After a long stagnation, is CNS starting to crack?

After being the industry’s graveyard for over 20 years, there is finally room for optimism in CNS (central nervous system) disorders. The void created in the field is now being filled by small companies which are using novel therapeutic (gene therapy, antisense, antibodies) and development (genetic validation in humans, biomarkers for patient selection) approaches. While clinical results are early and sparse they may represent the beginning of a new innovation cycle in CNS.

Tackling old problems with new tools

A lot of the progress stems from using new treatment modalities in order to modulate targets previously regarded as underuggable.

Gene therapy is making headways in rare genetic diseases led by Avexis’ (AVXS) SMA program and Abeona’s (ABEO) Sanfilippo programs. For these programs, as biology is already sorted out (driver of the disease is known) the challenge lies in delivery to the brain and duration. So far results with AAV9-based programs look promising but sample sizes are small and follow up is limited. If the data hold up, and AAV9 vectors can lead to meaningful protein production in the brain/CNS, the implications cannot be overstated. CNS is a particularly attractive domain for AAV-based gene therapy because neurons don’t replicate so the transduced cells should not be diluted with time.

Antisense drugs are proving to be a useful tool to modulate targets in the CNS (via repeated intrathecal administration). Biogen/Ionis’s (IONS) Spinraza, the first FDA approved treatment for SMA1, works by enhancing production of an alternative protein to SMN1 which is mutated in the disease. Last month, Ionis and its partner Roche reported that another antisense drug (IONIS-HTTRx) demonstrated target engagement, this time in Huntington disease. Although it is too early to conclude anything about efficacy, the dose-dependent reductions of mutant Huntingtin in the CNS are a remarkable achievement.

Antibodies are a successful class of drugs but their use has been predominantly limited to oncology and inflammation. After years of setbacks that included the never-ending beta amyloid farce and safety concerns with anti-NGF antibodies for pain, looks like anti-CGRP antibodies will become the first approved class of antibodies for a CNS indication (migraine). Multiple data readouts from Alder (ALDR) Amgen (AMGN), Lilly (LLY) and Teva (TEVA) validate this class and are likely to result in FDA approvals in 2018.

Promising small molecule CNS programs

The lion share of CNS drug development is still based on small molecule drugs. This segment has benefited from genetic research in humans that can validate targets more reliably (preclinical CNS models are notoriously hard to translate to humans) and more efficient drug discovery tools. Nav1.7 is the poster child for CNS target validation based on humans genetics (lack of pain sensation in people with Nav1.7 mutations) although the target has proven challenging to date.

My top three CNS picks in this group are:  Sage (SAGE) in depression, Xenon (XENE) in epilepsy and Minerva (NERV) in Schizophrenia and insomnia.

Sage – Potential breakthrough in depression

Despite a complete failure in preventing epileptic seizures, Sage’s GABA agonists are emerging as a novel class of antidepressants, potentially the biggest advancement in depression in decades.

In October, the company announced positive P3 data for SAGE-547 in PPD (post-partum depression), which will probably lead to the first FDA approval in this indication (especially in severe PPD where the effect was more pronounced). It is hard to estimate the commercial opportunity in severe PPD as there is no benchmark to rely on, but epidemiological studies and literature suggest an annual incidence of 40k in the US (1% of ~4M births) which translates to a global market opportunity of ~$500M.

In November the company announced positive P2 results for its proprietary next-gen oral drug (SAGE-217) in MDD (Major Depressive Disorder), a huge indication that has been in stagnation since the 1990’s. Results were highly positive with a 6.9-point placebo-adjusted decrease in HAM-D (a commonly used depression scale), the primary endpoint, with only two weeks of treatment. At day 15, the drug led to a remission (no signs of depression) rate of 64% vs. 23% in the placebo arm. The most striking pieces of information were the very fast onset (statistically significant benefit already after 1 day) and the long lasting effect (separation from placebo at 4 weeks, two weeks after the last dose).

These data are from a relatively small study (89 patients) and need to be corroborated by larger trials, but the emerging profile of SAGE-217 is very compelling. Efficacy is at the (very) high end of the spectrum of what is commonly seen in MDD and most depression drugs take several weeks to reach this effect. Safety profile continues to look benign and highly differentiated from approved antidepressants. Such a profile will likely make SAGE-217 an overnight blockbuster, if approved.

The primary risk for SAGE-217 is its long term safety profile, which is still an open question given the limited treatment period in the P2. The company claims it already has animal tox data covering chronic use but in CNS indications, even a low rate of severe adverse events can lead to termination.

SAGE-217 may compete with J&J’s esketamine (P2 data recently published) and Allergan’s (AGN) rapastinel. It is hard to compare across small studies that used different endpoints and design, but sage’s molecules appear to have a clear advantage on safety and route of administration ( Esketamine is given intranasally whereas  rapastinel is given IV).

Xenon – Genetically validated targets for epilepsy

Sentiment around Xenon is understandably poor following the setbacks in its pain programs with Genentech (on hold following preclinical tox findings) and Teva (failed multiple P2s). In 2018, investor focus will shift to the company’s epilepsy programs (XEN1101 and XEN901), which I view as attractive and de-risked based on genetic validation in humans. Xenon intends to pursue XEN1101 and XEN901 in rare genetic pediatric epilepsies in parallel to the general adult population, a strategy that has never been tried and will require the FDA’s blessing early on.  Long-term safety is another major overhang, as is the case with most CNS programs.


XEN1101 was in-licensed from 1st order Pharmaceuticals in 2017. The drug works by opening Kv7 ion channels, an approach validated by human genetics (inactivating mutations in Kv7 are seen in a severe form of childhood epilepsy) and clinical data (ezogabine, a predecessor of XEN1101 had been approved for epilepsy in 2011 but was later withdrawn due to safety issues). Ezogabine was a highly efficacious antiepileptic but prolonged use led to pigmentation in the eye and skin that at the time raised concerns about vision loss and skin toxicity, which in turn triggered a black box warning. Although later the FDA softened its language and stated that pigmentation had no adverse effect on vision or other tissues, sales of the drug plummeted until it was withdrawn from the market in 2017.

XEN1101 was developed by Valeant (who co-developed ezogabine with GSK) and was designed to avoid pigmentation, which is believed to stem from dimerization of the drug and binding to melanin. In contrast to ezogabine, XEN1101 does not dimerize or bind melanin so it should not lead to pigmentation (needs to be validated in the clinic). XEN1101 is also more potent and selective and appears to be superior to ezogabine in animal models.

Xenon recently initiated P1 in healthy volunteers with XEN1101 and incorporated an imaging endpoint that could demonstrate biological activity already in P1. A P2 in adult patients is slated to begin in H2/2018 with the potential to reach clinical proof of concept in 2019. The company plans to pursue a subset of pediatric patients with Kv7.2 mutations in parallel to the adult development program. This is a novel approach that has never been tested (typically, drugs are first approved in adults before moving to pediatric indications) so it remains to be seen if and how the FDA will address this. From a clinical perspective, treating these patients early could be critical for preventing long term cognitive damage. The company estimates there are ~1500 patients with Kv7.2 mutations in the US.


XEN901 is Xenon’s internally developed Nav1.6 inhibitor. Similarly to the case with XEN1101, the target is validated by activating mutations in Nav1.6 observed in a rare form of childhood epilepsy, making it a “driver” mutation. The drug is expected to enter P1 in Q1/2018 and its development plan is similar to that of XEN1101 (start with P2 in adults and in parallel pursue the rare pediatric indication).

Minerva – Traditional CNS pipeline with convincing P2 data

Although Minerva’s lead programs look “old-fashioned”, reminiscent of the traditional Pharma CNS pipeline, its lead programs, MIN-101 for schizophrenia and MIN-202 for insomnia and depression, have encouraging data sets that point to clinical differentiation.

In its P2, MIN-101 demonstrated a significant effect on negative symptoms (apathy, poor social functioning) as well as cognitive symptoms with a mild safety profile. While the effect was not dramatic, it appears clinically meaningful (no other drugs have a proven effect on negative symptoms, some drugs are used off label) and dose dependent. A P3 for MIN-101 started last month with data expected in 2019.


MIN-202 is Minerva’s orexin-2 inhibitor, originally in-licensed from Janssen in 2014. The drug recently started two P2 studies in MDD and insomnia patients based on preliminary positive efficacy in a small P1b study in MDD patients who also suffer from insomnia. Orexin 2 antagonist could be a safer alternative to traditional sleep medicines as well as Merck’s recently approved dual orexin 1/2 inhibitor, Belsomra.

Limited composition of matter patent protection is a significant issue since Minerva is not going after orphan diseases. The company hopes its formulation patents to prevent (or at least delay) generic competition. The primary issue with owning the stock in 2018 is the lack of catalysts but valuation is attractive ($242M with $130M in cash expected at year end).

Two additional CNS stocks to watch: Denali and Ovid

Two other CNS stocks on my watchlist are Denali Therapeutics (DNLI) and Ovid Therapeutics (OVID). Both employ a more “precision medicine” like approach by pursuing genetically validated targets and or biomarker-defined patient populations.

Denali’s lead program is DNL201, a LRRK2 inhibitor for Parkinson’s disease that just completed P1. While there is a strong rationale for targeting LRRK2 in Parkinson (activating mutations are linked to Parkinson), DNL201 (while at Genentech) generated a safety signal in monkeys that prompted the FDA to issue a partial clinical hold. Last month, the FDA lifted the hold, which will allow Denali to test the therapeutically optimal doses of the drug in patients. P1 data demonstrate very good target engagement, which further de-risk the program. The company is expected to start P2 with DNL201 or a backup compound later in 2018 so efficacy data is still 18-24 months away. The company has a broad preclinical pipeline for other diseases including Alzheimer’s and ALS, all programs will utilize a biomarker for patient selection.

Ovid’s lead program is OV101, an isoform-selective GABA agonist that was originally developed by Merck and Lundbeck for insomnia (generated P3 data). Similarly to Sage’s molecules, OV101 can activate a discrete subset of GABA receptors (extrasynaptic, delta subtype) but according to Ovid, OV101’s unique feature is its ability to activate the receptor even in the absence of the natural ligand, GABA. The company is pursuing 2 genetic disorders (Angelman and Fragile-X syndrome) where GABA levels are decreases, but the drug does not directly target the underlying genetic defect in these syndromes. Another factor to bear in mind is the short patent protection which means the company will have to rely on market exclusivity for orphan diseases. Ovid’s second drug is a CH24H inhibitor for rare genetic types of epilepsy, currently in a small P2.

In both cases, I amwaiting for a cheaper entry point given valuations. ($1.54B for Denali, $234M for Ovid).

Portfolio updates

I am buying more Xenon and initiating a new position in Minerva. While neither company is expected to generate clinical proof of concept data in 2018, I still think valuations are compelling. I am also initiating new positions in Zymeworks (ZYME), Madrigal (MDGL) and Viking (VKTX).


Portfolio holdings – January 14, 2018

portfolio - 14-1-2018 - after changes

biotech etfs - 14-1-2018

151 thoughts on “After a long stagnation, is CNS starting to crack?

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s