The first part was carried out partly in cooperation with a post.doc. from France. This part comprises three scientific publications from 2000 which can be linked to method development and image analyses of tissue architecture based on raw material originating from Germany. In the Commission’s opinion, these studies are of a less sensational nature from a scientific point of view than the main part of the PhD project. However, as accounted for in section 4.2.3, Sudbø disagrees entirely in these works having a less essential role, even if
there has been less attention surrounding these studies than the main part of the PhD project. Sudbø underlines that these studies in terms of time and work represented the main part of his PhD work, and that they are published in recognized journals. The Commission has evaluated the articles and has not found indications of errors or deficiencies related to them.
The main part of the PhD project comprises three articles published in New England Journal of Medicine from 2001 (NEJM 2001), Journal of Pathology from 2001 (J Pathol 2001) and Journal of Oral Oncology 2001 (Oncology 2001):
in patients with oral leukoplakia. N Engl J Med. 2001 Apr 26;344: 1270-8.)
Sudbø J, Bryne M, Johannessen AC, Kildal W, Danielsen HE, Reith A. Comparison of histological grading
and large-scale genomic status (DNA ploidy) as prognostic tools in oral dysplasia. J Pathol. 2001
Sudbø J, Ried T, Bryne M, Kildal W, Danielsen H, Reith A.
Abnormal DNA content predicts the occurrence of carcinomas in non-dysplastic oral white patches. Oral
Oncol. 2001 Oct;37:558-65.
The articles are based on the same patient material and must be seen in conjunction. The patient material was probably obtained by Sudbø and Reith in 1995-1996 and linked with Cancer Registry data for supplementary information in 1996. The data material and samples were thereupon probably analyzed in 1999. The publication and the dissertation were published in 2001.
This patient material and the findings presented in the publications mentioned are at the very center of Sudbø’s subsequent scientific career and the series of subsequent publications.
The six articles were then collected in the PhD dissertation itself, which also contains an independent compilation and explanation of the PhD project. Thus it is the three latter articles together with the dissertation itself and the patient material these are based on, which have been the subject of thorough investigations that will be described below. The starting point is the dissertation itself.
The patient material – an overview
It appears from figure 5 at page 40 in the PhD dissertation, included as figure 1 in this report, that Sudbø started using human biological material (paraffin blocks with biopsy samples, tissue specimens, etc.) and person identifiable data from 263 patients (cases). In essential, the patient material, according to New England Journal of Medicine 2001 and J Pathol 2001 as well as statements to the Commission, was obtained in the following way:
Patient material from the Department of Pathology (“Gade’s Institute”), the Faculty of Odontology, the University of Bergen (hereinafter: Gade)
Patient data from Gade and the Odontology were thereafter sent collectively to the Cancer Registry for linkage, i.e. a supplement of further data.
The Commission, on its part, has collected extensive documentation in the form of original and processed data files from various persons and instances, as well as statements before the Commission. The Commission has also made its own comparison of different data files in order to trace patients, tissue samples, check mutual linkage, etc., as part of its own internal control. The Cancer Registry has made similar investigations. The Commission was given access to the Cancer Registry’s very thorough investigations. By this, the Commission has clarified how the patient material for the PhD work was obtained, and how it originally looked, including which patients were included in the study and their disease history.
The Commission then compared a series of figures and facts presented in the publications to figures and facts in the Commission’s possession.
In the following, the Commission will point out along the way the errors, deficiencies and discrepancies discovered in the three mentioned articles and the dissertation itself, including the raw material on which these publications are stated to be based upon.
Figure 1 is taken from Jon Sudbø’s PhD dissertation and shows, among other things, the inclusion process. An almost corresponding figure is included in New England Journal of Medicine 2001.
Representation of the raw material in the publications
The total number
The dissertation states that originally one had access to material from 263 persons collected during the period 1976-1995. According to the articles and the dissertation, this material originated exclusively from archives at the Odontology and Gade.
This means that Sudbø originally had access to human biological material and personal data from altogether 263 different persons. The number of samples and tissue blocks and the like was much higher, however, because normally one has several samples and blocks from the same patient. Oncology 2001 refers to all 263 patients because it includes both leukoplakias (white patches in the oral cavity) and erythroplakias (red patches in the oral cavity). In New England Journal of Medicine 2001, this starting point is reduced to 242 (263-21) because patients with erythroplakias were excluded initially. In J Pathol 2001 the starting point is 196 (263-(21+46)), although this number is 217 in the dissertation. According to Sudbø, however, it was only for 196 persons that ploidy classification existed and which for that reason were treated in J Pathol 2001. Below, the Commission will try to identify when and from where these patients come and the qualities they possess.