Use of Data Monitoring Committees DMC in Device Trials: An FDA Division of Cardiovascular Devices DC

1
Use of Data Monitoring Committees (DMC) in Device
Trials An FDA Division of Cardiovascular Devices
(DCD) Perspective

• Bram Zuckerman MD, FACC
• Bram.zuckerman_at_fda.hhs.gov
• Director, Division of Cardiovascular Devices
• Center for Devices and Radiological Health

2
Introduction

• Most literature on Data Monitoring Committees
  (DMC) deals with drug trial applications
• DMCs have been employed at FDA Device Center
  (CDRH) over the last decade
• Today's goals
• Talk about use of DMCs at CDRH
• Comment on FDA DMC Guidance Document

3
CDRH and Data Monitoring Committees (DMCs)

• HHS Office of Inspector General recommended in
  1998 that FDA clarify appropriate role and
  procedure for DMCs
• In 2006 FDA issued the DMC Guidance
  www.fda.gov/cber/gdlns/clintrialdmc.htm
• The Guidance is applicable to CBER, CDER, and
  CDRH trials

4
What is a Clinical Trial Data Monitoring
Committee?

• A Clinical Trial Data Monitoring Committee (DMC)
  is a group of individuals with pertinent
  expertise that reviews on a regular basis
  accumulating data from an ongoing clinical trial
• The DMC advises the sponsor regarding the
  continuing safety of current participants and
  those yet to be recruited, as well as the
  continuing validity and scientific merit of the
  trial

5
Other Oversight Groups Interact with DMCs

• Clinical Endpoints Committee (CEC)
  independently reviews important endpoints
  reported by trial investigators to determine
  whether they meet protocol-specified criteria
• Institutional Review Board (IRB) responsible
  for evaluating a trial to determine whether
  risks to subjects are minimized and risks to
  subjects are reasonable in relation to
  anticipated benefits (21 CFR 56.111(a)(1) and
  (3))

6
Which Device Trials need DMCs?

• All clinical trials require safety monitoring (21
  CFR312.32 (c)) but this does mean that every
  trial needs a formal committee external to the
  trial organizers and investigators
• A DMC is required by FDA in the case of waived
  informed consent (21 CFR 50.24) (e.g., emergency
  research for CPR devices)
• A DMC is not needed or advised for every clinical
  trial, although it may prove valuable and Least
  Burdensome

7
Which Device Trials Need DMCs?

• Consider relevant ethical and scientific reasons
  (i.e., high risk to trial participants, long-term
  trial) for institution of a Data Monitoring
  Committee in large multisite studies
• For cardiovascular disease trials the primary
  (or secondary) endpoint is often mortality or
  major morbidity for pivotal trials of arrhythmia,
  heart failure, myocardial infarction, and
  anti-restenosis therapies
• Would a favorable or unfavorable result ethically
  and\or scientifically require early termination?

8
Which Device Trials Need DMCs?

• Will institution of a DMC allow for the use of
  optimal statistical and clinical trial
  methodologies?
• A) If differences in major response variables
  are unimpressive at an interim analysis is it
  justifiable in terms of time, money, and effort
  to continue?
• B) Alternatively, if the difference in the
  primary endpoint is less than expected a DMC
  might be useful for legitimately helping to
  modify inclusion\exclusion criteria or sample
  size

9
Which Device Trials Need DMCs?

• Special concerns about risks to trial
  participants
• A) Is the treatment to be tested novel, so that
  there is little prior information on clinical
  safety, or is there prior information that raises
  concerns about the potential for serious
  problems? (e.g., DMCs are useful in feasibility
  device trials as well as pivotal trials)
• B) Safety concerns are usually heightened in
  studies performed in potentially fragile
  populations (e.g., children or the elderly)

10
Which Cardiovascular Device Trials Need DMCs?

• Last decade has seen increasing importance of
  device therapy for treatment of cardiovascular
  disease
• Most of the prior bullet points apply to the
  cardiovascular device trials arena
• DCD use of DMCs has been extensive
• DCD recommends use of DMCs for feasibility and
  pivotal trials when dealing with complex product
  development

11
General Operational Issues

• DMCs should have well-defined standard operating
  procedures
• The DMC and Sponsor should be able to operate per
  the planned operational timeline
• Members should be carefully selected (experienced
  biostatistician, independent and objective expert
  clinicians, DMC Chair should have experience in
  clinical trials as well as the disease of
  interest)

12
General Operational Issues

• The plan for interim looks at data and the alpha
  spending function need to be prospectively
  defined in the IDE protocol
• Any plan for possible increase of sample size
  needs to be prospectively stated in the IDE
  protocol
• It is not uncommon in the closed section of a
  DMC meeting that the DMC may need to see
  unblinded trial results in order to effectively
  determine the risk/benefit profile at an interim
  analysis

13
Operational Problems Encountered by DCD

• In many cases DMCs have not been able to operate
  per the planned operational timeline
• As a result DCD often only gives conditional
  approval for IDE studies
• Full IDE approval is dependent on the Sponsor and
  DMC demonstrating that they can operate
  effectively
• DCD may not agree with evaluation strategy and
  conclusions reached by DMC and exercise its
  authority accordingly 21CFR 812.150(b)10

14
Major Reasons for Early Termination of a Trial

• The trial may show serious adverse effects in the
  entire intervention group or in a dominating
  subgroup
• There may be greater than expected beneficial
  effects
• It may become clear that a statistically
  significant difference by the end of the trial is
  improbable
• Logistical or data-quality problems may be so
  severe that correction is not feasible

15
The Early Termination of a Clinical Trial Can be
Difficult

• Issues involved may be complex because study
  results are often mixed
• Statistical stopping rules are useful guides in
  this process but should not be viewed as an
  absolute
• Examine differences in prognostic factors,
  possible role of bias due to non-blinding, impact
  of missing data, side effects and outcomes of
  secondary response variables, internal
  consistency across subgroups and between centers,
  outcomes of similar trials, and impact of early
  termination on general acceptance of results in
  clinical practice

16
DCD Experience with Early Trial Termination

• Sponsor Consultation with FDA on implications of
  early stopping may be helpful
• FDA will rarely, if ever, tell a sponsor which
  decision to make but only provide scientific and
  regulatory guidance on the possible implications
  of early termination
• Percusurge Panel Transcript (2/05/01) on FDA
  website provides one detailed DCD device example

17
Conclusions

• Use of DMC is recommended for many device trials
• General DMC principals apply to device and drug
  trials even though devices are not drugs
• Real time DMC implementation has been a challenge
  for many DCD trials
• A conditional IDE approval mechanism has been
  helpful for practically improving DMC performance