On the morning of Tuesday, June 27, Alder Biopharmaceuticals (ALDR) announced the results of its Phase 3 PROMISE 1 trial. The trial evaluated the efficacy of Alder's anti-CGRP drug Eptinezumab in the prevention of frequent episodic migraine. The drug met the primary and secondary endpoints with very high statistical significance (primary: p = .0001 for 300mg and p = 0.0179 for 100mg). It was consistent with previous trial data, and its top-line results met or exceeded that of competitors. As I write this, investors are being rewarded with a greater than 35% drop in share price.
Importantly, on June 26, there was a block ~2,750 $15 ALDR put option trade at about 2:40 PM, representing an approximately $500,000 "investment". The entire open interest before that trade was 1,100 options. I will write more about the potential significance of this towards the end of this article.
In the minutes after the data came out and the days after, "rushed analysis" utilizing less-than-accurate information (starting on Twitter) made its way to StockTwits and multiple news mill websites, such as bizjournals, Xconomy, Benzinga, and MarketWatch. Many of these claimed they knew what investors "were thinking", without sources - or by linking an article on another site that had no sources.
These talking points have begun to verge on the hysterical
Most of these articles and Twitter posts aligned into one mantra against the company's results, arising from deeply flawed thinking: "Alder's placebo-adjusted response number was worse than that of competitors, and with the finance overhang as well as Alder's late entry into the market, the company won't be able to capture significant market share."
Unfortunately, this argument stems from a deep misunderstanding of how doctors and patients decide which therapies to use (and which thus determines which therapy captures the most market share).
Here are 10 points to consider.
1) You cannot use a placebo-adjusted response rate to make a cross-trial comparison, especially when demographics and trial entry criteria are nearly identical.
Using placebo-adjusted response rates makes sense when the route of administration is the same but demographics are different. In this case, the route of administration is different, and demographics are substantially the same.
Here are the demographics for the 2a trials for both Lilly (LLY) and Alder:
In these P2 trials and in the P3 chronic migraine trials, Alder, Amgen (AMGN), and Teva (TEVA) used migraine days, while Lilly used migraine headache days. It seems these are the same. All companies use the International Classification of Headache Disorders (2nd Edition for Alder, 3rd for Lilly) to define their patient population, and all four companies hired David W. Dodick as a trial investigator. Entry criteria in the Phase 3 frequent episodic trials on clinicaltrials.gov (see: Alder, Amgen  , Lilly  , Teva) are nearly identical.
(I would note that female patients on Teva's trial were asked to use two forms of contraceptives and remain abstinent, to reduce the possibility that pregnancy might skew the results.)
General patient demographics such as age and gender are similar in this 2a. Are they same in the Phase 3? Or, maybe there any regional differences?
As demographics (other than baseline migraine days) are not yet available for all of these trials, we can use trial locations as a proxy for overall demographics. Here are some metrics showing how US trial centers intersected between Lilly, Amgen, and Alder in Phase 3 frequent episodic migraine trials, derived from clinicaltrials.gov trial location data (Teva was not included in the analysis due to time constraints):
Number of US locations shared (by zip code):
|With Alder||With Amgen||With Lilly|
|Alder (Out of 82 Alder locations)||-||25/82||31/82|
|Amgen (Out of 99 Amgen locations)||25/99||-||41/99|
|Lilly (Out of 129 Lilly locations)||31/129||41/129||-|
Average distance, in miles, of 95% of US trial locations (by zip code) to nearest competitors' locations, weighed by population:
|Alder / Lilly||Alder / Amgen||Lilly / Amgen|
|0.2744 miles||0.5459 miles||0.0972 miles|
38% of Alder trial locations are in the same zip code as Lilly trial locations. 30.4% are in the same zip code as Amgen locations. This provides further evidence that demographics in each of these trials were largely the same - and thus placebo-adjusted rate comparisons are not necessary.
2) The placebo response is far more pronounced and variable when dealing with neurological health issues.
It has been shown via brain scans that there is a real effect when someone believes they are receiving a drug, even if it is actually a placebo. In one case this article covered, patients experienced reduced pain intensity via natural dopamine release when they were tricked into thinking they were receiving a pain-relieving drug, even as they consumed a placebo.
Perhaps most strikingly, when a patient believes they are not receiving the drug (but actually are!!!), their pain intensity reduction behaves in some cases as if they had taken no drug at all:
This meta-analysis touched on another part of the placebo effect, which is especially relevant in IV infusions:
...certainly the awareness of the treatment, the presence of the therapist [which occurs during IV, but not during subQ], and the expectation of the outcome are likely to play a crucial role. Because all these factors are strongly influenced by the doctor–patient interaction (e.g., the doctor’s words), the patient’s knowledge about a therapy appears to be fundamental to producing the optimal therapeutic effects.
In the PROMISE 1 trial, 25% of patients received a placebo. Thus, patients rightly believed they were likely receiving treatment. With p-values as low as 0.0001 in the trial, however, it seems that placebo response (dopamine, etc.) is no substitute for a real anti-CGRP medication.
And, in fact, when I go to a doctor's office, I do not ask the doctor to give me a placebo 25% of the time. I ask for the best drug. And, that drug is Alder's drug. It is faster-acting, more convenient, and more effective in treating migraines. The results show that in frequent episodic patients (and in chronic patients with 15+ days of migraines a month), Alder's Eptinezumab could cut the number of missed days of work due to severe migraine in half. What matters is the outcome for me, the patient, not the placebo-adjusted response.
3) The IV placebo response is historically stronger than subQ's placebo response.
So, why again is Alder's IV placebo response so much higher than that of competitors, all running subcutaneous injection trials? It is not demographics, and it is (obviously) not the drug. It's actually just the fact that it is a 30-minute IV infusion (per Alder IR). The patient is getting a drug in a clean, sterile environment, with a doctor and/or nurse present. The memory of this 30-minute experience in a different location is much more salient than a quick self-injection in one’s own home. In cases of a self-injection, the patient may wonder whether the "drug" is really getting into their system.
It’s not just an idea: in Botox migraine treatment studies, there was a very high placebo response rate (75%). And, there is scientific evidence that IV placebo responses are higher than subQ placebo responses. For example, this meta analysis of TNF inhibitor trials shows that the placebo response is better in IV than in subQ by about 36.9% (.319/.233 - 1):
|Route||Prevalence||95% conf. interval||Q-HET||Df-het||p-value-het||i^2|
Subgroup analyses for the method of injection showed 31.9% vs. 23.3% for the IV route non-specific effect response vs. the SubQ route.
Here are the drug and placebo responses in all four Phase 3 frequent episodic migraine trials, for weeks 1-12, using best dose. It is important to note that we only currently have 3-month data for Alder's trial (and other metrics got better with time), although that is likely to change soon.
|Timeline||12 weeks||Once a month for 6 months||Once a month for 6 months||Once a month for 3 months, or one dose|
|Dose size||300mg, 100mg, 30mg||140mg, 70mg||240mg, 120mg||225mg (once a month) or 675mg (once)|
|baseline||8.6||8.0 / 8.3||9.1 (average)||9.1|
|drug, best response || |
-2.9 / -3.7
(-36% / -44%)
-4.7 / -4.3
(-52% / -47%)
[placebo baseline was 8.4)
-1.8 / -1.8
(-23% / -22%)
-2.8 / -2.3
(-31% / -25%)
We can see here that Alder's placebo response was 77% higher than Amgen's, and 25.5% higher than Lilly's, both of which used subQ (baseline-adjusted basis: 64% & 32.1%). We can thus see that, per the TNF inhibitor meta-analysis discussed previously, not only is IV placebo naturally "better" than subQ placebo but also placebo responses are extremely variable even among subQ trials - and even between two essentially identical trials from the same company.
Not only that, we can also see that Amgen's first frequent episodic trial reported results significantly worse than the second one, despite a trial design that seemed substantially similar. All of this points to the fact that frequent episodic trials, even with 800+ people, are prone to substantial amounts of variability.
Average reduction in migraine days is thus looking more and more like a data point that is not a good gauge of true effectiveness.
But in any case, because a large fraction of migraine sufferers might not have the migraine due to CGRPs, this number is not commercially relevant. It is only an endpoint that the FDA is interested in when approving this kind of drug.
4) Investors should examine only commercially relevant endpoints to gauge potential market size and sales potential.
A better methodology, as what Alder, (and now Lilly, Amgen) is doing, is to see how many patients have reduced their migraines by 50%+, 75%+, and 100%. This weeds out those non-responders whose migraines are not due to CGRP sensitivity.
Here are the numbers we have so far - from Alder, Amgen, and Lilly. We do see a lot of missing numbers here - Alder just received the top-line data for their frequent episodic Phase 3, while Amgen has only recently begun to describe data in this way.
|Dosing schedule|| |
Once every 3 months, IV
|Two self-injections per month||One self-injection per month|
|Dose size - freq. episodic (mg)||300/100/placebo PROMISE 1|| |
140/70/placebo STRIVE and 70/placebo ARISE
240/120/placebo EVOLVE-1 & 240/120/placebo EVOLVE-2
|Dose size - chronic (mg)||300/100/30/10/placebo||-|| |
Reduction from baseline
|Frequent episodic 100%||14.9% / 8.6% / 5.9%||-||-|
|Frequent episodic 75%||31.5% / 30.8% / 20.3%||-||-|
|Frequent episodic 50%||-|
|Chronic 75%||-||8.8% (240mg) / 4.5% (placebo)|
|Chronic 50%||-||27.5% / 27.6% / 15.4%|
|Frequent episodic 75%||29.7% / 22.2% / 16.2%||-||-|
|Frequent episodic 50%||53.6% / 49.8% / 37.4%||-||-|
|Frequent episodic 100%||40.1% / 33.5% / 24.8%||-||-|
|Frequent episodic 100%||-||-|
|Frequent episodic 75%||-||-|
|Frequent episodic 50%||-||-|
|Any given month:|
|Frequent episodic 100%||20.6% (average)||-||-|
|Chronic 100%||27% - 41%|
5) Alder's chronic migraine Phase 2b shows greatly improved efficacy over the competition.
We do not yet have full Alder frequent episodic migraine results, but let's look at chronic data - that is, 15 or more migraine days per month: these patients are severely disabled. Here, Alder's data is superior to that of Lilly's. For example, in the chronic migraine patient population, the number of Alder patients with a 50%+ reduction in migraines was double that of Lilly, and for 75%+, this number was more than triple.
Although the comparison is not perfect, we do have Teva's average reduction numbers, and Amgen's as well. Teva reported -4.3 in chronic migraine between weeks 1-12 - and that was in headache days. Amgen's was -6.6, in migraine days. This is compared to Alder's ~-8.5. Alder's average chronic migraine reduction in its well-powered P2b (600 patients) was double that of Teva's and 30% higher compared to Amgen.
Can there be any doubt that Alder heavily beats Teva, Amgen, and Lilly in chronic migraine relief? And, the market, per my back-of-the-napkin calculations, is pricing in just about $80m in sales by 2021. Insane!
There are 3 million patients in the chronic migraine category, per Alder's research:
They will all prefer Alder's drug. To be completely honest, I don't think that the price will be $26.70 per yearly treatment. Current migraine IV cocktail infusions cost $2,000 according to research by Damian Garde. Botox therapy costs about $450 and lasts 12 weeks. If even 1/6th of the target 5 million patient population spends $8,000 a year to prevent their debilitating migraines (consider the lost productivity of chronic migraine patients), this is a $6.67 billion market opportunity. Some sell-side valuations price in $1B in sales for Alder by 2025 (using higher per-patient costs), at a 15% market share.
With its superior chronic data, and zero competitors in the IV segment, I believe Alder will capture 100% of that segment, and I believe that segment will be at least 30% of the overall market. Thus, per my calculations, with yearly sales of $2B by January 2025, Alder is worth approximately $99 a share:
2B * 15 [P/E] * .85 ^ 7.5 [discount rate for 7.5 years] * .75 [accounting for COGS, marketing, SG&A] * .75 [dilution discount] = $4.99B, or $99 a share.
But that's not all.
6) Compared to the competition, Alder's drug is much longer-lasting, has a fast response time, and convenient quarterly administration - with a subQ product planned for those who would like to self-administer (see this, 1:05:30).
Alder's drug, efficacy of which has been shown to last for at least six months after one dose, has an incredibly fast response time which beats competitors by miles. Alder's most recent chronic migraine trial shows a similar reduction in migraine the day after treatment. Over half of those experiencing migraines the day of treatment no longer have them the day after, compared to only 20% of placebo patients.
Competitors take several months to reach peak efficacy:
The reasoning behind Alder's efficacy is not magic. It has to do with the design and particular characteristics of the drug. For example, its half life is much longer than that of competitors - 1 month - and thus it stays in the body longer, preventing the drug efficacy troughs of shorter half-life drugs, and thus leading to its superior efficacy in treating chronic migraines. The IV administration route, of course, increases the bio-availability of any given drug, and Alder's Eptinezumab is no exception.
7) The hyper-responders. As shown in Alder's PR on 6/27, not only can we group responders into 75%+ and 50%+ migraine reduction categories but we can also look at "hyper-responders". In Alder's PROMISE 1 frequent episodic trial, 20.6% of patients had no migraines at any given month. Data is still being accumulated on the number of patients who had no migraines in the trial, but as stated previously, Alder significantly beat Lilly in 0 migraines for chronic (8% at 300mg). Lilly did not even report this data after showing 8% of patients having 75%+ reduction versus Alder's 33%.
We do not know yet if Alder's hyper-responders are the same as Lilly's - due to genetics and slightly different molecular structures, we could be looking at several different groups of patients best treated by different anti-CGRP drugs. Per the conference call on 6/27, a biomarker is on Alder's wish list and would really help differentiate and better define the market share of each drug.
8) Alder showed better results with a second dose. Alder's anti-CGRP seems to be effectively curing migraines over time. The separation from placebo increases at week 13-24 compared to week 1-12. This is because Alder is resetting years of receptor sensitivity to CGRPs that migraine sufferers have built up. Migraines get worse over time because of this reason:
The migraines started in childhood and have gotten worse as she’s grown older. Since 2008, they have incapacitated her at least 15 days per month, year-round.
On May 5, UBS (Ami Fadia, Marc Goodman, Uy Ear) stated, "[anti-] CGRPs may not be much more efficacious vs Botox, but they are more convenient (1 monthly sub-q vs. ~30 quarterly injections)." Clearly, the evidence is there - no improvement on weeks 13-24, right...? Botox does not effectively cure migraines - it only treats them. The company that scammed the American taxpayer (and Genocea investors) is advising us on biotechnology!
9) IV administration has both patient and commercial ramp-up advantages. Many patients would prefer IV rather than once or twice-monthly subQ self-administered injections. Already, emergency IV clinics exist to take care of chronic migraine sufferers [source 1, source 2, source 3]. These clinics give a mix of not very effective drugs (and with side-effects) to patients. Alder's fast-acting anti-CGRP drug can be part of this mix. Thus, the claim that IV is inferior to subQ isn't true: there's already a user base in place, which means IV commercialization will be quick and cheap even if Alder gets to market after some of its competition.
Alder's IV will use the same "buy and bill" mechanism to receive reimbursement that Botox already enjoys (as described here), which should make ramp-up faster than that of competitors who will require individual prescription.
Practices may purchase BOTOX® from Allergan and then bill the health plan when the vial is used. With this method, the practice keeps inventory and carries the financial responsibility for BOTOX®.
What's the current treatment like?
Every 10 weeks, Sundquist gets 32 bee sting-like injections of the nerve-numbing botulism toxin into her face and neck. She also visits a neurologist in Philadelphia, Pennsylvania, who gives her a continuous intravenous infusion of the anesthetic lidocaine over 7 days. The lidocaine makes Sundquist hallucinate, but it can reduce her attacks, she says - she recently counted 20 migraine days per month instead of 30. Sundquist can also sometimes ward off an attack with triptans, the only drugs specifically designed to interrupt migraines after they start.
Alder's drug doesn't inflict hallucinations and is administered once every few weeks in a safe, clean, and calming doctor's office (hence that dopamine-releasing placebo effect!). Alder's Eptinezumab showed a clean safety profile comparable to placebo.
Doctors' Economic Incentive To Prescribe IV
And, that base network of doctors and clinics is economically inclined to prescribe Alder's drug. As mentioned earlier, Damian Garde of Stat News (article here/twitter post here) found that migraine cocktail infusions cost $2,000, and doctors get a cut. This is a very important aspect of successful commercialization, as GSK (GSK) found out in 2013 with its Bexxar drug. Doctors were not financially motivated to prescribe the non-Hodgkin’s lymphoma treatment and completely overlooked its efficacy profile, leading to poor sales and a withdrawal from the market.
10) Alder's production method gives it a significant competitive advantage. Compared to all its competitors, Alder's drug will be very cost effective to produce due to its MabXpress antibody technology. With healthcare costs constantly in the minds of the public and three competitors, this will be a massive positive in getting approved insurance coverage at a lower price point, even while yielding higher margins.
That 2,700 Put Option Trade...
I did not forget that 2,700 put option trade. Someone bought $500,000 of put options the day before results, when there was previously open interest of only 1,100. It's not a hedge, and it is not a coincidence, and it is not a gamble. What are the possibilities?
Likely, this was a leak in the data that was illegally traded on, and the person surmised what the trading action would look like. (call the SEC).
An interesting alternative - and I have seen similar kinds of things happen to multiple companies in the market in the past (see page 12, section 6, paragraph 2 here, and see the price action following this event) - is an unscrupulous hedge fund that was invited to a funding round for Alder: last year, several days after good Phase 2b results, Alder executed a successful financing round at relatively high prices. They were likely hoping to do the same this year, and perhaps one of those invited to the new share offering decided to short the shares. Thus, an entity short a large number of shares can instantly cover at a low price without driving up market prices and reap huge profits. In my view, a significant part of the trading action can be explained by this possibility.
Regardless, the number of hysterical (and reckless) articles that came of late regarding this stock is breathtaking. It is a sign of our times when an article on Benzinga uses another article on MarketWatch as a source for what investors "noted", while the MarketWatch article likely used popular Twitter posters who always get on the "Top" tab as a source. But how do they get on the "Top" tab? By having many followers and posting a lot -perhaps not necessarily having the time to research whether their conclusions are actually accurate. What they have learned is that quantity over quality gets eyeballs. And then, unfortunately, high-frequency traders that rely on Twitter and articles bring a stock down to its knees, creating a chain reaction. Although things may be just this simple in terms of the short-term stock price, it is rarely an accurate portrayal of events, especially in biotech.
Alder's mean reduction results were in line with previous results, in line with sell-side analyst expectations (except for the UBS and Credit Suisse fraudsters), and in line with competitors' data.
Yes, the company will dilute (or take out a loan). Because that is how biotech companies fund themselves. But in this author's opinion, not after they launch a vigorous defense of their stock price - perhaps like Aerie Pharmaceuticals (AERI) did early last year in the wake of a short attack against it - it is now trading 4 times higher. The company has shown to care about investors, only doing dilutive financing at relative peaks. Paraphrasing from the latest conference call:
We will be opportunistic in doing an offering as circumstances permit.
The financing overhang "worry" is not a new one. It's not new before the 27th, and it's not new in the market: CLVS, AERI, PTLA, SRPT: these companies' valuations all fell to absurdly low levels following one (or more) "trade-able" event(S), and the financing overhang was cited on all four of these as a reason for not investing, despite major commercial potential and a competent management team. And then, after all retail investors capitulated, these companies' valuations recovered to where they should be.
The ferocity of the downward move on clean, good data is fairly unique but presents an attractive opportunity to sophisticated investors.
As long as the stock price is down, the false narrative about placebo adjusted response stays stuck, and what is just a gut reaction soon becomes a sort of reality. But as the price recovers little by little, the false narrative starts to break down, and soon, in a different but similar feedback effect, the price is back to where it should be.
Oh, and one more tidbit: institutional investors are, at present, heavily invested in Alder Biopharmaceuticals: According to Nasdaq, institutional ownership is 104.90%. The stocks stated above had these elevated levels of ownership when they were in the toilet as well, although currently we see somewhat lowered numbers: 92.11% for AERI, 89.78% for CLVS, 87.08% for PTLA, and 73.95% for SRPT.
The latest results show that Alder's drug, Eptinezumab, can compete well in the lower-intensity "frequent episodic" segment of the market, with a mean percent decrease from baseline that is higher than any of its competitors.
Eptinezumab has shown in its 588-patient 2b trial to have best-in-class treatment for the "chronic migraine" segment of the market, with significant efficacy differences over all competitors. It is the only drug with a very rapid migraine relief. It will be the cheapest drug to manufacture and will capture the IV sub-segment, and compete vigorously in the subQ sub-segment.
As a holder of Alder Biopharmaceuticals, I am not worried about any of the things that make a biotech stock worth holding: its science, its excellent management and employees, and its commercial prospects.
Disclosure: I am/we are long ALDR.
I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.