Gilead Sciences Gets Ambushed By The Patent Troll, AbbVie

| About: Gilead Sciences, (GILD)
This article is now exclusive for PRO subscribers.


AbbVie has been granted five US patents that specifically claim use of sofosbuvir and ledipasvir, the drug combination that Gilead sells under the trade name Harvoni.

Gilead has been granted no patents for the Harvoni combination, although an application was filed in Sep2011, a month prior to ABBV’s earliest application.

AbbVie cannot market Harvoni since it does not own patents on either individual drug in the combination, but can potentially compel royalty payments and damages from Gilead.

Each side has filed lawsuits – resolution could take years.

Gilead (NASDAQ:GILD) is angry - real angry. AbbVie (NYSE:ABBV) has ambushed them, obtaining five "method of use" and "utility" patents for the drug combination marketed as Harvoni. Gilead has zero such patents. By marketing Harvoni, Gilead is currently infringing the ABBV patents.

The first question that comes to mind is: how could it happen that the company that is now selling Harvoni has no granted "method of use" patent on the combination therapy while the company that cannot sell this drug combination has five? The answer: ABBV concocted a clever (or devious, depending on your point of view) strategy for convincing the patent office that they have invented a novel computer model to predict effective combination therapies.

The difficulty in obtaining patents on most combination therapies is overcoming obviousness. It makes sense that combining two drugs, especially those that work by different mechanisms, to treat a disease might be better than using only one. In addition, combination therapy is standard for treating viral diseases such as HIV, so it is a no-brainer to anticipate combination therapy would be effective for HCV as well. Because of these constraints, the inventor who applies for a patent on a drug combination for HCV needs to demonstrate a novel or unexpected outcome would result. AbbVie overcame the obviousness hazard by crafting an argument that a sophisticated computer model they developed allowed them to conduct virtual clinical trials to test combinations of drugs for treating HCV. The outcome of these virtual clinical trials was the identification of a few drug combinations that would be most effective, in the shortest amount of time, while not requiring interferon or ribavirin - all without the need to subject patients to many ineffective combinations by trial and error through actual clinical trials. While this all may be a fairy tale, ABBV created a compelling story in their patent applications and convinced the patent office they had a novel, unexpected, and useful invention.

AbbVie used their virtual clinical trial model not only to convince the US Patent and Trademark Office (PTO) of its novelty, but also persuaded them to grant expedited review. This led to issuance of the first two patents before Gilead had a chance to respond. Gilead was ambushed, caught completely flat-footed by an aggressive AbbVie. Ironically, the more mundane "method of use" patent application from Gilead, which was filed a month prior to ABBV, has languished at the patent office, with no action yet taken. Gilead sums up the situation in the 10-Q filed earlier in November:

We own published and pending patent applications directed to the use of combinations for the treatment of HCV, and, specifically, to combinations of sofosbuvir and ledipasvir.

In other words, they have filed patent applications, and the applications have published, but they do not have any granted patents.


Gilead has thrown the first punch. Gilead sued in December 2013, claiming ABBV

falsely represented to the U.S. Patent Office that they invented methods of treating hepatitis C that were actually invented by Gilead and Pharmasset.

Throughout their lawsuit, which in some places borders on an unprofessional rant, Gilead accuses AbbVie of


egregious misconduct

illicit plot

false representation

illegal scheme


intentional and willful violation of the Patent Laws of the United States

knowingly failing to disclose material prior art, and

a specific intent to deceive.

In response, on February 18, 2014, and again in March (now consolidated), AbbVie filed a suit against Gilead for infringement of U.S. Patent Nos. 8,466,159 and 8,492,386 based on Gilead's anticipated commercialization of the combination therapy of sofosbuvir and ledipasvir (now known as Harvoni) for the treatment of HCV. In addition to refuting Gilead's arguments, in their lawsuit AbbVie comments on the Gilead lawsuit:

The [Gilead] complaint slanders AbbVie's HCV drugs, its corporate integrity, and the integrity of its scientists and its attorneys, and overall seeks to disparage AbbVie's scientific reputation. Indeed, it makes numerous statements that it would not be permitted to make outside a privileged litigation pleading without incurring liability for slander or reprisal from FDA.

ABBV argues that the commercialization of Harvoni by Gilead (illegally) will result in a loss of revenue from AbbVie's own combination therapy.

As a result of its infringement of AbbVie's patents, Gilead will owe substantial damages, including but not limited to the profits that AbbVie would have made through the sale of its own patented combination HCV products. And AbbVie will ask for those damages to be tripled, as provided for under 35 U.S.C. § 285, and an award of attorneys' fees. Such remedies are particularly appropriate here, given Gilead's deliberate conduct.

Thus, the patent war between Gilead and AbbVie is officially underway. It has already gotten ugly - and the fun is just beginning.

The ABBV Patents

As most of readers of this article know, the recently approved drug Harvoni is a combination of two drugs, the nucleoside inhibitor sofosbuvir and the NS5A inhibitor ledipasvir.

AbbVie has been granted five US patents, 8,466,159, 8,492,386, 8,680,106, 8,685,984, and 8,809,265, for the use of antiviral medications in combination for the treatment of HCV in 12 weeks without the need for interferon or ribavirin.

Each ABBV patent claims their own combinations, then also claims a few other combinations, most notably sofosbuvir and ledipasvir. Two patents claim use without interferon and two patents claim use without interferon and ribavirin. The last patent claims use of sofosbuvir with any NS5A inhibitor.

All patents have a priority date of Oct 21, 2011. The first patent was granted on June 18, 2013 and the most recent on Aug 19, 2014. More patents may be on the way.

The 4 claims related to Harvoni (sofosbuvir (PSI-7977) and ledipasvir (GS-5885)) of the '386 patent are copied below. These claims describe the current use of Harvoni to treat genotype 1a HCV patients.

13. A method of treatment for HCV, comprising administering at least two direct acting antiviral agents (DAAs) to an HCV patient infected with HCV genotype 1, wherein said treatment does not include administration of either interferon or ribavirin to said patient, wherein said at least two DAAs comprise PSI-7977 and GS-5885, and wherein said treatment as for 12 weeks.

14. The method of claim 13, wherein said patient is a treatment-naïve patient.

15. The method of claim 13, wherein said patient is infected with HCV genotype 1a.

16. The method of claim 14, wherein said patient is infected with HCV genotype 1a;

ABBV - A Patent Troll

We often hear of patent trolls in the tech industry, but rarely see this from major pharma companies. However, the term applies perfectly to AbbVie's strategy to either prevent Gilead from selling Harvoni or compelling royalties. Clive Thompson defines "patent troll" as follows on page 64 of his book, Smarter Than We Think:

They're people who get a patent for something without any intention of actually producing the invention - it is purely so they can sue, or soak, people who go to market with the same concept.

ABBV cannot commercialize Harvoni since they do not own any patents to the individual drugs that make up the combination, sofosbuvir and ledipasvir. However, it is perfectly legal to apply for and obtain "method of use" and "utility" patents for products that a company does not own and this is what ABBV has accomplished with its five patents. In their lawsuit against Gilead, AbbVie offers a precedent from a landmark case in 1988, Kingsdown Medical Consultants v. Hollister, Inc. (bold and italics are those of AbbVie):

It should be made clear at the outset of the present discussion that there is nothing improper, illegal or inequitable in filing a patent application for the purpose of obtaining a right to exclude a known competitor's product from the market; nor is it in any manner improper to amend or insert claims intended to cover a competitor's product the applicant's attorney has learned about during the prosecution of a patent application. Any such amendment or insertion must comply with all statutes and regulations, of course, but, if it does, its genesis in the marketplace is simply irrelevant and cannot of itself evidence deceitful intent.

863 F.2d 867, 874 (Fed. Cir. 1988)

The above quote perfectly summarizes the ABBV strategy and may well form the basis of their litigation approach. There is a decided edginess in AbbVie's approach to obtaining these patents, something we may not have expected from a conservative Midwestern pharma company. It's an edginess that may play out in the head to head competition with Gilead once AbbVie's HCV combination is approved in December.

How did ABBV get here?

Let's back up to the beginning to understand how AbbVie got to this point, outscoring Gilead 5 - 0 in Harvoni patents. Most of the following narrative is based on information provided in the ABBV and GILD patent applications, the granted ABBV patents, and the lawsuits each side have brought forward.

First of all, let's review how patent applications work. When an inventor has conceived a novel idea and has reduced it to practice, but has not yet ironed out all the details, they can file what is called a provisional patent. This is a placeholder or a stake in the ground regarding the idea and establishes the initial priority date of the invention. This provisional patent application is not examined by the patent office.

Over the next 12 months, the inventor can add to the provisional patent to flesh out the details that were not fully worked out when the provisional was filed. At the end of 12 months, the inventor must decide whether to move ahead with a full filing (called a non-provisional patent) or let the application lapse. If they decide to move forward with the non-provisional patent application, the application is officially filed and is published 6 months later - that is 18 months after the original provisional filing. During these 18 months, the inventor has kept the invention a secret. It should be noted that this publication is not a patent - just a patent application that is an announcement to the public of the invention. By the time the patent is examined by the US Patent Office, the granted claims of the patent might look completely different from those in the original patent application.

ABBV filed the first two provisional patents on HCV combination therapy on Oct 21, 2011. The patents are virtually identical, except one is for treatment without interferon and the other is for treatment without interferon or ribavirin. Their filing date was exactly one month before Gilead announced they were acquiring Pharmasset for $11B. At the time of filing, then, ABBV would not have known about the Gilead impending purchase. The original 2 provisional patent applications filed by ABBV (U.S. Provisional Application Nos. 61/550,352 and 61/550,360) are available via the Patent Application Information Retrieval (PAIR) service offered by the US Patent Office.

Each of the 2 provisional patents includes 32 broad claims that cover hundreds of potential combinations of the dozens of oral drug candidates that were in development by various companies. The provisional patent applications broadly included sofosbuvir and GS-5885 for potential use in combination therapy but not that specific combination.

No mention is made in either provisional patent application of the computer model or virtual clinical trials - this material was added during the 12-month period from provisional filing to full filing. This is going to be a key point in the litigation since new information added in the non-provisional patent may not have the benefit of the provisional filing date. This can affect patentability if information that is considered prior art is made public after the filing date of the provisional but prior to the filing date of a non-provisional application.

ABBV continued with their aggressive filing of patent applications:

  • Provisional patent application Nos. 61/600,276 and 61/600,468 titled "Methods for Treating HCV," was filed on February 17, 2012, which outlined their virtual clinical trial model for the first time. This date is important since it came after Gilead had acquired Pharmasset on Nov 21, 2011. In the press release, Gilead signaled its intention to pursue combinations with sofosbuvir and, in particular, GS-5885, or ledipasvir. They further outlined their clinical development plans for Sovaldi, including plans to study the combination with ledipasvir with 12-weeks of treatment, during the 4Q11 earnings conference call held on Feb 2, 2012.
  • Provisional patent applications 61/656,251 and 61/656,253 were filed on June 6, 2012 that specifically included the sofosbuvir and ledipasvir combination for the first time. As mentioned above, by this time it was well known that Gilead was advancing the ledipasvir and sofosbuvir combination into late stage clinical trials.
  • ABBV filed provisional patent application 61/645,696 titled "Solid Compositions" on May 11, 2012, which claims oral dosage forms of the Harvoni combination as well as oral dosage forms of other combinations of competitor compounds.
  • On September 4, 2012, Abbott filed utility patent applications, 13/603,022 and 13/603,006 claiming Harvoni in claims 18-21 and Sovaldi in claims 29 and 30. Abbott requested expedited review and the PTO agreed.
  • On October 19, 2012, Abbott filed another utility patent application, 13/656,012 claiming 12-week methods of treatment of sofosbuvir with any NS5A inhibitor.

To date these applications have resulted in three additional patents being granted that cover Harvoni, including 8,680,106 issued on Mar 14, 2014, 8,685,984 issued on Apr 1, 2014, and 8,809,265, issued on Aug 19, 2014. All patents claim a priority date of Oct 21, 2011, the date when the first two provisional patents were filed.

In Europe, AbbVie has pursued the same strategy, filing a patent application covering the use of sofosbuvir and ledipasvir for 12-week treatment of HCV genotype 1. (EP 2583677 A3, WO2013059630A).

How ABBV Ambushed Gilead

A provisional patent application is kept a secret for 18 months. So while AbbVie was busy filing all these applications that covered Gilead's combination, Gilead had no idea this was occurring.

During the summer and fall of 2011, Gilead was actively pursuing Pharmasset and its key HCV drug, PSI-7977, now known as sofosbuvir or Sovaldi. As negotiations were ongoing, Gilead had the foresight to file their own provisional patent application WO 2013/040492 A2 entitled "Methods for Treating HCV," which discloses the combination of sofosbuvir and ledipasvir. Gilead filed its application on Sep 16, 2011, just over a month before ABBV filed its application.

While the filing dates of these provisional patents are important, first to file was not the law for establishing priority in 2011 (it is now, since 2013). In the US, priority was established based on who first reduced the invention to practice. Determining priority is a tricky business, which usually involves going back to memos and even notebook entries written by scientists. This system is fraught with problems and is readily manipulated, which is why the US finally converted to the "first to file" system used everywhere else in the world.

Gilead and the rest of the world first learned of AbbVie's first patent applications upon publication on April 25, 2013, of US 20130102557. This application did not look much like the original provisional application. The computer modeling and virtual clinical trials was new subject matter that had been added during the 12-month provisional period. The original provisional application had 32 broad claims while the published non-provisional application had many specific claims for combinations that were in late stage clinical development, including the Harvoni combination.

Since AbbVie had requested expedited review of these applications, by the time of publication, the applications had already advanced through examination and just 6 days later, on May 1, 2013, the US Patent Office issued notices of allowances for the two ABBV patents. The two patents were formally issued on June 18, 2013 and July 23, 2013.

The Gilead patent application, which was filed a month prior to AbbVie's, was published on March 21, 2013. This is an important date since ABBV should have been aware of the Gilead patent application before their own patent application published. In other words, they would have understood an interference existed (two applications that have overlapping claims) but they chose not to inform the patent office and thus, the ABBV patents sailed through the final examination and issued in the summer of 2013. Gilead wrote several letters to ABBV notifying them of the Gilead patent application but ABBV ignored the letters and did not notify the PTO of a conflicting application.

Handicapping the Litigation in the US

I am not a lawyer so I cannot begin to predict how the litigation might turn out. I am also likely missing many legal nuances on which the case could turn. However, after studying the background, I think it is safe to say this is not a frivolous case nor a slam dunk for Gilead. As a Gilead investor, I wish I could say it was, but ABBV has built, and is continuing to build, an IP portfolio on Harvoni and they show every intention of mounting a vigorous defense. ABBV has already shown itself to be a tough and creative opponent.

Key points that will be considered as the lawsuits move forward include:

1. Obviousness

2. Priority

3. Playing by the rules

1. Obviousness

The ABBV non-obvious argument

The novelty that underpins the ABBV combination patents, according to the company, is based on development of a proprietary, sophisticated computer model that examined clinical trial data on dozens of individual compounds. The model predicted which combinations and doses would be effective in HCV patients in a treatment period of 12 weeks or less without the need for interferon or ribavirin. AbbVie claims their model was not only able to predict and inform development of AbbVie's proprietary compounds in combination, but could also predict the success of drug candidates developed by others that had never been administered in combination.

According to the company, these virtual clinical trials allowed them to quickly identify the best combinations of drugs with the shortest dosing regimens, without requiring patients to test regimens that might ultimately prove to be ineffective. The company claims they were the first to recognize that short duration, interferon-free treatment of HCV was achievable - contrary to current thinking at the time.

According to ABBV, when they discussed running clinical trials with two of their oral drugs with no interferon, their consultants and the FDA expressed doubts it would be effective within a 12-week period, if at all. Nonetheless, AbbVie initiated in Oct 2010 a small 12-week, interferon-free, study in HCV genotype 1 patients with a combination of two of its drug candidates, ABT-450 and ABT-072, and ribavirin. According to ABBV:

The clinical trial was a success. Nine out of eleven patients in AbbVie's clinical trial achieved sustained undetectable levels of HCV after the 12-week treatment regimen. The results of this clinical study validated AbbVie's modeling approach, showing that combination treatment with certain DAA's for short duration, interferon-free could achieve SVR in patients with HCV genotype 1.

AbbVie followed this up with a larger study with a different combination of ABT-450 and ABT-333, again with ribavirin but with no interferon. This combination cured 95 percent of patients receiving the high dose of ABT-450. On October 21, 2011, AbbVie announced these results publicly and secretly filed its first two provisional patent applications. AbbVie summarizes their patent position by claiming

It was unexpected that an interferon-free and ribavirin-free treatment using a combination of two or more DAAs, and for a duration of no more than 12 weeks, can achieve significant SVR. It was also unexpected that an interferon-free and ribavirin-free treatment using a combination of two or more DAAs, and for a duration of no more than 12 weeks, may achieve significant SVR in interferon non-responders (e.g., null responders).

The essence of the ABBV argument is that the computer modeling led to an unexpected discovery - that short term, all-oral therapy could be highly effective in treating HCV if you chose the right combination of drugs. ABBV claims they were the first to make this insight and all other companies, including Gilead, have followed their lead.

Counterpoint - The Obvious Argument

Combination therapy is the predominant treatment for HIV, so it would be obvious to combine drugs that are mechanistically different to treat another viral infection, HCV. In addition, given the significant side effects of IV interferon, which causes a high level of patient discontinuation, it was well recognized that the primary goal of combination therapy would be all-oral therapy and the avoidance of interferon.

Roche appears to be the first company to report combination therapy in the absence of interferon. In April 2009, Roche reported the results of a 14-day trial with 88 patients of a protease inhibitor R7227 with a nucleoside inhibitor R7128, in the absence of interferon. This combination produced an average 5 log reduction in viral count. This was subsequently published as a Lancet paper in 2010.

In a Boehringer Ingelheim study, published in Oct 2010, 3 drugs were tested in combination in the absence of interferon. In this 32 patient study, after one month treatment, all patients in the high dose group (17/17) had an undetectable viral load <25 IU/ml.

The 45th Annual Meeting of the European Association for the Study of the Liver (EASL) in April 2010 described combination studies in the absence of interferon from Pharmasset and Idenix.

On January 10, 2011, Pharmasset and Bristol Myers-Squibb announced a 24-week proof-of-concept study of PSI-7977 (Sovaldi) in combination with BMS-790052 (daclatasvir), an NS5A inhibitor. This combination ultimately has turned out to be highly effective in treating HCV in 12-weeks without interferon.

It seems clear that by the time ABBV filed their combination provisional patents in October 2011, that many companies were pursuing combination therapies with the goal of eliminating the use of interferon. However, most of these early combination studies never evolved into meaningful treatments, so it can still be argued that ABBV's model could have been useful in predicting which combinations were most likely to be successful.

2. Priority

Gilead was first to file for a combination patent that included Harvoni, with their provisional patent filed on September 16, 2011. ABBV filed on October 21, 2011. Thus, the Gilead PCT publication may qualify as prior art under 35 U.S.C. § 102(e) because the subject matter of Gilead combination "was described in an application for patent, published under section 122(b), by another filed in the United States before the invention by the applicant for patent …."

However, as discussed above, priority in the US at that time was not based on first to file, but who made the discovery first. For filings occurring this close, determining priority may require looking back at memos and meeting minutes from the two companies.

On the other hand, the primary novelty of the ABBV patent is their computer modeling and in silico clinical trials. This material was not included in the first provisional filings in Oct 2011. This information was added later, and in fact, ABBV filed two additional provisional patents to describe the computer model, one in February 2012 and one in June 2012, the latter of which was the first that specifically claimed sofosbuvir and ledipasvir. While it is acceptable to add supporting data to a provisional filing, brand new material that is added may not get the benefit of the original priority filing date. This could well become a key point of contention and the most vulnerable aspect of the ABBV case, since by the time the February 2012 provisional was filed, Gilead had already disclosed their plans to advance sofosbuvir and ledipasvir into combination clinical trials, which would be considered prior art.

3. Playing by the Rules

In their lawsuit, Gilead repeatedly accuses ABBV of illegal and unethical activities. Based on the information provided in Gilead's lawsuit, I see no merit in Gilead's claims of illegality. When ABBV filed their initial patents they were not aware that Gilead was going to acquire Pharmasset and advance sofosbuvir and ledipasvir as a combination therapy. Filing a patent to protect their own combination therapy that was in development, as well as attempting to cover other combinations of competitor compounds, is nothing more than a good business practice. To actively pursue the Gilead combination once they realized Gilead likely had a best-in-class therapy that would outcompete their own, is again not illegal and is an acceptable, although perhaps unsavory, business practice to protect your product against a competitor.

The one questionable activity of ABBV was their failure to notify the PTO of the Gilead patent application that was published about 5 weeks before the PTO allowed the claims for the first ABBV patent. But as noted in the following publication, precedent from the Kingsdown case in 1988 declared that

an intent to deceive cannot be inferred solely from the failure to disclose a material reference. (Source)

There could be many reasons ABBV did not notify the PTO - and since the time is only 5 weeks, it is easy to claim it was simply overlooked. Gilead first notified ABBV of the Gilead patent one day after the claims were granted (May 2, 2013) but before the final fees were paid and the paperwork finalized. Thus, ABBV skates on the edge here.

Summing it Up

AbbVie has staged a stealth intellectual property attack on Gilead's Harvoni, taking advantage of a patent system that allows patent applications to remain secret for 18 months after the initial filing. After filing 2 patent applications in October 2011, ABBV has continued to aggressively file multiple applications, unknown to Gilead or the rest of the world. In addition, by requesting and receiving expedited review for these applications, ABBV was able to obtain two methods of use patents on the Harvoni combination therapy just days after Gilead knew they had even filed any applications.

Gilead had the foresight to file a patent application on their potential combinations while they were negotiating to acquire Pharmasset. They filed a patent application 5 weeks prior to ABBV, but ABBV was able to advance their patents through the examination phase and obtained 2 patents that cover Harvoni during the summer of 2013. Meanwhile, Gilead's combination patent has languished at the PTO.

Will the ABBV patents be upheld? Can ABBV compel royalties and damages from Gilead? Will the uncertainty of the patent situation with ABBV constrain the GILD share price? No crystal ball here but what seems sure is that neither side will back down and this will drag on for years. The weakest point in the ABBV case appears to be that their touted computer model was not included in their original filing and was only added after Gilead had announced their plans for ledipasvir and sofosbuvir as a combination treatment. ABBV may not be able to hold on to the October 2011 priority date, which would make them vulnerable to prior art.

While the litigation will likely take many years to resolve, the surprising aggressiveness of ABBV to challenge Gilead on intellectual property makes me believe they will creatively market their "ABBV combination" once it is approved in December. I think the share price of Gilead has stalled over the past few weeks due to uncertainty of the ABBV launch and how ABBV pricing may influence formulary decisions. Many months ago ABBV management suggested they would not compete on price - but now we know they are the masters of subterfuge. ABBV has had many months to devise and develop a marketing plan - expect the unexpected.

In my opinion, the worries regarding the HCV patent portfolio and concerns regarding the sustainability of the Sovaldi/Harvoni pricing are already baked into the GILD share price, with GILD undervalued on most metrics vs. its pharma peers. A disruptive marketing approach by ABBV would cause a short term GILD share price dip that will afford a great opportunity to buy one of the world's best run companies.

Disclosure: The author is long GILD.

The author wrote this article themselves, and it expresses their own opinions. The author is not receiving compensation for it (other than from Seeking Alpha). The author has no business relationship with any company whose stock is mentioned in this article.