AlzeCure Pharma — Well-balanced pipeline for Alzheimer’s and pain

AlzeCure Pharma is a clinical-stage biotech company based at the Novum Science Building associated with the Karolinska Institute in Stockholm, Sweden, with c 10 employees, who combined have over 100 years of experience from large pharma companies. The current management team formed part of AstraZeneca’s neurology and pain division and were involved in the development of both symptomatic and disease-modifying drugs for AD. When AstraZeneca decided to focus on other indications, that team partnered with Professor Bengt Winblad (one of the world’s most cited researchers in neurodegenerative diseases) at the Karolinska Institute in 2012. The Swedish Alzheimer’s Foundation provided early funding and soon thereafter a commercial company was incorporated in 2016. Since then, it has raised c SEK100m in financing rounds and listed on Nasdaq First North Premier in November 2018. The IPO proceeds were c SEK200m (priced at SEK14 per share).

We value AlzeCure at SEK729m or SEK19.3 per share (cash of SEK112m at end Q420, no debt; 12.5% discount rate). We modelled ACD856 (NeuroRestore) as a symptomatic AD drug with a cognition-improvement effect superior to that of existing generic AD drugs. We do not value Alzstatin program yet, but will reconsider once it will be close to or in the clinical development. Using the same model as for ACD856, but changing assumptions to a higher penetration rate and better pricing, we estimate potential peak sales for a Alzstatin disease-modifying AD drug could reach c $15bn (more details in the Valuation section). We include the clinical stage ACD440 asset from the Painless platform in our valuation, but similarly will reconsider TrkA- negative allosteric modulator (NAM) once it makes progress.

AlzeCure reports no income and in FY20 it reported SEK71.6m in operating loss, of which most (SEK62.4m) was booked as R&D expense. In FY19 the operating loss was SEK50.9m indicating that spending increased due to more R&D activities. We expect the spending to increase somewhat further in 2021 as both clinical trials are in full swing and forecast SEK82.0m in operating loss. Existing cash of SEK112m at the end of FY20 should provide funding into 2022. We note the existing budget is sufficient to reach the readouts from both clinical trials.

AlzeCure is exposed to the drug-development risks typically associated with pure play biotechs, namely clinical development delays or failures, IP protection, regulatory risks, competitor successes, partnering setbacks and financing and commercial risks. Its focus on developing drugs for AD could be considered higher risk, but also higher reward in our opinion. However, having projects that could have potential beyond AD does mitigate this risk to an extent. While AlzeCure has a sufficient cash runaway to complete the ongoing clinical trials for ACD856 and ACD440, additional funding will be required to progress these and facilitate the progression of preclinical candidates. As the company has guided, assets will likely need to be partnered at some point as later-stage CNS studies can be costly due to large populations and study duration, hence the ability to successfully negotiate a deal is a key sensitivity.

AlzeCure has three platforms: NeuroRestore, Alzstatin and Painless (Exhibit 1). Assets within the two platforms targeting AD are wholly-owned. The Painless platform consists of two assets that are unrelated from the mechanism of action perspective. ACD440 (Painless) was in-licensed at rather beneficial terms (Exhibit 1 notes), while TrkA-NAM (Painless) is wholly-owned and has an interesting discovery history. Pharmacologically speaking, TrkA-NAMs have the opposite activity of NeuroRestore assets (which are TrkA/TrkB positive allosteric modulators or PAMs). AlzeCure somewhat serendipitously discovered these NAM molecules while working on NeuroRestore. Modulators help either activate or deactivate receptors via a separate site, so do not compete with inhibitors or agonists.

Two Phase I clinical trials are up and running with ACD856 and ACD440. The existing funding is sufficient to complete them, so the data readouts expected over the next 18 months will provide significant catalysts. In parallel, progression of preclinical assets can also be expected, which will add to the news flow. The company’s overall strategy is to create value by advancing assets from early preclinical through to mid-stage clinical development and then to seek partnering deals.

AlzeCure’s NeuroRestore platform comprises drugs (ACD856, ACD857 and ACD 855) designed to improve cognitive function in neurological disorders, but could also have the potential to show a disease modifying effect. These are novel, first-in-class, small molecule drugs that all exhibit the same mechanism of action: namely they enhance the signalling of neurotrophins, including NGF and BDNF through positive modulation of the TrkA and TrkB receptors, respectively. NGF and BDNF are part of the family of neurotrophin proteins that signal via tropomyosin receptor kinases (TrkA, TrkB and Trk C) and the p75 neurotrophin receptor (p75NTR); this signalling cascade helps to regulate neuronal survival, differentiation, neurogenesis and synaptic plasticity.

A variety of preclinical in vitro and in vivo models have established NGF/TrkA or BDNF/TrkB signalling as therapeutic target in numerous neurological disorders. However, the utility of using NGF or BDNF (recombinant or mimetics) as a systemically delivered drugs is limited due to their restricted ability to cross the blood-brain barrier, poor plasma stability and promiscuous binding across multiple receptors subtypes, which could have undesirable side effects. An approach to use small molecule TrkA and TrkB modulators, therefore, has gained interest. Based on the preclinical data so far presented by AlzeCure, ACD856 has overcome these limitations.

Having successfully completed a microdosing Phase 0 study for the lead compound ACD856 in June 2020, AlzeCure has initiated a Phase I trial with ACD856 in November 2020. The trial is currently recruiting patients and initial safety and tolerability data from single ascending dose (SAD) part of the study are expected by mid-2021, with multiple ascending dose (MAD) data by end-2021. In an extension phase, a separate signal-detection study will be conducted to assess early cognitive effects of the compound, with data expected during H222.

The Phase I trial will establish safety and tolerability, as well as confirm its pharmacokinetic (PK) profile. This latter will be particularly relevant, in our view, given predecessor compound ACD855 was initially developed as an AD drug. However, Phase I data indicated that a longer than expected half-life meant it would be unsuitable for a systemic, oral drug. In May 2019, the development was redirected to focus on eye diseases, while the back-up compound ACD856 was selected as the lead asset. We are reassured by the successful completion of the microdosing Phase 0 trial, following which the company reported that ACD856 has a suitable PK profile. ACD857 is another compound within the NeuroRestore platform that could be also used as a back-up compound or could be developed to address other relevant CNS indications where cognitive function is disturbed, such as certain sleep disorders, traumatic brain injury and Parkinson’s disease (PD). The requisite IND enabling, preclinical studies started in December 2020.

During 2020, AlzeCure published preclinical data with ACD856 at the AAT-ADPD (March 2020) and at CTAD (November 2020) conferences that indicate a more optimal PK profile has been achieved with ACD856, which should be amenable to once-daily dosing. We highlight the preclinical data disclosed at these conferences support the proposed development of ACD856 as a cognitive enhancing drug:

Loss of synaptic function is one of the hallmarks of AD and strongly correlates with the cognitive decline that occurs during disease progression. It has been extensively highlighted that the accumulation of toxic protein aggregates, such as Aβ oligomers/plaques or tau tangles, play a significant role in this neuronal degeneration through triggering toxic inflammatory responses (Koffie et al, 2011; Pepeu and Giovannini, 2017). More specifically, the loss of cholinergic neurons is a cardinal feature of AD, as evidenced by three of the FDA approved drugs, which all aim to correct cholinergic tone through inhibition of acetylcholinesterase. These cholinergic neurons are lost extensively in the basal forebrain and hippocampus and are strongly dependent on NGF for neuronal survival, with evidence indicating that Aβ can disrupt NGF metabolism. (Pepeu and Giovannini, 2017; Latina et al, 2017). Hence the view is that ACD856, which can positively modulate TrkA, could in theory correct this dysregulated NGF signalling and slow the cognitive decline. We highlight two seminal clinical studies that, despite their limitations, highlight interest in the area, as the settings in these trials were fairly complicated:

ACD856’s ability to also positively modulate TrkB provides it with an additional mechanism of action since BDNF has also been closely linked with the worsening of cognition in AD. BDNF is a polymorphic gene meaning that more than one allele occupies the gene’s location (eg genes that control hair colour are polymorphic, hence the wide range of colour pattern). The majority of polymorphisms are silent and they do not alter the function of a gene. In the BDNF case, a genetic single nucleotide polymorphism is found to occur that causes a slight change in the BDNF protein sequence (Val66Met). This specific polymorphism has been shown to result in a reduction of BDNF secretion in the hippocampus (Egan et al, 2003). Exhibit 3 highlights results from a longitudinal study of the Wisconsin Registry for Alzheimer's Prevention, in which patients underwent BDNF genotyping and cognitive assessments. Carriers of the BDNF-Val66Met polymorphism occurred in c 30% of the population and exhibited a more severe decline in both verbal learning/memory (Exhibit 3 left, p = 0.002) and speed/flexibility (right, p = 0.02), compared to BDNF Val/Val homozygotes.

Findings from a separate study of the Dominantly Inherited Alzheimer Network (DIAN) have further supported the BDNF-Val66Met polymorphism theory. Preclinical AD patients carrying BDNF-Val66Met exhibited worse memory performance. Especially notable from this study was the finding that carriers had lower hippocampal glucose metabolism and increased cerebrospinal fluid (CSF) levels of tau and p-tau, suggesting the BDNF-Val66Met polymorphism could exacerbate AD pathogenesis downstream of Aβ. The authors concluded: ‘strategies designed to increase CNS BDNF levels may be a viable therapeutic alternative or addition to those which seek to reduce the neurotoxic effects of Aβ’ (Lim et al, 2016; Rogaeva and Schmitt-Ulms, 2016). In our view, this indicates that ACD856 could even have a disease modifying effect for AD.

The consensus in Alzheimer’s is that disease modifying therapy (DMT) would represent a breakthrough. However, since there are no such drugs and development of DMTs has been spectacularly unsuccessful, a clear definition of what constitutes DMT and how it differs from symptomatic therapy is also lacking (J Cummings and N Fox, 2017). According to the recent EMEA guidance document, ‘symptomatic improvement is defined as a treatment effect that does not change the overall course of the disease’, whereas ‘a disease modifying therapy is one which delays the underlying pathological or pathophysiological disease processes’, which would correlate with ‘a significant effect on adequately validated biomarkers’. The choice of biomarker as well as the type of analysis is left open.

Disease modification can be understood as different outcomes that range from cure to slowing of the natural course of the disease and increasing patient survival. This concept is shown in Exhibit 4. Until more data become available for ACD856, it is unclear if AlzeCure will be attempting to show a disease modifying effect. Data expected from the Phase I extension cohort during H222 will look initially for a signal that ACD856 has a cognitive enhancing effect. We believe biomarker data will likely be required to show ACD856 has a disease modifying potential.

The revenue potential of new entrants in the AD market would depend on the benefit relative to the currently approved symptomatic drugs. This could range from direct competition with existing drugs if efficacy is similar, to a market disruptor with a multibillion-dollar potential if the drug modifies the progression of the disease. There could also be a variety of scenarios in between these extremes, in our view:

Four pharmacologic treatments have been approved by major regulators and are established in clinical practice that can moderately improve symptoms. However, these do not slow AD’s progression nor improve patient survival. These are acetylcholinesterase inhibitors: donepezil, rivastigmine and galantamine, which are the standard first-line therapy and the only approved drugs for mild disease. Memantine (NMDA receptor antagonist) is approved for moderate to severe disease but is often used off-label in mild disease as well. Treatment options vary widely across geographies.

With regards to market size, currently the trend of ageing populations and the increasing prevalence of AD is mostly offset by the increasing use of lower-priced generics, as branded drug protection expires. However, due to a high unmet need and the vast patient population, any new, effective drug could disrupt the market. In our view, any new disease modifying drug would have a realistic potential to reach annual peak sales that are a multiple of the current combined sales of all generic AD drugs (in the Valuation section we show a scenario where a disease modifying drug could reach $15bn in peak sales). We note that Institute for Clinical and Economical Review (ICER) is due to publish a draft report on 5 May 2021 that will assess the comparative clinical effectiveness and value of aducanumab, and might better define pricing expectations for DMTs in AD.

From our research, we believe that AlzeCure has a first-in-class molecule with ACD856. Although we have identified a variety of developmental drugs with similar mechanisms of action to ACD856, we do not currently believe there to be anything that we would consider a direct competitor.

ATH-1017, a small molecule designed to increase HGF/Met activation, which can be viewed as a similar neurotrophin signalling cascade to NGF/TrkA and BNDF/TrkB, has also been implicated in the cognitive decline associated with AD and PD. Athira Pharma is developing it as a daily, subcutaneous injection and recently initiated two Phase II–III clinical trials in patients with mild-to-moderate AD. Based on its interactions with the FDA, Athira believes the Phase II/III LIFT-AD trial may provide registrational data if key secondary efficacy endpoints are meet (ADAS-Cog and ADAS-CGIC). Athira was valued at c $760m as of 16 March 2021.

PharmatrophiX recently completed a Phase IIa proof-of-concept trial for its p75NTR modulator LM11A-31, which again can be considered to work in a similar manner. We understand that the trial was completed in June 2020 and an analysis was expected Q420 but data have not yet been published.

The traditional amyloid hypothesis postulates that Aβ protein is the causative factor in AD. However, this has been a subject of much debate over the last decade, with multiple clinical trial failures of Aβ-targeting therapies seemingly vindicating the scepticism. As the science evolved, a rather simplistic view that removing Aβ plaques should lead to improvement has changed too. More recent theory updates focus on so called toxic forms of Aβ.

Drug candidates from the Alzstatin platform have potential as disease modifying drugs for AD, in our view, although both projects (ACD679 and ACD680) are in an early stage of development. These have a distinctly different mechanism from the Aβ-targeting antibodies, such as Biogen’s aducanumab, which have recently captured the attention of investors. These are small molecules capable of modulating γ-secretase, a target involved upstream in the production cascade of toxic Aβ-aggregates. Modulation of γ-secretase produces shorter, less toxic Aβ-peptides whilst not changing overall levels of Aβ. This is a novel approach where modulation rather than full inhibition of γ-secretase is expected to circumvent the safety issues previously found by large pharma with γ-secretase inhibitors. Given that safety issues have plagued the development of full γ-secretase targeting agents in the past, establishing a clean safety profile will be important. So, preclinical and Phase I safety data for ACD679 and ACD680 will be a meaningful catalyst for AlzeCure, in our view.

The Alzstatin platform has recently gained significant attention, primarily due to external factors; namely, the upcoming US FDA decision on whether it will approve Biogen’s aducanumab for AD, which has pushed the amyloid hypothesis back into focus. Should the FDA approve Biogen’s aducanumab, it will be a landmark decision in Alzheimer’s research, making it the first new AD drug approved in c 20 years and more importantly the first disease modifying drug. As described below, the data, however, were rather mixed, so the stakes for Biogen are high.

In short, the traditional amyloid hypothesis postulates that insoluble Aβ oligomers/plaques are the key neurotoxic culprit in AD and drugs targeting these could have a disease modifying effect. Vast resources in the industry have been aimed at intervening in the amyloid cascade to prevent or slow these toxic effects, however, no drugs have yet been successfully developed and approved. Aβ-targeting antibodies has been extensively pursued and we highlight three notable examples that are still being actively developed:

We have highlighted these three Aβ-targeting agents as they are the most advanced assets still being actively developed. Notable historical Phase III failures include: solanezumab (Eli Lilly), bapineuzumab (Pfizer, Janssen), gantenerumab (Roche, MorphoSys) and crenezumab (Roche, AC Immune).

Upstream targets in the amyloid cascade have also become a focus, namely β-secretase (BACE) and γ-secretase, which are two enzymes that process amyloid precursor protein (APP) into the Aβ monomer units that aggregate and result in the build-up of toxic Aβ-oligomers/plaques. As per the traditional amyloid hypothesis, inhibition of these upstream processes should prevent the formation of Aβ aggregates, pathological progression and the worsening of symptoms. BACE inhibitors were viewed as having great potential and many large pharma players had heavily invested in this strategy. However, 2018 alone saw three BACE inhibitors failing Phase III studies: verubecestat (Merck), atabecestat (Janssen) and lanabecestat (Eli Lilly, AstraZeneca).

Similarly, γ-secretase inhibitors were initially viewed as having a good potential, but failed to meet expectations, particularly with Eli Lilly’s semagecstat casting a shadow of safety concerns on the class. Semagacestat’s Phase III IDENTITY trials stopped prematurely in August 2010 due to the drug apparently worsening disease progression. We note one causal factor that was been attributed to this has been the high dose used in the trials resulted in sustained, complete inhibition of γ-secretase. Beyond processing APP into smaller Aβ monomer units, γ-secretase is thought to play a critical role in processing other important proteins such as Notch. It is thought that impairment of Notch signalling results in negative cognitive effects. The question that therefore remains is whether AlzeCure’s approach with Alzstatin, modulating rather than inhibiting γ-secretase, can spare critical signalling pathways like Notch and circumvent the safety issues previously found with semagacestat, but still maintain a similar effect on reducing downstream levels of toxic Aβ aggregates.

The question that therefore remains is whether AlzeCure’s approach with Alzstatin, modulating rather than inhibiting γ-secretase, can circumvent the safety issues previously found, but still maintain a similar effect on reducing downstream levels of toxic Aβ aggregates. Not all Aβ monomer units are equal and it is thought the longer units (Aβ1-40 and AB1-42) are more susceptible to aggregation, whereas shorter fragments (Aβ1-37 and Aβ1-38) are less prone to this, less neurotoxic and could in fact have an anti-amyloidogenic effect (Boy et al, 2019). The Alzstatin programme has been derived from and built on learning management took from an analogous programme at AstraZeneca, which generated compounds that achieved this: modulating γ-secretase to produce shorter, less toxic Aβ-peptides while not changing overall levels of Aβ and sparing Notch signalling. Worth noting is that in familial AD, mutations in the γ-secretase complex result in virtually the opposite effect of Alzstatin’s compounds. Familial AD is characterised by an early onset (between 30 and 60 years of age) and is inheritable.

We note others have also pursued this concept and conducted clinical studies, but none of these programmes seem to be active anymore:

ACD679 and ACD680 are progressing through pre-clinical development and updates on their progress can be expected during 2021. These assets are aiming for the disease modifying effect in AD, which is the ‘holy grail’ indication in a major CNS indication. We do not value the Alzstatin programme yet because of its early stage, but once the assets reach clinical development, value creation can be rapid.

Ongoing evolution of assets within the Painless platform will further diversify AlzeCure’s pipeline and, in our view, de-risks the investment case as the market opportunity for these innovative, non-opioid based analgesics could be significant. The platform currently consists of in-licensed asset ACD440, which is being developed as a topical treatment for neuropathic pain (Phase Ib), and TrkA-NAM for osteopathic pain (early preclinical).

ACD440 is a topical vanilloid receptor 1 (VR-1) antagonist currently in a Phase Ib trial. The VR-1, more commonly known as TRPV1 or the capsaicin receptor, has been a well-established target for the treatment of pain and has been implicated in the transduction of pain (nociceptive, inflammatory or neuropathic). In fact, topical capsaicin has been used as a homeopathic remedy for more than a century with the presumption that the ability to induce persistent receptor activation (a burning sensation) results in receptor desensitization and induction of an analgesic effects. We note several capsaicin-based topical treatments (Qutenza, Zacin and Axsain) are approved by regulators and prescribed for the treatment of pain associated with a variety of conditions, including postherpetic neuralgia, diabetic peripheral neuropathy and osteoarthritic pain.

The roots of this program stem back to AstraZeneca, which originally discovered and developed the compound as an oral drug in various pain indications. In January 2020, it was in-licensed from Acturum Life, which acquired rights to the drug-candidate when taking over the research sites in Södertälje, Sweden, previously occupied by AstraZeneca. We note the oral formulation of this had undergone extensive development at AstraZeneca but was discontinued. We understand this was related to the undesirable side effects resulting from systemic exposure, which appear to be a side effect of the class, but also coincided with when AstraZeneca exited from CNS research. Looking at a broader picture, extensive efforts have been made into developing VR-1 antagonists by various other companies (Amgen, Abbott, GSK, Merck/NeuroGen, Glenmark/Eli Lilly), but all of them were stopped in early clinical development, likely due to the same undesirable side effects resulting from systemic, oral exposure. Namely, hyperthermia and impairment of noxious heat sensation. Such high interest could be a result of the highly lucrative and large pain indications (osteoarthritis, dental pain, migraine).

ACD440 is a topical reformulation of the AstraZeneca’s original compound, which we believe could be a solution to the undesirable side effects, while retaining the efficacy seen in the oral formulation trials. In January 2021, AlzeCure initiated a relatively inexpensive (SEK10–15m) placebo-controlled Phase Ib trial, conducted in 24 healthy volunteers, for ACD440. We anticipate safety and tolerability to be reported during mid-2021, as well as some proof-of-mechanism data, which could act to validate the topical reformulation and its use in neuropathic pain. Thereafter, Alzecure has guided that preparations for a Phase IIa trial could start in H221 to enable trial initiation in 2022, which would give a better sense of the forms of neuropathic pain it will target. As ACD440 is in-licensed, we understand the initiation of a Phase IIa for ACD440 will trigger payment of an undisclosed milestone to Acturum Life, however, the deal terms have been described as very attractive by AlzeCure.

The potential indication currently is neuropathic pain, which is a symptom of a broad group of different underlying conditions. The key characteristic of this indication is the fact that this is a vast population of patients (diabetic nephropathy, herpetic neuralgia, cancer chemotherapy, autoimmune conditions, etc). Another key characteristic is the fact that in less than half of the patients pain control is unsatisfactory, so the unmet need is high (Sommer and Gruccu, 2017).

AlzeCure’s TrkA-NAM program stems from the high throughput screening efforts that generated compounds in the NeuroRestore platform, where it discovered selective negative allosteric modulators (NAM) for the TrkA receptor (note that the NeuroRestore programme has positive modulators, PAMs). Pharmacologically, these can be viewed as having inhibitory activity on NGF/TrkA neurotrophic signalling, which has been implicated to play a role in the pathogenesis of inflammatory pain (Skeratt et al, 2016).

Although a different mechanism of action, large pharma have committed significant resources to develop anti-NGF targeting antibodies for pain, with Pfizer/Lilly’s tanezumab undergoing c 21 and Regeneron/Teva’s fasinumab around seven Phase III trials. Development of these has been complicated, however, and safety concerns have plagued this class, with the FDA placing a clinical hold on Pfizer’s tanezumab trials in June 2010 due to some patients experiencing a rapidly progressing, severe form of osteoarthritis leading to joint replacements. Despite this and bolstered by vast market, such as osteoarthritic pain (even an imperfect drug with a smaller market share could be lucrative), large pharma pushed for an FDA assessment. An FDA AdComm is planned for tanezumab on 24–25 March 2021 ahead of its potential approval for moderate-to-severe osteoarthritic pain. Similarly, long-term safety data for fasinumab are expected H121 that could define its path to approval.

Despite these challenges, the extensive efforts from large pharma to develop these anti-NGF targeting antibodies highlight the significant commercial need to develop effective, non-opioid based treatments. AlzeCure believes the addressable population for TrkA-NAM is c 250–300m patients and that with its downstream mechanism of action, TrkA-NAM could potentially offer safety improvement over these anti-NGF antibodies, by selectively attenuating NGF/TrkA signalling while sparing NGF/p75NTR signalling. AlzeCure’s TrkA-NAM program is in an early preclinical development stage and successfully completed preclinical efficacy studies in December 2020. An oral candidate drug is expected to be nominated during H221.

The compounds in the NeuroRestore platform all stem from a high throughput screening effort conducted by AlzeCure at the Karolinska Institute back in 2013. AlzeCure optimised and developed these, filing primary, composition of matter patents for compounds in December 2017, which should provide exclusivity and market protection out until at least December 2037. We note that given the length of time it takes to conduct late-stage CNS clinical trials, patent term extensions will likely be filed once market authorisations have been granted and these could extend patent protection by up to five years. We understand AlzeCure plans to file formulation patents for ACD 440 that could provide it with additional IP protection.

AlzeCure is exposed to the drug development risks typically associated with pure play biotechs, namely clinical development delays or failures. Additional sensitivities exist around IP protection, regulatory risks, competitor successes, partnering setbacks and financing and commercial risks. Its focus on developing drugs for AD could be considered higher risk, but also higher reward in our opinion. Having projects that could have potential beyond AD does mitigate this risk to an extent.

While AlzeCure currently has a sufficient cash runaway to complete the ongoing clinical trials for ACD856 and ACD440, additional funding will be required to progress these as well as facilitating the progression of preclinical candidates. As the company has guided, assets will likely need to be partnered at some point as later-stage CNS studies can be costly due to large populations and study duration, hence the ability to successfully negotiate a deal is a key sensitivity, as well as the partner’s willingness to progress these projects to commercialisation. We have assumed a deal in our valuation once meaningful Phase II proof-of-concept data are available, but we have limited visibility on the timing and terms.

Future drug pricing and market dynamics are invariably a key sensitivity, but can be hard to predict, especially if competitors are successful. Our current scenario for ACD856 assumes that a cognitive effect is seen that is superior to standard of care, which would warrant premium pricing paid to existing, generic AD drugs. If a disease-modifying effect will be obtained, market penetration and pricing should be higher. Conversely, if ACD856 has an efficacy that is only comparable to existing generic AD drugs, then it would invariably compete directly with these, although it could still be successful due to the vast market size.

We value AlzeCure at SEK729m or SEK19.3 per share, based on a risk-adjusted net present value (rNPV) analysis, which includes an SEK112m in cash (no debt) at the end of FY20. A breakdown of our rNPV valuation, which uses a discount rate of 12.5%, is shown in Exhibit 8. We note that for NeuroRestore, we use a rather conservative success probability of 5%. In most other indications, the success probability for Phase I stage asset would be around 10%. Due to the lack of successful attempts to develop novel AD drugs so far, the success probability for these projects is significantly lower. However, lower success probability is built in the market’s expectations, in our view. The attractive economics (large peak sales even if the drug provides only symptomatic relief) more than offset that, in our view.

Our valuation only includes the two clinical-stage assets, ACD856 for AD and ACD440 for neuropathic pain. We currently do not value any other potential indications or projects in the preclinical stages, but will reconsider them if AlzeCure will progress with them. We note that although the projects in the Alzstatin program are still in preclinical stage, they could move into the clinic within the next 24 months. This is a drug candidate for AD with potentially disease-modifying properties, so a major indication. Using the model, we developed for NeuroRestore and tweaking the assumption, we calculate that such a drug could reach c $15bn in peak sales (details below).

In our model we use the US and the top 14 wealthy European countries (top five, Benelux, Scandinavia, Switzerland and Austria). One of the key questions is which setting is the best for a symptomatic drug across the severity spectrum of AD. Minimal cognitive impairment (MCI) and the early-stage AD settings are a natural target for Alzstatin, as the disease-modifying approach in AD is increasingly moving towards earlier stages. Treating moderate to severe patients with disease modifying therapy can be too late. It is rational to assume that symptomatic treatment would likely be accepted in later stages.

But as we discussed, ACD856 could potentially show disease-modifying activity, so it could be used in different stages of AD. Even if it shows a symptomatic-only effect, it could still be useful in early-stage AD, and even in MCI stage, if the drug is safe. For all these reasons we use the mild AD patient population as the target population, as this setting will likely overlap in future development paths for ACD856 and Alzstatin. According to different sources:

This translates into the target populations we use in our model: 1.2 million in the US and 980 thousand in European countries, approximately 2.2 million mild AD patients in total treated.

Overall, we view current market size estimates and even existing treatment paradigms in different countries as not limiting the potential of novel effective AD drugs, as they would likely disrupt the status quo and expand the treated population.

We assume a price of US$6,000 per patient per year in the US (a premium to branded generics currently) and a 50% discount in Europe. We assume market penetration of 30% with peak sales of US$4.6bn achieved in seven years after launch in 2030. We assume that AlzeCure will out-license ACD856 during the Phase II stage in 2024 and that until then the company will invest c $17m in development. After the out-licensing we assume that the partner will take over the development (although various other arrangements, such as co-development, are possible). We assume Phase III start in 2026 and launch in 2030. Licensing deals that we view as comparable are listed in Exhibit 9. We use the average values: an upfront payment of US$55m and total milestone payments of US$550m spread across the R&D (one-third of the total amount) and commercial phases (two-thirds of the total amount). We assume AlzeCure will receive a tiered royalty rate of 14–18%.

Due to the highly fragmented pain relief market and the fact that ACD440 is still in the early stages of clinical development, we believe a bottom-up approach of estimating peak sales is not feasible at the moment. The topical pain relief market has been dominated by generic NSAIDs, primarily the brand Voltaren (diclofenac, Novartis), which had sales of $360m in 2020 (EvaluatePharma), although it has reached $1bn+ in the past. It is not a direct comparison as it is perceived as a branded generic available OTC and used for many forms of pain, such as trauma. The most comparable drug in terms of setting, namely, neuropathic pain, would be Lyrica (pregabalin, Pfizer), which is one of the standard of care drugs. Lyrica sales in neuropathic pain plateaued at c $3bn before patent expiry in 2019. However, it is an oral formulation, so not a direct benchmark drug as well, in our view. Keeping in mind this range, we use what could be considered a conservative assumption of $500m in peak sales assumed to be reached in seven years after the launch in 2028.

We assume that AlzeCure will out-license during the Phase II stage in 2023 and that until then the company will invest c $10m in development. We assume Phase III start in 2025 and launch in 2028. Licensing deals that we view as comparable are listed in Exhibit 9. We include an upfront payment of US$5m and total milestone payments of US$120m spread across the R&D (one-third of the total amount) and commercial phases (two-thirds of the total amount). We assume AlzeCure will receive a tiered royalty rate of 14–18%.

Alzstatin projects are still in the preclinical development, which is the main reason we do not include them in our valuation. However, Alzecure has chosen the drug candidate and currently is conducting in vivo toxicology studies, so it could move into the clinic sometime in 2022/2023. Given the large indication (potentially disease-modifying effect in AD), we have used our model to estimate potential peak sales for a disease-modifying AD drug.

We use the same mild AD patient population and most of other assumptions as for NeuroRestore. All timelines are delayed by two years. The two variables that are most likely to differ, in our view, are the penetration rate and the pricing. We have doubled the penetration rate for a successful disease modifying drug to 50% from 30% and also the price tag to $12k from $6k in the US (50% discount in Europe). Keeping other assumptions the same (timelines, R&D costs, licensing deal probability, etc), we calculate potential peak sales of a DMT could reach c $15bn. As a sense check, EvaluatePharma collects consensus forecasts with a time horizon until 2026. For aducanumab the current estimate is $4.2bn, however it is still in an exponential growth stage at that time. We believe it is widely expected that a truly disease modifying AD drug would reach $10bn+.

AlzeCure reports no income and in FY20 it reported SEK71.6m in operating loss, of which most (SEK62.4m, of which SEK495k is amortisation and depreciation) was booked as R&D expense. In FY19 the operating loss was SEK50.9m indicating that spending increased due to more R&D activities. We expect the spending to increase somewhat further in 2021 as both clinical trials are in full swing and forecast SEK82.0m in operating loss.

In our valuation model we assume out-licensing deals for ACD856 in 2024 and for ACD440 in 2023. The assumed deal terms are described above, however, we don not include the upfront payments in our financial forecasts, as these are still risk-adjusted.

AlzeCure had cash of SEK112m at the end of FY20, which according to our model should provide funding into 2022. We assume am illustrative long-term liability of SEK 57m in 2022 (as per our research principles in lieu of equity funding). Notably, its existing budget is sufficient to reach the readouts from both clinical trials, as well as other potential catalysts from the preclinical R&D pipeline: