International Stem Cell — Update 16 May 2016

International Stem Cell (ISCO) is an early-stage biotechnology company developing therapeutic, biomedical and cosmeceutical applications for its proprietary form of pluripotent stem cells – human parthenogenetic stem cells (hpSCs). ISC California (the predecessor to ISCO) was created in 2006 and went public in December 2006 through a reverse merger, changing its name to International Stem Cell in January 2007.

With its hpSC technology, ISCO has created 15 stem cell lines to develop different cell types: liver cells, neural cells and three-dimensional eye structures. The company’s technology platform does not require the use of fertilized eggs, embryos or foetal tissue to create stem cells, improving product cost and consistency while reducing the controversy associated with other types of stem cells. Its lead candidate is a set of neural cells currently undergoing Phase I/IIa clinical testing for Parkinson’s disease (PD). The company’s portfolio includes preclinical candidates to address ischemic stroke and spinal cord injuries, metabolic liver disease, and retinal and corneal blindness.

ISCO’s commercial operations leverage its hpSC technology and generate revenues to partially offset R&D spending for therapeutic development. LSC develops and sells skincare products and LCT produces human cell culture products for testing.

We value ISCO based on a risk-adjusted net present value (rNPV) methodology, using a 10% discount rate and 90% probability for the skincare/biomedical businesses and a 12.5% discount rate and 7.5% probability for the PD candidate, due to its early development stage. Our model does not ascribe any value to the rest of ISCO’s therapeutic pipeline as the projects are largely dormant while the company focuses on its lead candidate. Our rNPV value is $27m, or $9.60 per share on an undiluted basis.

The company had $0.5m in cash on the balance sheet at 31 December 2015. This was supplemented with a March 2016 offering that provided $2.5m in cash and extinguished $3.8m in debt ($6.3m total proceeds). Its cash burn is just over $0.5m/month and likely to increase as clinical trials progress. Minority investors should note that under the current capital structure, there are c 19m potential common shares from convertible preferred stock, options and warrants that are outstanding, on top of the 2.8m common shares outstanding, which potentially creates sizeable dilution. In addition, the company’s convertible preferred shares are subject to anti-dilution protection, creating further dilution under an array of circumstances. We project financing needs as illustrative long-term debt of $76m by 2023, with an estimated $13.5m required by 2017.

As with any early-stage biotechnology company, ISCO faces numerous risks and uncertainties, especially preclinical risk. The investment case rests largely on the successful execution of the PD cell therapy clinical trials and the company’s ability to attract a licensing/development partner to continue clinical development and, if approved, commercialization. The company’s therapy for PD is in Phase I/IIa human trials and, due to its early-stage nature, has not been shown definitively to be efficacious. However, preclinical data showed improvements in both dopamine production and stable Parkinson’s factor scores (ParkScore) in a primate study. Another risk stems from the capital structure, which, as mentioned, potentially creates sizeable dilution risk for minority investors.

International Stem Cell (ISCO) is an early-stage biotechnology company developing therapeutic, biomedical and skincare applications for its proprietary form of pluripotent stem cells – human parthenogenetic stem cells (hpSCs). With its hpSC technology (see below), ISCO has created 15 stem cell lines, each of which is a different HLA type. From this, it creates different cell types such as liver cells, neural cells and three-dimensional eye structures. Its lead candidate is a set of neural cells currently undergoing Phase I/IIa clinical testing for Parkinson’s disease.

According to the Michael J. Fox Foundation, stem cell research has the potential to have a significant impact on the development of disease-modifying treatments for PD and considerable progress has been made in creating dopamine-producing cells from stem cells.

There are two naturally occurring types of stem cells, each with their own positives and negatives:

A key focus of stem cell research is the development of pluripotent stem cells – stem cells that can be coaxed to become virtually any type of cell in an organism. There are two key types of engineered pluripotent stem cells.

Induced pluripotent stem cells (iPSCs) are engineered from older, fully specialized cells such as skin cells. Scientists encourage these limited cells to act like embryonic stem cells again, with the ability to become any type of human cell. This is a complex technique that has only recently been developed and is the subject of much ongoing research.

Parthenogenetic stem cells are created from unfertilized human eggs (oocytes) activated by chemical or physical means (parthenogenesis) to make the cells pluripotent. Activation results in a parthenote from which pluripotent parthenogenetic stem cell lines can be derived and further differentiated into specific cell types.

ISCO’s technology platform is based on hpSCs differentiated using chemical means. There are several techniques for turning an oocyte into a parthenote. One example uses a chemical catalyst such as SrCl2, ethanol, Ca2+ ionophore, or ionomycin. This is followed by another chemical, for instance 6-DMAP (a broad protein synthesis inhibitor) or cytochalasin B or D (inhibitors of actin filaments polymerization), which blocks second polar body (PB2) extrusion. Thus, the resulting parthenote is a “pseudodiploid” heterozygous embryo containing the two sister chromatids of each maternal chromosome present in the MII oocyte.

Basic parthenogenetic cells are expanded, characterized and cryopreserved into a master cell bank under cGMP conditions. These cells are then chemically directed to differentiate into a pure population of neural stem cells (ISC-hpNSC) under feeder-free conditions. The differentiated cells are then grown in an incubated environment, characterized for the presence of neural markers and the lack of pluripotent markers, and tested for microbial and viral contaminants before being used.

Currently ISCO owns the largest published collection of human parthenogenetic stem cell lines (hpSCs) with 15 lines total; each line can be differentiated into multiple cell types. According to the article Parthenogenesis and Human Assisted Reproduction (Bos-Mikich et al, Stem Cells International 2016): “the therapeutic use of pluripotent stem cells, such as human embryonic stem cells (hESCs) or phESCs, is still in its infancy. The wider application of hESCs is limited due to their genetic background, which will most likely be divergent from a potential patient. phESCs may overcome this limitation presented by hESCs in autologous histocompatible transplantations, as they should be isogenic with the gamete donor.”

Parkinson’s disease is a progressive, irreversible neurodegenerative disorder. It arises from a lack of dopamine in the brain, owing to the death/damage of dopamine-generating cells in the substantia nigra located deep in the mid-brain just above where the spinal cord connects to the brain. The cause of cell death is currently unknown.

According to the PDF, as many as 7-10 million people are affected by PD worldwide. The American Parkinson Disease Association estimates there are around 1.5 million Americans with PD; the PDF estimates there are nearly one million US patients, with 60,000 diagnosed a year. The European Parkinson’s Disease Association estimates there are around 1.2 million patients affected in Europe. Prevalence increases with age, with the typical age of onset more than 50 years old.

PD patients are evaluated using the Unified Parkinson Disease Rating Scale (UPDRS), which consists of three parts: mentation, behavior, mood; activities of daily living; and motor symptoms. Of a total possible score of 199 points (0 = no disability, 199 = worst/total disability), motor symptoms can account for just over half (108 points at worst) of the patient rating. When patients begin to take L-DOPA, motor improvements of 25% (off-patient baseline so -7 points on an initial baseline of 28) have been cited. However, long-term studies suggest that after four or five years patients see modest improvement (6 points on the UPDRS scale) on L-DOPA compared to baseline.

There are a number of different classes of therapeutic molecules available for PD; however, these therapies only address the symptoms and do not slow or halt progress of the disease. The main families of drugs useful for treating motor symptoms are levodopa, dopamine agonists and MAO-B inhibitors.

Levodopa is one of the most commonly used treatments for PD symptoms. However, after five to 10 years of treatment, between 50-70% of PD patients develop levodopa-induced dyskinesia (LID), which is characterized by involuntary random and jerky movements similar to those motor symptoms for which the levodopa was originally prescribed. Current management includes adjusting levodopa dosing and/or adding a dopamine receptor D2 agonist to the regimen to spare levodopa. Amantadine (Symmetrel by Endo Pharmaceuticals), an NMDA receptor antagonist, is the only orally active drug on the market that can be prescribed without special monitoring to improve PD-LID, although it can induce side effects including dizziness.

When medications are not enough to control symptoms, surgical techniques such as deep brain stimulation (DBS) can relieve the associated movement disorders; however, patients undergoing DBS frequently develop side effects such as short-term memory loss.

Much of the advanced small molecule pipeline for PD focuses on enhancements to existing therapies such as ways to provide more consistent delivery of levodopa. However, the clinical challenge may be that the mechanisms of L-DOPA in the brain are more complicated than originally believed. For example, L-DOPA is released on an intermittent or as needed basis in the brain, not at a consistent level. In addition, the oral dosage of L-DOPA needed to achieve therapeutic benefits is considerably higher than would be needed under more direct delivery methods, which may contribute to long-term L-DOPA resistance. However, there are several drug candidates in the pipeline for L-DOPA-induced dyskinesia, which, if effective over the long term, may reduce or delay the need for more expensive and invasive therapies.

There are several cell therapy and gene therapy treatments looking to address the cause of Parkinson’s. Cell therapies introduce new neural cells to supplement or replace cells damaged or destroyed by the disease. According to BioCentury, there are currently six cell therapy candidates for PD in the pipeline from discovery to Phase II, of which the most advanced is Living Cell Technologies’ NeutrophinCell (NTCELL), an implantable choroid plexus cell product that contains specialized brain cells, which produce and secrete neurotrophins and cerebrospinal fluid (CSF). In a Phase IIa clinical study, NTCELL was injected in four patients under guidance by neuroimaging into the affected area of the brain. NTCELL decreased UPDRS by an average of 16 points after 58 weeks, representing a three- to four-year reversal of neurological deterioration.

Gene therapy seeks to repair damaged neurons or coax the body into creating new neural cells, as opposed to adding new neural stem cells as in cell therapy. BioCentury lists seven gene therapy candidates in development for PD. Furthest along in gene therapy research is Voyager Therapeutics’ VY-AADC01, for advanced PD with human POC expected in H216. VY-AADC01 is an AAV serotype 2 vector encoding dopa decarboxylase (DDC; AADC) that is delivered to the posterior putamen using image-guided, convection-enhanced delivery (CED) to treat PD. VY-AADC01 is in an actively recruiting, open-label Phase Ib trial.

ISCO presented results of its PD preclinical studies in October 2015 at Neuroscience 2015 in Chicago. The preclinical studies on 18 non-human primates showed that at 12 months, the transplanted cells had integrated into the dopamine fibres and dopamine levels post-mortem were significantly higher in the transplanted group versus the control group.

On 7 March, ISCO began enrolment in its Phase I clinical study for PD. The Phase I clinical study is a dose escalation safety and preliminary efficacy study of ISC-hpNSC, intracranially transplanted into patients. The procedure is invasive. Doctors create an opening in the skull and deliver the cells into the deep brain via an injection over a four- to five-hour period. As with other studies that involve cell transplantation into the brain, patients may suffer adverse effects such as wound infection and ischemia at the transplantation site among others.

The open-label, single-center, uncontrolled clinical trial will evaluate three different dose regimens of 30-70m neural cells. A total of 12 participants with moderate to severe Parkinson’s disease will be treated. Patients must be between 30 and 70 years of age. Patients will be monitored to evaluate the safety and biologic activity of ISC-hpNSC for a year. A PET scan will be performed at baseline, as part of the screening assessment, and at six and 12 months after surgical intervention. Clinical responses compared to baseline will be evaluated using various neurological assessments such as UPDRS, Hoehn and Yahr and other rating scales. The study will be performed at Royal Melbourne Hospital in Australia. If the pilot proves successful, larger Phase II trials will start in 2017 or 2018 in Australia, Europe and the US. As ISCO operates a fully licensed tissue bank with GMP, it will not require additional certification to produce cells for these additional trials.

We do not anticipate sales from ISCO’s PD therapy until 2024, with the company identifying a licensing partner sometime after Phase II trials. Our model includes a 50/50 share of R&D costs for the PD product between ISCO and its licensee, milestone payments of $10m in 2021, $15m in 2022 and $30m on US FDA approval (anticipated in 2024) and a 12% royalty on sales to ISCO. Our model calls for the licensee to assume or reimburse ISCO for PD production costs, which, according to ISCO’s management, are currently in the range of $1,000-2,000 per treatment. We also anticipate that the licensee will assume all marketing and distribution costs.

We assume a treatment price of $20,000 (excluding surgical costs) in the US and $15,000 outside the US. We look for first year sales of $16.8m in 2024 based on 993 patients treated worldwide, at an average price of c $17,000, resulting in royalties of just over $2m to ISCO. We forecast peak sales in 2032 of $2.8bn, based on treatment of 176,927 patients (approximately 1.5% of the PD population) at an average price of c $16,000 resulting in royalties to ISCO of $334m. Our peak sales number for the US in 2032 equates to approximately nine surgeries pa for each of the c 4,900 neurosurgeons in the US. According to the World Health Organization (WHO), in 2004 there were 33,193 neurosurgeons in 103 countries surveyed. On our 2032 peak procedure number of 176,927, this equates to 5.3 procedures per surgeon pa. We note that ISCO’s treatment is invasive, requiring surgeons to create an opening in the skull to administer the cell therapy directly into the brain. The invasiveness may limit market acceptance if there are less invasive options available.

Lifeline Skin Care (LSC) develops, manufactures and markets a line of luxury skincare products sold in the US and internationally through a branded website, professional channels (including dermatologists, plastic surgeons, medical, day and resort spas) and a network of distributors including Amazon and DermStore.

Products sold include cleansers, exfoliators and a range of specialized moisturizers and serums. LSC is a very small player in the $111bn global skincare market (source: Euromonitor), but its focus on Asia-Pacific and premium-priced skincare products has enabled it to grow faster than the 3.3% CAGR for the overall market. The skincare division grew revenues at a CAGR of 17% in 2012-15, reporting revenues of $3.5m in 2015, flat vs 2014, and operating margins of 11.9%.

Lifeline Cell Technology (LCT) develops, manufactures and commercializes over 130 human cell culture products, including frozen human “primary” cells and the reagents (called media) needed to grow, maintain and differentiate the cells. Cell types include endothelial, epithelial, fibroblasts, melanocytes, stem and smooth muscle among others. Products are sold through the company’s website. According to a 2015 survey by Lab Manager, sales of life sciences reagents/chemicals have been the fastest growing part of the laboratory products industry, growing an estimated 6.6% in 2014. Over 2012-15, revenue from ISCO’s biomedical business grew at a CAGR of 19%, or nearly 3x the industry rate. ISCO’s biomedical division had sales slightly above $4m in 2015, up 15% from 2014. Operating margins were just over 31% in this business in 2015, up from 21.7% in 2014 and 12.4% in 2013. We assume the margin expansion is due to efficiencies of scale.

Development risk: the company’s therapeutic products are still in Phase I and patient recruitment has just begun. In addition, the science behind stem cell therapies is in its infancy, with researchers studying not only how to produce cells reliably, but also how stem cells may work in various therapies.

Financing and partnership risk: ISCO will continue to need new capital to conduct operations and develop its therapeutic products for the foreseeable future. Its cash burn in 2015 was approximately $0.53m/month; however, the commercial operations offset the cash burn by approximately $0.13m/month. ISCO will seek to out-license its cell therapy treatment for PD following the completion of Phase I/IIa studies in the 2018 time frame, although we believe potential partners will seek additional confirmation before licensing the product. The timing of any partnership agreement, along with related milestones, will affect our valuation model.

Dilution risk: to date, ISCO has relied primarily on a combination of convertible preferred shares, warrants and options to fund its growth. While management has not converted or sold its sizeable holdings (and says that it does not intend to), investors need to consider the possibility of significant dilution risk at some point in the future. Currently, there are 19m potential common shares from convertible preferred stock, options and warrants that are outstanding, on top of the 2.8m common shares outstanding. Investors should note that the convertible preferred shares are subject to anti-dilution provisions under certain circumstances, creating further dilution potential.

Management voting control: management controls approximately 70% of the total voting power of the company, which may deter some investors.

Stem cell regulatory risk: stem cell research is controversial and is currently subject to intense scrutiny. Any future legislation that is unfavorable to the development of hESC technology, hpSC technology, or nuclear transfer technology could adversely affect ISCO’s operations.

Our model calls for the skincare business to grow to sales of $4.8m by 2025, while the biomedical business grows to $10.5m over the same period. Sales in both businesses will be driven by overall growth in their respective industries (discussed in more detail in the Financials section), as well as modest market share gains in these multi-billion dollar global business segments. On a combined basis, before corporate G&A, we forecast a profit margin in the high teens to low 20s, peaking at 23% in 2025.

As the first human trials are still ongoing, we have narrowed our primary treatment population to those with newly diagnosed disease (5% by 2030 in the US and lesser amounts in the EU and RoW) and a smaller portion (1-2%) of patients with existing disease. We are forecasting peak sales of $2.8bn in 2032, resulting in peak royalty revenues to ISCO of $334m, in addition to $55m in milestone and upfront payments.

We value ISCO based on a risk-adjusted net present value (rNPV) methodology with a 10% discount rate and assign a 90% probability to the commercial (skincare and biomedical) businesses and a 12.5% discount rate and 7.5% probability to the PD therapy given its early stage of development. Our model does not ascribe any value to the rest of ISCO’s therapeutic pipeline, as the projects are largely dormant while the company focuses on its lead candidate. Our intrinsic value is $27m, or $9.60 per share. Our valuation reflects cash of $0.5m at 31 December.

We examined peer multiples for the skincare and biomedical businesses to estimate how those businesses might be valued on a standalone basis. P/Es on latest earnings for consumer products and cosmetics companies span a wide range from the high teens for Nu Skin Enterprises (18x) to the high 20s/low 30s for Estee Lauder (30x), Revlon (34x) and L’Oréal (26x). In the laboratory supply business, P/Es on latest earnings range from the low double digits for Enzo Biochemical (14x) to the high 20s/low 30s for Agilent Technologies (30x).

Our model calls for revenues in the cosmetic/skincare business to grow from $3.5m in 2015 to $4.8m by 2025, a CAGR of 3.2%, in line with the 3.3% CAGR estimate for the global skincare market from Euromonitor.

According to a 2015 survey by Lab Manager, sales of life sciences reagents/chemicals have been the fastest growing part of the laboratory products industry, growing an estimated 6.6% in 2014. In 2012-15, ISCO’s biomedical business (which consists of cells and reagents/chemicals) grew at a CAGR of 19%, or nearly 3x the industry rate. We look for revenues in this business to grow at a CAGR of 10% through 2025. Operating margins were just over 23% in this business in 2015. On a blended basis, we forecast operating margins (excluding corporate G&A) for the combined skincare/biomedical supply business in the high teens to low 20s, peaking at 23% in 2025.

We expect the company to post a loss of $5.5m in 2016, growing to $15.0m in 2022 as it moves into costly Phase II/III trials. We expect the company to start generating profit from 2024. We project financing needs as long-term debt of $76m by 2023. These financing needs include our assumptions of a 50% R&D cost share with any potential licensing partner and milestones of $10m in 2021, $15m in 2022 and $30m on US FDA approval. However, we note that the company has so far used preferred equity to cover costs. On 20 April, ISCO filed an S-1 registering 8m of the underlying common shares from the March private placement. Under the current capital structure, there are c 19m potential common shares from convertible preferred stock, options and warrants that are outstanding, on top of the 2.8m common shares, creating sizeable dilution potential.