| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 Novartis Institutes for Biomedical Research, Cambridge, Massachusetts
Correspondence and requests for reprints should be addressed to David M. Rodman, M.D., Translational Medicine Head, Respiratory Novartis Institutes for Biomedical Research, 400 Technology Square, Cambridge, MA 02139. E-mail: david.rodman{at}novartis.com
ABSTRACT
Proof-of-concept (PoC) trials are an important component of the new therapeutic development paradigm, contributing to the goal of accelerated progression from bench to bedside. Developing a simplified PoC strategy for lung stem or progenitor cell therapy can be helpful in focusing initial efforts and identifying methodologic hurdles in humans at a relatively early stage of development. In this article, the use of lung radiation injury as a model of epithelial injury and regeneration is proposed as a potential PoC strategy. Strengths of the approach include well-understood biology, predictive preclinical and translational models, a tractable human trial design, and the potential to improve outcome in an area of high unmet medical need. While significant hurdles remain, this approach could provide the first interpretable results within 4 years of the decision to proceed, thus greatly accelerating the translation of lung stem cell therapy to humans.
Key Words: radiation • fibrosis • inflammation
The modern concept of drug development incorporates the rapid progression of preclinical research through the early phases of human studies. The goal is to build confidence in the approach and identify critical issues for development that can only be addressed in humans. At the core of this approach is the concept of proof-of-concept (PoC) trials, which are performed during or soon after the first-in-human safety and tolerability studies.
PoC is a term that can have many meanings, but at its core is the basic goal of testing the therapeutic agent in a relevant disease model or disease population in a small, rapid mechanistic study that validates the approach and provides the information necessary for quality decision making (1). PoC can be conceptualized as building confidence and significantly "de-risking" the project, so that approaches that are unlikely to be successful are eliminated before large-scale investment in an expensive and potentially risky registration effort is made. A key secondary benefit of PoC is generating data in humans that can be used to validate preclinical assumptions and improve future translation from bench research to clinical development.
ADAPTING PoC TRIALS TO THE DEVELOPMENT OF LUNG STEM CELL THERAPY
Lung stem or progenitor cell therapy is a particularly challenging concept on which to model PoC trials. The aspirational goal of lung stem cell therapy may be for lung regeneration in individuals with emphysema or abnormalities in lung development, gene correction for inherited disorders such as cystic fibrosis or surfactant protein-B deficiency, and resolution of normal lung function after a severe insult such as acute lung injury (2–4). However, none of these goals lend themselves readily to a short, well-controlled, confidence-building trial. All entail multiple levels of complexity, such as the need to design a substrate for lung growth, correct a genetic defect in the progenitor cell, or to attain high levels of long-term engraftment. Given these hurdles, it might seem that translation of stem cell therapy to lung disease is decades away from reality.
However, by reducing the level of complexity, one can develop an approach to PoC that could be addressed in a matter of several years, rather than decades. Three general areas must be addressed in the preclinical stage of development to allow this to occur: (1) optimization of stem/progenitor cell isolation and manipulation; (2) validation and refinement of the therapeutic approach; and (3) assessment of toxicology and safety in a translational preclinical model, ideally in a nonhuman primate.
Whether the source of cells is autologous bone marrow–derived progenitors, umbilical cord blood stem cells, fetal stem cells, or some other source is a critical question. Also important are the conditions used to maintain, manipulate, expand, and differentiate the cell population before delivery. At present, these are the critical concerns facing the field and will likely be so for some time (5). However, one can propose a paradigm that minimizes the hurdles involved. For instance, one can put aside for the present the problem of large-scale production, and focus on using autologous bone marrow–derived progenitor cells for initial PoC. This would avoid many methodologic hurdles as well as the need for allogenic tolerance. Second, one can focus on a paradigm that involves only a limited region of the lung (rather than diffuse disease), potentially one that can be restricted to a segment of lung that can be readily accessed by bronchoscope. This would reduce potential safety concerns (because cell-based therapy also has the risk of failing to help if cells are too differentiated or it may even do harm to healthy cells) and provide a reasonable method for delivery and sampling. Finally, one could identify a translational paradigm that requires regeneration of the most accessible cell population, epithelium, and has measurable therapeutic effects in a period of weeks, rather than months to years.
The ideal translational paradigm incorporates a predictive animal model of the human condition that can be used for fine-tuning the approach, development of biomarkers, assessment of toxicity, and therapeutic index. This would include identification of the minimum anticipated biological effect level (MABEL) and the no adverse effect level (NOAEL). These parameters can then be used to set the safe starting dose, the likely therapeutic dose, and the maximum dose likely to be safe and achievable. The translational model can also validate the biomarkers and clinical measures that will be used in the PoC to establish mechanistic and clinical evidence of engraftment and efficacy.
RADIATION-INDUCED LUNG INJURY AS A MODEL FOR PoC
A potential pathway forward that satisfies all of the requirements for simplification and translation is that of radiation-induced lung injury (6). Although this paradigm is not currently the focus of commercial cell-based therapy programs, there is unmet medical need because, for some radiation oncology indications, lung toxicity is dose limiting and thus reduces the probability of cure. Amelioration of lung damage could also be useful for therapy for industrial and military personnel exposed to lethal sources of radiation. Although obtaining autologous bone marrow might be difficult in the latter circumstance, for many neoplastic indications it should be feasible. Figure 1 shows a scheme for what is understood to take place after radiation exposure in both rodents and humans. After external beam radiation, there is an initial period of apoptotic cell death affecting primarily epithelial and endothelial cells (7). After the period of parenchymal cell death, a macrophage- and neutrophil-predominant inflammatory response occurs (8). This phase lasts several weeks and resolves spontaneously. Subsequently, after a period of weeks to months, progressive fibrosis occurs. Comparisons between mouse strains have demonstrated that these two phases can be distinguished, with some strains relatively resistant to inflammation, but not fibrosis, and vice-versa (9). Therapies that reduce the inflammatory response, such as systemic corticosteroids, are relatively ineffective at preventing the late fibrotic response. This could be an important observation, because mesenchymal progenitor cells appear to confer a strong antiinflammatory effect that may not require long-term engraftment but may provide clues to developing nonsteroidal antiinflammatory therapies for radiation injury (10).
|
A potential mechanism through which progenitor cells may be instructed to differentiate into alveolar epithelial cells in the irradiated lung was recently reported (14). Radiation-treated lung was found to lead to apoptosis of alveolar type II cells with shedding of microvesicles of approximately100-nm diameter from type II cells which, when incubated with whole bone marrow, resulted in differentiation into a cell with type II cell markers. Thus, it is possible that the introduction of mesenchymal progenitors into the radiation-injured lung can lead to rapid differentiation and repopulation of the alveolar epithelial surface, limiting injury and fibrosis. In contrast, introduction of progenitors at later time points exposes them to transforming growth factor-β and other profibrotic signals, resulting in exacerbation of injury (15). If this hypothesis can be validated, then the need to differentiate progenitors ex vivo could be obviated, thus accelerating the pace of translation.
Although the focus of this discussion is on radiation-induced injury, it is likely that the timing of progenitor cell introduction into the lung may be critical to success in acute lung injury, fibrotic lung disease, and other situations where the local environment contains strong local signals for cell recruitment and differentiation. In the later stages of acute lung injury and fibrosis, profibrotic cytokines like transforming growth factor-β may predominate and drive progenitor cells toward a myofibroblast phenotype, worsening the disorder. In contrast, early acute lung injury may be similar to the early phase after radiation injury, and provide a fertile environment for epithelial cell differentiation. Chronic disorders such as chronic obstructive pulmonary disease and cystic fibrosis may present different challenges, and characterization of local effects on progenitor cell differentiation will be critical to moving these therapies forward.
Figure 2 shows the steps necessary to use the radiation injury PoC strategy. In yellow are the steps necessary to prepare the cell-based therapy for use in translational studies. These studies include optimization of isolation and processing of the cells as well as tests in the rodent model of radiation injury. These studies would be directed toward optimization using rodent reagents as well as identification of biomarkers suitable for translation to nonhuman primate studies and human trials.
|
The final phase of the strategy is the first in human and PoC trial. This would take place in radiation-treated individuals, initially with relatively low doses of cells designed to establish safety followed by progressively higher doses. An adaptive design enabling the accumulation of safety data during an accelerated dose-escalation protocol could be applied to allow the rapid progression to the PoC phase of the trial. Figure 3 shows a hypothetical time line for progression from preclinical studies to first interpretable results within 4 years of optimizing the cell production procedures. Implicit in the design is that the first set of reagents will not be entirely optimized, so a back-up program with improved reagents and methodology would be planned to follow approximately 2 years behind the lead program, building on knowledge gained from the first-in-human studies. Figure 4 shows a proposed set of decision criteria for the PoC trial. Using these criteria, sufficient confidence in the potential for the approach could be gained to justify the substantial additional investment that would be required to move from PoC through the scale-up processes and phase II/III studies necessary to make lung stem cell therapy widely available. Although this program would be narrowly focused on developing cell-based therapy to prevent radiation injury, the knowledge gained could be readily applied to the even more challenging development of treatment of established lung diseases.
|
|
At present, there are numerous small academic trials in progress using some form of progenitor cell therapy for a variety of conditions. The majority use bone marrow–derived cells with various techniques for purification, amplification, and so forth. Although a great deal of excitement surrounds these trials, one must be cautious in anticipating and interpreting their results. A positive trial in the absence of strong mechanistic data and documentation and quantification of engraftment would be a tenuous foundation upon which to base further development activities.
Many trials include limited controls, lack demonstration of engraftment, and may be confounded by transient antiinflammatory effects of the therapy that are observed in preclinical studies. The latter point deserves some comment. It appears that mesenchymal progenitor cells when infused systemically can produce a potent, transient antiinflammatory effect (17). Although this may be beneficial, it is not the principal objective of regenerative stem cell therapy and could divert the field from its primary objective. Furthermore, until the mechanism is established, it would be uncertain if the observed benefits could be more reliably, safely, and economically produced by pharmacologic antiinflammatory therapies, such as the use of an IL-1RA or recombinant IL-10, both of which are potential mediators of this effect (17, 18). (Although the antiinflammatory effect must be accounted for in any trial design, there may be value in determining the mechanism, potential therapeutic benefit, and role in modulating successful engraftment). It is critical that the PoC trials incorporate carefully constructed controls, include some biological demonstration of engraftment, and include strong mechanistic biomarkers that identify the mechanism of action before a substantial commitment to further development activities is made.
Using radiation pneumonitis as a PoC strategy is just one of a number of potential strategies that could be used. The potential to directly translate between an appropriate preclinical safety and efficacy model in nonhuman primates and humans is one of primary strengths of this proposal. The ability to target and sample the lung epithelium via the airway is another methodologic strength of the approach. A third strength is the possibility that a relatively low-tech approach to progenitor cell isolation and expansion might be effective, shortening the time required to optimize the reagent itself. Despite potential difficulty in generalizing from this PoC model to other regenerative paradigms, if the model proves robust and informative then, as stem cell therapeutics evolve, the model could continue to be used as a first validation step along the pathway to the ultimate goal of correcting genetic disorders and regenerating lung.
ACKNOWLEDGMENTS
The author thanks Steven Albelda, M.D., for suggesting the possibilities of the radiation model, and Susan Majka, Ph.D., for technical advice on mesenchymal progenitor cell isolation and delivery.
FOOTNOTES
The opinions expressed in this article are those of the author and do not represent those of Novartis Institutes for Biomedical Research or Novartis Pharma AG.
Conflict of Interest Statement: D.M.R. is an employee of Novartis Institutes for Biomedical Research.
(Received in original form February 5, 2008; accepted in final form June 25, 2008)
REFERENCES
| ||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
