(M1130-01-03) A Clinical Study to Assess the Bioequivalence of Topical Diclofenac Products Using Dermal Open Flow Microperfusion

Purpose: In recent years, an in vivo dermal sampling method known as dermal open flow microperfusion (dOFM) has repeatedly demonstrated that it can accurately and sensitively compare the cutaneous bioavailability of drugs applied on the skin. In dOFM BE verification studies, equivalent and non-equivalent products have been dosed in parallel on the same subject and the bioequivalence (BE) of different acyclovir and lidocaine/prilocaine products has been assessed based on cutaneous pharmacokinetics (PK) endpoints. To evaluate dOFM’s capabilities to monitor the cutaneous PK of highly lipophilic and highly protein-bound drugs, we conducted another BE verification study using diclofenac sodium. This study comprised of a pilot and a pivotal study. The aim of the pilot was to optimize parameters for the pivotal study such as the topical product dose amount and duration of application, and to assess the suitability of the study design parameters by monitoring for any evidence of lateral diffusion between adjacent application sites and measuring drug concentrations in the systemic circulation that may redistribute to the skin. The suitability of a non-equivalent test product to be discriminated as a negative control for BE was also evaluated. In the pivotal study, BE was evaluated by comparing the PK endpoints for a reference gel product with those from an approved generic gel product (positive BE control), and those from a non-equivalent solution (negative BE control).

Methods: Voltaren (diclofenac sodium) topical gel 1% was used as reference and diclofenac sodium topical gel 1% as generic product. Pennsaid (diclofenac sodium) topical solution 2% served as non-equivalent test product. In the pilot study 6 subjects received each 20 dOFM probes across nine application sites; one site on the arm and four on each thigh. To evaluate dOFM’s ability to measure differences in bioavailability, three different doses (2, 10 and 50 mg/cm²) of the reference gel product were applied, and to assess lateral diffusion, three dOFM probes were positioned in untreated sites next to the treated application site. The non-equivalent solution was applied to one application site on one thigh. The site on the arm was untreated to evaluate any systemic drug redistribution. In the pivotal study 16 subjects received each 27 dOFM probes across seven application sites; one untreated site on the arm, and three application sites on each thigh. On each thigh the reference and the generic gel products were applied at a dose of 20 mg/cm², each to one application site, and to the third application site 10 mg/cm² of the non-equivalent solution was applied (Figure 1). All products were removed six hours post-application, and continuous dOFM sampling was performed for 25 hours. Dermal diclofenac was quantified with high performance liquid chromatography tandem mass spectrometry. The dermal PK endpoints area under the concentration-time curve (AUC) and maximum concentration (C_max) were used to evaluate BE by means of the scaled average bioequivalence (SABE) statistical approach.

Results: In the pilot study, the bioavailability from different doses of the reference gel was well differentiated. Dermal PK profiles from dOFM probes in untreated sites showed no quantifiable amount of diclofenac, indicating no significant lateral diffusion or systemic redistribution.

The PK profiles derived from the non-equivalent solution were well differentiated from those of the reference product. Results from the pilot study confirmed that the study design parameters were appropriate for the pivotal study; a reasonably complete PK profile was captured within the sampling period. To evaluate the impact of anomalous data on the results, a sensitivity study was performed to evaluate how inclusion or exclusion of anomalous data may alter conclusion from the study. In the pivotal study, when excluding anomalous data, the generic gel product was accurately found to be bioequivalent to its reference gel and the point estimates for each endpoint lay inside the BE limits (Table 1). However, when anomalous data were not excluded, these products did not meet the BE criteria. The negative control was sensitively discriminated and found not to be bioequivalent regardless of whether the anomalous data were included or excluded.

Conclusion: We demonstrated the ability of dOFM to measure the dermal PK of diclofenac, a highly lipophilic and highly protein-bound drug. Our results also demonstrated that dOFM was sufficiently sensitive to discriminate differences in bioavailability between different products (gel versus solution). Notably, we learned that an order of magnitude higher bioavailability from the solution product may cause lateral contamination of an adjacent probe treated with the gel product. This appeared to produce anomalous results that impacted the accuracy of the study to assess BE. The sensitivity study demonstrated that if such effects were removed or better controlled by randomization, the pivotal study would accurately determine the gel products to be bioequivalent. These results for diclofenac, and those from prior studies indicate that dOFM has the potential to support a demonstration of BE for a wide range of different topical drug products.