Validation of the Cleaning Process for a Semi-Automatic Capsule Filler

Introduction
Conducting two clinical trials simultaneously involves the use of a semi-automatic filling machine (SAFM) (INCAP SE, Bonapace) to produce 100- and 200-mg caffeine (CAF) capsules, 10-mg prednisolone (PRD) capsules, and placebo (PCB) capsules. Good Preparation Practices 2023 require measures such as cleaning procedures to prevent cross-contamination. The objective of this study was to validate the cleaning process for the machine.

Materials and Methods
This study was conducted according to the A3P guide on cleaning validation, in line with Annex 15 of Good Manufacturing Practices (GMP). It was based on the assessment of residual surface contamination (RSC) and cross-contamination (CC) of the PCBs. The assessment of RSC required prior validation of the sampling and extraction method (SEM) with a target extraction recovery (ER) of ≥ 70% for both active pharmaceutical ingredients (API) through induced contamination tests on surfaces representative of the SAFM materials in contact with the powder: stainless steel and plastic. A 25 cm² surface area was contaminated by depositing a defined mass of API; a swab impregnated with a fixed volume of solvent was applied to the surface and then dissolved in a defined volume of extraction solvent. To evaluate the RSC, samples were taken from ten areas of the SAFM following a capsule testing campaign for each API, after minor cleaning (MC) and thorough cleaning (TC). The CRS limit threshold (LT) was defined by calculating the Maximum Allowable Carry Over (MACO) in accordance with GMP, considering the Permissible Daily Exposure (PDE) and the actual surface area of the SAFM in contact with the powder. For the CC assessment, the contaminant mass limit in the capsules was set by dividing the PDE by the number of doses per day. All analyses were performed by high-performance liquid chromatography coupled with diode array UV detection, according to validated methods.

Results
After optimization of the MPE, the mean ER values on steel and plastic were 90.2 ± 4.3% and 90.7 ± 6.3% (CAF); 93.0 ± 5.4% and 93.7 ± 5.0% (PRD). The calculated CRS limits were 1.56 mg/cm² (CAF); 0.43 µg/cm² (PRD). The calculated CC limit was 2.1 µg/capsule (PRD) and 7.55 mg/capsule (CAF). The most contaminated area after MC was the drainage channel: 12.21 µg/cm² (CAF); 4.56 µg/cm² (PRD). All CRS were acceptable after TC. No CC was detected in the capsules of any PCB.

Discussion/Conclusion
The methodology developed validated the RSE cleaning process. Due to its low toxicity, the calculated CRS LT for CAF was high; therefore, for maximum safety, a stricter LT was set at 10 µg/cm². No CC was to be detected in the capsules. The LODs of the methods could detect API masses below the established LT. This approach could be extended to other API and automated systems.