Microbial Monitoring for Biopharma Manufacturing

Biomanufacturing relies on the use of living cells as factories to produce biotherapeutics, or as therapeutics themselves. Most biologic drugs are highly specific in their mechanistic function, making them popular for the treatment of cancers, autoimmune diseases and infections.

Production and characterization of these drugs are exceedingly more complex than conventional small molecule drugs that are chemically synthesized. Cells must be grown and expanded under aseptic conditions. The use of good manufacturing practices (GMP) and strict quality control are applied to mitigate the risk of product contamination.¹

The World Health Organization (WHO) defines adventitious agents as contaminating microorganisms of cell culture or source material that have been unintentionally introduced into the manufacturing process of a biological product. This includes bacteria, fungi, mycoplasma, parasites and viruses.² These contaminants can be introduced through starting materials, such as cell substrates, or by environmental exposure, such as equipment, handling or personnel.

Contamination is a significant issue for bioprocessing as it results in loss of time, money and effort. Yearly revenue losses are estimated between $100–300 million due to contamination.³ Contaminated products that are not detected during production must also be pulled off shelves, which may result in shortages of life-saving drugs and vaccines.⁴

The risk of contamination necessitates the need to monitor for microbial contamination during bioprocessing to mitigate both product lot failures as well as adverse reactions to patients.

Detecting contamination during bioprocessing

Bioprocessing Residue Detection Solutions

Discover reagent kits for the detection of residual components across a range of biologics. 
View eBook

Advertisement

Bioprocess monitoring is a critical aspect of manufacturing as it ensures the final product meets the desired specifications, quality and safety requirements. Monitoring of contamination is typically achieved by the analysis of process parameters such as pH, temperature, dissolved oxygen, nutrient levels and cell density and viability. These variables are already monitored during processing as optimized conditions are required to produce quality products.⁵

Multiple established methods exist to measure microbial contamination:

• Enzyme-linked immunosorbent assay (ELISA): assesses the presence and concentrations of specific proteins or antibodies and can also be used to indicate the presence of bacterial contamination.
• Polymerase chain reaction (PCR): can be used to amplify specific DNA sequences of interest, particularly for those of microbial contaminants.⁶
• Process Analytical technology (PAT): may be able to detect contamination by statistical or mathematic techniques if process deviations are detected.^5,7
• Limulus Amebocyte Lysate (LAL): used to test for endotoxins that are released by gram-negative bacteria.⁸

Figure 1: Methods to detect microbe contamination in a bioprocess. Credit: Aron Gyorgypal, Adobe Illustrator.

Contamination’s influence on product quality and safety

Rapid Contamination Detection in Short Shelf-Life Biopharmaceuticals

Explore the benefits of rapid microbial detection.

View Poster

Advertisement

What are some of the key signs of cell culture contamination? “Because microbes multiply much faster than mammalian cells, the culture could quickly become significantly more turbid with a high concentration of bacteria. The microbes would also rapidly consume the sugar and oxygen in the culture, effectively starving the mammalian cells,” says Dr. Erica Berilla, senior research scientist in the Office of Pharmaceutical Quality Research (OPQR) within the Food and Drug Administration’s (FDA) Center for Drug Evaluation and Research.

Berilla’s research focuses on new manufacturing technologies used in the production of biologics and the methods used to monitor those manufacturing processes. Such projects include, but are not limited to, end-to-end continuous manufacturing, automated in-process sampling and microbial monitoring during biomanufacturing processes. This research can help guide and inform the FDA’s product quality assessors who may be encountering these new technologies during their assessments of protein drug substance products.

Berilla’s team wanted to understand microbial contamination in terms of product quality while also building an understanding of early contamination detection. To do this, they intentionally contaminated a Chinese hamster ovary (CHO) cell bioreactor that was used to produce a model monoclonal antibody (mAb).⁹ Near-infrared (NIR) spectroscopy was used to analyze the culture in 15-minute intervals to generate a partial least square (PLS) model. Though the model was not sensitive to detect low-level contamination it could detect contaminant in the later stages.

“We found that cell cultures contaminated with Mycoplasma arginini (M. arginini) plateaued in their growth once the mycoplasma concentration was maximal, which usually took two to three days after mycoplasma introduction… We noted significant increases in ammonium and depletion of arginine that was not seen in the control.” This result was perhaps to be expected, given that there is a competition for nutrients within the cell culture between the mammalian and the bacterial cells. Also, because bacteria are more metabolically efficient, there will be an increase in their proliferation and production of metabolic byproducts that are toxic to mammalian cells.

These experiments also gave insights into how the quality of mAbs is affected by M. arginini contamination. Principal component analysis revealed a correlation between contamination and increased antibody charge variants and hypoglycosylation – qualities that influence therapeutic efficacy and clearance.¹⁰

To mitigate the lost time and cost associated with manufacturing, there needs to be more advances in detection methods, Barilla says: “Technologies such as next-generation sequencing (NGS) and spectroscopy have the potential for use in earlier detection of adventitious agent events during bioprocessing. Further development and validation of these technologies could lead to more rapid/real-time testing.”

NGS hold potential for contamination monitoring as indicated by the 2024 ICH-Q5A(R2) guidance document published by the FDA, which has encouraged the use of the technology. In a recent study by Hiarai et al., the performance of NGS was evaluated for in vitro contaminant detection in biologic products.¹¹ Here the authors inoculated a model adenovirus into Vero cells and checked for contamination with increasing dilutions to find the limit of detection (LOD). The authors found that the LOD for this assay was roughly 2.49x10^-4 copies/ng-RNA, or about 1:10⁷ dilution from their original inoculate. This proved to be extremely sensitive, however there were some drawbacks such as the risk for false positives that may arise when multiplexing the assay. While these results are promising, the technology will need to be validated for contaminant monitoring before it can be implemented for routine use.

Emerging methods for detecting microbial contaminations

There are many different methods currently used to detect contamination. The method of choice will vary depending on the type of culture that is being run, such as CHO-cell-producing mAbs, human-derived cells for cell therapies or the production of viral vectors. “The first key factor to look at is the unit operation along with the sensitivity required of the assay,” says Dr. Richa Pandey, assistant professor of biomedical engineering at the University of Calgary.

Pandey's research group is focused on point-of-care diagnostics, but she has also taken an interest in using her expertise in biosensor technology, applying it to biomanufacturing. She recently published a review article that highlights the current methods for the analysis of biological contaminants within the biomanufacturing industry.¹² Pandey argues the need for continuous monitoring systems for microbe detection that can be enabled by optical and electrochemical sensors and biosensors equipped with microfluidics.

“Optical sensors work by interacting light with thin films that are functionalized with nanoparticles that, once interacted with by the antigen of interest, produce a signal. These sensors are quite advantageous as they are sensitive and require low volumes of analyte,” says Pandey.

An example of this technology was showcased by Zandieh et al., who demonstrated that the use of silver nanocolumns functionalized with a polycationic peptide, polymyxin B (PmB), to recognize and interact with lipopolysaccharide (LPS) a highly toxic endotoxin produced by gram native bacteria.¹³ The methodology allowed sensitivity as low as 340 pg/mL of endotoxin. Although the conventionally used LAL assay may be more sensitive, it suffers from the need for sample prep, which makes this optical sensor methodology more attractive for a streamlined and fast bioprocess contamination screening.

Figure 2: Representation of an optical (left) and electrochemical (right) biosensor for detecting contamination such as toxins produced by microbes in a bioprocess. Credit: Aron Gyorgypal, Adobe Illustrator.

Electrochemical-based sensors are another methodology that have been used to detect bioprocess contaminants. “Unlike the optical sensor, an electrochemical sensor measures binding events through either change in conductance, resistance or capacitance on its surface,” says Pandey.

The use of aptamer-based electrochemical sensors has gained interest in recent years. Their popularity is derived from their high affinity to a variety of targets, such as proteins, small molecules and living cells. Kim et al. used an endotoxin-specific aptamer to detect bioprocess microbial contaminants.¹⁴ This methodology showed reduced processing time against the conventional LAL, although with less sensitivity. Studies like this one showcased the use of aptamers for the detection of not only endotoxins, but also whole pathogens, and can be used to measure different compounds in a bioprocess culture.^15,16

Indeed, the use of optical and electrochemical sensors does have some advantages over current practices, given that they allow for faster detection with little to no processing. “Of course, there is merit to the current methods used for contaminant detection, but there still needs to be advances to make them more sensitive, as well as with automation to allow for at-line or in-line monitoring,” says Pandey.

Moving towards automated monitoring

Although the current gold standard technologies for contamination detection are sensitive and provide specific detection, they are still limited to batch monitoring. The current priority is to detect contamination quickly and accurately, without the need for off-line timely assays. The production of a real-time monitoring system capable of continuous contaminant monitoring would allow for early intervention that would prevent the spread of adventitious agents and reduce the risk for product recalls.