Celyad — Update 12 November 2015

Celyad has two core competences: autologous cell therapy development (in cancer and cardiac indications) and cardiac medical devices. Allogeneic therapies are being developed with a granted US patent in the area. Projects at clinical or near-clinical stage are shown in Exhibit 1. Device projects are discussed in previous notes.

In January 2015, Celyad bought the OnCyte business. OnCyte has rights to NKG2D CAR-T technology for cancer therapy from the group of Professor Charles Sentman at Dartmouth College, US. A product, CAR-T NKG2D, is in a 21-patient Phase I US clinical study NCT02203825. This is currently in its initial safety phase, with the first cohort of three patients having completed their three-month safety follow ups after an infusion of one million cells. There are a further three possible intravenous dose levels planned. Cohort 2, now underway, will receive three million cells, Cohort 3 10 million cells and Cohort 4 30 million cells assuming no limiting side effects are noted. The dose used for an expansion of the study will probably be either Cohort 3 or 4 depending on the maximum tolerated dose. Patients can have either acute myeloid leukaemia, myelodysplastic syndrome (the AML precursor) or multiple myeloma.

There are a number of patents covering NKG2D CAR T-cell therapy. Of these, granted patent US7994298 B2 (priority 31 August 2005) discloses the core discovery. Note that there are general concerns on overlapping patents in the CAR area and not much IP has yet been granted.

The newly granted US patent US9181527 was filed on 30 April 2012. The US government has asserted some rights to this patent as it provided grant funding. This patent is crucial as it protects a method for modifying T-cells, which can then be cultured to produce lines of allogeneic CAR T-cells.

A major obstacle to widespread CAR T-cell therapy is the current need to use the patient’s own immune cells: autologous therapy. Cells must be harvested from the patient, transported to a centralised facility, the NKG2D genes inserted, the altered cells cultured in sterile conditions and then rigorously tested before they can be sent back to the hospital for infusion into the patient. All this is an expensive process and takes weeks. In addition, it may not work for all patients. Immune cells from patients with advanced cancer, who may have had chemotherapy, often do not respond well to laboratory manipulations and culture. Cancer patients are often elderly and may suffer from other diseases that may limit the potential of their immune cells to be used in CAR-T-cell approaches. These factors mean that autologous CAR-T-cell therapies are inherently variable and expensive. A reliable, allogeneic therapy that can be sourced quickly, used ‘out of the freezer’, at a lower cost with more consistent results is therefore desirable on medical grounds and valuable commercially as it vastly expands the potential market.

However, using T-cells from another individual could result in them attacking healthy tissues in the host and triggering a generalised immune response. This is called graft vs host disease and is a difficult condition to manage requiring immune suppression. Celyad acquired rights to the patented work of Professor Sentman when it acquired the OnCyte business. The method disclosed in the granted patent is termed TCR Inhibitory Molecules (TIMs). TIMs are in preclinical development.

T-cells, the killer cells of the immune system, recognise non-self-cells by binding them with T-cell receptors (TCR). TCRs self-assemble in the membranes of T-cells when several different component proteins come together. Professor Sentman discovered that TIM technology could prevent the TCRs assembling or functioning correctly. Some functional TCRs may still form, but there must be a minimum level of functional TCRs for a T-cell to be activated to destroy its target cell. Professor Sentman used various viral vectors to insert the TIM genes into T-cells. Selecting the best viral vector and negotiating any required licences will be necessary for full development.

TIMs are either short hairpin RNA or dominant negative proteins (Exhibit 2). Both require new genes to be inserted into the T-cells, normally as part of the NKG2D CAR gene construct used to target the T-cells to the cancer. In development, Celyad will need to confirm that TIM genes are stable and remain active during the cell culture process.

C-Cure has been extensively reviewed (2014 outlook note); trials are summarised in Exhibit 3.

The CHART-1 Phase III is due to reach its primary data point in April 2016, with headline results (after analysis) due in mid-2016. The endpoint is a composite measure of heart function. There are two secondary follow-up endpoints at 52 and 104 weeks looking at efficacy and safety.

The CHART-2 study could start in late 2015 meaning that dosing should start in H116. The trial is scheduled to take about 30 months to recruit and run, so an H218 result is possible. This is an FDA-approved, part-US trial with a six-minute walk primary endpoint; this is simpler that the CHART-1 endpoint. The trial is delayed while the FDA considers whether to allow C-Cathez use in CHART-2. As the dose range is believed to be very wide, enough cells can, in theory, be delivered by existing catheters. However, it would give comparability to CHART-1, to use C-Cathez.

In 2014 Celyad and Medisun, a new company, set up a JV to develop C-Cure in the Chinese and other Asian markets. This agreement was terminated in August 2015 and a €60k loss recorded. A new agreement was then put in place, under which Medisun will still fund the Chinese trials at a cost of €20m but Celyad will organise and run them, mostly in Hong Kong, and seek regulatory approval. The royalty to Celyad will range from 10% to 30% based on the total revenues of C-Cure. In addition, there will be a profit share of 20-25% based on total revenues less royalties. This agreement will last for an initial period of 15 years. Clinical material will be prepared in Belgium.

Celyad has diversified its risk with major cardiac and cancer programmes based on its autologous cell competences. Both Phase III CHART studies are fully funded. Valuations are sensitive to the changed $/€ rate as the US is the most valuable market.

CHART-1 data in mid-2016 remain the major sensitivity. C-Cure has the potential to be the first cardiac regenerative cell therapy from 2017 and could create a major new market with direct EU sales. The clinical benefit will need to justify a high price for C-Cure as autologous therapies are expensive to manufacture. Hence, commercial success is not necessarily automatic. The C-Cure Phase II offers support, but was not an exact parallel to CHART-1 as the Phase III uses a complex hierarchical endpoint. Use of another catheter adds uncertainty to the outcome.

The CAR programme opens a major long-term opportunity. The initial indications of AML and MM have clear market opportunities. Progress appears slower than we expected but the first cohort has now completed safety follow up so the project is progressing; these therapies always start slowly for safety reasons. A move into sold tumours would be a major opportunity The CAR sector may become the major value driver for the business. Celyad is setting up the US infrastructure required. Adding allogeneic versions of CAR products through TIM technology should cut costs, make them easier to use and extend biological exclusivity. Celyad could also license TIM technology to gain value and perhaps to access other technologies itself.

Values, based on risk-adjusted sales to 2030, have been adjusted to the $/€ rate of 1.1 with 9.31m shares as of August 2015. US sales of C-Cure are now assumed to start in 2020 after regulatory review in 2019. Overall, these adjustments leave the indicative fair value at €99 per share, formerly €98. This is based on valuing Celyad on the basis of C-Cure and then adding in additional cash flows from the CAR projects using AML and MM alone. Direct funding for trials by Celyad allows most of the profits to be retained after marketing costs. A 50:50 US marketing joint venture is assumed in the US with direct sales in Europe. The CHART-1 probability remains at 45%. The US success probability remains at 30% due to uncertainty over C-Cathez, unchanged from the last note. CAR products are at 15% Phase I probabilities; as yet, little Phase I information has been released.

Adding in up to six solid tumours may generate a further €514/$566m of additional value. As the solid cancer indications are still uncertain, they are not included in the valuation estimate. Allogeneic versions of these products (not specifically valued as too early in development) could have longer biological protection and should greatly expand sales. There may be possible licensing fees from the granted TIM patent (not forecast).

Celyad’s cash use in 2015 and 2016 depends on the range of clinical trials undertaken and will also be affected by short-term data, for example CAR therapies and solid tumours. Cash at 30 June was €124m after €109m in cash from financing activities in H115. Local funding in the US is supporting the establishment of the new manufacturing facility in Rochester and Medisun pays the €20m cost of the Chinese trials. We have not assumed any significant C-Cure marketing investment in 2016, but a successful CHART-1 study would need pre-launch investment. Year-end 2015 cash is expected to be around €107m, depending on clinical trial investments and working capital; creditors rose by €2.4m to €6.4 at the interim stage, but a grant from 2014 worth €1.1m was received. The US listing has effectively hedged the US currency costs of its cancer programme and CHART-2 cardiac trial. Cash burn in 2016 will depend on the outcome of the CHART-1 study (for example, if pre-marketing and C-Cure production investments need to be made) and on the extent and pace of any solid cancer CAR therapy trials. Our financial forecasts, revised following the interims, are shown in Exhibit 5.

Note that a large contingent payment liability of €36.3m fair value in respect of Oncocyte. US$50m of this would be due in development and regulatory milestones on NKG2D CAR products, plus a further US$21m on other products. There are also sales-related milestones of US$80m and royalties of between 5% and 8%. The acquisition has been subject to purchase price allocation, so all assets and liabilities are booked at fair value. The intangible assets have risen accordingly to €56.8m, of which Oncocyte comprises €46.9m.