG-PET
FLT-PET
Day
Day -14 to -1
(screening)
0
Diagnostic CT
FDG-PET
FLT-PET
14
Diagnostic CT
FDG-PET
FLT-PET
56
Every 56 day
Erlotinib Rx
until progressive disease,
intolerable toxicity, or up
to 1 year
Determine
progression free
survival
Continued follow-up until
death, or up to 1 year
following enrollment of
last patient
Overall
Survival
23
FLT in a multi-center setting:
Regulatory path
• FLT is not approved by FDA
• Filed an IND for FLT
• Benefited from NCI’s Cancer Imaging
Program having already filed an IND for
FLT
• Needed to ensure quality of FLT
manufacturing process and product
– Challenging to monitor multiple sites using
different processes
24
FLT in a multi-center setting:
Quality/reliability of imaging agent
• In the U.S., decided to work with a
commercial producer/distributor
– Ensured adequate control of manufacturing
process and product quality
– Significantly limited geographic area of
possible clinical sites
25
Ensuring Image acquisition
consistency/quality at clinical sites
• Identified an expert imaging group to develop:
– An Imaging Charter describing image acquisition
procedure
– Pre-specified image analysis approach
• At each clinical site, an imaging physician
(radiology/nuclear medicine) formally identified as a
sub-investigator on the study
• Representatives of central imaging group visited
each imaging site to train and evaluate site
personnel
• Used case report forms to collect critical parameters
for image quality (e.g. radiotracer uptake time)
– Provided feedback to imaging sites
26
Summary
• Molecular imaging has tremendous potential
value for clinical drug development
• Impact on drug development has been limited, in
part, due to challenges of multi-center clinical
studies
• Example of FDG/FLT-PET study of erlotinib in
NSCLC
– Illustrates challenges drug developers face
– Initial results to be reported at the 13th World
Conference on Lung Cancer in July
27
What we would like to see in the future
• For imaging agents requiring an IND, a
mechanism that provides:
– Broader choice of clinical sites
– Shorter time to initiate clinical studies
– Confidence in quality/consistency of imaging
agent at multiple clinical sites
– Confidence in quality/consistency of image
acquisition procedures at multiple clinical sites
• SNM Clinical Trials Network may address
these needs
28
Thank you
29