NeuroVive Pharmaceutical — Licensing deal and R&D pipeline progress

NeuroVive is a global biopharma company based in Sweden with broad expertise in mitochondrial medicine. The company traces its roots to the discovery by co-founder and current CSO Eskil Elmér and his colleagues that ciclosporin has potent neuroprotection properties. Impaired mitochondrial function has been shown to have a profound effect on secondary cell injury. NeuroVive was founded in 2000 (then called NeuroPharma) and has since maintained its focus on mitochondrial medicine. As of end-H118, it employed around 10 employees. However, the company emphasises its wide network of academic partners and R&D organisations, which facilitates innovative and flexible studies to advance the R&D programmes. NeuroVive owns 82.5% of the Hong Kong-based subsidiary, NeuroVive Pharmaceutical Asia, and has a 10% equity stake in UK-based, privately owned Isomerase Therapeutics (acquired in 2016), a drug discovery expert in the area of microbial natural products. Isomerase helped NeuroVive to expand its pipeline with novel mechanism of action compounds, including new-generation succinate prodrugs and sanglifehrin-based cyclophilin inhibitors. NeuroVive shares were listed on Nasdaq Stockholm in April 2013. The company’s shares also trade on the US OTC Markets Group’s OTCQX segment.

We value NeuroVive at SEK1.64bn or SEK17.9/share, based on a risk-adjusted NPV model using a 12.5% discount rate, including SEK51.9m net cash at end-Q218. We include five of the company’s seven programmes/assets, but for the time being exclude the earliest assets NVP025 and NVP022 as NeuroVive is yet to provide more details about the further development plan or in vivo proof-of-concept data. NeuroVive’s strategy is to develop core products internally and out-license non-core projects, which is reflected by assumed licensing deals in our model.

NeuroVive had cash and cash equivalents of SEK51.9m at the end of Q218 and was debt free. We have fine-tuned our estimates and expect operating losses of SEK83.7m and SEK130.1m in 2018 and 2019 respectively. NeuroVive does not provide guidance on timing for its cash reach (instead it breaks down the milestones that can be achieved with current cash) but, according to our model and based on current R&D plans, cash reach is into 2019, beyond several R&D events. Notably, we do not take into account revenues from any potential licensing-related income in our financial forecasts.

NeuroVive is subject to the usual risks associated with drug development. The company is mainly an early-stage drug developer, and value creation in the foreseeable future will therefore depend on successful R&D progress and any potential partnering activities. The biggest near-term development sensitivity is related to the most advanced products, NeuroSTAT and KL1333. NeuroVive continues to be in an arbitration dispute with CicloMulsion AG regarding the royalty payments in the original licensing deal, when NeuroVive in-licensed CicloMulsion technology. NeuroVive subsequently discontinued the CicloMulsion product for cardiovascular indications and, so far, the outcome in the arbitration case remains uncertain.

NeuroVive specialises in mitochondrial medicine, with its most advanced assets in traumatic brain injury and genetic mitochondrial diseases. The company employs a dual business model:

NeuroVive has made recent progress across all of its development programmes (Exhibits 1 and 2).

KL1333 is a small molecule in-licensed from Yungjin Pharm (a diversified South Korean pharmaceutical company) in May 2017. It targets genetic mitochondrial diseases and is intended for chronic oral use in a variety of mitochondrial diseases. KL1333 was originally developed by KT&G Life Sciences, which was acquired by Yungjin in January 2017.

Yungjin Pharm is developing the drug in South Korea and Japan, and NeuroVive has rights to the rest of world (more detailed deal terms can be found in our initiation report). Recently, NeuroVive announced positive results from the KL1333 Phase I trial in healthy volunteers performed by partner Yungjin Pharm in South Korea (n=60). The researchers found that KL1333 has a favourable PK profile and no serious adverse events were observed.

Following this, NeuroVive plans to start its own Phase Ib study in Europe (UK sites) in H218, with results expected in H119. It has not disclosed the full trial design yet, but said that it will be a multiple ascending dose (MAD) trial with KL1333 in healthy volunteers as well as patients with mitochondrial genetic disorders, focusing on MELAS and mitochondrial DNA deletion syndromes (PEO, KSS). According to management, NeuroVive’s selected CRO has begun working and the trial is expected to commence in H218.

Exploratory Phase II trial(s) should start within 12–18 months, followed by a confirmatory trial that is expected to be of limited size due to the rarity of the conditions in this orphan drug development space. Open-label studies may also be considered for ultra-rare conditions, which could lead to an early market introduction in 2023, with expansion to the main groups of mitochondrial diseases in subsequent years.

KL1333 is a novel NAD+ modulator, which interacts with NQO1 (NAD(P)H:quinone oxidoreductase 1) as a substrate and regulates the levels of nicotinamide adenine dinucleotide (NAD+). NAD+ is a coenzyme necessary for many cellular metabolism processes including mitochondrial function and production of ATP in the electron transport chain. Increased NAD+ could therefore improve mitochondrial function in various mitochondrial diseases. NAD+ can be synthesized de novo (from scratch) or via a salvage pathway, which uses compounds acquired via diet (B3 vitamin niacin). Newer third-party data suggested that NADH conversion to NAD+ can also be performed by enzymes such as NQO1 (Gi-Su Oh et al, 2013) (Exhibit 4).

A recent peer-reviewed publication from Yungjin researchers described how KL1333 increases intracellular NAD+ via the enzyme NQO1 in an in vitro model of cell cultures taken from MELAS patients. The results include:

These results suggest that KL1333 could be used to treat MELAS effectively by modulating intracellular NAD+ levels via NQO1 and may be more effective than idebenone for the treatment of mitochondrial diseases. The indication for KL1333, as an activator of the SIRT1/AMPK/PGC-1α signalling network, could also be extended to age-related and metabolic diseases such as neurodegeneration, diabetes and non-alcoholic fatty liver diseases.

Santhera’s idebenone (Raxone, synthetic analogue of Coenzyme Q10), which to our knowledge was the first and remains the only drug approved specifically for mitochondrial disease LHON, also binds to NQO1 and increases NAD+. Idebenone is a synthetic analogue of Coenzyme Q10, while KL1333 is a novel NAD+ modulator, so differentiated from idebenone. Raxone sales in 2017 were CHF23m, while the consensus estimate for 2024 is CHF628m, which also includes the second indication for which Raxone is being developed, Duchenne muscular dystrophy (EvaluatePharma).

NeuroSTAT is NeuroVive’s innovative, patent-protected formulation of ciclosporin without the use of Cremophor or ethanol, hence differentiated from other formulations of ciclosporin in the market (see our initiation report for detailed analysis of NeuroSTAT’s competitive edge). There is still no neuroprotective treatment available for TBI. In May 2017, NeuroVive announced positive findings from both the preclinical and clinical studies it had been conducting at that time and started preparations for the proof-of-concept Phase IIb.

One of these, the Phase IIa Copenhagen Head Injury Ciclosporin (CHIC) study, explored NeuroSTAT in severe TBI patients. The trial was conducted at Rigshospitalet in Copenhagen, Denmark and was complemented by an experimental large animal (piglets) study conducted in collaboration with the University of Pennsylvania, US. The purpose of both studies was to accumulate pharmacokinetic/pharmacodynamic (PK/PD) data, confirm safety and understand which factors are important in designing efficacy trials. TBI varies significantly in each case and can be characterised as ‘no two similar cases exist’. Therefore, defining a target patient population and which measures are the best to evaluate efficacy need careful consideration before starting efficacy trials.

The Phase IIa CHIC trial was an open-label study. It aimed to recruit 20 severe TBI patients and explore PK/PD, safety and exploratory outcomes tests, such as electroencephalography changes. Two dosing regimens were used: 5mg/kg/day and 10mg/kg/day. Preliminary disclosed findings show that dose-dependent concentration levels can be measured in the blood and that NeuroSTAT reaches the CNS (target tissue), which means it passes the blood-brain barrier. The safety profile was confirmed.

On 4 October 2018, NeuroVive, in collaboration with researchers at the University of Florida released a biomarker analysis from this study, which appeared to provide initial signs of NeuroSTAT’s effect on the secondary brain injury cascade. Four novel biomarkers, GFAP (Glial fibrillary acidic protein), UCH-L1 (Ubiquitin carboxy-terminal hydrolase L1), NF-L (Neurofilament Light) and Tau, were analysed in cerebrospinal fluid samples taken from the severe brain trauma patients in the CHIC study. The results showed that the administration of NeuroSTAT had a positive longitudinal effect on the levels of these biomarkers potentially alleviating secondary brain injury.

This was a non-controlled and open-label study; therefore, no concrete conclusions can be made at this point. On the other hand, it was a Phase IIa study with representative patients and CSF samples were taken on several occasions during the trial, so it represents a good starting point to understand how to best use the biomarkers in upcoming randomised trials, in our view. Having biochemical markers in the trial could lead to better understanding as to which patients NeuroSTAT is best suited to (patient stratification: mild, moderate, severe brain trauma). They also provide a sensitive way to evaluate NeuroSTAT’s treatment effect (follow-up) and make prognosis more accurate (outcomes). Detailed data from the CHIC study will be published at a future date.

All four biomarkers analysed in the study come from cutting-edge research, and recently the FDA approved GFAP and UCH-L1 blood tests for diagnosis of mild TBI. Until recently, no biochemical measures existed in practice to evaluate TBI patients; this mainly depended on physical examination and imaging studies. Given that NeuroSTAT’s mechanism of action is to limit secondary injury to brain tissue, the availability of biochemical biomarkers adds a new dimension of patient evaluation.

Until recently, only headline findings have been released from the experimental preclinical study with large animals (piglets), but a new peer-reviewed paper published in the Journal of Neurotrauma in July 2018 described the findings in detail. The importance of this study is that, while it is preclinical, it was carried out in a highly controlled manner similar to clinical trials, ie blinded and placebo controlled. There is a high degree of translatability of findings into humans due to the substantial structural, anatomical similarities between human and pig brains (similar gyral pattern and distribution of grey and white matter). In addition, the outcome was measured using MRI and the upcoming Phase IIb trial is likely to use same method.

The model involved standardised focal cranial impact resulting in a brain injury while the piglets were sedated. The study consisted of three different sub-studies, all of which were randomised. The first two sub-studies established the PK/PD profile, proving that NeuroSTAT reaches the CNS in a dose-dependent manner and established the recommended dose of 20mg/kg/day for the efficacy part of the trial.

The third sub-study evaluated the efficacy of the recommended dose after five days of treatment (randomized, blinded). A total of 37 animals underwent randomisation, of which 24 received the full dose (n=11 NeuroSTAT; and n=13 placebo; the remaining 13 animals did not receive the full dose due to various technical issues, such as failed intubation or anaesthesia). The findings include:

NeuroVive is making preparations for the Phase IIb trial and has now received positive opinions from both the FDA and EMA. The details of the trial design have not yet been released, but important elements include novel biomarker endpoints and a subpopulation of TBI patients with defined trauma. The latter is especially beneficial for NeuroVive, in our view, as TBI can present a very diverse pathology. The ability to select patients with similar trauma-features greatly improves the consistency of the study. NeuroVive appears to be the first company to adopt a precision medicine approach in the TBI field, also highlighted by the recent partnerships with University of Florida for biomarker development and the TRACK-TBI consortium.

Submission of the IND application is the next step. The precise timing of the start of the study has not been announced yet. We believe NeuroVive’s current cash position is sufficient to continue the preparations, but it will need new funds or a partner to start and run the study. Our model includes patient enrolment start in 2019. Enrolment rate is inherently unpredictable in acute treatment studies, but NeuroVive has indicated plans for a multi-centre design in Europe and the US. We expect data readout to be possible by 2021, which can be a substantial value trigger for the programme and enable progress to a confirmatory Phase III trial thereafter.

In November 2017, NeuroVive announced that it had selected the lead compound NV354 for the NVP015 programme, which is focusing on the development of succinate prodrugs targeting complex I deficiency (Exhibit 7). Mitochondrial complex I deficiency is the most prevalent defect in the respiratory chain in paediatric mitochondrial diseases (around 50%) and clinically presents as a group of syndromes, which can be caused by changes in either the nuclear or mitochondrial genome.1 In vivo preclinical proof-of-principle results for the lead compound are expected in 2018. See our initiation for a more detailed background on the NVP015 programme and mitochondrial complex I deficiency.

As part of the NPV015 discovery programme, NeuroVive evaluated many other succinate prodrugs. On 18 June 2018, the company announced that it had out-licensed a subset of these succinate prodrugs to private biotech BridgeBio, based in California, US. BridgeBio plans to develop these compounds for the localized treatment of Leber’s Hereditary Optic Neuropathy (LHON) in its new subsidiary Fortify Therapeutics. The total deal value could reach $60m (a ‘limited’ amount will be provided upfront for research funding). The disclosed deal terms are shown in Exhibit 8.

BridgeBio has a unique business model which involves acquiring rare genetic disease assets and forming subsidiaries around the assets for further development, a so-called hub and spoke corporate structure. It now has 19 assets in its subsidiaries. BridgeBio believes that this new approach to drug development will facilitate focused R&D at the level of each asset. Most recently, in June 2018, it in-licensed Phase II asset infigratinib from Novartis to treat rare cancers, and created a new subsidiary around it, QED Therapeutics, along with a $65m commitment.

Only a subset of NVP015 chemistry has been out-licensed which, BridgeBio believes, has best potential specifically for local delivery to the eye in LHON patients. The NPV015 programme could potentially cover various different mitochondrial diseases and NeuroVive stressed that it is continuing its internal programme for NVP015 outside local delivery for LHON. We believe the new partnership:

The aforementioned Raxone (Santhera Pharmaceuticals) is the only drug currently marketed for LHON. It was approved by the EMA in 2015 and fully reimbursed in eight European countries, but is not yet approved by the FDA. Other late-stage programmes in the LHON space include GS010 (GenSight), an NADH-ubiquinone oxidoreductase chain 4 gene therapy currently in Phase III, and Ocuvia (Stealth BioTherapeutics), which also improves the function of the electron transport chain currently in Phase II. All three of these therapies have orphan drug designation from the EMA and/or FDA.

The NVP025 programme targets mitochondrial myopathies, which are a subgroup of neuromuscular diseases with hallmark symptoms being muscle weakness, exercise intolerance, fatigue and heart problems, often accompanied by neurological symptoms such as dementia, movement diseases, stroke-like episodes, deafness and blindness. NVP025 is an early programme with the goal of developing a compound that prevents the weakening of muscle fibres associated with these diseases. The NVP025 mechanism of action is different to that of succinate prodrugs or KL1333 and comes from NeuroVive’s sangamide class of compounds. In January 2017, NeuroVive signed an agreement with Karolinska Institutet to study a potent sangamide in preclinical models in mitochondrial myopathy. The Karolinska Institutet research team previously published results showing that another cyclophilin inhibitor, ciclosporin, prevented muscle fibre weakness in an experimental model of mitochondrial myopathy. The company expects to select a lead drug candidate in 2018.

NV556 is a novel cyclophilin inhibitor originating from NeuroVive’s sangamide class compounds, which are derivatives of sanglifehrin A. Sanglifehrin A, like ciclosporin, inhibits cyclophilin but binds to a different site from ciclosporin. NV556 has a direct antifibrotic mechanism of action in the liver. The asset has undergone extensive preclinical development, has favourable drug-like properties and confirmed antifibrotic effect in several animal models.2 It is the most advanced asset in the portfolio for out-licensing and NeuroVive aims to license it as soon as possible.

NeuroVive initiated another programme targeting NASH, in which the company is testing model compounds. Few details had been disclosed about NVP022, until the company reported data at the Liver Meeting organised by the AASLD on 20-24 October 2017. NVP022 is a protonophore, which acts as a ‘mild’ liver-targeted uncoupler. Protonophores can transport protons across mitochondria membranes and disrupt the proton gradient that is needed to produce ATP in mitochondria. Mitochondrial protonophores have been shown in the past to positively affect NASH and diabetes biomarkers, but had limitations due to the risk of overdosing easily (maximum uncoupling effect). NVP022 is a mild, liver-targeted uncoupler and could have a direct effect on metabolic components of NASH. NeuroVive is currently evaluating compounds with the expectation of selecting a lead compound in 2018.

The addition of the HCC indication to NeuroVive’s portfolio was an evolution of the company’s research in NASH, in addition to the well-known feature that cyclophilins are overexpressed in HCC. NVP024 is an early preclinical programme, in which NeuroVive is testing novel model compounds. The mechanism of action is also based on sanglifehrin A and cyclophilin inhibition, but the company is using its second-generation compounds. See our initiation report for more detail. Currently, NeuroVive continues early preclinical development.

NeuroVive is subject to the usual risks associated with drug development, including clinical development delays or failures, regulatory risks, competitor successes, partnering setbacks and financing risks. NeuroVive is mainly an early-stage drug developer, and in the foreseeable future value creation will therefore depend on successful R&D progress and any potential partnering activities. The biggest near-term development sensitivity is related to the most advanced products – NeuroSTAT and KL1333, which are both expected to be developed in house. The company employs a dual strategy and has a portfolio of assets for out-licensing. Typically, the timing of licensing deals is difficult to forecast. NeuroVive continues to be in an arbitration dispute with CicloMulsion regarding the royalty payments in the original licensing deal. NeuroVive in-licensed CicloMulsion technology, which was used in post-operative acute kidney injury and myocardial infarction studies, but has discontinued the development of CicloMulsion product for cardiovascular indications. The outcome of the arbitration case remains uncertain.

Our updated valuation of NeuroVive is largely similar at SEK1.64bn or SEK17.9/share compared to our last published SEK1.62bn or SEK17.7/share. As previously, in our valuation we include clinical-stage NeuroSTAT (traumatic brain injury) and KL1333 (genetic mitochondrial disorders) and the advanced preclinical products. We continue to exclude NVP025 (mitochondrial myopathy) and NVP022 (NASH) for the time being, as both are at an early stage. We maintain all our previous valuation assumptions.

We have already included a licensing deal for NVP015 in our model and for the time being we do not make any adjustment to it. This is primarily because our risk-adjusted NPV includes the potential of NVP015 in a broad set of mitochondrial diseases with impaired complex I function. Only a subset of NPV015 chemistry has been out-licensed and only for local treatment of one indication, therefore we consider the potential $60m in milestones (not risk-adjusted, not discounted) to be already reflected in our calculated, risk-adjusted and discounted present value for the whole NVP015 project of $33m.

NeuroVive reported other income of SEK1.5m comprising a research contribution from BridgeBio of SEK876k and a research grant from Vinnova (a Swedish innovation agency) of SEK576k in. Total operating H118 expenses were SEK39.6m, of which SEK23.0m was spent on R&D. Given that NeuroVive plans to initiate two clinical studies in the near term, we expect R&D costs to pick up. Consequently, we maintain our increasing R&D cost estimates of SEK56.6m and SEK103.2m for 2018 and 2019 respectively. We have slightly fine-tuned our other estimates (other operating income/expenses and net financial loss) with minimal net effect on EPS.

Cash at end-Q218 was SEK51.9m which, according to our model, should fund the company’s activities into 2019. Further capital will be required to complete the NeuroSTAT Phase IIb trial, but results from the Phase I MAD study with KL1333 should be in H119, which could provide a catalyst for the share price. NeuroVive indicated that it is also working towards securing co-funding for the Phase IIb NeuroSTAT trial, as well as non-dilutive funding from a potential licensing deal for NV556 (NASH) in H119 and potential further research grants. None of the milestones from the BridgeBio deal is included in our model, which could provide additional runway. Furthermore, warrants exercisable in November 2018 could bring in another SEK37.3m gross in cash (at a strike price of SEK3.80).