Acute Exacerbation of Chronic Bronchitis Issues in Drug Development

							Acute Exacerbation of Chronic
 Bronchitis - Issues in Drug
        Development
     Susan D. Thompson, M.D.
           FDA, DAIDP
        November 20, 2002

                                1
             AECB - Outline
•   Issues in diagnosis
•   Study design considerations
•   Inclusion/Exclusion criteria
•   Outcome assessment and timing
•   Statistical issues
•   Conclusions


                                    2
   AECB: Definition of Disease
     and Issues in Diagnosis
• Contrast with acute bronchitis: viral etiology
• AECB occurs in patients with chronic bronchitis
  (CB); CB is a subset of patients with chronic
  obstructive pulmonary disease (COPD)
• Common disease: 5-10% of all U.S. antibiotic
  prescriptions
• Positive sputum culture is not diagnostic of
  AECB, nor does a bacterial isolate necessarily
  document the etiology of AECB
                                                    3
        AECB - Study Design
          Considerations
• Concomitant medication/therapies: shown
  to have independent therapeutic efficacy in
  treatment of AECB:
  – Steroids
  – Bronchodilators: beta-adrenergic agonists and
    anticholinergic agents



                                                    4
       AECB - Study Design
     Considerations - Placebo-
      Controlled Trials (PCT)
• Past 40 years: 1100 patients enrolled in
  randomized, PCT of antibiotic treatment of
  AECB; none of identical design
• Differences in the definition of acute
  exacerbation
• Lack of standard outcome measures (PEFR,
  duration of AECB, PaO2, symptom score,
  overall severity score)
                                               5
       AECB - Study Design
      Considerations - PCT - 2
• Lack of reproducible rating system for
  severity
  – “Winnipeg criteria”: cough, sputum
    production, sputum purulence; I most severe
  – Winnipeg criteria not validated, no effect of
    antibiotics; found cardiopulmonary status and
    more than 4 exacerbations/year to be predictive
    of severity - i.e., historical parameters

                Anthonisen NR et al, Ann Int Med,   6
                1987;106:196-204; Ball et al, Q J
         AECB - Study Design
        Considerations - PCT - 3
•   Patient population not uniform
•   Varying outcomes!
•   Most performed more than 10 years ago
•   Conclude: Patients with more severe illness
    may benefit most from antibiotics, but this has
    not been conclusively demonstrated, nor have
    validated severity criteria been established;
    “narrow-spectrum” antibiotics preferred
                                                      7
AECB: Inclusion and Exclusion
          Criteria
• Guidance: PFT/ABG suggested, not required;
  sputum culture and history of CB required
• CB, COPD with AECB
  – Smoking history    -Age
  – Low FEV1
• Control for concomitant interventions and
  cigarette smoking

                                               8
AECB: Study Populations
                   Anthonisen      NDA

Mean age (y)       67.3  9.0    59 (range
                                  15-88)

Smoking history      93.6%        61.9%

FEV1 (% pred.)     33.9  13.7     N/A

Sputum > 30 mL/d     26.5%         N/A

Type 1 or 2 sx       80.9%         N/A


                                             9
  AECB: Evaluation,Timing of
   Assessment, and Outcome
• Current FDA TOC for AECB: Clinical
  response (return to baseline) at TOC = 1-2
  weeks after completion of therapy
• Alternative 1o outcome variables: time to
  resolution of symptoms, symptom/severity
  score, deterioration.
• Microbiological endpoint not appropriate


                                               10
       AECB: Statistical Issues
• Determine D2: AECB has very low attributable
  mortality and morbidity. Thus D2 (loss of efficacy w/r
  control) is relatively large and greater than 20%
• Determine D1: estimation of the benefit (if any) of
  active control over placebo.
   – Meta-analysis with determination of D1 is not a valid
     approach for AECB due to the limitations of the
     existent PCT’s: differences in study design, outcome,
     patient population, and endpoints would not allow a
     definitive estimation of the benefit of active control
     over placebo.
                                                              11
  AECB: Statistical Issues 2
– Placebo-controlled trials
   • Drug versus placebo
   • Early escape: Rigid discontinuation criteria (day 2-
     3), e.g., deterioration or progression based on
     objective criteria
   • Only high risk patients - but how to define?
   • Only low risk patients - but how to define?




                                                        12
    AECB: Statistical Issues 3
• Options for future AECB trials:
  – Superiority trial - Standard of care vs
    experimental drug
  – Non-inferiority trial for all or for a subset of
    AECB
     • only in severely ill
     • different deltas by strata
     • 3 arm (placebo, new drug, old drug)

                                                       13
         AECB: Conclusions
• Selection of appropriate study design is
  critical for AECB trials
  – patient population
  – concurrent therapies
  – choice of endpoints
• Placebo-controlled or superiority trial
  designs should be conducted for antibiotic
  trials in AECB.
                                               14
15
            AECB Questions
• Are there methods to select a patient
  population more likely to benefit from
  antimicrobial therapy? Please discuss:
  – Is it more appropriate to look at patients with
    acute exacerbation of chronic obstructive lung
    disease (as defined by PFTs) than at all patients
    with chronic bronchitis?
  – Other criteria such as patient age

                                                    16
          AECB - Questions - 2
• Please discuss the effects of potential confounders
  of the measurement of antimicrobial effect in trials
  of AECB such as:
   – Should concomitant medications (e.g., beta agonists,
     anticholinergic agents, steroids) be standardized in the
     protocols? Do use of these agents differ across
     geographic regions?
   – Current smoking status
   – The patient’s prior history of exacerbations (e.g., are
     patients with more exacerbations per year more likely
     to fail any therapy?).                                   17
18