Testing for Host Cell Proteins

In this day and age, many biological drugs are produced at scale in cell culture. This method has many benefits but one of the big challenges it presents is purifying the drug from the cell culture and in particular making sure the final product has no remaining host cell proteins (HCPs).

Host cell protein detection is “the use of established analytical techniques: ELISA, LC-MS, etc., to detect the residual process-related protein impurities that are produced by the host organism during biotherapeutic manufacturing and production,” explains Brad J. Williams, Ph.D. Principal MS Applications Specialist at Waters.

Regulatory agencies around the world mandate and monitor the quantification, qualification, and removal of HCPs. And for good reason. “Many HCPs are benign, but some are immunogenic; some may interact with the drug substance, and others, like proteases and lipases, can reduce effective product dosage through direct action on the drug or its stability by interfering with the formulation buffer. Thus, they must be removed to the lowest level possible,” says Alla Zilberman, Ph.D., VP for Technical Marketing at Cygnus Technologies.

There are several options for the type of assay used for HCP detection but an HCP ELISA is the gold standard. Orthogonal methods include mass spectrometry, antibody affinity extraction, and 2D-PAGE (namely Western blot).

Developing the HCP quantification and removal assays is difficult, time consuming, and costly. “It takes 12–14 months to develop a well-qualified and broadly reactive HCP ELISA. It also takes a lot of expertise to develop a broadly reactive HCP antibody and then turn this antibody into a specific and sensitive immunoassay. This process is significantly more complicated as compared to developing an antibody to a single analyte protein,” according to Dr. Zilberman.

It is no surprise then that assay development runs in parallel to drug development. “Typically, two or more different biopharmaceutical departments perform the purification and analytical measurements. Both groups need to work closely together to efficiently purify and detect HCPs throughout the purification process,” adds Dr. Williams.

As drugs move through the various stages of development, the HCP detection method required becomes more stringent and exacting. “Typically, customers start with a generic, off-the-shelf HCP ELISA for preclinical process development through classification as an Investigational New Drug (IND) and into manufacturing drug substance (DS) candidate batches for Phase 1 and 2 clinical trials. When a DS candidate moves into Phase 3, the customer can either qualify and validate the generic kit if it proves to be a great fit for their process. Or they can develop a process-specific HCP immunoassay,” explains Dr. Zilberman.

Dr. Williams adds that “discovery experiments are focused on HCP identification and relative quantitation. Whereas, at the quality control level the HCPs are quantified using high-throughput chromatography combined in tandem with quadrupole mass spectrometry.”

Emerging fields

As biomanufacturing, and in particular the realm of biomanufacturing therapeutics, expands into new territories the need for HCP detection methods is also expanding. Technologies such as gene therapy, therapeutic mRNA/DNA, and hybrid protein-oligonucleotide conjugates will require not only using the current HCP detection tool kit but also expanding upon and innovating existing HCP detection methods. Work detecting HCPs in some of these emerging contexts is already underway. Dr. Williams points out that “recently LC-MS based HCP detection methods have been used to characterize lentiviral vectors and adeno-associated virus particles used for gene therapy.”

Limitations and future improvement

ELISA is the gold standard in HCP detection for a reason. The list of benefits includes ease of use once developed and wide dynamic range. Dr. Zilberman also notes that “ELISA is effective in relatively high levels of product protein, which makes it ideal for detection of HCP proteins at ng/mg concentration relative to drug substances.” Plus, it can be combined with liquid chromatography-mass spectrometry, which can detect many HCPs within a single assay.

But ELISA’s also have limitations, for example, they do not give information about which HCPs are present or which HCPs are going undetected. They also require a lot of expertise to develop for each drug. And they are prone to analyte interference, which can give rise to false positives or unreliable quantification.

Dr. Williams says an ideal HCP detection method would “involve simple sample preparation, and be selective, sensitive, robust, high throughput, and highly reproducible.”

In working toward this goal, many improvements have been made recently. In particular sample-preparation protocols have been significantly streamlined over the last couple of years. Improved methods include antibody precipitation, HILIC, and MWCO filter-based techniques. These methods “significantly reduce the level of the high concentration biotherapeutic while concentrating the HCPs. Identifying the ‘preferred’ sample-preparation protocol among the HCP community is something that needs to be determined for the future,” Dr. Williams notes.

Many of the previously discussed methods for HCP detection used fragmented or denatured proteins as input due to assay limitations and large proteins. This is another area that could be improved. Dr. Williams points to HCP detection at the intact protein level as a promising future direction for the field. He says that working with intact proteins “could streamline HCP detection while minimizing sample preparation and increasing throughput.”

The most recent generation of mass spectrometers have also greatly improved. Dr. Williams notes that “they have improved sensitivity, mass spectral resolution, and ion mobility separation that can enhance the overall selectivity (mass spectral clarity) to enable the detection of low-level HCPs.”

Ultimately, he is very optimistic that HCP detection methods are going to continue to improve. “I feel that continued improvement in many technological areas including sample preparation, separation science, mass spectrometry, and informatics are on a trajectory to reach the ideal HCP workflow within the next five years,” he says.