Newron Pharmaceuticals — Dyskinesia deal delivered

Xadago is licensed, outside Japan and Asia, to Zambon, a private Italian company. Zambon is strong in Europe with 20% of its sales in the home Italian market. It launched Xadago in Europe during H115, with Newron receiving double-digit royalties on sales. Our royalty rate estimate is in about 12%. Total product royalties rose in 2020 to €5.24m versus €4.75m in 2019.

In June 2020, the US Xadago rights transferred to Supernus Pharma, a quoted US company. Supernus cites that the US Parkinson’s disease (PD) market is anticipated to grow from $1.5bn to $6.2bn by 2026. Newron reported that Supernus is developing Xadago sales but so far there is no overall gain after a mid-2020 dip. Under the sublicence, Newron shares 50% of Zambon’s US royalties. Given the importance of the US market, this sharing arrangement might be a drag on royalty growth. From 2017, Eisai has had the exclusive rights to market safinamide in Japan, as well as to develop and market safinamide in Asia. Safinamide (the generic name for Xadago but branded as Equfina) was approved in Japan in late 2019. Newron does not disclose sales by region, so detailed forecasting is not feasible.

Xadago is indicated for the treatment of adult patients with idiopathic PD as add-on therapy to a stable dose of levodopa (L-DOPA) alone or in combination with other PD medicinal products in mid- to late-stage fluctuating patients. Dopamine is a major brain neurotransmitter. A symptom of the neurodegeneration seen in PD is lower levels of dopamine. Hence, boosting dopamine levels relieves PD symptoms. It does not stop the fundamental disease progression.

One of the issues with standard oral dosing of dopamine (given as the prodrug L-DOPA) is inconsistent blood levels, which lead to ‘off’ periods when PD symptoms return. MOA-B inhibitors are used as one of several possible therapeutics to limit off time. Xadago is a monoamine oxidase type B (MOA-B) inhibitor. MOA is the enzyme that breaks down dopamine, so reducing its MOA activity gives more consistent brain dopamine levels. Other treatments to deal with this include direct continuous delivery of L-DOPA gel to the small intestine via an external pump, but this is very expensive. Deep brain stimulation is also used.

In the MOA-B class, the leader is Azilect (rasagiline, Teva), approved by the FDA in 2006 and EMA in 2005. Rasagiline is an irreversible MOA-B inhibitor, which means it destroys the enzyme activity after it binds. Other MOA-B drugs like Xadago and Zelapar (selegiline) are reversible. Rasagiline started to become generic in 2016. It has a wide label for use alone or in combination with other PD medications. Azilect had peak sales in 2014 of $519m (EvaluatePharma) but generics are now under $50m in Europe and the US combined. Due to a late approval, it is a branded product in Japan (Takeda), accounting for about half of global sales. Selegiline seems never to have gained a significant market share with very low sales.

A noted complication of L-DOPA treatment, the main PD therapy, is dyskinesia: uncontrolled tremors that can occur both when treatment is working (‘on’) and in ‘off’ periods. Dyskinesia is a very complex condition and medical opinion is still uncertain about why it occurs and how to treat it. It develops progressively in many patients after several years of L-DOPA therapy. This is called LID and is the subject of further Xadago development by Newron.

Under an agreement reached in March 20201, Newron and its partner Zambon agreed to run a trial to formally gain an indication in LID. This is particularly important for US marketing and could lead to a significant rise in sales as it could provide a validated therapeutic gain over standard generics. Newron notes that anecdotally, doctors observe less dyskinesia in Xadago treated patients, but this effect needs to be validated and quantified. If successful, Newron would receive a €4m milestone on approval and a marginally enhanced royalty rate on all Xadago sales.

The trial will be funded 50:50 by Newron and Zambon. Newron will run the study and expects its share to be under €10m. The trial is stated by Newron to be a six-month, double-blind, placebo-controlled study enrolling 223 patients. Patients will have pre-existing dyskinesia and will need to complete daily diaries to record dyskinesia using a Dyskinesia Rating Scale. This study will take about two years to recruit, run and analyse and involve 30–40 centres. It is expected to start in late 2021 so could announce data by early 2024 and the new indication might be fully marketed by 2025.

Newron has finished recruiting for the four-week exploratory safety trial (Study 008) in 138 patients with chronic schizophrenia who are taking an established therapy (NCT04461119). Doses tested were 7.5mg and 15mg given twice a day versus a placebo arm. Results are now expected by the end of March 2021. Given that evenamide has a novel mode of action on voltage-gated sodium channels (in effect, it suppresses abnormal activity (rapid firing) while allowing normal nerve function (firing at longer intervals)). If fully developed, it might have wider indications.

Newron now plans to start two pivotal studies in H221. These comprise Study 003 in patients with inadequate, worsening response to antipsychotic agents plus Study 004 in patients who are resistant to clozapine. Each is expected to recruit about 150 patients who will be dosed for eight weeks each. The trials are estimated by Newron to cost €30m.

These trials could both complete by the end of 2022. In that event, marketing could be possible by 2024. This would be earlier than we last projected (formerly 2025) and so boost value. However, we await the next set of data before adjusting our financial model. Although the trials can be funded to late 2022, assuming no delays, there are then further regulatory and marketing costs, so we anticipate a deal with an upfront and milestones to help fund the project.

One limitation on deal value is the limited patent life. The core patent expires in 2028 but patent extensions and data protection (requiring any generics to run a trial) will give an extended protection period. As before, we assume US protection up to 2033.

Newron is seeking a partner for the more general indication of inadequate response to current atypical antipsychotic agents. This is a general schizophrenia indication for patients who are not well controlled on their current anti-psychotic medicine, about 70% of cases. In Newron’s estimates, this could have a market potential of over US$1bn a year. In the general anti-psychotic market, the leading agents are now generic, but evenamide will add onto current therapy rather than seek to replace it. Both the general and clozapine studies need to be successful for FDA approval in this wider indication and safety would need to be acceptable.

Newron is considering direct marketing of evenamide for the focused indication of clozapine-resistant adjuvant treatment. Clozapine is currently the most effective antipsychotic agent, but about 30–50% of patients become resistant, giving no further options. Newron estimates there are 20,000 clozapine-refractory patients in the US. An approval in this indication would give a solid market base that Newron estimates could be worth several hundred million dollars. Note that meeting the primary endpoint in the single clozapine-resistant study would potentially suffice for this indication.

Reported revenues in FY20 were €5.26m (FY19: €7.04m comprising €4.76m plus a €2.28m milestone). There was 10.1% Xadago royalty growth that reflected steady European performance but with a mid-year dip in US royalties, noted by Newron, due to the marketing partner change over.

Costs in FY20 were predominantly R&D of €14.8m (€17.4m) after lower clinical costs. R&D is reported after R&D tax credits of €1.4m (FY19: €5m). General and administrative costs were €8.14m (FY19: €9.86m). There were other costs of €0.33m (€0.63m). There was a tax charge of €1.3m as a one-off expense related to the revaluation of IP. The reported loss of €20.96m (FY19: €20.21m) includes €1.46m of non-cash share option expenses (FY19: €2.13m). Depreciation and amortisation was €0.2m, mostly on leased assets. There were various interest charges. The €2.2m of financial expense (€2.2m interest plus warrant charges) was an accrual of European Investment Bank (EIB) interest.

The reported operating cash outflow in FY20 (inc tax and interest) was €15.59m (FY19: €21.98m). Cash fell from €22.05m in December 2019 to €13.21m. Newron has other financial assets of €18.04m (FY19: €17.11m). This gave €31.24m of cash and investments in December 2020.

In 2020, Newron drew a €7.5m EIB loan (after drawing €17.5m in 2019) and still has €15m available to draw down. This takes the outstanding EIB loan on the balance sheet to €25.67m as of 31 December 2020 (including the fair value of warrants). In our model, we show the EIB loan being fully drawn during 2022 taking the balance sheet loan liabilities (plus warrants) to around €41m.

In addition, over the two-year period 2021–22, there should be over €10m of Xadago royalties plus tax credits of about €5–6m. In Newron’s estimate, this gives potentially €61m of funding. Of this, trials will require at least €40m and general expenses and core R&D probably at least €20m. This indicates that Newron, depending on cash flow timings and trial speed, has resources until about the end of 2022. An evenamide deal with a substantial upfront by late 2021/H122 should offset some clinical costs and provide more cash. These extra resources could be used to acquire new projects for internal development. If a deal is not available, further funding will be required.

Note that in compiling forecasts for Newron, there are non-cash accounting adjustments that make precise forecasting of reported numbers difficult. In particular, Italian tax credits change and are used to offset personal taxes rather than being cash payments. Newron holds €12.58m of non-current tax credit receivables and aims to use €3.3m (in short-term receivables) in 2021. It can only claim up to €3m per year.

There are 17.8m shares in issue (unchanged) plus 1.85m employee options and 0.5m warrants held by the EIB as part of the loan.

There is funding and anticipated cash flows to support full development until late 2022, partly from an unused €15m EIB loan facility. An evenamide deal would extend this. Our current model (dated 1 January 2020) indicates a value of about CHF121m (CHF6.8/share). This included CHF49m of cash, now CHF34m (current rate of CHF1.11/€). Offsetting the lower cash level as of 1 January 2021 is the removal of 2020 costs from the calculation. As these values are similar, we intend to fully revise the value model once we have the Study 008 data.