Biotech portfolio update – Sage, Marinus and Immunogen

A new lucrative indication for Sage (and potentially Marinus)

Last week, Sage (SAGE) surprised the market by announcing preliminary but impressive results for SAGE-547 in patients with postpartum depression (PPD, also known as postnatal depression). Four patients with severe PPD experienced a dramatic improvement in their depression score from an average of 26.5 to 1.8. In other words, these patients entered the study with a severe debilitating depression and became symptom-free within 2-3 days. Despite the preliminary nature of the results, they generated a clear efficacy signal that merits evaluating SAGE-547 in a randomized trial, to be started later this year.Potential beyond epilepsy – Before the announcement, SAGE-547 was viewed as an anti-seizure drug but now investors are starting to realize the drug and related next-gen derivatives may treat a wider spectrum of neurological diseases, including depression. The decision to go after such a severe patient population resonates with Sage’s strategy in epilepsy. In both cases, the effect seen in a handful of patients without a control arm was so quick and profound that it enabled the company to validate the indications and de-risk the program.

Marinus gaining visibility – Sage’s announcement sent Marinus’ (MRNS) shares up almost 30%. This is the first time that investors link between the two companies, whose lead agents have a common mechanism of action (GABA-A activation) and are almost identical in terms of chemical structure (SAGE-547 is a formulation of allopregnanolone). In contrast to Sage’s strategy, ganaxolone has been tested in placebo-controlled trials of hundreds of patients with milder forms of epilepsy, demonstrating a relatively modest benefit. Another important difference between the two drugs is that SAGE-547 is given as an injection and ganaxolone is an oral drug.


Source: Marinus presentations

Valuation gap still significant – The notable gap in valuation ($2.4B vs. $181M) in favor of Sage reflects the strength of SAGE-547’s data package. Going after the severe end of the spectrum allowed Sage to get more credit and share appreciation based on very small studies. The complete lack of indication overlap makes it impossible to compare the two drugs and it is hard to speculate how ganaxolone would perform in Sage’s indications and vice versa. A preclinical poster presented by Sage suggests SAGE-547 is more potent in some assays, although both drugs demonstrated good efficacy. Sage also has a portfolio of next-generation molecules with better pharmacologic properties, including orally bioavailable drugs but these agents are still preclinical.

My working hypothesis (which I discussed two months ago) is that although SAGE-547 may be a better drug, there is a reasonable likelihood that the two drugs are comparable. SAGE-547 cannot be pursued as a classic anti-seizure drug because it is an injectable agent but an IV formulation of ganaxolone (expected to enter the clinic this year) can be tested in Sage’s high profile lucrative indications and potentially compete with SAGE-547.

Marinus is well positioned in depression – Last week’s announcement makes things even more interesting since PPD (as well as other forms of severe depression) may require prolonged treatment preferably without hospitalization, where ganaxolone’s oral bioavailability is a clear advantage. Realizing this, Sage intends to eventually pursue PPD with one of its oral next-generation compounds, which are expected to enter the clinic by year-end 2015 (phase I will likely be healthy volunteers). Marinus, on the other hand, can easily start a clinical program with ganaxolone which is in phase III trial for epilepsy and has a safety database comprising more than 1,000 patients. Ganaxolone’s well characterized safety profile (which enabled Marinus to pursue pediatric indications) cannot be over-estimated in indications like depression, where safety is often more important than efficacy.

For the first time, Marinus can leapfrog Sage and beat it to market in one of Sage’s indications. This may be the reason why for the first time, Marinus and Sage traded in tandem.

SAGE vs. MRNS - chart

It is important to note that Marinus is still a high risk company and will remain one until it can provide an efficacy signal in depression. Even if Marinus pursues PPD, results are likely 9-12 months away. By then, the company will have results from its pivotal trial in epilepsy.

IMGN – Breaking the ADC curse?

Immunogen (IMGN) was one of the clear winners of ASCO 2015 based on preliminary data for mirvetuximab soravtansine (IMGN853). The company reported a 53% (9/17) response rate (including one complete response) in patients with folate-receptor alpha (FRα)-positive platinum-resistant ovarian cancer. This may trigger the long awaited shift in sentiment around ADCs following two challenging years, as I discussed several months ago.

Impressive activity… – This level of single agent activity is rarely observed in solid tumors, let alone with antibody-based therapies. In fact, this might be one of the highest response rates observed with antibodies or ADCs in solid tumors. Roche’s Kadcyla, which utilizes another ADC technology from Immunogen, is the only ADC approved for a solid tumor. In its phase I in HER2+ breast cancer, Kadcyla demonstrated a 37.5% (6/16) response rate at clinically relevant doses.

…but program is still risky – Acknowledging the high response rate, mirvetuximab is still a high risk program with multiple open questions. To begin with, it remains to be seen if this high response rate will be replicated in larger studies. Typically, oncology drugs generate a lower response rate in large studies compared to that observed in small trials. In addition, using a more rigorous assessment that requires a second scan to confirm responses, only 5 of 9 responders qualify as confirmed responses. Of the remaining 4, 2 responses were short-lived and two are awaiting confirmation. Therefore the confirmed response rate is currently 29.5% (5/17) with a potential to increase to 41% (7/17).

Response duration critical for accelerated approval – Although responses in 6 of 9 patients were still ongoing, the data set is too immature to conclude anything about response duration. Durability of response is a crucial factor in getting an accelerated approval, which Immunogen intends to pursue using a single arm phase II. The non-official bar for accelerated approval is a 30% confirmed response rate and a 6-month durability of response. At ASCO, only one response was ongoing for 6 months while the rest of responses had a shorter duration due to progression or limited follow up.

IMGN853 - Swimmers plot

Source:  Moore K ,  ASCO 2015

Pivotal trial planned for H2:2015   Immunogen plans to start a pivotal single-arm phase II in  FRα+ ovarian cancer. Although specifics were not given, the study will probably enroll 100 patients with response rate as the primary endpoint. Given the excitement around mirvetuximab and the lack of promising experimental agents in ovarian cancer, enrollment should be very quick with potential data and accelerated approval in Q4 2016 and mid-2017, respectively.

In parallel, Immunogen plans to evaluate mirvetuximab in combination regimens for earlier stage patients.  The company is also evaluating mirvetuximab in additional FRα+ tumors including endometrial and lung cancer. So far data in these indications have been limited and did not include meaningful clinical activity.

Significant commercial opportunity – FRα+ ovarian cancer represents a ~$600M opportunity in the US assuming 14,200 annual deaths, 60% of cases with FRα+ and a $70k cost per patient. If brentuximab gets to earlier lines of treatment, the US potential increases to $1B. Immunogen believes that ~2500 patients will be initially eligible for the drug based on the indication pursued for accelerated approval, which represents a $175M opportunity.

Limited direct competition – One of the most compelling features of mirvetuximab is the lack of other ADCs against FRα in clinical development. Although other ADC companies will likely pursue FRα based on the recent data, Immunogen should have the market to itself for several years should mirvetuximab gets approval in 2017. Endocyte (ECYT) is developing small molecule drug conjugates (SMDC) targeting FR but to date single agent activity has been minimal.

There are other experimental agents in development for ovarian cancer that may compete with with mirvetuximab. These include Roche’s Anti-NaPi2b ADC (DNIB0600A), which generated a 50% response rate in NaPi2b+ ovarian cancer. Results from a randomized trial comparing DNIB0600A to Doxil are expected this year. Immunogen’s Mirvetuximab and Roche’s DNIB0600A have the potential to have a meaningful impact in ovarian cancer as both FRα and NaPi2b are expressed in the majority of cases.

Portfolio updates

We are selling our positions in Genmab (GEN.CO) and Morphosys (MOR.DE) for a profit of 972% and 404%, respectively. Even in light of daratumumab’s imminent approval and, Genmab’s valuation of >$5B more than factors in daratumumab’s 2B-$3B sales potential. The decision to sell Morphosys was based on the recent setbacks with Roche’s gantenerumab and MOR202, which was returned by Celgene.

We are adding new positions in Xenon (XENE) and Trevena (TRVN) which are developing promising pain products. Both companies are expected to report data from well-powered randomized trials for their respective programs in the coming months. We are also initiating a position in Ocata (OCAT) based on exciting data for its cell therapy in dry AMD.

Portfolio holdings – June 14, 2015

portfolio holdings - 14-6-2015 - after changes

Biotech ETFs - 14-6-2015

34 thoughts on “Biotech portfolio update – Sage, Marinus and Immunogen

  1. hi Ohad

    thanks for your update!

    do you know whether MRNS intends to evaluate ganax in PPD? so far they have strictly focused on epilepsy



  2. Ohad
    As usual – very balanced and informative write-up.
    It is noticable that you try to diversify away from cancer. From the last 10 buys only one is cancer. Is it bc there are no new exciting names or the valuation of cancer companies is too high?
    Any opinion about CEMP. They seems to have advanced antiviral platform.
    About OCAT – is it a concern almost zero institutional ownership? Also the mgmt has shaky reputation. The CEO e.g. was fired last year for misleading BOD and investors.


  3. Chris (AUPH) – I still view it as an attractive opportunity and plan to get in.

    Alex (ESPR) – PCSK9 antibodies have a stronger effect on LDL-C but ETC-1002 is oral and may have a broad metabolic impact.

    andre – You are right, I am trying to diversify the portfolio beyond oncology even though I am still very much interested in oncology companies. While oncology is still a dominant sector there have been tremendous advances in other fields that are gaining momentum (ophthalmology, rare diseases, and certain CNS indications like pain). In addition, most of oncology drug development (especially immuno-oncology) is being conducted by large biopharmas. Lastly, there is the valuation issue as there are several good companies (BPMC, AGIO etc.) that are simply too expensive imo.
    Re CEMP, I don’t know them well.
    OCAT caught my eye following their publications in good peer reviewed journals. I have to admit I not completely familiar with their history but that’s obviously a concern. If they can corroborate P1 findings in the ongoing studies all will be forgotten.

    Peter – I am still exploring this field, no concrete opinion…



  4. Hey Ohad
    thanks for this, I always read your blogs with great interest. I am already an investor in XENE and so I am happy you adding it.

    One of the cos I have been a long-time investor of that nobody on your blog seems to talk about or have interest in is HTBX. They have just released news about the broadening of their technology to deliver with a single off-the-shelf drug co-stimulatory immunotherapy that can go after multiple targets or antigens and generate a t-cell response. It seems to me like and elegant solution, especially considering how everything is moving toward combination therapy. But the technology is the early stages… not yet in the clinic. Their current phase 2 drugs are an earlier version of the technology that does not enable co-stimulation.



  5. Ohad, as you have added two “pain” new companies, i was wondering if you have had a look at Collegium Pharmaceutical (COLL)


  6. Hi Ohad,

    What do you make of AAVL’s P2 results today? Also, do you have any opinion of LPTN’s bioactive-lipid neutralization mabs or prospects in neuropathic pain?



  7. Dan (HTBX) – Don’t know them well but the new platform looks interesting but as you wrote still early. Many others are working on combining co-stimulatory agents with cancer vaccines so differentiation is an important issue.

    lgonber (COLL) – Personally, I am not a big fan of abuse-deterrent formulations of opioids. I prefer novel MOAs like XENE.

    Wildbiftek (AAVL) – The reduction in Lucentis rescue injections was only 50% (2 vs. 4), which is somewhat disappointing but the visual acuity improvement looks much better than anyone could have expected (11.5 letters). So overall I thought results are positive because patients get less injections and have a dramatic improvement in visual acuity. Can’t explain the after-hours drop, will listen to the webcast maybe they revealed something disturbing there…



  8. HI Ohad

    AAVL: guess the unexpected high improvement in VA is based on the really high “loss” in VA in the control arm, which is seen as unusual based on historic Lucentis treatments/studies as I understand.
    Main concern though guess is that the reduction is only 50% in rescue injections.

    At first glance the results are somewhat mixed based on the quite high expectations, so the drop seems logical… also not down 70-80% as this is clearly far from being a fail either.
    Interesting on a side not is this reoccuring 9.1% in the control arm, which was apparently 1 individual out of the 11 that improved this arm relatively seen…


  9. Looks disappointed as medium number of injections is 2 but CI is 1-6, which is covering not overlapping control arm CI 3-5. Also 2 vs 4 looks inferior to Phase 1 0.33 vs 3.


  10. Same thoughts on AAVL. Will be great if you can share any new insight you’ll have after listening to the webcast.

    What’s your take on the news regarding AERI?



  11. Just finished listening to the call. To begin with, there was a strong discrepancy between management’s comments and the tone (which was quite depressed). The results are very surprising to me as I had expected clear success or failure but they are mixed.

    The BCVA benefit is very strong numerically but the control arm had a sharper decrease than what would be expected with Lucentis alone. This can be explained by the a tough patient population or by a statistical fluke.

    50% reduction in rescue Lucentis shots was disappointing (I hoped to see a 75%-80%) but combined with the BCVA figures, there is clearly an efficacy signal.

    Retinal thickness change is negative and hard to explain, definitely a reason for concern.

    Bottom line – Although there are clearly red flags in the data set, I don’t plan to sell tomorrow. If the stock goes down 50% , market cap is $500M with $250M in cash. The technology could still have long term value to a potential acquirer (e.g. REGN, Roche).



  12. Hey Ohad
    seems that management could have done a better job at AAVL and and been a little more informed, especially considering the mixed results that have have everybody scratching their heads. I do not understand how the management of a $1B company fails to have solid answers regarding results.

    It looks like the shares might fall 50% how do you decide if you are going to add on the weakness? what would you like to see?



  13. Dan (AAVL) – Agree about management’s behavior as a contributor but the negative reaction to the data is based on real issues (although I don’t think they justify a 50% fall…).
    I don’t plan to add shares today unless the stock goes to cash levels. The next couple of months are going to be quiet in terms of news flow so I don’t see major near term catalysts.



  14. Thanks for the update on the portfolio.I also got into XENE and TRVN..loos for an entry point for AUPH and LOXO. Very disappointed with AAVL but again you win some and you lose some.

    I do have a question regarding Clovis. Atlhough AZN has numerical advantage over Clovis’s drug however since the patient population is different will FDA through any wrench in approval of AZN drug? I don’t think so but I just wanted to see your what opinion is.

    Another question is on AERI now with FDA approving end point changes from 27mm Hg to 25mm Hg. Does that changes your opinion? Is it a buy now?

    Thank you for posting your opinion!



  15. Amol (CLVS) – That’s a good question I don’t have a concrete answer to. In general, the FDA requires companies to have US patients in their pivotal trials but I don’t know if there is a defined %. I believe there are drugs that got approved based on data which had been predominantly generated ex-US (GSK’s pazopanib for example) so I don’t think this will limit them.

    Re: AERI – Yes that’s quite positive but I don’t plan to add more as this program became much riskier.



  16. Ohad,

    LOXO is on fire. Do you know what’s going on and still wait for a good entry? Do they have any catalyst in near future?



  17. Hi Ohad, regarding AAVL, they are around cash level trading at $15. Would you considerado is a good price to get in? Do you expectativas a bounce after the big fall? Thanks


  18. Cloud (LOXO) – I don’t know the reason for the move. Could be anti-tumor activity observed in one or two patients with Trk fusion. They already disclosed the enrollment of a patient in March so efficacy data should be available for that patient. I am still waiting for a good entry point (after missing a 90% increase…).

    Lgonber (AAVL) – Although I think AAVL became a good acquisition target for companies with interest in ophthalmology (there is still value in the platform beyond AVA-101) , it is hard to envision a big move in the near time frame given the hostility towards management. I plan to sell at the next portfolio update and get back in after the summer.

    Steve – Sorry, don’t know GHDX well. FMI is an exception to me as I view them as an integral part of cancer drug development and continuum of care.



  19. Hi Ohad,

    Always enjoy reading your blog. Your pain stock targets are interesting. I am on the fence regarding Cara versus Trevena. Cara’s kappa opoid drugs seems less potent but with less CNS effect versus Trevena mu opoid drugs which seem to be more potent but with more side effects (relative to kappa opoids but certainly less than plain old morphine) . I’m just wondering if you looked at Cara and decided Trevena class of drugs were better. Thoughts? Optionality of Cara and uremic pruritus could be nice as well.



  20. Mike (CARA/TRVN) – Thanks. I am far from being a pain expert but TRVN’s bunionectomy data set was so impressive especially vs. morphine (although side effects appear more common) that I decided to go with them. In acute pain, my assumption is that getting rid of pain has the highest priority,even at a price of nausea/vomiting.
    Agree about the uremic pruritus study.



  21. Ohad
    Do you see Xene’s upcoming phase 2 trial result as a toss up (50/50 chance of success)? The stock seems to be very undervalued despite its multiple partnership.


  22. Cloudy (XENE) – I find it hard to assign a probability there . The osteoarthritis study is a high risk one because the drug is applied topically and penetration is an issue. On the other hand the company saw good penetration with their formulation in animals so I am cautiously optimistic.

    I view XENE is a core long term hold based on the validated pain target plus a couple of other early stage programs.



  23. Ohad, what do you think about the Phase III results in AML for CPXX? The Market reaction seems rather muted. The stock still seems very cheap to me with a market cap still under $100 million. Does the lack of statistical significance in the data revealed portend problems in the OS data to be reported in 1Q16?


  24. Richard (CPXX) – I find what they did highly unusual and questionable as the primary endpoint is overall survival. Why not wait for the OS analysis?
    The numerical difference they show is encouraging but I would take it with a grain of salt as:
    a) the difference is not huge (statistics?)
    b) In AML there are no good surrogate endpoints
    c) History taught us that adding more chemo doesn’t help.



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 )

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