Sandell D, Mitchell JP. Considerations for Designing In Vitro Bioequivalence (IVBE) Studies for Pressurized Metered Dose Inhalers (pMDIs) with Spacer or Valved Holding Chamber (S/VHC) Add-on Devices.
J Aerosol Med Pulm Drug Deliv 2014;
28:156-81. [PMID:
25089555 DOI:
10.1089/jamp.2014.1150]
[Citation(s) in RCA: 1] [Impact Index Per Article: 0.1] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/12/2022] Open
Abstract
BACKGROUND
The choice of analytical test methods and associated statistical considerations are considered for the laboratory testing of pressurized metered dose inhaler-spacer/valved holding chamber (pMDI-S/VHC) combinations for in vitro bioequivalence (IVBE).
METHODS
Four scenarios are presented for comparing TEST ("second entry" or "generic") versus REF ("innovator"): (1) innovator and second entry product pMDI alone without any S/VHC (baseline comparison); (2) innovator and second entry pMDI product with the same S/VHC; (3) innovator pMDI product with existing S/VHC and second entry product with a different S/VHC; and (4) introduction of a second, different S/VHC to be used with a given innovator pMDI product. The following aspects should be reviewed in the preparatory stage of designing experiments to establish IVBE: (a) the inclusion of delayed inhalation; (b) the utilization of age-appropriate flow rates; and (c) the use of anatomically appropriate face models for evaluation of devices with a facemask. Statistical considerations that fit in with such experimental methods include: selection of pMDI batches and S/VHC lots; choice of sample size and acceptance criteria; bracketing or worst case approaches; and balanced/paired designs. A stepwise approach for selection of impactor stage groupings is presented, and an approach to determine realistic acceptance criteria based on REF product characteristics is suggested.
RESULTS
An example of an efficient statistical design of experiment is provided for each scenario, together with alternate approaches for calculation of confidence intervals for the mean TEST/REF relationship. It is important to appreciate that the optimal design depends on balancing numerous considerations and will thus likely differ from case to case; hence, the designs presented here should be seen as illustrations rather than the only option available. More effective approaches may be found that suit a particular case at hand.
CONCLUSIONS
The information provided will assist in developing correlations in support of IVBE for these add-on devices.
Collapse