Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 7-Day Trial for You or Your Team.

Learn More →

Aspirin use and lung cancer in men

Aspirin use and lung cancer in men British Journal of Cancer (2003) 89, 1705 – 1708 & 2003 Cancer Research UK All rights reserved 0007 – 0920/03 $25.00 www.bjcancer.com ,1,2 3 3 1,2,4,5 1,2,4,5 CN Holick , DS Michaud , MF Leitzmann , WC Willett and E Giovannucci 1 2 Department of Nutrition, Harvard School of Public Health, 677 Huntington Avenue, Boston, MA 02115, USA; Department of Epidemiology, Harvard School of Public Health, 677 Huntington Avenue, Boston, MA 02115, USA; Nutritional Epidemiology Branch, Division of Cancer Epidemiology and Genetics, National Cancer Institute, Bethesda, MD 20892, USA; Channing Laboratory, Department of Medicine, Brigham and Women’s Hospital, Boston, MA 02115, USA; Harvard Medical School, Boston, MA 02115, USA We examined prospectively the relation between regular aspirin use and lung cancer risk in the Health Professionals Follow-Up Study. Of 49 383 US men aged 40–75 years who completed biennial self-administered questionnaires that assessed aspirin use beginning in 1986, 328 developed lung cancer during 601 453 person-years of follow-up through 31 December 2000. No information on aspirin dose was available. Controlling for current age, smoking status, and age at starting to smoke regularly, the relative risk (RR) of total lung cancer for regular users of aspirin (twice or more per week) at baseline compared to nonusers was 1.13 (95% confidence interval (CI) ¼ 0.89–1.43). Results were similar for non-small-cell lung cancer (RR¼ 1.16; 95% CI¼ 0.88–1.54). No apparent dose-dependent association was observed for the frequency of aspirin use and lung cancer risk (P for trend¼ 0.64), and results remained null when consistent use of aspirin over time was examined. These findings do not suggest that regular aspirin use is associated with a reduced lung cancer risk. British Journal of Cancer (2003) 89, 1705 – 1708. doi:10.1038/sj.bjc.6601343 www.bjcancer.com & 2003 Cancer Research UK Keywords: aspirin; lung cancer; chemoprevention; prospective studies; epidemiology Evidence from animal studies suggests that aspirin and possibly MATERIALS AND METHODS other nonsteroidal anti-inflammatory drugs (NSAIDs) including Study population cyclooxygenase-2 (COX-2) inhibitors may influence the develop- ment and progression of lung cancer (Duperron and Castonguay, The Health Professionals Follow-Up Study (HPFS) is an ongoing 1997; Rioux and Castonguay, 1998; Yao et al, 2000). As COX-2 prospective cohort study of 51 529 US male dentists, optometrists, enzymes are constitutively expressed in lung neoplastic tissue (Xu, osteopaths, podiatrists, pharmacists, and veterinarians aged 40–75 2002), reduced synthesis of prostaglandins from arachidonic acid years. The primary objective of the study is to evaluate a series of by inhibition of COX-2 enzymes may have preventive or hypotheses relating nutritional and lifestyle factors to the therapeutic effects in lung carcinogenesis. incidence of cancer, heart disease, and other major diseases. At Clinical and epidemiological studies, however, have not baseline in 1986, enrollees returned a mailed questionnaire that consistently supported for the hypothesis that regular use of assessed information on lifestyle factors, aspirin and other NSAID aspirin and other NSAIDs reduces lung cancer incidence but are use, diet, and medical and smoking histories. Follow-up ques- limited in several key areas. These include residual confounding tionnaires are mailed biennially to the entire cohort to update due to broad categories of smoking habits (Paganini-Hill et al, information on potential risk factors and ascertain new cases of 1989; Schreinemachers and Everson, 1994; Langman et al, 2000; disease. Akhmedkhanov et al, 2002; Moysich et al, 2002); misclassification due to a single assessment of aspirin use (Paganini-Hill et al, 1989; Schreinemachers and Everson, 1994; Harris et al., 2002); and small Assessment of aspirin use number of lung cancer cases observed (Peto et al, 1988; Paganini- Regular use of aspirin (e.g. Anacin, Bufferin, Alka-Seltzer), Hill et al, 1989; Schreinemachers and Everson, 1994; Akhmedkha- acetaminophen (e.g. Tylenol), and other NSAIDs (e.g. Motrin, nov et al, 2002). Indocin, Naprosyn, Dolobid) separately defined as two or more Given the limitations of previous literature and uncertainty times per week was ascertained in 1986 and updated every 2 years regarding the protection of aspirin against lung cancer risk, we thereafter. More detailed questions on aspirin use began in the examined the association between regular use of aspirin and lung 1992 questionnaire. Frequency of aspirin use was assessed by the cancer incidence in a large cohort of US. male health professionals average number of days each month aspirin was taken (none, 1–4, with long-term aspirin use and detailed information on smoking 5–14, 15–21, or 22 or more). Information on aspirin dose was not history. available. Reasons for aspirin use were assessed in 1993 by a supplemen- tary questionnaire sent to a random sample of 211 participants, *Correspondence: CN Holick; E-mail: [email protected] who reported taking aspirin from 1986 to 1990. Of the 211 men, Received 19 June 2003; revised 27 August 2003; accepted 1 September 185 (88%) responded and reported one or more of the following Epidemiology Epidemiology Aspirin use and lung cancer CN Holick et al reasons: cardiovascular disease, 25%; to decrease risk for carcinoma). Potential interaction by smoking status category cardiovascular disease, 58%; headache, 25%; joint or musculoske- (never, past, and current separately) and baseline aspirin use was letal pain, 33%; and other or unknown reasons, 7%. We were assessed using a cross-product term in multivariate models. Tests unable to examine long-term use of acetaminophen and other of linear trends for increasing categories of aspirin use were NSAIDs separately because of the small number of men who used conducted by assigning the median value of aspirin use for them regularly (approximately 6% of the cohort). categories and treating these as a single continuous variable. Reported P-values are based on two-sided tests. Identification of lung cancer cases On each biennial questionnaire participants were asked whether RESULTS they had been diagnosed with lung cancer during the previous 2 years. The follow-up rate with respect to the incidence of cancer At baseline in 1986, about one-third of men reported regular was 96% of the total possible person-years. After receiving aspirin use defined as two or more tablets per week (Table 1). permission from identified cases (or next of kin for decedents), Aspirin users tended to be slightly older, to have smoked, and to hospital records and pathology reports were obtained and have started smoking at an earlier age. Dietary patterns were very reviewed by a physician for histological confirmation. Information on lung cancer cell type was available for approximately 80%. Deaths in the cohort were ascertained through family members Table 1 Characteristics (means and proportions) of the HPFS cohort and the National Death Index (Stampfer et al, 1984). Cases by aspirin use at baseline in 1986 diagnosed between date of 1986 questionnaire return and 31 December 2000 were included in this report (n¼ 328). Aspirin Characteristic Nonuse Use Statistical analysis Follow-up for each participant was calculated from the date of Participants (n) 34 841 14 542 Age (years) 53.7 56.4 return of the 1986 questionnaire until the date of lung cancer Height (in) 70.0 70.1 diagnosis, date of death from any cause, or 31 December 2000, Body mass index (kg m ) 24.9 25.1 which ever came first. At baseline in 1986, we excluded 2069 men who reported a history of cancer other than nonmelanoma skin Smoking history cancer by 1986, 40 men who incorrectly completed the ques- Smoking status (%) tionnaire, and 37 men who later wished to be removed from the Never 45.8 41.9 cohort. Only those with complete information on aspirin use at Past (years) baseline were included in the analyses. The analytic cohort o10 11.4 14.9 consisted of 49 383 men representing 601 453 person-years of 10+ 27.7 31.9 Current (cigarettes per day) follow-up. Cox proportional hazards models were used to estimate 1 – 4 0.9 1.1 relative risks (RRs) and 95% confidence intervals (CI). 5 – 14 1.5 1.9 To take account for changes in aspirin use over time and best 15 – 24 3.2 3.1 represent long-term use of aspirin, our analyses were conducted 25 – 34 1.4 1.6 using consistent aspirin use. For example, lung cancer from 1986 35 – 44 1.0 1.2 through 2000 was related to aspirin use reported on the 1986 45+ 0.4 0.5 questionnaire, that from 1988 through 2000 to consistent use Age (years) at starting to smoke (%) reported in the 1986 and 1988 questionnaires, that from 1990 o15 5.7 6.9 through 2000 use reported on the 1986, 1988, and 1990 15 – 19 18.9 23.4 20 – 29 20.8 23.6 questionnaires, and that from 1992 through 2000 use reported 30+ 47.6 44.0 on the 1986, 1988, 1990, and 1992 questionnaires. Consistent Years since quitting 8.3 8.1 aspirin users were compared to nonusers (i.e., participants who consistently reported no aspirin use during the same specified time Intakes (per day) periods). Fruit and vegetable 5.3 5.3 Potential confounders were specified a priori based on a review b-Carotene (mg) 5150 5039 of putative risk factors for lung cancer and included age (year); a-Carotene (mg) 949 918 marital status; body mass index (kg m ); use of multivitamins, Lycopene (mg) 10 374 10 321 vitamin A, vitamin C, vitamin E, b-carotene, and selenium Multivitamin use (%) 39.4 50.0 Supplement use (per day) supplements (yes or no); intake of fruit and vegetables; energy- Vitamin C (mg) 57.6 67.7 adjusted a-carotene, b-carotene, lycopene, lutein/zeaxanthin, and Vitamin E (IU) 17.6 22.7 b-cryptoxanthin; alcohol use; and family history of lung cancer. Vitamin A (IU) 7.9 9.3 Smoking history was categorised as the combination of age at Selenium (mg) 6.2 7.5 starting to smoke regularly (less than 15, 15–19, 20–29, and more d b-Carotene (mg) 1.9 2.6 than 30 years old) and smoking status, which included current smoking (1–4, 5–14, 15–24, 25–34, 35–44, and 45 or more Family history of lung cancer (%) 5.7 6.0 cigarettes per day), past smoking with time since quitting (less Disease history (%) than 10 years and 10 or more years), and never smoked. Asthma 5.2 5.4 Emphysema or chronic bronchitis 1.4 1.6 Multivariate models included current age, age at starting to smoke Rheumatoid arthritis 2.0 3.1 regularly, and smoking status (included time since quitting and Other arthritis (e.g. degenerative) 7.3 12.2 dose), plus the additional confounders mentioned previously that a b were assessed by evaluating whether their inclusion into the Standardised to the age distribution of the study population. Aspirin use was multivariate model changed the risk estimate by more than 10%. defined as aspirin use two or more times per week. Values do not add up to 100% d e Additional analyses were restricted to non-small-cell lung cancer because of missing values. Use of supplement on a regular basis. Data from the 1996 follow-up questionnaire. (NSCLC) (squamous carcinoma, adenocarcinoma, and large cell British Journal of Cancer (2003) 89(9), 1705 – 1708 & 2003 Cancer Research UK Aspirin use and lung cancer CN Holick et al Table 2 Relative risk of lung cancer by aspirin use in the HPFS, 1986– 2000 Aspirin use Variable 1986 1986+1988 1986+1988+1990 1986+1988+1990+1992 Follow-up period 1986 – 2000 1988 – 2000 1990 – 2000 1992 – 2000 Total cases Nonusers, cases/person-years 204/428 688 96/202 457 66/126 065 22/44 436 Users, cases/person-years 124/172 765 64/93 431 41/52 730 28/37 478 Age-adjusted RR (95% CI) 1.19 (0.95 – 1.49) 1.11 (0.80 – 1.53) 1.07 (0.72 – 1.60) 1.18 (0.66 – 2.10) Multivariate RR (95% CI) 1.13 (0.89 – 1.43) 0.98 (0.70 – 1.36) 0.88 (0.58 – 1.34) 0.89 (0.47 – 1.67) Non-small-cell carcinomas Nonusers, cases/person-years 140/428 717 69/202 466 45/126 073 13/44 441 Users, cases/person-years 87/172 789 49/93 441 33/52 733 24/37 479 Age-adjusted RR (95% CI) 1.25 (0.95 – 1.64) 1.20 (0.82 – 1.74) 1.27 (0.80 – 2.01) 1.76 (0.88 – 3.54) Multivariate RR (95% CI) 1.16 (0.88 – 1.54) 1.02 (0.69 – 1.49) 0.98 (0.61 – 1.58) 1.10 (0.52 – 2.35) Aspirin use was defined as aspirin use two or more times per week. Users consistently reported regular aspirin use (1986; 1986 and 1988; 1986, 1988, and 1990; 1986, 1988, 1990, and 1992); nonusers consistently reported no aspirin use during the specified time periods. Separate proportional hazards models were analysed for each of the four c d follow-up periods. Proportional hazards models adjusted for current age. Proportional hazards model adjusted for current age, age at started to smoke regularly, and smoking status (includes dose and time since quitting). similar except for multivitamin and supplement use, reflecting DISCUSSION the tendency of aspirin users to also take supplements. Family In this prospective study among male health professionals, twice history of lung cancer was similar among users and nonusers of or more weekly aspirin use was not associated with risk of total aspirin. lung cancer or NSCLC, even for long-term consistent users. Also, Overall, we observed no significant association between aspirin no dose–response association was observed for frequency of use at baseline and total lung cancer incidence (Table 2). Age- aspirin use and lung cancer incidence. adjusted associations were attenuated after adjustment for age at Several epidemiologic studies have reported an increased starting to smoke regularly and smoking status but did not incidence of lung cancer associated with several inflammatory- change appreciably after additional inclusion of the other related lung diseases including asthma and chronic bronchitis, covariates. The association between duration of aspirin use and suggesting that inflammation may play an important etiologic role lung cancer risk became slightly inverse with evidence of more (Wu et al, 1995; Mayne et al, 1999; Brownson and Alavanja, 2000; consistent aspirin use (consecutive RRs of 1.13, 0.98, 0.88, and Brenner et al, 2001). The reduction of inflammatory prostaglan- 0.89), but the results were not statistically significant. After dins by inhibition of COX-2 by NSAIDs, including aspirin, might adjustment for current age, age at starting to smoke regularly, have a chemopreventive effect. However, in a recent study of and smoking status the RR of total lung cancer for users of hydroxytoluene-induced early lung adenocarcinoma in mice, acetaminophen and other NSAIDs separately at baseline compared aspirin attenuated pulmonary inflammation but was ineffective to nonusers was 1.26 (95% CI¼ 0.81–1.96) and 1.07 (95% at preventing lung tumorigenesis (Kisley et al, 2002). In other CI¼ 0.69–1.66), respectively (N¼ 23 cases among users for both animal studies, COX-2 was also consistently expressed in normal compounds). bronchoalveolar and alveolar epithelium of the lung (Bauer et al, Previous studies (Akhmedkhanov et al, 2002; Moysich et al, 2000; Wardlaw et al, 2000), casting some doubt on a specific role of 2002) have examined the association between aspirin use and COX-2 in the development of lung cancer. NSCLC. We limited the analysis to NSCLC (227 cases) and Results from previous clinical and epidemiological studies of compared baseline aspirin use with nonuse of aspirin (Table 2); the aspirin use and the risk of lung cancer have been mixed. An early multivariate risk was 1.16 (95% CI¼ 0.88–1.54). Furthermore, no trial among British physicians found a statistically nonsignificant evidence of an association was observed with increasing consis- 36% lower mortality rate from lung cancer among users of aspirin tency of aspirin use and risk of NSCLC. compared to nonusers (Peto et al, 1988). A prospective study based We also evaluated the association between frequency of aspirin on data from the National Health and Nutrition Examination use, first assessed in 1992, and lung cancer risk, but no dose- Survey found among men a statistically significant 46% lower lung dependent association was observed. The multivariate RRs of lung cancer risk among users of aspirin during the 30-day period cancer for increasing frequency of aspirin use (0–4, 5–21, and 22 preceding recruitment into the cohort compared with nonusers or more days per month) were 1.00, 0.62, and 1.21 (95% CI¼ 0.68– (Schreinemachers and Everson, 1994). Two recent case–control 2.16), respectively (P for trend¼ 0.64). Similar nonsignificant studies (Harris et al, 2002; Moysich et al, 2002) found statistically associations for any given amount of aspirin use were observed for significant reductions in lung cancer risk for aspirin users the non-small-cell types (P for trend¼ 0.40). compared to nonusers. In contrast, some prospective studies No evidence of interaction was observed between categories of reported no overall differences in lung cancer mortality (Thun et al, smoking status and baseline aspirin use for lung cancer risk. After 1993) or incidence (Paganini-Hill et al, 1989) between aspirin users controlling for age, the RR of lung cancer for aspirin users vs and nonusers, consistent with our results. Furthermore, in two nonusers among never smokers was 1.37 (95% CI¼ 0.70–2.69); case–control studies, lung cancer risk was not significantly lower for past smokers the RR was 1.10 (95% CI¼ 0.81–1.50), and for regular aspirin use defined as 4 days per week for 3 months for current smokers the RR was 1.27 (95% CI¼ 0.79, 2.06) compared to nonuse (Rosenberg et al, 1991) or seven prescriptions after additional adjustment for age at starting to smoke of aspirin and other NSAIDs received in 13–36 months before regularly. However, the relatively low number of lung cancer cases diagnosis compared to no prescriptions (Langman et al, 2000). In a within each smoking category (never: 40, past: 179, and current: recent case–control study, a nonsignificant reduction in total lung 104; data missing: five cases) gave limited power to detect cancer risk (odds ratio (OR)¼ 0.66; 95% CI¼ 0.34–1.28, adjusted interactions. & 2003 Cancer Research UK British Journal of Cancer (2003) 89(9), 1705 – 1708 Epidemiology Epidemiology Aspirin use and lung cancer CN Holick et al for smoking (never, past, current) and educational status (attended limitation in the current study, the lack of information on aspirin college, attended graduate school)) was seen among those who dose will result in any potential misclassification of the risk reported aspirin use 3 or more times per week for at least 6 estimate towards the null. Furthermore, the current study has months. Similar reductions in total lung cancer risk were observed strengths regarding the other issues raised. The potential for recall for 5 or more years of aspirin use (OR¼ 0.68; 95% CI¼ 0.31–1.51) or selection bias is greatly minimised, if not eliminated, due to the (Akhmedkhanov et al, 2002). We used a roughly similar exposure prospective study design. To minimise confounding by smoking, definition of aspirin use (i.e. aspirin use two or more times per smoking habits were modeled to best predict lung cancer. The week), and our findings on duration of aspirin use are in fair relative homogeneity of this population of male health profes- agreement. However, we found no significant association between sionals decreases the likelihood of residual confounding by aspirin use at baseline and total lung cancer incidence (RR¼ 1.13; smoking or other factor. With repeated measurements of aspirin 95% CI¼ 0.89–1.43, adjusted for current age, age at starting to use we account for changes in aspirin exposure over time and smoke regularly, and smoking status (includes dose and time since reduce the potential for misclassification of aspirin use. The lack of quitting)); the risk of total lung cancer associated with consistent association observed between aspirin use and lung cancer risk in aspirin use for 6 years was slightly inverse (RR¼ 0.89; 95% the present study might reflect measurement error in aspirin use. CI¼ 0.47–1.67) but the results were not statistically significant. However, using our definition of regular aspirin use, we observed a No association was evident between regular aspirin use and markedly decreased risk of colorectal cancer and adenoma in this NSCLC in the current study. To date, two case–control studies same population (Giovannucci et al, 1994), consistent with other have examined the association between regular aspirin use and studies. This suggests that we are able to detect important NSCLC. In one study, results for NSCLC among women taking associations with aspirin use in this study. We were unable to aspirin three or more times per week for at least 6 months assess the long-term use of acetaminophen and other NSAIDs due (adjusted OR¼ 0.39; 95% CI¼ 0.16–0.96) were stronger that those to the insufficient number of cases among users of these when all histological types combined were considered (adjusted compounds (range of cases among users¼ 0–8). OR¼ 0.66; 95% CI¼ 0.34–1.28) (Akhmedkhanov et al, 2002). In Our findings do not support the hypothesis that regular aspirin contrast, a second study observed similar significant risk use is associated with a decreased risk of lung cancer. Prevention reductions for NSCLC among men and women who were regular of lung cancer should be practiced primarily through smoking users of aspirin (adjusted OR¼ 0.62; 95% CI¼ 0.45–0.86) and prevention and cessation. total lung cancer risk (adjusted OR¼ 0.57; 95% CI¼ 0.41–0.78) (Moysich et al, 2002). We note a number of potential limitations in the epidemiologic ACKNOWLEDGEMENTS studies evaluating the role of aspirin and other NSAIDs in lung cancer risk, including (1) recall bias and selection bias potential in Supported by research Grant CA55075 from the National case–control studies; (2) inefficient control for smoking history; Institutes of Health, Department of Health and Human Services. (3) small numbers of cases; and (4) misclassification due to the We are indebted to Barbara Vericker, Jill Arnold, Elizabeth Frost- insufficient assessment of long-term aspirin use or lack of detailed Hawes, Mira Kaufman, Stacey DeCaro, and Mildred Wolff for information on aspirin dose. Although we acknowledge the dosage expert help. REFERENCES Akhmedkhanov A, Toniolo P, Zeleniuch-Jacquotte A, Koenig KL, Shore RE Paganini-Hill A, Chao A, Ross RK, Henderson BE (1989) Aspirin use and (2002) Aspirin and lung cancer in women. Br J Cancer 87: 49–53 chronic diseases: a cohort study of the elderly. BMJ 299: 1247–1250 Bauer AK, Dwyer-Nield LD, Malkinson AM (2000) High cyclooxygenase 1 Peto R, Gray R, Collins R, Wheatley K, Hennekens C, Jamrozik K, Warlow (COX-1) and cyclooxygenase 2 (COX-2) contents in mouse lung tumors. C, Hafner B, Thompson E, Norton S, Gilliland Doll R (1988) Randomised Carcinogenesis 21: 543–550 trial of prophylactic daily aspirin in British male doctors. Br Med J (Clin Brenner AV, Wang Z, Kleinerman RA, Wang L, Zhang S, Metayer C, Res Ed) 296: 313 –316 Chen K, Lei S, Cui H, Lubin JH (2001) Previous pulmonary diseases Rioux N, Castonguay A (1998) Prevention of NNK-induced lung and risk of lung cancer in Gansu Province, China. Int J Epidemiol 30: tumorigenesis in A/J mice by acetylsalicylic acid and NS-398. Cancer 118–124 Res 58: 5354 –5360 Brownson RC, Alavanja MC (2000) Previous lung disease and lung cancer Rosenberg L, Palmer JR, Zauber AG, Warshauer ME, Stolley PD, Shapiro S risk among women (United States). Cancer Causes Control 11: 853–858 (1991) A hypothesis: nonsteroidal anti-inflammatory drugs reduce the Duperron C, Castonguay A (1997) Chemopreventive efficacies of aspirin incidence of large-bowel cancer. J Natl Cancer Inst 83: 355–358 and sulindac against lung tumorigenesis in A/J mice. Carcinogenesis 18: Schreinemachers DM, Everson RB (1994) Aspirin use and lung, colon, and 1001– 1006 breast cancer incidence in a prospective study. Epidemiology 5: 138–146 Giovannucci E, Rimm EB, Stampfer MJ, Colditz GA, Ascherio A, Willett Stampfer MJ, Willett WC, Speizer FE, Dysert DC, Lipnick R, Rosner B, WC (1994) Aspirin use and the risk for colorectal cancer and adenoma in Hennekens CH (1984) Test of the National Death Index. Am J Epidemiol male health professionals. Ann Intern Med 121: 241–246 119: 837–839 Harris RE, Beebe-Donk J, Schuller HM (2002) Chemoprevention of lung Thun MJ, Namboodiri MM, Calle EE, Flanders WD, Heath Jr CW (1993) cancer by non-steroidal anti-inflammatory drugs among cigarette Aspirin use and risk of fatal cancer. Cancer Res 53: 1322–1327 smokers. Oncol Rep 9: 693–695 Wardlaw SA, March TH, Belinsky SA (2000) Cyclooxygenase-2 expression Kisley LR, Barrett BS, Dwyer-Nield LD, Bauer AK, Thompson DC, is abundant in alveolar type II cells in lung cancer-sensitive mouse Malkinson AM (2002) Celecoxib reduces pulmonary inflammation but strains and in premalignant lesions. Carcinogenesis 21: 1371– 1377 not lung tumorigenesis in mice. Carcinogenesis 23: 1653–1660 Wu AH, Fontham ET, Reynolds P, Greenberg RS, Buffler P, Liff J, Boyd P, Langman MJ, Cheng KK, Gilman EA, Lancashire RJ (2000) Effect of anti- Henderson BE, Correa P (1995) Previous lung disease and risk of lung inflammatory drugs on overall risk of common cancer: case–control cancer among lifetime nonsmoking women in the United States. Am J study in general practice research database. BMJ 320: 1642–1646 Epidemiol 141: 1023–1032 Mayne ST, Buenconsejo J, Janerich DT (1999) Previous lung disease and Xu XC (2002) COX-2 inhibitors in cancer treatment and prevention, a risk of lung cancer among men and women nonsmokers. Am J Epidemiol recent development. Anticancer Drugs 13: 127–137 149: 13–20 Yao R, Rioux N, Castonguay A, You M (2000) Inhibition of COX-2 and Moysich KB, Menezes RJ, Ronsani A, Swede H, Reid ME, Cummings K, induction of apoptosis: two determinants of nonsteroidal anti-inflam- Falkner KL, Loewen GM, Bepler G (2002) Regular aspirin use and lung matory drugs’ chemopreventive efficacies in mouse lung tumorigenesis. cancer risk. BMC Cancer 2: 31 Exp Lung Res 26: 731–742 British Journal of Cancer (2003) 89(9), 1705 – 1708 & 2003 Cancer Research UK http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png British Journal of Cancer Springer Journals

Aspirin use and lung cancer in men

Download PDF
4 pages

Loading next page...
 
/lp/springer-journals/aspirin-use-and-lung-cancer-in-men-zPNc41CbTx

References (44)

Publisher
Springer Journals
Copyright
Copyright © 2003 by The Author(s)
Subject
Biomedicine; Biomedicine, general; Cancer Research; Epidemiology; Molecular Medicine; Oncology; Drug Resistance
ISSN
0007-0920
eISSN
1532-1827
DOI
10.1038/sj.bjc.6601343
Publisher site
See Article on Publisher Site

Abstract

British Journal of Cancer (2003) 89, 1705 – 1708 & 2003 Cancer Research UK All rights reserved 0007 – 0920/03 $25.00 www.bjcancer.com ,1,2 3 3 1,2,4,5 1,2,4,5 CN Holick , DS Michaud , MF Leitzmann , WC Willett and E Giovannucci 1 2 Department of Nutrition, Harvard School of Public Health, 677 Huntington Avenue, Boston, MA 02115, USA; Department of Epidemiology, Harvard School of Public Health, 677 Huntington Avenue, Boston, MA 02115, USA; Nutritional Epidemiology Branch, Division of Cancer Epidemiology and Genetics, National Cancer Institute, Bethesda, MD 20892, USA; Channing Laboratory, Department of Medicine, Brigham and Women’s Hospital, Boston, MA 02115, USA; Harvard Medical School, Boston, MA 02115, USA We examined prospectively the relation between regular aspirin use and lung cancer risk in the Health Professionals Follow-Up Study. Of 49 383 US men aged 40–75 years who completed biennial self-administered questionnaires that assessed aspirin use beginning in 1986, 328 developed lung cancer during 601 453 person-years of follow-up through 31 December 2000. No information on aspirin dose was available. Controlling for current age, smoking status, and age at starting to smoke regularly, the relative risk (RR) of total lung cancer for regular users of aspirin (twice or more per week) at baseline compared to nonusers was 1.13 (95% confidence interval (CI) ¼ 0.89–1.43). Results were similar for non-small-cell lung cancer (RR¼ 1.16; 95% CI¼ 0.88–1.54). No apparent dose-dependent association was observed for the frequency of aspirin use and lung cancer risk (P for trend¼ 0.64), and results remained null when consistent use of aspirin over time was examined. These findings do not suggest that regular aspirin use is associated with a reduced lung cancer risk. British Journal of Cancer (2003) 89, 1705 – 1708. doi:10.1038/sj.bjc.6601343 www.bjcancer.com & 2003 Cancer Research UK Keywords: aspirin; lung cancer; chemoprevention; prospective studies; epidemiology Evidence from animal studies suggests that aspirin and possibly MATERIALS AND METHODS other nonsteroidal anti-inflammatory drugs (NSAIDs) including Study population cyclooxygenase-2 (COX-2) inhibitors may influence the develop- ment and progression of lung cancer (Duperron and Castonguay, The Health Professionals Follow-Up Study (HPFS) is an ongoing 1997; Rioux and Castonguay, 1998; Yao et al, 2000). As COX-2 prospective cohort study of 51 529 US male dentists, optometrists, enzymes are constitutively expressed in lung neoplastic tissue (Xu, osteopaths, podiatrists, pharmacists, and veterinarians aged 40–75 2002), reduced synthesis of prostaglandins from arachidonic acid years. The primary objective of the study is to evaluate a series of by inhibition of COX-2 enzymes may have preventive or hypotheses relating nutritional and lifestyle factors to the therapeutic effects in lung carcinogenesis. incidence of cancer, heart disease, and other major diseases. At Clinical and epidemiological studies, however, have not baseline in 1986, enrollees returned a mailed questionnaire that consistently supported for the hypothesis that regular use of assessed information on lifestyle factors, aspirin and other NSAID aspirin and other NSAIDs reduces lung cancer incidence but are use, diet, and medical and smoking histories. Follow-up ques- limited in several key areas. These include residual confounding tionnaires are mailed biennially to the entire cohort to update due to broad categories of smoking habits (Paganini-Hill et al, information on potential risk factors and ascertain new cases of 1989; Schreinemachers and Everson, 1994; Langman et al, 2000; disease. Akhmedkhanov et al, 2002; Moysich et al, 2002); misclassification due to a single assessment of aspirin use (Paganini-Hill et al, 1989; Schreinemachers and Everson, 1994; Harris et al., 2002); and small Assessment of aspirin use number of lung cancer cases observed (Peto et al, 1988; Paganini- Regular use of aspirin (e.g. Anacin, Bufferin, Alka-Seltzer), Hill et al, 1989; Schreinemachers and Everson, 1994; Akhmedkha- acetaminophen (e.g. Tylenol), and other NSAIDs (e.g. Motrin, nov et al, 2002). Indocin, Naprosyn, Dolobid) separately defined as two or more Given the limitations of previous literature and uncertainty times per week was ascertained in 1986 and updated every 2 years regarding the protection of aspirin against lung cancer risk, we thereafter. More detailed questions on aspirin use began in the examined the association between regular use of aspirin and lung 1992 questionnaire. Frequency of aspirin use was assessed by the cancer incidence in a large cohort of US. male health professionals average number of days each month aspirin was taken (none, 1–4, with long-term aspirin use and detailed information on smoking 5–14, 15–21, or 22 or more). Information on aspirin dose was not history. available. Reasons for aspirin use were assessed in 1993 by a supplemen- tary questionnaire sent to a random sample of 211 participants, *Correspondence: CN Holick; E-mail: [email protected] who reported taking aspirin from 1986 to 1990. Of the 211 men, Received 19 June 2003; revised 27 August 2003; accepted 1 September 185 (88%) responded and reported one or more of the following Epidemiology Epidemiology Aspirin use and lung cancer CN Holick et al reasons: cardiovascular disease, 25%; to decrease risk for carcinoma). Potential interaction by smoking status category cardiovascular disease, 58%; headache, 25%; joint or musculoske- (never, past, and current separately) and baseline aspirin use was letal pain, 33%; and other or unknown reasons, 7%. We were assessed using a cross-product term in multivariate models. Tests unable to examine long-term use of acetaminophen and other of linear trends for increasing categories of aspirin use were NSAIDs separately because of the small number of men who used conducted by assigning the median value of aspirin use for them regularly (approximately 6% of the cohort). categories and treating these as a single continuous variable. Reported P-values are based on two-sided tests. Identification of lung cancer cases On each biennial questionnaire participants were asked whether RESULTS they had been diagnosed with lung cancer during the previous 2 years. The follow-up rate with respect to the incidence of cancer At baseline in 1986, about one-third of men reported regular was 96% of the total possible person-years. After receiving aspirin use defined as two or more tablets per week (Table 1). permission from identified cases (or next of kin for decedents), Aspirin users tended to be slightly older, to have smoked, and to hospital records and pathology reports were obtained and have started smoking at an earlier age. Dietary patterns were very reviewed by a physician for histological confirmation. Information on lung cancer cell type was available for approximately 80%. Deaths in the cohort were ascertained through family members Table 1 Characteristics (means and proportions) of the HPFS cohort and the National Death Index (Stampfer et al, 1984). Cases by aspirin use at baseline in 1986 diagnosed between date of 1986 questionnaire return and 31 December 2000 were included in this report (n¼ 328). Aspirin Characteristic Nonuse Use Statistical analysis Follow-up for each participant was calculated from the date of Participants (n) 34 841 14 542 Age (years) 53.7 56.4 return of the 1986 questionnaire until the date of lung cancer Height (in) 70.0 70.1 diagnosis, date of death from any cause, or 31 December 2000, Body mass index (kg m ) 24.9 25.1 which ever came first. At baseline in 1986, we excluded 2069 men who reported a history of cancer other than nonmelanoma skin Smoking history cancer by 1986, 40 men who incorrectly completed the ques- Smoking status (%) tionnaire, and 37 men who later wished to be removed from the Never 45.8 41.9 cohort. Only those with complete information on aspirin use at Past (years) baseline were included in the analyses. The analytic cohort o10 11.4 14.9 consisted of 49 383 men representing 601 453 person-years of 10+ 27.7 31.9 Current (cigarettes per day) follow-up. Cox proportional hazards models were used to estimate 1 – 4 0.9 1.1 relative risks (RRs) and 95% confidence intervals (CI). 5 – 14 1.5 1.9 To take account for changes in aspirin use over time and best 15 – 24 3.2 3.1 represent long-term use of aspirin, our analyses were conducted 25 – 34 1.4 1.6 using consistent aspirin use. For example, lung cancer from 1986 35 – 44 1.0 1.2 through 2000 was related to aspirin use reported on the 1986 45+ 0.4 0.5 questionnaire, that from 1988 through 2000 to consistent use Age (years) at starting to smoke (%) reported in the 1986 and 1988 questionnaires, that from 1990 o15 5.7 6.9 through 2000 use reported on the 1986, 1988, and 1990 15 – 19 18.9 23.4 20 – 29 20.8 23.6 questionnaires, and that from 1992 through 2000 use reported 30+ 47.6 44.0 on the 1986, 1988, 1990, and 1992 questionnaires. Consistent Years since quitting 8.3 8.1 aspirin users were compared to nonusers (i.e., participants who consistently reported no aspirin use during the same specified time Intakes (per day) periods). Fruit and vegetable 5.3 5.3 Potential confounders were specified a priori based on a review b-Carotene (mg) 5150 5039 of putative risk factors for lung cancer and included age (year); a-Carotene (mg) 949 918 marital status; body mass index (kg m ); use of multivitamins, Lycopene (mg) 10 374 10 321 vitamin A, vitamin C, vitamin E, b-carotene, and selenium Multivitamin use (%) 39.4 50.0 Supplement use (per day) supplements (yes or no); intake of fruit and vegetables; energy- Vitamin C (mg) 57.6 67.7 adjusted a-carotene, b-carotene, lycopene, lutein/zeaxanthin, and Vitamin E (IU) 17.6 22.7 b-cryptoxanthin; alcohol use; and family history of lung cancer. Vitamin A (IU) 7.9 9.3 Smoking history was categorised as the combination of age at Selenium (mg) 6.2 7.5 starting to smoke regularly (less than 15, 15–19, 20–29, and more d b-Carotene (mg) 1.9 2.6 than 30 years old) and smoking status, which included current smoking (1–4, 5–14, 15–24, 25–34, 35–44, and 45 or more Family history of lung cancer (%) 5.7 6.0 cigarettes per day), past smoking with time since quitting (less Disease history (%) than 10 years and 10 or more years), and never smoked. Asthma 5.2 5.4 Emphysema or chronic bronchitis 1.4 1.6 Multivariate models included current age, age at starting to smoke Rheumatoid arthritis 2.0 3.1 regularly, and smoking status (included time since quitting and Other arthritis (e.g. degenerative) 7.3 12.2 dose), plus the additional confounders mentioned previously that a b were assessed by evaluating whether their inclusion into the Standardised to the age distribution of the study population. Aspirin use was multivariate model changed the risk estimate by more than 10%. defined as aspirin use two or more times per week. Values do not add up to 100% d e Additional analyses were restricted to non-small-cell lung cancer because of missing values. Use of supplement on a regular basis. Data from the 1996 follow-up questionnaire. (NSCLC) (squamous carcinoma, adenocarcinoma, and large cell British Journal of Cancer (2003) 89(9), 1705 – 1708 & 2003 Cancer Research UK Aspirin use and lung cancer CN Holick et al Table 2 Relative risk of lung cancer by aspirin use in the HPFS, 1986– 2000 Aspirin use Variable 1986 1986+1988 1986+1988+1990 1986+1988+1990+1992 Follow-up period 1986 – 2000 1988 – 2000 1990 – 2000 1992 – 2000 Total cases Nonusers, cases/person-years 204/428 688 96/202 457 66/126 065 22/44 436 Users, cases/person-years 124/172 765 64/93 431 41/52 730 28/37 478 Age-adjusted RR (95% CI) 1.19 (0.95 – 1.49) 1.11 (0.80 – 1.53) 1.07 (0.72 – 1.60) 1.18 (0.66 – 2.10) Multivariate RR (95% CI) 1.13 (0.89 – 1.43) 0.98 (0.70 – 1.36) 0.88 (0.58 – 1.34) 0.89 (0.47 – 1.67) Non-small-cell carcinomas Nonusers, cases/person-years 140/428 717 69/202 466 45/126 073 13/44 441 Users, cases/person-years 87/172 789 49/93 441 33/52 733 24/37 479 Age-adjusted RR (95% CI) 1.25 (0.95 – 1.64) 1.20 (0.82 – 1.74) 1.27 (0.80 – 2.01) 1.76 (0.88 – 3.54) Multivariate RR (95% CI) 1.16 (0.88 – 1.54) 1.02 (0.69 – 1.49) 0.98 (0.61 – 1.58) 1.10 (0.52 – 2.35) Aspirin use was defined as aspirin use two or more times per week. Users consistently reported regular aspirin use (1986; 1986 and 1988; 1986, 1988, and 1990; 1986, 1988, 1990, and 1992); nonusers consistently reported no aspirin use during the specified time periods. Separate proportional hazards models were analysed for each of the four c d follow-up periods. Proportional hazards models adjusted for current age. Proportional hazards model adjusted for current age, age at started to smoke regularly, and smoking status (includes dose and time since quitting). similar except for multivitamin and supplement use, reflecting DISCUSSION the tendency of aspirin users to also take supplements. Family In this prospective study among male health professionals, twice history of lung cancer was similar among users and nonusers of or more weekly aspirin use was not associated with risk of total aspirin. lung cancer or NSCLC, even for long-term consistent users. Also, Overall, we observed no significant association between aspirin no dose–response association was observed for frequency of use at baseline and total lung cancer incidence (Table 2). Age- aspirin use and lung cancer incidence. adjusted associations were attenuated after adjustment for age at Several epidemiologic studies have reported an increased starting to smoke regularly and smoking status but did not incidence of lung cancer associated with several inflammatory- change appreciably after additional inclusion of the other related lung diseases including asthma and chronic bronchitis, covariates. The association between duration of aspirin use and suggesting that inflammation may play an important etiologic role lung cancer risk became slightly inverse with evidence of more (Wu et al, 1995; Mayne et al, 1999; Brownson and Alavanja, 2000; consistent aspirin use (consecutive RRs of 1.13, 0.98, 0.88, and Brenner et al, 2001). The reduction of inflammatory prostaglan- 0.89), but the results were not statistically significant. After dins by inhibition of COX-2 by NSAIDs, including aspirin, might adjustment for current age, age at starting to smoke regularly, have a chemopreventive effect. However, in a recent study of and smoking status the RR of total lung cancer for users of hydroxytoluene-induced early lung adenocarcinoma in mice, acetaminophen and other NSAIDs separately at baseline compared aspirin attenuated pulmonary inflammation but was ineffective to nonusers was 1.26 (95% CI¼ 0.81–1.96) and 1.07 (95% at preventing lung tumorigenesis (Kisley et al, 2002). In other CI¼ 0.69–1.66), respectively (N¼ 23 cases among users for both animal studies, COX-2 was also consistently expressed in normal compounds). bronchoalveolar and alveolar epithelium of the lung (Bauer et al, Previous studies (Akhmedkhanov et al, 2002; Moysich et al, 2000; Wardlaw et al, 2000), casting some doubt on a specific role of 2002) have examined the association between aspirin use and COX-2 in the development of lung cancer. NSCLC. We limited the analysis to NSCLC (227 cases) and Results from previous clinical and epidemiological studies of compared baseline aspirin use with nonuse of aspirin (Table 2); the aspirin use and the risk of lung cancer have been mixed. An early multivariate risk was 1.16 (95% CI¼ 0.88–1.54). Furthermore, no trial among British physicians found a statistically nonsignificant evidence of an association was observed with increasing consis- 36% lower mortality rate from lung cancer among users of aspirin tency of aspirin use and risk of NSCLC. compared to nonusers (Peto et al, 1988). A prospective study based We also evaluated the association between frequency of aspirin on data from the National Health and Nutrition Examination use, first assessed in 1992, and lung cancer risk, but no dose- Survey found among men a statistically significant 46% lower lung dependent association was observed. The multivariate RRs of lung cancer risk among users of aspirin during the 30-day period cancer for increasing frequency of aspirin use (0–4, 5–21, and 22 preceding recruitment into the cohort compared with nonusers or more days per month) were 1.00, 0.62, and 1.21 (95% CI¼ 0.68– (Schreinemachers and Everson, 1994). Two recent case–control 2.16), respectively (P for trend¼ 0.64). Similar nonsignificant studies (Harris et al, 2002; Moysich et al, 2002) found statistically associations for any given amount of aspirin use were observed for significant reductions in lung cancer risk for aspirin users the non-small-cell types (P for trend¼ 0.40). compared to nonusers. In contrast, some prospective studies No evidence of interaction was observed between categories of reported no overall differences in lung cancer mortality (Thun et al, smoking status and baseline aspirin use for lung cancer risk. After 1993) or incidence (Paganini-Hill et al, 1989) between aspirin users controlling for age, the RR of lung cancer for aspirin users vs and nonusers, consistent with our results. Furthermore, in two nonusers among never smokers was 1.37 (95% CI¼ 0.70–2.69); case–control studies, lung cancer risk was not significantly lower for past smokers the RR was 1.10 (95% CI¼ 0.81–1.50), and for regular aspirin use defined as 4 days per week for 3 months for current smokers the RR was 1.27 (95% CI¼ 0.79, 2.06) compared to nonuse (Rosenberg et al, 1991) or seven prescriptions after additional adjustment for age at starting to smoke of aspirin and other NSAIDs received in 13–36 months before regularly. However, the relatively low number of lung cancer cases diagnosis compared to no prescriptions (Langman et al, 2000). In a within each smoking category (never: 40, past: 179, and current: recent case–control study, a nonsignificant reduction in total lung 104; data missing: five cases) gave limited power to detect cancer risk (odds ratio (OR)¼ 0.66; 95% CI¼ 0.34–1.28, adjusted interactions. & 2003 Cancer Research UK British Journal of Cancer (2003) 89(9), 1705 – 1708 Epidemiology Epidemiology Aspirin use and lung cancer CN Holick et al for smoking (never, past, current) and educational status (attended limitation in the current study, the lack of information on aspirin college, attended graduate school)) was seen among those who dose will result in any potential misclassification of the risk reported aspirin use 3 or more times per week for at least 6 estimate towards the null. Furthermore, the current study has months. Similar reductions in total lung cancer risk were observed strengths regarding the other issues raised. The potential for recall for 5 or more years of aspirin use (OR¼ 0.68; 95% CI¼ 0.31–1.51) or selection bias is greatly minimised, if not eliminated, due to the (Akhmedkhanov et al, 2002). We used a roughly similar exposure prospective study design. To minimise confounding by smoking, definition of aspirin use (i.e. aspirin use two or more times per smoking habits were modeled to best predict lung cancer. The week), and our findings on duration of aspirin use are in fair relative homogeneity of this population of male health profes- agreement. However, we found no significant association between sionals decreases the likelihood of residual confounding by aspirin use at baseline and total lung cancer incidence (RR¼ 1.13; smoking or other factor. With repeated measurements of aspirin 95% CI¼ 0.89–1.43, adjusted for current age, age at starting to use we account for changes in aspirin exposure over time and smoke regularly, and smoking status (includes dose and time since reduce the potential for misclassification of aspirin use. The lack of quitting)); the risk of total lung cancer associated with consistent association observed between aspirin use and lung cancer risk in aspirin use for 6 years was slightly inverse (RR¼ 0.89; 95% the present study might reflect measurement error in aspirin use. CI¼ 0.47–1.67) but the results were not statistically significant. However, using our definition of regular aspirin use, we observed a No association was evident between regular aspirin use and markedly decreased risk of colorectal cancer and adenoma in this NSCLC in the current study. To date, two case–control studies same population (Giovannucci et al, 1994), consistent with other have examined the association between regular aspirin use and studies. This suggests that we are able to detect important NSCLC. In one study, results for NSCLC among women taking associations with aspirin use in this study. We were unable to aspirin three or more times per week for at least 6 months assess the long-term use of acetaminophen and other NSAIDs due (adjusted OR¼ 0.39; 95% CI¼ 0.16–0.96) were stronger that those to the insufficient number of cases among users of these when all histological types combined were considered (adjusted compounds (range of cases among users¼ 0–8). OR¼ 0.66; 95% CI¼ 0.34–1.28) (Akhmedkhanov et al, 2002). In Our findings do not support the hypothesis that regular aspirin contrast, a second study observed similar significant risk use is associated with a decreased risk of lung cancer. Prevention reductions for NSCLC among men and women who were regular of lung cancer should be practiced primarily through smoking users of aspirin (adjusted OR¼ 0.62; 95% CI¼ 0.45–0.86) and prevention and cessation. total lung cancer risk (adjusted OR¼ 0.57; 95% CI¼ 0.41–0.78) (Moysich et al, 2002). We note a number of potential limitations in the epidemiologic ACKNOWLEDGEMENTS studies evaluating the role of aspirin and other NSAIDs in lung cancer risk, including (1) recall bias and selection bias potential in Supported by research Grant CA55075 from the National case–control studies; (2) inefficient control for smoking history; Institutes of Health, Department of Health and Human Services. (3) small numbers of cases; and (4) misclassification due to the We are indebted to Barbara Vericker, Jill Arnold, Elizabeth Frost- insufficient assessment of long-term aspirin use or lack of detailed Hawes, Mira Kaufman, Stacey DeCaro, and Mildred Wolff for information on aspirin dose. Although we acknowledge the dosage expert help. REFERENCES Akhmedkhanov A, Toniolo P, Zeleniuch-Jacquotte A, Koenig KL, Shore RE Paganini-Hill A, Chao A, Ross RK, Henderson BE (1989) Aspirin use and (2002) Aspirin and lung cancer in women. Br J Cancer 87: 49–53 chronic diseases: a cohort study of the elderly. BMJ 299: 1247–1250 Bauer AK, Dwyer-Nield LD, Malkinson AM (2000) High cyclooxygenase 1 Peto R, Gray R, Collins R, Wheatley K, Hennekens C, Jamrozik K, Warlow (COX-1) and cyclooxygenase 2 (COX-2) contents in mouse lung tumors. C, Hafner B, Thompson E, Norton S, Gilliland Doll R (1988) Randomised Carcinogenesis 21: 543–550 trial of prophylactic daily aspirin in British male doctors. Br Med J (Clin Brenner AV, Wang Z, Kleinerman RA, Wang L, Zhang S, Metayer C, Res Ed) 296: 313 –316 Chen K, Lei S, Cui H, Lubin JH (2001) Previous pulmonary diseases Rioux N, Castonguay A (1998) Prevention of NNK-induced lung and risk of lung cancer in Gansu Province, China. Int J Epidemiol 30: tumorigenesis in A/J mice by acetylsalicylic acid and NS-398. Cancer 118–124 Res 58: 5354 –5360 Brownson RC, Alavanja MC (2000) Previous lung disease and lung cancer Rosenberg L, Palmer JR, Zauber AG, Warshauer ME, Stolley PD, Shapiro S risk among women (United States). Cancer Causes Control 11: 853–858 (1991) A hypothesis: nonsteroidal anti-inflammatory drugs reduce the Duperron C, Castonguay A (1997) Chemopreventive efficacies of aspirin incidence of large-bowel cancer. J Natl Cancer Inst 83: 355–358 and sulindac against lung tumorigenesis in A/J mice. Carcinogenesis 18: Schreinemachers DM, Everson RB (1994) Aspirin use and lung, colon, and 1001– 1006 breast cancer incidence in a prospective study. Epidemiology 5: 138–146 Giovannucci E, Rimm EB, Stampfer MJ, Colditz GA, Ascherio A, Willett Stampfer MJ, Willett WC, Speizer FE, Dysert DC, Lipnick R, Rosner B, WC (1994) Aspirin use and the risk for colorectal cancer and adenoma in Hennekens CH (1984) Test of the National Death Index. Am J Epidemiol male health professionals. Ann Intern Med 121: 241–246 119: 837–839 Harris RE, Beebe-Donk J, Schuller HM (2002) Chemoprevention of lung Thun MJ, Namboodiri MM, Calle EE, Flanders WD, Heath Jr CW (1993) cancer by non-steroidal anti-inflammatory drugs among cigarette Aspirin use and risk of fatal cancer. Cancer Res 53: 1322–1327 smokers. Oncol Rep 9: 693–695 Wardlaw SA, March TH, Belinsky SA (2000) Cyclooxygenase-2 expression Kisley LR, Barrett BS, Dwyer-Nield LD, Bauer AK, Thompson DC, is abundant in alveolar type II cells in lung cancer-sensitive mouse Malkinson AM (2002) Celecoxib reduces pulmonary inflammation but strains and in premalignant lesions. Carcinogenesis 21: 1371– 1377 not lung tumorigenesis in mice. Carcinogenesis 23: 1653–1660 Wu AH, Fontham ET, Reynolds P, Greenberg RS, Buffler P, Liff J, Boyd P, Langman MJ, Cheng KK, Gilman EA, Lancashire RJ (2000) Effect of anti- Henderson BE, Correa P (1995) Previous lung disease and risk of lung inflammatory drugs on overall risk of common cancer: case–control cancer among lifetime nonsmoking women in the United States. Am J study in general practice research database. BMJ 320: 1642–1646 Epidemiol 141: 1023–1032 Mayne ST, Buenconsejo J, Janerich DT (1999) Previous lung disease and Xu XC (2002) COX-2 inhibitors in cancer treatment and prevention, a risk of lung cancer among men and women nonsmokers. Am J Epidemiol recent development. Anticancer Drugs 13: 127–137 149: 13–20 Yao R, Rioux N, Castonguay A, You M (2000) Inhibition of COX-2 and Moysich KB, Menezes RJ, Ronsani A, Swede H, Reid ME, Cummings K, induction of apoptosis: two determinants of nonsteroidal anti-inflam- Falkner KL, Loewen GM, Bepler G (2002) Regular aspirin use and lung matory drugs’ chemopreventive efficacies in mouse lung tumorigenesis. cancer risk. BMC Cancer 2: 31 Exp Lung Res 26: 731–742 British Journal of Cancer (2003) 89(9), 1705 – 1708 & 2003 Cancer Research UK

Journal

British Journal of Cancer – Springer Journals

Published: Oct 28, 2003

There are no references for this article.