Another non-unanimous PNAS retraction, for potential anti-cancer drug, after company’s method proves unreliable

There’s another non-unanimous retraction in the Proceedings of the National Academy of Sciences (PNAS) this week, and this one involves an anti-tumor antibody that may not be what the authors originally thought it was.

According to the notice for “The antibody zalutumumab inhibits epidermal growth factor receptor signaling by limiting intra- and intermolecular flexibility:”

The undersigned authors wish to note the following: “In our study we employed Protein Tomography (PT), an electron tomography method commercialized by Sidec AB, for structural analysis of proteins. Following our publication, doubts were raised with respect to the validity of Sidec PT, and we therefore conducted an extensive validation study with Sidec AB to determine the fidelity of PT for structural analysis of therapeutic antibodies and antibody–antigen complexes in solution. In two independent double blind experiments, PT was found to be highly unreliable in distinguishing structural features of the molecules and complexes studied. First, approximately 90% of the identified protein density maps could not be interpreted due to complex morphology or low quality. Second, among the remaining objects, a high number of the protein images observed did not match with sample composition resulting in misinterpretation of sample identities. These disappointing results led us to reanalyze our previously acquired PT data in situ using a weighted backprojection reconstruction method with IMOD (1) rather than COMET (2), which also indicated our previous analysis to be unreliable. With the current PT methods, we were thus not able to validate tomograms neither obtained from vitrified proteins in solution nor aldehyde-fixed, stained cells. Therefore the undersigned authors no longer feel confident of the PT data presented in Figs. 3, 4, 5, and S4 of our study. The authors stand behind the supporting biochemical data provided in the manuscript and believe the proposed model to be plausible, the validity of which, however, should be addressed with other methods. We found it important to notify our colleagues of the specific technical flaws in our publication and apologize for for any inconvenience caused. We hereby retract this manuscript.”

The paper has been cited 25 times, according to Thomson Scientific’s Web of Knowledge.

Just as in last week’s PNAS retraction, not everyone signed. This time, the odd woman out was Anne von Euler, who was apparently at Sidec, the company whose product “found to be highly unreliable” when the researchers tried to repeat the experiments, when the paper was originally published. She is now listed as a member of the department of molecular biology and functional genomics at Stockholm University.

The study’s corresponding author, Paul W.H.I. Parren, is senior vice president and scientific director of Genmab –which last year put its plans to conduct late-stage trials of zalutumumab for squamous cell carcinoma of the head and neck on hold, after failing to find a partner to develop and sell the drug. CEO Jan van de Winkel told Genetic Engineering & Biotechnology News last June:

We have always been focused on ending our investment in zalutumumab so that we can concentrate our resources to progress other high-value programs in our pipeline. The product has shown good results in our clinical studies without any serious adverse effects, and I believe it has therapeutic potential. However, we were committed to taking a decision on the future of the product by the end of June and therefore we will now shelve Zalutumumab but make it available for partnership should the opportunity arise.

We haven’t been able to reach von Euler or Parren for comment, but we’ll update with anything we hear back.

Update, 11:30 a.m. Eastern, 3/22/12: Please see an update with comments from von Euler and Parren.

Hat tip: John Timmer

11 thoughts on “Another non-unanimous PNAS retraction, for potential anti-cancer drug, after company’s method proves unreliable”

  1. In the interwebs we’ll all live forever. Sidec AB is still listed as active on many business sites, although it went into liquidation in June 2010. But papers that have used their technology keep on popping up even afterwards – at least a dozen since 2010… How many more will be retracted and how many would just sit there forever, like something that nobody wants to touch with a 6-foot pole?

    You see, that’s what I hate about complicated instruments that always give you “data” no matter what – even when “data” is Nth derivative of a signal that nobody can see. I had a postdoc once who for a month studied kinetics of a random noise in a stop-flow machine. Very complicated reaction pathway it was, until she figured to add nucleotide that was needed to drive the reaction. Didn’t miss a bit, went right on fitting them curves again.

  2. I thought Sidec published a retraction of their method or at least announced that it was bogus- but I can’t seem to find it on a quick search. Basically the math was sleight of hand and they were modeling noise. There was nothing nefarious about it, it just wasn’t carefully validated enough. It will have done quite a bit of damage to those that have used it.

  3. If I’m reading the retraction notice correctly, “PT was found to be highly unreliable”–not the best business model for long term stability. It is no wonder that Sidec AB went in to liquidation. It may take several years for all the retractions to come through.
    I seem to recall many other “promising” methods that fell by the wayside, including thermography in the late seventies.
    “The mills of the gods grind slow but exceedingly fine” is how I think that one goes. Sad that individual careers don’t survive this sort of mistake.
    Is that like a hallucination when you see something in a random noise signal? Does obsession with technique and loss of overall perspective lead to this sort of misperception?

    1. Oh, no, nothing like hallucinations. Just inability to see physical reality that gives rise to numbers. I’d mark it to poor training and lacking the habit to think on the job. Give you a simpler example – a person is going to spec the protein prep using specific extinction coefficient. Having plugged it into Nanodrop, the person is calmly writing down the number 3.56 mg/ml – despite the fact that from the spectrum clearly visible on the screen is it bloody obvious that the prep is heavily contaminated by nucleic acids, and protein-specific extinction calculations thus rendered meaningless.

      1. That’s a simpler example?…a protein preparation contaminated with nucleic acids and your “simple laboratory instrument” still gives you a number you can write down. You’re overworked and underpaid.

        I was thinking more of the type of misperception that occurs when the evidence is vague, for example when Mars was observed through the telescope and thought to have “canali” (channels)–the eye translating a tiny, blurred image into a series of defined structures.

        That’s what the protein tomogram looked like. I think. I’m going to look into this.

  4. Actually, this is a major problem. Back in 2005, Sidec came out with Protein Tomography(TM) (basically a computer program which takes raw scanning electron microscopy data and turns it into a 3-D image of the structure of a protein) and they appear to have done well in speculative circles, associating with another mysterious company, FEI.
    However, i can’t find references to very much research done with it. One paper is about the structure of the receptor for levetiracetam(an anti-epileptic drug), a synaptic vesicle membrane protein. Likely it was used in the drug industry and the results never saw the light of day.
    I did find indications that if you want to do research with it, you can pay a nice fee to use their technique(If you fill out a form giving your institutional affiliation, they’ll send you a “White Paper” about it.) And they offered a “prize” of up to 500K Euros to researchers offering an experimental design, to “get acquainted with the technique.”
    This explains why the one author didn’t sign the retraction–she worked for Sidec (but no longer? Did she quit?)

    If I’m reading this paper right, the technique is no good, meaning Sidec is going to vanish. I can’t read it any other way. Amazing that it took seven years and significant expenditure of grant money by clients of Sidec to “validate” their proprietary technique. Weren’t they supposed to validate it themselves before marketing it as the greatest thing since sliced bread?

    This makes the monkey very puzzled indeed.

    1. Sidec AB is gone, and there is quite a few papers done using their technology – as I said about a dozen after they went under in 2010. Overall number of citations of just one of their method papers (cited in retraction note) is in the 60s.

  5. Monkey is, as usual, late. “I am Ozymandias, king of kings. Look upon my works, ye mighty, and despair.” Monkey puzzles over ruins. The method papers read like advertisements, not objective descriptions of methods. Someone has thrown away a lot of grant money to the benefit of these scientific confusion enhancers; “nothing nefarious about it”… but a lot of money changed hands over this “Nth derivative of a signal nobody can see.”
    More retractions to come, or monkey won’t be happy.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.