CHARACTERIZATION OF THE PRODUCT

Because of their targeted mechanisms of action, means of manufacture, and species specificity, biologic agents must be well characterized by potency, identity/quality, purity, and stability. Typically, ex vivo assays are used to assess potency, and it can be challenging to adapt one or multiple assays to be practical or feasible for routine use in a commercial manufacturing process. Requirements for identity and purity must be well defined, and these as well as stability may change as the manufacturing process is refined. Because of the potential variability introduced by changes in the working cell bank, it is ideal to define the process and initiate scale-up of manufacturing as soon as proof of concept of clinical activity has been obtained. Finally, formulation and route of administration should be decided early in the development process. FDA may allow manufacturers of biologic products to make manufacturing changes without conducting additional clinical efficacy studies but only if “comparability” data demonstrate that the product after the manufacturing change is still identical by assays for purity, potency, and biologic activity. Changes in either formulation or manufacturing process may force “bridging” studies to demonstrate that the newer product is not only identical by potency, purity, and identity as well as pharmacokinetic (PK) profile, but that it also performs similarly in the clinic. Such bridging studies may introduce significant delays in clinical development programs.

Manufacturing Process

Although manufacturing and testing practices typically evolve during product development, scale up is risky if it changes the product characteristics. Often a source of regulatory concern, it should be performed sooner rather than later, for example, before phase 3, and must be dealt with proactively ( Fig. 33-1 ).

Flow diagram for manufacturing processes of biologic agents.

Glycosylation occurs in the endoplasmic reticulum and Golgi apparatus as a post-translational modification of protein production. It is absent in Escherichia coli but present in yeast and highly conserved among mammalian cells (mouse myeloma, baby hamster, Chinese hamster ovary [CHO] and human), according to the cell line selected for manufacture and working cell bank. Of the tumor necrosis factor-α (TNF-α) inhibitors, adalimumab and etanercept are manufactured in CHO cells, whereas the aglycosylated F(ab)’-PEG product, certolizumab pegol, is manufactured in E. coli .

LESSON 1: DIFFERENCES IN GLYCOSYLATION PATTERNS DUE TO PRODUCTION IN MICROBIAL VERSUS MAMMALIAN CELL LINES: GRANULOCYTE MACROPHAGE/GRANULOCYTE COLONY STIMULATING FACTOR

Five different granulocyte macrophage (GM)/granulocyte colony stimulating factor (G-CSF) molecules have been produced in various expression systems. Bacterial expression ( E. coli ) yields aglycosylated molecules (molgramostim and filgrastim), yeast expression ( Saccharomyces cerevisiae ) yields O-glycosylated molecules (sargramostim) and mammalian expression (CHO cell line) yields fully glycosylated molecules (regramostim and lenograstim).

• •
  

  In prospective randomized placebo controlled trials, only glycosylated CSFs have been associated with improved patient survival (sargramostim and lenograstim).

  
  
• •
  

  In vitro and in vivo data generated from comparative studies have documented differences in biologic activity among CSFs.

  
  
• •
  

  A literature review of clinical trials evaluating sargramostim (glycosylated) and molgramostim (aglycosylated) concluded that molgramostim was associated with a higher incidence of adverse effects than sargramostim, as shown in the following table.

  
  
  
  

  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  

  Adverse Effect	Pooled Median Frequency (%)
  	Sargramostim	Molgramostim
  Fluid retention	8.3	18.4
  Dyspnea	13.4	55.2
  Fever	21.7	40.7
  Myalgia/bone pain	16.0	28.5
  Rash	14.3	12.5

   
  
  

Other manufacturing platforms include transgenic plants and animals as well as live animal platforms such as mammary glands of living goats. Glycosylation largely determines tertiary, even quaternary structure; binding to the selected epitope as well as immunogenicity. Several examples exist in which changes in the working cell bank due to scale up issues resulted in products with differing glycosylation profiles, resulting in loss of efficacy despite identity by high performance liquid chromatography (HPLC) and other in vitro assays.

LESSON 2: RISKS OF CHANGING MANUFACTURING PROCESSES DURING CLINICAL DEVELOPMENT: PRIMATIZED ANTI CD4 MONOCLONAL ANTIBODY AND LENERCEPT

Monoclonal antibody (mAb) effector function and clearance are affected by glycosylation patterns, which depend on the production cell line [working cell bank] and culture media.

• •
  

  This was illustrated with a nondepleting primatized anti-CD4 mAb following a change in the working cell bank due to scale up requirements, as well as removal of products of bovine and human origin in the manufacturing process. Although the inserted coding construct remained identical, a new CHO cell line adapted to growth in cell-free media was used, resulting in sustained rather than transient T-cell depletion, infusion reactions, and less efficacy.

  
  
• •
  

  Similarly, attempts to improve manufacturing yield of the p55 soluble TNF-α receptor I (sTNF RI-Fc) construct lenercept resulted in inter-batch differences in glycosylation patterns with loss of efficacy, despite identity in assays of purity, potency, and specificity.

  
  
• •
  

  Differences in glycosylation patterns accounted for rapid initial phases of clearance, altering the half life and analytical ultracentrifugation (AUC) of the molecule.

  
  
• •
  

  Production processes can also affect immunogenicity, when aggregates form after lyophilization (see Lesson 3 ).

Changes in manufacturing processes, and even the final formulation, may also introduce variability, which is exemplified by the alterations in final manufacture of erythropoietin, which resulted in cases of pure red cell aplasia ascribed to increased immunogenicity of the altered product.

LESSON 3: SUBTLE CHANGES IN MANUFACTURING PROCESSES CAN SIGNIFICANTLY AND DETRIMENTALLY ALTER IMMUNOGENICITY, DEMONSTRATED BY AGGREGATE FORMATION WITH RECOMBINANT ERYTHROPOIETIN

Since 1998, approximately 200 suspected cases of pure red cell aplasia (PRCA) due to neutralizing antibodies have been reported in patients with chronic renal failure receiving erythropoietin treatment. The majority of cases were associated with the use of epoetin alfa distributed outside of the United States, manufactured at a different site than that distributed within the United States. Immunogenicity of this preparation appeared to occur when polysorbate 80 and glycine were substituted for human serum albumin in the formulation to comply with new regulations from the European regulatory authorities.

It is believed that the new formulation was less stable, so that aggregates might form under adverse storage circumstances, such as prolonged exposure to increased temperatures. The presence of aggregates increase the likelihood of antibody formation following subcutaneous administration, and antierythropoietin antibodies observed in patients with PRCA appear to be directed against the protein moiety of the molecule. It is not known whether subtle differences between the carbohydrate moieties of epoetin alfa and beta have had any impact on this observed immunogenicity.

Preclinical Data

With a well-characterized mechanism of action, animal models are considered to have high relevance to clinical use of biologic agents. Nonetheless they must be viewed with caution in terms of their interpretability and predictive value to human disease. Owing to the lack of species homology for many products, murine-specific analogs may be used in disease models in rodents. With sufficient homology regarding target binding or mechanism of action, pharmacodynamic (PD) effects in primates may be informative. Clearly, differences in the underlying immune systems may result in unexpected or difficult to interpret findings. Most products designed for treatment of RA have been studied in adjuvant-induced (AIA) or collagen-induced (CIA) arthritis in murine systems: rat or mouse. CIA in rhesus monkey offers the potential opportunity to use the human product, but the arthritis in this model is often unpredictable and very severe—and forces use of a valuable animal resource. Gene therapy products have been extensively studied in antigen-induced arthritis in rabbits; B-cell targeted therapies by demonstrating B-cell depletion in non-human primates ( Table 33-3 ).

Table 33-3

Only gold members can continue reading. Log In or Register to continue

Share this:

• Click to share on Twitter (Opens in new window)
• Click to share on Facebook (Opens in new window)

Related

Related posts:

Management of Rheumatoid Arthritis: Principles and Strategies for Antirheumatic Pharmacotherapy Juvenile Arthritis Management of Medium and Small Vessel Vasculitis, Including the Antineutrophil Cytoplasmic Antibody–Associated Diseases Gout and Crystal Deposition Disease Mechanisms and Treatment Strategy of Complex Regional Pain Syndromes Regional Disorders of the Neck, Shoulder, Arm, and Hand

Stay updated, free articles. Join our Telegram channel

Join