The role of the DSMB by pC7J9D


									The role of the DSMB
         SE Asia
  Influenza Clinical Research
      Network (ICRN)

    David Lalloo- Chair
DSMB – Data Safety Monitoring Board

 • Also termed DMC= Data Monitoring Committee
   in some countries
 • Independent body appointed for most clinical
 • Major role is to ensure safety of participants
 • Assures validity of results by overseeing conduct
   of a trial
 • Protects investigators
 • Advises investigators (often the Steering
   committee) and the sponsor
              Who sits on a DSMB
• Clinicians and statisticians
• Sometimes ethicists and community
• Other members to address specific issues

• This DSMB
  –   3 clinicians (including chair)
  –   2 statisticians
  –   1 ethicist
  –   I virologist
  –   Executive Secretary
How do DSMBs function?

• Meet before trial start, sometimes in joint
  meeting with trial steering group

• Decide on
   –   Role
   –   Frequency of meetings
   –   What data they need to see
   –   What rules they will use to make decisions

• Often formalise agreement in a “charter”
Typical DSMB meeting

• Members get confidential data reports from
  designated statistician
• Meet Open session
• Review general progress and overall data (may
  include report from trial team)
• Closed session: review all data & discuss
  issues. Focus on safety, often look at efficacy
• Make a decision, may need more data
• Write a report
Role of a DSMB
• Monitor data throughout a trial
• Only group that has access to unblinded data
  before the end of a study
• Analyzes data by treatment arm
  – Are there concerns about safety?
  – Is there already enough proof that the intervention
• Also considers
  – New evidence that might be relevant to the trial
  – Practicalities -is it still feasible to continue?

 • Possible decisions:
   – Continue with trial as planned
   – Stop: safety problem
   – Stop: efficacy established
   – Stop: new knowledge (usually from other trials
     suggesting risks)
   – Stop: futile. Trial unlikely to show a result
   – Modify trial design
Issues to consider in interim analyses

 • Need to be absolutely sure that you are making
   the correct decision
 • Need to adjust the p value if you are doing
   multiple interim analyses
 • Various different statistical methods for doing
 • Establish these “stopping rules” at the beginning
   of the study
Examples of DSMB decisions
• SAPIT study
  – Compares three different time points for starting ART
    in those with TB.
  – One arm (starting ART after TB treatment completed)
    was stopped by the DSMB in October 2008 because
    of a 55% higher mortality

  – Compared artesunate to quinine for severe malaria.
    Trial designed to randomise 2000 patients.
  – Stopped after interim analysis when 1500 patients
    entered: artesunate reduced mortality by a third
Examples of DSMB decisions

• Cellulose sulfate microbicide trial stopped in
• Interim analysis suggested increased risk of
  harm (HR 2.23; P = 0.02).
• Other trials stopped on basis of interim analysis
• When trial finally analysed, HR 1.61 (95% CI
  0.86 to 3.01; P = 0.13)
 • DSMBs have important role in the running of clinical
 • Ensure that decisions made on sound clinical and
   statistical bases
 • If properly constituted
    – help to ensure the safety of participants in a trial
    – help the investigators to make sure that the maximum
      information is obtained from a trial

 • Reassure participants, sponsors, ethics committees and
   general public that a trial is being conducted safely and

To top