1946 · June 29, 2000 The New England Journal of Medicine THE EFFECT OF CELECOXIB, A CYCLOOXYGENASE-2 INHIBITOR, IN FAMILIAL ADENOMATOUS POLYPOSIS G IDEON S TEINBACH , M.D., P H .D., P ATRICK M. L YNCH , M.D., J.D., R OBIN K.S. P HILLIPS , M.B., B.S., M ARINA H. W ALLACE , M.B., B.S., E RNEST H AWK , M.D., M.P.H., G ARY B. G ORDON , M.D., P H .D., N AOKI W AKABAYASHI , M.D., P H .D., B RIAN S AUNDERS , M.D., Y U S HEN , P H .D., T AKASHI F UJIMURA , M.D., L I -K UO S U , P H .D., AND B ERNARD L EVIN , M.D. A BSTRACT Background Patients with familial adenomatouspolyposis have a nearly 100 percent risk of colorectalcancer. In this disease, the chemopreventive effectsof nonsteroidal antiinflammatory drugs may be re-lated to their inhibition of cyclooxygenase-2. Methods We studied the effect of celecoxib, a se-lective cyclooxygenase-2 inhibitor, on colorectal pol-yps in patients with familial adenomatous polyposis.In a double-blind, placebo-controlled study, we ran-domly assigned 77 patients to treatment with cele-coxib (100 or 400 mg twice daily) or placebo for sixmonths. Patients underwent endoscopy at the begin-ning and end of the study. We determined the num-ber and size of polyps from photographs and video-tapes; the response to treatment was expressed asthe mean percent change from base line. Results At base line, the mean (±SD) number ofpolyps in focal areas where polyps were counted was15.5±13.4 in the 15 patients assigned to placebo,11.5±8.5 in the 32 patients assigned to 100 mg of cele-coxib twice a day, and 12.3±8.2 in the 30 patients as-signed to 400 mg of celecoxib twice a day (P=0.66for the comparison among groups). After six months,the patients receiving 400 mg of celecoxib twice a dayhad a 28.0 percent reduction in the mean number ofcolorectal polyps (P=0.003 for the comparison withplacebo) and a 30.7 percent reduction in the polypburden (the sum of polyp diameters) (P=0.001), ascompared with reductions of 4.5 and 4.9 percent, re-spectively, in the placebo group. The improvementin the extent of colorectal polyposis in the group re-ceiving 400 mg twice a day was confirmed by a pan-el of endoscopists who reviewed the videotapes. Thereductions in the group receiving 100 mg of celecoxibtwice a day were 11.9 percent (P=0.33 for the com-parison with placebo) and 14.6 percent (P=0.09), re-spectively. The incidence of adverse events was sim-ilar among the groups. Conclusions In patients with familial adenomatouspolyposis, six months of twice-daily treatment with400 mg of celecoxib, a cyclooxygenase-2 inhibitor,leads to a significant reduction in the number of co-lorectal polyps. (N Engl J Med 2000;342:1946-52.) ©2000, Massachusetts Medical Society. From the University of Texas M.D. Anderson Cancer Center, Houston(G.S., P.M.L., N.W., Y.S., T.F., L.-K.S., B.L.); the Imperial Cancer Re-search Fund, St. Mark’s Hospital, London (R.K.S.P., M.H.W., B.S.); theNational Cancer Institute, Bethesda, Md. (E.H.); and G.D. Searle, Skokie,Ill. (G.B.G.). Address reprint requests to Dr. Steinbach at the University ofTexas M.D. Anderson Cancer Center, Division of Cancer Prevention, Box203, 1515 Holcombe Blvd., Houston, TX 77030, or at [email protected] authors were Louis Godio, Ph.D. (G.D. Searle, Skokie, Ill.), SherriPatterson, B.A. (M.D. Anderson Cancer Center, Houston), Miguel A.Rodriguez-Bigas, M.D. (Roswell Park Cancer Institute, Buffalo, N.Y.),Susan L. Jester, M.S. (G.D. Searle), Karen L. King, M.S. (G.D. Searle),Marta Schumacher, M.B.A. (National Cancer Institute, Bethesda, Md.),James Abbruzzese, M.D. (M.D. Anderson Cancer Center), Raymond N.DuBois, M.D., Ph.D. (Vanderbilt University Medical Center, Nashville),Walter N. Hittelman, Ph.D. (M.D. Anderson Cancer Center), Stuart Zim-merman, Ph.D. (M.D. Anderson Cancer Center), Jeffrey W. Sherman,M.D. (G.D. Searle), and Gary Kelloff, M.D. (National Cancer Institute). UMAN colon cancer develops in a step-wise fashion from normal mucosa to ad-enomatous polyps to carcinoma. Muta-tion in the adenomatous polyposis coli( APC ) gene commonly occurs early in the develop-ment of sporadic adenomas. 1 Patients with familialadenomatous polyposis have an inherited germ-line APC mutation 2 that results in hundreds of adenom-atous polyps and a nearly 100 percent risk of coloncancer. Management includes prophylactic procto-colectomy, or colectomy followed by sigmoidoscop-ic surveillance and rectal polypectomy. Because theadenoma-to-carcinoma sequence in familial adenom-atous polyposis resembles sporadic colon carcinogen-esis, 1 studies of familial adenomatous polyposis maycontribute to the prevention of sporadic adenomasand colon cancer.Nonsteroidal antiinflammatory drugs (NSAIDs)reduce the incidence of carcinogen-induced colon tu-mors in rodents. 3,4 NSAIDs are associated with a re-duced incidence of and mortality from sporadic ad-enoma and colon cancer in epidemiologic studies. 5-8 In early clinical studies 9,10 and small, randomized, pla-cebo-controlled trials, 11-13 sulindac caused the regres-sion of colorectal adenomas in patients with familialadenomatous polyposis. However, the gastrointesti-nal toxicity associated with conventional NSAIDs maylimit their long-term use for cancer prevention. 14 NSAIDs are inhibitors of the cyclooxygenaseenzyme family, which catalyzes the metabolism ofarachidonic acid to prostaglandins, prostacyclin, andthromboxanes. The cyclooxygenase-1 isoform is con-stitutively expressed in most tissues, where it medi-HDownloaded from www.nejm.org at Stanford University on September 22, 2004.Copyright © 2000 Massachusetts Medical Society. All rights reserved.THE EFFECT OF CELECOXIB, A CYCLOOXYGENASE-2 INHIBITOR, IN FAMILIAL ADENOMATOUS POLYPOSIS Volume 342 Number 26 · 1947 ates physiologic functions such as gastric mucosal cy-toprotection and regulation of platelet aggregation.Its inhibition may account for many of the commonside effects of NSAIDs, including gastric ulcerationand gastrointestinal hemorrhage. 14,15 The cyclooxy-genase-2 isoform is induced in response to cytokinesand growth factors and is expressed in inflammatorydisease, premalignant lesions (such as colorectal ad-enomas), and colon cancer. 16-18 Its inhibition has notbeen associated with gastric ulceration. 15,19-21 How-ever, the long-term effects of