PickyBicky said:
Point conceded. I was not aware of those particular examples.
Not a dis at all. You're exactly right....these are only a few examples, limited to a very small number of histologies (the number of cases of melanoma each year are a log lower than lung ca). The burden of disease is tremendous and treatment options (this is specifically for widely diseminated disease...not local or regional) are dreadfully small. I agree with the above post: we should not let "chatter" that "the cure" has been found but miserly MDs and big pharma are holding back because of profit motives become widely disseminated. It's cruel to patients, and it's potentially detrimental to current research efforts and funding. If anything, the opposite is true: pharma and MDs claiming results of efficacy which are not supported by the data. Just like at the IMclone scandal, for example. Another example: forged data in preventative measures of oral cancer (see below)
From Science, Jan. 27th, 2006:
SCIENTIFIC MISCONDUCT:
Fraud Upends Oral Cancer Field, Casting Doubt on Prevention Trial
Jennifer Couzin and Michael Schirber
The world of oral cancer research is reeling after one of its stars, Norwegian oncologist Jon Sudbø, admitted this week through his attorney to falsifying data in three seminal papers published by top medical journals. A fourth paper is under suspicion after editors at the New England Journal of Medicine (NEJM) found that it contains a pair of duplicate images. For one of the papers, in The Lancet, Sudbø also appears to have claimed funding from a nonexistent grant.
The revelations have put on hold a multimillion-dollar oral cancer prevention trial, sponsored in part by the U.S. National Cancer Institute. The affair has also raised questions about whether researchers in multi-institutional collaborations should do more to double-check the validity of data collected by others. The fraud is all the more unsettling given the recent fabrications by South Korean researcher Woo Suk Hwang in stem cell science (Science, 13 January, p. 156).
Fraud exposed. Cancer researcher Jon Sudbø acknowledged faking data in three of these papers, and journal editors found a duplicated image in the fourth.
"Something like this, coming so hard on the stem cell revelation, is almost catastrophic," says Fadlo Khuri, an oncologist at Emory University in Atlanta, Georgia. Sudbø's results, he says, "are among the most important findings of the last decade [in] understanding the biology" of oral cancer.
The Norwegian Radium Hospital, where Sudbø is based, has launched an investigation led by Anders Ekbom of the Karolinska Institute in Stockholm. Sudbø's 38 published articles will be reviewed, as will the role of his co-authors, one of whom is his twin brother and another his wife. Results are expected in a couple of months. "We don't have any suspicions that the other authors knew," says Stein Vaaler, director of strategy at the hospital, which has already found that hundreds of patient records were fabricated in the Lancet paper.
Some papers in question identified those at greatest risk of oral cancer, a disease often preceded by noncancerous mouth lesions. Just 20% to 30% of individuals with lesions develop oral cancer, confounding prevention efforts.
The earliest paper to contain false data, according to Sudbø's attorney, Erling Lyngtveit, appeared in NEJM in April 2004. It reported that 26 of 27 individuals with aneuploid mouth lesions, so called because they contain abnormal numbers of chromosomes, developed aggressive oral cancer and were more likely to die of the disease than were those with other types of lesions. Lyngtveit confirmed that Sudbø did not have access to death information on which the study's conclusion was based. (Sudbø is currently on sick leave and has not spoken publicly.)
That 2004 study built on one that appeared 3 years earlier in NEJM that identified aneuploid mouth lesions as unusually hazardous. Eighty-four percent of study volunteers with the lesions developed oral cancer. On 20 January, NEJM released an "Expression of Concern" stating that one of the paper's images of a mouth lesion is a magnified version of another in the same article. The journal, says a spokesperson, is awaiting the results from the Radium Hospital's investigation before determining how to handle both studies.
Two other reports that Sudbø's attorney told Science contain fabrications were published in the 20 March 2005 issue of the Journal of Clinical Oncology and the 15 October 2005 issue of The Lancet. The first concluded that smokers with mouth lesions, if told they were at high risk of oral cancer, were likelier to quit than were those without detectable lesions. The second, in The Lancet, claimed to draw on archived health records to show that long-term use of anti-inflammatory drugs reduced the risk of oral cancer.
That study was the first to attract suspicion. Several weeks ago, Camilla Stoltenberg, director of epidemiology at the Norwegian Institute of Public Health, noticed that the Lancet study relied on a database not yet available to researchers, and she alerted the Radium Hospital on 11 January. An internal investigation by the hospital concluded that Sudbø "fabricated all the data in the article," which included names, genders, diagnoses, and other variables for 908 people. The paper also cites funding from a Norwegian Cancer Society grant even though the proposal was rejected, says society spokesperson Terje Mosnesset.
An immediate casualty of the fraud may be a 360-person trial of the anti-inflammatory Celebrex, along with another drug, in healthy people with aneuploid mouth lesions. The cancer prevention trial garnered roughly $9 million from the National Cancer Institute in Bethesda, Maryland, and was to be led by Sudbø and Scott Lippman of the M. D. Anderson Cancer Center in Houston, Texas, who was a co-author on the 2004 NEJM paper and the Lancet paper. "Everything has to be put on hold," says M. D. Anderson Vice President for Research Administration Leonard Zwelling.
The hospital, he adds, will consider new ways to handle large population studies in which its researchers analyze results but may not see the raw data. "Should we have an independent board" to examine those data, Zwelling wonders.
Meanwhile, oral cancer experts are grappling with the fabrications and whether the aneuploid work will stand. Notes Richard Jordan, an oral pathologist at the University of California, San Francisco, aneuploid lesions weren't "100% predictive, but [they] were the best that anyone heard of."
-PB
