|
|
Background Overexpression of cyclooxygenase 2 (COX-2) has been associated with colorectal adenomatous polyps and cancer, prompting researchers to propose its inhibition as a chemopreventive intervention.
Methods The Prevention of Colorectal Sporadic Adenomatous Polyps trial was a randomized, placebo-controlled, double-blind study of the COX-2 inhibitor celecoxib given daily in a single 400-mg dose. At 107 centers in 32 countries, we randomly assigned 1561 subjects who had had adenomas removed before enrollment to receive celecoxib (933 subjects) or placebo (628 subjects) daily, after stratification according to the use or nonuse of low-dose aspirin. The primary outcome was detection of adenomas at either year 1 or year 3 by colonoscopy and was compared among the groups with the use of the Mantel–Cox test.
Results Colonoscopies were performed at year 1 on 88.7 percent of the subjects who had undergone randomization and at year 3 on 79.2 percent. Of the 557 subjects in the placebo group and the 840 subjects in the celecoxib group who were included in the efficacy analysis, 264 and 270, respectively, were found to have at least one adenoma at year 1, at year 3, or both. The cumulative rate of adenomas detected through year 3 was 33.6 percent in the celecoxib group and 49.3 percent in the placebo group (relative risk, 0.64; 95 percent confidence interval, 0.56 to 0.75; P<0.001). The cumulative rate of advanced adenomas detected through year 3 was 5.3 percent in the celecoxib group and 10.4 percent in the placebo group (relative risk, 0.49; 95 percent confidence interval, 0.33 to 0.73; P<0.001). Adjudicated serious cardiovascular events occurred in 2.5 percent of subjects in the celecoxib group and 1.9 percent of those in the placebo group (relative risk, 1.30; 95 percent confidence interval, 0.65 to 2.62).
Conclusions The use of 400 mg of celecoxib once daily significantly reduced the occurrence of colorectal adenomas within three years after polypectomy. (ClinicalTrials.gov number, NCT00141193
[ClinicalTrials.gov]
.)
The regression and prevention of adenomas have become primary aims of investigators, who have used nonsteroidal antiinflammatory drugs (NSAIDs), including aspirin, to reduce the incidence and recurrence of adenomas.5,6,7 Since being identified as distinct from the ubiquitous cyclooxygenase (COX) 1 isoform,8 COX-2 has been shown to be overexpressed in colorectal cancers and adenomas in humans9,10 and was demonstrated in knockout mice to be a controlling factor in the formation of adenomas.11 Researchers have therefore sought to assess the effects of such selective COX-2 inhibitors as celecoxib on the formation of adenomas in hopes that these agents could overcome the gastrointestinal adverse events of nonselective NSAIDs that placed limits on effective therapy.
Three international, multicenter studies of the use of COX-2 inhibitors to prevent sporadic adenomas, including the present study, were launched in 1999 and 2000.12,13 The randomized Adenoma Prevention with Celecoxib (APC) trial included 2035 patients in a study of two twice-daily doses of celecoxib, and the randomized Adenomatous Polyp Prevention on Vioxx (APPROVe) trial included 2586 patients in a study of rofecoxib. We undertook a randomized, placebo-controlled, double-blind study of 400 mg of celecoxib once daily in subjects who had undergone colonoscopy and polypectomy at baseline.
Methods
Study Design
We performed this randomized, placebo-controlled, double-blind study of three years of treatment with a once-daily dose of 400 mg of celecoxib (Celebrex, Pfizer). With guidance from the steering committee, the investigators developed the study protocol, which was approved by central and local ethics committees and overseen by an independent data and safety monitoring board. Investigators and an independent organization (Covance, Princeton, N.J.) collected data, Pfizer maintained the database, and a team consisting of both Pfizer and independent statisticians conducted the analyses. The cardiovascular safety committee independently reviewed and adjudicated all serious adverse events and performed an independent analysis. The authors vouch for the completeness and veracity of the data and data analyses.
Within three months before enrollment, all subjects had undergone a colonoscopy, documented to the cecum, in which all polyps that were found were removed. Enrolled patients, all of whom had had adenomas detected and removed, received placebo during a single-blind, one-month run-in period. When the period ended, eligible subjects were stratified according to country and to the use or nonuse of low-dose aspirin, and then they were randomly assigned in a 3:2 ratio by an interactive voice-response system to receive 400 mg of celecoxib or placebo once daily. Colonoscopies were performed 12 months (year 1) and 36 months (year 3) after randomization. The investigators completed enrollment in one year (from March 2001 to March 2002) and performed the last colonoscopy in May 2005. Meetings of investigators before enrollment and site inspections for quality control during the trial were designed to promote uniform practice and quality assurance.
Safety evaluations during the study included physical examinations and clinical laboratory tests during the run-in period and at one and three years after enrollment. After they underwent randomization, subjects made return visits at 3, 6, 12, 18, 24, 30, 36, and 37 months for assessment of pill counts, as well as for reporting of concomitant medications and adverse events (which were also gauged during bimonthly telephone assessments). Safety data were reviewed every six months by the independent data and safety monitoring board, for whom an independent statistician issued unblinded reports for review. After reports in September 2004 of increased cardiovascular risk with the use of rofecoxib, the steering committee for the Prevention of Colorectal Sporadic Adenomatous Polyps (PreSAP) trial requested that an independent cardiovascular safety committee perform an analysis of cardiovascular events. Results from this analysis in December 2004 showed no significant increase in cardiovascular risk14; however, an increase detected in a similar, concurrent analysis of the APC trial by the same cardiovascular safety committee led to suspension of the administration of the study drug in the PreSAP trial on December 17, 2004.13
Study Population
We enrolled 1738 subjects at 107 centers in 32 countries on six continents. Before study participation, all subjects provided written informed consent. Eligible subjects were 30 years of age or older and had undergone colonoscopy within three months before enrollment that showed either 1 adenoma that was at least 6 mm or 2 to 10 adenomas of any size. Inclusion criteria required photographic documentation that the cecum was reached, the colon was clean, and the adenomas that were removed were of the size charted. Subjects had to show 80 percent adherence to the 1-month run-in placebo medication and had to be willing to abstain from long-term use (defined as use for more than 21 days throughout the year) of all NSAIDs, including COX-2 inhibitors other than the blinded study drug but excluding cardioprotective doses of oral aspirin (
162.5 mg daily or 325 mg every other day). Subjects were ineligible if they had a history of familial adenomatous polyposis, hereditary nonpolyposis colorectal cancer, inflammatory bowel disease, or invasive cancer within five years before enrollment. Also ineligible were subjects with renal, hepatic, or bleeding disorders; subjects hypersensitive to COX-2 inhibitors, NSAIDs, salicylates, or sulfonamides; subjects who had undergone large-bowel resection (excluding appendectomy); and subjects who had used NSAIDs, excluding cardioprotective doses of aspirin, three or more times a week within the 60 days before randomization.
Study End Points
The primary efficacy end point was detection by colonoscopy of at least one colorectal adenoma at year 1, at year 3, or both. All polyps removed during colonoscopies were submitted to local and central pathology laboratories for blinded review. An adjudicator resolved discrepancies in a blinded fashion.
Secondary efficacy end points one year and three years after enrollment included the number of colorectal adenomas, the size of the largest ones measured, and the total adenoma burden (the sum of the diameters of all the adenomas) over the three-year period. In addition, for consistency with the APC trial, end points included detection of advanced adenoma with any of the following characteristics: size of at least 1.0 cm according to measurements obtained endoscopically, villous or tubulovillous histologic features, high-grade dysplasia, intramucosal carcinoma, or invasive cancer.
Safety end points were based on investigator-reported adverse events, serious adverse events, laboratory measurements, and physical examinations. Adverse events were coded according to criteria from the Medical Dictionary for Regulatory Activities (MedDRA), version 8.1. Adjudicated cardiovascular end points were categorized as previously described.15
Statistical Analysis
The trial required 1500 subjects (900 in the celecoxib group and 600 in the placebo group) to ensure 90 percent power to detect at least a 35 percent relative reduction in the formation of adenomas in the celecoxib group as compared with the placebo group, assuming that 36 percent of subjects in the placebo group who were taking low-dose aspirin and 42 percent of subjects in the placebo group who were not taking low-dose aspirin would have newly detected adenomas at year 1, at year 3, or both. It was anticipated that 35 percent of the subjects would be taking low-dose aspirin. The sample size further allowed for subjects to withdraw from the study and assumed a two-sided type I statistical error of 0.05.
Efficacy analyses included all subjects who had undergone randomization. The effect of celecoxib in reducing the proportion of subjects in whom adenomas were detected at year 1, at year 3, or both was estimated with the use of the Mantel–Cox test, a life-table extension of the Mantel–Haenszel test,16,17,18 with stratification for the use or nonuse of aspirin. The method for calculating relative risk was described by Kleinbaum et al.19 Subjects considered at risk at year 1 included those who underwent colonoscopy at year 1, as well as those who underwent colonoscopy only at year 3 and who were classified as having had no adenomas detected at year 1. Subjects considered at risk at year 3 were those who had undergone colonoscopy at year 3 and in whom no adenomas were identified before year 3. The occurrence of advanced adenomas was analyzed according to the same method. Separate analyses of the occurrence of adenomas at year 1 and at year 3 were performed with the use of a Mantel–Haenszel test, stratified according to the use or nonuse of aspirin. The number of adenomas, the size of the largest adenoma, and the total adenoma burden detected during the three years were analyzed with van Elteren's test,20 an extension of the Wilcoxon rank-sum test, stratified according to the use or nonuse of aspirin. There were no interim analyses of efficacy. All reported P values are two sided.
Investigator-reported adverse events were summarized for all treated subjects, regardless of adherence to study protocol. Additional summaries were provided for prespecified categories of adverse events. Serious adverse cardiovascular events were adjudicated and categorized as previously described.13
Results
Baseline Characteristics
Baseline demographic characteristics, colonoscopic findings, cardiovascular risk factors, and body-mass index were similar in the groups (Table 1). The proportion of subjects with a history of cardiovascular events (atherosclerotic or cerebrovascular disease) was 11.1 percent in the placebo group and 13.7 percent in the celecoxib group.
|
Of 1738 subjects screened for eligibility, 1561 underwent randomization, 628 to the placebo group and 933 to the celecoxib group; 107 (17.0 percent) in the placebo group and 160 (17.1 percent) in the celecoxib group took low-dose aspirin. In the placebo group, 554 subjects (88.2 percent) completed the colonoscopy at year 1 and 498 (79.3 percent) completed the colonoscopy at year 3; in the celecoxib group, 831 subjects (89.1 percent) completed the year 1 colonoscopy and 739 (79.2 percent) completed the year 3 colonoscopy (Figure 1). In all, 495 subjects in the placebo group (78.8 percent) and 727 in the celecoxib group (77.9 percent) received the study medication for 80 percent or more of the study period. Forty-two percent of subjects who underwent colonoscopy at year 3 did so after administration of the drug was suspended on December 17, 2004; 93 percent of those colonoscopies were performed within three months after suspension. Total treatment exposure was 2331 patient-years (mean, 2.49 years) for celecoxib and 1570 patient-years (mean, 2.50 years) for placebo, reflecting the 3:2 ratio of randomization.
|
Of the 557 subjects in the placebo group and the 840 subjects in the celecoxib group who were included in the efficacy analysis, 264 and 270, respectively, were found to have at least one adenoma at year 1, at year 3, or both (Table 2). Overall, the three-year estimated cumulative rate of adenomas, calculated on the basis of the Mantel–Cox method, was 33.6 percent for subjects receiving celecoxib and 49.3 percent for subjects receiving placebo (relative risk, 0.64; 95 percent confidence interval, 0.56 to 0.75; P<0.001) (Table 2). The results were similar for subjects taking low-dose aspirin at baseline (relative risk, 0.61; 95 percent confidence interval, 0.43 to 0.88; P=0.007), as well as for those not taking low-dose aspirin at baseline (relative risk, 0.65; 95 percent confidence interval, 0.55 to 0.76; P<0.001). Fifty-six subjects in the placebo group and 42 in the celecoxib group were found to have advanced adenomas as follows: 40 in the placebo group and 25 in the celecoxib group at year 1, and 16 in the placebo group and 17 in the celecoxib group at year 3. The estimated cumulative proportion of subjects with advanced adenomas at year 3 was 5.3 percent in the celecoxib group and 10.4 percent in the placebo group (relative risk, 0.49; 95 percent confidence interval, 0.33 to 0.73; P<0.001). Among subjects who had any new adenoma detected, the mean size of the largest adenoma and the mean adenoma burden were significantly lower in the celecoxib group than in the placebo group; the difference in the mean number of adenomas was not significant (Table 2).
|
In the placebo group, the only adverse events related to colonoscopy were three cases of bleeding and one case of hypotension. All four occurred at baseline. Adverse events related to colonoscopy reported in the celecoxib group were 10 cases of bleeding, 2 perforations, 3 cardiovascular complications (tachycardia and bradycardia), and 1 vasovagal episode. Eleven occurred at baseline, four at year 1, and one at year 3.
Investigator-reported serious adverse events occurred in less than 20 percent of either group (Table 3). Thirty-five subjects died of cardiovascular causes or had myocardial infarction, stroke, or congestive heart failure adjudicated by the cardiovascular safety committee; these consisted of 2.5 percent of the celecoxib group (23 of 933) and 1.9 percent of the placebo group (12 of 628) (relative risk, 1.30; 95 percent confidence interval, 0.65 to 2.62), with estimated rates of 9.4 and 7.2 events per 1000 patient-years for celecoxib and placebo, respectively. Of these 35 subjects, 12 came from a subgroup of 198 with a history of cardiovascular or cerebrovascular events (relative risk, 1.55; 95 percent confidence interval, 0.42 to 5.76), and 23 from a subgroup of 1363 without a medical history of cardiovascular or cerebrovascular events (relative risk, 1.14; 95 percent confidence interval, 0.49 to 2.65; P for interaction=0.59), indicating no significant difference in relative risk between subjects with and those without a history of previous cardiovascular events.
|
Investigator reports of gastrointestinal ulceration or hemorrhage were similar (celecoxib group, 12.1 percent; placebo group, 10.4 percent), and no significant difference was seen in hemoglobin and hematocrit laboratory values over the three-year period (mean hematocrit values in the placebo and the celecoxib groups were 43.9 and 43.8, respectively, at year 1, and 44.2 and 44.1, respectively, at year 3) (see Supplementary Appendix 1, available with the full text of this article at www.nejm.org).
A summary of the number of patients with serious adverse events categorized by the MedDRA systems organ class is provided in Supplementary Appendix 2. Those systems organ classes with the most patients were gastrointestinal disorders; cardiac disorders; benign, malignant, and unspecified neoplasms; surgical and medical procedures; and nervous system disorders. Within the systems organ classes, there were no significant differences in incidence between the celecoxib and the placebo groups.
Discussion
In this randomized, placebo-controlled, double-blind clinical trial, 400 mg of celecoxib once daily was associated with a relative risk of 0.64 for adenomas detected during a three-year period. A reduced risk was apparent at the first follow-up colonoscopy (year 1) and persisted at the second (year 3). Likewise, the cumulative rate of detection of advanced adenomas was lower in the celecoxib group, which had a relative risk of 0.49. The size of the largest adenoma and the adenoma burden were significantly lower in the celecoxib group, but the mean number of adenomas among those with any adenoma was not significantly different between the groups. The relative risk of adenomas during the three years among subjects who took low-dose aspirin was 0.61, which was similar to the relative risk among all subjects.
Celecoxib may be effective against existing adenomas perhaps missed on an index colonoscopy and against the formation of new adenomas, as evidenced by reductions in the newly diagnosed adenomas at year 1 and year 3. This effect is consistent with experimental evidence in animals and humans and with epidemiologic data.
The study did not directly address whether celecoxib affects the risk of a first adenoma or progression to invasive cancer. Colorectal cancer was diagnosed in six subjects in the celecoxib group (0.6 percent) and one subject in the placebo group (0.2 percent) (Table 3). Because five of these seven cases were diagnosed at the year 1 colonoscopy (stage 0 in two subjects and stage 1 in three), these five cases might represent either incomplete excision or disease that was undetected at the index colonoscopy. The remaining two cases of colorectal cancer were diagnosed at year 3 (stage 2A and stage 4). Three of the seven cases were in subjects whose mothers had colorectal cancer.
Delineation of clinical benefit requires further research. As the type of adenoma that responds to celecoxib becomes characterized, treatment with celecoxib may be found to reduce the frequency of colonoscopy necessary in certain subjects prone to adenomas. Also, determining the persistence of the effect of celecoxib on the prevention of adenomas and advanced adenomas is expected to be an especially important aspect of the follow-up of these subjects. Celecoxib may also have effects on other enzymes, such as 15-lipoxygenase-1,21 which could mediate its biologic effects. Further assessment of the relative risks and benefits of celecoxib in patients with adenomas is expected not only to promote greater safety for patients but also to encourage greater patient–physician dialogue, thus improving overall patient care.
Neither gastrointestinal adverse events nor changes in the laboratory values of hemoglobin were significantly different between groups. Cardiovascular safety results, however, need to be considered with those previously reported for the APC trial.13 A significant increase in adjudicated serious cardiovascular events with the use of celecoxib in the APC trial (an increase in risk by a factor of two or three for a composite end point of myocardial infarction, stroke, congestive heart failure, or cardiovascular-related death)13 prompted suspension of the administration of celecoxib in both the APC trial and our study. In our study, the relative risk of such events with the use of celecoxib as compared with placebo was 1.30 (95 percent confidence interval, 0.65 to 2.62).
The relative risk of renal or hypertensive adverse events associated with celecoxib was 1.35 (95 percent confidence interval, 1.08 to 1.69). Nevertheless, once-daily celecoxib was not associated with an increase in mean blood pressure, a finding in contrast to the results in the APC trial reported by Bertagnolli et al. elsewhere in this issue of the Journal.22
In our study, subjects came from six continents and the study protocol required photographic documentation of polyp size and cecal intubation to ensure conformity across clinical sites. The simplicity of once-daily dosing may have promoted enrollment and sustained adherence to the regimen of study medication. Subjects required documentation of adenomas at baseline and were at high risk for further adenomatous polyps. Because this study included a diverse population, our findings probably have broad applicability.
In conclusion, we found that celecoxib significantly reduced the risk of colorectal adenomas. These findings, which are consistent with other evidence linking COX-2 with colorectal neoplasia, indicate that selective inhibition of COX-2 may reduce colorectal tumorigenesis. Deriving specific recommendations for the clinical use of this treatment approach requires further investigation.
Supported by Pfizer, which also provided celecoxib and matching placebo.
Dr. Arber reports having served as a consultant to, owning stock in, and having received lecture fees and grant support from Pfizer. Dr. Levin reports having served as a consultant to Pfizer and having received consulting fees from Enterix and grant support from Pfizer for this study. Drs. Lechuga, Gerletti, Macdonald, Eagle, Bhadra, and Rosenstein and Ms. Tang report being employees of Pfizer. No other potential conflict of interest relevant to this article was reported.
We are indebted to Gary Gordon, Jeffrey Sherman, Ivan Horak, Daniel R. Vlock, and Aby Buchbinder for their dedication and support during the early phases of the trial; to John Sanocki and Wei Shen for programming support; to Angela Alvarez-George and Deidre Sullivan for data management; and to Beth W. Allen, Michele Norton, and Briton Shell for assistance with manuscript preparation.
Source Information
From the Integrated Cancer Prevention Center, Tel Aviv Sourasky Medical Center, and Sackler School of Medicine, Tel Aviv University, Tel Aviv, Israel (N.A.); Pfizer, New York (C.J.E., M.J.L., P.G., J.T., R.B.R., K.M., P.B.); the Institute for Clinical and Experimental Medicine, Prague (J.S.); Faculty Hospital Brno Bohunice, Brno (P.D.), Faculty Hospital Královské Vinohrady, Prague (J.H.); and Central Military Hospital, Prague (M.Z.) — all in the Czech Republic; Petz Aladár County Hospital, Gyõr, Hungary (I.R.); Statistics Collaborative, Washington, D.C. (R.F., J.W.); Memorial Sloan-Kettering Cancer Center, New York (A.G.Z.); Brigham and Women's Hospital, Boston (S.D.S.); and the Division of Cancer Prevention and Population Sciences, the University of Texas M.D. Anderson Cancer Center, Houston (B.L.).
Address reprint requests to Dr. Levin at the Division of Cancer Prevention and Population Sciences, the University of Texas M.D. Anderson Cancer Center, 1155 Pressler St., Unit 1370, Houston, TX 77030-4009.
References
The following persons participated in the PreSAP trial: Steering Committee — N. Arber, W. Atkin, C.J. Eagle, A. Dannenberg, R. DuBois, B. Levin, R. Stockbrugger, B.C.Y. Wong, A.G. Zauber; Statistical Team — A.G. Zauber, R. Rosenstein, J. Tang, P. Bhadra, J. Wittes, R. Fowler; Medical Monitors — M.J. Lechuga, P. Gerletti; Central Pathology Review — Diagnostic Cytology Laboratories (Indianapolis); Third-Party Pathology Reviewer — K. Geisinger; Project Director — C.J. Eagle; Data and Safety Monitoring Board — S. Winawer (chair), J. Brinker, G. Elfring (independent statistician), J.P. Faivre, J. Lee, A.I. Neugut; Co-Lead Principal Investigators: N. Arber (Tel Aviv Sourasky Medical Center and Sackler School of Medicine, Tel Aviv University, Tel Aviv, Israel), B. Levin (the University of Texas M.D. Anderson Cancer Center, Houston); Principal Investigators — Australia: J. Croese (Mater Misericordiae Hospital, Pimlico, Queensland), V.D. Watters (Geelong Hospital Ryrie, Geelong); Belgium: M. De Vos (Ghent University Hospital, Ghent), J. Deviere (Erasme University Hospital, Brussels), D. Urbain (Vrije Universiteit Brussel, Brussels), H. Piessevaux (Université Catholique de Louvain, Brussels); Brazil: F. Oliveira Ferreira (Hospital do Câncer–A.C. Camargo, São Paulo), I. Maguilnik (Hospital de Clinicas de Pôrto Alegre, Pôrto Alegre); Canada: F. Bursey (Health Sciences Center, Saint John's, Newf.), R. Enns (St. Paul's Hospital, Vancouver, B.C.), R.N. Fedorak (University of Alberta, Edmonton, Alta.), J. Howard (London Health Sciences Centre, London, Ont.), H. Pluta (Gastroenterology and Hepatology Clinic, Abbotsford, B.C.), J. Simon (Hotel-dieu Hospital, Kingston, Ont.), A. Weiss (Vancouver Hospital and Health Sciences Center, Vancouver, B.C.); Chile: F. Lopez (Pontificia Universidad Catolica de Chile, Santiago), J. Valenzuela (Clinica Las Condes, Santiago); China: L. Cheng (General Hospital of the Chinese People's Liberation Army, Beijing); J. Qian (Peking Union Medical College Hospital, Beijing); Czech Republic: P. Dite (Faculty Hospital Brno Bohunice, Brno), J. Dosedel (Hospital of Merciful Sisters St. Charles Borromeo, Prague), J. Hajer (Faculty Hospital Královské Vinohrady, Prague), P. Hulek (Faculty Hospital Hradec Králové, Hradec Králové), J. Janku (Hospital Liberec, Liberec), J. Spicak (Institute for Clinical and Experimental Medicine, Prague), M. Zavoral (Central Military Hospital, Prague); Denmark: S. Lauerberg (Aarhus Amtssygehus, Aarhus); Finland: S. Niemelä (Oulu University Hospital, Oulu); France: R. Benamouzig (Hôpital Avicenne, Bobigny), G. Lledo (Cabinet Médical Lledo, Lyon), S. Chaussade (Groupe Hospitalier Cochin, Paris), J. Sahel (Hôpital Ste. Marguerite, Marseille); Germany: B. Birkner (Munich), D. Mueller, S. Boehm (Philipps-Universität Marburg/Lahn, Marburg), H.G. Dammann (Klinische Forschung Hamburg, Hamburg), A. Dettmer (Munich), R.R. Fink (Freising), H.D. Janisch (Erlangen), C. Klein (Kuenzing), J.F. Riemann (Klinikum der Stadt Ludwigshafen, Ludwigshafen), A. Roempp, C. Weber (Universitätsklinikum Ulm, Ulm), B. Wiedenmann (Universitätsklinikum Charité, Berlin); Hong Kong: F. Chan (Chinese University of Hong Kong, Shatin), B.C.Y. Wong (University of Hong Kong, Hong Kong); Hungary: I. Rácz (Petz Aladár County Hospital, Gyõr), L. Simon (Tolna County Hospital, Szekszárd); Ireland: P. Macmathuna (Mater Misericordiae Hospital, Dublin), F.E. Murray (Beaumont Hospital, Dublin), E. Quigley (Cork University Hospital, Wilton); Israel: N. Arber (Tel Aviv Sourasky Medical Center and Sackler School of Medicine, Tel Aviv University, Tel Aviv), S. Bar-Meir (Sheba Medical Center, Tel-Hashomer), A. Eliakim (Rambam Medical Center, Haifa), A. Fich (Soroka University Medical Center, Beersheba), Z. Fireman (Hillel Yaffe Medical Center, Hadera), E. Goldin (Hadassah Medical Center, Jerusalem), Y. Niv (Rabin Medical Center, Petah-Tikva), H. Shirin (Wolfson Medical Center, Holon); Italy: A. Andriulli (Casa Sollievo della Sofferenza, San Giovanni Rotondo, Foggia), V. Casale, M. Crespi (Polo Oncologico Regina Elena, Rome), R. Cestari (Ospedali Civili, Brescia), C. Crosta (Istituto Europeo di Oncologia, Milan), G. Frosini (Policlinico Le Scotte, Siena), S. Morini (Ospedale Nuovo Regina Margherita, Rome); the Netherlands: R.W. Stockbrugger (Academisch Ziekenhuis Maastricht, Maastricht); Norway: G. Huppertz-Hauss, J. Sauar (Medical Department Sykehuset Telemark, Skien); Peru: O. Frisancho (Hospital Nacional Edgardo Rebagliati Martins, Lima), J. Watanabe (Policlinico Peruano Japones, Lima); Poland: E. Butruk (Medical Centre for Postgraduate Education, Institute of Oncology, Warsaw); Portugal: C. Nobre Leitão (Instituto Português de Oncologia, Lisbon); Russian Federation: G.P. Arutyunov (Hospital #4, Moscow), O.N. Minushkin (Hospital #51, Moscow), O. Naumov (City Clinical Hospital #24, Moscow), B.K. Poddubny (Blokhina Cancer Research Center, Moscow), S.G. Shapovaliants (Hospital #31, Moscow), V.V. Veselov (State Scientific Center for Coloproctology, Moscow); Singapore: K.G. Yeoh (National University Hospital, Singapore), K.M. Fock (Changi General Hospital, Singapore); Slovak Republic: I. 
uri
(University Hospital, Bratislava), A. Vavrecka (St. Cyril's and Method's Hospital, Bratislava); South Africa: R. Jobson (Sandton Clinic, Johannesburg), B. Shmeizer (Sunninghill Hospital, Johannesburg), H.R. Schneider (Milpark Hospital, Johannesburg), J.P. Wright (Kingsbury Hospital, Cape Town); Spain: J.I. Arenas (Hospital Nuestra Señora de Aránzazu, San Sebastián), A. Castells (Hospital Clínic de Barcelona, Barcelona), J. Herrerias (Hospital Universitario Virgen Macarena, Seville), A. Obrador (Hospital Son Dureta, Palma de Mallorca), L. Rodrigo (Hospital Central de Asturias, Oviedo); Sweden: B. Kollberg (Mag-Tarm-centrum, Stockholm), L. Pahlman (Akademiska Sjukhuset, Uppsala); Switzerland: A. Hadengue (Hôpitaux Universitaires de Genève, Geneva); Taiwan: T.-M. King (Veterans General Hospital, Kaohsiung); United Kingdom: W. Atkin (Coordinator) (Imperial College, London), K. Vellacott (Royal Gwent Hospital, Gwent), B. Saunders (St. Mark's Hospital, Harrow, Middlesex), N. Mortensen (John Radcliffe Hospital, Oxford); United States: M.T. Bennett (Medical Associates Research Group, San Diego, Calif.), S.A. Cohen (Thomas Jefferson University Hospital, Philadelphia), D.Y. Graham (Veterans Affairs Medical Center, Houston), D.M. Kruss (Veterans Affairs Medical Center, North Chicago, Ill.), S.J. Meltzer (University of Maryland School of Medicine–Baltimore, Baltimore), N.N. Ravendhran (Digestive Disease Associates, Baltimore), M.L. Schubert (McGuire Veterans Affairs Medical Center, Richmond, Va.), D.S. Weinberg (Fox Chase Cancer Center, Philadelphia), K.L. Woods (Medical Center Endoscopy, Houston); Uruguay: H. Cohen (Clinicas Hospital, Montevideo).
| |||||||||||||||||||||||||||||||||||||||||||||||||
Related Letters:
Risks and Benefits of Celecoxib to Prevent Colorectal Adenomas
Andersohn F., Suissa S., Garbe E., Solomon S., Wittes J., Arber N., Bertagnolli M., Hawk E., Levin B., the APC and PreSAP Trial Investigators , Psaty B. M., Potter J. D.
Extract |
Full Text |
PDF
N Engl J Med 2006;
355:2371-2373, Nov 30, 2006.
Correspondence
This article has been cited by other articles:
HOME | SUBSCRIBE | SEARCH | CURRENT ISSUE | PAST ISSUES | COLLECTIONS | PRIVACY | TERMS OF USE | HELP | beta.nejm.org Comments and questions? Please contact us. The New England Journal of Medicine is owned, published, and copyrighted © 2009 Massachusetts Medical Society. All rights reserved. |
Previous
