Sonnenberg2.tripod.com

Chemoprevention of Colorectal Cancer by Aspirin: SAUD SULEIMAN,* DOUGLAS K. REX,‡ and AMNON SONNENBERG**Department of Veterans Affairs Medical Center, Albuquerque, New Mexico; and ‡Division of Gastroenterology, Department of Medicine,Indiana University School of Medicine, Indianapolis, Indiana diagnostic test, alternative strategies to prevent colorec- tal cancer have been sought. Regular intake of nonste-roidal anti-inflammatory drugs (NSAIDs), including as- Background & Aims: The aim of the study is to compare pirin, was found to cause regression of colonic polyps and the cost-effectiveness of aspirin and colonoscopy in the reduce the risk for developing colorectal cancer by 40%- prevention of colorectal cancer. Methods: A Markov pro- 50%.8-19 Chemoprevention of colorectal cancer by means cess is used to follow a hypothetical cohort of 100,000 of regular intake of aspirin or other NSAIDs may repre- subjects aged 50 years until death. Four strategies arecompared: (1) no intervention, (2) colonoscopy once per sent a viable option to reduce morbidity and mortality 10 years and every 3 years in subjects with polyps, (3) from colorectal cancer. NSAIDs are associated with mul- chemoprevention with 325 mg of daily aspirin, and (4) tiple side effects, involving predominantly hemostasis combination of the second and third strategies. The and the gastrointestinal (GI) tract, which can lead to various strategies are compared calculating incremental excessive consumption health care resources.20-23 Before cost-effectiveness ratios (ICERs). Results: The expected introducing a strategy of cancer prevention through number of colorectal cancers is 5904 per 100,000 sub- NSAIDs, therefore, one has to weigh the benefits of jects. Colonoscopy prevents 4428 colorectal cancers chemoprevention against their costly side effects. The and saves 7951 life-years at an ICER of $10,983 per aim of the present study is to compare the cost-effective- life-year saved compared with no intervention. Aspirin ness of regular aspirin intake with colonoscopy once per prevents 2952 colorectal cancers and saves 5301 life- 10 years for the prevention of colorectal cancer. Aspirin years at an ICER of $47,249 per life-year saved com- is chosen as a representative medication for the class of pared with no intervention. The cost of aspirin therapy NSAIDs because it is the most widely used and the least plus management of aspirin-related complications wasreported to be $172 per year per patient. Varying the annual aspirin-related costs between $50 and $200results in ICER changes between $4617 and $57,080, with the 2 strategies breaking even at $70. Applyingaspirin chemoprevention plus colonoscopy screening The cost-effectiveness of 3 cancer prevention strategies concomitantly yields an ICER of $227,607 per life-year are compared with a strategy of nonprevention. A previously saved compared with screening colonoscopy alone.
published model of a Markov process is used to follow a Conclusion: As compared with colonoscopy once per 10 hypothetical cohort of 100,000 persons aged 50 years until years, the use of aspirin to prevent colorectal cancer death.5 A cohort of 50-year-old persons is subjected to the saves fewer lives at higher costs. The high complication following 4 prevention strategies: (1) no intervention; (2) 1 cost and the lower efficacy of aspirin render screening colonoscopy every 10 years or, in case of adenomatous polyps, colonoscopy a more cost-effective strategy to prevent every 3 years until polyps are no longer found; (3) no colonos- copy but chemoprevention with 325 mg of daily aspirin; and(4) combination of the second and third strategy, that is,colonoscopy every 10 or 3 years plus daily aspirin. As in the Colorectal cancer ranks second among the causes of previous model, the time frame of the analysis is divided into cancer death in the United States.1 More than 95% equal increments of 1 year, during which the subjects transi- of colorectal cancer arises from adenomatous polyps.2Colonoscopy by way of polypectomy reduces colorectal Abbreviations used in this paper: ACER, average cost-effectiveness cancer, and this modality has been established as a cost- ratio; GI, gastrointestinal; ICER, incremental cost-effectiveness ratio.
2002 by the American Gastroenterological Association effective method for the prevention of colorectal can- cer.3-7 Because colonoscopy is an expensive and invasive Individual transitions between different states are associated with costs, the costs being estimated from the perspective of athird party payer. The costs of colorectal cancer therapy,colonoscopy, polypectomy, and their related complications relyon published cost data and Medicare reimbursements in theyear 2000.5 Published cost estimates for the medical care ofsubjects with colorectal cancer range between $25,000 and$45,000.7,30-32 These costs include expenditures for diagnosis,surgery, radiation, and chemotherapy. We use the most recentdata available from Lee et al.32 Smalley et al.22 published dataon the excess costs from GI disease associated with NSAIDs,comparing a cohort of 46,000 nonusers with 5,000 regular Figure 1. (Left) A Markov state diagram of screening for colorectalcancer (CRC) by repeat colonoscopy every 10 years in case of normal users. Assuming a 3% interest rate, their 1989 costs data were colonoscopy or every 3 years after polypectomy. (Right) A Markov updated to an average net present value of $154 per patient per state diagram for CRC prophylaxis using daily aspirin. The arrows year. Adding the annual drug cost of $18 spent on 325 mg of symbolize transitions between the various states. s/p, status post.
aspirin per day, the total cost of chemoprevention amounts to$172 per patient per year. These estimates were used as tion from one state of health to another. The Markov chains baseline cost. Similar cost data are also available from other underlying the preventive strategies are depicted in Figure 1.
sources. Based on 527 Medicaid patients treated over 786 In the colonoscopy model, all subjects start with a colonoscopy treatment quarters with NSAIDs, Bloom21 estimated that in at age 50. Subjects can then transition among 4 different 1982, $66 was spent per quarter on adverse GI drug reactions.
states, that is, (1) a state after a negative colonoscopy without These numbers translate to $449 per patient per year in 2000.
polyps, (2) a state after colonoscopy plus polypectomy, (3) a Lastly, in a Canadian population of 5,268 NSAID users, GI state after developing colorectal cancer, and (4) death from adverse events during a 2-year follow-up cost $134 (1995 colorectal cancer or other causes. In case of aspirin prophylaxis, Canadian dollars), resulting in a current estimate of $78 per all subjects are started on a daily dose of aspirin. Subsequently, year.23 Although the absolute Canadian price levels for all they can transition among 3 different states, that is, (1) remain medical interventions were lower than in the United States, disease-free on aspirin prophylaxis, (2) develop a colorectal the percentage of GI-related costs of NSAID therapy was cancer, or (3) die from colorectal cancer or other causes. The found to be similar in both countries.21-23 In a sensitivity various Markov models of the 4 strategies for cancer preventionare simulated on Excel spreadsheets (Microsoft, Redmond, Table 1. Transitions and Costs Used in the Markov Model The transitions among the various states are governed by chance, the annual transition rates being shown in Table 1. To facilitate comparison with other strategies of cancer preven- tion, the same transition probabilities as in previous models of colorectal cancer are used.5,24 The Markov model uses an Efficacy of colonoscopy in preventing CRC annual 1% incidence polyp rate to calculate the number of polypectomies and repeat colonoscopies after polypectomy.25 The annual age-specific incidence rate of colorectal cancer is taken from published statistics of the Surveillance, Epidemi- ology, and End Results Program.26 The population in each state is also subjected to natural attrition by the annual age- specific death rate of the US population.27 Colonoscopy, polypectomy, or aspirin prevent colorectal cancer by reducing its incidence. In addition, early detection of colorectal cancer through colonoscopy lowers cancer-related mortality. The National Polyp Study3 showed an efficacy of colonoscopy in reducing the incidence of colorectal cancer ranging between 76% and 90%. As other studies have sug- gested an efficacy of only 49% to 59%,4,28,29 a median value of 75% is chosen as a baseline rate for the present analysis with a range of 50%-75% used in a subsequent sensitivity analysis.
The 50% efficacy of aspirin has been reported in preventing colorectal cancer.8-10 In the sensitivity analysis, this efficacy NOTE. Expenditures include professional fees and facility costs.
Table 2. Outcome of Various Programs to Prevent Colorectal Cancer NOTE. Numbers in the table relate to a cohort of 100,000 persons aged 50 and followed on the average for 28.5 years. Future life-years savedand future costs are discounted using an annual rate of 3%.
CRC, colorectal cancer; ICER, incremental cost-effectiveness ratio of costs per saved life-year.
analysis, the baseline cost of chemoprevention is varied be- 50% efficacy prevents 2,952 colorectal cancers and saves 5,301 life-years. The ACERs and ICERs are both larger Effectiveness of screening is measured in terms of life-years than those of colonoscopy. Chemoprevention as an added saved through prevention of colorectal cancer and improved measure to colonoscopy is assumed to prevent an addi- survival of earlier cancer stages. The amount of life-years saved tional 50% of all cancers failed to prevent through through prevention corresponds to the difference in life-years colonoscopy alone. The ACERs and ICERs are smaller lost from cancer-related deaths between each 2 Markov models, than those of chemoprevention alone, but much larger 1 with and 1 without a preventive strategy. All costs arising than those of colonoscopy alone. The ICER of combined from screening colonoscopy, chemoprevention, and the care ofcolorectal cancer and all future life-years saved are discounted chemoprevention plus colonoscopy, compared with at an annual rate of 3%.33 The average cost-effectiveness ratio colonoscopy alone, exceeds $200,000 per life-year saved (ACER) corresponds to the total costs in the patient cohort (Table 2). On the other hand, if aspirin prevention was divided by the total amount of life-years saved. The incremen- already implemented in the general population, the ad- tal cost-effectiveness ratio (ICER) compares each screening dition of 1 colonoscopy per 10 years to prevent even more strategy with the previous less effective option, including a deaths from colorectal cancers would result in a relatively strategy of no screening. The ICER is calculated as the differ- low ICER of $34,836 compared with chemoprevention ence in costs divided by the corresponding difference in effec- alone. In other words, chemoprevention on top of screen- ing colonoscopy is not cost-effective, whereas screeningcolonoscopy on top of chemoprevention still represents a Table 2 shows the outcomes of modeling 4 dif- Figure 2 shows the results of a sensitivity analysis ferent strategies to prevent colorectal cancer. Without varying the cost of chemoprevention and the preventive any prevention, the expected number of colorectal can- efficacy of both colonoscopy and daily aspirin. Under cers in a cohort of 100,000 subjects is 5,904, and the baseline conditions, the costs of chemoprevention need to only costs incurred relate to the care of colorectal cancer.
fall below $70 to become more cost-effective than Colonoscopy once per 10 years at 75% efficacy prevents colonoscopy. Even under the assumption of highly effi- 4,428 colorectal cancers and saves 7,951 life-years. The cacious (75%) chemoprevention and a comparatively in- cost of colonoscopy screening in this cohort amounts to efficacious (50%) colonoscopy, the threshold is only $223,780,829. This includes facility fees, physician fees, $150. This value still lies below the actual costs of $172 and the costs incurred for the care of associated compli- cations. The ACER is $28,143 per life-year saved. The Under baseline assumptions, colonoscopy combined ICER is calculated at $10,983 per life-year saved when with chemoprevention prevents 87.5% of all colorectal compared with no intervention. Chemoprevention at cancers. In a second sensitivity analysis, the combined sent a conceptually interesting but presently not cost-effective strategy. The relatively large costs associatedwith the adverse effects of NSAIDs render 1 colonoscopyevery 10 years a more cost-effective alternative. A com-bined prevention strategy using both colonoscopy plusdaily aspirin could increase the overall effectiveness of acancer prevention program. However, the incrementalcost-effectiveness of chemoprevention combined withcolonoscopy alone would be rather high, costing morethan $100,000 per additional life-year saved as comparedwith colonoscopy alone.
In the assessment of cost-effectiveness, our analysis was based on the comparison of 4 management options, Figure 2. Influence of cost and efficacy of chemoprevention on itsICER compared with no prevention. The intersection between the lines that is, no prevention, colonoscopy alone, aspirin alone, of colonoscopy and chemoprevention delineates the threshold costs and combination of colonoscopy plus aspirin. The argu- of chemoprevention to become less or more expensive than colonos- ments against or in favor of a particular strategy are based on the comparison of 2 ICERs rather than their absolutevalues. In terms of absolute values, however, all ICERsfall within a range that is still considered economically efficacy is varied between 50% and 100%. The lower feasible. To put values of ICERs in perspective, it has value corresponds to the efficacy of colonoscopy alone become customary to compare the cost-effectiveness of with a 0% contribution of chemoprevention, whereas various health care interventions by "league tables."35 100% corresponds to a 25% or 50% efficacy of chemo- Different types of medical procedures and therapies are prevention added to a 75% or 50% baseline efficacy of ranked in a league table by their cost-effectiveness. For colonoscopy alone. On the x-axis of Figure 3, the addi- instance, it has been estimated that endoscopic surveil- tional efficacy of chemoprevention (beyond the baseline lance of Barrett's esophagus would cost $98,000 per efficacy of colonoscopy) is varied between 0% and 50%.
quality adjusted life-year gained. Cervical cancer screen- The 2 curves represent 2 baseline efficacy rates of ing may cost as much as $250,000 per life saved.36 colonoscopy. The lower curve representing 75% baseline The present decision analysis uses a previously pub- efficacy of colonoscopy ends at 25% efficacy of chemo- lished model of a Markov process to assess the impact of prevention, because otherwise the joint efficacy would various screening strategies on the prevention of colorec- rise beyond 100%. As the figure shows, a higher addedefficacy of chemoprevention makes it a more cost-effec-tive option. If endoscopy and aspirin combined were ableto prevent all cancers, the incremental cost-effectivenessof aspirin would vary between $50,000 and $100,000,dependent on the preventive efficacy of colonoscopyalone. Such expectations seem unrealistic, and with moremoderate success of chemoprevention, its cost-effective-ness ratio (as compared with colonoscopy alone) is morelikely to fall into the $100,000 to $200,000 range. Thisholds true even if due to the reduced incidence rate ofpolyps under chemoprevention, the interval of surveil-lance colonoscopy after polypectomy is increased to 5years.
The notion is quite appealing that by taking a daily aspirin pill, one could reduce the risk of colo-rectal cancer and avoid unpleasant, repeated, and costlycolonoscopy. As the present analysis shows, unfortu- Figure 3. Influence of chemoprevention efficacy on the ICER of che- nately, chemoprevention of colorectal cancer may repre- moprevention plus colonoscopy compared with colonoscopy alone.
tal cancer.5 As compared with colonoscopy, the previous To reduce the side effects of NSAIDs, new specific model showed screening by annual fecal occult blood cyclooxygenase-2 inhibitors have been developed.37,38 testing to cost less, but also save fewer life-years. A Results from laboratory studies suggest that, like con- screening strategy based on flexible sigmoidoscopy every ventional NSAIDs, these newer compounds with a lesser 5 or 10 years was not cost-effective compared with the GI toxicity may also protect against colorectal can- other 2 screening methods. Because under baseline as- cer.39-41 The cost saved by the drugs' safer profile and sumptions, screening through fecal occult blood testing lesser side effects, however, becomes spent on the drug and decennial colonoscopy were comparable, fecal occult itself. A year's supply of daily 100 mg celecoxib (Cele- blood testing was not again included in this study. The brex; Searle, Skokie, IL) costs about $600. These costs incidence rates of colon polyps and the length of the time exceed by far the threshold for chemoprevention to be- intervals between each 2 endoscopic examinations had come a cost-effective alternative. It is conceivable, how- already been shown to exert a lesser influence on the ever, that in the future other and cheaper drugs with few outcome of the analysis. The present analysis also relies or no adverse effects on the GI tract will make chemo- on other results of the foregoing cost-effectiveness study.
prevention of colorectal cancer a cost-effective option.
Because low compliance of colonoscopy, for instance, Several economic analyses have dealt with the use of reduces both the overall number of cancers prevented and aspirin in the secondary prevention of recurrent myocar- the total costs in a linear fashion, it does not affect the dial infarction, transient ischemic attack, or stroke. In cost-effectiveness ratio of colonoscopy. A low compliance addition to being clinically efficacious, these studies sug- rate of aspirin intake would reduce its efficacy and render gest that aspirin therapy also represents a cost-effective it even less cost-effective than the other means of cancer management option.42-44 In patients who are already put on aspirin to prevent recurrent cardiovascular disease, Chemoprevention alone is not cost-effective compared therefore, the prevention of colorectal cancer adds to the with colonoscopy alone for 2 reasons. The first reason overall cost-effectiveness of the preventive strategy.
relates to the overall high annual costs associated with Three large randomized clinical trials have shown that regular intake of nonsteroidals. The common occurrence aspirin is efficacious in the primary prevention of cardio- of adverse effects during the treatment with NSAIDs vascular disease.45-47 It reduces the overall occurrence of renders this drug class costly, even if the expenses for the cardiovascular death, nonfatal myocardial infarction, and drug itself may appear relatively cheap. The annual costs stroke by 15%-30%. These results were achieved in of chemoprevention, including the expenditures for the populations of hypertensive patients or patients who NSAIDs and their side effects, would need to fall below presented with an increased risk for cardiovascular dis- a threshold that seems by today's standards unrealisti- ease. Chemoprevention in the general population would cally low. The second influence to determine the out- be associated with a lesser benefit, but excess GI bleeding come of chemoprevention relates to its efficacy in pre- and hemorrhagic cerebrovascular events. It is currently venting colorectal cancer. Even if chemoprevention alone being debated among cardiologists and public health were as efficacious as colonoscopy and averted, for in- researchers whether, in the general population as well, stance, 75% of all colorectal cancers, its overall annual the benefit of aspirin outweighs its adverse effects.48,49 costs would need to be less than $100 to $150 to become Data derived through meta-analysis and modeling sug- a cost-effective alternative. Again, such assumptions gest that primary prevention of cardiovascular disease seem overly optimistic. Chemoprevention represents may be cost-effective only in subpopulations with an only a viable, yet expensive means to reduce the occur- increased risk.49-51 Because cardiovascular disease is rence of colorectal cancer if added to an endoscopic much more frequent than colorectal cancer, the issue of screening program. Rather than estimate the possible cancer prevention becomes completely overwhelmed by occurrence of various GI adverse events and the health the cardiovascular side. From a strictly cancer-related care expenditures associated with individual complica- perspective, however, chemoprevention alone does not tions, the present model relies on actual average cost represent a cost-effective alternative to colonoscopy. If data. The cost data from several large patient populations primary cardiovascular prevention with aspirin were al- treated with NSAIDs fall within the same order of ready used in the general population, the added screen- magnitude and support the contention that similar costs ing through colonoscopy would still represent a cost- could be expected if chemoprevention were to be imple- effective strategy, because it would save many more lives We conclude that in the prevention of colorectal can- rectal polyps in familial adenomatous polyposis. Gastroenterol- cer, colonoscopy once per 10 years saves more lives at 19. Spagnesi MT, Tonelli F, Dolara P, Caderni G, Valanzano R, Anas- lower overall costs than daily aspirin. The high cost of tasi A, Bianchini F. Rectal proliferation and polyp occurrence in chemoprevention and its relatively low efficacy render patients with familial adenomatous polyposis after sulindac screening colonoscopy the more cost-effective option to treatment. Gastroenterology 1994;106:362-366.
20. Trujilllo MA, Garewal HS, Sampliner RE. Nonsteroidal anti-inflam- reduce morbidity and mortality related to colorectal matory agents in chemoprevention of colorectal cancer. At what cancer. Chemoprevention used in addition to endoscopic cost? Dig Dis Sci 1994;39:2260 -2266.
screening provides a viable strategy to save more lives 21. Bloom BB. Direct medical costs of disease and gastrointestinal than through colonoscopy screening alone, although this side effects during treatment for arthritis. Am J Med 1988;84(suppl 2A):20 -24.
benefit is associated with an ICER that may be prohib- 22. Smalley WE, Griffin MR, Fought RL, Ray WA. Excess cost from itively high for most health-care systems.
gastrointestinal disease associated with nonsteroidal anti-in-flammatory drugs. J Gen Intern Med 1996;11:461- 469.
23. Rahme E, Joseph L, Kong SX, Watson DJ, LeLorier J. Gastroin- testinal health care resource use and costs associated with 1. American Cancer Society. Cancer facts and figures, 1998. Publi- nonsteroidal antiinflammatory drugs versus acetaminophen. Ret- cation No 5008.98. Atlanta, GA: American Cancer Society, 1998.
rospective cohort study of an elderly population. Arthritis Rheum 2. Peipins LA, Sandler RS. Epidemiology of colorectal adenomas.
` F. Cost-effectiveness of a single colonos- 3. Winawer SJ, Zauber AG, Ho MN, O'Brien MJ, Ho MN, Gottlieb LS, copy in screening for colorectal cancer. Arch Intern Med 2001;in Sternberg SS, Waye JD, Schapiro M, Bond JH, Panish JF. Preven- tion of colorectal cancer by colonoscopic polypectomy. The Na- 25. Williams A, Balasooria B, Day D. Polyps and cancer of the large tional Polyp Study Workgroup. N Engl J Med 1993;329:1977- bowel: a necropsy study in Liverpool. Gut 1982;23:835- 842.
26. Ries LAG, Kosary CL, Hankey BF, Miller BA, Harras A, Edwards BK ¨ller AD, Sonnenberg A. Protection by endoscopy against death (eds). SEER Cancer Statistics Review, 1973-1994, National Can- from colorectal cancer: a case control study among veterans.
cer Institute. NIH Publication No. 97-2789. Bethesda, MD: NIH, Arch Intern Med 1995;155:1741-1748.
` F, Inadomi JM. The cost-effectiveness of 27. National Center for Health Statistics. US decennial life tables for colonoscopy in screening for colorectal cancer. Ann Intern Med 1989-91. DHHS Publication No. PHS-98-1150-1, Volume 1, No 1.
Hyattsville, MD: U.S. Department of Health and Human Services, 6. Eddy DM. Screening for colorectal cancer. Ann Intern Med 1990; ¨ller AD, Sonnenberg A. Prevention of colorectal cancer by 7. Lieberman DA. Cost-effectiveness model for colorectal cancer flexible endoscopy: a case control study of 32,702 veterans. Ann screening. Gastroenterology 1995;109:1781-1790.
8. Kune GA, Kune S, Watson FL. Colorectal cancer risk, chronic 29. Selby JV, Friedman GD, Quesenberry CP, Weiss NS. A case illnesses, operations, and medications: case control results from control study of screening sigmoidoscopy and mortality from the Melbourne colorectal cancer study. Cancer Res 1988;48: colorectal cancer. N Engl J Med 1992;326:653- 657.
30. Winawer SJ, Fletcher RH, Miller L, Godlee F, Stolar MH, Mulrow 9. Thun MJ, Namboodiri MM, Heath CW Jr. Aspirin use and reduced CD, Woolf SH, Glick SN, Ganiats TG, Bond JH, Rosen L, Zapka JG, risk of fatal colon cancer. N Engl J Med 1991;325:1593-1596.
Olsen SJ, Giardiello FM, Sisk JE, Van Antwerp R, Brown-Davis C, 10. Greenberg ER, Baron JA, Freeman DH Jr, Mandel JS, Haile R.
Marciniak DA, Mayer RJ. Colorectal cancer screening: clinical Reduced risk of large bowel adenomas among aspirin users.
guidelines and rationale. Gastroenterology 1997;112:594 - 642.
J Natl Cancer Inst 1993;85:912-916.
31. Provenzale D, Wong JB, Onken JE, Lipscomb J. Performing a 11. Sandler RS, Galanko JC, Murray SC, Helm JF, Woosley JT. Aspirin cost-effectiveness analysis: surveillance of patients with ulcer- and non-steroidal anti-inflammatory agents and risk for colorectal ative colitis. Am J Gastroenterol 1997;93:872- 880.
adenomas. Gastroenterology 1998;114:441- 447.
32. Lee JG, Vigil HT, Leung JW. The hospital costs for diagnosis and 12. Giovannucci E, Rimm EB, Stampfer MJ, Colditz GA, Ascherio A, treatment of colorectal cancer (abstr). Gastrointest Endosc Willett WC. Aspirin use and the risk for colorectal cancer and adenoma in male health professionals. Ann Intern Med 1994;121:241-246.
33. Weinstein MC, Stason WB. Foundations of cost-effectiveness for 13. Giovannucci E, Egan KM, Hunter DJ, Stampfer MJ, Colditz GA, health and medical practices. N Engl J Med 1977;296:716 -721.
Willett WC, Speizer FE. Aspirin and the risk of colorectal cancer in 34. Siegel JE, Weinstein MC, Torrance GW. Reporting cost-effective- women. N Engl J Med 1995;333:609 - 614.
ness studies and results. In: Gold MR, Siegel JE, Russel LB, 14. Suh O, Mettlin C, Petrelli NJ. Aspirin use, cancer, and polyps of Weinstein MC, eds. Cost-effectiveness in health and medicine.
the large bowel. Cancer 1993;72:1171-1177.
New York: Oxford University Press, 1996:276 -303.
15. DuBois RN, Giardiello FM, Smalley WE. Nonsteroidal anti-inflam- 35. Drummond MF, O'Brien B, Stoddart GL, Torrance GW. Methods matory drugs, eicosanoids, and colorectal cancer prevention.
for the economic evaluation of health care programmes. 2nd ed.
Gastroenterol Clin North Am 1996;25:773-791.
New York: Oxford University Press, 1997.
16. DuBois RN, Smalley WE. Cyclooxygenase, NSAIDs, and colorectal 36. Provenzale D, Schmitt C, Wong JB. Barrett's esophagus: a new cancer. J Gastroenterol 1996;31:898 -906.
look at surveillance based on emerging estimates of cancer risk.
' JM, Rubio E, Planas R, Tarrech JM, Bordas JM.
Am J Gastroenterol 1999;94:2043-2053.
Effects of long-term sulindac therapy on colonic polyposis. Ann 37. Laine L, Harper S, Simon T, Bath R, Johanson J, Schwartz H, Stern S, Quan H, Bolognese J. A randomized trial comparing the 18. Labayle D, Fischer D, Vielh P, Drouhin F, Pariente A, Bories C, effect of rofecoxib, a cyclooxygenase 2-specific inhibitor, with Duhamel O, Trousset M, Attali P. Sulindac causes regression of that of ibuprofen on the gastroduodenal mucosa of patients with osteoarthritis. Rofecoxib Osteoarthritis Endoscopy Study Group.
Framework. Thrombosis prevention trial: randomised trial of low- Gastroenterology 1999;117:776 -783.
intensity oral anticoagulation with warfarin and low-dose aspirin 38. Silverstein FE, Faich G, Goldstein JL, Simon LS, Pincus T, Whel- in the primary prevention of ischaemic heart disease in men at ton A, Makuch R, Eisen G, Agrawal NM, Stenson WF, Burr AM, increased risk. Lancet 1998;351:233-241.
Zhao WW, Kent JD, Lefkowith JB, Verburg KM, Geis GS. Gastro- 46. Hansson L, Zanchetti A, Carruthers SG, Dahlof B, Elmfeldt D, intestinal toxicity with celecoxib vs nonsteroidal anti-inflammatory Julius S, Menard J, Rahn KH, Wedel H, Westerling S. Effects of drugs for osteoarthritis and rheumatoid arthritis: the CLASS intensive blood-pressure lowering and low-dose aspirin in pa- study: a randomized controlled trial. Celecoxib Long-term Arthritis tients with hypertension: principal results of the Hypertension Safety Study. JAMA 2000;284:1247-1255.
Optimal Treatment (HOT) randomised trial. HOT Study Group.
39. Boolbol SK, Dannenberg AJ, Chadburn A, Martucci C, Guo XJ, Ramonetti JT, Abreu-Goris M, Newmark HL, Lipkin ML, DeCosse 47. Collaborative Group of the Primary Prevention Project (PPP). Low- JJ, Bertagnolli MM. Cyclooxygenase-2 overexpression and tumor dose aspirin and vitamin E in people at cardiovascular risk: a formation are blocked by sulindac in a murine model of familial randomised trial in general practice. Lancet 2001;357:89 -95.
adenomatous polyposis. Cancer Res 1996;56:2256 -2260.
48. Hebert PR, Hennekens CH. An overview of the 4 randomized trials 40. Sheng H, Shao J, Kirkland SC, Isakson P, Coffey RJ, Morrow J, of aspirin therapy in the primary prevention of vascular disease.
Beauchamp RD, DuBois RN. Inhibition of human colon cancer Arch Intern Med 2000;160:3123-3127.
cell growth by selective inhibition of cyclooxygenase-2. J Clin 49. Sanmuganathan PS, Ghahramani P, Jackson PR, Wallis EJ, Ram- say LE. Aspirin for primary prevention of coronary heart disease: 41. Kawamori T, Rao CV, Seibert K, Reddy BS. Chemopreventive safety and absolute benefit related to coronary risk derived from activity of celecoxib, a specific cyclooxygenase-2 inhibitor, meta-analysis of randomised trials. Heart 2001;85:265-271.
against colon carcinogenesis. Cancer Res 1998;58:409 - 412.
50. Augustovski FA, Cantor SB, Thach CT, Spann SJ. Aspirin for 42. Shah H, Gondek K. Aspirin plus extended-release dipyridamole or primary prevention of cardiovascular events. J Gen Intern Med clopidogrel compared with aspirin monotherapy for the preven- tion of recurrent ischemic stroke: a cost-effectiveness analysis.
51. Troche CJ, Tacke J, Hinzpeter B, Danner M, Lauterbach KW.
Cost-effectiveness of primary and secondary prevention in car- 43. Ebrahim S. Cost-effectiveness of stroke prevention. Brit Med Bull diovascular disease. Eur Heart J 1999;19(suppl C):C59 - 65.
44. Sarasin FP, Gaspoz JM, Bounameaux H. Cost-effectiveness of new antiplatelet regimes used as secondary prevention of stroke Received June 1, 2001. Accepted August 2, 2001.
or transient ischemic attack. Arch Intern Med 2000;160:2773- Address requests for reprints to: Amnon Sonnenberg, M.D., M.Sc., Department of Veterans Affairs Medical Center 111F, 1501 San Pedro 45. The Medical Research Council's General Practice Research Drive SE, Albuquerque, New Mexico 87108. e-mail: sonnbrg@unm.edu.

Source: http://sonnenberg2.tripod.com/chemoprev.pdf