If patents on brand-name drugs are the lifeblood of Big Pharma, then the mysterious Columbo may be the drugmakers' poison pill.

Columbo is the online handle for a man who won't publicly identify himself. But Article One Partners Holdings LLC, the Manhattan startup created to allow anyone who wants to check the validity of patents on products far and wide, has already paid him two $50,000 awards because he found "prior art" that Article One believes could invalidate the patents on two of the biggest-selling drugs in the world -- Crestor, the cholesterol drug produced by AstraZeneca plc, and Cymbalta, the antidepressant sold by Eli Lilly and Co.

Columbo says he is a researcher by trade, having worked in academia as a librarian and as a research analyst in the business world. "I enjoy research, and I especially like learning new methods, databases and strategies," he says in an e-mail account of his professional life. "Research is an art and a science. It involves the creativity of an artist and the method of a scientist."

Article One takes its name from Article One of the U.S. Constitution, enumerating the government's power to secure "for limited times to authors and inventors the exclusive right to their respective writings and discoveries." In all, the company has studied 52 patents, some initiated by Article One and others sponsored by corporate clients looking to validate their patents or undercut a competitor's. Investors poured $3 million into Article One's launch, and while chief executive Cheryl Milone (pictured) says the private company was not profitable last year, it expects to be in 2010.

When one of the searchers from around the world finds some possible prior art, Milone says the firm's experts and the patent law firm of Darby & Darby PC examine it to determine, at least in their minds, whether a patent might be invalid or possibly provide more proof that it is valid.

Its strategy then calls for licensing that information to whichever parties may want it, possibly generic-drug makers hoping to shorten the length of time for which a pharma holds exclusive rights to a brand-name drug.

Milone would not say what it has done with the information that Columbo discovered about Cymbalta, which IMS Health Inc. says produced $3.2 billion in sales for the year ended in September 2009, or Crestor, which brought in $4.9 billion. But it might be of particular use to eight generic-drug makers that are contesting the validity of AstraZeneca's Crestor patent, which extends to 2016, in a trial in Wilmington, Del., that is scheduled to start Feb. 22.

"We are in contact with potential clients," Milone says.

Fran Weber, an executive vice president at Article One who oversees the company's drug patent research, says that Columbo found information "in an academic journal" that appeared to undercut the Crestor patent, and "we gave it to an independent law firm and independent scientific analyst. They agreed that these references gave support for an invalidity."

Milone says the firm is impartial. "We're agnostic on who owns patents. We don't have an agenda." Indeed, the firm says it has closed numerous patent studies involving a variety of products, including such widely used drugs as Pfizer Inc.'s Lipitor and Effexor XR, Novartis AG's Lotrel and Takeda Pharmaceuticals North America Inc.'s Prevacid, acknowledging that the information collected by its researchers "does not form an independent basis for an invalidity position related to the subject patents at this time."

Milone says, "We have a number of clients we are talking to about strengthening their patents to better evaluate their patents to eliminate uncertainty." She says the firm has four Fortune 100 companies as clients, as well as one of the Big Three automakers, several law firms and top U.S. telecom companies.

In some of its earliest patent reviews, Article One made its information public to anyone who wanted to read it, in effect showcasing the skills of its citizens-of-the-world researchers. In one instance, after Northeastern University organic chemistry doctoral candidate Amy Kallmerten, 24, found potentially invalidating prior art related to Merck & Co.'s asthma drug Singulair, Article One submitted the information to the U.S. Patent and Trademark Office and asked for a re-examination of Merck's patent.

Kallmerten, who is headed to law school later this year and a career in patent law, says she studied the molecules comprised by Singulair in late 2008, "trying to break it apart so I could then search" for articles that might show information that predated Merck's patent. Eventually, she found scientific articles showing "the molecule was out there ... not all that different from the Merck one. Any person of ordinary skill would be able to develop it, that [the Merck patent] was not inventive.

"At first I didn't really know that I'd found anything," Kallmerten says. But she took the information to Article One. Weber says that Kallmerten and a researcher using the online handle of Arteriel, who was working independently of Kallmerten, produced various "pieces of the puzzle." Article One decided that the information might undercut the Merck patent on the blockbuster medication, one that IMS says produced $4.8 billion in revenue for the pharmaceutical in the past year. Article One awarded Kallmerten $35,000 and Arteriel $15,000 for their discoveries.

After Article One's request for the patent re-examination, the Patent Office proposed a 15-month cut in the Merck patent exclusivity on Singulair from the existing August 2012 expiration date. But after Merck submitted further information on its development of Singulair, the Patent Office upheld the original 2012 date. The firm also won a patent infringement case against Teva Pharmaceuticals USA Inc., the generic-drug maker seeking to produce a Singulair copy to sell.

Despite the eventual outcome, Milone says, "we felt the ability of our community was a very successful outcome for the public. It enabled the Patent Office to make a better decision."

So just who are these patent searchers Article One uses? Milone says that more than half of the 3,800 researchers have advanced science or technology degrees.

But a company spokesman says the mysterious Columbo "doesn't want anything out there that might divulge who he is," not even his age or where he went to school.

But Columbo did offer some e-mailed thoughts on his searching in response to questions posed to him by The Deal and passed along to him by the Article One spokesman.

"I approach each patent study differently, but some basic things are constant," Columbo says. "The first thing I do is to go forward in time, not backwards. This approach is counterintuitive, and it's a kind of cold-case detective approach. The patent authors put together some ideas. A patent was granted. Someone thinks that there could be questions about it. So let's put the evidence to the test and see what the results and research outcome are.

"If the goal is to find prior art, why would I first go forward in time to gather clues and details? It's an individual preference in strategy.

"For me, reading a patent is like reading Shakespeare. I need help. So I look for scholarly and even 'popular' journal articles that describe the drug upon which the patent is based. I read these articles because they describe, sometimes in layperson's terms, the drug's blueprint," Columbo says.

"There still are intellectual hurdles to clear, but the articles help me to understand what the drug is about: its discovery, structure and development. If reading a patent is like reading Shakespeare, then reading articles about a drug is like reading explanatory notes on Shakespeare," Columbo adds.

"Once I have a better handle on what I'm looking for, it's easier to find what I'm looking for. There are no guarantees that you'll find the needle in the haystack, but your chances are greater if you have a good grasp of the problem.

"And I don't assume anything at the outset," he says. "The patent could be valid, it could be invalid. I give the patent the benefit of the doubt. This approach allows for objectivity in research. I employed these philosophies and practices in the Cymbalta and Crestor studies. I do research consulting work as a source of income, outside of and unrelated to AOP, but AOP's been my focus lately. I have no guarantee that I will win or partially win any given study, but I give it my best shot."

And what did he do with the $100,000 he has won from Article One Partners?

"I bought my Labrador retriever a new doghouse," Columbo says. "I paid some student loans. And I put the rest in savings."

See the rest of the Healthcare Special Report: