Stem Cell Therapies Walk A Risky Regulatory Road

As the most advanced stem cell therapies, known collectively these days as "regenerative medicine," edge into the clinic, much about the field remains unresolved - including basic regulatory questions and whether the therapies, which hope to restore circulation to ischemic tissue or function to damaged nerves, for example, will even work in humans.

Regenerative medicine companies can be differentiated based on the therapies they're pursuing, the cells they're using and the manufacturing processes they employ. But one thing they all have in common is the need to prove to regulators working with rapidly evolving science and little experience to guide them that the new products are safe.

After all, stem cell therapies involve putting millions of living cells into someone's body, and most of FDA's regulatory framework is premised on things that are more predictable and where the science is much more solid. Their path to approval goes through CBER's Office of Cellular, Tissue and Gene Therapies, which learned much from its experience with gene therapies that is transferable to regenerative products (though much is still new and unknown).

One basic assumption likely to transfer, for example, is the need for long-term follow-up, which for gene therapies is a 15-year requirement.

"It's cutting edge technology, and frequently the science gets ahead of us," Donna Przepiorka, medical officer in OCTGT's Division of Clinical Evaluation and Pharmacology/Toxicology, said during an April Drug Information Association forum on the scientific, regulatory and ethical considerations surrounding the clinical development of stem cell therapies.

CBER refers to existing regulations, guidances and policies and draws on the medical literature and external experts and standards to inform its oversight of cell therapies, she said. The field is advancing very rapidly, and for that reason, "the advice we give you tomorrow may not be the same as the advice we gave you yesterday," Przepiorka said.

Przepiorka and other OCTGT officers at the DIA meeting, as well as at a similarly oriented session during the Biotechnology Industry Organization annual meeting in May, implored cell therapy developers to contact the agency early and often before the pre-IND stage (as well as for the more formal pre-IND meetings) to talk about CMC and other preclinical issues in the evolving regulatory environment.

CMC Challenges For Stem Cell Therapies

Those consultations appear to be paying off. The agency receives about 50 new IND applications for cell therapies a year, and over the course of time the proportion being put on hold has gone down, Przepiorka said. It is currently running at 4 percent.

Still, well over half of the IND holds are due to questions about or inadequate descriptions of manufacturing, including testing within manufacturing, Donald Fink, OCTGT Division of Cellular and Gene Therapies, observed.

Chemistry, manufacturing and controls remain challenging for stem cell companies looking to commercialize their products. Companies are largely on their own to create and validate assays that best characterize their products and demonstrate consistency, potency and purity of the batches, as they ramp up from research to the clinic and on to potential commercialization.

The impurity profile is important too. "You may have other cell types there that may or may not contribute to the sustainable activity of the target," Fink said. "I've never seen a 100 percent single purity population of cells, so you don't have to worry that that's what we expect, but we do expect that whatever those profiles are, they remain consistent lot to lot."

"Do your own qualification of reagents through your own testing," Fink advised. "Build and implement a qualification program early."

Lions And Tigers And Bears: Pitfalls Loom

Indeed, Przepiorka's regulatory wish list includes technology to enable validated assays for purity, potency and identity for testing of stem cell products and to assess the biodistribution and fate of the product in patients in vivo. Also on the list are biologically relevant animal models that provide useful information about the safety of the human product. Each of these items is now evaluated case by case as products come to the attention of the agency.

Another challenge, says Przepiorka, is that "we are frequently uncertain of all the things the cells can do in the body, simply because the science is not there, and, therefore, it may be difficult even to identify a potency assay."

One particularly troubling possibility is that the cells may "stick around" for a very long time, or they might migrate to an unintended place in the body or differentiate into something "we don't want, including tumors," Przepiorka said. Also, there may be immunologic reactions, especially with allogenic cells (derived from donors other than the patient). Conversely, the therapeutic cells could enter the body and die fruitlessly.

For those reasons and more, sponsors need to start early to "get a handle on what your product does and how it behaves," Fink stressed. Sponsors must be ready to characterize the products at all stages of development when a BLA is the goal.

Last October, Fibrocell Science had a setback on that score when OCTGT advisors failed to support the BLA for its injectable cultured autologous (self-donated) fibroblasts for treating facial wrinkles. The product had been marketed safely for four years before FDA determined that all somatic cell therapies should go through the approval process, and it met all the endpoints in two subsequent Phase III trials with strong p-values.

But despite the product's demonstrated efficacy, an FDA advisory committee was troubled about the absence of clinical or preclinical information on what happens to the cell product once it is injected (¹ (Also see "Fibrocell's Wrinkle Drug Underlines Challenges Of Seeking Aesthetic Claims" - Pink Sheet, 19 Oct, 2009.)). "The lack of data on cell fate led the committee to seriously question whether safety had been established for this product," Przepiorka said.

On the other hand, it's so far so good for Aldagen, which has three cell therapies in late stage trials, two autologous angiogenic studies in ischemia and an allogenic therapy for accelerating engraftment in pediatric cord blood transplants. At the DIA forum, Andrew Balber, Aldagen's chief science officer, advised sponsors to "do your science and bring everything to bear on your argument to the agency to get your IND."

The agency will want to see dose responses even though they are rarely seen in the clinic, he advised. Sponsors also must demonstrate phase-appropriate risk control for reagents selected as part of CMC and manufacturing, he said. In addition, Aldagen had to convince regulators of the rationale behind its platform, which selects for cells that express a high level of the enzyme aldehyde dehydrogenase, so-called "brite cells," which the firm believes have a better chance of engraftment, he said, stressing that mechanistic trials "may not get you approved, but they help you design your product."

Animals - Bad Models, But How Else To Predict?

Because of the dangers presented by stem cells' potential to migrate within the body, their likely tumorigenicity and their ability to differentiate into other tissue types, researchers don't have the opportunity to go into healthy humans with a Phase I study, so those questions must be addressed as much as possible through animal experimentation. At the end of the day, however, the differences between an immune-compromised rodent and a human are too great to extrapolate the results to human experience of the cell product, stakeholders at the DIA forum agreed.

Even with the uncertainty, however, comparability between analogous human and animal cells is an important step toward understanding the safety of the proposed therapy, Wei Liang, from FDA's OCTGT/DCGT, argued.

Preclinical studies are needed to predict human clinical trials. "If you do nothing in animals, you will have no predictions," she said, adding that animal models are good for testing some endpoints, like tumorigenicity.

For a first-in-man trial, regulators are going to look for a great deal of supportive animal information, particularly on dose escalation, David Maybee, DCGT medical officer, said. Animal studies aside, those first in human trials also should have stopping rules along the lines of studying only a few patients and then moving on based on those results, he said.

Indeed, it was rat data that supported the potential efficacy of Geron's lead compound OPC1 in humans, but also led to two clinical holds on the product (see sidebar; " ² Geron Aims For Fall Return To The Clinic ").

Knowing Your Stem Cells' Pedigree

Another regulatory issue unique to developing regenerative therapies is being able to trace the derivation of the stem cell line(s) involved back far enough to ensure safety.

The prevailing perception is that companies working with stem cells should follow the new NIH guidelines to ensure the provenance of their cells is acceptable to FDA, but that isn't necessarily so.

While the sources of stem cells have been a point of great ethical debate, "I'm not so sure" that a NIH safe harbor is necessary, Robert (Skip) Nelson, an ethicist in CBER's Office of Pediatric Therapeutics, said at the DIA event. "I suspect the FDA would be agnostic about the source of the stem cells, barring any obvious ethical indiscretions pertaining to the source of those cells," he said.

What companies developing allogenic therapies do have to comply with is the existing regulations regarding tissue screening, Raj Puri, DCGT director, said during the panel on commercializing regenerative therapies at the recent BIO meeting in Chicago.

In fact, NIH guidelines don't include donor eligibility, the screening/testing of donors for relevant communicable disease, which has been a requirement under FDA tissue regulations since 2005, CBER's Fink pointed out at the DIA event. "Bear that in mind," he warned. "You're going to need donor eligibility information even in embryonic stem cells."

Sponsors need to "be able to trace through from anonymization of your donor material, if that is necessary, all the way through the manufacturing process, up until the product is given to a patient, so that if something goes wrong, you can trace your way back through the process," Fink said.

Hoping To Bypass Some Of That

Two regenerative medicine companies say they have limited their risk of coming up against such issues as cell migration and immunogenicity: Pluristem Therapeutics and Garnet BioTherapeutics. Both are working with mesenchymal stromal cells (MSCs), which are attracted to an injury site by inflammation and once there secrete a cascade of cytokines that induces the repair process. Each firm also feels that it has solved the fundamental manufacturing problem of generating enough stem cells fast enough and under the proper standards to be commercially effective.

Pluristem is winding up two Phase I safety studies of its cell therapy PLX-PAD in end-stage limb ischemia patients who have exhausted their options and are about to lose a limb to amputation. As it regularly does with these early clinical forays, FDA limited the U.S. study to 12 patients. Patients in that study were injected with about an ounce of cell product via 30 injections in the affected limb, William Prather, Pluristem's senior VP-corporate development, pointed out during an interview.

The data available so far show not only safety but an increase in blood flow, Prather said. Roughly 10 of the 12 people who have reached their three-month follow-up have efficacy. "They can walk," he said.

The adult cells are derived from placentas, a renewable source that might ordinarily be thrown away, Prather explained. They are then expanded, or grown, on a series of polystyrene discs stacked in a totally self-contained bioreactor with only their own growth factors to encourage growth. In that way the company introduces no substances that have to be tested and accounted for to regulators. The system is extremely cost-effective and produces mass quantities of pure cells, Prather said.

The company has rodent data using Luciferase-stained cells (luminescence tagging) that show the locally injected cells migrate to the site of injury and disappear in about three weeks, he said. "The regulatory bodies have come a long way, even in a year," he said. "They are comfortable now realizing our cells go away; they're not going to be available for tumorigenesis."

Garnet has developed a proprietary manufacturing process that renders its MSCs incapable of differentiating into any cell type except the one needed for the firm's lead product, tropic support cells that aid healing of surgical wounds with less scarring, CEO Gerri Henwood said in an interview.

For Garnet, the process ensures regulators that the cells won't "morph into bone cells that show up in your liver," Henwood said. In addition, the cells disappear in 28 days. "They go in, they do their job, and they're gone." The last thing a reviewer wants to learn 10 years from now is that this cell has been identified in a tumor line, she said. "We've removed that possibility." To prove its cells don't roam, Garnet injected one model species' cells into another, and they didn't migrate more than a centimeter from where they were injected, Henwood said.

Garnet also has a proprietary manufacturing process. The firm has qualified three donors for its bone-marrow-derived cells that allow production of "many, many, many millions of doses" from one specimen, Henwood said. The process allows more than 40 doublings, above the upper limit at which cell lines typically start to show senescence, and end-of-process testing shows all the cells are genetically and "performance-wise" the same as earlier expansions, she said.

Garnet has permission from FDA to enroll five women undergoing breast reconstruction surgery in a Phase II study to see whether its product can create healing with less scarring. The target is the hip-to-hip abdominal incision surgeons make so they can reach up under and do the reconstruction without scarring the skin in the breast area. The patients are their own control, with six intradermal injections of product on one end of the incision and six of a vehicle on the other. For those five women, a total of 60 million MSCs will be needed.

In the end, "the thing is to try and make discrete the regulatory issues, so it is easier to assess risk, and, therefore, easier to demonstrate safety that could be acceptable for regulatory action," Henwood said.

The Information Consistency Challenge

When it comes to regenerative medicine, "it's hard even to compare published research studies because there is no consensus around some of the terms or scientific concepts," says Michael Werner, a partner in the Washington, D.C., law office of Holland & Knight and co-founder of the Alliance for Regenerative Medicine.

ARM was formed last September as a lobbying group to present a unified voice for the industry on Capitol Hill and among federal agencies, as well as to serve an educational function (³ (Also see "Regenerative Medicine Sector ARMs Itself With New Advocacy Group" - Pink Sheet, 15 Mar, 2010.)).

"It's about having an organization that can be there to support the industry as it's going though this critical phase in its evolution," Werner said in an interview. One item on ARM's agenda is to create processes and venues where industry and academic researchers can engage with FDA in a pre-competitive way to discuss issues such as the state of the science or creating standards for safety. ARM is a "way of making the path to commercialization easier on each individual company," Werner said.

FDA is also working on that goal. Beyond a list of ongoing working groups and one-off workshops, the agency is set to host a new series of workshops with NIH, CBER's Puri said.

The Ethics Of Trials For Products That May Not Be Picked Up

Both industry and academic researchers worry about the hype/hope proposition that brings patients to their doors. And the investigators don't take lightly the act of enrolling a patient with a degenerative disorder, such as amyotrophic lateral sclerosis (Lou Gehrig's disease), or one who is destined to live a very long time with an accident-induced paralysis, in a clinical trial that involves an invasive technology delivering cells directly to the spinal cord.

"We have a responsibility," cautioned Jonathan Glass, director of Emory University's ALS Center. "No matter how bad it is, we can make it worse." Glass is primary investigator for a site on a Phase I trial sponsored by Neuralstem that involves injecting processed embryonic neural progenitor stem cells into the spinal cord in a highly invasive procedure, with the goal of getting replacement neurons to grow and restore the nerve/muscle connection.

Neuralstem announced May 24 that the trial's safety monitoring board had reviewed safety data on the first cohort of three patients and approved moving on to a second three-patient group.

Key among concerns expressed by Glass and Jeff Rothstein, a professor of neurology at Johns Hopkins, is whether these dicey human experiments will actually lead to products. Academic researchers don't usually trouble themselves with whether the start-up company sponsoring the research has investors or the hope of a partnership that will enable scale-up to Phase III testing and commercialization.

"Most small molecules move on," said Rothstein. But, "there is no indication this field has any legs in neurologic disease, with real companies that really are investing that can take it into a Phase III trial."

Others have wondered, given CBER's request for 15 years of follow-up on gene therapy trials, who would be responsible for following stem cell therapy patients for such a long period if the company that sponsored the therapy goes broke.

Those fears are not without foundation. Regenerative medicine firms are scrambling to find business models. And while the economic crisis may be easing off for other fields, this one is more risky than most, with companies well into clinical trials still looking for pharma partners ("Regenerative Medicine Faces Funding Crisis," ⁴ (Also see "Despite Advances, Regenerative Medicine Faces Funding Crisis" - Scrip, 1 Feb, 2010.)).

- Shirley Haley ( ⁵ [email protected] )