Where Commercial Stem Cell Clinical Trials Have Been, and Where They're Going: A Quantitative AnalysisSeptember 7, 2013
This article was revised (and retitled) on Sept. 8, 2013 to include two additional trials pointed out to us by sharp-eyed readers, including Irv Arons. Thank you! Please report errors or omissions to trials(at)BusaConsultingLLC(dot)com
Part I: Introduction
The year 1998 witnessed two important American business landmarks: Google was incorporated, and Osiris Therapeutics launched the world's first industry-sponsored stem cell clinical trial. Over the intervening 15 years Google has reported gross revenues totaling $200 billion, while ethical stem cell therapeutics have generated revenues that -- on this scale, at least -- closely approximate zero.
This is, of course, an unfair comparison if ever there was one. Internet search engines are not government regulated, whereas biologic drugs are. No one dies or is horribly injured if a search engine isn't effective or acts in unanticipated ways, but the same can't be said for a drug. Nonetheless, this comparison serves to illustrate a key point that technology investors know only too well: the payoff time for an investment in stem cell drugs is painfully long (though we still don't really know how long), due in large measure to the lengthy clinical trial process they face on their road to market.
Nonetheless, and despite both chronic capital insufficiency and more than its fair share of controversy, the stem cell therapeutics industry continues to grow explosively as its players labor to deliver blockbuster therapies to the clinic. Systematically taking stock of their efforts raises a host of important questions for investors, entrepreneurs, granting agencies, lawmakers, regulators, and scientists alike. Where have we been? Where are we headed? Which combinations of stem cell types and disease indications are achieving traction in clinical trials? Which aren’t? Which combinations have been studied to death, and which are under-served by clinical trials today? What are the odds that a company entering a Phase 1 trial today can see its candidate all the way through a successful Phase 3 trial? How much time does a company need in order to do so? What are the newest trends, and disappointments? The fates of aspiring stem cell therapy companies -- and thus the fate of the industry and, ultimately, the technology itself -- hinge on correctly answering these and similar questions.
Up to now there has been no way (that we know of, anyway) to answer these questions with anything like quantitative rigor. But last March, SCSI: The Stem Cell Stock Index introduced our automated stem cell Clinical Trials Monitor, employing computerized text mining and other technologies to track, in real time, the progress of stem cell trials sponsored by the 16 SCSI component companies. We’ve been pleasantly surprised by readers’ response to the Monitor, so we are now in the process of expanding its coverage to include, as nearly as possible, every commercially sponsored stem cell clinical trial (with a few exceptions noted below). We're bringing to bear on the problem of identifying all the world's industry-sponsored stem cell trials a host of proprietary tools and processes first developed for the Monitor and now being enhanced and refined, from natural-language processing and text mining, to corporate databases, to assemblies of expert knowledge.
This article is the first in a series presenting our initial high-level analysis of trends and statistics gleaned from our application of these new tools. Stay tuned for additional installments in this series, which should be published here approximately weekly. We expect to update this series annually.
Our analysis relies exclusively on the U.S. National Institutes of Health's ClinicalTrials.gov (CTG) registry for its raw data. The quality, breadth, depth, and organization of CTG’s data greatly exceeds that of any other national or international registry, thus uniquely supporting sophisticated analysis. We believe that the comprehensiveness of our trial coverage does not greatly suffer from this restriction, since the commercial significance of the U.S. healthcare market drives the vast majority of serious commercial trial sponsors to voluntarily register their ex-U.S. trials with CTG. Fully 50% of actively recruiting clinical trials registered with CTG are ex-U.S., and thus not FDA regulated. If anything, this analysis' reliance upon CTG as its sole data source probably serves an important filter function, automatically excluding the majority of the all too common fly-by-night stem cell marketing schemes that masquerade as 'clinical trials' and are commonly registered only overseas.
We employed a multi-faceted iterative strategy to comprehensively identify CTG-registered, industry-sponsored stem cell clinical trials. Companies in this space were identified via Busa Consulting's proprietary Stem Cell Industry Database of over 200 stem cell-focused corporations worldwide, which is continuously updated via both manual curation and text-mining of press releases, job ads, and other industry-related publications. Additionally, expert knowledge was applied to include the names of companies not present in the Stem Cell Industry Database because they are not stem cell-focused companies, but which are known to be current or past stem cell trial sponsors (examples include Baxter Healthcare, Pfizer, Novartis, and GlaxoSmithKline, among others). Along with this company-focused search strategy, a parallel strategy employed search terms from a rich list of keywords and phrases conceptually associated with stem cell therapeutics. The search results yielded by these and additional strategies were then text-mined to tentatively classify each record as either true-positive or false-positive. False-positives included, for example, trials involving conventional unmanipulated bone marrow or cord blood transplantation subsequent to myeloablative conditioning. The results of this true/false classification were manually checked for accuracy by domain-knowledgeable curators.
All relevant CTG records identified by the above process were downloaded in XML-tagged format, automatically parsed and databased, and the resulting database was manually cleaned and annotated by domain-knowledgeable curators. One important aspect of this curation involved the deletion of trials sponsored by companies which, in our editorial judgement, fail a reasonable seriousness of purpose test (most often due to either their commercial activities marketing unproven and unregulated stem cell 'therapies,' their insignificant R&D expenditures, or both). Companies excluded from this analysis on seriousness-of-purpose or other grounds include Ageless Regenerative Institute, Adistem, Alliancells (a subsidiary of Zhongyuan Union Stem Cell Bioengineering Co., Ltd.), Bioheart, Cytopeutics, International Stemcell Services, MD Stem Cells, Retinal Associates of South Florida, RNL Bio (now K-StemCell), Shenzhen Beike Bio-Technology, Thermogenesis and its partner, TotipotentSC (which we have discussed at length in two previous Organelles essays), TheraVitae, and Translational Biosciences.
Additional trials were excluded from consideration if their only relationship to stem cells involved now-conventional unmanipulated bone marrow or cord blood transplants. Such transplants were, however, included in this analysis if they involved significantly manipulated cells (selected, expanded, drug-treated, or genetically engineered).
For completeness, this analysis also includes trials of small-molecule, protein-based, or DNA-based drugs intended to modulate the therapeutic performance of endogenous or exogenous stem cells (this includes a total of 3 trials; to be discussed in a subsequent installment). Also included are stem cells used extracorporeally to treat patients' blood or tissue (3 trials; to be discussed in a subsequent installment). In a few instances (5 trials) our analysis also includes trials sponsored exclusively by hospitals, medical centers, or universities, (i.e., not explicitly sponsored by a corporation) if sufficient information is available in the trial's record to unambiguously attribute the cells employed to a specific company's clinical development program. For convenience, we treat such trials as being 'sponsored' by these companies.
1. Basic Statistics Describing the Trials in Our Database
For CTG records dated through the end of August, 2013, our analysis identified a total of 196 industry-sponsored or collaborated CTG-registered stem cell clinical trials that meet our inclusion criteria, dating back to Osiris Therapeutics' original study launched in 1998, sponsored by a total of 64 companies. Some of those companies have been acquired by others and now function as subsidiaries, or else are divisions of larger corporations. Examples include:
- Angioblast Systems (acquired in 2010 by Mesoblast Ltd.)
- Harvest Technologies (acquired by Terumo in 2011)
- Samsung Medical Center (a subsidiary of Samsung Group)
- Max Neuroscience Institute (a subsidiary of Max India)
- Aldagen (acquired by Cytomedix in 2012)
- Cordis Corp., Centocor Inc., and Janssen Research and Development LLC, all subsidiaries of Johnson & Johnson
Our analysis thus includes a total of 60 parent companies. Just over half (55%) are headquartered in the U.S., followed by Korea (13%) and Israel (7%). The remaining countries included in our data set (Germany, Japan, India, Belgium, Switzerland, Great Britain, Spain, China and Australia) each are home to less than 4% (1 or 2) of the remaining companies. The fraction of trials in our database expressed as a function of their parent companies' headquarter countries (Figure 1) largely parallels this same distribution, with the notable exception of Australia, which is home to only 1 (1.7%) of the companies included in our analysis (Mesoblast) but accounts for 7.7% of all trials.
Figure 1: Percentage of trials by parent companies' headquarter countries. Abbreviations: AU, Australia; BE, Belgium; CH, Switzerland; CN, China; DE, Germany; ES, Spain; GB, Great Britain; IL, Israel; IN, India; JP, Japan; KR, Republic of Korea; US, United States
The sheer magnitude of the contribution of U.S.-based companies apparent in Fig. 1 is, of course, due in part to our reliance upon a U.S.-centric data source (CTG), but America's dominance is likely real, not merely an artifact. Our proprietary Stem Cell Industry Database reflects a similarly predominant contribution by U.S.-based companies to the larger ethical stem cell industry (including companies that are not clinically focused), despite the fact that the assembly of that database relies upon a much more international collection of data sources.
One hundred and twenty-three (63%) of the trials included in this study are under the jurisdiction of the U.S. Food & Drug Administration (FDA) by virtue of having one or more trial sites located in the U.S.; the remaining 73 (37%) are conducted exclusively ex-U.S. and are thus not FDA-regulated.
A clinical trial's current status is recorded by CTG as either Not yet recruiting (registered but still being organized, and thus not yet treating patients), Recruiting (treatment of patients is underway), Active, not recruiting (the trial's full complement of patients have been recruited and treated, but follow-up observations are still in progress), Completed (the trial has reached its designed end), Suspended (recruitment and treatment are temporarily halted due to safety, business, or other considerations) Withdrawn (a registered trial is abandoned by the sponsor prior to its initiation), or Terminated (the trial is abandoned by its sponsor subsequent to its initiation for safety, business, or other reasons). Companies' efforts to update their trials' statuses in CTG in a timely manner are frequently deficient (this is particularly true for ex-U.S. trials voluntarily registered with CTG), so this parameter is, at best, a lagging indicator. Figure 2 illustrates the reported most current status distribution of the 196 trials in our analysis. Two-thirds (66%) of all industry-sponsored stem cell trials ever registered with CTG are currently in progress (Not yet recruiting, Recruiting, or Active, not recruiting).
Figure 2: Distribution of most recently reported trial statuses.
Clinical testing of a candidate therapeutic normally (but not always) progresses through three somewhat distinct phases, each phase constituting a distinct trial. According to the FDA's definition (as summarized by CTG) so-called Phase 1 trials emphasize the demonstration of the candidate therapeutic's safety; their goal is to determine what the drug's most frequent and serious adverse effects may be. Phase 1 trials are typically too small (in terms of the number of patients enrolled) to demonstrate the efficacy (or lack thereof) of the candidate therapeutic with statistical rigor, but the clinical observations collected during such trials may provide suggestive evidence, and companies frequently rely upon such evidence in deciding whether or not to advance a candidate beyond Phase 1. Phase 2 trials are intended to gather preliminary data on effectiveness (whether the drug works in people who have a certain disease or condition). Safety also continues to be evaluated in Phase 2 trials. Finally, Phase 3 trials gather still more data regarding effectiveness and safety, and are typically large enough to enable a definitive statistical judgement regarding the candidate's efficacy. Typically, only candidate therapeutics that successfully complete Phase 3 trials are qualified to seek government approval for marketing. Sometimes Phases 1 and 2 are combined (so-called Phase 1/2 trials), as Phases 2 and 3 may also be (Phase 2/3 trials).
Figure 3: Distribution of trials by phase, broken down by last reported status.
As Figure 3 illustrates, the majority of trials are Phase 1 or 1/2, and this is true both for all-time trials (60% of all trials) as well as for currently in-progress trials (59% of in-progress trials). Only 9 Phase 2/3 or 3 trials (5% of all trials) have been completed to date (see below).
The number of patients actually enrolled or anticipated to be enrolled in the trials in our database totals 10,894 (an average of 56 patients per trial), but average enrollment varies substantially by trial phase and, to an extent, differs between FDA-regulated non-FDA-regulated trials (although the difference is not statistically significant) (Figure 4). There appears to be a tendency for non-FDA trials to experience their major enrollment increases at Phase 2, whereas FDA-regulated trials defer the major expansion of enrollment until Phase 3.
Figure 4: Average patients per trial by trial phase, for FDA-regulated (blue) and non-FDA-regulated (ex-U.S.; red) trials.
Over all trials (both FDA- and non-FDA) average enrollment by phase is 20 (Phase 1), 32 (Phase 1/2), 75 (Phase 2), 84 (Phase 2/3), and 170 (Phase 3).
The number of all-time trials per sponsoring company ranges from a low of 1 (24 companies, or 40% of all companies) to a high of 15 (1 company; 2%). Seven companies combine to account for more than a third of all trials in the database:
- Mesoblast (including subsidiary Angioblast Systems): 15 (8%)
- Osiris: 14 (7%)
- Aastrom Biosciences Inc.: 10 (5%)
- Cytomedix (via its Aldagen subsidiary): 8 (4%)
- Medipost Co. Ltd.: 7 (4%)
- Stempeutics Research Pvt. Ltd.: 7 (4%)
- TCA Cellular Therapy, LLC: 7 (4%)
Four of these top seven companies are U.S.-based (Osiris, Aastrom, Cytomedix, and TCA Cellular Therapy).
Eight companies in our analysis have completed a total of 9 Phase 2/3 or 3 trials, including Osiris (2 trials) and Aastrom, Anterogen Co., Ltd., Cardio3 BioSciences, Medipost, TiGenix S.A.U. (via its Cellerix subsidiary), Johnson & Johnson (through its Cordis subsidiary), and Pharmicell Co. Ltd. (1 trial each). Three of these are Korean companies (Anterogen, Medipost, and Pharmicell), 3 are U.S.-based (Osiris, Aastrom and Johnson & Johnson), and 2 are Belgian (TiGenix and Cardio3).Tweet