Masterstudiengang "Drug Regulatory Affairs"
Evaluation of Non-clinical Requirements for Medicinal Products containing Monoclonal Antibodies as an Active Substance based on review of EPARS during last 10 years (2010-2019) ***
Dr. Ekaterine Schäfer (Abschlußjahr: 2020)
Although the non-clinical testing program of monoclonal antibodies takes into consideration all sections of the pre-clinical development program for conventional small molecules, the non-clinical testing program needs to be specifically adapted for mAb therapeutics. This flexibility is also recommended by the ICH-S6 guideline, applying a case by case, science-based assessment approach.
The establishment of relevant animal models to assess pre-clinical safety and to establish FIH dosing range is a key element for the preclinical programme of the monoclonals. The identification of relevant species was supported by a number of different assays in almost all cases. There are some mAb therapeutics, for which no or more than one cross-reactive animal species exist. But more common are products with only one pharmacologically active species (about 2/3 of cases). The non-human primate (NHP) cynomolgus monkey (Macaca fascicularis) was an active species in almost all cases, with only few exceptions.
The comprehensive primary pharmacodynamic in vivo and/or in vitro studies were conducted for all reviewed products. In vivo studies for the detection of pharmacological effects in products containing a new active substance were conducted in well-adjusted model systems. Active species were used for the purpose of validating the reactivity of these species only.
Although in vivo studies on primary pharmacodynamic issues are not mandatory for biosimilar products, for more than half of the products (58%) the in vivo studies were conducted.
Secondary pharmacodynamic studies were performed for more than half (60%) of the products with the new active substance. The common reasons for omission include strong specificity of the antibodies to the respective targets, which indicates the low potential for unintended activity, the absence of an adequate test system or already existing knowledge of the pleiotropic activities of the antibody.
Although dedicated safety pharmacology studies were conducted in part of the cases, the incorporation of safety pharmacology endpoints in repeat dose toxicity studies was common for the majority of products, particularly for biosimilars, which is in line with the 3Rs principles.
Basic pharmacokinetic data were available for all products. Nevertheless, PK data do not comprise the tests covering absorption, distribution, metabolism and excretion (ADME), which would be common for small molecules. The data on distribution, metabolism and excretion are presented in 31%, 10% and 13 % of the products, respectively. There is a high percentage of EPARs not containing any information on the availability of these studies (35%, 22%, 22% respectively), making these interpretations less significant.
The use of relevant species was essential for the evaluation of pharmacokinetic issues. In all cases, whenever the relevant species was identified, they were used for obtaining pharmacokinetic data (except one biosimilar product).
A large proportion of the products, for which anti-drug-antibodies (ADA) status has been evaluated (84%), underline its importance for the interpretation of pharmacokinetic and other pre-clinical data.
Only for a part of the products containing a new active substance (43%), the separate single dose toxicity studies were performed as proposed by the ICH-S6 guideline. For biosimilars dedicated single dose toxicity studies were not conducted in 88% of the products.
For nearly all products (97 %) with just a few exceptions (3 cases) the extensive repeat dose toxicity studies were conducted. The omission of the repeated dose toxicity study was justified by the lack of active species. The majority of the longest pivotal chronic toxicity studies with products containing a new active substance were 26 weeks in duration (61%). Less frequently, studies were 39 weeks in duration (15%). If the repeat dose/chronic toxicity studies were of shorter duration (13 weeks, or in some cases 4 weeks), this was scientifically well justified.
The use of pharmacologically active species in the toxicological studies is crucial for the validity of the study: in all cases, when the relevant species was identified, they were used in toxicity studies. Such cases were 56 products out of 63 with the new active substance and 18 out of 26 biosimilars.
For 40 cases (63%) of the products with the new active substance the toxicity studies were performed only in one species. Two species (rodent and non-rodent) for toxicity testing were used in 19 cases (30%) in line with the ICH-S4 guideline, but only in 8 cases of them the both used species were active for the test products.
The proportion of non-human primates used in toxicity studies was high particularly for products containing new active species. The cynomolgus was used for toxicity testing in 94 % of products with a new active substance, while chimpanzee, rhesus and marmoset were used only in isolated cases. Rats were the most used rodents in toxicity studies and were used for 17 products, in five cases of them as an active species. Rat was followed by mouse (10x), guinea pig (3x) and hamster (1x).
From a high proportion of products (89 %) lacking genotoxicity studies it can be concluded, that the evaluation of the genotoxic potential was not substantial for authorisation, which is in line with the ICH-S6 guideline. The studies were conducted only for products containing antibody-drug conjugates or a radioactive labelled active substance.
For a vast majority (93%) of the products reviewed in this thesis, no carcinogenicity studies were conducted, which was from case to case well justified, using the Weight of the Evidence Approach.
For the products containing a new active substance dedicated fertility and early embryonic Development (FEED) studies were performed for 35% of the products, for another 40% the endpoints of FEED were incorporated in the repeat dose toxicity studies. For a vast majority of biosimilar products (85%) the FEED studies were not performed.
The dedicated embryo-foetal development (EFD) studies were performed in 48% of the products containing a new active substance. In 24% of the products the FED studies were fully integrated in the ePPND. Considering all studies (EFD+ePPND) for the products containing a new active substance (45 products) pharmacologically active species were used in 35 cases (78%). Cynomolgus was used in 29 cases out of them.
In 30% of products containing a new active substance the dedicated pre- and postnatal development (PPND) studies were performed. In further 25% the PPND were fully integrated in the ePPND. In the majority of cases (29 out of 35) PPND studies were conducted using active species.
In the EPARs of 25 biosimilar products it was clearly stated that no FED and PPND studies were carried out and it was also explained that these studies are not required for biosimilars.
Only for 9 out of 63 products containing a new active substance specific juvenile animal studies (JAS) were conducted. The data on juvenile toxicity was obtained as part of other studies in further 9 cases.
The proportion of dedicated local tolerance studies was high (30 %) in products with the new active substance. In most cases (63% of products with a new active substance and 69% of biosimilars) the injection site was observed in repeat dose or other toxicity studies. Rabbit was a prominent test species in local tolerance studies (used in 15 of 22 cases, 68%) even if it was not a cross-reactive species.
The data of immunotoxic potential of biopharmaceuticals containing monoclonal antibodies as an active substance were presented in 30% of the products, whereas for a vast majority of these cases the immunotoxicity relevant endpoints were incorporated in other toxicity studies.
Tissue cross-reactivity studies directed to the human tissues were performed in 63 out of 89 products. For the remaining 26 products, the EPARs provided no information whether these studies were carried out. The proportion of cross-reactivity studies for products containing new active substances was higher than for biosimilars (81% vs. 46%).
For biotechnology-derived products a non-clinical toxicology assessment of process or product related impurities is not required according to the ICH-S6 and ICH-M4 guidelines. Only the description of the purification processes to remove impurities should be provided. In line with this rule the extensive specification of the test material (including investigations of the product and process related impurities) were provided in the quality section of the EPARs of all products. Only for five products the assessment of toxic potential of impurities was carried out.
The studies on Dependence, Metabolites and Phototoxicity were not evaluated since these studies are irrelevant for the products containing monoclonals as an active substance.
For all products containing protein-drug conjugates or radioactive labelled mAb, standard environmental risk assessments (ERA) were carried out. For all other products no ERA was performed. This is in line with the specific guideline (EMEA/CHMP/SWP/4447/00).
EPARs are not a primary source of information on the drug product and do not always contain the complete detailed information about the medicinal product. In fact, the proportion of missing data ranged from individual cases to more than half of the products, which made the assessment challenging and at times made it difficult to arrive at the conclusion for the products in question.
Pages: 69, Annexes: 15, pages: 76