In an unusual if not unprecedented situation, Osteoceutics has claimed the wrong chemical structure in its granted patent, US 8,673,932. The drug candidate has since entered Phase I/II clinical trials, and Oncoceutics was seemingly unaware of the blunder until a recent publication by The Scripps Research Institute emerged. In that publication, the Scripps identified the correct structure, which it has since gone on to license to Sorrento Therapeutics. In a final blow, Sorrento Therapeutics has filed a US patent application for the correct structure. How did this happen? And what are the consequences?
It is every patentee’s worst nightmare. On 21 May 2014, C&EN News reported an unprecedented drug discovery blunder, the magnitude of which is perpetuated by the granted patents and licensing arrangements pertaining to the implicated drug candidate. Drug discovery chemist and science blogger, Derek Lowe, discussed the issue further on In the Pipeline. The issue has left many scientists and intellectual property experts to speculate as to the potential outcome. Though what is certain, is that the consequences could prove devastating for the parties involved. The take home message according to Kim D. Janda, chemistry Professor at The Scripps Research Institute; “One lesson from this has got to be: don’t leave your chemists behind”.1
TIC10 was first reported as an antiseizure medication in a 1973 German patent (2150062) owned by C. H. Boehringer Sohn, a predecessor of Boehringer Ingelheim. The patent also disclosed the synthesis of TIC10. At some stage, the National Cancer Institute (NCI) incorporated TIC10 into one of its publicly accessible databases, and listed the chemical structure according to that shown in the Boehringer Ingelheim patent (Figure 1).
Click here to view image.
Figure 1: Linear structural isomer as shown in the 1973 Boehringer Ingelheim patent, The NIC database, and the Oncoceutics patent (US 8,673,932 B2).
In 2013, Wafik S. El-Deiry’s laboratory at Pennsylvania State University reported that TIC10 demonstrated potential anticancer activity.2 Mass spectrometry was used to confirm that the structure of the compound was that of the linear structure listed in the NCI database. The use of the compound as an anticancer therapy was patented by the group on 18 March 2014 (US Patent No. 8,673,932 B2). The patent was licensed to Oncoceutics, a US biotech company co-founded by El-Deiry, for further development. Oncoceutics followed the method disclosed in the Boehringer Ingelheim patent to synthesise TIC10, and the drug candidate is currently in Phase I/II clinical trials as ONC201, a potential anticancer agent.
Oncoceutics’ patent, US Patent No. 8,673,932 B2 consists of 72 Figures reporting the biological activity of TIC10, and claim 1 reads:
- A method of medical treatment of a subject having brain cancer, comprising: administering to the subject a pharmaceutical composition comprising a pharmaceutically effective amount of a compound of formula (I):
Click here to view image.
or a pharmaceutically acceptable salt thereof; and a pharmaceutically acceptable carrier.
The only chemical structure disclosed in the patent is, and repeatedly, the linear structure.
Meanwhile, Professor Kim D. Janda’s research group at The Scripps Research Institute (TSRI) commenced studies on TIC10 for its possible use in combination therapy. In doing so, the group synthesised TIC10 from scratch, that is, not according to synthetic method disclosed in the Boehringer Ingelheim patent. Surprisingly, the group found the synthesised compound to be inactive as a potential anticancer agent!
Experienced chemists will know that this situation occurs all too often. In many cases, the original screening compound is not clean, and any activity can be attributed to some sort of contamination. Janda’s group analysed the TIC10 sample from the NCI and found that this was not the case; the sample was clean and it was biologically active. The group subjected the NCI’s TIC10 sample to in-depth spectral analyses, and eventually x-ray crystallography confirmed the compound as the angular structural isomer (Figure 2).
Click here to view image.
Figure 2: Angular structural isomer identified and published by Janda’s group in Angewandte Chemie, and as shown in the Sorrento patent application.
Janda’s group published its findings last month.3 The results show that the Oncoceutics patent has a terrible problem: the wrong chemical structure is patented. To exacerbate the matter, Janda’s group has since licensed its technology to Sorrento Therapeutics, who promptly submitted a patent application disclosing the newly assigned TIC10 angular structural isomer.
Click here to view image.
Figure 3: Cartoon reprinted from the 2014 Angewandte Chemie publication by Jacob et al.4
How Did This Happen?
The error can be traced back to the original Boehringer Ingelheim patent, which disclosed the synthesis of the bioactive but structurally misidentified compound. What was demonstrating biological activity, in fact, turns out to be the angular structural isomer, not the linear structural isomer as described and claimed. In fairness, given the patent was granted in 1973, it is highly probable that spectroscopic analyses available at the time were unable to distinguish between the linear and angular structural isomers. It is not uncommon to retrospectively analyse structures published around this time and discover that they are incorrect, particularly where structural isomers are involved.
Upon incorporating TIC10 into its database, it is unknown if the NCI performed further structural analyses. It is altogether possible, and some would argue reasonable, for the NCI to have merely performed analyses to confirm the linear structure. If this is what happened, then the angular structural isomer’s identity had little chance of being revealed.
The buck then falls to Oncoceutics. Drug discovery chemists are aware that identifying a ‘hit’ compound, one that is active against your biological target, is only half the battle. What should follow are detailed chemical analyses of the hit compound, in this instance TIC10, to assign its structure and confirm its purity. Ideally, the compound should be resynthesised via an alternate synthetic route and again assessed for activity against the biological target. Validating the hit compound to this extent is not an unreasonable expectation in drug discovery ventures. Rather, it is simply good practice, and a way to safeguard against situations akin to what Oncoceutics now find themselves embroiled in. Greater care, time, and effort should have been invested before the compound was patented, and most certainly, before the compound entered clinical trials.
As it stands, Oncoceutics’ chief business officer, Lee Schalop, has told C&EN that the “chemical structure is not relevant to Oncoceutics’ underlying invention”, and that plans for the clinical trials are moving forward. Oncoceutics released the following statement:
“Our invention is the discovery that the lead compound, ONC201, which was originally synthesised by Boehringer Ingelheim, distributed by the National Cancer Institute (NCI), and manufactured by Oncoceutics, has unexpected activity against human cancer. The fact that the drawing included in the Boehringer Ingelhem patent, the NCI literature characterising the compound, and ultimately our patent does not accurately represent the isomeric details of the structure does not impact Oncoceutics’ underlying invention”.5
Really? If the patent is held invalid due to the incorrect structure being disclosed, then one does not need to be an intellectual property expert to imagine that this would impact, very heavily, upon Oncoceutics’ underlying invention and commercial position.
An Intellectual Property Perspective
The drug discovery blunder raises two key questions:
- Will the Oncoceutics patent be held as valid?
- Will the Sorrento application be granted, regardless of whether the Oncoceutics patent is held as valid?
Several US patent attorneys have weighed in on the debate with varying opinions. Some believe that the mistake may not be fatal for Oncoceutics.6 Others believe that Oncoceutics’ patent may prevent Sorrento’s application from being approved, as the identification of the correct chemical structure may have been within the skill of a person working in the field.7
If the case were to play out under Australian patent law, Oncoceutics may have significant problems that would probably invalidate its granted patent. Could Oncoceutics claim a clerical error in that the structure was printed in error? Probably not. El-Deiry’s group didn’t know the angular structure existed. It could only be deemed a clerical error if they knew of the alternate structure. In addition, the structure is depicted several times in the patent as the linear isomer. A clerical error would not be a viable basis for amendment. The Patents Act 1990 requires that a patent specification must disclose the invention in a manner that enables a person skilled in the art to perform the invention, and that the subject matter of the claims must fulfil the promise of the invention (sufficiency and fair basis, s 40(2) and s 40(3)). Quite simply, if a person skilled in the art were to attempt to perform the invention according to Oncoceutics’ disclosure, they would not arrive at a chemical compound capable of reproducing the biological activity disclosed in the 72 Figures of the granted patent. On this basis alone, the patent should be held invalid.
Whether the invention of the Sorrento application would be considered obvious to a person skilled in the art in view of Oncoceutics’ patent is debatable. Pragmatically, given the mistake has existed since the 1973 Boehringer Ingelheim patent, it would appear as though Sorrento’s invention really may not have been obvious to a person skilled in the art. However, it’s not that simple. In considering Sorrento’s application, it may be found that it would be obvious to a person skilled in the art to try to synthesise the angular structural isomer in the search for the biologically active compound with a reasonable expectation of success. If so, then the claims of Sorrento’s application cannot be granted. Conversely, the 2009 Australian decision in Lundbeck v Alphapharm ( FCAFC 70) found that an enantiomer of Citalopram was patentable despite the other enantiomer previously being patented. This decision was founded in the difficulty of synthesis of the second enantiomer, and would support the patenting of Sorrento’s invention.
This, here, is the key matter. Will a prior disclosure, albeit an incorrect disclosure, be considered to render the current invention obvious? We will have to wait and see. No doubt many, both scientists and intellectual property experts alike, will be eagerly awaiting the outcome.
The Take-Home Message
The question must be asked – how could it have gotten this far? Yes, mistakes in science are inevitable, but it is hard to recall an instance where the wrongly identified structure has advanced as far as clinical trials. If nothing else, the blunder highlights the value of the chemist capable of performing high-level structural analyses during the drug discovery process. Recent times have seen cuts to scientific funding, with many biopharmaceutical companies consequently slashing jobs across chemistry departments. Janda believes that fuelling the recent blogging and internet discussion around TIC10 has been the “backlash from so many pharmaceutical companies having laid off their chemists and disbanded their medicinal chemistry divisions”.8 “Chemistry in pharmaceutical companies these days has taken a real back seat,” Janda said, “Lots of chemists have been laid off recently, and there’s a real place and need for chemistry”.9 Oncoceutics could have most certainly benefited from greater chemist involvement in their drug discovery efforts. If Oncoceutics had done so, the problem would have been identified much earlier in the pipeline, avoiding what will no doubt be a tricky intellectual property dilemma. As so perfectly stated by Janda: “Now we see how ditching the chemistry can really bite you in the ass”.10