BioProcess International

by Audrey Chang, Gail Sofer, and Sandra Dusing

Regulatory documents that address clinical biotherapeutics such as monoclonal antibodies and recombinant DNAderived proteins are written in broad generalities to accommodate variability in the types of products, patient populations, dosage, and product indications they cover. However, as a result of this flexibility, sponsors are often confused about what is required for regulatory compliance. A good first step is to read the regulations that might be applicable (see the “Regulatory Documents” box). Second, perform a risk–benefit analysis and evaluate appropriate patient safety issues. Consider what technological advances will be beneficial. Finally, make a plan and arrange to discuss it with the appropriate regulatory agency.

In a pre-IND (investigational new drug application) meeting with the US Food and Drug Administration, be prepared to discuss facility and responsibility issues, source and processing materials, upstream and downstream processes, and the analytical methods that will allow decisions to be made about the suitability of raw materials, the manufacturing process, and the safety of the final product. You will also need to discuss specific risk factors and your efforts to minimize them.

Here we address some specific risk factors and analytical methods used to detect and minimize those risks for products produced by mammalian cell culture.

Endogenous and adventitious risks Risk factors may be inherent in the cells used to make a product and in raw materials such as cell culture media, or they may be adventitious.

Potential endogenous risks from cell banks and production cells include those associated with viral agents such as retroviruses, host cell proteins and nucleic acids, and other impurities. Retroviruses may be infectious or noninfectious, depending on the cell type. In either case, they must be removed during processing. Host cell proteins, host nucleic acids, and impurities added during processing — detergents, endonucleases, and resin leachables such as protein A — must also be consistently removed from the product.

Adventitious agents are those not expected to be present. An FDA speaker noted that for phase I, the agency expects to see data on the prevention and control of contamination by adventitious microbial agents (viruses, bacteria, fungi, mycoplasma) and transmissible spongiform encephalopathy (TSE) agents (1).

The European Union’s GMP Annex 13 on the manufacture of investigational medicinal products notes that “virus inactivation/ removal and removal of other impurities of biological origin should be no less rigorous than for licensed products” (2).

Generally, during clinical development, products and their impurities are not as well understood as when products are licensed. As a result, more inprocess and final product testing is usually applied to provide consistently safe biotherapeutics. The use of relatively rapid methods such as polymerase chain reaction (PCR) and generic enzyme-linked immunosorbent assays (ELISAs) can aid in process development and permit process improvements to be made during clinical studies without increasing safety risks. Such methods can be used as in-process tests for determining clearance during downstream processing, and they are suitable for assaying residuals in final purified bulk materials and the final product. Rapid methods for quantifying DNA, host cell proteins, proteins added during processing, and retroviruses are being used today to enhance product development and patient safety.

Nucleic acids
DNA is regulated as a cellular contaminant rather than as a risk factor that requires removal to extremely low levels (3). The World Health Organization and the European Union allow for up to 10 ng of residual DNA per dose (4). The FDA has published that an acceptable residual amount of DNA is 100 pg/dose (5). However, for products requiring large doses (such as monoclonal antibodies) amounts greater than 100 pg/dose, up to 10 ng/dose, may be accepted after discussion with the appropriate FDA division reviewer.

Currently used assays measure either total DNA or species-specific DNA. The Threshold system from Molecular Devices determines total DNA. Methods that detect species-specific DNA include slot blot and quantitative PCR (Q-PCR). The latter two methods require the use of specific probes. Of the two, Q-PCR is more sensitive, and it is highly reproducible. For Chinese hamster ovary (CHO) DNA standards, Q-PCR is reproducible over a 6-log dynamic range. Validation studies carried out in two laboratories by three operators using two machines and involving nine individual runs on different days have shown that this assay has an accuracy of 0.3 log10, a quantitation range from 100 pg to 3 fg, and a detection limit of 1 fg. Q-PCR is ideally suited for assessing the amount of DNA that is removed during downstream processing, lot release of clinical batches, and assessing comparability when process changes are made. Unlike viral clearance studies, samples determining DNA clearance can be obtained from at-scale runs.

Regulatory documents

Points to Consider in the Manufacture and Testing of Monoclonal Antibody Products for Human Use; US Food and Drug Administration, February 1997

Guidance for Industry: IND Meetings for Human Drugs and Biologics, CMC Information; US Food and Drug Administration, May 2001

Guidance for Industry: Content and Format of INDs for Phase 1 Studies of Drugs, Including Well-Characterized, Therapeutic, Biotechnology-Derived Products; US Food and Drug Administration, October 2000

ICH Q7a: Good Manufacturing Practice Guide for Active Pharmaceutical Ingredients, November 2000

Good Manufacturing Practices Annex 13: Manufacture of Investigational Medicinal Products, Draft 1 November 2001, European Commission

Host cell proteins
Determining the consistency and quantity of host cell proteins in unprocessed bulk, recovery and downstream processing steps, and final product can be problematic in analysis and comparability studies during clinical trials. Host cell proteins are often highly immunogenic, and slight changes in the proteins and their amounts can affect patient safety in the clinic.

Assays for host cell proteins typically require development of an ELISA. Briefly, you do that by first making cells without the gene for the product — null cells — and then culturing them at the scale and under the conditions for which a license application will be submitted. The cell lysate, unprocessed bulk, or partially purified product is injected into rabbits or other large animals to produce antibodies that are screened for reactivity and pooled. The antibodies are used as capture and detection reagents for ELISA development.

Problems with ELISA: It is obvious that this technique has some inherent problems. The host cell protein pattern in the null cells is assumed to be the same except for the absence of the product. Some host cell proteins are more immunogenic than others, and there may be some for which antibodies are not produced. The host cell proteins are likely to change when cell culture conditions and scale changes are made. The ELISA that has been developed against the parental cell strain is unlikely to accurately measure specific host cell proteins that are modified by changes in culture.

When the inherent problems in the assay and its development are compounded with process changes, the utility and validity of the assay results may be questioned. As a result, before the process is finalized it is common to use generic assays that use antibodies directed against a broad population of host cell proteins derived from commonly used prokaryotic and eukaryotic host cell substrates, such as E. coli and CHO. These generic assays can reduce development costs, provide sufficient data to assess comparability of host cell protein patterns when process changes are made, and be used for lot release of phase I and II clinical products. The CHO ELISA, for example, is sensitive at ng/mL levels and has been validated over a dynamic range of 2–75 ng/mL with a quantitation limit of 2 ng/mL. The SP2/0 ELISA has a broader dynamic range, from 2 to 200 ng/mL (Cygnus Technologies).

Other process impurities
The removal of cell culture media proteins must be demonstrated by assay. Albumin is a commonly used protein, and ELISAs are available for both bovine and human serum albumin. After cell culture steps, processing additives may be used, and they often vary in size and complexity. Small molecules such as guanidine hydrochloride, detergents, and even proteins may be added to the process. For example, Benzonase (Merck) is a nuclease that is sometimes added to initial feedstreams to reduce DNA-related viscosity for downstream processing. Proteins are also used as ligands for purification. Removal of potential leachables from protein ligand affinity chromatography such as protein A is typically measured in at least one subsequent downstream step. ELISAs can be used to demonstrate where in the process protein A and Benzonase are removed. They are also suitable for lot release of clinical products.

Use Defined Materials: To detect substances for which validated assays are not readily available, sponsors benefit from using defined materials. If assays are available but not validated, their use can still reduce the time required for IND approval. During development, optimize those assays for the sample to be tested. All too often that aspect of development is ignored, which can result in a clinical hold because unknown materials in the final product certainly have the potential to affect patient safety. When designing the process for clinical manufacturing, it is advantageous to evaluate every component used and perform a risk assessment. A combination of inprocess and lot release testing will generally be necessary to ensure provision of a safe product.

Viruses
The potential for viral contamination of biotherapeutics is a significant concern. Although to our knowledge no biotechnological product has ever caused transmission of an infectious virus, viral contamination of processes and even final product have occurred. Examples of contamination include the murine minute virus (MMV) in CHO cell cultures and reovirus in the production of monoclonal antibodies.

Viral clearance studies can provide a high degree of confidence in the ability of the manufacturing process to produce a safe product. Because PCR is a molecular assay, it can be used to generate data quickly to assess process development parameters. Instead of waiting and hoping the process will provide enough clearance to begin clinical trials, a sponsor can screen the process while it’s under development. This can prevent costly clinical delays. PCR has been proposed by researchers in the FDA’s Division of Monoclonal Antibodies as an ideal method for investigating and optimizing retroviral clearance in purification processes (6).

Furthermore, the use of PCR in conjunction with infectivity assays has enhanced understanding of viral clearance mechanisms. For a given viral clearance step, it is now feasible to attribute clearance to removal, inactivation, or a combination of both. PCR will detect both active and inactive virus; whereas infectivity assays such as TCID50 (tissue culture infectious dose) and PFU (plaque forming units) will pick up only infectious units. Table 1 illustrates the clearance of a murine leukemia virus (X-MuLV) on a hydrophobic interaction column. The spiked load detected by Q-PCR is greater than that detected by the TCID50 because PCR picks up both inactive and active species. The table illustrates that infectivity is completely removed, although some of the viral sequence remains.

Table 1: Clearance of X-MuLV by a hydrophobic interaction column.
The>= indicates complete inactivation by the TCID50 assay.

Samples	Total Virus (log10) TCID50	Total Virus (log10) Q-PCR	Clearance (log10) TCID50	Clearance (log10) Q-PCR
Spiked load	7.4 ± 0.43	10.51 ± 0.07
Peak	=3.10 ± 0.43	3.71 ± 0.16

Q-PCRcan be used over a wide dynamic range and can accommodate frozen samples. It is being used to quantify retroviral particle load in CHO cell cultures (7, 8). Electron microscopy (EM) has traditionally been used and is the official method. Sponsors should discuss the appropriateness of PCR as a replacement for EM with their reviewers. Table 2 compares EM, CHO particle reverse-transcriptase PCR (RT/PCR), and quantitative product-enhanced reversetranscriptase PCR (Q-PERT). With increased sensitivity, smaller sample volumes, higher sample throughput, good reproducibility, and robustness, PCR assays are ideally suited for process development and cell line selection. PCR is also a rapid method for detecting the presence of adventitious viruses such as MMV, which could be introduced at the cell bank or culture stages or during downstream processing.

Whereas the sensitivity of assays for DNA and host cell proteins may enable detection at pilot or manufacturing scale, virus clearance studies are always performed at small scale in a spiking study. The same holds true for TSEs.

Table 2: CHO retrovirus-like particle (RVLP) assay comparisons.

Assay	EM CHOp	Q-RT/PCR	Q-PERT
Method	Particle count	Real time RT-PCR amplification	Real time RT activity detection
Target	Particle	CHOp RNA	RT activity
Volume of sample	mLs to L	140 µL	100 µL
Duration	five days	one day	one day
Specificity	RVLP	CHO particles	Retroviral RT
Sensitivity	106 particles/mL	103 particles/mL	106 RT molecules/mL
Throughput	Low	High	High

CHOp: CHO particle; RT: reverse transcriptase; PERT: product-enhanced RT

TSES
The best strategy for the prevention of TSE transmission is clearly appropriate sourcing of raw materials. In some cases, a full history of cell lines is unavailable, and for some regions of the world, sponsors have had to perform TSE clearance studies. This is clearly not an ideal situation. The Western blot, however, does provide a good method for screening the process to demonstrate where TSEs could be removed. Data have been presented to show the correlation of the Western blot with the infectivity assay, which requires many animals and up to 200 days or more to complete (9).

Assay validation
Although all the assays discussed here should be validated, it is essential to use appropriate positive and negative controls with each assay and to establish appropriate dilutions for test materials, if necessary. Each time the test article changes, it is essential to ensure that the assays still provide a suitable test system. The assays should also be maintained for future use. Even if a given assay is dropped as an inprocess or product test, it will be useful for comparability studies and for evaluating deviations and out-ofspecification results. The use of generic, readily available, validated assays saves time and permits the production of consistently safe biotherapeutics. New assays are continually being developed and can enhance process and product understanding for early clinical studies as well as for licensed product. Manufacturers of licensed biotherapeutics are expected to keep up with technological advances. The best time to begin using new assays, however, is in development so they can be part of a license application package.

References
1 Joneckis, C. The Road to Approval: CMC Common Pitfalls. Presented at the BIO Annual Meeting, Boston, MA, June 2002.

2 Good Manufacturing Practices Annex 13: Manufacture of Investigational Medicinal Products. July 2003, European Commission; http://pharmacos.eudra.org/F2/eudralex/ vol-4/pdfs-en/anx13en.pdf.

3 Griffiths, E. WHO Expert Committee on Biological Standardization Highlights of the Meeting of October 1996. Biologicals 1997, 25(3): 359–362.

4 CPMP Position Statement on DNA. CPMP/BWP/382/92

5 Points to Consider in the Manufacture and Testing of Monoclonal Antibody Products for Human Use; US Food and Drug Administration, February 1997.

6 Brorson, K; et al. Use of a Quantitative Product-Enhanced Reverse Transcriptase Assay to Monitor Retrovirus Levels in MAb Cell-Culture and Downstream Processing. Biotechnol. Prog. 2001, 17(1): 188–196.

7 De Wit, C; Fauatz, C; and Xu,Y. Realtime Quantitative PCR for Retrovirus-Like Particle Quantification in CHO Cell Culture. Biologicals 2000, 28: 137–148.

8 Brorson, K; et al. Evaluation of a Quantitative Product-Enhanced Reverse Transcriptase Assay to Monitor Retrovirus in MAb Cell-Culture. Biologicals 2002, 30: 15–26.

9 Lee, DC; et al. A Direct Relationship Between Partitioning of the Pathogenic Prion Protein and Transmissible Spongiform Encephalopathy Infectivity During the Purification of Plasma Proteins. Transfusion 2001, 41(4): 449–455.

Audrey Chang, PhD, is senior director of biologics and LADS, Gail Sofer is director of regulatory services, and Sandra Dusing, PhD, is director of immunoassay and clinical trials testing services at BioReliance, 14920 Broschart Road, Rockville, MD 20850, gsofer@bioreliance.com, fax 1-301-610- 2590.

To make any comments on this article, or to ask a question of the author, please contact the publisher. If you would like to submit an article please subscribe to our PL Intelligence service.

The opinions expressed in the articles published in this section do not necessarily reflect those of Pharmalicensing or UTEK Corporation. No actions including proposals to or agreements with other companies should be taken by any reader without obtaining specific business or legal advice. Neither the publisher nor the authors accept any liability for any actions or activities undertaken by any reader or other third party as a consequence of these articles or for any errors or omissions therein.