©2021 by The International Society for Stem Cell Research. All rights reserved.

No part of this document may be produced in any form without written permission of The International Society for Stem Cell Research.

3.3.3 Efficacy Studies

Given the therapeutic goals of stem cell-based interventions, preclinical studies should demonstrate evidence of therapeutic effect in a relevant animal model for the clinical condition and the tissue physiology to be studied. Mechanistic studies utilizing cells isolated and/or cultured from animal models or human tissues, both diseased and controls, are critical for defining the underlying biology of the cell-based interventions. However, a complete understanding of the biological mechanisms at work after stem cell-based intervention is not a prerequisite to initiating trials, especially when trials involve serious and untreatable diseases for which efficacy and safety have been demonstrated in relevant animal models and/or in conclusive human studies with the same cell source.  Further, in rare cases, appropriate animal models may not exist. In these cases, in vitro studies may be used to support the rationale for potential efficacy.

Efficacy Evidence for Initiating Trials
Recommendation Trials should generally be preceded by compelling preclinical evidence of clinical utility in well-designed studies. Animal models suited to the clinical condition and the tissue physiology should be used, unless there is evidence of efficacy using similar products against similar human diseases, or if it is not feasible to establish appropriate or predictive animal models.

Rigorous preclinical testing in animal models is especially important for stem cell-based approaches because cell therapies have distinctive pharmacological characteristics. Before clinical testing, preclinical evidence should ideally provide the following: 

  1. Mechanism of action. Preclinical studies should establish evidence connecting a cell-based intervention’s therapeutic activity in animal models to a pathophysiological process. These studies establish the localization of transplanted cells and provide evidence that the predicted localization is tied to the proposed mechanism of action. 
  2. Optimal conditions for applying the stem cell-based intervention (for example, dose, co-interventions, delivery).
  3. Ability to modify disease or injury when applied in suitable animal systems, and under conditions that are similar to expected trials (see design principles under Section, Study Validity).
  4. Sufficient magnitude and durability of disease modification or injury control to be clinically meaningful.

In cases where an intervention is substantially similar to one that has already been tested in humans, trial evidence may reduce the demand for preclinical evidence.

Animal Studies
Recommendation Appropriate animal models should be selected which allow the assessment of efficacy and safety of the stem cell-based intervention. Safety testing should include assessment of the delivery procedure or surgical technique used for implantation of the cells.

Immune-deficient rodents or those manipulated to have humanized immune systems can be especially useful to assess human cell transplantation outcomes, engraftment in vivo, stability of differentiated cells, and cancer risk. Many small animal models of disease can faithfully reproduce aspects of human diseases, although there are considerable limitations. Small animal studies should also attempt to correlate cell number and potency required for large animal studies and subsequent trials.

Large animals may better represent human physiology as they are often genetically outbred, maintained in more diverse environments, and anatomically more similar. They may provide the opportunity to test co-interventions used in trials (for example, adjunctive immunosuppressive drug therapy) methods of introduction or the compatibility of surgical devices and cell products. They also may be essential to evaluate issues of manufacturing scale up, or anatomical factors that are likely to mediate a therapeutic effect (for example, bone, cartilage, or tendon in a load-bearing model). Trials involving risky or novel approaches should generally be supported by evidence from large animal models in cases where such models better recapitulate human disease and human anatomy (e.g., cardiomyocyte transplantation).

The need for invasive studies in non-human primates should be evaluated on a case-by-case basis, performed only if trials are expected to present high risk, and where non-human primates are expected to provide information about cell-based interventions not obtainable with other models. All studies involving the use of non-human primates must be conducted under the close supervision of qualified veterinary personnel with expertise in their care and their unique environmental needs. Particular care should be taken to minimize suffering and maximize the value of studies by using rigorous designs and reporting results in full.