samedan logo
home > ebr > summer 2011 > species selection
European Biopharmaceutical Review

Species Selection

Demonstrating relevance of species for non-clinical safety testing of monoclonal antibodies is both critical and complex. A number of factors must be considered in order to ensure that the selected species appropriately identifies possible human safety concerns.

It is imperative that the safety of investigational new drugs, including therapeutic monoclonal antibodies (mAbs), is demonstrated in non-clinical safety studies in order to obtain approval from regulatory agencies for initiating clinical trials. According to the relevant International Conference on Harmonisation (ICH) safety guidelines, goals of non-clinical safety studies include a characterisation of pharmacological activity of the drug and of potential toxic effects with respect to target organs, relationship to exposure, and reversibility (1,2). This information is also important for estimating a safe starting dose for clinical trials and the identification of safety biomarkers for monitoring in the clinic. Although the goals of safety evaluation for mAbs and biologics in general are similar to those for small molecules, the approach taken to identify species for safety testing is very different. In both cases, regulatory guidelines state that non-clinical s afety testing should be carried out in one rodent and one non-rodent species. However, in the case of biopharmaceuticals like mAbs, many additional and unique factors need to be taken into account as only pharmacologically relevant species should be used (1). Due to the exquisite specificity and consequent high species selectivity of mAbs, this often means that only one relevant species is identified and in those cases a single species toxicology programme is justified.

The importance of species relevance for non-clinical safety testing of biologics is underscored by a recent publication on the properties of the anti-CD28 superagonistic mAb TGN1412 (3). In this publication, it is shown that despite identical amino acid sequence of the epitope and expression of CD28 on human and cynomolgus monkey T cells, the nonclinical safety studies in monkeys did not predict the serious adverse events observed in the first-time-in-human trial with TGN1412 (4) – likely because of differential expression of CD28 on a T cell subset, CD4+ T effector memory cells. This editorial will describe which factors need to be taken into consideration when selecting nonclinical species for safety testing of mAbs in order to promote best practice and ensure the data generated represents a risk assessment that is relevant for human safety.

Determinants of Species Relevance

In order to determine species relevance, a tiered approach is recommended, as illustrated by the species selection ‘funnel’ in Figure 1. Conventional toxicology species such as the mouse, rat, rabbit and dog, are often unsuitable for non-clinical studies with mAbs for two main reasons: the mAb may not be pharmacologically active in these species; and immunogenicity may limit exposure after repeated dosing.While there is a significant amount of data to show that immunogenicity in the nonclinical species does not predict for immunogenicity in humans (5), antidrug antibodies can mediate enhanced clearance or neutralisation of the therapeutic mAb, or even lead to adverse effects due to immune complex deposition and consequent inflammatory responses. Therefore, the validity of the non-clinical safety assessment may be compromised if a mAb is immunogenic in the nonclinical species. The cynomolgus monkey is the most commonly used non-rodent species for non-clinical studies with mAbs because of the high degree of cross-reactivity to the monkey orthologue that can generally be attained, the tolerability of human mAbs in this species (including the reduced immunogenicity of mAbs in this species versus other non-rodent species), and the availability of background data (6).

It is possible and prudent to begin considering which species may be relevant for the testing of a mAb as early as the target selection stage. Sequence identity is the first criterion applied to species selection as it provides an early read on the likelihood of a mAb against a human target binding to the orthologue in the nonclinical species.While high sequence identity builds confidence, low sequence identity erodes confidence with the understanding that the primary amino acid sequence of a full protein may not reflect the sequence identity at the eventual binding epitope of the mAb.

Alongside sequence identity, target expression – including the presence of the same ligand(s) or receptor(s) and signalling pathways – form part of the first tier of assessments to determine species relevance. This is of particular importance for species selection for mAbs because their toxicity is generally mediated through on-target rather than off-target effects, and manifests itself as exaggerated pharmacology. Tissue cross-reactivity studies with the clinical candidate may form part of the expression analysis. However, given the limitations that immunohistochemistry techniques present, this should not be used to select non-clinical species for safety studies, but rather aid in the interpretation of toxicological events seen in non-clinical safety studies, and potentially assist in determining their relevance to man (7,8). Finally, with regard to target expression, it should be kept in mind that non-clinical safety testing generally involves studies in healthy young animals and many disease targets are expressed at a low level, or are not expressed at all in these animals.While targets can be up-regulated in non-clinical disease models, there is industry acceptance that these are not a true model for human disease, are poorly understood, and generally acute in nature. Therefore, at best these allow for hazard identification or proof of principle, but not for risk assessment.

Once a panel of lead mAbs is generated, binding affinities (Kd) of the mAbs to the human target and nonclinical orthologues are compared using methods such as surface plasmon resonance. A similar Kd builds confidence in species selection, whereas a dissimilar Kd erodes confidence or even excludes a species. However, binding alone does not guarantee the desired activity, and mAbs with the most similar Kd to human are selected to progress to the next level of assessments using in vitro cell-based potency assays. In order to adequately compare the activity of the mAb in human cells and cells from the non-clinical species, the in vitro potency assays should be physiologically relevant and mechanistically similar.

Another important factor to consider when selecting non-clinical species is the functionality of the mAb Fc domain. The Fc domain mediates binding to the neonatal Fc receptor (FcRn) on endothelial cells and consequent recycling through the endosomal compartment is responsible for the unique and prolonged serum half-life of antibodies (9). In addition, the Fc domain can mediate antibody-dependent cell mediated cytotoxicity (ADCC) and complement fixation (9). These Fc effector functions are dependent not only on the immunoglobulin isotype, the density of the target on the cell, and binding affinity to Fc receptors and complement factors, but also on the type of Fc receptors expressed on the immune cell(s) of interest (10). Thus, when selecting non-clinical species for safety testing, due consideration needs to be given to the activity of the human mAb isotype in the non-clinical species. Furthermore, the properties of engineered Fc domains – for example designed to prolong antibody half life or enhance ADCC activity – must be taken into consideration.

With regard to in vivo characteristics, a relevant species for non-clinical safety testing is one in which the mAb is pharmacologically active. Therefore, it is essential that in addition to pharmacokinetic (PK) endpoints, a molecular marker of pharmacodynamic activity (PD) is identified in order to demonstrate appropriate pharmacology in the species of choice. This allows for the PK/PD relationship for the mAb and its link to toxicity (if any) to be established. PD can be measured directly or indirectly and can represent measures of target engagement – for example, receptor occupancy, level of free or total (free and mAb-complexed) ligand – target activation, or cellular or physiological responses to mAb treatment. If a mAb is pharmacologically active in the nonclinical species and PD measures beyond target engagement can be demonstrated, the predictivity of the non-clinical safety assessment for human safety is likely to be high, whilst this predictivity may decrease when PD can be demonstrated by target engagement only. Demonstration of PD effects forms the final criterion for identifying relevant non-clinical species.

In some cases, despite considerable effort to identify pharmacologically relevant species, none are identified due to insufficient homology of the target across species, lack of expression of the target in naïve, healthy animals, or complete absence of the target and/or the target pathway in species other than humans. When no pharmacologically relevant species can be identified, three options remain for non-clinical safety assessment:
  • Testing the mAb in a transgenic mouse that expresses the human target
  • Testing a surrogate mAb or homologue directed against the non-clinical orthologue; unless there is a robust scientific justification for using a non-rodent species, a rodent species should be used
  • Testing the mAb in an array of in vitro assays relevant for safety evaluation, for example using human primary cells and/or immortalised cell lines. In vitro dose-response data can then be integrated with projected clinical exposures to set first time in human doses
The alternative path taken should be based on the target, the mechanism of action of the mAb, and technical feasibility. It is important to note that all of the above alternatives for determining non-clinical safety have their own drawbacks (11). For example, a surrogate mAb must be extensively characterised and compared to the clinical candidate (binding epitope, isotype, potency in vitro and so on) and produced to similarly high manufacturing standards as the clinical candidate. At the same time, from a regulatory perspective, non-clinical safety studies with surrogates are considered for hazard identification only, as the surrogate is not the clinical candidate (8). Similarly, a transgenic mouse model must be extensively characterised to demonstrate functionality of the transgene. Furthermore, transgenic models are associated with a considerable immunogenicity risk, and may not be suitable for targets that interact with multiple ligands or accessory proteins (as this could potentially require multiple transgenes and become too challenging technically). A solely in vitro assay-based approach to safety testing precludes evaluation of risk due to chronic exposure or reproductive toxicity and thus may not be a viable option for non-life-threatening indications. In addition, there will be little data to support a starting dose other than that based on the minimal anticipated biological effect level (MABEL) which can greatly extend the duration of the first-time-in-human clinical trial (12). The value of the target and the risk:benefit ratio in the proposed patient population should therefore be carefully considered before embarking on an alternative route, and early consultation with regulatory agencies on the path chosen is essential for any of the alternative approaches to be successful.


Selection of a relevant non-clinical species forms the foundation of the non-clinical testing of therapeutic mAbs. Only relevant species can provide insight into possible hazards/safety risks in humans. Although no species can be considered equivalent to humans in terms of safety prediction, careful selection of relevant species for non-clinical safety testing as outlined above has proven to be extremely valuable in reducing risk to humans in clinical trials, and the extreme rarity of serious or lifethreatening events in the large number of clinical trials conducted every year is testament to the effectiveness of this approach.

  1. ICH S6: Preclinical safety evaluation of biotechnology-derived pharmaceuticals
  2. ICH M3(R2): Non-clinical safety studies for the conduct of human clinical trials and marketing authorisation for pharmaceuticals
  3. Eastwood D, Findlay L, Poole S, Bird C, Wadhwa M, Moore M, Burns C, Thorpe R and Stebbings R, Monoclonal antibody TGN1412 trial failure explained by species differences in CD28 expression on CD4+ effector memory T-cells, Br J Pharmacol 161: p512, 2010
  4. Suntharalingam G, Perry MR, Ward S, Brett SJ, Castello-Cortes A, Brunner MD and Panoskaltsis N, Cytokine storm in a Phase 1 trial of the anti- CD28 monoclonal antibody TGN1412, N Engl J Med 355: p1,018, 2006
  5. Bugelski PJ and Treacy G, Predictive power of preclinical studies in animals for the immunogenicity of recombinant therapeutic proteins in humans, Curr Opin Mol Ther 6: p10, 2004
  6. Chapman K, Pullen N, Graham M and Ragan I, Preclinical safety testing of monoclonal antibodies: the significance of species relevance, Nat Rev Drug Discov 6: p120, 2007
  7. Leach MW, Halpern WG, Johnson CW, Rojko JL, MacLachlan TK, Chan CM, Galbreath EJ, Ndifor AM, Blanset DL, Polack E and Cavagnaro JA, Use of tissue cross-reactivity studies in the development of antibody-based biopharmaceuticals: history, experience, methodology, and future directions, Toxicol Pathol 38: p1,138, 2010
  8. ICH S6(R1) Addendum: preclinical safety evaluation of biotechnologyderived pharmaceuticals
  9. Raghavan M and Bjorkman PJ, Fc receptors and their interactions with immunoglobulins, Annu Rev Cell Dev Biol 12: p181, 1996
  10. Nimmerjahn F and Ravetch JV, Fcgamma receptors as regulators of immune responses, Nat Rev Immunol 8: p34, 2008
  11. Bussiere JL et al, Alternative strategies for toxicity testing of species-specific biopharmaceuticals, Int J Toxicol 28: p230, 2009
  12. EMEA/CHMP/SWP/28367/07 Guideline on strategies to identify and mitigate risks for first-in-human clinical trials with investigational medicinal products

Read full article from PDF >>

Rate this article You must be a member of the site to make a vote.  
Average rating:

There are no comments in regards to this article.

Lolke de Haan is Director of Toxicology within Biologics Safety Assessment (BSA) at MedImmune, UK. Prior to joining MedImmune he was a discovery toxicologist at AstraZeneca, supporting small and large molecule drug development projects. Lolke has an academic background in immunology and molecular biology, and received a PhD in Medical Sciences from the University of Groningen, The Netherlands. Email:

Simon Henderson is a Toxicology Project Leader (TPL) within BSA at MedImmune, UK. Prior to joining MedImmune he gained experience in drug development at Huntingdon Life Sciences. Originally with a background in bioanalysis and DMPK, Simon initially managed non-clinical safety programmes for small molecules and biopharmaceuticals before specialising exclusively in biologics as a Biopharmaceutical Product Development Manager. Simon received a PhD from the University of Liverpool. Email:

Mary McFarlane is a TPL within BSA at MedImmune, UK. Prior to joining MedImmune, she was a TPL at GSK, where she gained experience in drug development of both small molecules and biologics. Previously, Mary worked for AgrEvo Agrochemicals as a toxicologist and managed a small team of toxicologists. Mary received an MSc and PhD in Toxicology from the University of Surrey, UK and is a Eurotox Registered Toxicologist. Email:

Alison Scott is Toxicology Programme Coordinator within BSA at MedImmune, UK. Prior to MedImmune she held a Senior Study Manager position at Huntingdon Life Sciences, where she was responsible for performing regulatory toxicology studies with small and large molecule drugs. Alison has an MSc in Applied Toxicology from the University of Surrey. Email:

Jacintha Shenton is a TPL within BSA at MedImmune, UK. Previously, she held a TPL position in the Department of Immunotoxicology at Bristol-Myers Squibb, Syracuse, NY. Jacintha received her PhD in Toxicology from the University of Toronto. Jacintha represents MedImmune on the ILSI-HESI Immunotoxicology Technical Committee and recently completed her tenure on the Executive Committee of the Immunotoxicology Specialty Section of the Society of Toxicology. Email:

Lolke de Haan
Simon Henderson
Mary McFarlane
Alison Scott
Jacintha Shenton
Print this page
Send to a friend
Privacy statement
News and Press Releases

SONOTEC® launches industrial clamp-on flow meter for plastic tubes and pipes for the use in hazardous areas of Zone 1 according to ATEX/IECEx

The new SEMIFLOW Ex1 Set opens a new product category of intrinsically safe ultrasonic flow sensors with control gear for non-contact liquid flow measurement on small and midsize rigid plastic tubes to ensure safe operation in hazardous areas. SEMIFLOW CO.66 PI Ex1 sensors are protected against explosion hazard by gases, vapors, and liquids according to the gas group IIB. The device protection level (EPL) is "Gb" for the use in Zone 1 according to ATEX/IECEx.
More info >>

White Papers

Orthogonal Approaches for the Analysis of Protein Sequence

Ludger Ltd

One of the most important classes of biopharmaceuticals is monoclonal antibodies. They are expressed from living cells, and are consequently subject to complex biochemical pathways. Not all of these routes are fully understood, and many are sensitive to the subtlest of environmental changes. Changes such as these can affect the final biopharmaceutical sequence, structure and post-translational modifications, not to mention any changes that may occur during further purification. This often means that the final product from one batch may be different from the last in another batch. Each batch is a complicated mix of similar molecules, and analysis of the degree of batch-to-batch variation is therefore very important to the establishment of the drug's safety for medicinal use and patient administration.
More info >>

Industry Events

BIO-Europe Digital

26-29 October 2020, Digitally delivered

Global connections are more important than ever. With the same powerful one-to-one partnering and valuable content, BIO-Europe Digital continues to help you drive your life science partnering forward.
More info >>



©2000-2011 Samedan Ltd.
Add to favourites

Print this page

Send to a friend
Privacy statement