Collplant Holdings — Solid foundations with potential for growth

CollPlant, an Israel-based, regenerative medicine company, is focused on developing and commercialising tissue repair products with its plant-based technology. The company’s genetically engineered plants produce human recombinant collagen, which is initially targeted for use in orthopaedic and wound healing products. Current products on the market include a soft tissue repair matrix for treating tendinopathy and a wound repair matrix to promote rapid optimal healing of acute and chronic wounds. Additional applications, at an earlier stage, include bone void filler for spine fusion and Trauma and rhCollagen-base Bioink for 3D printing of tissues and organs. CollPlant conducts its product research and development work at offices and research laboratories in the Weizmann Science Park in Ness-Ziona, Israel. The agricultural research and development and extraction activities for its rhCollagen are conducted in North Israel. The company became public in May 2010 through a reverse merger with Portfolio Green. It trades on the Tel Aviv Stock Exchange (TASE) and has raised c $56m to date.

We value CollPlant at $61m, or 48c per share, based on a risk-adjusted NPV analysis, with a 12.5% discount rate. We model cash flows for each of the current product lines, VergenixSTR and VergenixFG, from their current stages through to potential expiration of market exclusivity (ie patent expiry in 2032). We note this excludes a contribution for the earlier R&D programmes and the inherent value of the rhCollagen platform technology, which has the potential to be developed into a number of other applications. This offers upside potential if the programmes progress into the clinic and/or CollPlant secures partners or funding. We also employ relatively conservative assumptions, which could bring further upside as we get more clarity on uptake and spread.

We forecast FY16 net cash of NIS3m and a funding requirement in FY17, despite CollPlant raising c.NIS 7m on TASE in Feb 2017, for which we model an illustrative loan of NIS15m. In the absence of a significant fund-raising (although including the illustrative NIS15m long-term debt in FY17), we expect the main areas of spend to be its R&D activities during FY17 and FY18. We forecast R&D spend of NIS13.3m and NIS15.3m and an SG&A spend of NIS3.6m and NIS4.0m in 2017 and 2018 respectively. This is lower than preceding years as its main products have moved into the commercialisation stage (via distributors and partnerships) and now that the NASDAQ offering has been abandoned, as there was an increase in costs associated with this.

Key sensitivities include funding requirements and execution of the tendinopathy and wound care product sales (including reimbursement) and the rate of clinical progress in orthobiologics. Commercial success in each area will depend on the success of the regulatory process, reimbursement and surgeon uptake. Development of the pipeline and technology is subject to additional funding and/or partnerships to conduct further studies and/or to grow salesforces that could prove dilutive to current shareholders. CollPlant has received government grant funding to develop its technology, which comes with various stipulations on the return of the funding amount and location of manufacturing and knowledge. Finally, while CollPlant performs the extraction process from the tobacco plant, the purification process that produces rhCollagen is carried out by a subcontractor in Israel (in-house capability being developed) and the end products are manufactured by a subcontractor in the US. Although strict protocols are in place, it should be noted that the process is not completely in house and there could be risks in execution and delivery.

CollPlant is a regenerative medicine company. It develops and commercialises medical devices for regenerating human tissues based on a proprietary plant-based genetic engineering technology to produce recombinant human collagen (rhCollagen) from tobacco plants. Initially, CollPlant introduced five human genes encoding heterotrimeric type I collagen to tobacco plants. Once this is achieved the ongoing production process has four broad steps: growing the tobacco plants that contain the type I collagen genes in greenhouses, harvesting them, extracting the collagen and purifying it.

The resulting rhCollagen is, according to the company, identical to type I human collagen, which can then be used in a number of biological applications in particular tissue repair. Tissue engineering and regenerative medicine is an advancing therapeutic area, which is generating demand for collagen fibres that mimic native collagen fibre structure and function and can then be produced in large quantities and processed effectively. CollPlant’s rhCollagen reportedly offers this and has a number of features1 that we believe could differentiate it from existing animal and human tissue options, including:

The investment story is built on the versatility of the technology described above. The potential application is broad and the adoption of biological, as opposed to standard treatments such as synthetic implants, is driven by the need for earlier intervention and cost savings for longer-term healing solutions. CollPlant is initially focused on the wound care and orthobiologics markets, but also has a potential pipeline of other applications including cardiovascular, ophthalmic, 3D bio-printing and other extracellular proteins (such as elastin, fibronectin). We estimate that the current target market for CollPlant is c $15bn. This is based on Smith and Nephew’s estimate that the advanced wound care market is $8.5bn (CAGR 4-5% 2015-18) and Transparency Market Research’s estimate of the global orthobiologics market of $5.5bn in 2019 (CAGR 5.9% from 2013 to 2019). A significant commercial advantage of CollPlant’s products is that they can be stored and transported cost-effectively as many potential competitor products are cryopreserved, which is more costly and can create structural integrity problems. CollPlant is operating in highly innovative areas where commercial traction could make these products attractive to larger medical devices companies that need to enhance their portfolios.

Type I collagen is the most abundant form of collagen in the human body. It is the dominant constituent of connective tissue and is best suited for applications associated with regenerative medicine because of its essential role in the healing process of bones, skin and tendons. Type II and type III collagen are primarily found in articular cartilage and granulation tissue (artery walls, skin, intestines, uterus) respectively.

The human body has five genes that encode heterotrimeric type I collagen: two are structural genes (collagen α1 and α2 protein chains) and three are enzyme genes (prolyl-4-hydroxylase [P4Hα and P4Hβ] and lysyl hydroxylase 3 [LH3] enzymes). CollPlant introduces the two structural genes into one tobacco plant and the three enzyme genes into another. These are cross-bred to obtain a parent plant that contains the five genes. The plant essentially becomes a biofactory, able to synthesise the proteins. The plant produces procollagen, which is extracted, then the c and n propeptides are cleaved, mimicking the process in a human, and are enzymatically converted to atelocollagen using a plant-derived protease. The protein is purified through a cost-effective industrial process, taking advantage of collagen’s unique properties that make it soluble at a very low pH.

Tobacco plants have been chosen for the medium for production because they:

The production process is outlined below in Exhibit 6. CollPlant conducts the growing and extraction part of the process in house and currently subcontracts the purification and product manufacturing. The company has indicated that scale-up is easily achieved and that the in-house purification capability will be implemented in 2017.

Orthobiologics uses cell-based therapies and biomaterials to help musculoskeletal injuries heal more rapidly. These products are made from substances that are naturally found in the body, which dynamically interact with the musculoskeletal system to facilitate the healing of bone, cartilage, meniscus, tendons and ligaments affected by disease or injury. Key market players include DePuy Synthes, Medtronic, Sanofi, Stryker and Zimmer Biomet, although unusually for orthopaedics the market is highly fragmented with a number of small players. According to GlobalData, the worldwide orthobiologics market is currently $6.7bn, which is growing (CAGR 7% 2016-20), driven by an ageing population, obesity, active demographics, innovative technology and emerging geographic areas. The orthobiologics market can be split into five broad segments: bone allografts, bone graft substitutes, growth factors, stem cell therapy and viscosupplementation. CollPlant is targeting the bone graft segments with its bone void filler and the growth factor segment with VergenixSTR, its tendinopathy treatment with PRP. Bone grafting is the predominant segment of the orthobiologics market. According to a Global Industry Analysts report, it is valued at $850m and is forecast to grow to $3.2bn by 2022. The PRP market was reported by Transparency Market Research as $160m in 2015 and is expected to grow to $452m by 2024, a CAGR of 12.5%.

VergenixSTR is a soft tissue repair matrix intended for the treatment of tendinopathy (micro tears in collagen fibres in the tendon) by promoting healing and repair of tendon injuries in a variety of tendons including the elbow, rotator cuff, patellar, Achilles and hand. The product combines cross-linked rhCollagen with autologous (extracted from the patient) PRP. The platelets contain growth factors that stimulate tissue generation and repair.

The predominant approach to tendinopathy is rest and physiotherapy. In approximately 5-10% of cases an intervention is needed. Current treatments offered for tendinopathy include a local steroid injection, which is the standard of care, shock wave therapy and injection of PRP alone. Steroid injections have been shown to damage the tissue, so repeat injections are not recommended. Equally, PRP alone requires more than one injection due to an issue relating to platelets localisation: it tends to dissipate quickly.

The main advantage of VergenixSTR over PRP alone is its ability to localise platelets in the vicinity of the injured tissue. VergenixSTR is injected into the tendon, where it forms a viscous gel matrix. This provides a temporary reservoir and scaffold for PRP. It essentially holds the PRP in place at the point of injury and potentially supports cell proliferation via sustained release of growth factors. The company indicates that, rather than being rapidly degraded (within 24 hours), VergenixSTR remains in place for up to a month. This issue of localisation, where the viscous gel matrix remains in place at the point of injection for a longer period of time, becomes increasingly important with the extension of VergenixSTR’s indications to its application in joints as opposed to tendons. This is because there is a large amount of synovial fluid in joints, which increases the problem of the injected substance remaining in situ as it is more easily ‘washed’ away. Exhibit 9 shows in vivo data demonstrating the clot degradation, comparing steroids with VergenixSTR.

Direct competition with VergenixSTR comes from companies that specifically sell steroid and PRP kits. These include Biomet, Harvest Technologies Corporation, MTF Sports Medicine and Arteriocyte Medical Systems.

VergenixSTR received its CE approval in October 2016 and CollPlant announced in November that it had signed an exclusive distribution agreement with Arthrex to commercialise in Europe, Middle East and Africa (EMEA). We believe there are two possibilities for commercialising this product in the US. CollPlant could introduce it there alone, or it could secure a commercialisation partner in the US, possibly via an extension of the partnership announced with Arthrex in EMEA or with another player. We assume launch in the US c.24 months following the launch in Europe via a partner. We do not currently include commercialisation costs in our model if CollPlant were to commercialise in the US alone. For an overview of the clinical data and commercialisation steps, see Exhibit 10.

This is at an earlier stage of development and CollPlant are currently seeking a partner (following the recent discontinuation of its joint development with Bioventus) to develop any product along the regulatory pathway and to commercialise. Bone grafts are pieces of bone or bone substitute that can replace damaged or diseased bone in a number of joints and bones in the body. Bone grafts have four mechanistic functions: structural, osteoconduction, osteoinduction and osteogenesis. These provide mechanical support and scaffold for bone formation, induce differentiation of progenitor cells into bone-forming cells, or osteoblasts, and directly contribute cells for bone formation respectively. Bone grafts can be autologous (graft taken from the recipient), allogeneic (grafts from cadaveric donors), xenogeneic (animal) or synthetic:

CollPlant’s bone void filler is a novel resorbable scaffold composed of rhCollagen and synthetic minerals that mimic bone structure. It is osteoconductive, but designed to be charged with a growth factor, which will create an osteoinductive/osteogenic implant, so it will stimulate bone and tissue growth. It is intended to be used as a one-time, easy-to-use treatment.

The initial target markets for its bone void filler will be in trauma and spinal fusion. GlobalData indicates that in 2013 there were 1.8 million bone-grafting procedures performed worldwide, c 400 million related to trauma and one million related to spinal fusion. The market is very fragmented with, according to CollPlant, over 50 DBM products currently available. Interestingly, however, GlobalData also indicates that fewer than 10 have shown clinical evidence of their benefits.

There has been a shift in treatment from traditional wound dressings towards advanced therapies that aim to optimise the wound healing environment. Advanced wound care products include products such as bioengineered skin and skin substitutes, along with dressings incorporating growth factors. In a healing wound, a cascade of events occurs that includes platelet accumulation, inflammation, fibroblast proliferation, cell contraction, angiogenesis and re-epithelization, ultimately leading to scar formation and wound remodelling. Collagen plays an important role in each of these phases of wound healing and, as a result, there are a number of biological wound care dressings available that incorporate tissue-derived collagen.

The overall wound care market is $14bn, according to Derma Sciences. The size of the advanced wound care market, according to Smith & Nephew, is $8.5bn. The wound care market remains dominated by basic products, such as gauze and bandages. This part of the market is low margin and experiencing price erosion. Innovative, new products such as dermal substitutes, growth factors and debridement agents have pricing power and are the growth part of the market. The wound care biologics market is a small but fast-growing segment of the wound care treatment landscape, forecast (according to Research and Markets) to grow at c 21% (CAGR) from 2016 to 2020. The primary market drivers are the ageing population and an increasing emphasis on earlier intervention and cost reduction. The US is expected to lead the market for the foreseeable future due to the high incidence of chronic and acute wounds, favourable healthcare reimbursement policies and an ageing population. In addition to this, there is a more conservative use of tissue-derived products, greater hurdles in the production of processed human tissue, and religious and sociological considerations, which have an impact on the use of, for example, porcine-derived products.

The major players in the advanced wound care market include Smith & Nephew, Acelity, Covidien, and ConvaTec. However, the remainder of the market is the largest segment (36%) and very fragmented.

Wounds are generally classified as acute or chronic. Acute wounds progress through the normal stages of wound healing and demonstrate definite signs of healing at approximately the four-week mark. Chronic wounds do not progress in this manner and ‘stall’ at one of the stages of healing. The value of the products currently used for acute wounds is low. Chronic wounds, such as diabetic foot ulcers, vascular ulcers and pressure ulcers by definition do not heal easily and as such there is an unmet medical need. Alongside the unmet medical need is the significant cost to the healthcare system of chronic wound treatment. This is mainly associated with nursing resources, hospital admittance and complications such as infection.

CollPlant indicates that it will target a broad number of indications including chronic wounds, diabetic foot ulcers (DFUs), pressure, venous and surgical wounds. DFU and venous leg ulcers (VLU) are particularly difficult and expensive to heal. Pressure ulcers are also prevalent, developing in 15-30% of all hospital patients (acute and chronic), and cause a 7% increase in mortality.2

Wound care is a complex market, primarily because it is not usually the focus of any single physician speciality but is handled in and as part of other specialities such as vascular surgery, gynaecology and dermatology. As a result, consensus about treatment protocols is challenging and exacerbated by variations in reimbursement and budgets. More importantly, patients often have co-morbidities and are affected by adverse socioeconomic factors, which can undermine compliance and effectiveness.

Wound care procedures are predominantly performed in an outpatient setting in Europe, with a well-developed system of specialised wound nurses. In the US, it tends to be in wound care centres and Healogics and Restorix are the largest commercial operators. However, procedures are also performed in surgery centres and physicians’ offices.

CollPlant’s advanced wound care product (VergenixFG) is intended for the treatment of surgical incisions and wounds, including diabetic ulcers, burns, bedsores and other chronic wounds that are difficult to heal. It is in a form of a dry powder that creates a gel when mixed with saline. This is applied to the wound after debridement, via a cannula. Once applied, a standard secondary dressing is applied to cover the wound. The gel is applied once and is intended as a single treatment, which potentially reduces follow-up visits. The gel provides a scaffold of collagen, which is thought to provide the wound with collagen to initiate healing by enabling cellular infiltration and capillary growth. The Vergenix-provided collagen degrades, leaving the endogenous native collagen to continue normal wound healing.

VergenixFG is the most commercially advanced product, having been on the market since mid-2016 with the first commercial patients treated in August 2016. For an overview of the clinical data and commercialisation steps, see Exhibit 15 below. The next steps for this product include developing traction in Europe via the establishment of a distribution network. As the product gains traction in Europe, the company will build up its clinical data then intends to find a collaborator to enter the US market. We expect this could be in c.24 months.

CollPlant protects its proprietary technology via a combination of patents, trade secrets and trademarks. Its global patent portfolio comprises eight patent families. The patents cover methods of creating collagen-producing plants and processing recombinant collagen. Core patents around processing start to expire in 2025; however, CollPlant states that protection extends until 2032. It is important to note that in addition to IP there are other significant barriers to entry to compete with a like-for-like product due to the level of knowledge required to be able to replicate the technology and the significant infrastructure requirements. For example, there are restrictions on genetically engineered tobacco plant growth in places such as the US.

Key sensitivities include funding requirements and execution of the tendinopathy and wound care commercial strategy, and the rate of clinical progress in wound care and orthobiologics. Commercialisation of the tendinopathy and wound care products is dependent on raising the visibility of VergenixFG among key opinion leaders (KOLs), which are typically conservative in adopting new technologies. While existing study data demonstrate promising results, larger studies may be needed to differentiate the wound care product from the range of advanced wound care products available. CollPlant’s other recently marketed product, VergenixSTR, is in an area where standard treatment is not to intervene as injection can cause unwanted side effects. Earlier uses of its technology such as the bone void filler potentially address a significant innovation gap; however, its development requires a partner to move forward, which is less assured following the announcement that Bioventus recently discontinued its joint development agreement. Development of the pipeline and technology is subject to additional funding to conduct further studies and/or to grow salesforces that could prove dilutive to current shareholders. We are not aware of any competitors that produce human collagen from plants or recombinant type I human collagen. The industry is characterised by rapidly evolving technology and intense competition, and its products will compete with several alternative tissue-derived or synthetic products. CollPlant has received government grant funding to develop its technology, which comes with various stipulations on the return of funding amounts and location of manufacturing and knowledge. As a result, the company could be affected by any change in governmental regulations or conditions. Finally, while CollPlant performs the extraction process from the tobacco plant, the purification process that produces rhCollagen is carried out by subcontractors in Israel (the in-house capability is being developed) and the products are manufactured by a subcontractor in the US. Although strict protocols are in place, it should be noted that the process is not completely in house and there could be risk in execution and delivery.

We value CollPlant at $61m, or 48c per share, based on a risk-adjusted NPV analysis. The breakdown of our rNPV valuation, which uses a 12.5% discount rate, is shown in Exhibit 16. We model cash flows for each of the current product lines: VergenixSTR and VergenixFG from their current stage through to potential expiration of market exclusivity (ie final core patent expiry in 2032).

We do not yet include the potential of VergenixFG for other applications such as for the treatment of surgical incisions, which the company has indicated it may target. There is also potential uplift to the selling price once a reimbursement code is established in each jurisdiction. We also note that rhCollagen is a platform technology and there are a number of other potentially valuable applications. We do not currently ascribe a value to CollPlant’s other R&D activities based on its rhCollagen such as 3D bioprinting; however, any future developments and progress into the clinic would offer additional upside to our base case valuation.

CollPlant could be an acquisition target either on a product-line basis or for its platform technology as a whole. M&A has been a significant factor in the orthopaedic sector, which has led to considerable consolidation. For example, LifeCell was acquired by Allergan for $2.9bn in 2016, Synthes was acquired by J&J/DePuy ($21.3bn in 2011), alongside other large deals such as Zimmer Biomet ($13.4bn in 2014) and Wright Medical and Tornier ($3.3bn in 2014). We expect a gradual shift away from these scale-based acquisitions towards transactions that enhance value through innovation and enable a focus on category leadership and portfolio depth. In the longer term, as CollPlant gains sales traction and builds an evidence base around its technology, it could attract M&A value. This could be enhanced as a result of the high barriers to entry due to IP protection, internal knowledge and regulatory restrictions on infrastructure and location. We note that the average price paid (we have removed deals that were significantly higher outliers) in recent orthopaedic deals is four times sales (see Exhibit 17 below). We have specifically selected deals that were in the orthobiologics and/or spine space.

CollPlant reported net cash of NIS8m at Q316. We forecast FY16 net cash of NIS3m and a funding requirement in FY17, (despite raising NIS7m in Feb 2017) for which we model an illustrative loan of NIS16m, which on our current forecasts gives a cash horizon to 2019. We expect CollPlant to carry out a significant round of fund-raising in 2019, if not earlier, to drive the business towards profitability. In the absence of a significant fund-raising (although including the illustrative NIS16m debt financing in FY17), we expect the main areas of spend to be its R&D activities during FY17 and FY18. This could include expansion of its product pipeline, additional indications for existing product candidates and improvement in yield and process. We currently forecast SG&A spend to return to lower levels now that the NASDAQ offering has been abandoned, as there was an increase in costs associated with this. We forecast R&D spend of NIS17.8m, NIS13.3m and NIS15.3m and an SG&A spend of NIS 7.3m, NIS3.6m and NIS4.0m in 2016, 2017 and 2018 respectively. It should be noted that the SG&A spend does not include the increased cost that would be necessary if CollPlant decided to take steps to commercialise VergenixSTR or VergenixFG itself rather than with a partner or through a distributor respectively.

CollPlant has received R&D grants from the Office of the Chief Scientist (OCS) of the Ministry of Economy and Industry. Up to June 2016 it received $7.2m. In turn, CollPlant has to pay a royalty of 3-5% of any revenues derived from products that resulted wholly or in part through the grant money. This is up to a total of the money received. An additional stipulation is that CollPlant is restricted in terms of manufacturing and transferring know-how outside Israel. The company would need to gain approval from the authority (OCS) if it were to do this.

We forecast revenue of NIS1.2m in 2017 and NIS2.9m in 2018, broadly split between VergenixFG and VergenixSTR. We expect VergenixSTR to achieve a higher portion of the revenue split as it is being sold via a partnership with an orthopaedic company (Arthrex), which has an established sales team in place across Europe. VergenixFG will be commercialised via distributors, which can take longer to get in place and is more segmental in terms of geographic coverage. Our current forecasts do not include any potential milestone payments from its current partnerships, potential OCS grant awards or potential future partnerships. We do not include any tax in our forecasts as the company has indicated that for the foreseeable future there will not be a requirement to do so. We incorporate the payment of tax in four years’ time (2020), at the Israeli tax rate of 25%. We expect to review this as we gain greater clarity on the progress of product commercialisation.

The company announced at Q316 that it was enacting a reverse share split of 3:1. We have restated the number of shares in the model to reflect this. Also, it is worth noting that it has a number of warrants and options, currently an additional 31.8m at an average exercise price of NIS1.9, and as such out of the money. It also has 10.96m options at a range of NIS0.9-3.9. If all warrants and options were exercised, the maximum dilution would c 25%.