- Email: [email protected]
Preclinical safety testing for cell-based products using animals
Biologicals xxx (2015) 1e4
Contents lists available at ScienceDirect
Biologicals journal homepage: www.elsevier.com/locate/biologicals
Meeting report
Preclinical safety testing for cell-based products using animals James W. McBlane* Medicines & Healthcare Products Regulatory Agency, 151 Buckingham Palace Road, London SW1W 9SZ, UK
a r t i c l e i n f o
a b s t r a c t
Article history: Received 17 February 2015 Received in revised form 21 April 2015 Accepted 1 May 2015 Available online xxx
The objectives of preclinical testing include to show why there might be therapeutic benefit in patients and to provide information on the product's toxicity. For cell-based products, given even once, there may be long term exposure and this could imply, unlike for conventional drugs, that all preclinical studies may be needed prior to first human use. The duration of exposure to cells should be studied in animals to guide toxicity assessments. Distribution of cells after administration by a route resembling that intended in humans should be studied to understand potential risks. Risk of tumour formation with the product may also need to be characterised. To the extent that this information can be generated by in vitro testing, studies in animals may not be needed and limitations on the capability of preclinical data to predict human toxicity are recognised: species-specificity make some cell products act only in humans and a human cell-product might be expected to be rejected by immunocompetent animals. Does this suggest testing in immunosuppressed animals or of development of an animal-cell product supposedly similar to the human cell product? No single answer seems to fit every situation.
Keywords: Advanced therapy medicinal products Cell therapy Clinical trials Marketing authorisation Preclinical testing Tumourigenicity
1. Introduction This article is the second of two reflecting two talks given by the author at a meeting1 in Kyoto, Japan in March 2014. The first addressed the European regulatory framework for the regulation of cell therapies: this article addresses issues relating to preclinical testing to support development of such products, with a particular focus on in vivo studies in animals. The theme of the meeting was to share experience from different territories in the international regulation of such products. In particular, the meeting aimed at discussing how to ensure development of good quality, safe and effective cell therapy products throughout the world. This article discusses use of animals in preclinical testing of human cell-based therapies. In the development of any type of medicinal product, containing a novel active agent, preclinical testing is conducted in order to provide evidence for expectation of therapeutic benefit in patients, to provide information on what toxicity the drug might possess and to indicate doses for each such
Abbreviations: ADME, Absorption, distribution, metabolism & excretion; MHRA, Medicines & Healthcare products Regulatory Agency; WHO, World Health Organization. * Tel.: þ44 20 3080 6381. E-mail address: [email protected]. 1 Challenges towards sound scientific regulation of cell therapy products, March 7e8th, Kyoto International Conference Center, Kyoto Japan.
effect: it also aims to identify agents which should not be given to humans at all, either because of their inherent toxicity which can translate into a lack of any evident safety margin, or perhaps because the agent in question has unsuitable kinetics; preclinical testing also aims to provide further information to aid understanding of an effect recognised but poorly delineated, whether toxic or beneficial. Limitations of testing in animals are acknowledged [1] and where in vivo studies are not useful or may well be potentially misleading, their omission is justified. This article addresses: aims of preclinical testing contrasts between studies for a small chemical drug or a cell therapy product regulators' expectations for preclinical data for cell therapy products circumstances where an absence of any in vivo testing is appropriate.
2. Compare and contrast: data supporting a first human trial with a novel agent that is a small chemical drug as compared to a cell therapy product By the time of the first clinical trial, a typical preclinical dataset for a novel small molecule chemical drug will likely include the following:
http://dx.doi.org/10.1016/j.biologicals.2015.05.002 1045-1056
Please cite this article in press as: McBlane JW, Preclinical safety testing for cell-based products using animals, Biologicals (2015), http:// dx.doi.org/10.1016/j.biologicals.2015.05.002
2
Meeting report / Biologicals xxx (2015) 1e4
- primary pharmacology e in vitro and in vivo studies supporting the intended therapeutic action; - secondary pharmacology e in vitro studies into inherent properties of the drug, but which are not the basis of its intended therapeutic effect; - safety pharmacology e effects on the vital systems, cardiovascular, central nervous and respiratory systems e with interest on effects at likely maximal intended clinical doses; - pharmaco- & toxicokinetics e exposure, absorption, distribution, metabolism and excretion; - general toxicity and local tolerability e nature, dose/exposure associated with effects, and reversibility - genotoxicity e mutagenicity, clastogenicity and in vivo genotoxicity. This set of data will typically support relatively short term clinical studies into the safety of the product in healthy humans which may also, perhaps, provide clinical data on either pharmacodynamic effects relevant to the intended therapeutic effect, or otherwise give some indication of the potential for efficacy with testing in patients. For later clinical development, in patients with less close monitoring than applied in early clinical studies, data from reproductive toxicity studies (fertility, embryofetal development and prenatal and postnatal development) and from other toxicity studies (eg phototoxicity, immunotoxicity, dependence, longer term general toxicity studies, metabolite toxicity studies) might be expected. Carcinogenicity studies may be required usually by the time an application for approval to market the product is to be made. These expectations are set out in regulatory guidance [2]. Whereas many small molecule drugs and some non-cell therapy biological products might fit well into this set of studies, to begin the development of a cell therapy product by intending to implement this plan, making modifications as necessary for a cell therapy product, is an inadvisable approach. Rather, a fresh approach should be adopted, based on defining what questions preclinical testing needs to address prior to clinical testing and focussing only on studies that will enable only clinical testing that is both reasonable and safe.
2.1. Primary pharmacodynamics One key question to answer in preclinical development of a cell therapy product is what evidence suggests there is a reasonable expectation of benefit? Some of the type of primary pharmacology data generated for small molecules is usually not relevant here: there is no parallel with the type of information describing drugreceptor or whether the drug is an agonist, partial agonist, antagonist or inverse agonist e cells of a cell therapy product likely secrete multiple molecules covering all these actions. Nevertheless, evidence supporting use of the cell therapy product must be provided. In some instances, this can be from use cell products in animals with spontaneous disease, rather than from experimentallyinduced pathological states in experimental animals. Instances where such data are used are rare: usually, proof of concept data are provided from studies that characterise disease and its cause in humans and from effects noted in experiments in animals that recapitulate some of these features: it may even include specific explorations into the understanding of where experimental system in animals or veterinary pathology differs from that in humans, either for anatomical or pathological reasons. It is also relevant to note and contrast the typically chronic nature of human disease, compared to the usually acute nature of an induced change in experimental studies in animals.
For chemical drugs, defining the dose expected for clinical therapeutic benefit in human patients is based on testing with that molecule. Data on drugs of similar chemical and pharmacological classes can be applied to set proposed human dosing in context, but cannot substitute for data generated with that specific drug in preclinical development. In contrast, with cell therapies, clinical data with similar products are of greater relevance than testing in animals, whether for dose selection or for safety considerations. However, such an approach raises the question of how similar is the product to be used in the proposed trial with that used in clinical studies, results from which are the basis of the dose selection and the claim for the expectation of safety. Where it is feasible to adopt this approach of crossreference to clinical data with products that are claimed to be similar, this is encouraged. The nature of clinical data can also be variable e where a dedicated clinical study with a defined product has been done, it can easily be recognised that this type of systematic data is of greater use in providing a basis to make a decision going forward, than are anecdotal data, perhaps from a case series of patients that have been treated with product that may either be known to have been produced by variable means or the provenance of which is even more uncertain. Clinical testing with cell therapy products will almost certainly always start with testing in patients with the condition for which the product is being developed; it is almost impossible to imagine use of such products in healthy subjects. This has the consequence that it is probably not reasonable to use doses that are expected to be inactive. The clinical starting dose should therefore be in the range that may be expected to have therapeutic activity and for safety reasons, should likely be positioned at the lowest end of the range of doses that are considered active. Where the product is unlike any for which there is previous clinical experience and/or there are major difficulties in projecting a clinical dose based on studies in animals, then the ability to identify a potentially active dose may be compromised and, there being very little evidence to support higher doses, safety should be the prime consideration driving dose selection, even if this has the consequence that patients in initial cohorts are later recognised to have received an inactive dose. Thus, in contrast to small molecule drugs, where clinical testing is escalated from an initial dose expected to be safe as it has no activity into an anticipated therapeutic range, testing with cell therapy products should first characterise what is expected to be an active dose and then seek to demonstrate why this is expected to be safe: this shows the different mindset appropriate to such products. 2.2. Secondary and safety pharmacology Potential for unintended effects and for risks of adverse effects on function of vital systems (central nervous, cardiovascular and respiratory systems) should be understood, but for cell products, separate studies are rarely justified and should not be done unless there is a specific consideration that requires this testing. It is likely that general toxicity testing will suffice. In contrast, most small molecules should have some dedicated testing for potential effects on these systems, and perhaps also gastrointestinal and renal system functions [3]. 2.3. Kinetics e distribution and persistence For small molecules, there is interest in understanding and characterising the exposure and elimination of a small molecule drug. This has led to a set of studies that address its absorption, distribution, metabolism and excretion (ADME) with results from animals applied to model reasonable expectations in humans. Toxicokinetic investigations provide data on exposure with
Please cite this article in press as: McBlane JW, Preclinical safety testing for cell-based products using animals, Biologicals (2015), http:// dx.doi.org/10.1016/j.biologicals.2015.05.002
Meeting report / Biologicals xxx (2015) 1e4
particular emphasis on the maximal exposure in animals that is not considered to cause adverse effects with an intent that human exposure would not exceed this, without exceptional reasons. For cell therapies, again, as for the overall preclinical development programme, it is a hindrance to think of applying this thinking to a cell therapy product. Rather, the starting points are the questions of understanding of how the product will be applied to, or otherwise reach, its site of action, where it will distribute from there and for how long after its administration it will be present; risks attached to these issues can then be delineated. Within preclinical development, it is of relevance to understand the persistence of cells in order to be better able to understand how its effects are produced e if beneficial effects can be detected at a timepoint when the administered cells are no longer present, this suggests the effects arise not directly from those administered cells but from an endogenous process kick-started by the administered cells. Of note, this information may be understood as of use to predicting a human profile; in contrast, the author judges that this information is primarily to understand the results from preclinical testing and not primarily to make a projection as what may happen in humans. However, results should nonetheless have a direct bearing on the nature of monitoring conducted in clinical studies e both what should be monitored, possibly how it should be monitored, how frequently and for how long. 2.4. General toxicity Severe, sometimes fatal, adverse reactions to cell therapy products have been reported in clinical testing [4,5]. Toxicity testing in animals prior to human use aims to identify the type of toxicity that may arise so that clinical monitoring can be implemented to protect patients and to indicate doses for such effects, noting that not all findings in animals are of relevance to humans [6]. Where appropriate, it can sometimes suggest that the product should not be given to humans because a consistent argument for expecting benefit at safe doses cannot be made. However, the objective of general toxicity testing of cell therapy products is the same as for a chemical drug ie to identify what type of toxicity and determine if projected clinical doses for therapeutic action can be judged to be expected to be safe. Judgement is called for by the developer and regulators in each instance where it is argued that testing in animals should not be done. Some cell therapy products are active only in human systems and for such products, animal testing is not relevant and should not be done, as it may either lead to a false sense of expectation of safety in humans, or may indicate the potential for effects that are, in fact, of no relevance to humans. In other cases, a transgenic murine system may be established which is capable of recapitulating some elements of the disease and use of the product; these can be complex. Regulators should recognise that in some circumstances, it is appropriate to conclude that there is no in vivo animal testing that can rationally support later development: these instances are rare and not limited to cell therapy products (eg some monoclonal antibodies can lack activity in all species except humans). Despite the lack of preclinical safety data, these products may yet offer great potential, typically for a small number of patients who otherwise would have no specific therapeutic options. Dialogue with regulators to facilitate understanding is encouraged such that clinical development can be supported.
3
tumours and, secondly, does its administration as proposed in this trial put patients at unreasonable risk of tumours? Where a product is subject to ex vivo expansion, there is a natural concern that needs to be addressed which is whether the expansion steps result in any risk of cell immortalisation. Proof of absence of uncontrolled proliferation, clonogenic potential and characterisation of telomere length is advised from in vitro studies. In vivo testing of potential of a human cell therapy product to form tumours in vivo after its single subcutaneous administration to immune-incompetent animals is of interest: if this is negative with testing supported by use of a positive control cell lines, no hindrance to clinical testing is identified. If such testing is positive, the cell line may be judged to have an unreasonable safety profile. Alternatively, it may be that development of tumours in animals can be argued to not predict what is likely to happen in humans. One other element should be mentioned: guidance from the World Health Organisation (WHO) [7] discusses tumourigenicity testing for cell substrates: this guidance aims to address testing of the production process to ensure consistent product quality where a product is produced by cell-based mans. This is not the same as studies to address risk of whether a cell-based product poses a risk of tumourigenicity. 3. Some specific points for consideration Some regulatory bodies offer scientific advice procedures, with the intent of assisting developers of medicinal products. The aim of such meetings is for the regulators to indicate, which among a range of options, might result in the developer presenting data that would support a decision to approve an application or to identify issues at regulatory review that might pose challenges and how the developer can anticipate these. Some issues discussed in these procedures are considered below. Question: Should animal cells be used in preclinical testing? Comment: This may reflect the stage of development. At the stage of exploratory research, it could be that initial data supporting the proof of principle is generated by testing with animal cells. However, if the product to be placed on the market is a human cell product, development studies should focus on this product, not on an animal-cell-based product.
2.5. Testing for tumour risk
Question: Should a large species be used in preclinical testing? Comment: There is no single answer to this question. In some circumstances, there will be data only from testing in rodents and this can suffice: clinical studies can be supported by preclinical testing only in mice, for instance. In other circumstances, it may only be possible to generate proof of concept data from testing in large animals or there may be an existing veterinary product and clinical development of a similar product is being undertaken. Thus, applicants may be able to present data from testing in dogs, horses, goats, pigs, sheep or other species. In other cases, there may be intent to deliver the product to an unusual anatomical site with a novel device, and so testing to support this is done in large animals in order to demonstrate its feasibility and expectation for its success when applied to humans because dosing so in a rat or mouse could not be done with the device. Whilst each case therefore needs to be considered on its own merits, it is clear that testing in a large animal species should not be done unless such testing is essential because its absence would create a deficiency that might pose risks to patients.
Two considerations should be faced in terms of tumour risk assessment: does this product have inherent potential to elicit
Question: Do preclinical studies indicate whether immunosuppressive drugs be used in clinical trials?
Please cite this article in press as: McBlane JW, Preclinical safety testing for cell-based products using animals, Biologicals (2015), http:// dx.doi.org/10.1016/j.biologicals.2015.05.002
4
Meeting report / Biologicals xxx (2015) 1e4
Comment: In many instances, studies are done in immunoincompetent animals because immuno-competent animals would reject a human cell product, so invalidating the testing or causing effects that are difficult to translate to humans. In other cases, immunosuppressive drugs are given to normal animals to prevent human cell rejection. Does this suggest that immunosuppressive drugs should be used in clinical use of the product? Probably this question should only be addressed clinically; that is, if it is feasible to avoid use of immunosuppressive drugs, clinical evidence should be generated to determine whether responses are better, worse or not different, depending on use, or not of immunosuppressive therapy. This can be addressed systematically in clinical trials that adopt dosing in the presence of immunosuppressive therapy and in its absence.
4. Preclinical development of cell therapies From the foregoing, it can be summarised that at the time of first clinical trial application, and excepting a small number of products for which it can be argued that this testing is not relevant, a typical preclinical development programme will include data on the following. - primary pharmacology e evidence supporting the expectation for therapeutic benefit - biodistribution & cell persistence e likely to be human cells given to animals, not in the context of a disease model - general toxicity e after a single dose by a route that is intended to be by the clinical route with follow up determined with reference to results from data on cell persistence - where the clinical dosing is to be repeated, repeat dose general toxicity data may be expected - in vitro and perhaps in vivo studies on risk of tumour formation.
5. Conclusions In assessing an application, whether for a clinical trial or for a marketing authorisation, the assessor has to judge whether the data supplied support the claim by the applicant that approval is reasonable and safe. Specific challenges posed by cell therapy products are different from small molecules and require a different approach. Nevertheless, across different types of cell therapies, similar themes emerge [8]. In considering this, the primary considerations can be identified as:
1 Is the product to be used clinically the same as that used in the preclinical testing programme? What is the nature of the evidence for this? Where an applicant cites data from a published paper, how can the assessor judge that those results can be directly applied to the product that is the subject of the application under consideration? If there are differences, does recognition of this have an impact on the decision? 2 Is the basis for selection of the first human dose sound? As the initial studies are in patients, use of doses expected to be beneficial should first be characterised e this is likely to refer to both clinical and preclinical data, but for novel products it may be that there are no previous relevant clinical data. The developer should then consider what evidence is presented for the expectation that these doses will be safe. 3 Testing in animals should be designed to provide only information that is essential in order to be able to safely test the product in patients; it is the nature of preclinical testing that it becomes of lesser relevance once clinical data are generated. 4 For cell therapy products, engagement with regulators, for instance in scientific advice meetings, is encouraged. However, this should not be considered a routine step of a means of obtaining an opinion on a potential application: rather it should aim to present issues where there are several options available for progression, each with likely detrimental elements that argue against their use. References [1] Broichhausen C, Riquelme P, Ahrens N, Wege AK, Koehl GE, Schlitt HJ, et al. In question: the scientific value of preclinical safety pharmacology and toxicology studies with cell-based therapies. Mol Ther d Meth Clin Dev 2014;1:1e12. [2] ICH M3 (R2) Non-clinical safety studies for the conduct of human clinical trials and marketing authorization for pharmaceuticals. [3] ICH S7A note for guidance on safety pharmacology studies for human pharmaceuticals CHMP/ICH/539/00. [4] Morgan RA, Yang JC, Kitano M, Dudley ME, Laurencot CM, Rosenberg SA. Case report of a serious adverse event following the administration of T cells transduced with a chimeric antigen receptor recognizing ERBB2. Mol Ther 2010;18(4):843e51. [5] Linette GP, Stadtmauer EA, Maus MV, Rapoport AP, Levine BL, Emery L, et al. Cardiovascular toxicity and titin cross-reactivity of affinity-enhanced T cells in myeloma and melanoma. Blood 2013;122(6):863e71. [6] Rosenberg SA. Of mice, not men: no evidence for graft-versus-host disease in humans receiving T-cell receptor-transduced autologous T cells. Mol Ther 2010;18(10):1744e5. [7] WHO evaluation of cell substrates for the production of biological: revision of recommendations. 2009. Report of the WHO Study Group on cell substrates for the production of biologicals. [8] Schneider CK, Salmikangas P, Jilma B, Flamion B, Todorova LR, Paphitou A, et al. Challenges with advanced therapy medicinal products and how to meet them. Nat Rev Drug Discov 2010;9:195e201.
Please cite this article in press as: McBlane JW, Preclinical safety testing for cell-based products using animals, Biologicals (2015), http:// dx.doi.org/10.1016/j.biologicals.2015.05.002
Contents lists available at ScienceDirect
Biologicals journal homepage: www.elsevier.com/locate/biologicals
Meeting report
Preclinical safety testing for cell-based products using animals James W. McBlane* Medicines & Healthcare Products Regulatory Agency, 151 Buckingham Palace Road, London SW1W 9SZ, UK
a r t i c l e i n f o
a b s t r a c t
Article history: Received 17 February 2015 Received in revised form 21 April 2015 Accepted 1 May 2015 Available online xxx
The objectives of preclinical testing include to show why there might be therapeutic benefit in patients and to provide information on the product's toxicity. For cell-based products, given even once, there may be long term exposure and this could imply, unlike for conventional drugs, that all preclinical studies may be needed prior to first human use. The duration of exposure to cells should be studied in animals to guide toxicity assessments. Distribution of cells after administration by a route resembling that intended in humans should be studied to understand potential risks. Risk of tumour formation with the product may also need to be characterised. To the extent that this information can be generated by in vitro testing, studies in animals may not be needed and limitations on the capability of preclinical data to predict human toxicity are recognised: species-specificity make some cell products act only in humans and a human cell-product might be expected to be rejected by immunocompetent animals. Does this suggest testing in immunosuppressed animals or of development of an animal-cell product supposedly similar to the human cell product? No single answer seems to fit every situation.
Keywords: Advanced therapy medicinal products Cell therapy Clinical trials Marketing authorisation Preclinical testing Tumourigenicity
1. Introduction This article is the second of two reflecting two talks given by the author at a meeting1 in Kyoto, Japan in March 2014. The first addressed the European regulatory framework for the regulation of cell therapies: this article addresses issues relating to preclinical testing to support development of such products, with a particular focus on in vivo studies in animals. The theme of the meeting was to share experience from different territories in the international regulation of such products. In particular, the meeting aimed at discussing how to ensure development of good quality, safe and effective cell therapy products throughout the world. This article discusses use of animals in preclinical testing of human cell-based therapies. In the development of any type of medicinal product, containing a novel active agent, preclinical testing is conducted in order to provide evidence for expectation of therapeutic benefit in patients, to provide information on what toxicity the drug might possess and to indicate doses for each such
Abbreviations: ADME, Absorption, distribution, metabolism & excretion; MHRA, Medicines & Healthcare products Regulatory Agency; WHO, World Health Organization. * Tel.: þ44 20 3080 6381. E-mail address: [email protected]. 1 Challenges towards sound scientific regulation of cell therapy products, March 7e8th, Kyoto International Conference Center, Kyoto Japan.
effect: it also aims to identify agents which should not be given to humans at all, either because of their inherent toxicity which can translate into a lack of any evident safety margin, or perhaps because the agent in question has unsuitable kinetics; preclinical testing also aims to provide further information to aid understanding of an effect recognised but poorly delineated, whether toxic or beneficial. Limitations of testing in animals are acknowledged [1] and where in vivo studies are not useful or may well be potentially misleading, their omission is justified. This article addresses: aims of preclinical testing contrasts between studies for a small chemical drug or a cell therapy product regulators' expectations for preclinical data for cell therapy products circumstances where an absence of any in vivo testing is appropriate.
2. Compare and contrast: data supporting a first human trial with a novel agent that is a small chemical drug as compared to a cell therapy product By the time of the first clinical trial, a typical preclinical dataset for a novel small molecule chemical drug will likely include the following:
http://dx.doi.org/10.1016/j.biologicals.2015.05.002 1045-1056
Please cite this article in press as: McBlane JW, Preclinical safety testing for cell-based products using animals, Biologicals (2015), http:// dx.doi.org/10.1016/j.biologicals.2015.05.002
2
Meeting report / Biologicals xxx (2015) 1e4
- primary pharmacology e in vitro and in vivo studies supporting the intended therapeutic action; - secondary pharmacology e in vitro studies into inherent properties of the drug, but which are not the basis of its intended therapeutic effect; - safety pharmacology e effects on the vital systems, cardiovascular, central nervous and respiratory systems e with interest on effects at likely maximal intended clinical doses; - pharmaco- & toxicokinetics e exposure, absorption, distribution, metabolism and excretion; - general toxicity and local tolerability e nature, dose/exposure associated with effects, and reversibility - genotoxicity e mutagenicity, clastogenicity and in vivo genotoxicity. This set of data will typically support relatively short term clinical studies into the safety of the product in healthy humans which may also, perhaps, provide clinical data on either pharmacodynamic effects relevant to the intended therapeutic effect, or otherwise give some indication of the potential for efficacy with testing in patients. For later clinical development, in patients with less close monitoring than applied in early clinical studies, data from reproductive toxicity studies (fertility, embryofetal development and prenatal and postnatal development) and from other toxicity studies (eg phototoxicity, immunotoxicity, dependence, longer term general toxicity studies, metabolite toxicity studies) might be expected. Carcinogenicity studies may be required usually by the time an application for approval to market the product is to be made. These expectations are set out in regulatory guidance [2]. Whereas many small molecule drugs and some non-cell therapy biological products might fit well into this set of studies, to begin the development of a cell therapy product by intending to implement this plan, making modifications as necessary for a cell therapy product, is an inadvisable approach. Rather, a fresh approach should be adopted, based on defining what questions preclinical testing needs to address prior to clinical testing and focussing only on studies that will enable only clinical testing that is both reasonable and safe.
2.1. Primary pharmacodynamics One key question to answer in preclinical development of a cell therapy product is what evidence suggests there is a reasonable expectation of benefit? Some of the type of primary pharmacology data generated for small molecules is usually not relevant here: there is no parallel with the type of information describing drugreceptor or whether the drug is an agonist, partial agonist, antagonist or inverse agonist e cells of a cell therapy product likely secrete multiple molecules covering all these actions. Nevertheless, evidence supporting use of the cell therapy product must be provided. In some instances, this can be from use cell products in animals with spontaneous disease, rather than from experimentallyinduced pathological states in experimental animals. Instances where such data are used are rare: usually, proof of concept data are provided from studies that characterise disease and its cause in humans and from effects noted in experiments in animals that recapitulate some of these features: it may even include specific explorations into the understanding of where experimental system in animals or veterinary pathology differs from that in humans, either for anatomical or pathological reasons. It is also relevant to note and contrast the typically chronic nature of human disease, compared to the usually acute nature of an induced change in experimental studies in animals.
For chemical drugs, defining the dose expected for clinical therapeutic benefit in human patients is based on testing with that molecule. Data on drugs of similar chemical and pharmacological classes can be applied to set proposed human dosing in context, but cannot substitute for data generated with that specific drug in preclinical development. In contrast, with cell therapies, clinical data with similar products are of greater relevance than testing in animals, whether for dose selection or for safety considerations. However, such an approach raises the question of how similar is the product to be used in the proposed trial with that used in clinical studies, results from which are the basis of the dose selection and the claim for the expectation of safety. Where it is feasible to adopt this approach of crossreference to clinical data with products that are claimed to be similar, this is encouraged. The nature of clinical data can also be variable e where a dedicated clinical study with a defined product has been done, it can easily be recognised that this type of systematic data is of greater use in providing a basis to make a decision going forward, than are anecdotal data, perhaps from a case series of patients that have been treated with product that may either be known to have been produced by variable means or the provenance of which is even more uncertain. Clinical testing with cell therapy products will almost certainly always start with testing in patients with the condition for which the product is being developed; it is almost impossible to imagine use of such products in healthy subjects. This has the consequence that it is probably not reasonable to use doses that are expected to be inactive. The clinical starting dose should therefore be in the range that may be expected to have therapeutic activity and for safety reasons, should likely be positioned at the lowest end of the range of doses that are considered active. Where the product is unlike any for which there is previous clinical experience and/or there are major difficulties in projecting a clinical dose based on studies in animals, then the ability to identify a potentially active dose may be compromised and, there being very little evidence to support higher doses, safety should be the prime consideration driving dose selection, even if this has the consequence that patients in initial cohorts are later recognised to have received an inactive dose. Thus, in contrast to small molecule drugs, where clinical testing is escalated from an initial dose expected to be safe as it has no activity into an anticipated therapeutic range, testing with cell therapy products should first characterise what is expected to be an active dose and then seek to demonstrate why this is expected to be safe: this shows the different mindset appropriate to such products. 2.2. Secondary and safety pharmacology Potential for unintended effects and for risks of adverse effects on function of vital systems (central nervous, cardiovascular and respiratory systems) should be understood, but for cell products, separate studies are rarely justified and should not be done unless there is a specific consideration that requires this testing. It is likely that general toxicity testing will suffice. In contrast, most small molecules should have some dedicated testing for potential effects on these systems, and perhaps also gastrointestinal and renal system functions [3]. 2.3. Kinetics e distribution and persistence For small molecules, there is interest in understanding and characterising the exposure and elimination of a small molecule drug. This has led to a set of studies that address its absorption, distribution, metabolism and excretion (ADME) with results from animals applied to model reasonable expectations in humans. Toxicokinetic investigations provide data on exposure with
Please cite this article in press as: McBlane JW, Preclinical safety testing for cell-based products using animals, Biologicals (2015), http:// dx.doi.org/10.1016/j.biologicals.2015.05.002
Meeting report / Biologicals xxx (2015) 1e4
particular emphasis on the maximal exposure in animals that is not considered to cause adverse effects with an intent that human exposure would not exceed this, without exceptional reasons. For cell therapies, again, as for the overall preclinical development programme, it is a hindrance to think of applying this thinking to a cell therapy product. Rather, the starting points are the questions of understanding of how the product will be applied to, or otherwise reach, its site of action, where it will distribute from there and for how long after its administration it will be present; risks attached to these issues can then be delineated. Within preclinical development, it is of relevance to understand the persistence of cells in order to be better able to understand how its effects are produced e if beneficial effects can be detected at a timepoint when the administered cells are no longer present, this suggests the effects arise not directly from those administered cells but from an endogenous process kick-started by the administered cells. Of note, this information may be understood as of use to predicting a human profile; in contrast, the author judges that this information is primarily to understand the results from preclinical testing and not primarily to make a projection as what may happen in humans. However, results should nonetheless have a direct bearing on the nature of monitoring conducted in clinical studies e both what should be monitored, possibly how it should be monitored, how frequently and for how long. 2.4. General toxicity Severe, sometimes fatal, adverse reactions to cell therapy products have been reported in clinical testing [4,5]. Toxicity testing in animals prior to human use aims to identify the type of toxicity that may arise so that clinical monitoring can be implemented to protect patients and to indicate doses for such effects, noting that not all findings in animals are of relevance to humans [6]. Where appropriate, it can sometimes suggest that the product should not be given to humans because a consistent argument for expecting benefit at safe doses cannot be made. However, the objective of general toxicity testing of cell therapy products is the same as for a chemical drug ie to identify what type of toxicity and determine if projected clinical doses for therapeutic action can be judged to be expected to be safe. Judgement is called for by the developer and regulators in each instance where it is argued that testing in animals should not be done. Some cell therapy products are active only in human systems and for such products, animal testing is not relevant and should not be done, as it may either lead to a false sense of expectation of safety in humans, or may indicate the potential for effects that are, in fact, of no relevance to humans. In other cases, a transgenic murine system may be established which is capable of recapitulating some elements of the disease and use of the product; these can be complex. Regulators should recognise that in some circumstances, it is appropriate to conclude that there is no in vivo animal testing that can rationally support later development: these instances are rare and not limited to cell therapy products (eg some monoclonal antibodies can lack activity in all species except humans). Despite the lack of preclinical safety data, these products may yet offer great potential, typically for a small number of patients who otherwise would have no specific therapeutic options. Dialogue with regulators to facilitate understanding is encouraged such that clinical development can be supported.
3
tumours and, secondly, does its administration as proposed in this trial put patients at unreasonable risk of tumours? Where a product is subject to ex vivo expansion, there is a natural concern that needs to be addressed which is whether the expansion steps result in any risk of cell immortalisation. Proof of absence of uncontrolled proliferation, clonogenic potential and characterisation of telomere length is advised from in vitro studies. In vivo testing of potential of a human cell therapy product to form tumours in vivo after its single subcutaneous administration to immune-incompetent animals is of interest: if this is negative with testing supported by use of a positive control cell lines, no hindrance to clinical testing is identified. If such testing is positive, the cell line may be judged to have an unreasonable safety profile. Alternatively, it may be that development of tumours in animals can be argued to not predict what is likely to happen in humans. One other element should be mentioned: guidance from the World Health Organisation (WHO) [7] discusses tumourigenicity testing for cell substrates: this guidance aims to address testing of the production process to ensure consistent product quality where a product is produced by cell-based mans. This is not the same as studies to address risk of whether a cell-based product poses a risk of tumourigenicity. 3. Some specific points for consideration Some regulatory bodies offer scientific advice procedures, with the intent of assisting developers of medicinal products. The aim of such meetings is for the regulators to indicate, which among a range of options, might result in the developer presenting data that would support a decision to approve an application or to identify issues at regulatory review that might pose challenges and how the developer can anticipate these. Some issues discussed in these procedures are considered below. Question: Should animal cells be used in preclinical testing? Comment: This may reflect the stage of development. At the stage of exploratory research, it could be that initial data supporting the proof of principle is generated by testing with animal cells. However, if the product to be placed on the market is a human cell product, development studies should focus on this product, not on an animal-cell-based product.
2.5. Testing for tumour risk
Question: Should a large species be used in preclinical testing? Comment: There is no single answer to this question. In some circumstances, there will be data only from testing in rodents and this can suffice: clinical studies can be supported by preclinical testing only in mice, for instance. In other circumstances, it may only be possible to generate proof of concept data from testing in large animals or there may be an existing veterinary product and clinical development of a similar product is being undertaken. Thus, applicants may be able to present data from testing in dogs, horses, goats, pigs, sheep or other species. In other cases, there may be intent to deliver the product to an unusual anatomical site with a novel device, and so testing to support this is done in large animals in order to demonstrate its feasibility and expectation for its success when applied to humans because dosing so in a rat or mouse could not be done with the device. Whilst each case therefore needs to be considered on its own merits, it is clear that testing in a large animal species should not be done unless such testing is essential because its absence would create a deficiency that might pose risks to patients.
Two considerations should be faced in terms of tumour risk assessment: does this product have inherent potential to elicit
Question: Do preclinical studies indicate whether immunosuppressive drugs be used in clinical trials?
Please cite this article in press as: McBlane JW, Preclinical safety testing for cell-based products using animals, Biologicals (2015), http:// dx.doi.org/10.1016/j.biologicals.2015.05.002
4
Meeting report / Biologicals xxx (2015) 1e4
Comment: In many instances, studies are done in immunoincompetent animals because immuno-competent animals would reject a human cell product, so invalidating the testing or causing effects that are difficult to translate to humans. In other cases, immunosuppressive drugs are given to normal animals to prevent human cell rejection. Does this suggest that immunosuppressive drugs should be used in clinical use of the product? Probably this question should only be addressed clinically; that is, if it is feasible to avoid use of immunosuppressive drugs, clinical evidence should be generated to determine whether responses are better, worse or not different, depending on use, or not of immunosuppressive therapy. This can be addressed systematically in clinical trials that adopt dosing in the presence of immunosuppressive therapy and in its absence.
4. Preclinical development of cell therapies From the foregoing, it can be summarised that at the time of first clinical trial application, and excepting a small number of products for which it can be argued that this testing is not relevant, a typical preclinical development programme will include data on the following. - primary pharmacology e evidence supporting the expectation for therapeutic benefit - biodistribution & cell persistence e likely to be human cells given to animals, not in the context of a disease model - general toxicity e after a single dose by a route that is intended to be by the clinical route with follow up determined with reference to results from data on cell persistence - where the clinical dosing is to be repeated, repeat dose general toxicity data may be expected - in vitro and perhaps in vivo studies on risk of tumour formation.
5. Conclusions In assessing an application, whether for a clinical trial or for a marketing authorisation, the assessor has to judge whether the data supplied support the claim by the applicant that approval is reasonable and safe. Specific challenges posed by cell therapy products are different from small molecules and require a different approach. Nevertheless, across different types of cell therapies, similar themes emerge [8]. In considering this, the primary considerations can be identified as:
1 Is the product to be used clinically the same as that used in the preclinical testing programme? What is the nature of the evidence for this? Where an applicant cites data from a published paper, how can the assessor judge that those results can be directly applied to the product that is the subject of the application under consideration? If there are differences, does recognition of this have an impact on the decision? 2 Is the basis for selection of the first human dose sound? As the initial studies are in patients, use of doses expected to be beneficial should first be characterised e this is likely to refer to both clinical and preclinical data, but for novel products it may be that there are no previous relevant clinical data. The developer should then consider what evidence is presented for the expectation that these doses will be safe. 3 Testing in animals should be designed to provide only information that is essential in order to be able to safely test the product in patients; it is the nature of preclinical testing that it becomes of lesser relevance once clinical data are generated. 4 For cell therapy products, engagement with regulators, for instance in scientific advice meetings, is encouraged. However, this should not be considered a routine step of a means of obtaining an opinion on a potential application: rather it should aim to present issues where there are several options available for progression, each with likely detrimental elements that argue against their use. References [1] Broichhausen C, Riquelme P, Ahrens N, Wege AK, Koehl GE, Schlitt HJ, et al. In question: the scientific value of preclinical safety pharmacology and toxicology studies with cell-based therapies. Mol Ther d Meth Clin Dev 2014;1:1e12. [2] ICH M3 (R2) Non-clinical safety studies for the conduct of human clinical trials and marketing authorization for pharmaceuticals. [3] ICH S7A note for guidance on safety pharmacology studies for human pharmaceuticals CHMP/ICH/539/00. [4] Morgan RA, Yang JC, Kitano M, Dudley ME, Laurencot CM, Rosenberg SA. Case report of a serious adverse event following the administration of T cells transduced with a chimeric antigen receptor recognizing ERBB2. Mol Ther 2010;18(4):843e51. [5] Linette GP, Stadtmauer EA, Maus MV, Rapoport AP, Levine BL, Emery L, et al. Cardiovascular toxicity and titin cross-reactivity of affinity-enhanced T cells in myeloma and melanoma. Blood 2013;122(6):863e71. [6] Rosenberg SA. Of mice, not men: no evidence for graft-versus-host disease in humans receiving T-cell receptor-transduced autologous T cells. Mol Ther 2010;18(10):1744e5. [7] WHO evaluation of cell substrates for the production of biological: revision of recommendations. 2009. Report of the WHO Study Group on cell substrates for the production of biologicals. [8] Schneider CK, Salmikangas P, Jilma B, Flamion B, Todorova LR, Paphitou A, et al. Challenges with advanced therapy medicinal products and how to meet them. Nat Rev Drug Discov 2010;9:195e201.
Please cite this article in press as: McBlane JW, Preclinical safety testing for cell-based products using animals, Biologicals (2015), http:// dx.doi.org/10.1016/j.biologicals.2015.05.002