Geron Presents New Data That Document Progress in Development of Therapeutic Products from Human Embryonic Stem Cells.

MENLO PARK, Calif. -- Geron Corporation (Nasdaq:GERN) today announced the presentation of new data that document the company's progress in the development of first-in-class therapeutic products from human embryonic stem cells (hESCs) for the treatment of degenerative diseases.

Presented by Geron scientists and collaborators at the International Society for Stem Cell Research's (ISSCR) annual meeting in Toronto, the data pertained to four products in development: (1) GRNOPC1, Oligodendroglial Progenitor Cells for acute spinal cord injury, (2) GRNIC1, Islet Clusters for diabetes, (3) GRNCM1, Cardiomyocytes for myocardial infarction and (4) osteoprogenitor cells for bone fractures and osteoporosis. Other presentations at the meeting described advances in hESC culture and derivation techniques and further characterization of hESC-derived hepatocytes for liver failure and drug metabolism testing.

"The progress sets Geron apart in terms of leveraging our human embryonic stem cell platform into the development of multiple therapeutic products," said Thomas B. Okarma, Ph.D., M.D., Geron's president and chief executive officer. "Our most advanced product, GRNOPC1 for acute spinal cord injury, is in multiple IND-enabling studies. Its in vivo mechanism of action includes myelination and trophic support of damaged spinal cord neurons, resulting in significant locomotor recovery of spinal cord injured rodents. As our work progresses with other hESC-derived cell types such as cardiomyocytes, islets, hepatocytes and osteoprogenitor cells, we are hopeful that these cell types will also demonstrate significant, long-term regenerative capacity in animal models, setting the stage for clinical trials in patients."

GRNOPC1 - Oligodendroglial Progenitor Cells for Acute Spinal Cord Injury

Neurotrophic Effects of GRNOPC1: Additional Mechanism of Action In Vivo

Data show that partial repair is observed in the spinal cord of injured rats transplanted with GRNOPC1. Rats transplanted with GRNOPC1 showed evidence of increased remyelination of axons and enhanced axonal survival and growth. GRNOPC1 has been shown to secrete neurotrophic factors which can improve neuronal survival and stimulate axon growth. Moreover, at least one known neurotrophic factor secreted by the cells, TGF-Beta2, has been identified that mediates this effect.

In new studies presented by scientists at Geron and the University of California, Irvine, GRNOPC1 was shown to have a second activity that may also contribute to its observed efficacy. GRNOPC1 stimulates axonal survival and growth both in tissue culture and in the rodent injured spinal cord. When GRNOPC1 was transplanted directly into the injury site of contused rats, treated animals had 2.5 times more corticospinal tract neurons near the lesion site as compared to controls. The data suggest that GRNOPC1 may not only prevent death of injured neurons, but could restimulate growth of axons in the lesion area, potentially allowing new pathways for nerve conduction past the injury.

Significant progress has been made in identifying the factors produced by GRNOPC1 that may be responsible for these effects on axonal survival and growth. Conditioned medium produced by GRNOPC1 cultures was found to contain neurotrophic factors that significantly increased neuronal survival and axonal growth in vitro. Moreover, the axonal growth stimulating activity of GRNOPC1 was characterized at the molecular level and shown to result in part from secretion of TGF-Beta2 protein, a known neurotrophic factor. The trophic effects were partially inhibited by a neutralizing TGF-Beta2 antibody, suggesting that GRNOPC1 secretes other additional neurotrophic factors as well.

"These studies show that GRNOPC1 impacts two of the major defects that occur at the spinal cord injury site," stated Jane S. Lebkowski Ph.D., Geron's senior vice president of regenerative medicine. "In addition to inducing remyelination, GRNOPC1 produces neurotrophic factors that reduce deterioration of axons and stimulate their partial regrowth, potentially enabling development of alternative circuitry through the lesion site."

cGMP Production of GRNOPC1 from Geron's Master Cell Bank

Data show the feasibility of scalable cGMP production of GRNOPC1 from Geron's master cell bank (MCB) of H1 hESCs, as well as the long-term survival of GRNOPC1 made from the MCB after transplantation into the rodent injured spinal cord.

Geron scientists presented data on the production of GRNOPC1 under current good manufacturing practices (cGMP) from its Master Cell Bank (MCB) of pathogen-free H1 hESCs. The MCB was established without the use of feeder layers and utilized serum-free media containing human or recombinant-derived reagents. The MCB was characterized for pathogens from any human or animal source, including the mouse feeder cells upon which the H1 hESC line was originally derived. No evidence of contamination of the MCB by human or animal pathogens was detected. Cells from the MCB were used to produce GRNOPC1. When injected into the lesion site of spinal cord injured rodents, GRNOPC1 produced from the MCB migrated throughout the lesion, survived and maintained their appropriate phenotype in vivo for at least nine months. GRNOPC1 has now been produced from the MCB under cGMP, the standard that will be required for manufacturing of products for human clinical testing. At the current scale of production, up to 2500 doses of GRNOPC1 can be produced per run, and further scale-up is quite feasible.

Reduced Immunological Recognition of GRNOPC1

Data show that human immune cells have little direct immunoreactivity with GRNOPC1. These findings suggest that temporary, low-dose immune suppression should enable transplanted GRNOPC1 to survive without immune rejection. When the injury heals, the transplanted cells should be protected by the blood-brain barrier, allowing immunosuppression to be withdrawn.

Geron scientists and collaborators have previously shown that undifferentiated hESCs possess unique properties that qualify them as "immunologically-privileged" -- they are not recognized by human immune cells in vitro (Stem Cells, 2004; 22:-448-456). New data presented by Geron scientists extend these prior observations to GRNOPC1 differentiated from hESCs. Both undifferentiated hESCs and GRNOPC1 express moderate levels of HLA A, B and C antigens (Class I HLA markers) but lack surface expression of HLA DR, DP and DQ (Class II HLA markers). Both HLA Class I and II molecules are the major determinants of acceptance or rejection of grafts into different or "allogeneic" recipients. In in vitro mixed lymphocyte reactions (an assay used to detect immune recognition of GRNOPC1 by human immune cells) immune cells taken from disparate donors showed only weak reactivity to GRNOPC1. In addition, GRNOPC1 was only weakly susceptible to lysis mediated by human NK cells from multiple donors. Moreover, of 10 sera from normal healthy volunteers representing all four blood groups, eight samples showed no antibody reactivity to GRNOPC1 and only two had low, but detectible, cytotoxicity against the cells. These observations were repeated with multiple production runs of GRNOPC1.

"The data also show that the H1 hESCs, which were originally derived using mouse feeders, are not irreversibly contaminated with mouse antigens such as sialic residues that could potentially stimulate human immune rejection of these hESCs or their derivatives," stated Dr. Anish Majumdar, Geron's senior director of cell therapy research and a presenter of the studies.

Background on GRNOPC1

Geron's most advanced hESC-based therapeutic, GRNOPC1, is a population of living oligodendroglial progenitor cells shown to significantly improve locomotor recovery when transplanted into the damaged spinal cords of rodents. The transplanted human cells enabled replacement of axonal myelin sheaths in vivo, repairing the damage associated with acute spinal cord injury (Journal of Neuroscience, 25(19): 4694-4705, 2005). Myelin sheaths function to insulate neurons and promote efficient conduction of electrical impulses. Axons are the portion of the neuron that act as a conduit for impulse conduction.

GRNIC1 - Islet Clusters for Diabetes

A study highlights the procedure that is first to generate fully functional human islet progenitor cells with sufficient purity and activity for rigorous testing in animal models of diabetes.

Dr. Majumdar presented in detail the production, enrichment, and characterization of hESC-derived Islet Clusters (GRNIC1) for use in treating diabetes. The GRNIC1 production procedure generates pancreatic endocrine cell clusters that express pancreatic islet transcription factors as well as the hormones insulin, glucagon and somatostatin in physiological proportions. In the presentation at ISSCR, Dr. Majumdar showed that these clusters have biochemical and physical attributes of islets, budding off as discrete cell aggregates that can be enriched by simple size exclusion. Previous studies have shown that transplantation of GRNIC1 can positively impact survival in rodent models of diabetes and produces measurable levels of human insulin in the animals' blood.

GRNCM1 - Cardiomyocytes for Myocardial Infarction

Studies add to the growing body of evidence that Geron's method to produce ventricular cardiomyocytes (GRNCM1) can be scaled, and yields cells which can be cryopreserved and thawed. They also show that the method generates ventricular cardiomyocytes that display appropriate gene expression patterns and responsiveness to cardiac drugs, as well as engraft and survive in the damaged zone of an infarcted rodent heart, providing the rationale for their use to treat infarct-induced heart failure.

Significant and important improvements to Geron's previously published method of producing hESC-derived cardiomyocytes (GRNCM1) were introduced in two presentations by Dr. Joseph Gold, director of stem cell biology, and other Geron scientists. A new method to scalably produce cardiomyocytes was described that has several advantages over previous methods: (1) It uses serum-free media and a novel combination of defined growth factors, (2) It does not require embryoid body formation or co-culture with other cell types; (3) It directly generates populations of cardiomyocytes at up to 80% purity without further purification and (4) It produces yields of cardiomyocytes sufficient for large animal and eventually human use. These GRNCM1 cells spontaneously contract, express appropriate cardiomyocyte transcription factors, respond normally to cardiac drugs, have normal electrophysiological properties and can be cryopreserved and thawed with high viability. This new production method enables the scalable manufacturing of cardiomyocytes for use in preclinical testing in large animal models of myocardial infarction and heart failure. Moreover, this new method of GRNCM1 production can be used successfully with undifferentiated hESCs maintained without feeder cells or fibroblast conditioned medium, resulting in a production process that can be scaled and performed under cGMP conditions.

When transplanted into the infarct zone of infarcted nude rats, these cardiomyocytes engrafted and survived. All animals had robust human cardiac grafts at four weeks after transplant, surviving in the face of the ischemic and reperfusion injury induced by coronary artery ligation and release.

In another presentation by collaborators from the Imperial College of London, hESC-derived cardiomyocytes were configured into a drug screening assay that measures contractile frequency and amplitude changes in response to various cardioactive drugs. Administration of Isoprenaline in the presence and absence of Beta1 and Beta2 adrenergic receptor antagonists showed that hESC-derived cardiomyocytes are comparable to adult human ventricular myocytes in that Isoprenaline responses can be mediated by both Beta1 and Beta2 adrenergic receptors. Carbachol, a muscarinic agonist, appropriately reduced both the basal and isoprenaline-induced contractile frequency, demonstrating appropriate muscarinic receptor activity as well.

Hepatocytes for In Vitro Drug Screens and the Treatment of Hepatic Failure

A study shows the functionality of hESC-derived hepatocytes that could alter the landscape of pharmaceutical drug discovery by defining early in drug development the toxicity and hepatic metabolism of new drugs before they enter expensive clinical trials. Moreover, these cells may find use for both extra-corporeal liver assist devices to treat acute hepatic failure as well as transplantation to restore hepatic function in vivo.

Geron scientists and collaborators at the Roslin Institute and CXR Biosciences in Scotland presented data on their method to produce functional hESC-derived hepatocytes. The hESC-derived hepatocytes express liver-specific genes including AFP, albumin and various CYP450 isozymes and stain for stored glycogen. These cells are now being scaled and formatted for in vitro ADME/TOX testing as well as transplantation into animal models of liver failure.

hESC-Derived Osteoprogenitor Cells

A report is first to show the in vivo bone-forming activity of hESC-derived osteoprogenitor cells. Scalable sources of functional bone-forming cells could have great impact on the treatment of hip fractures, non-union bone fractures and osteoporosis in the elderly.

Geron collaborators at the Medical School, University of Edinburgh, and the Roslin Institute in Scotland reported on methods to generate osteoprogenitor (bone-forming) cells from both hESCs and bone marrow-derived mesenchymal stem cells (MSCs). They compared the capacity of the two cell types of bone-forming cells to repair a full thickness, critical size calvarial defect generated in rats. Their results show significantly more bone formation in vivo in rats receiving hESC-derived osteoprogenitors compared to those receiving MSC-derived osteoprogenitors. The hESC-derived cells were detected throughout the calvarial defect and were highly integrated into the matrix carrier used to deliver the cells to the lesion.

Geron's Intellectual Property in the hESC Field

Geron, which funded the pioneering work that resulted in the initial derivation of hESCs, continues its leadership in the field, which has resulted in the allowance or issuance of 56 patents and an additional 227 patent filings worldwide. "Geron has developed enabling technology which has general applicability in the development of cell therapies," said David Earp, Ph.D., J.D, Geron's senior vice president of business development and chief patent counsel. "The resulting patents protect Geron's commercial opportunities. They also create a licensing opportunity. We have announced two such transactions in the past two weeks."

Geron is a Menlo Park, Calif.-based biopharmaceutical company that is developing and intends to commercialize first-in-class therapeutic products for the treatment of cancer and degenerative diseases, including spinal cord injury, heart failure, diabetes and HIV/AIDS. The products are based on Geron's core expertise in telomerase and human embryonic stem cells. For more information, visit www.geron.com.

This news release may contain forward-looking statements made pursuant to the "safe harbor" provisions of the Private Securities Litigation Reform Act of 1995. Investors are cautioned that such forward-looking statements in this press release regarding potential applications of Geron's human embryonic stem cell technology constitute forward-looking statements that involve risks and uncertainties, including, without limitation, risks inherent in the development and commercialization of potential products, uncertainty of clinical trial results or regulatory approvals or clearances, need for future capital, dependence upon collaborators and maintenance of our intellectual property rights. Actual results may differ materially from the results anticipated in these forward-looking statements. Additional information on potential factors that could affect our results and other risks and uncertainties are detailed from time to time in Geron's periodic reports, including the quarterly report on Form 10-Q for the quarter ended March 31, 2006.