On 18 January 2006, the Rikshospitalet–Radiumhospitalet Medical Center and the
University of Oslo (UiO) jointly appointed a special commission to conduct an
independent investigation in accordance with detailed terms of reference.
The background for the investigation was that a researcher employed by these
institutions, Jon Sudbø, had admitted fabricating the raw data used for a scientific article
published in the renowned medical journal The Lancet in October 2005.
1.2 The investigation
Early in the investigation it became clear that the entire body of Sudbø's scientific work
from 1993-2006 (at least 38 publications) would have to be scrutinized, and that the coauthors
(60 altogether) would in reality also have to be subject to investigation. All the
authors received a letter requesting them to submit a voluntary written statement, which
they all did. Moreover, information was gathered from relevant institutions and other
relevant partners. Special mention should be made of the findings from the thorough
investigations made by the Cancer Registry of Norway. The Commission also met with
individuals and representatives of institutions, including Jon Sudbø. Furthermore, the
Commission has obtained documents and other information from several other sources.
Available data lists, etc., and published research results have been correlated and
compared. Accordingly, the Commission was generally able to judge whether, and the
extent to which, the underlying data on which the publications are based are genuine. As
its main principle, the Commission has found it appropriate to apply a standard of
proof based on a qualified preponderance of probability as a condition for accepting a
particular fact as grounds for the report.
Jon Sudbø began his PhD project in 1993 under the supervision of Albrecht Reith.
The PhD project consists of two separate parts. One part involves theoretical and
applied works on tissue architecture in cancerous tumors and normal tissue. The
Commission has not found indications of research flaws related to these works.
As reflected in his subsequent research, most of his PhD project involved
characterizing the early stages of oral cancer. The research question was whether and, if
so, to what extent, different types of classifications of white patches in the oral cavity
were indicative of a high risk for developing oral cancer. The doctoral dissertation and
related publications give an affirmative response to this question, asserting that a
classification based on DNA content can with great accuracy predict the subsequent
development of cancer.
First published in the highly respected New England Journal of Medicine in 2001,
this sensational finding was based on DNA analyses of 150 patients with leukoplakia (i.e.
'white patches' that may be early stages of oral cancer) in the oral cavity. In 2004, a
second article was published in the New England Journal of Medicine, based on further
investigations of the same 150 patients. Based on their own investigations and those
made by the Cancer Registry of Norway, the Commission's point of departure is that
there are serious problems associated with this crucial patient material. For instance, the
same patient appears several times. As far as the Commission can determine, the material
consists of 141 different patients at the most, since several patients are represented by
several tissue samples that collectively add up to 150. Further, the Commission has found
that 69 of the 141 patients included in the study should have been excluded because they
had been diagnosed with oral cancer before or at the same time as the leukoplakia was
diagnosed. For these patients, it was not possible to study the future development of
cancer, since they already had cancer. This error alone is so serious that the results and
the conclusions are invalid. The Commission has also uncovered several other
inconsistencies. For example, the age distribution in the data files is not consistent with
the underlying patient material. Further, the Commission has noted that the reported 150
DNA analyses are to some extent repetitions of data from a far smaller number of
patients. The reporting on how DNA analyses and the classification of leukoplakia were
conducted (by several observers) is also incorrect and misleading.
Consequently, the Commission has determined that the data underlying parts of
the PhD project, as well as several other publications, are not sufficiently consistent with
the actual facts the Commission has found it reasonable to take into account. The internal
affairs investigation conducted by the Cancer Registry of Norway has arrived at the same
The Commission is of the opinion that the errors and defects that have been
exposed are too numerous, too great and too obvious to be attributed to random errors,
incompetence or the like; and that the raw data therefore appear to have been fabricated,
manipulated and adapted to the desired findings.
The consequence of this is that the doctoral dissertation and three related original
articles must be retracted. In addition, subsequent publications must be retracted where
they are based on the same raw material, as most of them are. On the same grounds, the
Commission also questions one other original article. Further, the Commission has
questioned an original article published in the Journal of Clinical Oncology 2005, inter
alia in the light of circumstances partially acknowledged by Sudbø. The most recent
original article published in The Lancet in 2005 has been retracted, since it is, in its
entirety, based on fabricated raw data. Jon Sudbø has admitted this.
This means that the bulk of Jon Sudbø's scientific publications are invalid due to
the fabrication and manipulation of the underlying data material.