Amarin Corporation (NASDAQ:AMRN) is a small, late-stage, pharmaceutical company developing a treatment for cardiovascular disease resulting from hypertriglyceridemia (high fat levels in the blood). Amarin's only drug candidate, AMR101, is a prescription grade omega-3 fatty acid capsule consisting of ≥96% eicosapentaenoic acid ethyl ester (EPA) that has successfully completed two phase 3 trials and is now in the midst of a third.
AMR101: the best in-class drug for lowering triglycerides
AMR101 has completed phase 3 trials for two indications. The MARINE trial targeted very high triglyceride (NYSE:TG) levels ranging from 500-2000 mg/dL while the ANCHOR trial targeted high triglyceride levels (200-500 mg/dL) for patients already on statins, a class of cholesterol lowering drugs that includes Pfizer's blockbuster Lipitor. Amarin estimates a population base of 4 and 36 million for the MARINE and ANCHOR indications respectively. In Dec '11, Amarin initiated another phase 3 trial, REDUCE-IT, to determine whether AMR101 decreases major adverse cardiovascular events in high-risk patients currently on statins with triglyceride levels 150-200 mg/dL. The REDUCE-IT indication represents 34 million people not addressed in the ANCHOR and MARINE indications. All three AMARIN trials are FDA special protocol approved, indicating that their trial designs are acceptable for drug approval if endpoints are met. However, REDUCE-IT will likely take at least 6 years to complete. The FDA has indicated that for a new drug application (NDA) submission for the ANCHOR indication, Amarin must first get approval for MARINE, which is likely, and have the REDUCE-IT study substantially underway. Joseph Schwartz from the investment bank Leerink Swann estimates annual sales of ARM101 to top $3 billion by 2024 were Amarin's trials successful (Nat. Med. 18(1), 6, 2012).
Importantly, the JELIS trial in Japan has already shown positive results for an EPA-based drug in cardiovascular disease. Similar to the REDUCE-IT trial, the JELIS trial compared EPA plus statin-treated patients to statin-only patients, with a major coronary event as the primary endpoint. The trial showed that the EPA plus statin-treated group had 19% lower rates of major coronary events than the statin-only group (Lancet 369, 1090-98, 2007). Sub-group analysis of the JELIS study indicated that EPA treatment suppressed coronary artery disease risk by 53% for patients with triglycerides ≥ 150md/dL and HDL-C < 40 mg/dL (Athersclerosis 200, 135-140, 2008). The JELIS study results and the fact that EPA is currently marketed in Japan as Epadel by Mochida Pharmaceuticals with annual sales of $400M in 2009 and 2010 are encouraging for AMR101. Indeed the JELIS trial is one of the reasons why the FDA did not require phase 2 trials for AMR101.
GlaxoSmithKline's competing drug, Lovaza, with US sales of $1B per annum is currently marketed for very high triglycerides. However, AMR101 is superior to its competitor in that it does not increase low-density lipoprotein cholesterol (LDL-C) or "bad cholesterol" like Lovaza. In fact, to greatly expand its target market, GlaxoSmithKline sought approval for conditions with triglyceride levels as low as 200mg/dL, but was rejected by the FDA twice because it increases LDL-C. Unlike Lovaza, which is composed of 36% docosahexanoic acid (DHA) and 46% EPA, AMR101 is almost pure EPA (>96%). There exists a link between DHA and increased LDL-C, which might explain the LDL-C effect of Lovaza (J. Nutr. 142, 99-104, 2012). In contrast, the MARINE trial showed that AMR101 lowered placebo-corrected triglyceride levels by 33% and 20% for the 4g and 2g per day doses respectively, without significantly increasing LDL-C. In ANCHOR, AMR101 lowered triglycerides by 21% and 10%, at 4g and 2g per day respectively, while LDL-C decreased 6.2% at the 4g dose. In both the ANCHOR and MARINE trials, AMR101 met its secondary endpoints of lowering important lipid markers, including but not limited to, apolipoprotein B (Apo B), lipoprotein-associated phospholipase-A2 (Lp-PLA2), and high-sensitivity C-reactive protein (hsCRP). Apo B is considered a biomarker for residual cardiovascular risk. High levels of Lp-PLA2 enzyme have been implicated in atherosclerosis, while hsCRP is an inflammatory biomarker that is predictive of cardiovascular events. These additional benefits of AMR101 are an additional strength. In both the MARINE and ANCHOR trials AMR101 had a safety profile similar to placebo with no treatment-related serious adverse events.
Omthera's competing omega-3 drug, EpanoVa, targets the 500-2000mg/dL patient population. In Apr '12, Omthera released data from its phase 3 EVOLVE study for EpanoVa. EpanoVa seemed as efficacious as AMR101 given that daily 4g and 2g doses decreased triglyceride levels by 31% and 26%, respectively. However, similarly to GSK's Lovaza, EpanoVa is composed of 55% EPA and 20% DHA (docosahexanoic acid), suggesting that EpanoVa could also raise LDL-C. Omthera did not release data regarding changes in LDL-C. Omthera's positive efficacy and omission of the LDL-C data suggested that it was in the same class as GlaxoSmithKline's Lovaza, thus sparking a bullish run on Amarin's share price in late April.
Generic versions of Lovaza are also potential competitors of AMR101, but will still suffer from the problem of increasing LDL-C and thus will not have access to the 200-500 mg/dL market of AMR101. On May 29th, 2012, a Delaware court upheld two patents for Lovaza. This ruling was against Par Pharmaceutical Cos. and Teva Pharmaceutical Industries, Ltd., which had filed abbreviated new drug applications (ANDAs) in 2009 to bring generic Lovaza to the market to treat very high triglyceride levels. The ruling means that Par Pharmaceutical and Teva are unable to sell generic Lovaza. Apotex Inc. was also named in this suit, but settled in March 2011, receiving a license to sell generic Lovaza starting in the first quarter of 2015.
Another omega-3 based competitor, Trygg Pharma's AKR-963 is currently pursuing a phase 3 trial, but has not released any results. Amarin's other long-term competitors include Neptune (NASDAQ:NEPT), Resolvyx Pharmaceuticals, and Catabasis Pharmaceuticals, which are all developing omega-3 treatments for high triglyceride patients. NEPT's CaPRE, a krill-derived mixture of phospholipids, EPA, and the problematic DHA, only just initiated phase 2 trials (NCT01516151 & NCT01455844). Neither Resolvyx Pharmaceuticals nor Catabasis Pharmaceuticals have entered phase 2 trials. Thus, AMR101 represents the best-in-class omega-3 drug amongst its competitors, and will likely hit the market well before its lagging competitors.
Concerns over Amarin's patent protection for AMR101 have been significant drivers of share price volatility. In the past year, Amarin's stock price has repeatedly dropped following initial rejections of a number of patents related to AMR101 market exclusivity. However, recent developments have shown AMRN to be on increasingly firm footing with regards to patent protection. The main patents AMRN has been working on are the '889, '620, '598, and '885. The '889 is a methods patent for treating very high triglycerides, and the '620, '598, and '885 are all iterations of Amarin's composition patents.
The '889 (full number 12/702889) the key patent followed most closely by investors. This patent, if awarded, would protect AMR101 until 2030 and covers the use of AMR101 for treating patients with high triglyceride levels (over 500mg/dL) to lower triglycerides and Apo-B. The application was first completed at the end of 2010, with the first non-final rejection coming in June 2011. Remember that multiple non-final rejections before getting a patent approved is normal; there is a lot of back-and-forth between the company and the USPTO before things are worked out to the USPTO's satisfaction. A response to the non-final action was filed in late June, and a final rejection came in August 2011. As is common after a "final" rejection, Amarin filed a "Request for Continued Examination" (RCE) in September 2011. The RCE allows the applicant to ask for continued examination of a patent application for a fee. The RCE was granted and another non-final rejection was given in October 2011. Amarin responded in January 2012 and received another non-final rejection in late February 2012.
To understand this back-and-forth over the '889 patent, it's important to look back at the USPTO examiner's rationale for each decision. The initial patent rejection was on the grounds that the '889 patent was using concentrated EPA to lower triglycerides and this technique had already been used in Katayama et. al (Prog. Med. (21), 457-467, 2001), and therefore was not novel. The key here was that the previous work had reported a decrease in non-HDL-C, which the examiner took to mean that a decrease in LDL would be predictable/obvious. As this does not necessarily follow, Amarin gave additional information in its response prompting the examiner to look further.
In the next rejection, the examiner argued that whatever discoveries the '889 patent is based on still come from previous art because of the use of high concentration EPA to reduce triglycerides. However, Amarin argued that they made a completely unexpected discovery - the LDL neutrality or LDL reduction under certain conditions - which differentiates them from the prior art the examiner was citing.
Earlier this year, Amarin responded to the late 2011 rejection with the claim that the MARINE trial/population is different from the Katayama population that the examiner referenced from "prior art," and so is a novel population for the application of AMR101. In late April, Amarin initiated a meeting with the supervisor of the current examiner at the USPTO. Also remember, the same examiner has been working on this particular patent through all of the previous non-final rejections. Speaking directly with the supervisor may be a positive sign, as it will bring more focus to the patent application and deal with notions that the examiner is biased against Amarin.
Most recently, on May 16th, Amarin submitted its response to the non-final rejection of late February. This submission was addressed to the supervisor they met with in late April, which could mean that he has taken over the case from the previous examiner. Amarin has narrowed the '889 claims somewhat, with the triglyceride lowering being taken out of claim 1. Claim 1 now refers to lowering ApoB in high triglyceride patients, treating them with the goal of reducing Apo-B without increasing LDL-C (Apo-B and LDL-C are now directly linked). Claims 7 and 9 now just focus on reducing triglycerides. The previous rejections were all based on the obviousness/lack of novelty of the patent's claims; with the changes in patent claims, especially the linking of Apo-B and LDL-C, it will be much harder for the examiner to make that argument. He cannot target lowering triglycerides as being non-novel/based on prior art. Amarin's response seems to refer to that examiner; page 22 mentions that the "Applicant submits that the Examiner has impermissibly substituted his own opinion for that of at least one expert in field," in response to the claim that Apo-B lowering would be expected.
The other key patents Amarin is currently pursing are the composition patents (12/951,620; 12/052,598; 12/769,885), the most recent being the '620. In late March 2012, AMRN was awarded the '598 patent for "Highly Purified Ethyl EPA and Other EPA Derivatives." This patent is also known as the "EPA with no DHA in a capsule application." This key patent grants Amarin significant protection from competition by generic fish-oil formulations until 2022. The '885 patent is for a composition of at least 95% in a capsule, which is less restrictive than the '598. It is not as important as the '889 patent, but will add some strength to AMR101's intellectual property portfolio. The '620 is a little broader, for a composition comprising EPA with less than 5% any other fatty acid that competes with EPA for binding to a biological site of action. This is more general, not specifically excluding DHA but instead referring to any other competing fatty acid. This patent could be useful, if awarded, for preventing production of generics that incorporate non-DHA fatty acids.
Other patents discussed by the investor community include the 12/815569, 13/272520, 12/888994 patents. The '569 patent is a composition and methods patent for lowering triglycerides with ultra-pure EPA without increasing, but potentially decreasing, LDL-C levels in patients on statins. The patent targets the anchor population triglyceride levels, 200-500 mg/dl. Jefferies has said there was overreaction in Amarin stock after the USPTO posted a non-final action in early April '12, as it is still early in the review process. Patent uncertainty has been a source of volatility in Amarin share price. The restriction requirement issued for '569 is not concerning and was probably anticipated by Amarin; combining all the claims into one patent was probably a strategic decision used, for example, to expedite filing and establish earlier priority dates. The '520 patent targets a similar patient population as '569 (statin plus anchor triglyceride range) but with broader claims to reduce triglycerides, non-LDL, Lp-PLA2, VLDL, Apo-B, total cholesterol and without increasing LDL-C with 96% pure ethyl-EPA. However, the '520 patent received a "final rejection" in March which could be appealed. Patent '994 is for a composition of an omega-3 fatty acid and a hydroxy-derivative of a statin and its method of use to treat cardiovascular disease including, but not limited to, hypertriglyceridemia. The claims cover >95% ethyl EPA and also ethyl EPA mixed with ethyl DHA. The USPTO has issued a requirement for election between product and method and requires a specific statin to be selected.
In addition, there has also been an update office action from the USPTO on patent '977, another strong composition of matter patent directly related to treatment of heart conditions. Lastly, Amarin is seeking new chemical entity status for AMR101 in order to get five years of market exclusivity. However, whether this will be awarded remains controversial as Lovaza also contains EPA.
Amarin has enough cash to operate through 2012
As of Dec '11 Amarin had $116.6M in cash and cash equivalents. This, with proceeds from the Jan '12 debt offering of $144.3M in exchangeable 3.5% senior notes, should be sufficient to fund operations for an additional 12 months including the preparation for the commercial launch of AMR101 and advancing the REDUCE-IT trial. Operational expenses were $44M in 2011 stemming from $21M in R&D and $22M in general and administrative costs. In 2010, operational expenses totaled $45M from $28M for R&D and $17M general and administrative costs, respectively. The decrease in R&D expenditure in 2011 versus 2010 was due primarily to completion of the ANCHOR and MARINE trials. Amarin forecasts an increase in R&D expenditure in 2012 driven in part by increased purchase of EPA from the active pharmaceutical ingredient supplier, Nisshin Pharma, and increases in expenses due to REDUCE-IT. Contractual obligations for 2012 include a minimal EPA purchase from Nisshin Pharma of $13.3M and $0.6M in operating lease. The 2012 expenditure for REDUCE-IT through contract research organization is estimated at $25M. Additionally, general and administrative costs are projected to increase in preparation for the commercialization of AMR101. Consistent with this, first quarter operational loss totaled $19M stemming from $14M in marketing, general and administrative costs, and $5M in R&D. The final cost to launch AMR101 will depend on the commercialization strategy; additional funding would be required should Amarin pursue self-commercialization.
Insider sales - confidence in company?
In the last few months, there have been reported stock sales by CEO Zakrzewski, Director James Healy, President John Thero and Abingworth LLP. The president of Amarin exercised 150,926 options in early April. The exercise price of these options was $1.35. He still holds 749,074 options, so this sale represents only 25% of his options holdings. CEO Zakrewski exercised 160,000 options in late March, with an exercise price of $1.35. He still has 515,000 options, so this sale represents only 24% of his options holdings. He still also holds 78,604 shares of Amarin. Director Healy had 8,513,388 shares of Amarin at the beginning of 2012. After recent sales, he still has 5,513,388 shares, 65% of that original position.
Abingworth LLP is a large holder in Amarin. In the last month, they have sold 6,308,341 shares of Amarin, leaving them with 1,599,168 shares. They have been a long time investor in Amarin, owning 17 million shares in early 2011. They have periodically sold large share blocks over the last 17 months, so we have no concern over their stock sales and low current share holdings. They still have a considerable position in the company.
While insider stock sales can be indicative of low confidence in a stock, we see no reason to be negative on Amarin based on recent insider sales. The leadership of the company still holds a sizable position, after rebalancing their holdings. We are confident in Amarin future prospects.
Amarin is relatively cheap
Amarin's share price has had its ups and downs in the last year. After the ANCHOR trial data was released on April 18th, the stock price jumped up from under $9 to $16+, progressing to almost $20 in the weeks after as investors speculated that the excellent data would soon lead to an acquisition. As the initial excitement wore off and concern regarding IP protection made shareholders more insecure, the stock price drifted lower. General market uncertainty and the debt crisis in late 2011 led to AMRN reaching prices lower than pre-ANCHOR data release, with the 52-week low of $5.99 in December 2011. However, after the Omthera data release, Amarin jumped from $10 to $12 and has stayed within the range of $10-12 since mid-March.
Based on previous interactions, USPTO responses to Amarin's responses post-non-final rejection have come at an average of 6 weeks post-submission. After Amarin's May 16th response, we could see a response from the USPTO in late June or early July. With Amarin's PDUFA date in late July, progress from the USPTO on the '889 patent would certainly boost their share price. Recently, Amarin's share price was heading back down to $10, but the recent ruling upholding patents on Lovaza (and limiting the risk of generic Lovaza) has boosted it back towards the upper end of its range.
As previously mentioned, GlaxoSmithKline's Lovaza had sales of $1B in the US in 2011. Lovaza is only approved for the very high triglyceride market; this represents only a fraction of AMR101's potential market reach. As the patents are issued and the IP protections made clear, Amarin could easily reach $20 again on news of FDA approval. A stock price in the low $20s would give AMRN a market cap of around $3B which seems reasonable given forecasted sales post AMR101-approval and considering the potential for expansion of their market after approval for the wider triglyceride indications and completion of the outcomes study.
Currently the FDA has set a PDUFA date of July 26th 2012 for the MARINE indication. Not only is FDA approval necessary to begin commercialization of AMR101 for the MARINE indication, but this approval is also the main requirement of an NDA for the 10-fold bigger market of the ANCHOR indication (the other requirement being that a clinical outcomes study (REDUCE-IT) with AMR101 be substantially underway, which is expected by end 2012). Given AMR101's positive results, we believe that the FDA decision for the MARINE indication will be favorable, thereby paving the way for entry into this billion dollar market.
Post-AMR101 approval, Amarin would be an excellent acquisition for a large pharma company looking to bolster its pipeline. CEO Zakrzewski is experienced at making large dollar value deals; during his 17-year career at drug giant Eli Lilly and Co he brokered deals valued at more than $2 billion. This bodes well for AMRN, which does not currently have the marketing and sales infrastructure needed to single handedly commercialize their new drug. A buyout would be advantageous for Amarin, offering the company a premium on the current stock price without requiring their investment in sales and marketing. We are optimistic about Amarin's prospects and believe that at $11.50 per share it is undervalued.
Disclosure: I am long AMRN.