Cell-based therapies represent a new therapeutic approach for stroke. In 2007, investigators from academia, industry leaders, and members of the National Institutes of Health crafted recommendations to facilitate the translational development of cellular therapies as a novel, emerging modality for stroke from animal studies to clinical trials. This meeting was called Stem Cell Therapies as an Emerging Paradigm in Stroke (STEPS) and was modeled on the format of the Stroke Therapy Academic Industry Roundtable (STAIR) meetings. Since publication of the original STEPS guidelines, there has been an explosive growth in the number of cellular products and in the number of new laboratory discoveries that impact the safety and potential efficacy of cell therapies for stroke. Any successful development of a cell product will need to take into consideration several factors, including the preclinical safety and efficacy profile, cell characterization, delivery route, in vivo biodistribution, and mechanism of action. In 2010, a second meeting called STEPS 2 was held to bring together clinical and basic science researchers with industry, regulatory, and National Institutes of Health representatives. At this meeting, participants identified critical gaps in knowledge and research areas that require further studies, updated prior guidelines, and drafted new recommendations to create a framework to guide future investigations in cell-based therapies for stroke.
Cell-based therapy is a potential new treatment approach for stroke. Over the past 20 years, there have been extensive efforts to develop and translate new stroke therapies, but there remains no proven treatment aside from tissue plasminogen activator for acute ischemic stroke. When neurological deficits persist, despite acute treatment, there is no Food and Drug Administration-approved therapy to enhance recovery. Given the difficulties of identifying new treatments for stroke and the promising results of cell therapy in animal stroke models, investigators from academia, industry leaders, the National Institutes of Health, and the Food and Drug Administration convened in 2007 to discuss research guidelines in the field following the format of the prior Stroke Therapy Academic Industry Roundtable (STAIR) meetings.1 This meeting was called Stem Cell Therapy as an Emerging Paradigm for Stroke (STEPS).2
Since publication of the first STEPS meeting, there has been an explosive growth in the number and types of cells under investigation for stroke. Cells have been prepared and isolated from a range of different tissues, including blastocysts, embryonic and fetal tissue, neural tissue, bone marrow, peripheral blood, umbilical cord, placenta, amniotic fluid, menstrual blood, dental pulp, and adipose. Induced pluripotent cells have emerged based on new technology to reprogram adult skin fibroblasts into pluripotent stem cells with the potential to differentiate into cells from all 3 germinal layers, including neurons and other cells that comprise the nervous system. Many types of cell-based preparations are composed of heterogeneous cell populations such as umbilical cord blood or the mononuclear fraction of bone marrow. Even some types of more purified populations of bone marrow such as marrow stromal cells may be heterogeneous depending on culture passage and isolation procedures. Not all types of cell-based preparations necessarily include stem cells and the field may be more appropriately termed cell-based therapy rather than stem cell therapy. Clinical trials testing cellular products in patients with stroke have emerged since the STEPS 1 publication and are mainly focused on the use of autologous mixed cell populations. The application of allogeneic, “off-the-shelf” cells to patients with stroke is poised for early-stage clinical testing. It is therefore timely and necessary to update preclinical and clinical trial guidelines for translating cell-based therapies for stroke. A workshop was held on crafting suggestions for preclinical studies that should be performed on any cellular product that is being developed as a potential therapeutic for stroke. A second workshop focused on suggestions for early-stage clinical testing of cellular products in patients with stroke. The recommendations from these workshops are described subsequently and agreed on by the participants listed at the end of this article following the format of the prior STAIR meetings.1
Updated Preclinical Guidelines
The prior STEPS document2 described recommendations on preclinical testing. We refer back to the original document regarding cell delivery (Table 1) and cell dosing (Table 2). We now provide modifications and add new recommendations regarding the following factors that apply to both ischemic stroke and intracerebral hemorrhage (Table 3).
The intended cellular product needs to be sufficiently described for several purposes, including cell identity and characteristics, conducting experiments by other groups for reproducibility, and evaluating safety risks. At a minimum, it is important to provide immunophenotyping in any peer-reviewed publication. For ex vivo expanded products and nonexpanded products, it is suggested to perform and publish transcriptional profiling as an open code approach to cell characterization. It is recommended that references be provided citing laboratories that have independently derived the same characterized cell therapy product using published methodologies. Guidance documents from the Food and Drug Administration on cell characterization ask for information about the identity, purity, viability, potency, stability, and dosage (www.fda.gov/cber/guidelines.htm).
We refer to the prior STEPS document and add the following recommendations. A stepwise approach would be useful to test a cellular product in models that address the heterogeneity of different types of stroke. Rodent models are well established and multiple strains and genetic backgrounds can be exploited. Large animal models may be helpful for specific situations in which they permit testing of specific neuroanatomical structures (white matter), specific types of imaging, or delivery options. Primates and other large animals also allow for testing in gyrencephalic brains. Animal models should be exploited to examine the effects of age (young versus old), gender, and comorbidities (hypertensive, diabetic, etc) on the therapy being investigated. These baseline conditions are important given that patients with stroke tend to be older and have vascular risk factors. Testing of cellular products in multiple, independent laboratories is crucial for reproducibility, robustness of effect, and to broaden the compendium of preclinical studies.
There are many types of focal ischemic stroke and intracerebral hemorrhage models causing injury in cortical or subcortical areas of the brain. Evaluation of a cell-based approach is important in multiple focal ischemic stroke or intracerebral hemorrhage models using appropriate histological and behavioral tests. We recommend models that produce deficits that persist up to 4 weeks after stroke.
Preclinical Safety Indices
Safety includes tumorigenicity, immune sensitization, biodistribution, persistence, and cell fate and these issues are referenced in the following guidelines from the Food and Drug Administration (www.fda.gov/cber/guidelines.htm). As stated in the STEPS 1 document, cell therapy studies should include measures for detecting tumor or ectopic tissue formation, overt behavioral abnormalities, and adverse physiological alterations according to Food and Drug Administration guidelines. The duration of safety testing will vary depending on the cell type, but exogenous cells that die within days to weeks after injection in vivo or that already have been proven safe in patients with other clinical disorders may not require long-term testing in animals. Other types of cells with high proliferative and differentiation profiles such as embryonic or neural progenitor cells will likely require more extensive and long-term monitoring such as histopathology to assess for overgrowth and tumor formation.3 Positive controls for tumor formation or overgrowth, when available, and relevance of immunosuppression regimens should strongly be considered. All adverse behaviors during the life of the animal after cell injection should be evaluated and tracked if observed. Acute toxicity of relevant organ systems should also be tested based on the delivery route. For example, the effects of cells on cerebrovascular blood flow or cerebral perfusion should be evaluated for an intra-arterial route of delivery.4 Pulmonary function should be evaluated for an intravenous delivery route for cells that accumulate within the first-pass filter of the lungs.5 Such tests might include respiratory rate and arterial blood gases. The rate of infusion is an important variable with respect to assessing these safety outcomes.
The primary goal of initial testing should be to address safety risks evaluated around cell identity, method of isolation, and expansion procedures. Once safety is established, functional end points should be the mainstay of primary outcomes. There are various behavioral outcomes within the domains of motor control, sensation, and cognition.6 Testing a cellular therapy using multiple different behavioral studies is favored to support robust efficacy. A battery of behavioral end points should be selected that are sensitive to the degree of injury, sites of damage, and severity of impairment.7,8 Testing should be performed multiple times in a longitudinal fashion for at least 1 month after treatment. Positive, neutral, and negative outcomes should be reported. It is also recommended to test cellular therapies in >1 laboratory to assess reproducibility of safety and efficacy.
It is important to establish a dose–response curve and determine an optimized dose and treatment schedule as well as the minimum threshold for observed benefit. The chosen preclinical regimen should correlate with the intended clinical protocol, including delivery route and treatment schedule regimen, with single and cumulative dose greater than anticipated in clinical testing. There are limited data available regarding serial dosing for benefit or with respect to immune sensitization; further research is therefore encouraged. Negative controls are the subject of much debate. At a minimum, we recommend the vehicle solution of the cellular product. Other controls include dead cells, although cellular debris might be less desirable compared with cells that remain intact but are nonfunctional. It has been shown that freeze–thawing of grafted cells can worsen outcome after stroke.9 If immunosuppression will be needed in a clinical trial, it is recommended to study the cellular product with immunosuppressive agents along with a separate group receiving the immunosuppressive agents alone. Consideration may also be given to applying clinically relevant rehabilitation to all treatment groups in functional testing.10 Finally, comparing different therapeutic cell products would contribute greatly in this emerging field.
Biodistribution and Cell Persistence
Studying cell deposition, migration, persistence, and fate in stroke models may have value relative to defining mechanistic pathways. Because engraftment of delivered cells remains low whenever it has been examined, methods to improve engraftment should be evaluated for those cellular products in which engraftment is necessary to achieve benefit. Noninvasive imaging to address these issues is insightful and could be developed as a surrogate biomarker for translation to the clinical arena.
Mechanisms of Action
Defining the underlying mechanisms of therapeutic action may contribute to timing and duration of therapy, accurate clinical end point selection, and appropriate biomarkers for treatment response. Epigenetics, tissue microarray, and other emerging technologies are providing insight into mechanism of action of cellular therapeutics. Studies should consider cell–host interactions, including the site of injury, immune system effects, interaction with parenchymal cells, and remodeling of the microenvironment. Such approaches may also rule out irrelevant pathways and give insight to clinical trial design. Although some studies suggest that certain cell types when injected into the brain after stroke may lead to differentiation of donor cells into host brain cells, the majority of exogenous cells under investigation at the present time exert so-called “nursing functions” to the injured brain such as cytoprotection or stimulation of endogenous repair mechanisms.11 Clarifying the mechanisms of action is generally useful but is not a prerequisite for proceeding to human clinical trials provided sufficient, encouraging, and reproducible preclinical evidence of efficacy exists.
Guidelines on Designing Early-Stage Clinical Trials
When to Start Clinical Trials
We encourage confirmation of pivotal preclinical results in at least 2 laboratories and 2 species (Table 4). Understanding the mechanism of action is not essential before initiating clinical trials but such information is desirable to plan strategies, including treatment regimen, route of administration, and outcome measures.
We highly encourage initial testing in patients with stroke, not healthy control subjects, and enroll patients who will be informative based on safety profile and the anticipated biological effect of the cellular product. The selection of heterogeneous (eg, all types of ischemic stroke) versus patients with homogeneous stroke (eg, middle cerebral artery stroke) depends on a number of factors. Including patients with heterogeneous stroke improves recruitment and provides more robust safety information, whereas a more homogeneous stroke population may be more desirable for detecting early efficacy signals or determining a biological target. The size and location of the infarct may be important to use as selection criteria, particularly when efficacy is a consideration. Inclusion and exclusion criteria may vary with the cell type, delivery, and treatment time window.
Route of Therapy and Biocompatibility of Devices
We refer to the STEPS 1 document2 and add that the route of delivery should be based on preclinical data regarding mechanism, biological target, and cell type. Assessing the biocompatibility of devices with the cell product is useful and important.12 More information can be found in the STEPS 1 guidelines.
Timing of Cell Therapy
Preclinical data and the proposed mechanisms of action should drive decisions regarding timing of therapeutic delivery. In addition to exploring the optimal timing for effective cell therapy, the window for enrollment should also consider any information regarding when after stroke the cell product is not effective. A well-defined therapeutic window in animals is therefore highly encouraged. Classifying the timing of injury into categories such as acute, subacute, and chronic based on biological activity will eventually be necessary, but, at the present time, there is insufficient knowledge to fully define these temporal categories.
Role of Imaging in Clinical Trials
It is important to clarify the intended purpose of imaging methods, which can be applied for various purposes, including patient selection, surrogate end points, safety, and exploration of mechanism (eg, repair measures). We advise incorporating imaging to establish the size and location of the infarct. When feasible, advanced imaging techniques may be considered for exploring the mechanisms of action or surrogates of activity of the cellular therapy. Several imaging biomarkers of recovery are actively being explored.13 Further studies are needed, however, to validate imaging end points as surrogate outcomes measures. To this end, a stroke recovery neuroimaging consortium is highly recommended. Imaging is also very useful in the preclinical setting to monitor biodistribution of delivered cellular products. Although there are no accepted techniques to label and monitor cells for clinical testing, several approaches are currently available, including iron, indium, thallium, gadolinium-based agents, etc.14,15 More investigation is urgently needed to develop safe and reliable labeling techniques for deployment in clinical trials. Whatever labeling approach is chosen, it is important to assess that the label does not impair viability of the cellular product. It is also recommended to test the effects of the label on various in vitro functional assays of the cellular product.
The decision to consider immunosuppression is based on a number of factors, including whether the cellular product is autologous or allogeneic. At present, it is unknown whether immunosuppression in a stroke clinical trial is necessary for some allogeneic cells that have been shown to exert immunomodulatory effects. Immunosuppression may be more relevant if long-term engraftment of the cellular product is thought to be required for effectiveness. If immunosuppression is used, there should be a robust monitoring plan and follow-up in all early-phase trials. Another consideration is HLA matching, the benefits of which are well known in transplantation biology.16,17
Controls in Cell Therapy Trials
Comparison of outcomes to a placebo arm may be useful, particularly for detecting initial evidence for efficacy, but no early-phase study would likely be sufficiently powered to detect a difference. However, in early-phase studies, safety issues are most important and control subjects reduce the sample size of informative patients. Placebo control subjects, nevertheless, may allow a reasonable comparison of safety outcomes with a similar population treated under the same conditions with the vehicle as patients treated with active cells. One way of addressing this issue is to use an uneven randomization scheme, which assigns a higher number of active to placebo subjects. We therefore recommend justification for incorporating a placebo arm in Phase I/IIa testing. An alternative approach is to use historical data from a database such as Virtual Stroke International Stroke Trial Archive (VISTA). Standard of care should be provided to all control patients. We recommend using American Heart Association guidelines for rehabilitation to ensure standardization of poststroke care. Capturing and controlling for confounding factors is highly encouraged.
Safety end points will likely be negotiated with regulatory agencies and should be driven around cell type, delivery routes, biodistribution of cells, and other preclinical data. Similarly, the duration of monitoring for safety end points needs to be negotiated with regulatory authorities. For cell types that die within days after administration, long-term monitoring beyond 6 months is likely unnecessary. Intravenous delivery of exogenous cells should be monitored for acute infusional toxicities and pulmonary complications. The selection of functional end points in stroke is the subject of much debate. The traditional outcome measures of the National Institute of Health Stroke Scale, modified Rankin Scale, or Barthel Index still have merit but other, more novel end points should be developed and considered. Domain-specific modalities such as language or hand function may also be suitable or even more desirable outcome measures in efficacy studies. Any novel outcome measures should be validated and peer-reviewed.
Cell-based therapies may represent a new therapeutic modality for stroke. Not all types of cell-based preparations necessarily include stem cells. Therefore, this emerging field may more appropriately be termed “cell-based therapy” rather than solely “stem cell therapy.” Nevertheless, all of these approaches fall under the rubric of “regenerative medicine,” which represents a cutting-edge approach to ischemic injury of the nervous system. To accelerate the field of cell therapy for stroke, we have updated the recommendations from the prior STEPS meeting and identified key translational barriers that need further study, including cell labeling, imaging, biodistribution of exogenous cells in patients, and identifying imaging biomarkers of stroke recovery (Table 5). Given the monumental failures of neuroprotective agents for acute stroke over the past 20 years, these guidelines are based, in part, on the lessons learned from those prior failures in the hopes of facilitating the successful development of cellular therapies for stroke from preclinical studies to early-stage clinical trials.
S.I.S. had received consulting fees from J&J, Celgene, and Aldagen. STC has received a speaking fee from Cortex Pharmaceuticals.
Sean I. Savitz, Larry Wechsler, Robert Deans, Michael Chopp, Tom Carmichael, and Donald Phinney.
Jaraslow Aronowski, Martin Bednar, Johannes Boltze, Cesar Borlongan, Tom Carmichael, Thomas Chase, Michael Chopp, Dale Corbett, Charles S. Cox, Steven Cramer, Robert Deans, Steven Fischkoff, Joseph Frank, David Greenberg, David Hess, Klaudyne Hong, Minako Koga, Theresa Jones, Armand Keating, Zaal Kokaia, Robert Mays, Keita Mora, Mark Pittenger, Donald Phinney, Paul Sanberg, Sean I. Savitz, Tim Schallert, John Sinden, Evan Snyder, Gary Steinberg, Larry Wechsler, Steven Victor, Alison Willing, Ernest Yankee, and Dileep Yavagal.
- Received September 6, 2010.
- Revision received November 24, 2010.
- Accepted November 26, 2010.
- © 2011 American Heart Association, Inc.
Recommendations for standards regarding preclinical neuroprotective and restorative drug development. Stroke. 1999;30:2752–2758.
Stem Cell Therapies as an Emerging Paradigm in Stroke Participants. Stem cell Therapies as an Emerging Paradigm in Stroke (STEPS): bridging basic and clinical science for cellular and neurogenic factor therapy in treating stroke. Stroke. 2009;40:510–515.
- Seminatore C,
- Polentes J,
- Ellman D,
- Kozubenko N,
- Itier V,
- Tine S,
- Tritschler L,
- Brenot M,
- Guidou E,
- Blondeau J,
- Lhuillier M,
- Bugi A,
- Aubry L,
- Jendelova P,
- Sykova E,
- Perrier AL,
- Finsen B,
- Onteniente B
- Walczak P,
- Zhang J,
- Gilad AA,
- Kedziorek DA,
- Ruiz-Cabello J,
- Young RG,
- Pittenger MF,
- van Zijl PC,
- Huang J,
- Bulte JW
- Biernaskie J,
- Chernenko G,
- Corbett D
- Janowski M,
- Walczak P,
- Date I
- El Khoury R,
- Misra V,
- Sharma S,
- Cox CS,
- Walker P,
- Grotta JC,
- Gee A,
- Suzuki S,
- Savitz SI
- Barbosa da Fonseca LM,
- Battistella V,
- de Freitas GR,
- Gutfilen B,
- Dos Santos Goldenberg RC,
- Maiolino A,
- Wajnberg E,
- Rosado de Castro PH,
- Mendez-Otero R,
- Andre C