Monday, April 13, 2015

Interested in the History of Biotech Companies? Don't start with Wikipedia.

I'm generally a big fan of Wikipedia and use it often for background research.  I've gotten more active this year in editing it, particularly around biographies of scientists.  For example, this year I've made major additions or edits to the entries for Walter Gilbert and Arthur Pardee and the , created entries for Martinas YcasBenno Müller-HillMonica Riley and Helen Donis-Keller. I also stumbled my way into a campaign of major revisions to the entry on Marie Antoinette, getting sometimes into a revision war with one other editor (which we resolved with a truce).  Along the way I've gotten almost adept at writing Wikipedia references and discovered a bizarre recurrent vandalism of Wally's page in which the vandal changes his name and personal details.  Recently, I've discovered a whole category of flawed entries: those on companies in the biotechnology industry.

Wikipedia is not the ultimate storehouse of all human knowledge, but it is a good starting place.  Therefore, I like Wikipedia entries that are polished, readable, accurate and represent comprehensive surveys of a subject.  My edits can often be seen in this light.  For example, the Walter Gilbert article was missing many of the more interesting parts of his career, such as his tumultuous time as CEO at Biogen and his purification of lac repressor. The ultimate in lack of comprehensiveness is to not exist, hence the various stubs I have created.  The fact that female scientists are underrepresented in Wikipedia also drove my creation of the Riley and Donis-Keller entries.  A lack of historical context spurred me to an edit on the demolished Gettysburg Cyclorama building.

Personal angles don't hurt: in addition to the Gilbert page I've enhanced the entry on my late uncle in the movie business.  An entire experiment was missing from the Galileo spacecraft entry; my father contributed to the Galileo probe heat shield experiment instrumentation, which is how I knew it was missing from the entry.  A long-standing interest in old train stations triggered my creation of an entry for one in Wilmington Delaware.   Monica Riley, who I was lucky to meet once,  was an important contributor to bioinformatics. If I do have a strong personal angle, I try to be utterly upright about it, marking the fact on the talk page, sticking to pure facts and being exquisitely careful in citations.

I recently reread the classic biotech book The Billion Dollar Molecule as a prelude to reading the sequel The Antidote.  They're both referenced in the Wikipedia entry on Vertex, but little else of history of the company is given.  You won't find any mention of Vertex successfully developing an HIV protease inhibitor or the fact they sold away the rights to it.  The tumultuous history of their Hepatitis C drug gets only a brief mention.

The Vertex entry seems, from a very unscientific scan of mine, to be representative of most entries on biotech companies either extant or defunct.  Too many read like company websites.  Few have information on previous trials and tribulations.  They are (at the time of this writing) dull, dry and incomplete.

For example, the page on Centocor (now Janssen Biotech) has very little of the history of the company other than the present, though there is a brief and incomplete timeline.  No mention of Centocor developing an early HIV diagnostic (one of my tasks as a summer intern was to clear out the last lab from that that effort).  An even bigger gap is the the lack of coverage of the patent battle royale with Xoma on antibodies to treat sepsis, a battle halted only because both drugs failed miserably in the clinic (but not before at least one of the companies shipped many doses to the U.S. military in conjunction with the First Gulf War).

Similarly, the page on Genetics Institute has no whiff of the tremendous patent fight with Amgen over EPO.  Amgen's page fails to mention that, as well as giving little coverage of their early wanderings in search of a product.  Millennium's page hasn't been updated to mark the demise of that moniker, nor does it make it clear that Velcade came via the Leukosite acquisition. Mention is made of the many strategic collaborations, but not the variety of spaces nor how fallow all those proved to be. Biogen's page has no mention of their early wanderings in search of a product (such as Hirulog, derived from leech spit), nor the very difficult troubles with Tysabri.  Of course it could be worse; neither Infinity Pharmaceuticals nor Codon Devices has an entry at all.

Increasingly, if Wikipedia isn't the first draft of history it is at least the notes for writing the first draft.  I feel having balanced, complete articles on biotechs is important.  First, it should be an embarrassment to anyone in the field that so many entries look like the work of corporate shills; it reflects badly on all of us.  Second, the various twists and turns are important parts of the history of the field and important object lessons for current and future participants.  Product ideas rise and fall, lawsuits are fought and won and lost, trials succeed and trials bomb.  Codon might have been short-lived and small, but it engendered more than a small amount of news and concern in that short span. Surely a Wikipedia that can have individual entries for thousands of mediocre sports players can accommodate hundreds of short-lived biotechs.

Of course, it is Wikipedia and all the faults above could succumb to a campaign of editing.  I might even try to engage in some, once I unbury myself from a mountain of other projects.  I write this in the hope that I can enlist others in this task, particularly since I'm not qualified to spot what is missing on most pages.  Only an army of interested individuals has any hope of fixing what is wrong with biotech companies in Wikipedia.


Jonathan Badger said...

Well, Wikipedia is always a battleground when it comes to anything current and controversial. It shines when things are far enough removed from current attention to allow objective reflection.

The issue with existing companies is that people involved with them are mostly interested in promoting them. not in giving an objective history. It would be nicer if people would do something with the defunct ones though. There are Wikipedia pages on defunct fast-food chains, surely defunct biotech companies are at least as significant.

I'm glad that you created an entry for Monica Riley though, as short as it is. I met her at the 1992 Molecular Evolution course at Woods Hole and in retrospect it is interesting how her work in metabolic modeling was so similar to what would later be called "systems biology".

Greg said...


I used to work down the hall from Monica when she was at the MBL and focusing on EcoCyc. I confess I never knew much about her early work until I read of it in Sydney Brenner's wonderful and too brief autobiography.

Thanks for making the entry.