TxCell — Update 31 May 2016

TxCell — Update 31 May 2016

TxCell

Analyst avatar placeholder

Written by

TxCell

Two valuable and versatile platforms

Initiation of coverage

Pharma & biotech

31 May 2016

Price

€5.04

Market cap

€65m

Cash (€m) at 31 March 2016

5

Shares in issue

12.9m

Free float (Dec 2015)

22.4%

Code

TXCL

Primary exchange

Euronext Paris

Secondary exchange

N/A

Share price performance

%

1m

3m

12m

Abs

(9.2)

0.8

(8.7)

Rel (local)

(10.9)

(3.3)

0.7

52-week high/low

€10.00

€4.21

Business description

TxCell is a pioneer in developing regulatory T-cell immune therapies against autoimmune and inflammatory disorders. The lead product in Crohn’s refractory disease is due to restart Phase IIb in mid-2016. A novel CAR Treg technology platform is in early development.

Next events

H1 FY16 results

September 2016

Analyst

Dr John Savin MBA

+44 (0)20 3077 5735

TxCell is a research client of Edison Investment Research Limited

TxCell offers a rare investment opportunity in the regulatory T-cell (Treg) area with major potential in inflammatory and autoimmune disorders. TxCell is restarting the Ovasave Phase IIb study in refractory Crohn’s disease; data is due by early 2018. A flexible CAR Treg platform is being actively developed with a leading academic partner to address possible markets like lupus nephritis and perhaps multiple sclerosis. The indicative value is €7.08/share before further funding of an expected €27m over 2016-17. By 2018 a scenario value, given good Phase IIb Ovasave data and partnering plus CAR progression, could be over €300m.

Year end

Revenue (€m)

PBT*
(€m)

EPS*
(c)

DPS
(c)

P/E
(x)

Yield
(%)

12/14

1.39

(8.7)

(82.6)

0.0

N/A

N/A

12/15

1.61

(10.8)

(87.4)

0.0

N/A

N/A

12/16e

0.0

(14.5)

(118.3)

0.0

N/A

N/A

12/17e

0.0

(18.3)

(148.8)

0.0

N/A

N/A

Note: *PBT and EPS are normalised, excluding amortisation of acquired intangibles, exceptional items and share-based payments.

Antigen-specific Tregs: Phase IIb underway

The lead product, Ovasave, uses an ovalbumin (egg white) trigger to activate regulatory T-cells. Once activated, these autologous, cultured cells are intended to control the inflammatory response in Crohn’s disease. The target market is about 100,000 patients who have failed on biological therapy. The Phase IIb (CATS29) is approved to restart with the redesigned protocol. Manufacturing has been transferred to MaSTherCell, a Belgium CMO. An efficient manufacturing system is being developed to obtain a commercial cost of goods and speed delivery times.

CAR Treg: Flexible and powerful platform potential

TxCell is developing a chimeric antigen receptor (CAR) regulatory T-cell platform. This could be an excellent basis for partnering and technology licensing; high deal values of over €300m are seen in the related CAR cancer area. TxCell has started to collaborate with the leading San Raffaele Scientific institute in Milan to develop a first CAR Treg product in lupus nephritis. An academic patent application is under option but not critical to development of the platform as TxCell is filing further patents.

Valuation: CAR Treg is the future but Ovasave crucial

The valuation of TxCell depends on Ovasave as the sole clinical-stage project. The overall probability is set at 28%, which is a combination of 33% clinical and 85% manufacturing probabilities. A core assumption is that a partner to fund Phase III is found in 2018 in a €175m deal plus 16.5% royalties. A niche orphan product for uveitis (eye inflammation) is in preclinical with a 7.5% probability but with direct sales potential. The CAR Treg platform could become the main value driver due to its versatility and deal potential and is assigned a nominal value of €20m; lupus nephritis in the initial, research-stage indication. This gives a prefunding value of €7.08/share with funding of €27m (net of current cash) until late 2017 needed. If CAR progresses and Ovasave enters Phase III with a partner in a deal worth at least €175m in 2018, the indicative market value could rise to over €300m.

Investment summary: Valuable and versatile platforms

Company description: Nice regulators

TxCell is a pioneer in regulatory T-cell (Treg) therapy and so offers a rare investment opportunity in this emerging technology space. Treg cells prevent harmful activity by the immune system so can be used, in concept, to control multiple chronic autoimmune and inflammatory diseases like Crohn’s disease, multiple sclerosis and lupus nephritis. If the resumed Ovasave Phase IIb delivers a clear and positive primary endpoint success by early 2018, this should validate the general concept of using Tregs to control autoimmune disease and inflammation, two major markets with remaining unmet medical needs. Management aims to enter platform and preclinical Treg deals in 2016 and 2017. The company is based in the south of France near Nice and has 50 employees. TxCell has raised €66m of capital including private equity; this was restructured in 2014 and is now reported at €32.5m in the 2015 accounts. The 2015 accounts show a cumulative loss of €20.9m.

Valuation: Based on Treg platform capability

TxCell’s value depends largely on Ovasave as the sole clinical-stage project. The overall Ovasave probability of success is 28%, which is a combination of 33% clinical and 85% manufacturing process probabilities. A core assumption is that a partner to fund Phase III is signed in 2018 in a €175m+ deal plus 16.5% royalties. An orphan product for uveitis (eye inflammation) is in preclinical with a 7.5% probability but with direct sales potential. The CAR Treg platform could become the main value driver due to its versatility and deal potential and is assigned a nominal value of €20m; lupus nephritis is the lead, research-stage indication. This gives a prefunding value of €7.08/share. If CAR progresses and Ovasave enters Phase III with a partner in a deal worth at least €175m in 2018, the indicative market value could rise to over €300m. TxCell also plans to enter at least one preclinical or platform deal per year from 2016.

Financials: Adding value needs additional funding

At 31 March 2016, TxCell had €5m cash with a €3m tax credit due. The 2015 core operating cash costs were €12.3m. Management has stated that TxCell expects a €15m cash burn over 2016. Cash expenditure will rise in H2 as the Phase IIb trial restarts; this study is costed by management at €15m over three years, with €4.5m assumed by Edison in 2016. There is investment into the CAR Treg programme, which should steadily rise as the platform develops.

Sensitivities: Expanding out of the niche

The major sensitivity relates to the resumed CATS29 Phase II and TxCell’s ability to fund the €15m study and ongoing R&D including the important CAR Treg cell platform. French tax credits are expected to pay about 30% of each year’s cost; this cash is received in the following year. The Crohn’s market is complex with uncertainties in patient numbers and significant regional market differences. A core value assumption is that an Ovasave partner is found by the end of 2018 at a significant overall deal value of at least €175m with at least a €25m upfront payment. The project may not progress into Phase III unless a deal occurs. Steroid resistant Uveitis (eye inflammation) is a niche market with a candidate in late preclinical with no trial timeline announced. TxCell plans to sell directly, capturing extra value, but will need to fund the trials. Finally, the new CAR Treg opportunity appears to be potentially applicable to many indications but as a technology platform it is hard to value and is at least three years from a clinical trial. TxCell patents are now being filed. There is no guarantee that an academic blocking CAR Treg patent application under option till June 2016 will be secured.

Initiation: Regulatory T-cell specialist

Regulatory T-cells (Tregs) naturally block autoimmune and inflammatory disorders and control the cell-killing T-cell immune response. The Treg area is underdeveloped and TxCell offers a rare investment opportunity, targeting major conditions like Crohn’s disease and, in future, Lupus through direct immune control. TxCell has two technology platforms, Exhibit 1.

TxCell originated in 2001 as a spin-out from INSERM1 to develop Tr1 Regulatory T-cell technology; this was discovered and published in 1997. The company ran a Phase I/IIa in patients with Crohn’s disease over 2008-2010 with Ovasave; established a manufacturing site in Besançon (near the Swiss border) in 2013; did a pre-IPO deal with Ferring (later transferred to Trizell) in early 2014; and started the Ovasave Phase II in late 2014. TxCell completed its IPO on Paris Euronext on 11 April 2014 at €5.58 per share raising €16.2m gross in cash and converting €3.5m of bonds to shares.

INSERM stands for Institut National de la Santé et de la Recherche Médicale.

TxCell was restructured from Q215 after Stéphane Boissel was appointed as CEO. Over 2015, TxCell paused the Phase II, closed the Besançon site, contracted out manufacturing to MaSTherCell, and bought out the Trizell constraining partnering deal. In November 2015, an option was signed (expiring at the end of June 2016) to license an original 2008 patent application on CAR Treg technology; note that TxCell is developing this regardless but the patent would give a good blocking position over the CAR Treg area generally.

The investment case for 2016 is consequently on a revised basis relative to the 2014 IPO. The company has regulatory approval (May 2016) to restart the CATS29 Phase IIb on Ovasave using a redesigned protocol. It is developing a commercial manufacturing process; plans to progress a second product into clinical development, and, ideally, sign a preclinical/platform deal by the year end. Further funding is required for the CATS29 trial and CAR technology platform development.

Exhibit 1: Technology platforms

Acronym

Technical basis

Comments

ASTrIA

Tr1 antigen-specific regulatory T-cells (Ag-Treg). The lead product, Ovasave, is activated by the antigen ovalbumin (the major egg white protein)

Antigen Specific Treg for Inflammation and Autoimmunity (ASTrIA) is an autologous cell technology where T-cells are harvested from blood, exposed to an antigen and the resulting antigen-specific Treg cells cultured and then infused back into the patient. The process takes about 12 weeks and is still relatively expensive. Automation and refinement of the process is underway. This aims to reduce the time required to five weeks and to lower the cost of goods. Manufacturing is done by MaSTherCell, a specialist Belgium based company.

ENTrIA

Chimeric antigen receptor regulatory T-cells (CAR-Treg)

Engineered Tregs for Inflammation and Autoimmunity (ENTrIA) is a TxCell platform being developed for cellular immunotherapy. The approach is that autologous T-cells are removed from the patient and genetically modified using a virus. The technology is similar to the CAR T-cell approach being developed in cancer but using regulatory cells to target autoimmune and inflammatory disorders. The concept was published in 2008 by the Weizmann Institute of Sciences, Israel; a patent application from the Weizmann is under option by TxCell until June 2016.

Source: Edison Investment Research based on TxCell statements and literature sources

Antigen-specific technology: Building the Tr1 pipeline

The immune system needs to protect against pathogenic bacteria and viruses in the intestine but also needs to tolerate the many benign bacteria and food components as a reaction to these stimuli can cause intestinal damage. In 1997, Groux et al published a paper in Nature on a new type of regulatory T-cell.2 Asseman 1998 provides a good commentary. The area has been recently reviewed by Zeng (2015).

At the time, Groux was working for a Californian biotechnology company, DNAX.

Groux showed that naïve CD4+ T-cells exposed to a specific food antigen produced a subset of T-cells, which he named T regulatory cells type 1 (Tr1). These prevented effector T-cells from responding to specific foreign food antigens so causing a pathological response to the intestine. Tr1 cells therefore have a crucial antigen-specific protective effect Tr1 cells are characterised by production of very high levels of IL-10. The high IL-10 levels stopped other T-cell cell types (Th1, Th2) from developing. Groux et al used ovalbumin as the antigen. Antigen presenting cells (APCs) ‘present’ specific antigens to naïve CD4+ T-cells so they become Tr1. Tr1 can also protect against autoimmune diseases for example Pot 2011.

As a natural cell, Tr1 cannot be patented but the process of generating Tr1 cells against a specific antigen can be protected. TxCell holds a granted process patent, EP1739166B1 (expiry 1 July 2025), and has filed a patent application, WO 2009068575 A1 (expiry 28 November 2028). The patent claims “at least one human Tr1 cell population directed against a food antigen from common human diet”. The ASTrIA method involves harvesting leukocytes from human blood and stimulating them with antigen. For Ovasave the antigen is ovalbumin. The resulting ovalbumin-specific Tr1 cells are then cultured, harvested, purified and frozen in single dose aliquots. A blood sample of 150ml is stated to produce between 30 and 50 doses: about a three-year supply. The cells are autologous, that is they are injected back into the donor, who is also the patient.

The method described in these two core patents can be applied to a variety of antigens. The pipeline is summarised in Exhibit 2. The immediate focus is Ovasave in inflammatory bowel disease (Foussat 2003) specifically Crohn’s. There is also published preclinical work using the ovalbumin model on central nervous system inflammation (Barrat 2002).

Exhibit 2: TxCell pipeline

Product

Indication

Stage

Timelines

Comments

Ovasave

Refractory Crohn’s

Phase IIb

Autologous Tr1 cells, in a 56-patient randomised, two-arm placebo-controlled European study. Restart by mid-2016, data Q118.

Regulatory approval received for the amended study protocol with MaSTherCell as the manufacturer. The open IND in the US could allow a US centre to participate. The trial is over 32 weeks per patient; the first six weeks are double blind.

Col-Treg

Uveitis

Preclinical

Start 2016-2017

Tr1 cells against Type II human collagen.

CAR Teg

Lupus nephritis

Research

Clinical trials possible from 2018-19, multiple possible products inc multiple sclerosis

CAR Treg platform being developed. Lupus nephritis, a severe complication of lupus, is the designated lead indication.

Source: Edison Investment Research based on TxCell corporate presentation January 2016

Mechanism of action of antigen specific Tr1

Exhibit 3 shows possible localised mechanisms of action (MoA) suggested by TxCell. As with most cell therapies, the cells probably exert an effect through multiple routes.

Exhibit 3: Potential mechanism of action

Mechanism

Notes

IL-10 secretion

The Tr1 cell type naturally expresses a high level of Interleukin -10 (IL-10). IL-10 is a potent anti-inflammatory cytokine that generally reduces immune activity. IL-10 was being developed by Schering-Plough as Tenovil (Ilodecakin) but was discontinued as trials showed that systemic IL-10 had no efficacy in Crohn’s disease, although it did help to prevent the disease re-occurring once it was controlled; Marlow 2013 has reviewed the area. Most Crohn’s patients have normal to high IL-10 levels. Scientific work has focussed on localised IL-10 as the key immune suppressing agent, for example Pot 2011.

IL-13 secretion

Tr1 cells express high levels of this cytokine. It is closely related to IL-4 so might stimulate Tr1 growth. There is no known association between IL-13 and Crohn’s disease (Biancheri 2014). IL-13 appears to have some anti-inflammatory proprieties, similar to IL-4, but this role is not well defined. IL-13 is known to be associated with allergic lung disorders, asthma and resistance to parasitic disease. Watson (1999) observed anti-inflammatory activity with suppression of eosinophil recruitment in an antigen-challenge Guinea Pig lung model.

CD39

CD39 is a membrane bound enzyme found on various cells including Tregs. It catalyses the conversion of extracellular ATP to AMP. An associated enzyme, CD73, then converts AMP to adenosine, which is anti-inflammatory. ATP is released in inflamed locations either directly by cells or as a result of cell lysis and promotes the inflammatory response. There is some literature on the role of CD39 in inflammatory bowel disease, for example Antonioli 2014. Friedman 2009 found that lack of CD39 increased susceptibility to inflammatory bowel disease. Gibson 2015 in a small clinical study (of whom 25 were Crohn’s patients) noted an inverse association between CD39 levels and disease severity with 8% showing CD39 at peak disease vs 22.5% in remission; this was statistically significant.

Source: Edison Investment Research, TxCell reports

The concept behind the use of ovalbumin is that, as a common food antigen, it will be frequently present in the intestine and some will be absorbed by special immune areas in the intestine wall called Peyer’s patches. These are specialised structures in the intestine wall that take in antigens to induce tolerance (food) or activate an immune response (pathogens). Antigens are passed to the mesenteric lymph nodes that protect the body from infection originating in the intestine.

For efficacy, ovalbumin has to pass through the intestine lining, be taken up and processed by APCs and then displayed as fragments in MHCII to stimulate the Ova-Treg cells to exert some form of localised immune regulation. The Ova-Treg cells need to migrate from the administration site to the intestine lining and the intestinal mesenteric lymph nodes. Ovalbumin does not affect the migration but T cells migrate to areas of inflammation due to other cytokine signals.

To get regular ovalbumin in the diet, the Phase I/IIa patients “ingested an ova-enriched diet in the form of a daily meringue cake”. Note that there is no clinical meringue dose response curve but that animal studies indicate that only a trace amount is required. A normal diet may be adequate.

It would seem likely that patients have frequently been exposed to ovalbumin and may already have endogenous Ova-Tregs. Native ovalbumin is resistant to proteolytic cleavage in the stomach (Walker 1978) and in mouse models, high levels of radiolabelled ovalbumin pass along the small and large intestines (Oliveira 2007). Oliveira et al also showed that ovalbumin locates to Peyer’s patches and moves to mesenteric lymph nodes. This is important for a systemic effect since Crohn’s disease commonly occurs in the lower small intestine and large bowel and rectum whilst most intact ovalbumin in the intestine will be in the upper bowel. Cooking of ovalbumin, as in meringue, exposes new proteolytic sites in the protein to enzyme action (Nyemb 2014) and bile salts and lipids in the duodenum (upper small bowl) increase digestion (Martos 2010). Golias et al (2012) found that cooking altered ovalbumin digestion and allergic response.

Ovasave Phase I data

The Ovasave Phase I ran between 2008 and 2010 and was published in 2012 (Desreumaux 2012). Details and comments on the study are in Exhibit 4.

Exhibit 4: Phase I design and outcomes

Aspect

Notes

Doses tested

.

The trial tested four single dose levels of autologous ova-Treg cells: 106, 107, 108 and 109.

Preclinical dose data

Groux 1997 noted in a mouse model that 2x105 cells prevented inflammatory bowel disease. A lab mouse is about 20 grams in weight: 3,500-fold difference to a 70kg human. Hence, one would expect a human dose of about 108-109 cells. We note that mice are very different to humans. For a rough dose comparison, in the TiGenix Phase III in fistulating perianal Crohn’s disease, the effective therapeutic dose was 1.2x108 cells injected into a highly concentrated zone, although this used different cells with a different indication and use.

Cohort sizes

Planned

The initial cohort had six patients with planned three per cohort then an extra six at the highest dose. The study was open label. That is normal for an early stage dose and safety study especially for cell therapies.

Actual

As the best dose seemed to be 106, this cohort was increased from six to eight. The 107 and 108 groups had three patients, with six in the 109 group.

Patient response to biological agents

Most (19/20) patients had failed on at least one anti-TNF therapy; in the eight patient 106 dose group, five (62.5%) had failed on three and the other three had failed on one.

Efficacy assessment

Patients were assessed using the Crohn’s Disease Activity Index (CDAI, see below) at weeks one, two, three, five, eight and 12. Response was a decrease in CDAI of greater than 100 points with remission defined as a score of 150 or less.

Biomarkers used

(see Chang 2015)

C reactive protein (CRP)

CRP is a marker of general inflammation. CRP is produced by the liver in response to inflammation and detected by a cheap blood test.

Calprotectin

Calprotectin is a protein released by activated neutrophils at an inflamed site. In inflammatory bowel disease it can be conveniently detected in faecal samples. Faecal calprotectin is specific to inflammatory bowel disease although levels do not necessarily correlate with clinical disease scores.

106 dose outcome

The data showed that 75% (6/8) of the 106 dose group showed a response, with three entering remission by week 5 and two remaining in remission at week 8. One patient, with a CDAI score of over 500, did not show any response at all (and did not provide Week 8 data). Note Crohn’s is a variable disease.

At five weeks (not prespecified), helped by an erratic-looking one-off score of 50, the average decrease was 143 points with a standard deviation of ±105 (the 95% confidence interval was between +63 and -349).

This was statistically different to the baseline score, p= 0.039 using the Wilcoxon signed rank test.

CRP levels showed high variability over the 12-week trial and there was no effect on calprotectin levels.

Higher dose outcomes

One patient in each of the 107 and 109 dose groups responded. No patient responded at 108. Immune products often have a low dose maximum as any strong immune responses are damped down by regulatory mechanisms.

Source: Edison Investment Research based on Desreumaux 2012

The Phase I and II studies use the CDAI score as an endpoint (Exhibit 5). This was devised in 1977 (Best 1976) and has been widely used since. The system has no specific upper limit, but a score under 150 points is deemed remission, and over 450 points is severe disease. Crohn’s is a condition subject to sudden flares then remission, so variability is expected. Note that the data is not controlled and could be subject to small number effects (in a small group, one or two outliers make a big difference to the average; in a large group, such outliers are averaged out and we see reversion to the mean).

Exhibit 5: Crohn’s Disease Activity Index (CDAI)

Criteria

Subgroup criteria

Multiplier

Number of liquid or soft stools each day for seven days

x 2

Abdominal pain (graded from 0-3 on severity) each day for seven days

x 5

General wellbeing, subjectively assessed from 0 (well) to 4 (terrible) each day for seven days

x 7

Presence of complications, Score one for each of:

x 20

the presence of joint pains (arthralgia) or frank arthritis

inflammation of the iris or uveitis

presence of erythema nodosum ((inflation of the fat deposits underlying the skin), pyoderma gangrenosum (ulcerating autoimmune skin disease) or aphthous ulcers (mouth ulcers / cankers)

anal fissures, fistulae or abscesses

other fistulae

fever during the previous week

Taking Lomotil or opiates for diarrhoea

x 30

Presence of an abdominal mass (0 as none, 2 as questionable, 5 as definite)

x 10

Haematocrit of <0.47 in men and <0.42 in women (haematocrit is the proportion of red cells in blood)

x 6

Percentage deviation from standard weight

x 1

Source: Best 1976

Exhibit 6 shows the average scores for the 106 group – the optimal identified dose, but this conceals some very erratic scores. A simplified per patient data plot is in Exhibit 6. The data in Exhibit 5 was used to set the Phase IIb endpoint at Week 6, with the assumption that efficacy declines thereafter as the ova-Treg cells die.

An issue with the CDAI is that it does not provide a measure of inflammation such as CRP. CDAI has been critiqued as a marker for testing biological therapies, for example Binion 2010.

Exhibit 6: Average response 106 dose

Exhibit 7: Individual patient responses, 106 dose group

Source: Desreumaux 2012. Note: Errors are one standard deviation.

Source: Edison Investment Research, Desreumaux 2012

Exhibit 6: Average response 106 dose

Source: Desreumaux 2012. Note: Errors are one standard deviation.

Exhibit 7: Individual patient responses, 106 dose group

Source: Edison Investment Research, Desreumaux 2012

Ovasave Phase IIb: CATS29

The CATS29 study is a randomised, double-blind, placebo-controlled study to evaluate the performance of a single 106 cell dose of Ovasave in refractory Crohn’s disease patients over six weeks, Exhibit 8. It is designated Phase IIb by TxCell. A second, open-label phase of the trial is then planned to provide additional safety data and possibly some uncontrolled data on efficacy and responses. Adding this open-label study where all are treated is designed to aid recruitment. The redesigned study protocol has regulatory approval and will restart soon with data in Q118.

The role of the CATS29 Phase II is to get enough clinical evidence to gain a strong partnering deal. The weakness of CATS29 is that it only tests a single dose and at only one dose level. Patients are not supposed to take other therapies but this might be hard to control and will result in them being counted as therapy failures if detected.

MaSTherCell, a contract manufacturing organisation (CMO) based in Belgium, was contracted to supply Ovasave during 2015 and technology transfer is underway. Regulatory approval of this move, expected by mid-2016, is needed before CATS29 can restart. A five-year strategic agreement was signed in December 2015 so that MaSTherCell will supply all TxCell cell products.

In July 2015, TxCell gained an FDA fast track and open IND for Ovasave, so it could add a US centre into the CATS29 study. PCT, a Caladrius subsidiary, was appointed in March 2016 as TxCell’s contract manufacturing partner in the US. It is likely that some US patients in one or two centres will be enrolled in CATS29 when it restarts.

Exhibit 8: Design of CATS29 Phase IIb study

Aspect

Comments

Initial design

The initial design tested three dose groups plus placebo and required 160 patients. The clinicaltrials.gov entry, NCT02327221, will be updated once the new design is approved. TxCell dosed fewer than 10 patients up to June 2015.

Patient selection

CATS29 patients have to score ≥250 points on CDAI to enter the study. These patients must have failed on at least one biological therapy and would be unresponsive to standard immune suppressing therapies.

Current design

In the new design, 72 eligible patients give 150ml of blood. This is processed into Tr1 ovalbumin specific antigen cells (Ova-Treg) in a process that currently taskes12 weeks. TxCell has to over recruit as patients need to meet the criteria of CDAI ≥ 250 (see below) with positive CRP and calprotectin markers both on enrolment and about three months later on dosing. Inevitably, some patients will fall below the 250 CDAI level, since Crohn’s is an episodic disorder, while their cells are being prepared so will not be dosed. In addition, not all patient blood samples will produce Ova-Treg cells in sufficient quantity or quality. TxCell requires 56 assessable patients for the endpoint.

Stage one

The initial six-week phase is a double-blind, placebo-controlled evaluation of a single 106 cell infused dose vs placebo The endpoint is a response based on the CDAI score assessed at week 6.

Primary endpoint

A response is defined as a reduction of ≥100 points in the Crohn’s Disease Activity Index (CDAI, Exhibit 5). TxCell expects a response difference of 70% treated vs 30% placebo.

Power

If all 28 patients per arm complete the study, the power is expected to be 80%. Statistical power is the probability that the clinical trial will have a significant (positive) result of p≤0.05 assuming that the real difference between treated and placebo is as anticipated. In other words, if Ovasave works, there is an 80% chance that this trial will be positive.

Stage two

This is an open-label phase in which all patients receive 106 cells at weeks 8, 16 and 24 with final assessment at week 32. There is a three-year safety follow up as this is a cell therapy product. This stage is designed to be attractive to patients and to aid recruitment since all will receive the stem cells.

Aspect

Initial design

Patient selection

Current design

Stage one

Primary endpoint

Power

Stage two

Comments

The initial design tested three dose groups plus placebo and required 160 patients. The clinicaltrials.gov entry, NCT02327221, will be updated once the new design is approved. TxCell dosed fewer than 10 patients up to June 2015.

CATS29 patients have to score ≥250 points on CDAI to enter the study. These patients must have failed on at least one biological therapy and would be unresponsive to standard immune suppressing therapies.

In the new design, 72 eligible patients give 150ml of blood. This is processed into Tr1 ovalbumin specific antigen cells (Ova-Treg) in a process that currently taskes12 weeks. TxCell has to over recruit as patients need to meet the criteria of CDAI ≥ 250 (see below) with positive CRP and calprotectin markers both on enrolment and about three months later on dosing. Inevitably, some patients will fall below the 250 CDAI level, since Crohn’s is an episodic disorder, while their cells are being prepared so will not be dosed. In addition, not all patient blood samples will produce Ova-Treg cells in sufficient quantity or quality. TxCell requires 56 assessable patients for the endpoint.

The initial six-week phase is a double-blind, placebo-controlled evaluation of a single 106 cell infused dose vs placebo The endpoint is a response based on the CDAI score assessed at week 6.

A response is defined as a reduction of ≥100 points in the Crohn’s Disease Activity Index (CDAI, Exhibit 5). TxCell expects a response difference of 70% treated vs 30% placebo.

If all 28 patients per arm complete the study, the power is expected to be 80%. Statistical power is the probability that the clinical trial will have a significant (positive) result of p≤0.05 assuming that the real difference between treated and placebo is as anticipated. In other words, if Ovasave works, there is an 80% chance that this trial will be positive.

This is an open-label phase in which all patients receive 106 cells at weeks 8, 16 and 24 with final assessment at week 32. There is a three-year safety follow up as this is a cell therapy product. This stage is designed to be attractive to patients and to aid recruitment since all will receive the stem cells.

Source: Edison Investment Research based on TxCell statements and presentations

Phase III, partnering and probabilities

A positive primary endpoint by Q118 should enable a partnering deal, perhaps by late 2018. Timing of such deals is hard to forecast. Assuming this timeline is met, the trial protocol and manufacturing system needs to be in place before any Phase III can start. However, it is probable that a small multi-dose Phase IIb will be required before a multi-dose Phase III. Note that a few patients have already received multiple Ovasave doses, two for up to a year. Management expect the Phase III to be funded by a global partner that will also manufacture.

Manufacturing of Tr1 autologous cells is a multistep process. TxCell is working hard to reduce the number of stages, optimise the yield of each stage and reduce manual interventions required. It is working with the UK Cell Catapult, a government-backed centre of expertise, and various equipment manufacturers. As a final concept, a fully automated system is desirable with Tr1 cell culture in sealed units. This should provide optimum yields, an ability to scale up throughput (each patient is one batch so the process is replicated), and get a low cost of goods, perhaps about 10% of the average revenue. However, it is not certain how automated the final system will be. The partner will need to transfer this optimised process to its own facility or designated CMO and gain regulatory approval, although they could use TxCell’s current contractors for the trials. A US product requires a US cell manufacturing site; this can take at least nine months to establish and validate. A deal in Q418 therefore implies a trial start in H219 or later.

A multi-dose Phase III with 12-month dosing could take three years, so if started in H219, data might be reported in late H222. We forecast sales from late 2023, but 2024 is realistic.

The success probability of Ovasave is hard to assess. Classical Phase IIb projects have a 20-40% probability. Although TxCell views the probability of success as being very high, the limited Phase I data (three clinical remission cases out of eight by week five) suggests a clinical 33% probability in line with our general caution about cell therapy. We have applied a further 85% probability of developing an automated manufacturing system on time and to acceptable regulatory and commercial standards. This gives a combined 28% probability.

Potential market for Ovasave in Crohn’s disease

The current definable market for Ovasave is in Crohn’s disease where patients have received biological therapy but have proved either refractory or relapsed or have experienced severe side effects. Exhibit 9 gives the Edison summary market assessment. Exhibit 10 shows the assumptions and sources used. Although Europe should have a similar or greater number of cases, lower diagnosis and biological therapy use rates compared to the US lower the market potential.

Exhibit 9: Edison market assumptions on Ovasave

Market segment

Sub segment

% of NA market

North America

% of Eur Market

Europe

% of JP market

Japan

Total

Total population

656,750

762,200

26,924

1,445,874

Adult cases

93%

610,778

95%

724,090

93%

25,578

1,360,446

Biological

16.7%

102,000

10.0%

72,409

20.0%

5,116

179,525

Share of biological treated cases accessible to Ovasave

Initial non-response

35%

35,700

35%

25,343

35%

1,790

62,833

Severe side effects

10%

6,630

10%

4,707

10%

333

11,670

Secondary non-response

20%

13,260

20%

9,413

20%

665

23,338

Potential Ovasave market

55,590

39,463

2,788

97,841

Source: Edison Investment Research based on TxCell data and literatures sources as in Exhibit 10

At least 25% of the above patients might have perianal disease (de Zoeten 2013), of whom half have fistulas, channels between the intestine and skin or vagina. It is not certain how Ovasave therapy would relate to these complications. The assumption made is that perianal disease does not affect Ovasave use. Stem cells have already been shown to be useful in Crohn’s: the Phase III study of Cx601 from TiGenix has shown that it accelerates complex perianal fistula healing.

A commercial analogy to Ovasave might be Entyvio (vedolizumab, Takeda) approved for Crohn’s disease by the FDA and EMA when other treatments have failed. The UK NICE organisation issued guidance in August 2015 that Entyvio is useful, but only at a discounted price (undisclosed), for active Crohn’s disease if biological therapies have failed. Entyvio has a UK list price of £2,050 per 300mg dose, costing about €25,830 in year one, then about €18,450 for subsequent years if the disease remains active and Entyvio continued to be effective. In the US, prices are higher, with Entyvio costing $43,371 for the first year (€37,750) and $33,733 subsequently. Remicade before biosimilar erosion costs $31,856 in year one and $24,777 subsequently. TxCell notes that other current biological agents cost €22,000.

Our forecast assumes (see Exhibit 10) an EU list price for Ovasave of €30,000 based on a premium to the cost of the first year of Entyvio, The US price assumed is €45,000. The average biological price will fall in future as biosimilars enter the market, but the Ovasave market segments should not be affected by this and may expand if biological use becomes more affordable and widespread. This would create more refractory patients - but maybe these will be more price sensitive. Inflation of 2% is assumed from 2018.

TxCell assumes that 80% of patients will receive regular six-weekly doses of Ovasave in year one, dropping to 60% in year two and 30% in year three. These appear to be arbitrary assumptions, but seem reasonable and have been followed by Edison. The current production system produces in effect three years’ worth of therapy in one run. This makes the economics for TxCell’s partner front-loaded: the first dose has a very high cost of goods as custom made but subsequent doses just carry storage and logistics costs. We also assume that a core of 25% of patients stay on Ovasave therapy in year four, declining at 25% year-on-year. It is known that some patients remain on anti-TNF therapy for long periods. Longer-term use will require a further cell batch to be made.

Exhibit 10: Crohn’s disease market data and assumptions

Data

Edison estimate

% cases

Sources and commentary

Adult prevalence of Crohn’s disease

1.44m

100%

One of the confounding factors in assessing prevalence is the difficulty of diagnosis as it can be difficult to separate ulcerative colitis from Crohn’s disease. The prevalence is high because the condition is long lasting, but the disease is episodic so not all patients need treatment all the time. Loftus 2002 found that “Most patients have a chronic intermittent disease course, while 13% have an unremitting disease course and 10% have a prolonged remission”. Incidence levels (new cases) are very low at about 10-12 per 100,000 per year, so about 5% of prevalence rates. This epidemiological data therefore needs to be treated with some caution.

The Center for Disease Control estimates 201 per 100,000; Kappelman 2013 using insurance claims found 241/100,000. The US population is approaching 320 million, with Canada at 35 million, so this gives about 610,000 adult Crohn’s cases; adults are 93% of all cases.

The current EU population (28 states) is 507 million, so about 940,000 cases are expected. However, this includes a number of states where higher priced therapy will be less available. Further, it excludes rich counties like Switzerland and Norway. Adjusting for these gives an accessible European population of about 412 mi, so 724,000 adult cases in theory. However, the rate of diagnosis varies across European countries; for example, in the UK Stone 2003 found 130 cases per 100,000 whereas Hein 2014 in Germany noted 322 per 100,000 using insurance data. Lucendo 2014 found 137/100,000 in Spain.

Japan claims a very low rate of 21.3/100,000 (Asakura 2009), about 10% of the likely European and US rate.

Proportion on biological therapy

205,000

13-20%

Use of biological agents is fairly stable and healthcare providers have often tried to optimise the dosing of such drugs due to their cost. Patients may cycle between agents, so there will be a degree of patient churn as patients drop out if in remission and new patients start. Edison uses a value of 15% on this, but there is no data to support this. The UK NICE recommends review after a year of therapy with the aim of discontinuing therapy if the patient is in remission. However, patients often wish to stay on therapy to avoid further attacks. The assumption is that only patients that have failed on at least one biological therapy will be eligible for Ovasave and these will tend to be moderate to severe disease categories.

The estimate for patients on biological therapy in the US is derived from market research from TxCell. Edison estimates about 102,000 Crohn’s patients on anti-TNF therapy in the US and Canada. The evidence for other countries does not appear as robust: TxCell data indicates that about 13.5% of European CD patients are using biological agents. Edison estimates about 98,000 patients. Japan might add a further 5,000. Japan has a novel system for regulating cell therapies, enabling faster approval and reimbursement.

This gives about 205,000 potential patients on biological therapy. This number will be subject to a natural level of churn as some patients gain prolonged remission and others start therapy. Lower biosimilar prices may also encourage biological use over the next decade.

Unresponsive

72,000

35%

Biological agents generally claim a response rate of about 65% (see Panaccione 2010 for a review). Hence, 35% are eligible for Ovasave therapy – several biological agents may be tried first. Roda 2016 noted a high level of unresponsive or relapsed patients and attributes this to the antibody responses against the monoclonal antibodies used. Ovasave will avoid that issue.

Side effect withdrawal

13,000

10%

The rate of withdrawal due to side effects of biological drugs in trials is about 10%, although it is not always clear why. In CLASSIC II (see below), the open-label withdrawal rate was 11.3%.

Relapsed

27,000

20%

Of the 65% who respond initially to biological therapy, TxCell estimated in 2014 that 20% would relapse. The CLASSIC II trial (Sandborn 2007) with Adalimumab looked at patients who were either initially in remission or responsive but not in remission. Of those in remission, 79-83% remained in remission at 56 weeks after maintenance therapy. Edison estimates about 27,000 possible prevalence cases.

Source: Edison Investment Research

Exhibit 11: Progression of Crohn’s disease and role of Ovasave

Source: TxCell presentation

Forecasting the Crohn’s disease market

The figures in Exhibit 9 are for prevalence. We assume that there is a turnover rate of 20% of biological patients per year; the incidence rate overall is about 5% of the prevalence rate. If the Ovasave market share is over 20%, the prevalence number reduces assuming patients try a course of Ovasave once only (note they receive multiple doses for up to three years from each custom manufactured batch). Turnover is due to the episodic nature of the disease and by the restraints imposed by healthcare providers to avoid substantial long-term costs with no clinical gain. Exhibit 11 shows the time course with patients often cycling between biological agents.

This might be offset commercially by longer periods of use. To model these potentially longer-term patients, we assume a fourth year of use by 20% of patients, with the number declining at 25% per year. All these are simplistic assumptions for the purposes of modelling.

The effects are shown in Exhibits 12 (EU) and Exhibit 13 (US) for the number of patients. Note that the x-axis shows years after launch, assumed to be 2023-4 for valuation, but uncertain.

Exhibit 12: European patient numbers

Exhibit 13: North America patient numbers

Source: Edison Investment Research. Note: Not probability adjusted.

Source: Edison Investment Research. Note: Not probability adjusted.

Exhibit 12: European patient numbers

Source: Edison Investment Research. Note: Not probability adjusted.

Exhibit 13: North America patient numbers

Source: Edison Investment Research. Note: Not probability adjusted.

Competition

At this stage in Ovasave’s development, it is not feasible to assess the competitive profile fully. The few possible alternatives are all yet to produce definitive data. In the cell area, Mesoblast has a Phase III study with Prochymal, due to report mid-2017, in resistant Crohn’s disease.

The main action in biosimilar development will not affect Ovasave and might expand the market. In the area of anti-TNF resistant Crohn’s disease, AbbVie has a major, 210-patient Phase II dose finding study running with a small molecule JAK inhibitor, ABT494 (NCT02365649). This reports in December 2017 and could reach the market before Ovasave, but the primary indication will be rheumatoid arthritis. It could be a powerful competitor if it works well, since as a small molecule it could be less expensive and easy to administer. Roche has a “dual-action anti-integrin” antibody, Etrolizumab, similar to Entyvio, which is targeting both anti-TNF naive and -refractory patients (NCT02394028). This trial reports in 2018 so could be marketed from 2020.

Uveitis: A new Phase I indication

Non-infectious uveitis is an inflammatory eye condition. It is mostly acute and low level, but serious cases can result in blindness. It is an orphan drug indication. Exhibit 14 has background details. TxCell has published a paper on its preclinical work in this indication (Asnagli 2015) using antigen- directed Treg cells. The Tr1 cells are targeted against Type II collagen.

Note that Collagen Type II is a self-antigen rather than a foreign food antigen like Ovalbumin. Self-protein fragments are normally presented in MHCI and recognised by effector T-cells (CD8+) while foreign antigens are brought into cells and presented in MHCII – although there is “crossover”, which is not well understood. This might be clinically important as antigen-specific Tr1 cells (CD4+) are only stimulated by antigens presented in MHCII. The clinical mode of action therefore relies on adequate Collagen Type II antigen presentation in MHCII.

An immune reaction to Type II collagen has been claimed in the middle eye disease pars planitis, an intermediate form of uveitis characterised by infiltration of white cells to the vitreous (these clumps of cells are known as “snowballs”). The concept is that Col-Tregs will migrate to the eye, be activated there by Type II collagen and then suppress inflammation. This theory needs to be tested clinically. TxCell has indicated that it is planning a Phase I study.

Exhibit 14: Uveitis background and epidemiology

Aspect

Notes

Eye structure

The eye capsule consists of three layers, the retina, the light sensing layer, underlain by the uvea, a pigmented layer containing cells and blood vessels that support the retinal cells and cornea, and the fibrous sclera. The uvea extends to the front of the eye and includes the iris.

Collagen Type II

Collagen is the ubiquitous, tough, triple-stranded structural protein that forms the matrix between cells. It gives strength and elasticity to tendons and is also in bone. There are many collagen types; Type II is a major component of the vitreous gel that fills the inner eye.

Inflammation of the uvea

This can be anterior (front of eye), intermediate around the mid-section of the eye or posterior, which attacks the retina. Acute anterior uveitis (AAU) is the most common type but the disease is heterogeneous. Lee et al (2014) have reviewed autoimmunity in uveitis. It seems that steroid refractory chronic non-infectious uveitis may have an autoimmune component but the percentage of uveitis cases is not clear.

Treatment

It is usually treated with steroid drops. Chronic uveitis is harder to treat and needs high dose systemic corticosteroids. Methotrexate can be used in chronic cases. Biological agents like anti-TNF biological drugs are used as third-line rescue therapy.

Risks

Most cases seem to resolve with local steroids, but chronic inflammation leads to permanent damage and blindness. 10% of sight loss cases are claimed as due to uveitis and 10-20% of severe visual handicaps – these figures cover infection and autoimmune cases.

US incidence

The incidence of uveitis is not well studied with only four medically controlled population studies. One estimated an incidence of 52 per 100,000 per year indicating 150,000 cases (Gritz 2004). The same study found a prevalence of 118/100,000. However, two others found rates half this: Acharya 2013 was 24/100,000 in Hawaii and a North Western study (Suhler 2008) found 25.6 cases/100,000 person-years and a crude prevalence of 69 cases/100,000 persons.

European incidence

European studies are more limited in scope and limited to referral centres, usually treating severe cases but not a systematic population survey. For example, in Barcelona Llorenç (2015) found an incidence of over 50/100,000 based on 1,022 cases. Of these, 52% were AAU. Some 25% of cases were due to autoimmune disease. In its orphan drug application, TxCell estimated that non-infectious uveitis affected approximately 48 per 100,000 people in the European Union. This was equivalent to a total of around 245,000 people. No sources were cited. The ceiling for orphan designation is five people in 10,000.

Aspect

Eye structure

Collagen Type II

Inflammation of the uvea

Treatment

Risks

US incidence

European incidence

Notes

The eye capsule consists of three layers, the retina, the light sensing layer, underlain by the uvea, a pigmented layer containing cells and blood vessels that support the retinal cells and cornea, and the fibrous sclera. The uvea extends to the front of the eye and includes the iris.

Collagen is the ubiquitous, tough, triple-stranded structural protein that forms the matrix between cells. It gives strength and elasticity to tendons and is also in bone. There are many collagen types; Type II is a major component of the vitreous gel that fills the inner eye.

This can be anterior (front of eye), intermediate around the mid-section of the eye or posterior, which attacks the retina. Acute anterior uveitis (AAU) is the most common type but the disease is heterogeneous. Lee et al (2014) have reviewed autoimmunity in uveitis. It seems that steroid refractory chronic non-infectious uveitis may have an autoimmune component but the percentage of uveitis cases is not clear.

It is usually treated with steroid drops. Chronic uveitis is harder to treat and needs high dose systemic corticosteroids. Methotrexate can be used in chronic cases. Biological agents like anti-TNF biological drugs are used as third-line rescue therapy.

Most cases seem to resolve with local steroids, but chronic inflammation leads to permanent damage and blindness. 10% of sight loss cases are claimed as due to uveitis and 10-20% of severe visual handicaps – these figures cover infection and autoimmune cases.

The incidence of uveitis is not well studied with only four medically controlled population studies. One estimated an incidence of 52 per 100,000 per year indicating 150,000 cases (Gritz 2004). The same study found a prevalence of 118/100,000. However, two others found rates half this: Acharya 2013 was 24/100,000 in Hawaii and a North Western study (Suhler 2008) found 25.6 cases/100,000 person-years and a crude prevalence of 69 cases/100,000 persons.

European studies are more limited in scope and limited to referral centres, usually treating severe cases but not a systematic population survey. For example, in Barcelona Llorenç (2015) found an incidence of over 50/100,000 based on 1,022 cases. Of these, 52% were AAU. Some 25% of cases were due to autoimmune disease. In its orphan drug application, TxCell estimated that non-infectious uveitis affected approximately 48 per 100,000 people in the European Union. This was equivalent to a total of around 245,000 people. No sources were cited. The ceiling for orphan designation is five people in 10,000.

Source: Edison Investment Research

At this time, the market for Col-Treg does not seem clearly defined by the literature. Uveitis is a disparate set of conditions so TxCell needs to define a target subgroup for initial clinical studies, possibly steroid refractory chronic non-infectious uveitis. Currently, it is hard to assess what proportion of uveitis cases could be treated with Tr1 therapy and for how long and at what price. Edison uses a TxCell guided figure of 15,000 cases per year but has not been able to verify this.

Chimeric antigen regulatory T-cells (CAR Tregs)

The big advantage of CAR Tregs (Exhibit 15) is that the genetically modified Treg cells can be easily cultured to give a homogeneous product. The CAR Tregs are targeted to the antigen so therapy design is much more flexible and direct.

TxCell is developing its own protected technology platform in this area and is building a strong science team to exploit this technology. In addition, TxCell has a licensing option, excisable up to 30 June 2016, over a patent application for CAR Treg cells (filed in 2007 by Professor Eshhar) from the Weizmann Institute of Science, Israel. Financial terms on the option have not been disclosed but are understood not to be material.

Exhibit 15: CAR Treg technology

Aspect

Comment

More powerful and specific than Tr1 cells

CAR Treg technology (ENTrIA) is more flexible than the ASTrIA platform. CAR Tregs are directly activated by the antigen so do not rely on the processing of the antigen by other immune cells (antigen processing cells, APCs); APCs only display certain protein fragment types. This new approach extends to multiple indications and should generate development partnerships.

Patent

The key patent (WO2008095141A2) is “Redirected, genetically-engineered T regulatory cells and their use in suppression of autoimmune and inflammatory disease” This is published as an application with an expiry date, if granted, of 30 January 2028. TxCell has an option to license this until the end of June 2016.

T-cell activation

A TCR recognises a specific protein when that protein has been processed into antigen fragments presented in the major histocompatibility complex (MHC). An individual only has certain MHC types. A CAR Treg directly binds a protein so can target either self or foreign targets, giving huge flexibility in therapeutic design.

CAR approach

To enable this precision of targeting, an artificial receptor is made using elements from the T-cell receptor complex. These use antibody binding regions. This requires that the autologous T-cells, from patients, are genetically modified, just as in CAR T-cell cancer therapy.

Engineering a CAR

The usual route to turn a Treg cell harvested from a patient into a CAR Treg cell is by using a lentivirus to incorporate the engineered genes. The genes are used to produce the CAR. The antibody region is external to the cell to detect its target and it signals through an internal domain: CD3 ζ. The CAR also requires an integrated co-stimulatory molecule. The internal signalling domain from the costimulatory protein CD28 is used, again parallel to CAR cancer methods; OX40 is also used in this role.

Example using CEA

As an example, Blat 2014 (part of work at Eshhar’s group) reported on a colitis preclinical model where a CEA-targeted CAR Treg was used to home into the site of inflammation to supress effector CEA T-cell activity and halt disease progression. CEA stands for carcinoembryonic antigen; a colon cancer marker. In the experiment, CEA, which is not present in adult tissues, was trigged by a chemical challenge. Effector CAR T-cells were also injected. This model therefore created an antigen and added the effector cell to attack the tissues and in effect create an autoimmune disease. If the CAR Treg cells were also given, there was no attack.

Mechanism

The mechanism by which Treg cells work is not clear but direct interaction with T-effector cells is likely and use of anti-inflammatory cytokines like IL10 seems implicated (Blat 2014).

Autologous therapy

CAR Treg therapy will be autologous. This is also currently the case with effector CAR T-cell therapies to avoid graft vs host disease. CAR Treg cells need to interact with the host immune system so will need to be autologous or closely matched to avoid rejection. CAR Tregs are likely to proliferate in response to antigen so a single dose might last for several months. This is still to be determined however.

Allogeneic therapy?

This seems a much harder challenge. In the related effecter CAR T-cell space, various solutions are being developed; for example, the Celyad TIM approach or promising Cellectis clinical technology.

Aspect

More powerful and specific than Tr1 cells

Patent

T-cell activation

CAR approach

Engineering a CAR

Example using CEA

Mechanism

Autologous therapy

Allogeneic therapy?

Comment

CAR Treg technology (ENTrIA) is more flexible than the ASTrIA platform. CAR Tregs are directly activated by the antigen so do not rely on the processing of the antigen by other immune cells (antigen processing cells, APCs); APCs only display certain protein fragment types. This new approach extends to multiple indications and should generate development partnerships.

The key patent (WO2008095141A2) is “Redirected, genetically-engineered T regulatory cells and their use in suppression of autoimmune and inflammatory disease” This is published as an application with an expiry date, if granted, of 30 January 2028. TxCell has an option to license this until the end of June 2016.

A TCR recognises a specific protein when that protein has been processed into antigen fragments presented in the major histocompatibility complex (MHC). An individual only has certain MHC types. A CAR Treg directly binds a protein so can target either self or foreign targets, giving huge flexibility in therapeutic design.

To enable this precision of targeting, an artificial receptor is made using elements from the T-cell receptor complex. These use antibody binding regions. This requires that the autologous T-cells, from patients, are genetically modified, just as in CAR T-cell cancer therapy.

The usual route to turn a Treg cell harvested from a patient into a CAR Treg cell is by using a lentivirus to incorporate the engineered genes. The genes are used to produce the CAR. The antibody region is external to the cell to detect its target and it signals through an internal domain: CD3 ζ. The CAR also requires an integrated co-stimulatory molecule. The internal signalling domain from the costimulatory protein CD28 is used, again parallel to CAR cancer methods; OX40 is also used in this role.

As an example, Blat 2014 (part of work at Eshhar’s group) reported on a colitis preclinical model where a CEA-targeted CAR Treg was used to home into the site of inflammation to supress effector CEA T-cell activity and halt disease progression. CEA stands for carcinoembryonic antigen; a colon cancer marker. In the experiment, CEA, which is not present in adult tissues, was trigged by a chemical challenge. Effector CAR T-cells were also injected. This model therefore created an antigen and added the effector cell to attack the tissues and in effect create an autoimmune disease. If the CAR Treg cells were also given, there was no attack.

The mechanism by which Treg cells work is not clear but direct interaction with T-effector cells is likely and use of anti-inflammatory cytokines like IL10 seems implicated (Blat 2014).

CAR Treg therapy will be autologous. This is also currently the case with effector CAR T-cell therapies to avoid graft vs host disease. CAR Treg cells need to interact with the host immune system so will need to be autologous or closely matched to avoid rejection. CAR Tregs are likely to proliferate in response to antigen so a single dose might last for several months. This is still to be determined however.

This seems a much harder challenge. In the related effecter CAR T-cell space, various solutions are being developed; for example, the Celyad TIM approach or promising Cellectis clinical technology.

Source: Edison Investment Research, sources as hyperlinks

Although of high potential, the CAR Treg area will take some years to develop into a set of clinical-stage projects. TxCell has entered that a collaboration with a leading European immunology institute, the San Raffaele Scientific institute in Milan. The collaboration aims to develop a Treg product for lupus nephritis. There is some evidence (Le Buanec 2011) that Interferon-alpha in lupus might downregulate the standard Treg immune repression allowing an autoimmune response. Autologous Tregs have already been used in small, usually academic trials to treat autoimmune disease such as Type I diabetes, for example, Bluestone 2015. Caladrius Bioscience has an autologous Treg trial (the Phase II T-rex) and FDA orphan designation in Type I diabetes.

CAR Treg cells are genetically modified, which adds regulatory complexity, but the effector CAR T-cell area in cancer provides regulatory precedents. No CAR Treg has been clinically tested, so initial trials, possibly in the 2019-2020 period, will be slow, patient by patient and safety focused. Development could then accelerate if CAR Tregs prove potent in lead indications. Other possible indications could include multiple sclerosis, according to TxCell.

Sensitivities: Early stage but with CAR Treg concept

The major sensitivity relates to the current CATS29 Phase II and TxCell’s ability to fund the €15m study and ongoing R&D including the important CAR Treg cell platform. French tax credits are expected to pay about 30% of each year’s cost; the cash is received in the following year. The Crohn’s market is complex to forecast, with uncertainties in patient numbers, about 100,000 possible, and significant regional market differences. Incidence is 5% of prevalence so the market renews slowly. The level of long-term Ovasave use is crucial to value but uncertain. Edison assumes some long-term use and a level of churn in patients starting and stopping biological agents. A core assumption is that a partner is found by the end of 2018. Ovasave may not progress unless this occurs.

Steroid resistant Uveitis (eye inflammation) is a niche market of about 15,000 cases per year where TxCell could sell Col-Treg directly. However, it will need to fund the trials. The evidence here is less clear overall and the product is in preclinical development with no disclosed timeline to a Phase I.

Finally, the new CAR Treg opportunity appears to be potentially applicable to many indications in theory but as a technology platform it is hard to value and is at least three years from a clinical trial, possibly for lupus nephritis, the current lead. The 2008 academic blocking patent under option till June may not be granted or secured although TxCell is filing new patents. The CAR platform does have a high deal potential, probably from 2017.

Valuation

The valuation of TxCell currently depends on Ovasave as the sole clinical-stage project. This combines two success probabilities; a clinical risk set at 33% multiplied by an 85% probability of achieving a commercial manufacturing system by late 2018. This gives a 28.05% combined probability. If the Phase IIb, which has yet to restart, shows a positive primary endpoint, the probability could rise to 45%. Edison has assumed a €25m upfront in a €175m overall deal with a 16.5% royalty. This is lower than the level of some deal upfronts paid for CAR T-cell cancer therapies but this is a less prominent indication although it offers a valuable, orphan market. Exhibit 16 shows the overall market (including Japan) by value (before probability adjustment).

Exhibit 16: Projected, non-probability adjusted sales projection for Ovasave

Source: Edison Investment Research

The second product, for uveitis, is still preclinical, so has a standardised 7.5% probability. Uveitis has a high value relative to its market as it might be marketed directly by TxCell. Partnering would limit this value. However, whereas costs for Ovasave truncate after partnering, TxCell is assumed to fund the full uveitis study programme to launch in 2024.

The important CAR Treg platform is valued at a nominal €20m, reflecting its importance to the company but noting that it is currently not possible to value it on the basis of candidate products. As a CAR area, it has the potential for substantive deals perhaps of €300m overall value although no substantive deal is expected before 2020.

This gives a current pre-dilution value of €7.08/share. A funding need of €27m net of current cash is projected until late 2017, with more possible thereafter. The valuation summary is in Exhibit 17. Cash (March 2016) of €5m could add €0.38/share.

Exhibit 17: Valuation summary

Item

Rate

Probability

NPV @ 12.5%

Revenues

Crohn's disease North America

28.05%

106

Crohn's disease Europe

28.05%

29

Crohn's disease Japan

28.05%

19

Uveitis

7.50%

10.4

Total Royalty NPV

 

 

 

163

Upfront and milestone NPV less Trizel share

 

 

22

Costs (inc Uveitis marketing) less tax credits

-74

Gross profit

 

 

 

111

Tax

24%

-33

Debt NPV (@1.5%)

-7

CAR Treg platform

Nominal

20.0

Pre-funding and dilution value

91

Value/share

Shares in issue

 

12.8873

€ 7.08

Diluted value/share

Warrants

1.77

14.6573

€ 6.22

Source: Edison Investment Research

The tax rate used is 24%. This is below the French standard rate of 33%, but TxCell will have ongoing costs from further projects to offset taxes.

An important aspect is the potential for value development. If the value of the CAR Treg platform rises to €50m nominal and Ovasave progresses to a Phase III with a partner and a deal worth at least €175m with a €25m minimum upfront, the indicative market value by late 2018 could rise to over €300m. Note that this is one of several possible value scenarios and is not a forecast. TxCell management expects significantly higher deal values if Proof of Concept is obtained on the Ovasave project in Phase IIb leading to overall validation of the regulatory T-cell concept.

Financials

At year-end 2015, TxCell had €9.2m cash; cash on 31 March 2016 was €5m with a €3m tax credit due. Financial estimates are in Exhibit 18. Cash expenditure will rise once the Phase IIb trial resumes in H216; this trial is costed by management at €15m with €4.5m assumed by Edison in 2016. The 2015 core operating cash costs (before funding) were €12.3m. TxCell has stated that the cash burn in 2016 will be €15m. This covers increased investment in the CAR Treg area and the resumed CATS29 trial costs; some cost savings are from closure of the manufacturing site in 2015. To cover all cash requirements and give a cash cushion for 2018, Edison assumes that €36m in cash will be needed over the next two years (after tax credit funding) with €9.2m cash available at the 2015 year end. This implies a funding gap of €27m. Edison treats future funding as illustrative long-term debt and additional capital of €7m in 2016 and €20m in 2017 is assumed.

TxCell has a liquidity contract worth €200k. As of 31 December 2015, €95k had been used to buy 16,280 shares leaving €105k in cash.

Ferring/Trizell deal and termination

In December 2015, the partnering deal with Trizell was terminated by agreement. TxCell paid €2m immediately with €2m due in late 2017 and a further €2m in late 2018. A further €9m payment is based on successful commercialisation and assumed to be paid in 2025, although some portion of any new partnering deal might be due. The buyout was expensive, but Trizell put little into development, €1m upfront with about €2m in 2015, and was not a strong future development partner. The buyout also enabled a smaller Phase IIb trial to be run with only one dose arm as opposed to the larger three dose plus placebo study originally planned.

The original 2014 deal with Ferring enabled the IPO to take place. It was billed as worth €76m. The agreement was transferred to Trizell in early 2015. Trizell and Ferring are both owned by the Dr Frederik Paulsen Foundation.

Exhibit 18: Financial summary

€ 000

2014

2015

2016e

2017e

Year End December

IFRS

IFRS

IFRS

IFRS

PROFIT & LOSS

Revenue

1,386

1,614

0

0

Tax refund

2,035

3,023

4,950

6,450

Cost of Sales

0

0

0

0

Gross Profit

3,421

4,637

4,950

6,450

EBITDA

(8,729)

(10,760)

(14,545)

(18,295)

Operating Profit (before amort. and except.)

(8,269)

(9,625)

(14,295)

(18,045)

Intangible Amortisation

0

0

0

0

Exceptionals

0

(1,189)

0

0

Share based payments

(1,615)

(483)

(500)

(500)

Operating Profit

(9,884)

(11,297)

(14,795)

(18,545)

Net Interest

4

15

5

5

Profit Before Tax (norm)

(8,725)

(10,745)

(14,540)

(18,290)

Profit Before Tax (FRS 3)

(8,265)

(11,282)

(14,790)

(18,540)

Tax

0

0

0

0

Profit After Tax (norm)

(8,725)

(10,745)

(14,540)

(18,290)

Profit After Tax (FRS 3)

(8,265)

(11,282)

(14,790)

(18,540)

Average Number of Shares Outstanding (m)

10.6

12.3

12.3

12.3

EPS - normalised (c)

(82.6)

(87.4)

(118.3)

(148.8)

EPS - (IFRS) (c)

(78.3)

(91.8)

(120.3)

(150.9)

Dividend per share (c)

0.0

0.0

0.0

0.0

Gross Margin (%)

NA

NA

NA

NA

EBITDA Margin (%)

NA

NA

NA

NA

Operating Margin (before GW and except.) (%)

NA

NA

NA

NA

BALANCE SHEET

Fixed Assets

1,543

6,938

7,438

7,538

Intangible Assets

8

5,907

6,407

6,507

Tangible Assets

1,404

876

876

876

Other

131

155

155

155

Current Assets

18,500

13,782

5,964

5,819

Stocks

0

0

0

0

Debtors

2,548

1,551

1,551

1,551

Cash

13,917

9,208

1,413

1,268

Other

2,035

3,023

3,000

3,000

Current Liabilities

(3,341)

(7,467)

(7,467)

(5,467)

Creditors

(1,946)

(5,859)

(5,859)

(3,859)

Short term borrowings

(1,395)

(1,608)

(1,608)

(1,608)

Long Term Liabilities

(1,990)

(1,664)

(8,641)

(28,641)

Long term borrowings

(1,627)

(1,641)

(8,641)

(28,641)

Other long term liabilities

(363)

(23)

0

0

Net Assets

14,712

11,589

(2,706)

(20,751)

CASH FLOW

Operating Cash Flow

(6,937)

(10,081)

(14,100)

(17,850)

Net Interest

4

15

5

5

Tax

0

0

0

0

Capex

(590)

(214)

(700)

(300)

Acquisitions/disposals

17

(5,879)

0

0

Financing

20,891

7,631

7,000

20,000

Other

0

3,813

0

(2,000)

Net Cash Flow

13,385

(4,715)

(7,795)

(145)

Opening net debt/(cash)

2,490

(10,895)

(5,959)

8,836

HP finance leases initiated

0

0

0

0

Other

0

(221)

(7,000)

(20,000)

Closing net debt/(cash)

(10,895)

(5,959)

8,836

28,981

Source: TxCell accounts, Edison Investment Research

Contact details

Revenue by geography

TxCell,
Les Cardoulines HT1, Allée de la Nertière,

06560 Valbonne - Sophia Antipolis
France

Tel:+33(0) 497 218 300

Fax:+33(0) 493 641 580

contact@txcell.com

www.txcell.com/index.php/en/

Contact details

TxCell,
Les Cardoulines HT1, Allée de la Nertière,

06560 Valbonne - Sophia Antipolis
France

Tel:+33(0) 497 218 300

Fax:+33(0) 493 641 580

contact@txcell.com

www.txcell.com/index.php/en/

Revenue by geography

Management team

CEO: Stéphane Boissel

Miguel Forte: Chief Operating Officer

Stephane studied management and finance at the University of Lyon, France, graduated at Paris-Dauphine and obtained his MBA from the University of Chicago (Booth GSB). He has worked for PWC and Lazard. He was CFO at Innate Pharma then from 2010 to 2014, he was Deputy-CEO of Transgene SA. Before joining TxCell, was CEO of Genclis, a molecular diagnostic company.

Miguel holds a MD and a PhD, and is a Specialist in Infectious Diseases with a certificate on Health Economics of Pharmaceuticals and Medical Technologies. He is Fellow of the Faculty of Pharmaceutical Medicine of the RCP (UK). He spent six years with Bristol-Myers before joining Nabi Pharmaceuticals as a Vice-President. In 2006 he became a Vice President at UCB, before joining TxCell as Chief Medical Officer in 2010 becoming Chief Operating Officer in 2015.

FD : Raphaël Flipo

Arnaud Foussat: Chief Scientific Officer

Raphaël graduated from business school (EDHEC, Nice) with a specialization in corporate finance, and he holds a Master’s Degree in tax and business law. He worked for PwC and then joined Lionbridge Technologies. He joined TxCell in 2013 as Chief Financial Officer.

Arnaud is a Graduate of Institut Pasteur of Paris, obtained his Ph.D. degree from the University of Paris VII. From 2000, he worked at INSERM on Tregs. He joined TxCell in 2004 and became Director of Research and Development in 2005, and Chief Scientific Officer in 2015.

Management team

CEO: Stéphane Boissel

Stephane studied management and finance at the University of Lyon, France, graduated at Paris-Dauphine and obtained his MBA from the University of Chicago (Booth GSB). He has worked for PWC and Lazard. He was CFO at Innate Pharma then from 2010 to 2014, he was Deputy-CEO of Transgene SA. Before joining TxCell, was CEO of Genclis, a molecular diagnostic company.

Miguel Forte: Chief Operating Officer

Miguel holds a MD and a PhD, and is a Specialist in Infectious Diseases with a certificate on Health Economics of Pharmaceuticals and Medical Technologies. He is Fellow of the Faculty of Pharmaceutical Medicine of the RCP (UK). He spent six years with Bristol-Myers before joining Nabi Pharmaceuticals as a Vice-President. In 2006 he became a Vice President at UCB, before joining TxCell as Chief Medical Officer in 2010 becoming Chief Operating Officer in 2015.

FD : Raphaël Flipo

Raphaël graduated from business school (EDHEC, Nice) with a specialization in corporate finance, and he holds a Master’s Degree in tax and business law. He worked for PwC and then joined Lionbridge Technologies. He joined TxCell in 2013 as Chief Financial Officer.

Arnaud Foussat: Chief Scientific Officer

Arnaud is a Graduate of Institut Pasteur of Paris, obtained his Ph.D. degree from the University of Paris VII. From 2000, he worked at INSERM on Tregs. He joined TxCell in 2004 and became Director of Research and Development in 2005, and Chief Scientific Officer in 2015.

Principal shareholders

(%)

BPI France

38.8%

Auriga Partners

30.4%

Seventure Partners

8.5%

Companies named in this report

Caladrius Biosciences Inc. (CLBS); TiGenix (TIGB)

Edison, the investment intelligence firm, is the future of investor interaction with corporates. Our team of over 100 analysts and investment professionals work with leading companies, fund managers and investment banks worldwide to support their capital markets activity. We provide services to more than 400 retained corporate and investor clients from our offices in London, New York, Frankfurt, Sydney and Wellington. Edison is authorised and regulated by the Financial Conduct Authority (www.fsa.gov.uk/register/firmBasicDetails.do?sid=181584). Edison Investment Research (NZ) Limited (Edison NZ) is the New Zealand subsidiary of Edison. Edison NZ is registered on the New Zealand Financial Service Providers Register (FSP number 247505) and is registered to provide wholesale and/or generic financial adviser services only. Edison Investment Research Inc (Edison US) is the US subsidiary of Edison and is regulated by the Securities and Exchange Commission. Edison Investment Research Limited (Edison Aus) [46085869] is the Australian subsidiary of Edison and is not regulated by the Australian Securities and Investment Commission. Edison Germany is a branch entity of Edison Investment Research Limited [4794244]. www.edisongroup.com

DISCLAIMER
Copyright 2016 Edison Investment Research Limited. All rights reserved. This report has been commissioned by TxCell and prepared and issued by Edison for publication globally. All information used in the publication of this report has been compiled from publicly available sources that are believed to be reliable, however we do not guarantee the accuracy or completeness of this report. Opinions contained in this report represent those of the research department of Edison at the time of publication. The securities described in the Investment Research may not be eligible for sale in all jurisdictions or to certain categories of investors. This research is issued in Australia by Edison Aus and any access to it, is intended only for "wholesale clients" within the meaning of the Australian Corporations Act. The Investment Research is distributed in the United States by Edison US to major US institutional investors only. Edison US is registered as an investment adviser with the Securities and Exchange Commission. Edison US relies upon the "publishers' exclusion" from the definition of investment adviser under Section 202(a)(11) of the Investment Advisers Act of 1940 and corresponding state securities laws. As such, Edison does not offer or provide personalised advice. We publish information about companies in which we believe our readers may be interested and this information reflects our sincere opinions. The information that we provide or that is derived from our website is not intended to be, and should not be construed in any manner whatsoever as, personalised advice. Also, our website and the information provided by us should not be construed by any subscriber or prospective subscriber as Edison’s solicitation to effect, or attempt to effect, any transaction in a security. The research in this document is intended for New Zealand resident professional financial advisers or brokers (for use in their roles as financial advisers or brokers) and habitual investors who are “wholesale clients” for the purpose of the Financial Advisers Act 2008 (FAA) (as described in sections 5(c) (1)(a), (b) and (c) of the FAA). This is not a solicitation or inducement to buy, sell, subscribe, or underwrite any securities mentioned or in the topic of this document. This document is provided for information purposes only and should not be construed as an offer or solicitation for investment in any securities mentioned or in the topic of this document. A marketing communication under FCA Rules, this document has not been prepared in accordance with the legal requirements designed to promote the independence of investment research and is not subject to any prohibition on dealing ahead of the dissemination of investment research.
Edison has a restrictive policy relating to personal dealing. Edison Group does not conduct any investment business and, accordingly, does not itself hold any positions in the securities mentioned in this report. However, the respective directors, officers, employees and contractors of Edison may have a position in any or related securities mentioned in this report. Edison or its affiliates may perform services or solicit business from any of the companies mentioned in this report. The value of securities mentioned in this report can fall as well as rise and are subject to large and sudden swings. In addition it may be difficult or not possible to buy, sell or obtain accurate information about the value of securities mentioned in this report. Past performance is not necessarily a guide to future performance. Forward-looking information or statements in this report contain information that is based on assumptions, forecasts of future results, estimates of amounts not yet determinable, and therefore involve known and unknown risks, uncertainties and other factors which may cause the actual results, performance or achievements of their subject matter to be materially different from current expectations. For the purpose of the FAA, the content of this report is of a general nature, is intended as a source of general information only and is not intended to constitute a recommendation or opinion in relation to acquiring or disposing (including refraining from acquiring or disposing) of securities. The distribution of this document is not a “personalised service” and, to the extent that it contains any financial advice, is intended only as a “class service” provided by Edison within the meaning of the FAA (ie without taking into account the particular financial situation or goals of any person). As such, it should not be relied upon in making an investment decision. To the maximum extent permitted by law, Edison, its affiliates and contractors, and their respective directors, officers and employees will not be liable for any loss or damage arising as a result of reliance being placed on any of the information contained in this report and do not guarantee the returns on investments in the products discussed in this publication. FTSE International Limited (“FTSE”) © FTSE 2016. “FTSE®” is a trade mark of the London Stock Exchange Group companies and is used by FTSE International Limited under license. All rights in the FTSE indices and/or FTSE ratings vest in FTSE and/or its licensors. Neither FTSE nor its licensors accept any liability for any errors or omissions in the FTSE indices and/or FTSE ratings or underlying data. No further distribution of FTSE Data is permitted without FTSE’s express written consent.

Frankfurt +49 (0)69 78 8076 960

Schumannstrasse 34b

60325 Frankfurt

Germany

London +44 (0)20 3077 5700

280 High Holborn

London, WC1V 7EE

United Kingdom

New York +1 646 653 7026

245 Park Avenue, 39th Floor

10167, New York

US

Sydney +61 (0)2 9258 1161

Level 25, Aurora Place

88 Phillip St, Sydney

NSW 2000, Australia

Wellington +64 (0)48 948 555

Level 15, 171 Featherston St

Wellington 6011

New Zealand

Frankfurt +49 (0)69 78 8076 960

Schumannstrasse 34b

60325 Frankfurt

Germany

London +44 (0)20 3077 5700

280 High Holborn

London, WC1V 7EE

United Kingdom

New York +1 646 653 7026

245 Park Avenue, 39th Floor

10167, New York

US

Sydney +61 (0)2 9258 1161

Level 25, Aurora Place

88 Phillip St, Sydney

NSW 2000, Australia

Wellington +64 (0)48 948 555

Level 15, 171 Featherston St

Wellington 6011

New Zealand

Ultra Electronics — Update 27 May 2016

Ultra Electronics

Continue Reading

Subscribe to Edison

Get access to the very latest content matched to your personal investment style.

Sign up for free