TxCell — Update 28 February 2017

TxCell is one of the few companies focusing on regulatory T-cell therapy (Tregs) for autoimmune and inflammatory indications. During 2016, the company made a big transition into the area of chimeric antigen receptor (CAR) targeting of Tregs, where it has a dominant position following the global licensing of a key 2008 patent. During 2017, four identified CAR Treg projects are expected to progress. A clinical proof-of-concept study in transplantation is planned by TxCell to reach the clinic in 2018 – this will be the first CAR Treg clinical study and so a key world scientific milestone. This could give validation of the CAR Treg concept by H120. TxCell’s antigen-specific Tr1 technology is expected to start a Phase IIb for Crohn’s disease in 2018.

TxCell is in a transition phase as the lead clinical project, Ovasave is on hold while the potentially much higher-value CAR Treg projects are still in preclinical. The value has been refocused onto CAR projects. The Ovasave probability is reduced to 18%, formerly 29.7% giving a total NPV €90m, formerly €170m. The CAR Treg projects are now each given a nominal value. In total, this adds to €30m (formerly €22m. Potential CAR Treg deals are valued NPV at €29m using the acquisition by Celgene of Delinia (a small company with an IL-2 fusion protein to up-regulate Tregs) as a benchmark. The high price, $300m upfront and potentially $475m in milestones was despite its preclinical status. After costs and tax, this gives an NPV of €74m, formerly $76m equal €3.75/share. Assuming full conversion of loans and warrants, the diluted value is €2.83/share.

At year-end 2016, TxCell had €3.5m cash. Management has stated that 2017 cash use will be €13m including a €2m payment to Trizell by late 2017 as part of the 2015 termination of the partnering deal. TxCell raised €11m in February 2017. This indicates year-end 2017 cash of €1.5m. Other funding up to February 2018 could be from the €10.8m of rights issue warrants exercisable at €2.60. TxCell also has further cash funding of €14.7m available to it from 2018 under the convertible loan arrangement. Partnering deals with upfront payments and milestones are expected as CAR Treg technologies develop.

The CAR Treg opportunity is developing well with four clear indications, two of which already have published academic support. As a technology platform CAR Treg (ENTrIA) has a high deal potential and TxCell is planning a clinical proof-of-concept study in transplant to start in 2018 with data by H120. This will be the first CAR Treg study and a milestone in the development of the technology TxCell has a strengthened position until 2028, with a granted European blocking patent. This is still under examination in the US. TxCell may need other CAR technology IP licences. The acquisition of preclinical-stage Delinia by Celgene for $300m upfront and potentially $475m in milestones indicates that the Treg area is viewed as ripe for development, even with only preclinical data. Ovasave, the older Tr1 ASTrIA platform candidate, is on hold but may start a new Phase IIb. However, it might be superseded by CAR Treg products. Finally, although TxCell’s 2017 cash needs are now covered, the source of 2018 and 2019 cash is still not known and may impact on future dilution, although any lucrative CAR Treg deals would add value and minimise share issues.

Regulatory T-cells (Tregs) naturally block autoimmune and inflammatory disorders and control the cell-killing T-cell immune response Singer et al (2014). The Treg area is underdeveloped and TxCell offers a rare investment opportunity, targeting major conditions like Crohn’s disease and, potentially in future, lupus nephritis and other immune indications. TxCell has two technology platforms: ASTrIA and ENTrIA, Exhibit 1. ASTrIA takes T-cells and adapts them naturally in culture to recognise specific antigens, Tr1 cells. ENTrIA inserts a modified TCR gene construct by genetic engineering so that the resulting CAR Tregs are tightly targeted to an antigen. Exhibit 2 shows a diagrammatic version of the technologies. ENTrIA technology runs parallel to, and is overtaking, ASTrIA technology. This note focuses on the background to the CAR Treg projects. TxCell’s pipeline is shown in Exhibit 3.

TxCell has appointed Dr Li Zhou, PhD, as VP of cell engineering. Dr Zhou previously worked for Novartis leading the discovery and engineering of CAR T-cells for cancer immunotherapy.

The concept of CAR Tregs was published in a patent in 2008 by the Weizmann Institute of Sciences, Israel, EP2126054. This patent has been granted by the European Patent Agency and was licensed by TxCell in June 2016. It is in process in the US. This patent gives broad coverage. To provide further layers of extended protection, TxCell is developing other CAR Treg intellectual property, firstly in the molecular technologies used to modify and activate Treg cells, and secondly in the manufacturing and processing technologies needed for commercialisation.

We view the CAR Treg platform as much more flexible and powerful than the established ASTrIA technology. This is because the Tregs can be precisely targeted to any antigen and the transformed cells selected, whereas in Tr1 approaches, a period of antigen exposure is needed and the level of Treg response is not guaranteed. CAR Tregs are therefore a better basis for multiple deals with partners interested in entering the new CAR Treg space. The basics of CAR design, a fast-evolving field in the cancer space, are reviewed in Sadelain et al (2013)).

We note that many aspects of CAR design in the cancer area are the subject of patent litigation, although many cancer CAR technologies may not apply to CAR Tregs. For example, CD28 co-stimulatory domains in ENTrIA (which increase CAR activity, see Zhong et all (2010)) are covered by a Juno patent; however, this expires in 2023 so will probably not affect TxCell. General aspects of CAR technology are patented. We cannot determine if TxCell will need to license other intellectual property and cannot forecast the cost of any licences if required. If licences are required but not obtained, this may delay or prevent the sale of TxCell products and reduce profits.

Three of TxCell’s CAR programmes (lupus nephritis, transplant and BP) already have leading academic collaborators and are therefore well placed to progress rapidly. Clinical trials are still some way off but the first ones are planned, by management, to start from 2018 onwards. More projects are likely to be announced during 2017 and the lead projects should deliver preclinical data as they move towards potential initial clinical evaluation in 2018.

The immune system uses HLA molecules (human leukocyte antigen) to distinguish cells that are healthy and self from infected or non-self (see Exhibit 4; Choo (2007) presents a good overview). The molecules of the HLA system are called the major histocompatibility complex (MHC). These sit on the cell surface and present short peptides (protein fragment) to T-cells, the active white cells of the immune system that patrol and defend the body, and to antibody-producing B-cells.

A Treg that recognises a particular MHC-peptide combination will down-regulate activated T-cells in the vicinity. Note that a normal Treg will not respond to MHCI (acute response) as they only have receptors for MHCII. CAR Treg technology now enables production of Tregs that respond to acute MHCI, like HLA-A2 CAR. Tregs also regulate the longer-term antibody response, Exhibit 5.

Tregs in human blood are a small part of the CD4 cell fraction, a typical adult value is 7.22-7.50%, but the range can be 4–9% so there is a lot of donor variability. Modifying harvested and sorted Tregs with a CAR gene construct and then growing (expanding) the cells is a practical way to get enough targeted cells for therapy. Other methods of Treg isolation often produce polyclonal mixtures, so requiring many more cells per dose as most will be ineffectual against the condition.

TxCell is developing CAR Tregs that recognise a specific antigen trigger. This can be an MHC type, as in transplantation, or a natural protein, as in multiple sclerosis. When the CAR Treg is activated by the signal, it will control any active CD8+ cytotoxic T-cells and conventional helper CD4+ cells in the neighbourhood so down-regulating the immune response.

The use of Tregs to control immune disorders is a well-established concept. Animal models have been tried in GvHD, multiple sclerosis, transplantation, colitis and arthritis; Trzonkowski et al (2015) reviewed the area in detail. Exhibit 6 has a summary of key indications. There is an active EU collaborative group in the area: A FACTT. However, clinical progress has been good but limited. This is because to harvest enough Tregs for therapy from blood is difficult since they need to be sorted to separate them from other T-cell types. Technically, this is possible but sorting at the sterile GMP standards needed for clinical use is very difficult and expensive. Non-specific Tregs target a wide range of antigen targets so will not necessarily have a therapeutic effect. Tregs are also slow to grow in culture.

The main area of research has been GvHD (see above), which can occur after a stem cell transplant to treat haematological cancers like leukaemia and myeloma. As an example, a clinical study (Martelli et al (2014) in 43 adult patients showed that adding 2.5m Tregs/kg to a stem cell graft reduced GvHD without compromising the ability of the granted immune system to attack and control any residual cancer (GvL). In Type 1 diabetes (T1D), a 2014 report (Marek-Trzonkowska et al (2014) in 12 Type 1 diabetic children showed improved levels of endogenous insulin production with two children ceasing to need insulin injections. The dose was 30m Tregs/kg. As a pure autoimmune disease, with serious life-changing consequences, this seems an ideal target for Treg therapies.

There are currently multiple academic sponsored studies running, mainly in GvHD but also T1D, transplantation, Crohn’s disease and multiple sclerosis, see Trzonkowski et al (2015). These tend to be small and slow, 10-20 patients, but can be informative. For example, the University of Bologna-sponsored, EU-backed NCT02749084 in severe GvHD will enrol 20 patients and report in 2020.

Industrial competitors are limited and many are not in the CAR Treg area but instead use either general Treg populations or sorted polyclonal Tregs against a target like an allograft (usually bone-marrow). Others are using mesenchymal or adipose stem cells. Exhibit 7 summarises projects.

Caladrius has a polyclonal Treg concept but only in Type 1 Diabetes so far. The Caladrius website comments that “GvHD, chronic obstructive pulmonary disease, MS, inflammatory bowel disease and steroid resistant asthma” could be further indications.

ILTOO is developing low-dose IL-2. This aims to stimulate Tregs but not other immune cells. This is the area that Celgene (via Delinia) is targeting but that project is still preclinical.

In other cell types, Mesoblast is perhaps the most advanced with an allogeneic mesenchymal lineage stem cell (MSC) approach. It is running a 60 patient Phase III for GvHD using intravenous remestemcel-L at a dose of 2m MSC/kg twice per week for four consecutive weeks. Data is due in H2 2017. Mesoblast has also reported promising Phase II safety and efficacy data in biologic refractory RA patients using a single dose of 1m/kg or 2 m/kg cells.

TiGenix has positive Phase III data (Panés et al. 2016) with Cx601 for the specific indication of complex perianal fistula healing in Crohn’s disease. It used 120m expanded adipose stem cells. This cell type is also being explored in severe sepsis.

TxCell sees the transplant indication as a good starting indication for its CAR Treg programme. Unless a transplanted organ or tissue is a close match for the HLA type of the host, there is always a risk of rejection where the host immune system attacks the transplant. Rejection is controlled by powerful drugs that suppress immune system activity. These drugs create a risk of serious infection. Mismatch means the number of HLA types (see Exhibit 4) that are the same between the donor and recipient. A perfect match is like one form an identical twin – the same genes. As the HLA system is very polymorphic and with six genes to match, it is hard to get exact matches: less than 1,000 in 2015 (Exhibit 8) out of 19,000 kidney transplants.

The exact role of different HLA types on transplant rejection is still not absolutely established. Most transplants are in kidney patients and the biggest group at risk of mismatch are those transplants where the donated organ comes from a deceased donor. This is because the operation is time critical and compromises on HLA matching have to be made given the extent of the waiting list. In the United States in 2016, there were 13,431 deceased donor kidney transplants

For live donors, the best match can be found (often but not always a family member) and the optimal transplant time determined. In the US in 2016 there were 5,629 live donor kidney transplants. Exhibit 8 shows the number of HLA mismatches. Exhibit 9 shows the same data as cumulative frequency. Only 24% of deceased donor grafts had 3 or fewer mismatches compared to 48% of living donor grafts. Nearly half the deceased donor grafts had 5 or 6 mismatches.

The effect of mismatching on transplant rejection is still debated. An analysis of the US transplant registry by Williams et al (2016) found the risk of deceased donor organ rejection was 13% higher with one mismatch and 64% higher with six mismatches. There was no HLA locus that gave a higher risk value. As this study covered nearly 190,000 patients, the dataset appears robust.

Earlier studies, for example Poli et all (1995) found that the HLA-DR locus is particularly important in triggering rejection if mismatched. HLA-DR is associated with the generation of an antibody response. The T-cell activating HLA A and B loci were also found to be important. An 8,000-patient study, Opelz (1985), found that matching HLA-DR and HLA-B in deceased donor transplants gave a 20% improvement in graft survival at one year than graphs with four mismatches. This study found little impact from matching HLA-A.

In lung cancer, a possible first target for the TxCell therapy (information supplied by management), HLA typing might be more important. In a study of the overall survival of 3,549 lung transplant patients, Quantz et al (2000) found that HLA mismatches at the HLA-A and the HLA-DR loci predicted one-year mortality and the total number of mismatches predicted three- and five-year mortality. However, the authors noted that the effect of each covariate was small, with 18% and 15% increases in risk, respectively. Other variables were strong predictors.

In the United States in 2016, there were 2,327 lung transplants – all, of course, from deceased donors. The Scientific Registry of Transplant Recipients data for 2015 show that over 80% of lung transplants had four or more HLA mismatches. Survival of lung transplant patients is not good at about 50% after five years.

It would therefore seem reasonable to assume that the majority of use of a CAR Treg therapy would be within the first year of transplantation to improve overall graft survival. There is a particular need with deceased donor transplants, as these tend to be poorly matched and can have other problems like delayed graft function where the kidney does not produce urine directly after transplantation.

The concept is that if a CAR Treg targets and binds to a non-self HLA in the graft, it will be activated locally and suppress endogenous cytotoxic T-cell helper T-cell (antibody) responses against the transplanted organ. The exact match of CAR Treg type should not matter since any activated CAR Treg should suppress any immune response, although this has to be established clinically.

The evidence that this can happen using CAR Tregs comes from a paper by Professor Levings of the University of British Columbia with preclinical work published (MacDonald 2016). TxCell has established a collaboration with this group to target solid organ transplantation. The Macdonald paper clearly shows that CAR Treg cells can control GvHD. This is, in effect, a model for control of transplant rejection and potentially an indication in its own right. The data show that mice which received HLA-A2 CAR Tregs mostly survived (red lines, one graft failed), while the others died (Exhibit 10A). In Exhibit 10B, the time to onset of GvHD was measured. No GvHD was seen in most of the mice given CAR Tregs against HLA-A2. GvHD appeared within a few days in mice not given CAR Tregs. Exhibit 11 has further analysis.

The clinical challenge will be when to use a CAR Treg therapy in transplantation. Transplant patients after they have received a graft are heavily immune suppressed, so isolation of Tregs to enable CAR Treg manufacture could be more difficult. In our view, the most likely design for a clinical trial would be to take blood from a transplant patient before the operation, process the CAR Tregs and give the CAR Treg cells about one to two months after the transplant.

Post-transplant, high doses of immune-suppressive drugs may hinder the engraftment of a CAR Treg therapy if given some weeks after the graft. One goal is to reduce the level of these toxic drugs to more tolerable levels. A paper by Siepret (2012), albeit in an animal model, indicates this might be feasible. However, a trial that reduced accepted doses of rejection-control drugs would need to find the lowest dose so some patients would experience rejection: that might be a hard trial to design! The common immune-suppressive drugs are now generic and relatively cheap.

The current preclinical project is based on the HLA-A2 CAR Treg discussed above. This will work in about 25% of cases.1 TxCell may need to develop other HLA-type CAR Treg constructs (see HLA discussion in Exhibit 4). If each CAR Treg HLA type were regarded as a new drug, the cost of development would be prohibitive. However, all of them would be orphan indications.

Finally, the design of clinical trials needs to be linked to measurable clinical outcomes. The long-term survival of grafts after a problem-free the first year is very good, so any clinical trial will have to be large in order to show a sustainable long-term advantage. There is a high rate of rejection and poor graft survival in lung transplantations, so any gains in survival would be potentially more easily seen than in kidney. Lung transplants are also done by a few specific centres so a clinical trial could be easy to organise and recruit. This is a possible clinical proof-of-concept indication as a result. However, the major market would be kidney transplant and clinical studies in that indication would be needed.

TxCell has a collaboration with a leading, prestigious European immunology institute, the San Raffaele Scientific Institute in Milan. The collaboration aims to develop a CAR Treg product for lupus nephritis. Currently, TxCell has not disclosed the antigen to which it may develop a CAR Treg construct. We note that there are some HLA associations between lupus and lupus nephritis (Almaani et al (2016)). The reason for this is unknown.

The lupus market has historically been very difficult to assess because the condition is episodic and easily confused with rheumatoid arthritis. In the US, the Centers for Disease Control and Prevention (CDC) published two studies in 2014 (Sommers (2014), Lim (2014)). Translated into patients in the US population of 320 million (source: 2015 US Census estimates), there may be about 160,000 prevalence cases in the US (Exhibit 12). In Europe, we estimate that accessible European markets may have a prevalence of about 170,000.

Of these groups, lupus nephritis can occur at any time, but the cumulative incidence is about 50% in people of African and Asian origin and 14% in Caucasians (Bastian et al (2002)). In between 11% and 33% of lupus nephritis cases, patients progress to end-stage renal disease, which can only be treated through dialysis and a kidney transplant (Mok (2005)). The epidemiology of these diseases is complicated, but an estimate could be for the US about 30,000 lupus nephritis cases, of which 5,000 might have end-stage renal disease (ESRD). The European figure will be lower, perhaps about 17,000 cases, of which about 2,000 may have end-stage renal disease.

At this time, with no data on the product concept, the utility of the product at different stages of the disease is unknown. It should also be noted that this is a prevalence market so treatments in any year will be lower if only one treatment is given; it is possible that repeated dosing will be needed and the cost of ESRD is very high, so this could justify long-term repeat use of Tregs.

A further collaboration with the Lübeck Institute of Experimental Dermatology will develop CAR Treg approaches in bullous pemphigoid (BP). The incidence of BP is poorly established with most data relating to hospital surveys. There is no US population survey. A UK study from 2008, Langham et al (2008), found 4.3 cases per 100,000 with a median onset age of 80, a bias towards females (60:40) and a death risk double that on age-matched controls. This indicates perhaps 12-15,000 European cases and perhaps 12,000 US cases, noting that the US population age structure is younger than the most developed European cases.

The disease is treated with steroids (UK guidelines). A small (86-patient) Canadian study (Heelan et al (2015)) found that the cost of treatment including intravenous immunoglobulin was C$7,000 per month., UK guidelines note that the evidence base for immunoglobulin therapy is very limited, it is not recommended and the cost of regular immunoglobulin is high, over £5,300 per month.

With a possible market of 25-30,000 cases a year indicated, this indication would be a niche orphan product. As Treg therapy is not likely to be cheap, this could still be a significant market and orphan status might limit the development costs. However, the market is likely to be very price sensitive given the age and likely associated morbidities of the patients, and the use of cheap and effective steroids as a first-line therapy. Against immunoglobulin treatment, Tregs could be competitive.

The global multiple sclerosis (MS) market has grown to an estimated US$21.1bn, Exhibit 13. Evaluate Pharma does not forecast any market growth.

The dominant product is a simple organic chemistry reagent, dimethyl fumarate marketed as Tecfidera by Biogen at £1,373 for a 28-day pack, £17,849 a year before discounts (NICE guidance). In the US, the wholesale yearly cost is $55,000, but the drug retails at around $7,200 per month, $86,400/year. Two similar-sized products with similar pricing are Copaxone and Gilenya. By category, beta interferons as a class dominate and effectively set the market pricing. Of the current products, about $11bn come off-patent before 2020. Dimethyl fumarate comes off patent in 2023. All these drugs control or limit progression and relapses but do not enable gain of function.

As an autoimmune disease, multiple sclerosis has been of interest as a possible Treg therapy for some time. A Phase I study (EudraCT number 2014-004320-22 is stated to be underway in Poland using expanded polyclonal Tregs. A recent paper by Schlöder et al (2017) indicates that dimethyl fumarate may improve the response of T-effector cells to Tregs. In that case, adding Tregs to dimethyl fumarate therapy might have a synergistic effect; Biogen is pursuing this with a small Phase IV trial NCT02461069.

For a CAR Treg therapy against MS, a suitable antigen target is needed. A paper from Uppsala University Fransson (2012) showed that in the EAE2 model of multiple sclerosis, CAR FOXP3 engineered Tregs “efficiently suppressed ongoing inflammation leading to diminished disease symptoms”. Interestingly, the therapy allowed regeneration of the myelin sheath in this well-used model so enabling the nervous system function to be regained. This cannot be extrapolated to human clinical outcomes (EAE in mice is not MS in humans), but if the same result was found it would change the way MS is treated. The antigen targeted by Fransson was myelin oligodendrocyte glycoprotein. The Fransson paper used intra-nasal delivery, although whether this is an ideal human route is unknown.

At the moment, TxCell has not disclosed a development plan for this indication, but it would seem to be a major target. Patients and payors are used to long-term therapy and use of repeat doses should not be a major issue. Pricing is also high currently. However, the advent of generics will alter the overall market so a Treg product will need to offer a significant therapeutic gain.

The rights issue which completed in February has given TxCell enough cash for 2017. This section reviews the status of the 2016 funding arrangement with Yorkville Advisors Global (YAG) and the current and potential equity position.

The August 2016 agreement is for up to €20m in cash from convertible loan notes (€0.1m each) plus a further €10m cash if all warrants are issued and exercised, Exhibit 14. A tranche of 30 loan notes (€3m face value) was drawn in August. A further €2m face value of loans was drawn in November 2016. This resulted in the issue of 0.69m warrants in H2. From 2018 to mid-2019, TxCell can draw down further loan notes of up to €15m in face value.

Of the 30 tranches issued in August, 17 (€1.7m) were converted in H2 leaving 13 outstanding. This is shown in our financial estimates as part equity funding and part long-term loan. Of the 20 November loan notes (€2m), all are outstanding and classed as a loan. Under a lock-up agreement, no tranches can be converted for three months following the February rights issue. As a scenario, if all issued but unconverted loans and warrants convert on a current price of €2 less the 7% discount, there will be 2.4m further shares in issue. The warrants would yield up to €2.5m in cash.

The February 2017 funding raised €11.01m gross and issued 5,549,300 shares plus warrants at €2 each. The warrants convert at a ratio of four warrants for three shares with each new share priced at €2.60. Consequently, if fully exercised by February 2018, the 5.5m warrants yield 4,125 new shares plus €10.8m cash. They are freely tradable.

Exhibit 15 shows the origin of the 19.42m shares in issue. If the outstanding loans and warrants convert, there could be 26m shares in issue. Management options currently total 1.87m and are additional to the possible 26m estimated. Note, this assume no further YAG loan drawdowns

The CAR Treg opportunity is developing well with four clear indications. As a technology platform CAR Treg (ENTrIA) has a high deal potential. A clinical proof-of-concept study in transplant to start in 2018 with data by H120 will be the first CAR Treg study and a milestone in the development of the technology TxCell has a granted European blocking patent until 2028; this is still under examination in the US. TxCell may need other CAR technology IP licences. The acquisition of preclinical-stage Delinia by Celgene for $300m upfront indicates that the Treg area is viewed as ripe for development, even with preclinical data. Ovasave, the older Tr1 ASTrIA platform candidate, is on hold and might be superseded by CAR Treg products. Finally, although TxCell’s 2017 cash needs are now covered, the source of 2018 and 2019 cash is still not known and may impact on future dilution, although any lucrative CAR Treg deals would add value and minimise share issues.

TxCell is in a period of strategic transition. The new CAR platform is crucial but preclinical and the clinical phase Ovasave project is on hold: it could renter clinical trials but this depends on strategic priorities in 2018. Accordingly, Edison has reworked the valuation estimate.

This gives a value estimate of €74m, Exhibit 16. Inevitably, this estimate carries a higher degree of variability than normal. However, it fits the structure of the current business and will be revised as the CAR Treg projects progress.

The value per share on the new shares in issue and fully diluted basis is in Exhibit 17.

The indicative core value before conversion of all loans and issued warrants is €3.75/share. Assuming that there is full conversion, the value is then €2.83/share. Note that 0.69m warrants have a five year duration. The rest are expected to convert in the next 12 months.

On 26 January 2017, Celgene announced that it was acquiring a recently funded start-up, Delinia, for $300m upfront with a further $475m in contingency payments and milestones. The Delinia lead preclinical product, DEL106 is a novel IL-2 engineered Fc fusion protein designed to preferentially up-regulate Tregs.3 The price for Delinia presumably reflects, in our view, the goal of a mass-produced, repeat-dose, off-the-shelf therapeutic applicable to major diseases like rheumatoid arthritis. However, all Tregs are stimulated, not just those that are disease targeted. If DEL106 is clinically successful, it could augment infusions of disease-targeted, antigen-specific Tregs. The Celgene deal shows that major companies are prepared to invest in Tregs for the right product concepts with limited validation.

At year-end 2016, TxCell had €3.5m cash. Convertible loans of €5m were taken out in 2016, of which €1.7m converted to equity. The issuing cost was 2% as a cash discount and there were fees estimated at about €225k in value paid as 77.7k of shares. A fee of €0.3m is due on commission for investor commitments on the rights issue. TxCell has indicated that operational cash use will be about €11m in 2017 as no clinical trials are planned. TxCell is due to pay a further €2m by late 2017 as part of the termination in 2015 of the partnering deal with Trizell. The 2017 cash outflow will be about €13m. Year-end 2017 cash should be about €1.5m. Financial estimates are in Exhibit 18.