Investment case

Karolinska Development traces its origins to 2003, when it was created to help finance the many medical and pharmaceutical innovations arising from the highly-renowned Karolinska Institutet. In 2014 a material shift in strategy saw the introduction of new management in 2015, with the aim of creating a Nordic life sciences investment vehicle that has a broader remit yet a more focussed portfolio. The portfolio now consists of ten companies that are developing novel treatments for life-threatening and seriously debilitating diseases.

The investment case rests on the real value of the portfolio and the progress within each of the companies. There is potential for significant value creation as exits (typically trade sales, IPOs, and/or licensing deals) are achieved at valuations more than typically modest carrying values. However, the complex nature of the Rosetta Capital venture, coupled with the uncertainties surrounding the convertible bond continue to overhang the investment case.

Valuation

We increase our valuation of Karolinska Development at SEK1,055m (equivalent to SEK16.65/share) from SEK1,042m (equivalent to SEK16.44/share) based on a sum-of-the-parts valuation of its five leading investments – Aprea, Modus Therapeutics, OssDsign, Promimic, and Umecrine Cognition. At this stage we assign no value to the remaining portfolio assets, nor the potential earn-outs, and hence they could be viewed as representing possible further upside.

Financials

Karolinska Development had a cash and equivalents position of SEK145.7m at Q118 (SEK169.6m at FY17 and SEK211.3m at Q117), and of this an estimated c SEK30m is earmarked for further follow-on funding rounds. The equity position was SEK247.4m at Q118 (SEK267.1m at FY17 and SEK62.8m in Q117). The convertible debt, which matures in December 2019, was valued at SEK379.2m at December 2017 (SEK394.4m in 2016) following the “set off” issue in March 2017.

Karolinska Development has a complex strategic alliance with Rosetta Capital, which owns 7.5% of the KDev Investment vehicle (this holds several of Karolinska Development’s more sizeable investments). The agreement with Rosetta Capital means that it has priority over dividends from KDev Investment and could potentially exercise a put option on its shareholding from March 2018, but, as we discuss later, we believe that it is unlikely to exercise this option.

Sensitivities

The investment case rests on the success of the investment strategy, in particular the ability to achieve increases in portfolio value over and above net new investment. Karolinska Development is most exposed to the business success of its five larger portfolio holdings. A significant success or set-back is likely to not simply affect the portfolio value but also impact investor sentiment.

As mentioned above, the Rosetta Capital dividend sharing structure and terms and the redemption of the convertible bonds in December 2019 remain major sensitivities. Much clearly depends on the progress of the key late-stage companies and the quantum and timing of any likely realisations or exits.

Karolinska Development: an exciting year ahead

Karolinska Development continues to make laudable progress in its goal to become the leading Nordic life sciences investment company. A wide-ranging overhaul in recent years means the portfolio now consists of nine companies that have highly differentiated and commercially attractive products. A multitude of significant value-inflection points are expected to materialise over the next 12-18 months, notably key results from five “proof of concept” clinical studies. The convertible debt does overhang the share price, as does understanding the complexities of the Rosetta Capital dividend share out. Nonetheless, our DCF valuation based primarily on the five late-stage investments is SEK1,055m (equivalent to SEK16.65/share), which is over twice the current share price.

The well-documented changes in investment philosophy and management has seen Karolinska Development undergo a material transformation over the past three years. The more rigorous, and pragmatic, investment strategy has resulted in the portfolio being streamlined from 21 companies at December 2014 to the current nine. These remaining investments have been selected because they are developing differentiated and commercially attractive products that have the potential to demonstrate convincing clinical and economic benefits. All are financed through to their next value inflection points, which are expected to arise within a period of twelve to eighteen months.

The rich vein of news flow is highlighted by the expectation that five clinical programmes within the therapeutic portfolio are due to report Phase II data during the remainder of 2018 and into early-2019. These data are commercially important as they are viewed as demonstrating “proof of concept” and successful outcomes often trigger licensing discussions with pharmaceutical partners or are used to support an IPO. Similar important realisation points are also approaching within the medtech portfolio, as both OssDsign and Promimic are generating recurring revenues and have significant commercial milestones in sight.

Looking further ahead, Karolinska Development has now developed a well-structured selection process to identify the most commercially attractive medical innovations. The philosophy is to use its existing network and collaborators to recognise and develop transformational research and technologies that originate from leading universities, teaching hospitals, and research institutes from across the Nordic region. Management is actively scouting out suitable investment opportunities to replace the maturing companies and so replenishing the portfolio.

Understandably, despite the evident progress achieved, investor concerns do remain. These mainly centre on the convertible debt, the strength of the equity position, and the complexity of the Rosetta Capital dividend split. Clearly positive clinical outcomes, followed by successful and sizeable financial realisations, would help alleviate these. The balance sheet is currently sound and has funding in place for the known investment and operational needs, with SEK145.7m in cash and equivalents at March 2018. In our view the equity position would benefit from further strengthening, with the timings of the first sizeable value realisations being highly pertinent. Nevertheless, we would argue that these concerns, although valid, are already discounted in the current share price.

The investment strategy and portfolio

Karolinska Development’s original remit was to fund and develop the technologies that arose from the various research programmes at the Karolinska Institutet. However, in 2014 a comprehensive review resulted in the investment philosophy being overhauled and the remit broadened to cover attractive therapeutic and medical technology (med-tech) innovation from universities and academic centres across the Nordic region.

Historically the programmes selected for investment tended to be early-stage with Karolinska Development taking a sizeable stake, usually a majority holding, and nurturing these embryonic businesses through the inevitably difficult first years. Now the strategy is much more flexible and allows for a non-controlling investment in more mature companies, where the reduced timescale to a value-inflection point means a financial return may be more material and, importantly, may be realised more quickly.

Exhibit 1: Karolinska Development’s investment approach in a nutshell

 Source: Trinity Delta

Ensuring all portfolio companies are properly financed to deliver on their next value milestones has become a key priority. This is achieved through not only direct funding, but also proactively syndicating investment with appropriate international life sciences funds. Such co-investment with experienced specialists also reduces the overall portfolio risk, helps widen the investment reach, and serves as valuable third-party endorsement for the commercial and technological attractiveness of the individual companies.

The changes in philosophy may appear subtle, but they are significant. The earlier stage businesses are actively nurtured through to the point when specialist operational management is appointed, whilst the more mature companies are supported financially and strategically (mainly through board appointments). All investments are made with deliberate consideration of the expected financing requirements that allow progress to the next defined value inflection point and potential exit. For pharmaceutical and biotech products this tends to be when meaningful clinical data is expected (eg the proof of concept is usually Phase IIa trial results), and for med-tech companies and other technologies it is when they become sustainably cash flow positive.

The outcome is a portfolio that has a better balance of early- and later-stage businesses; is better able to access appropriate financing; carries a reduced operating and investment risk; and has a greater profile among external specialist investors.

The portfolio review resulted in several divestments being made, with the number of active investments reduced from 21 in December 2014 to the current ten (Exhibit 2). An economic interest has been retained in most of the divested companies (including Axelar, Athera Biotechnologies, Clanotech, Inhalation Sciences, Lipidor, and Oncopeptides), with additional value expected to accrue as these businesses progress.

Exhibit 2: Karolinska Development’s Portfolio Investments

Source: Karolinska Development & Companies; Note: *These are held by KDev Investment, of which Rosetta Capital owns 7.5%

In the following pages we describe the investment cases for Aprea, Modus Therapeutics, OssDsign, Promimic, Umecrine Cognition, and Dilafor. Whilst all have important milestones and value inflection points due over the coming 18 months, it is worth highlighting that near-term value accretion also arises from other investments, for instance Oncopeptides and BioArctic in the past year.

Oncopeptides was successfully floated on Nasdaq Stockholm in February 2017 and Karolinska Development’s 5% earn-out agreement with Industrifonden has a current market value of c SEK 50m. Similarly, BioArctic successfully listed on Nasdaq Stockholm in October 2017, and Karolinska Development sold its shareholding in two tranches (3.2% to 0.6% in November 2017 and to 0.0% in April 2018) and received proceeds of SEK 35m and SEK12m respectively.

Aprea AB: coming of age of p53 targeting?

Aprea is a private Swedish company that is developing novel small molecules targeting the tumour suppressor protein p53 pathways. The de-activation of p53 is associated with uncontrolled cell growth and mutation of the p53 is seen in at least half of all solid tumours. Aprea’s lead candidate, APR-246, is a first-in-class compound that appears to reactivate the pathway and so induces cell death (apoptosis). APR-246 is currently in the Phase IIa element of the PiSARRO trail in relapsed ovarian cancer, which aims to examine 250 patients with relapsed high-grade serous ovarian cancer in Europe and the US. Promising results from the Phase Ib element were presented at ESMO in October 2016, with the important top-line data from the Phase IIa element due during 2018. A broad ranging research collaboration with Memorial Sloan Kettering to evaluate APR-246 in multiple solid tumour types with a variety of anti-cancer therapies was announced in March 2017.

p53 is a hot target in the fight against cancer

The p53 protein is normally present in all cell types, where it plays an integral role in regulating the cell cycle and is a central plank in the body’s defences against cancer. It is at the core of a complex network of proteins that monitor the genetic health of a cell; when damage is detected p53 orchestrates responses that, depending on the degree of mutation, either initiates repair mechanisms (cell-cycle arrest for DNA repair or senescence for growth regulation) or induces apoptosis (cell death).

p53 triggers apoptosis through multiple mechanisms, including mitochondrial and death receptor pathways, cytoskeleton changes, suppression of survival signalling, and induction of hypoxia. Due to this key function in stabilising cell mutation and proliferation, p53 has been described as the “guardian of the genome”.

The loss of p53’s suppressor function, either through mutation or inactivation of down-stream pathways, is present in most (arguably all) human cancers. Such frequency, regardless of patient age or tumour type, suggests a major role in a cancer cell’s life history. Importantly, both p53 mutation and pathway dysfunction are associated with poor clinical outcomes, with a close correlation to resistance to chemotherapy and radiation treatment.

The high number of known mutations possible in a single p53 protein means that the potential permutations of p53 mutants, each with a distinct molecular shape and biological properties, are too great to target specifically. Hence the focus has been on identifying areas that are common to a wide range of p53 mutations yet retain a major role in their key repair activities.

Understandably, p53 has been one of the most studied areas in cancer, with literally tens of thousands of papers published over the past 30 years. Yet, the complexity of the interactions is such that it was only recently demonstrated that rather than simply being transiently permissive for tumour initiation, persistence of p53 dysfunction is a continuing requirement for maintaining tumour growth.

The nature of the p53 interactions has generated a number of drug approaches, ranging from gene therapy to classical small molecules. The various methods can be envisaged as the reactivation of mutant p53, such as the stabilising of protein folding, and blocking the effects of p53 inhibitors, such as Mdm2 and MdmX.

APR-246 is in proof of concept clinical development

Aprea’s approach uses a small molecule to reactivate the p53 pathways by directly binding to mutant p53, which in turn allows the correct folding to occur and restores p53 activity. The lead compound, APR-246, belongs to a new chemical class called quinuclidinone that is converted to the active compound MQ (methylene quinuclidinone). MQ is a Michael acceptor that binds covalently to cysteines in the p53 core domain and allows the refolding and stabilisation of mutant p53 hence re-introducing its cell regulating activity. Interestingly, MQ also restores function to incorrectly folded and dysfunctional wild type p53.

APR-246 traces its origins to research at Karolinska Institutet led by Prof Klas G. Wiman in 2000. A targeted screening programme of a chemical library of the National Cancer Institute led to an initial compound, APR-017 (also known as PRIMA-1) being identified. Lead optimization resulted in the more potent and drug-like APR-246. These, and similar compounds (including back-up analogues), have extensive intellectual property rights, with broad protection through five patent families. Currently market exclusivity is to around 2035, with the patent estate continuing to be broadened.

Exhibit 3: Pre-clinical evidence of synergistic effect of APR-246

Source: Aprea

Pre-clinical testing showed APR-246 has inherent anti-cancer activity (with binding of the compound to P53 sufficient to induce apoptosis), but also demonstrates synergistic effects when used in combination with other chemotherapy (Exhibit 3); including platinum-based drugs (eg cisplatin), antimicrotubule agents (eg docetaxel & vinblastine), anti-metabolites (eg 5-FU), anti-tumour antibiotics (eg doxorubicin), and check-point inhibitors (such as anti-PD-1). Activity is also seen in treatment resistant tumours, which suggests reactivation of p53 re-sensitises cancer cells to the chemotherapy.

A Phase I study in 22 patients with haematological malignancies (mainly AML with 7 patients) and prostate cancer (also 7 patients) showed APR-246 (as mono-therapy) to be safe at the predicted therapeutic plasma levels, with a suitable pharmacokinetic profile, and able to induce measurable p53-dependent biologic effects, including anti-tumour activity.

The first phase of Phase Ib/II study (PiSARRO) in relapsed ovarian cancer was reported at ASCO in June 2016. This evaluated APR-246 (at escalating doses of 35, 50 and 67.5 mg/kg) in combination with carboplatin and pegylated liposomal doxorubicin (PLD), in 28 patients with recurrent p53 mutant platinum sensitive high grade serous ovarian cancer (HGSOC). The main toxicity attributable to APR-246, rather than chemotherapy, was dizziness (20/28 patients) and this resolved within 24 hours of infusion. Encouraging activity was seen in both partially and fully platinum-sensitive patients.

Exhibit 4: Graphic of patient responses with APR-246

Source: Aprea

Further data were presented at ESMO in October 2016 showing that of 22 patients with radiologically measurable lesions, 3 had confirmed complete response, 10 had confirmed partial response, 8 had stable disease and 1 was not evaluable. For the 2 patients with non-measurable disease, 1 had complete response and 1 had progressive disease (Exhibit 4). Although it should be remembered that only a limited number of patients have been treated so far, this equates to an impressive overall response rate (ORR) of 50%. Median progression-free survival (PFS) for the 22 evaluable patients, as measured by RECIST or GCIG, was 316 days (95% CI, 280-414 days) and was not influenced by the length of previous platinum treatment-free interval or dose cohort.

APR-246 showed linear pharmacokinetics with no accumulation and low intra-patient variability. There was no indication of interaction between APR-246 and chemotherapy, supporting the combination of APR-246 with carboplatin and PLD at relevant doses. The side-effect profile was similar to earlier findings with the most frequent treatment-emergent adverse events have been low grade GI (nausea/vomiting), CNS (dizziness and fatigue) and haematological (neutropenia and thrombocytopenia) events. A dose of 67.5mg/Kg has been carried into the next study phase.

The Phase IIa element, which enrolled its first patient in October 2016, sees up to 400 relapsed, p53-mutated high-grade serous ovarian cancer patients in Europe and the United States randomised into two equal treatment arms; these will evaluate up to six cycles of carboplatin and PLD, and carboplatin and PLD plus APR-246. Patients will be followed for safety, response (RECIST 1.1 and CA125 (GCIG criteria)), and several secondary endpoints, but the primary endpoint will be progression free survival (PFS). This “proof of concept” phase of the trial is due to complete recruitment in early-18 and to report top-line data before end-18.

APR-246 to be evaluated in other tumour types

Aprea is exploring APR-246 in other cancer types, with three Phase Ib trials in platinum resistant high grade serous ovarian cancer (HGSOC, first patient treated in August 2017), oesophageal cancer (first patient treated October 2017), and myelodysplastic syndrome (MDS, first patient treated May 2017 and promising initial results presented in April 2018) in combination with various chemotherapy regimens (Exhibit 5). Concurrently, preclinical work is also being performed to assess which other potential indications offer the most promise and are worthy of progressing into the early clinical stages. Additionally, APR-246 is currently administered as an i.v. infusion but related compounds, using the same chemical scaffold, are being developed for oral administration.

Exhibit 5: Aprea’s clinical trials programme

Source: Aprea, Trinity Delta

In March 2017 Aprea announced a broad ranging collaboration with Memorial Sloan Kettering Cancer Center to study the effects of reactivation of tumour suppressor protein p53 by APR-246. The aim is to evaluate and characterise the preclinical efficacy of APR-246 in combination with multiple other anti-cancer agents and across multiple tumour types. The Principal Investigator of the study is Taha Merghoub, Ph.D., and he will be conducting the research in collaboration with Jedd D. Wolchok, M.D., Ph.D., Chief of Melanoma and Immunotherapeutics Service, Department of Medicine.

Aprea successfully raised SEK437m ($51m, €46m) in a Series B financing round in March 2016. The funds allowed the completion of the Phase II element of the ovarian cancer trial as well as further exploratory trials in other cancer indications, including both solid and haematological tumours.

Assuming smooth progress, management expects market launch around 2020. From an investment perspective the key inflection point will happen when the Phase II trial results are known as this will effectively be the “proof of concept” and, we suspect, could rapidly lead to a material liquidity event. Although the most advanced indication of stage II-IV recurrent ovarian consists of only around 20,000 patients per annum, a successful outcome would likely lead to a broadening of this patient population as well as increase comfort for its potential use in the additional tumours that are being evaluated.

Karolinska Development currently owns 18.5% of Aprea (indirect holdings through KDev Investments and KCIF Co-Investment Fund), with other specialist investors including Versant Ventures (US), 5AM Ventures (US), HealthCap (Sweden), and Sectoral Asset Management (Canada).

Modus Therapeutics AB: pivotal results in H218

Modus Therapeutics (previously known as Dilaforette) is a private Swedish company that is developing sevuparin, a novel heparin-based polysaccharide, for treating Sickle Cell Disease (SCD). Sevuparin is currently in Phase II trials in Europe, the Middle East and the Caribbean. In H216 the size of the study was increased to a total of c 160 evaluable vaso-occlusive crisis (VOC) resolutions and in November 2016, following a planned safety review, the trial enrolment was expanded to adolescents aged 12 to 18. Top line data from this important proof of concept study is expected in H218. In February 2017 a financing round raised SEK32m (c $3.8m) from the existing investors: KDev Investments, The Foundation for Baltic & European Studies (Östersjöstiftelsen) and Praktikerinvest.

Exhibit 6: The effect of sickled red blood cells on blood flow

Source: National Heart Lung and Blood Institute

SCD is poorly treated and in need of effective agents

SCD (also known as sickle-cell anaemia) is a chronic inherited genetic disorder that affects the formation and shape of red blood cells. It was formally identified in the 1950s and was the first disease to have its genetic cause identified; a major milestone in human genetics research. SCD affects around 70,000 to 100,000 people in the US, predominantly of African and Hispanic descent. In Europe the estimates vary from 35,000 to 127,000 people, with the higher figure reflecting the quantum of migration from Sub-Saharan Africa. Larger populations are seen in the Middle East (the Arabian Gulf University reports there are 250,000-300,000 in the Gulf Co-operation Council area alone), Africa, India, and South America.

The hallmark of sickle cell disease is recurring episodes of severe, debilitating pain commonly known as sickle cell crisis or vaso-occlusive crisis (VOC). The intense pain experienced by patients is the result of obstruction of blood vessels by “sickled” red blood cells, which are rigid and highly adherent to the vessel walls and to each other. This obstruction leads to reduced blood flow to organs, including the bone marrow, not only causing severe pain, but also cumulative tissue damage and, ultimately, loss of vital organ function and significantly reduced lifespan (Exhibit 6).

The incidence of VOCs varies widely, with some patients rarely experiencing them whilst others have 10-plus per annum. US data shows there are between 80,000 to 100,000 hospitalisations related to VOCs, which supports the reported average incidence of 0.8 to 1.2 VOC per annum and average hospital stay of five to six days. There are few options to shorten the duration or reduce the severity of a crisis once underway, with treatment usually consisting of intravenous hydration to aid blood flow and opioid or similar potent analgesics to relieve pain. Blood transfusions and oxygen are also often required depending on the severity of the attack.

Hydroxyurea (also known as hydroxycarbamide), originally a chemotherapy agent, is the only FDA approved (1998) drug for VOCs. Hydroxyurea increases the production of foetal haemoglobin (HbF), the form of haemoglobin present in the foetus and small infants, which prevents sickling of red blood cells. Despite its demonstrable benefit/risk profile and low cost, the usage of hydroxyurea is patchy (particularly in Europe) with concerns of a raised risk of haematological malignancies being identified as a barrier to its use.

Sevuparin is showing promise in early clinical trials

Sevuparin is a new chemically modified heparin with low anticoagulant activity that has the potential to become a major treatment for SCD. It is an anti-adhesion agent that works through antithrombin III-dependent inhibition of thrombin as well as blockade of P-selectin-mediated adhesion. The specificity should allow a decrease in the vaso-occlusion seen during a VOC but with a low impact on the typical heparin anti-coagulation activity.

Preclinical data and early clinical studies show that sevuparin can have rapid and clinically relevant effects to prevent and resolve the micro-vascular obstructions of VOCs. Data presented at ASH (December 2016) shows it acts in a multi-cellular manner, blocking both firm adhesion by sickle red blood cells and L-selectin-mediated rolling adhesion of sickle-leukocytes, as well as interacting with the key adhesion receptor VCAM-1.

The Phase II study is now a circa 160 patient (up from the original 70 patients), multi-centre, double-blind, placebo-controlled study in hospitalised SCD patients experiencing VOCs. The patients are treated with i.v. infusion of sevuparin or placebo on top of standard pain medication, primarily opioids. This is an important proof of concept study that is designed to demonstrate reduced time to resolution of VOC – defined as freedom from parenteral opioid use and readiness for discharge from hospital.

The study is being performed in Bahrain, Lebanon, Netherlands and Turkey in collaboration with Ergomed (a specialist CRO) as part of a co-development deal. Ergomed is co-investing a proportion of its revenues from the clinical and regulatory activities of this trial in return for an equity stake (believed to be around 10%) in Modus Therapeutics. The results of this important trial were originally expected to be reported in late-2016; however, the patient expansion, coupled with slower than planned rate of patient enrolment (the study in now due to complete in H118), has seen this pushed back to H218.

Sevuparin has been granted Orphan Drug status in the US (March 2015) and in Europe (January 2015). The FDA designation brings user fee benefits, registration assistance and seven years of market exclusivity post-launch. With the EMA there are similar administrative and procedural benefits, but the period of market exclusivity is ten years. In November 2016 two US patents were granted that provide protection to at least December 2032, with equivalent patents pending in other major markets.

The costs, both economic and quality of life, of SCD are substantial and the clinical need for new treatments is clear. Against this backdrop, the commercial impact of a SCD treatment that reduces hospital stay and the use of opioid analgesics could to be substantial (the hospital setting work is known as the Resolve programme). Additionally, if further studies demonstrate sevuparin’s efficacy in early treatment (this is known as the Ease programme) then the subsequent label extensions would expand the market potential materially. Management is looking at more patient-friendly dosing methods, including formulations (similar to existing mini-heparin devices) that could be self-administered at home to prevent the VOC developing.

Sevuparin is also being explored for use in other clinical indications where micro-occlusion is a factor. Its anti-adhesive properties were shown in a Phase I/II open label study (published December 2017) in uncomplicated malaria. In malaria the red blood cells become adhesive due to infection by the parasite Plasmodium falciparum, which leads to clumping of the red blood cells and adherence to blood vessel walls. Sevuparin was shown to open clogged vessels and to rapidly inhibit parasite growth. The results would support further evaluations in malaria.

Non-dilutive funding in place through to possible IPO

Sevuparin is the result of initial research performed at the Karolinska Institutet and Uppsala University. Karolinska Development (through KDev Investments) currently owns 72% of Modus Therapeutics, with other investors including The Foundation for Baltic and East European Studies (Östersjöstiftelsen) and Praktikerinvest. A clinical collaboration agreement with Arabian Gulf University (Bahrain) provides up to $1.2 million in non-dilutive funding in return for royalties on sales in the MENA (Middle East and North Africa) region that are capped at $2.4m. Although no details are forthcoming, we believe that management has successfully sounded out the option of an IPO but a trade sale or suitable licensing opportunities have not been ruled out.

OssDsign AB: innovative implants by design

OssDsign is the result of the research collaboration between clinicians and scientists at Karolinska University Hospital and Uppsalla University to design and develop innovative implants for cranial and facial reconstruction. It was formed in 2011 and, following successful animal and human clinical studies, now uses its proprietary technology to custom-make implants specifically for individual patients. The focus has been on using a combination of novel bio-ceramic formulations, supported with a reinforcing titanium skeleton as necessary, with sophisticated computer-aided design, 3D printing and moulding to create individualised cranial and facial implants.

Addressing a clear and sizeable market need

The market for craniomaxillarfacial (CMF) implants is sizeable and growing, driven by greater needs for reconstructive surgeries as survival rates from head and neck cancers and road-traffic-accident injuries increase. The CMF market, in common with most orthopaedic indications, is crowded and highly competitive; being dominated by global players such as DePuy Synthes (part of Johnson & Johnson), Stryker Corp, and Medtronic as well as a broad number of specialised players (such as KLS Martin, Aesculap, Medartis AG, and TMJ Concepts).

Grand View Research estimates the global CMF segment to be worth around $1.2bn in 2016, rising to $1.7bn by 2022 (6.5% CAGR), with Markets & Markets reporting the value at $1.7bn, rising to $2.5bn by 2022 (6.9% CAGR). The report by Transparency Market Research shows similar conclusions. North America is the largest market (around 58%) and is expected to remain so as clearer re-imbursement pathways and greater surgeon awareness (aided by extensive educational and marketing campaigns) drive the uptake of newer techniques and materials. This receptive environment means North America remains the primary commercial focus for the R&D driven elements of the industry.

Global Data has specifically looked at Europe and notes that whilst demand will continue to grow (2.8% CAGR to 2020), the cost containment measures, coupled with inadequate surgeon training for such specialised procedures, will result in slower adoption rates for premium devices (including patient specific implants and virtual planning technologies). Similarly, whilst Asia-Pacific is expected to show significant growth (albeit off a smaller base), the opportunities for the higher performance technologies will remain restrained by constraints in re-imbursement and, importantly, surgical expertise.

The bespoke segment highlights OssDsign’s skills

OssDsign consciously operates in the bespoke segment, where complicated or difficult cases require individualised implants. Initially the focus is on revision or corrective surgery, where other implant types have failed to deliver the desired results; however, as surgeon confidence improves, this is expected to see greater usage of OssDsign’s products as first-line choices. OssDsign expects this specialised niche to grow to be worth between $100m to $200m.

Traditionally cranial defects were repaired with metal plates and/or bone cement, sometimes re-enforced with stainless steel grids. A major disadvantage of such methods is that a perfect fit and curvature can be difficult to obtain in large reconstructions, especially when the defect involves the cranio-facial junction. Traditional cranio-facial implant materials are also associated with high complication rates (requiring expensive revisions) and carry the lifetime risk of skin penetration and infection. This has led to the development of innovative materials (such as polymers and metallic alloys) that can significantly reduce such complications, albeit with variable clinical and commercial success. Pre-fabricated cranial implants, which are tailored to the specific defect of the individual patient, are now available in different composite and bio-materials.

Exhibit 7: OssDsign’s Cranial implant

Source: OssDsign

OssDsign’s products bring clear clinical benefits

OssDsign’s principal products are OSSDSIGN Cranial, for reconstructing complex cranial defect, and OSSDSIGN Facial, used in surgical augmentation and reconstruction of facial bone contour defects. These are individualised for each application and employ three-dimensional CT scans to create implants that are perfectly matched to the underlying or remaining bone structures. The use of bio-ceramic materials results in better blood flow, with improved vascularisation and tissue integration; improvements in soft tissue healing, with better aesthetic results; and sustained bone remodelling, as cell-mediated responses stimulate new bone formation.

An evaluation of 107 historic cranial implants used in 15 clinical sites since October 2014 has shown excellent outcomes, particularly in view of the severity and complexities of the cases reviewed. There were only six complications noted: one deep wound infection that required explantation of the device; two superficial wound infections that responded to antibiotic treatment; and three cases of wound dehiscences in incisions located directly over the implant in thin and fragile skin, which were resolved with local flap reconstruction without requiring explantation.

Addressing the US market is a clear priority

OssDsign markets the cranial and facial implants directly using a small, specialist sales team as well as through distribution partners. The complex nature of the surgical procedures means the implants are used by a limited number of dedicated medical centres and these can be identified and accessed relatively easily. The products have been launched in Scandinavia, Germany and the UK and in selected other markets (such as Singapore, Malaysia and Israel). In January 2017 distributors for Italy, Spain, Switzerland, Austria and The Netherlands were announced, which means the European market is now effectively covered.

Following the 510(k) clearance from the FDA in January 2017 a commercial presence has been established in the US. Matador Medical now acts as the master distributor, recruiting and managing a network of sales people to reach all the specialist centres and opinion leaders in the US. Training and education programmes were put in place to support a first launch in April 2017. In January 2018 a direct US subsidiary was also established. Preparations for the Japanese market are also progressing, with the timelines of the regulatory process suggesting a launch is possible in 2018. A strategy to access the Chinese market will be decided this year (most likely a strategic partnership). Similarly, product opportunities outside the bespoke CMF segment, including the Cranioplug, are expected to be partnered.

The management team was strengthened by the appointment of a new CEO, Anders Lundqvist, in March 2015 and CFO, Claes Lindblad, in April 2016. The previous CEO, Bo Qwarnström, who nurtured the company from its formation, has been retained as VP Business Development. Also in April 2016, the well-respected medtech executive Simon Cartmell joined as Chairman.

The shares are largely held by the four founders (Dr Thomas Engstrand, Prof Håkan Engqvist, Dr Jonas Åberg and Bo Qwarnström) and three institutional investors (Karolinska Development, SEB Venture Capital and Fourier Transform). Karolinska Development owns 25% of the shares when the indirect KCIF Co-Investment Fund holdings are included).

Promimic AB: performance enhancing coatings

Promimic is a Swedish biomaterials company that has developed a unique coating, known as HA^nano Surface, that is used to improve the integration of implants into bones. It was founded in 2004 as a spin-off from research performed at Chalmers University of Technology in Gothenburg, Sweden. The coating can be used on a wide range of substrates, including metals, ceramics, and polymers, and on a variety of structures, including complex and delicate implants. A strategic partnership with Danco Anodising has resulted in a commercial processing plant being established in the US to supply the North American and Chinese markets.

Better integration and enhanced bone growth

HA^nano Surface is based on the substance hydroxyapatite (HA). This is a form of calcium phosphate that is similar in chemistry, composition, and morphology to living bone tissue. HA is highly stable and its hexagonal structure and crystalline shape is identical to bone apatite. HA’s role and value in orthopaedics is well documented, with every comparative study with a non-coated matching device demonstrating that the speed and effectiveness of integrating into bone were dramatically improved. HA has excellent properties in terms of biocompatibility, bioactivity, osteoconductivity, and also has generally low toxicity and a non-inflammatory nature. Implants coated with HA have shown major improvements in crucial properties such as osteointegration, lubricity, and fatigue strength.

Understandably, this has led to the widespread use of HA in many orthopaedic applications. Many techniques have been explored to achieve effective and consistent HA films ranging from pulsed laser deposition to electron and ion beam sputtering; but thermal spraying, in particular, plasma spraying is the main commercial method for producing HA coatings. Spraying achieves thicknesses of 30-200µm and is a well understood process; however, control of the variables is quite complicated with even small changes in the processing parameters able to vastly affect the properties of the final coating.

The HA^nano Surface coating procedure is surprisingly simple. It is a patented wet chemical process where the HA crystals are formed in a coating liquid that is applied directly onto the implant in one of three different ways (dipping, spraying, or dripping). The excess liquid is removed through spinning and pressurised gas, which is then followed by a short heat treatment. The coating is nanometre thin (around 20 µm) and suitable for a variety of surfaces (including porous) and shapes (including complex structures and geometries). The process is more cost effective than other coating methods and its flexibility means it can be integrated, using easily available off-the-shelf equipment, into an implant manufacturer’s internal production lines.

The coating’s thinness allows the newly forming bone to attach directly to the micro-structure (topography) of the implant surface rather than to the coating itself. The thinness also means that there is no risk of cracking or flaking and the implant material’s characteristics are retained. From a user’s perspective there are no changes in the surgical instruments used or in the actual procedure. From a customer point of view, Promimic assists in all the stages from concept evaluation and development, through to preparation for commercial production and technical support post-launch.

Exhibit 8: The HA^nano Surface coating procedure

Source: Promimic

HA^nano Surface’s performance has been evaluated in over 20 pre-clinical in vivo studies, as well as validation exercises with clinicians and industry customers. HA^nano Surface is very versatile and can be employed with metals (such as titanium, stainless steel and cobalt-chrome alloys), ceramics (including aluminium oxide and zirconium oxide), polymers (notably polyether ether ketone – PEEK), and pyrocarbon. Its use with porous materials is noteworthy as the process allows uniform coverage, even within the smaller pores, with its hydrophilic nature creating an effective osteoconductive surface.

Promimic is transitioning from being a development phase company to a commercial business. The progress over the past few years has resulted in the customer base growing, with a solid pipeline of further evaluations underway. The business model is to sell non-exclusive licenses to the medium and large implant and dental companies. Exclusive licenses are limited to tightly defined sub-segments in terms of geography, application and implant material. The licenses will carry a royalty, typically 2%-5% of net sales, with a consumable element, the HA^nano Surface coating liquid, also sold. Most of the potential customer base is in North America and a US office has been opened in Austin, Texas in Q1 17 (Promimic Inc) to better address these opportunities.

The customer progression typically follows four separate steps:

• Concept evaluation – feasibility of coating the customer’s devices or implants. Providing advisory services in regulatory and production;
• Process Development – optimisation of the coating processes and design of the equipment for industrial scale production;
• Production preparation – technology transfer to production site, on site process validation, and initial batch runs with subsequent ramp up;
• Commercial Phase – provision of the coating liquid and continuing assistance with QC, training and technical support.

Three licensing agreements were signed in 2015-6. In 2015 Sistema de Implante Nacional (S.I.N), a leading provider of dental implants in Brazil, signed up and in January 2016 launched the first product using Promimic’s technology. The other agreements are with Amendia Inc. (2015) for use with its patient-focused spinal implant and Danco Anodizing (2016) where Danco has invested in creating a production line and is the preferred process partner for Promimic for the USA and China medical implant market. A number of evaluations are currently underway and further agreements, including with several major manufacturers, are expected in the near- and medium-term.

Although demographics and increased trauma injuries are driving demand, the orthopaedic market is crowded and highly competitive, with re-imbursement pressures mounting in most markets. Against this background, manufacturers are seeking novel technologies to help differentiate their product ranges and to reduce their manufacturing costs. In this context Promimic is well placed and the goal to achieve annual sales of around $20m over the medium to longer term appears realistic. Similarly, the licensing model suggests that the target of an operating margin of 75% is also reasonable and sustainable.

In January 2017 Mr Magnus Larsson, a proven executive with 15 years of sales and marketing experience in the dental implant industry, was appointed as CEO. Concurrently, the former CEO, Ulf Brogren, has moved to the USA to establish and lead the sales operations at Promimic Inc. in Austin, Texas. This, together with the strategic partnership with Danco Anodising, reflects the importance of the US-based implant manufacturers to Promimic near- and medium-term commercial future.

The major investors are Karolinska Development (34% via KDev Investments), Almi Invest Västsverige, K-Svets Venture, Chalmers Ventures and Innovationsbron.

Umecrine Cognition: recognised as first-in-class

Umecrine Cognition is a private Swedish company that was founded in 2006 as a spin-out from Umecrine AB (founded in 2000) and is based on the Karolinska Institutet Science Park in Solna, Sweden. It was Umecrine that undertook the original research that showed the negative effects of stress and sex hormones on the brain can be prevented. The related company, Asarina Pharma, was also spun out in 2006 to focus on menstrual related conditions.

Umecrine Cognition is developing compounds that act on endogenous CNS-active steroids (GABA-steroids), with the lead indication being for Hepatic Encephalopathy. GR-3027 is a first-in-class agent that acts to reverse the increased GABA_A signalling that is believed to cause the clinical symptoms of cognitive and motor function impairment. The strategy is to develop GR-3027 to proof of concept Phase II trials and then, assuming positive outcomes, seek a suitable commercial partner that has the skills and resources to complete the registration trials and market the product.

Hepatic encephalopathy (HE) is a serious complication in acute and chronic liver disease. It is characterized by impairments of consciousness, cognition, memory, and is associated with personality change and reduced motor skills. The resulting debilitation has direct costs in terms of reduced quality of life for the patient, as well as the associated costs of the care required. The prognosis following the first hospitalisation is poor; with a survival probability of 42% at one year and 23% at three years. HE is often classified into four stages:

• the first typically sees an inverted sleep-wake pattern (sleeping by day, being awake at night);
• the second is marked by lethargy and personality changes;
• the third shows confusion and a worsening of other symptoms; and
• the fourth is marked by a progression to coma.

The cause is a build up of ammonia (a product of protein digestion), and other related toxins, as the reduced liver function means it fails to clear from the blood stream. Ammonia crosses the blood-brain barrier easily where it is absorbed and metabolised by astrocytes and results in a raising of glutamate levels. These in turn lead to cerebral oedema (swelling) and impaired neural signalling. It has been postulated for some time that the resulting imbalance in the GABA neuro-transmitter system plays a material role in HE.

GR3027 is a GABA_A receptor modulating steroid antagonist (GAMSA) designed to antagonize GABA_A receptor activation by endogenous neuroactive steroids. Two animal models have shown it improved or normalised cognitive function and a two-part Phase I trial in healthy volunteers showed encouraging safety and tolerability. The single ascending dose study in 48 healthy volunteers demonstrated a clean safety profile and predictable pharmacokinetics. A second part of the study involving 18 patients, presented at the EASL Liver Congress in Amsterdam (April 2017), showed that GR3027 enters the CNS and can reverse the inhibitory effects of the endogenous neurosteroid allopregnanolone on brain function.

A Phase Ib/IIa study enrolled its first patient in March 2017 and completed the first stage in September 2017. The Phase Ib part in 18 healthy male patients has shown that GR3027 is well tolerated up to the highest dose of 100mg BID (twice daily for five consecutive days, with only mild adverse events and no-dose limiting toxicity, and with a favourable pharmacokinetic profile. The study has now advanced into the Phase IIa element, which is a double-blind, placebo-controlled trial evaluating up to 18 cirrhotic patients with evidence of HE (dosing yet to be disclosed). This should provide preliminary indications of the effect of GR3027 on cognitive function. Interim data from the Phase IIa element is expected in 2019. The clinical programme is funded by a syndicate of local investors who raised SEK12.6m during H116.

GR3027 is also being evaluated for the treatment of idiopathic hypersomnia (IH), a category distinct from narcolepsy, where the patient exhibits excessive day-time sleepiness despite normal night-time sleep and is not refreshed following a nap. GABA_A is known to be intimately involved in sleep onset and regulation and its antagonism could prove beneficial in limiting excessive sleepiness. The limited number of patients mean that it qualifies as an Orphan Drug indication, with the benefits in clinical trial design and registration support that entails. A Phase IIa trial was initiated in November 2017, following a SEK20m targeted fund raise from existing investors in October 2017. Preliminary results from this study are expected to be presented during 2018.

Karolinska Development is the major investor, holding 72% of the issued capital, with other sizeable investors including Norrlandsfonden, Fort Knox Förvaring, and PartnerInvest.

Dilafor AB: helping make labour easier

Dilafor AB is a private Swedish drug development company that is exploring pharmaceutical products for obstetric indications. The company was founded in 2003 by Professor Gunvor Ekman Ordeberg and colleagues and is based on the Karolinska Institutet Science Park in Solna, Sweden. The approach centres on exploiting the broad range of non-anticoagulant activities that heparin has, in this case to help promote cervical ripening and myometrial contractility (uterine contractions). The principal product is tafoxiparin, a modified form of heparin that is optimised for these indications. It is in Phase IIb clinical development to decrease the incidence of protracted labour both after induction of labour and after spontaneous onset of labour.

Labour is essentially a series of intense, repeated muscle contractions. The contractions help push the baby out of the uterus (womb) into the birth canal and beyond. Although in reality it is usually a continuous process, it is divided into three stages. Prolonged labour is often defined as when the total duration of childbirth is greater than around 20 hours, with a range of 18-24 hours depending on local practice. There are two main types, one when the latent phase of the first stage of labour is greater than circa 8 hours and the other when the active phase is greater than circa 12 hours. Little to no descent of the baby occurs during the latent phase, though contractions become more established, with increasing frequency, duration, and intensity. A prolonged latent phase may be physically exhausting and emotionally draining but rarely leads to complications. Active labour is the next phase and typically happens when the cervix dilates from 3-4cm to around 10cm. Prolonged labour at this phase does cause concerns.

Assessing the incidence of prolonged labour across the Western world is hampered by the varying clinical practises and classifications employed, with estimates ranging from 8% to over 20%. A common feature though is that nulliparous women (first-time mothers) are three to four times more likely to experience it than multiparous women (experienced mothers). Prolonged labour is seen as a primary cause in non-planned surgical deliveries (vacuum extraction, forceps, and caesarean section), with their attendant complications. Having ruled out an obstruction, especially in experienced mothers, a variety of physical and/or pharmaceutical measures are employed to try to improve cervical ripening and uterine contractions. The three main drug regimens used commonly are prostaglandin gel (vaginally), misoprostil (vaginally or oral), and oxytocin (IV infusion).

Heparan sulphate is a complex polysaccharide that is closely related to the anti-coagulant heparin. It is known to have a broad range of effects across the body, including helping in the ripening of the cervix and initiating and maintaining of uterine contractions, and a deficiency is thought to be an important contributor to the prolonging of labour. Tafoxiparin is a proprietary heparan sulphate mimetic that appears to optimise the effect of oxytocin in obstetric indications (effectively acting as an adjuvant).

Dilafor initiated a proof of concept Phase IIb clinical trial with tafoxiparin in women with protracted labour in six centres across Sweden in January 2017. This will enrol around 360 uncomplicated nulliparous mothers-to-be and examine tafoxiparin as an adjunct to oxytocin in primary slow labour, including prolonged latent phase and primary labour arrest. There are three active arms (low, medium, and high dose) and a placebo compactor. The trial is tentatively scheduled to complete in May 2018, with top line data likely around six months later.

A Phase IIa trial was performed in 263 first-time mothers over two years in 18 centres across Sweden. Although the primary endpoint of reduced labour time did not reach statistical significance, a sub-group analysis showed significantly fewer women were in labour for more than 12 hours compared to placebo and similarly also indicated that labour times were shorter in the induced labour sub-group. Additional Phase I studies were carried out in 2015 and 2016 in Europe and Asia, with the tolerability and safety data replicating the positive results seen in the original Phase I study.

In February 2014, Dilafor entered into licensing and partnership agreement with Lee´s Pharmaceuticals. Lee’s Pharmaceutical has the right to manufacture, develop and commercialize tafoxiparin for obstetrics and gynaecological indications in China, Hong Kong, Macau and Taiwan. Dilafor received an undisclosed upfront payment and is eligible for future development and sales milestones plus royalties on eventual net product sales. The results of the Swedish Phase IIb trial will be used by Lee’s Pharmaceuticals to guide their Phase II and III trial programme in China.

A financing round raising SEK51m ($5.9m) to fund the Phase IIb study was completed in October 2016. As a result of this financing, Karolinska Development decreased its indirect holdings via KDev Investments in Dilafor from 53% to 39%. The investors now include Lee’s Healthcare Industry Fund, Rosetta Capital IV, Östersjöstiftelsen, Praktikerinvest PE AB, and Pila AB.

Other portfolio investments

We have only detailed the six most advanced portfolio companies that are actively managed (although Dilafor has yet to be included in our valuation model). The remaining investments are either earlier stage or can be viewed as being more passive in nature. The earn-outs are linked to portfolio companies that have been divested following the strategic overhaul. The other portfolio companies include:

• Forendo Pharma is a private Finnish drug development company based on work from the University of Turku and Turku University Hospital. It has two active programmes addressing endometriosis. The first is a HSD1781 (hydroxysteroid dehydrogenase) enzyme inhibitor that promises a novel mode of action with the potential for disease modification. A lead compound has been identified and is being prepared for entry into the clinical phase of development during 2018. The second is an undisclosed dual HSD inhibitor that combines two mechanisms for the treatment of endometriosis and other endocrinological disorders. Endometriosis affects around 170m women globally, causing chronic pelvic pain, abnormal bleeding and infertility. Existing treatments, such as analgesics and hormone therapies, have limited efficacy and associated side-effects. Karolinska Development owns 15% of Forendo Pharma.
• Asarina Pharma is a private Swedish company spun out from the original Umecrine business in 2015 (it was previously known as Umecrine Mood). The lead compound, sepranolone, is a GABA_a modulating steroid antagonist (GAMSA) that is a first-in-class compound for pre-menstrual dysphoric disorder (PMDD). A 120 patient Phase IIa study showed good efficacy and safety and a Phase IIb trial started in April 2018. PMDD affects around 5 per cent of women during their fertile years and is linked to an altered sensitivity in the brain to an endogenous steroid produced following ovulation, resulting in monthly recurrent episodes of depression, anxiety and mood lability. Current treatment options include anti-depressants and oral contraceptives. Karolinska Development currently owns 2.5% of Asarina Pharma.
• Pharmanest is a private Swedish company focussed on a novel formulation of lidocaine, known as SHACT, for pain relief in obstetric and gynaecological indications. A Phase II trial in 218 women undergoing IUD placement has shown promising results with notable pain relief and discomfort reduction. Around 55m women worldwide have an IUD fitted each year. A licensing agreement with Gideon Richter for Europe, Latin America and certain other markets was signed in October 2017. Plans for US registration trials are expected to be finalised during 2018. Following an earn-out agreement activated in November 2017, Karolinska Development owns 10% of Pharmanest.

Other assets included the holding in BioArctic, which has now been fully divested following the realisation of SEK35m in November 2017 and SEK12m in April 2018, and the holding in XSpray, which was also divested over the past year for SEK13.3m. Other earn-outs still in place include Oncopeptides, Athera, Axelar, Lipidor, and Clanotech.

Sensitivities

Following the divestment of various sub-optimal investment positions and the issuing of the convertible debt in January 2015 (see below), Karolinska Development is reasonably well capitalised. Hence, the incremental change in value of Karolinska Development now primarily depends on the performance of its portfolio companies and its investment strategy.

Karolinska Development is exposed to all the risks that its portfolio companies face, including the development, regulatory, commercialisation and financial risk of young companies in the biotech and med-tech sectors. However, these risks are diversified across the ten portfolio companies, limiting both the downside and upside of events on Karolinska Development.

A key part of Karolinska Development’s new investment strategy is to work more broadly with third-party investors. This should result in Karolinska Development being offered more investment opportunities from which it can make attractive returns; it also transfers a material element of the continuing funding risk to the existing portfolio companies as successful funding rounds will depend on those companies finding investors other than Karolinska Development alone.

There are two issues that could significantly impact the returns of investors in subsequent years:

• the decisions of its strategic partner Rosetta Capital; and
• the actions of the holders of the convertible debt.

From 7^th March 2018, Karolinska Development could be obliged to buy out Rosetta Capital’s position in KDev Investments (see later in the Financials section), which could currently result in shares worth 10% of Karolinska Development’s market cap being issued to Rosetta Capital or the equivalent in cash. The option expires 7^th May 2018 and, at the time of publication, there is no indication what Rosetta Capital intends to do with this option. However, as detailed later, the appeal of exercising the option is limited.

In December 2019, convertible debt valued at SEK484m will need to be repaid (reduced from SEK586m following the outcome of the set-off issue in April 2017), be converted into shares (conversion price is SEK22) or be renegotiated. The majority of these, around 82.9%, are held directly or indirectly by the Chinese CP Group (see later).

Much clearly depends on the progress of the late-stage portfolio companies and the news flows over the next 12-18 months are crucial. There are five clinical programmes in four different companies that are expected to produce commercially relevant clinical data. These are Phase II trials that effectively act as proof of concept for the therapeutic compounds. Successful outcomes should result in material uplifts in valuation and often act as triggers for out-licensing, acquisition, or IPO. It will be the quanta and timings of the subsequent financial realisations that will shape Karolinska Development’s longer-term outlook.

Valuation

We consider a sum-of-the-parts DCF-based methodology (including rNPV where relevant) to be the most appropriate way to value Karolinska Development. As always, we employ conservative assumptions and have only valued the company’s five key late-stage investments, as indicated in Exhibit 9, to value Karolinska Development at SEK1,055m, or SEK16.65 per share. This compares to our previous valuation of SEK1,042m, or SEK16.44 per share after updating our model to reflect the Q118 results. The inherent value of the other investments in the portfolio and potential earn-outs from previous investments should be viewed as additional upside to our valuation.

The valuation of Karolinska Development is complicated by the ownership structure of Aprea, Dilafor, Modus Therapeutics and Promimic, which are owned by KDev Investments, in which Rosetta Capital is a minority shareholder with preferential dividend distribution rights (See Financials section). To take this into account, we have estimated the potential timing, value and likelihood of the exits for these four companies, before applying the dividend distribution structure, to value the dividends payable to Rosetta Capital.

Exhibit 9: DCF valuation of Karolinska Development

Source: Trinity Delta; Note: ¹ % owned by Karolinska Development directly or by KDev Investments; ² Companies owned by KDev Investments, in which Rosetta Capital owns 7.5% of the equity; ³ Karolinska Development’s fully diluted stake in OssDsign is now 20%, as OssDsign has revised its strategy to target a yet-to-be-disclosed larger opportunity; for the time being we continue to value OssDsign on its initial strategy with KDev Investments owning 28% of OssDsign; ⁴Net cash at Q118 includes the proceeds from the sale of its stake in BioArctic in April 2018 for SEK12m.

In comparison to our valuation of SEK1,055m (SEK16.65 per share), Karolinska Development’s total Fair Value of its investments at March 2018 was SEK729.0m (SEK11.5 per share), which with a potential distribution of SEK272.2m (SEK4.3 per share) to Rosetta Capital results in the net portfolio Fair Value of SEK456.4m (SEK7.2 per share). It should be noted that the current maximum distribution to Rosetta Capital is capped at circa SEK35m (see Financials section for the details of the distribution formula). For completeness, the book value of Karolinska Development’s investments at FY17 was SEK448m (SEK6.96 per share).

Karolinska Development follows the International Private Equity and Venture Capital Valuation Guidelines (IPEV) and IFRS 13 Fair Value Measurement to value the investments in its portfolio companies. This is a prudent and conservative approach in that the value of each holding is based on the post-money valuation of the company’s last funding round with a third-party investor regardless of the company’s progress; although values may be written down because of setbacks. For example, there was a write down in the value of its investment in Lipidor of SEK3.6m, following a financing round with only external investors at a lower valuation.

In certain cases where material progress has been made, such as Umecrine Cognition in Q417, an external third-party valuation is employed. This resulted in a value increase of SEK196m in Karolinska Development’s holding. No investments are assessed using simple DCF methods, which should only be used for companies with revenues that can be predicted with a higher degree of certainty. Consequently, we would expect our valuations of Karolinska Development’s investments to be materially higher than the book values.

Within Karolinska Development’s investment portfolio, we see most potential upside with Aprea. As Aprea’s lead product APR-246 advances through clinical development, the likelihood of success will increase (assuming that the clinical trials continue to deliver promising results). On top of this, we believe that our peak sales estimate could increase materially; our current peak sales estimate is based solely on its potential in ovarian cancer, but as changes in p53 (APR-246’s target) is associated with most tumours, APR-246 could be used to treat many other cancers. The company is also investigating the compound’s potential in oesophageal cancer, MDS and melanoma, but APR-246’s potential in these other indications is not yet included in our valuation.

The relatively high percentage holdings it owns in Modus Therapeutics and Umecrine Cognition means that Karolinska Development is particularly sensitive to progress in these companies. Positive outcomes in the Phase II proof of concept trials should result in a material uplift in valuations, but it is at the point of realisation (eg trade sale or IPO) that the true value will be seen.

There is also a degree of upside to our valuation from the investments in the other portfolio companies. These should be included in to our valuation in the future, once there is greater clarity about their commercial opportunities. Similarly, we have placed no value on the various earn-out agreements that Karolinska Development has on the divested investments.

Financials

Karolinska Development’s balance sheet is currently sound and has sufficient funding in place for the known investment and operational needs, with SEK145.7m in cash and equivalents at March 2018 (SEK211.3m in March 2017).

In 2017 a total of SEK34.1m (SEK4.8m in 2016) was invested in the portfolio companies, of which SEK32.5m (SEK4.5m in 2016) was cash. A similar amount is expected to be required this year. The equity position has improved considerably over the year, SEK267.1m at December 2017 (SEK29.8m at FY16) and SEK247.4m in March 2018 (SEK62.8m at Q117). However, over the near to medium term we believe it could benefit from further strengthening.

Obviously, the progress expected within the late-stage portfolio companies could mean the balance sheet is bolstered significantly in 2018/9 from successful realisations and exits. This will largely depend on the results of the key clinical trials with Aprea, Modus Therapeutics, and Umecrine Cognition as well as OssDsign and Promimic gaining meaningful commercial traction (particularly in the US) over the next twelve months.

Rosetta Capital – put option available from March 2018

Several of Karolinska Development’s major holdings (eg Aprea, Modus Therapeutics, Promimic, and Dilafor) are held by KDev Investments. This investment vehicle is the result of a strategic alliance between Karolinska Development and Rosetta Capital created in December 2012. Rosetta Capital bought 7.33% of the common shares for SEK110m (since increased to 7.53% following additional investments into KDev Investments) and preference shares worth SEK110m at the time of the deal, and 13.66% of the voting rights. As part of the agreement, KDev Investments is obliged to distribute dividends to Rosetta Capital in a “waterfall” as follows:

• 100% of dividends up to a total of SEK220m;
• 35% of dividends if total dividend payout is SEK220m-880m;
• 25% of dividends if total dividend payout is SEK880m-1,320m;
• 5% of dividends if total dividend payout is >SEK1,320m.

Rosetta Capital also has a put option, which can be exercised for a period of 60 days from 7^th March 2018 and would require Karolinska Development to acquire Rosetta’s position if Rosetta has not received a return equivalent to 2.5x the amount it has invested in KDev Investments. The price payable, in either cash or shares in Karolinska Development, would be based on the fair value of KDev Investments and the dividend distribution formula above, but with the value capped at 10% of the market cap of Karolinska Development.

For instance, the fair value of KDev Investments’s portfolio at March 2018 was SEK300.5m. This means the current potential distribution to Rosetta Capital is SEK272.2m, including investments worth SEK36.5m that are repayable to Rosetta Capital before the application of the formula. However, as the market cap is circa SEK400m, Karolinska Development would only have to repay SEK40.0m.

At the moment it appears that Rosetta Capital will be entitled to exercise its put option, but we consider that it is unlikely it will be exercised due to the current market cap of Karolinska Development. Should the option be exercised, we believe Karolinska Development would make any payment to Rosetta in shares, which although it would dilute existing shareholders, would strengthen its equity position.

Convertible bonds – repayment due in December 2019

In order to support the funding of the portfolio companies convertible bonds with a nominal amount of SEK 387m were issued in January 2015. These bonds are listed on Nasdaq Stockholm (ISIN: SE0006510103) and carry an annual interest rate of 8% (which compounds) with repayment due on 31 December 2019. The outstanding amount, including interest (which is added annually), was approximately SEK379m as of 31 December 2017 (against SEK394m at December 2016).

The bonds can be called for conversion into B-shares up until 30 June 2019 at a price of SEK22.00. Assuming these are held to maturity with the interest accrued, the amount repayable would now be around SEK484m with 22.08m new shares issued. CP Group currently holds around 82.9% of the convertible bonds and a further 7.54% of the equity (representing 6.23% of the votes). These are held indirectly via Sino Biopharmaceutical Limited and its subsidiaries.

As background, in February 2017 Karolinska Development announced its intention to offer to issue new B shares to its convertible bond holders though a set-off arrangement. The voluntary agreement involved new shares being issued at SEK6.17 per share. Convertible holders accepted to offset SEK67m of the debt and 10.9bn new B shares were issued. The level of acceptance was below our expectations; for context the maximum set-off would have seen proceeds of SEK451m, which compares to the current outstanding convertible debt, including accrued interest up until 31 December 2017, of circa SEK379m.

Our forecasts in Exhibit 10 do not include any potential transactions (investments or divestments) or assumptions regarding changes in the fair value of investments.

Exhibit 10: Summary of financials

Source: Karolinska Development, Trinity Delta  Note: Adjusted numbers exclude changes in the fair value of portfolio companies & exceptionals. Forecasts exclude potential investments or divestments.

Company information

Contact details

Karolinska Development AB
Tomtebodavägen 23A,
SE-171 65  Solna,
Sweden
Tel: +46 8 524 860 70

www.karolinskadevelopment.com

Key personnel

Source: Karolinska Development Note: A shares carry 10x voting rights of B shares