Preclinical safety testing of species-specific proteins produced with recombinant DNA-techniques. An attempt to transfer current experience into future testing strategies

Overview of the nonclinical quality and toxicology testing for recombinant biopharmaceuticals produced in mammalian cells

Biopharmaceuticals represent significant advances in therapeutic approaches for unmet medical needs, and increasingly, traditional pharmaceutical firms have been incorporating biotechnology capabilities into their product portfolios. There are some differences in the overall safety testing paradigms for small molecules and biopharmaceuticals, this safety testing including both quality and toxicology aspects. These differences are associated with both the manufacturing processes involved and the molecules themselves. For example, for biopharmaceuticals, living cells represent the factories for synthesizing complex molecular entities. As a result of this, safety testing for this class of drugs includes adventitious agent testing (e.g. viral, mycoplasma, transmissible spongiform encephalopathy agents) not normally needed for small molecules. Also, strategies for nonclinical toxicology testing of biopharmaceuticals differ from the paradigms used for small molecules and often need to be defined on a case-by-case basis, primarily taking into consideration species cross-reactivity attributes of the molecule of interest. Certain studies required for small molecules are not applicable to most biopharmaceuticals (i.e. genotoxicity testing, testing for interactions with the hERG channel). This manuscript provides an overview of both the quality and nonclinical toxicology testing for these mammalian-cell-derived products, two elements pivotal to the overall nonclinical assessment of the safety of these biopharmaceutical products.

Issues with biotechnology products in toxicologic pathology

The emergence of the biotechnology industry and introduction of drugs derived from recombinant DNA technology has generated many new issues in approaches to preclinical safety evaluation and extrapolation of results to risk assessment in humans. Products or therapeutic approaches for consideration include hormones, growth factors, cytokines, monoclonal antibodies, vaccines, blood products, antisense, and gene therapy. In many instances the application of standard safety tests conventionally used for small molecules are of limited value or are inappropriate. Studies should be designed to answer specific scientific questions rather than simply to fulfill regulatory requirements. Special consideration must be given to study design and species selection in terms of biological activity and species specificity, implications of immunological responses in the animal studies, and effects of systemic administration of molecules at clinically relevant doses. A full understanding of the clinical relevance of toxicological and pathologic findings associated with administration of these molecules to laboratory animals requires definition of the pathogenic mechanism of lesion induction.

Comparative pathology of recombinant murine interferon-gamma in mice and recombinant human interferon-gamma in cynomolgus monkeys

Interferon-gamma is a highly species-specific cytokine and has the most restricted host range of activity of the interferons. Recombinant human IFN-gamma was one of the first species-specific recombinant proteins to be thoroughly assessed in conventional safety models used for xenobiotics. Acute single-dose intravenous toxicity studies with rHuIFN-gamma were performed in rats, marmosets, and squirrel monkeys with no indications of toxicity. A complete series of subchronic toxicity studies and segment I and II reproductive studies in the rat revealed no evidence of toxicity at any of the doses tested. These results suggested that studies conducted in pharmacologically nonresponsive species may not be predictive of clinical toxicity. Human IFN-gamma is active on nonhuman primate cells, though not at the same level as on human cells. Multidose studies in cynomolgus monkeys with rHuIFN-gamma for 28 or 90 days were predictive of many of the dose-limiting clinical toxicities. Qualitative similarity was observed between toxicity studies employing rHuIFN-gamma in the cynomolgus monkey and rMuIFN-gamma in the mouse. The adverse effects seen in toxicity studies with cytokines and growth factors are often exaggerated pharmacological effects of the molecules, and therefore can only be studied in a responsive species. In situations in which a high degree of species specificity is encountered, studies employing a recombinant protein in a homologous species may provide a useful test system for preclinical safety assessment.

Effects of recombinant human alpha-interferon in a rodent cardiotoxicity model

Recombinant human alpha-interferon infused intravenously into rats at doses of 10(6) (1st infusion) and 10(5) IU (2nd and 3rd infusion) produced marginal evidence of liver damage but no serious toxicity. During the 2nd and 3rd infusions an increased incidence of cardiac arrhythmias of various types was observed. In 2 rats electrocardiographic evidence of myocardial ischemia was noted. No changes in myocardial structure were demonstrated by light and electron microscopy. With the high dose the decrease in body temperature resulting from anesthesia was significantly reduced. Antibodies to interferon were demonstrated in the majority of the animals.