November 1, 2014 (Vol. 34, No. 19)

Jenny Ekegren technical sales manager Cobra Biologics

Focus on Biomanufacturing Operations in a Small-Volume Aseptic Filling Environment

Fill/finish operations for biotherapeutics are a critical, but often overlooked, hurdle in the production process for these products, especially for early-phase clinical evaluation. From a processing perspective the products are complex and fragile; and terminal sterilization is not an option.

All products need to be sterilized by point-of-fill filtration which, from a regulatory perspective, is a challenge in itself. From an operational perspective, the products are often highly potent, hugely expensive, and complex to produce for early-phase studies. And, only small amounts of material are available. In some cases, about 1 L of drug substance may have been produced at costs in excess of $500,000.

As a result, losses incurred during the filling operation need to be kept to an absolute minimum, not least to ensure that sufficient filled drug product is available to support planned clinical studies.

It is essential that the approach taken by fill/finish facilities to produce the drug product addresses these issues and seeks to reduce process losses to an absolute minimum while retaining regulatory compliance with regard  to testing and quality assurance for these operations.


In 2009, regulatory authorities approved an advanced facility in Matfors, Sweden, for provision of fill/finish services for clinical and commercial supply, meeting the rigorous sterility requirements in accordance with aseptic manufacturing procedures. It was designed to perform inline 0.22 µm sterile filtration integrated in a fully disposable system to fill either vials (0.5–100 mL) or pre-filled syringes (0.5–3 mL).

From the outset, the facility was designed around the use of single-use filling systems to enhance operational flexibility, while retaining the highest levels of regulatory compliance. Figure 1 shows an overview of the fill/finish setup, including classification of surroundings. Shortly after a successful validation of the fill line, regular manufacture of a commercial small molecule API was running in the facility.

For this low-cost API, no special steps were taken to minimize drug product losses, and the system was flushed and the filter integrity tested with the bulk solution. In total, up to 1 L of bulk was used (and discarded) for these purposes.

Cobra Biologics acquired the facility in 2011, and the requests for fill/finish of biopharmaceuticals increased accordingly. Many of the products generated by Cobra are for early-phase clinical evaluation and, as such, are often products in relatively small quantities of <5 L, where existing process losses of 1 L would represent a loss of > 20% of the production batch. Clearly this would not be acceptable.

To better accommodate these high-value, small-volume fillings, a project was initiated to modify the existing fill process with the aim of minimizing product losses. 


Figure 1. Overview of fill/finish line design and summary of improvements leading to reduced fill volume losses.

Objectives

The initial project has focused on a detailed assessment of the equipment and filling procedures. Here, a number of challenges and constraints for the reduction of system volume losses were identified. To minimize revalidation work, staff requalification, and costs, major design changes were excluded in the first instance. Therefore, the project sought to use the existing disposable system design and with only minor changes—smaller fill bags and tubing with smaller diameter to enable the desired volume loss reduction.

With these constraints at hand and by comparing the design and capabilities of the fill/finish suite at the time of approval with current expectations on a successful small-volume, high-value fill, the goals were set for the project:

Initial Goals

  • Customers required automatic aseptic filling of large volume of low-cost drug substance
  • Manufacturing losses: approximately 1 L

Goals Today

  • Adapting existing automatic fill/finish systems
  • Adopting disposable solutions to customer-specific needs: small volume of high-cost drug substance
  • Improving working procedures
  • Manufacturing losses: approximately 100 mL

Method and Results

The entire fill process, including staff, equipment, and process flow was mapped and key target points during both filtration and filling to be addressed were identified:

Key target points during filtration

  • Saturation of the filter
  • Remaining substance in compounding bag
  • Losses during filtration

Key target points during filling

  • Losses due to filling weight adjustment
  • Losses during set up of the filling tubes
  • Loss of fluid due to failing check weights at end of filling
  • Loss of fluid trapped in the tubes and bag (line losses)


In a series of experiments, various modifications to address the above key points were evaluated. As a start, smaller sized formulation and filling bags were installed as well as tubing with smaller diameter and a sterile filter with a smaller filter area. Work with the staff determined the best way to practically handle the bags and tubing, and the first steps were taken to minimize the line losses in terms of incomplete emptying of those (Figure 1).

Although just one of two fill needles available in the line was used, the number of fill weight controls was reduced to half. To address the issue of losing drug product solution when saturating/rinsing the sterile filter before the fill, a procedure where first water for injection and then buffer of an appropriate composition was used to flush the filter was developed. How far into the fill, correct fill weight controls could be obtained was also challenged to make sure as much as possible of the drug product solution was filled before the weight control failed.


To verify the effect of the modifications, three test runs were performed; the first with water and the other two with buffer. Between 1.0 L and 1.5 L of water or buffer was sterile filtered into the 10 L filling bag and then filled into around 400 vials, size 10 mL, with a fill volume of 3 mL per vial. The results are shown in Table 1. The total loss of drug product solution could in all runs be reduced to less than 100 g of solution, compared to up to 1 L loss with the unoptimized methodology.

Successful improvement of the fill process, based on a commercially available filling equipment and tubing sets, resulted in a more than 10-fold decrease in product waste. This was achieved by careful mapping of the fill process followed by small changes in the working procedures and the choice of disposables.

Having optimized the filtration process, Cobra’s ongoing project will be to further minimize volume losses, enhancing customer product yields, by thorough review of drug product sampling and validation procedures.


Table 1. Losses during filtration and filling for runs 1-3.


























Jenny Ekegren (jenny.ekegren@cobrabio.com) is technical sales manager at Cobra Biologics.

This site uses Akismet to reduce spam. Learn how your comment data is processed.