4SC — Update 21 March 2016

4SC is a Munich-based biotech focused on developing small-molecule drugs for cancer. The company’s R&D pipeline is based on its expertise in cell signalling pathways and epigenetic regulation, which play a crucial role in cancer development. Lead product resminostat will commence a Phase II study in cutaneous T-cell lymphoma (CTCL) in H116. Resminostat is also partnered with Yakult Honsha in Japan, where Phase II trials for front-line HCC and second-line NSCLC have been initiated. Resminostat has completed a Phase II trial in second-line HCC, a single-agent Phase II study in Hodgkin’s Lymphoma, and a Phase I trial in colorectal cancer (CRC). 4SC Discovery, a wholly owned subsidiary, has a number of early-stage collaborations (eg BioNTech, CRELUX). 4SC was founded in 1997, listed on the Frankfurt stock exchange in December 2005, and as of year-end 2015 has 59 full-time employees.

Our risk-adjusted NPV renders a valuation of €143m or €7.55 per share. This includes an rNPV for resminostat across multiple indications and regions, which we lowered to €106m (vs €122m) after reducing the probability of success for NSCLC in Japan from 40% to 10%, to more accurately reflect the lower probabilities of success in this notoriously hard-to-treat indication and a lower likelihood, we suggest, that Yakult will advance development in NSCLC, in comparison to the larger opportunity in Japan for HCC. This has been offset, however, as we have introduced a specific value (following the announcement of further promising data) for 4SC-202’s potential in AML (€34m), while retaining a carrying indicative value for 4SC-205 at €10m; previously 4SC-202 and 4SC-205 were assigned €25m together. This represents fair value for the stock ahead of initiation of the potentially pivotal resminostat CTCL Phase II trial, Yakult Phase II data in 2016, and advancement of 4SC-202 and 4SC-205 through fresh financing and/or partnerships.

4SC is subject to sensitivities typical of biotech drug development, including the unpredictable nature of clinical trials, the success or failure of competitors, changing market dynamics and a reliance on partnerships to commercially exploit its products. The company reduced its financial risk with a €27.5m net capital increase in July 2015, enabling the company to progress resminostat (ex-Japan) into pivotal development for CTCL. 4SC has an improved but limited free float (c. 38%) with a single large shareholder, Santo Holding, holding a c. 48% stake.

We estimate 4SC’s net cash at €21m (€23m cash and €1.9m debt) at the end of December 2015, sufficient to fund the majority of the planned CTCL Phase II study. We model two-year (2016-17) R&D and SG&A spend of €21m and €6.9m respectively, which suggests a cash reach into 2018. Our current assumptions do not include any potential financing (or partnership) to progress 4SC-402 or 4SC-405 clinical programs, or potential further development of resminostat for solid tumours in EU/US (assuming positive Yakult study results in 2016).

4SC’s near-term investment case hinges on the successful development and commercialisation of its epigenetic lead candidate resminostat for CTCL. 4SC’s €27.5m net capital increase in July 2015 has provided vital funds to advance the programme and internal focus is now on initiating the EU Phase II trial in CTCL. Clinical data in 2016 in HCC and NSCLC from the Yakult collaboration may allow 4SC to revisit the prospects for resminostat in other cancers outside of Asia. Epigenetics, described in more detail later in the note, represents a discrete and effective therapeutic mechanism to target certain cancers, while also holding potential to enhance the outcome of other approaches, especially in combination with cancer immunotherapies. Resminostat and 4SC-202 both act on epigenetic targets.

Resminostat has shown efficacy and safety in HCC and HL

Resminostat shows encouraging anti-tumour activity with good tolerability. Resminostat has to date been tested in patients in Phase I in a range of blood and solid tumours, both as monotherapy and in combination with chemotherapies in the EU and Asia. It was generally well tolerated, with adverse effects in line with or better than other HDACs, principally GI effects (nausea and diarrhoea), fatigue and thrombocytopenia. Importantly, there were no severe liver, cardiovascular or GI bleed effects. Phase II trials in relapsed Hodgkin's lymphoma (HL) as monotherapy and in second-line treatment of HCC in combination with sorafenib have completed and showed encouraging anti-tumour efficacy.

Resminostat Phase II trials have been initiated in its lead indication of first-line HCC as well as NSCLC with partner Yakult Honsha. The first readout in HCC in Japan is expected during 2016, and should the data be positive we would expect Yakult to proceed to pivotal Phase III studies in HCC in Japan. Yakult is also carrying out Phase I trials in GI cancers (pancreatic and biliary duct) in Japan, while the deal signed with new partner Menarini in April 2015 may allow for further HCC trials of resminostat in Asia (ex-Japan), dependent on positive data.

4SC is currently preparing a Phase II trial of resminostat in CTCL. It is scheduled to start in summer 2016 pending agreement from the regulatory authorities. Success in this trial and subsequent regulatory approval would make resminostat the first HDAC for CTCL in Europe.

The randomised double-blind placebo-controlled phase II trial will test resminostat in approximately 150 CTCL patients who have progressed beyond Stage IIB, have failed oral bexarotene treatment and have stable disease (or a partial or complete response) after four cycles of 'salvage' chemotherapy. Patients meeting these criteria will be randomised to receive resminostat plus best standard of care vs. best standard of care alone (the placebo arm) for 24 months, plus 12 months follow-up. The study will likely run until H119.

The level of adverse events (AE) will clearly be key to an EU approval given the failure of Merck & Co's vorinostat (Zolinza), which was withdrawn after the CHMP expressed doubts over the clinical benefit and the level of side effects including thromboembolic events (4.7% of patients had pulmonary embolisms). Grade 3/4 adverse events for vorinostat in a Hodgkin’s lymphoma (HL) phase II trial included anaemia in 32%, thrombocytopenia in 16% and lymphopenia in 12% of patients. Resminostat demonstrated much safer AE levels in HL, with anaemia in 8% and thrombocytopenia in 14% of patients.

The clinical positioning that 4SC is targeting for resminostat appears similar to other HDAC inhibitors which have been approved in the US, that is, second/third-line treatment after at least one prior systemic treatment (in this case oral bexarotene and gemcitabine). Compared with the monoclonal antibody products we would expect resminostat to have a pricing advantage, given the significantly higher costs of goods of biological products.

4SC have presented preclinical data on resminostat’s activity as an immunomodulatory, It has been found to enhance natural killer (NK) cell recognition and killing, tumour associated antigen expression and presentation and reduces unspecific immunosuppressive mechanisms. The data also indicated a potential synergy between resminostat and other immunotherapy approaches e.g. checkpoint inhibitors (PD-1/PD-L1), antibodies and immunostimulants.

There is potential, therefore, that Resminostat may not only reprogram cancer cells but also strengthen the immune system’s defence mechanisms against cancer cells. This could offer a major developmental strategy to explore the combination potential further for resminostat with, for example, PD-1/PDL-1 blockade as the immune priming qualities of resminostat are believed to enable increased efficacy of PD-1/PDL-1 blockade. This would potentially enable expansion into indications such as none and low-immunogenic tumours.

Preliminary biomarker data from SHELTER (Phase II in HCC) and SAPHIRE (Phase II in Hodgkin’s Lymphoma), including c 80 subjects, suggests that baseline gene expression levels of ZFP64 (zinc finger protein 64) may predict the clinical response to resminostat. With the caveat that this was a post hoc analysis, the results show a positive correlation between baseline ZFP64 mRNA expression in peripheral blood cells and clinical benefit – patients with higher baseline ZFP64 expression showed longer OS benefit (median near doubling) than those with low expression. Importantly, 4SC’s results (first presentation at ICLA meeting, 12-15 September 2013) show that only resminostat downregulates ZFP64 expression, whereas sorafenib apparently does not – ie no difference was observed in the ZFP64 correlation with OS for resminostat alone or in combination with sorafenib. Importantly, ZFP64 is relatively easy to measure by real-time PCR (polymerase chain reaction). We also note that the Yakult Phase II studies in HCC/NSCLC are measuring ZFP64 levels throughout, which should provide greater insight into the validity of ZFP64 as a potentially predictive biomarker for resminostat.

4SC-202 is 4SC’s second epigenetic drug which selectively inhibits 3 HDAC isoforms 1,2 and 3 ,and another epigenetic target, LSD1. 4SC-202 is once daily and orally administered and although two other LSD1 inhibitors are in Phase I/II trials with Oryzon and GlaxoSmithKline, 4SC-202 is the only dual acting epigenetic product acting on both HDAC and LSD1.

On the strength of the positive results achieved to date, 4SC continues to consider a range of options in relation to a potential Phase II development of 4SC-202, and possibly in collaboration with potential partners 4SC-202 has the potential to change our view of cancer therapies by developing treatments that target cancer stem cells. This could offer a broad range of novel therapeutic opportunities in haematological and solid cancer types and include:

This is an oral, small molecule inhibitor of Kinesin spindle protein (KSP) also known as Eg5 and one of only two in early clinical development (the other being Array Biopharma’s filanesib):

Phase I study – Phase I is ongoing. A comprehensive safety and tolerability profile has already been established, and an outstanding pharmacokinetic profile was demonstrated. Biomarker analyses have also confirmed 4SC-205’s desired mechanism of action. The study has since been broadened and is currently investigating a new, innovative dosage regime in a study amendment.

The images below illustrate the basic idea behind epigenetics. There are around 25,000 genes in the human cell, of which around one third are switched on at any one time. The genetic information (genotype) carried in the cells of the caterpillar and the butterfly is identical in each form yet the expression of that gene information results in a completely different animal form (phenotype). The approximately 200 cell types in the human body contain the same DNA sequence, yet they behave very differently when it comes to expressing this genetic information.

Epigenetics is concerned with what switches certain genes on or off. For example, the factors that determine one cell becoming a beating myofibroblast cell of the heart and another a pancreatic cell that secretes insulin. There is a difference between the genetic code that describes us and the genes that are active at any location at any point in time. This difference, which is not encoded in the DNA sequence, is determined by chemical and structural changes of chromatin and classified under the term epigenetics (‘in addition to the DNA sequence’). Through these different chromatin states, a human genome, which comprises approximately 25,000 genes, can adopt a large number of epigenetic variants known as epigenomes.

The two-meter long strands of DNA, with a total of 3.2 billion base pairs contained in our genome are located inside a small cell nucleus with a diameter of just 10 micrometres. They are compacted to reduce their length by a factor of more than 1:200,000. The DNA strands located in a nucleus are not freely floating around but are wrapped in and protected by protein globules known as histones. Five major families of histones exist: H1/H5, H2A, H2B, H3 and H4. Histones H2A, H2B, H3 and H4 are known as the core histones, while histones H1/H5 are the linker histones. They form doublets and assemble to form an octamer. Each octamer can coil 147 base pairs of the DNA double helix and constitutes the basic building block of chromatin: the nucleosome. Densely packaging the entire genome thus requires approximately 20 million nucleosomes (Exhibit 3).

The DNA histone polymer with all its chemical changes is known as chromatin (Exhibit 4). Chromatin can assume many different packaging degrees that compact right down to the metaphase chromosomes, which become visible during cell division. The lowest packaging degree (open chromatin), in which the chain of nucleosomes is shaped like beads on a string allows for gene activity, whereas a higher packaging degree (closed chromatin) prevents access to the DNA strands and switches off genes. Chromatin therefore assumes a double task with regard to organising our genetic information: it compacts and protects the DNA double helix, and it influences the activity of the genes by regulating the access to the DNA strands.

Think of histones as the artists of gene expression. Genes are the colours on an artist’s palette and contain all the information necessary for the painting. Histones bring those basic colours to life and create the masterpiece.

The ability of the histones to perform this task is explained by the complexity of the histone protein modifications. Although there are only five histone family types, there are currently over 70 types of protein modifications that can take place in various permutations across 20 million nucleosomes.

The histones have long N-terminal tails that protrude from the nucleosome and these are subject to post-translation modifications (Exhibit 4). Modifications of the tail include methylation, citrullination, acetylation, phosphorylation, SUMOylation, ubiquitination and ADP-ribosylation. Modification by enzymes occurs primarily on the N-terminal tails, but can also occur in the globular region of the nucleosome.

These modifications influence the expression of genes and therefore the human health state is very dependent upon the fine tuning of histone activity. Any detrimental changes to histone activity can lead to certain disease states such as cancer. Histone activity is regulated and affected by N-tail modifications, so changes in the activity of the modifying enzymes that catalyse methylation or acetylation for example are responsible for histone activity and consequently gene expression.

There are around 1x1014 cells in the human body, around 200 different cell types and more than a million regenerations per second. In the DNA of each cell there are around 3x109 base pairs with a precision of DNA repair of approximately 109. In each cell cycle there are around three base pair mistakes, which mean that throughout the cells of the body there are estimated to be 3x106 mistakes or mutations every second. Although almost all mutations are resolved there may be around one mutation per second that is fixed and these accumulate over our lifetime increasing the risk of mutations that can lead to cancer.

A recent paper by Luzzatto and Pandolfi in the NEJM highlighted the combined role of stochastic mutation (explained above) and environmental mutagens in the development of cancer. They note the low prevalence of cancer of the small bowel, despite the size of the organ and rapid epithelial turnover (cell replication and turnover). However, among patients with Crohn's disease, the risk of small-bowel adenocarcinoma is 20 to 30 times that among patients without Crohn's disease. This disparity could not be explained purely by huge differences in random mutations, but instead by the inflammatory microenvironment to manipulate an apparently genetically stable system. So in addition to normal base pair mutations in the DNA, there are other factors that can promote the development of cancer. Epigenetics is one of these factors.

Exhibit 5 below illustrates the more targeted anti-cancer approaches that have emerged over the last 10-15 years, of which there are many. The types of therapeutic entities include small molecules and monoclonal antibodies. More recent times have seen the emergence of cancer immunotherapy with checkpoint inhibitors and transgenic T-cells (CAR T cells). Epigenetic therapy could also be considered as a more recent mechanism that goes beyond the simple ‘genome instability and mutation’ shown in the hallmark diagram (Exhibit 5). It can be considered an entirely new and innovative addition to the armamentarium against cancer, which acts on cancers that arise due to a large influence from changes in chromatin.

Many cancers are associated with epigenetic mechanisms of the enzymes that carry out post-translation modifications of the histones. Consequently there are a number of different approaches (Exhibit 6) which block acetylation and methylation of the N-terminal histone chains.

4SC has two epigenetic candidates in development; resminostat and 4SC-202. Resminostat is a histone deacetylase inhibitor (HDACi) that acts on acetylation of the histone chain. It has already demonstrated safety and efficacy and is in phase II clinical trials for HCC (partnered) and CTCL. The second candidate is 4SC-202, a histone demethylase inhibitor, which will move into phase II in SCLC and/or a haematological cancer, should fresh funding and/or partnerships be secured.

The importance of epigenetics as a target is supported by a range of other HDAC inhibitors that have proven effective across multiple cancer types, including a number of drugs currently on the market (Exhibit 7). In addition to HDAC there are other epigenetic-targeted therapies that are already being exploited in the market or are being studied in clinical trials. These include inhibitors of histone demethylase (HDMi), histone methyltransferase (HMTi), histone acetyltransferase (HATi), and deubiquitinating enzyme (DUBi). Early studies have indicated a role for many candidates either as monotherapies or to enhance the activity of other drugs with different but complimentary mechanisms of action.

Although some products targeting epigenetics have reached the market, the approach is still in relatively early stages of development. This is an opportunity for 4SC since resminostat appears to have a safer and more effective profile in CTCL and HCC, with almost no clinical competition for these indications within its epigenetic class.

For context, in terms of the potential for successful candidates in this field, Velcade (bortezomib) achieved global sales of almost $3bn in 2014 (in multiple myeloma and mantle cell lymphoma) and ranks in the world’s top 50 drugs in terms of sales.

4SC is subject to sensitivities typical of biotech drug development, including the unpredictable nature of clinical trials, the success or failure of competitors, changing market dynamics and a reliance on partnerships to commercially exploit its products. The company reduced its financial risk with a €27.5m net capital increase in July 2015, enabling the company to progress resminostat (ex-Japan) into pivotal development for CTCL. 4SC has an improved but limited free float (c. 38%) with a single large shareholder, Santo Holding, holding a c. 48% stake. The future development of 4SC-202 and 4SC-205 is entirely dependent on securing fresh financing and/or a partner.

Our risk-adjusted NPV renders a valuation of €143m or €7.55 per share. This includes an rNPV for resminostat across multiple indications and regions, which we have lowered to €106m (vs €122m) after reducing the probability of success for NSCLC in Japan from 40% to 10%, to more accurately reflect the lower probabilities of success in this notoriously hard-to-treat indication and a lower likelihood, we suggest, that Yakult will advance development in NSCLC, in comparison to the larger opportunity in Japan for HCC. The recent failure of Peregrine Pharmaceuticals’ Phase III study with bavituximab in NSCLC patients is just the latest in a long line of setbacks for novel drug development in this setting. This has been offset by the introduction of a specific value for 4SC-202’s potential in AML (€34m), while retaining a carrying indicative value for 4SC-205 at €10m; previously 4SC-202 and 4SC-205 were assigned €25m together. We have modelled 4SC-202 for its possible use in AML, based on Oryzon Genomics’ LSD1 inhibitor candidate in Phase I/II with a 15% probability and peak sales of €810m, with potential deal parameters at a 50% discount to Oryzon’s partnership with Roche. We use a 12.5% cost of capital and include end-2015 net cash estimate of €19m, offset by €27.8m in discounted operating expenses (Exhibit 8). We note there is near-term upside potential upon EU regulatory clearance and initiation of the resminostat CTCL Phase II trial, which we conservatively risk adjust at 20%, for illustration, raising the probability to 35% upon start of the study would increase the rNPV of this indication to €17m.

The 2015 net cash balance sheet forecast is based on reported cash of €24.5m at 30 September 2015 (boosted by the €27.5m in net proceeds from the July capital increase), and the €1.9m Santo shareholder loan. The capital increase was essential to launch the revised clinical development programme for resminostat in CTCL, with a Phase II study to start in Europe in H116. We note that the major life science fund Wellington Partners invested €5m in the capital increase and is a new cornerstone investor in 4SC (6.6% holding).

As of 30 September 2015 there was €1.9m outstanding on the Santo loan (€1.5m plus 8% pa interest), which is repayable by end 2016; we currently assume this remaining balance will be repaid in cash in 2016, but acknowledge that it may be refinanced or met through the issue of new equity.

FY15 revenues have been lowered to €3.6m to be more in line with the run-rate for the first nine months of 2015 at €2.9m. This is offset by a lower R&D and SG&A expense forecasts for FY15, at €6.9m (vs €8.4m) and €3.2m (vs €3.6m), respectively. 4SC will report its FY15 results on 30 March.

Our financial forecasts for 2016/17 include a significant increase in R&D expenses as the Phase II resminostat study gets underway (€21m over both years), and the model predicts sufficient cash into 2018. The timing and value of potential milestones from Yakult and Menarini in 2016/17 could help improve the cash position and ease the need for fresh funds in 2018. As such, the outcome of the Yakult Phase II HCC study in Japan this year could be significant if a deal and/or further capital injection becomes an option on the back of positive results.