Pharmacy Division Ramathibodi Hospital

Home
Pharmacist
About
ÃͺÃÙéàÃ×èͧÂÒ
¤Ø¡ѹàÃ×èͧÂÒ


Association Between Acetaminophen or Nonsteroidal Antiinflammatory Drugs and Risk of Developing Ovarian, Breast, or Colon Cancer

Christoph R. Meier, Ph.D., M.Sc., Stephen Schmitz, M.D., M.P.H., Hershel Jick, M.D.

Pharmacotherapy 22(3):303-309, 2002. © 2002 Pharmacotherapy Publications

Abstract and Introduction

Abstract

Study Objective. To explore the association between exposure to acetaminophen (paracetamol) or nonsteroidal antiinflammatory drugs (NSAIDs) and risk of developing ovarian, breast, or colon cancer.
Design. Retrospective case-control study.
Setting. General practice offices.
Subjects. Four hundred eighty-three women with ovarian cancer and 1877 women matched for age, years of medical history in computer record, general practice attended, and calendar time; 3706 women with breast cancer and 14,155 matched control subjects; and 635 women with colon cancer and 2434 matched control subjects.
Intervention. United Kingdom-based General Practice Research Database was searched for women aged 50-89 years with a first-time diagnosis of ovarian, breast, or colon cancer and for matched controls to assess prescription analgesic exposure.
Measurements and Main Results. Regular acetaminophen exposure (>/= 30 prescriptions) was associated with a slightly decreased risk of developing breast (odds ratio [OR] 0.8, 95% confidence interval [CI] 0.7-1.0) but not ovarian (OR 1.0, 95% CI 0.6-1.5) or colon (OR 1.0, 95% CI 0.7-1.4) cancer. Regular NSAID exposure was associated with a reduced risk of colon (OR 0.5, 95% CI 0.3-0.9) but not ovarian or breast cancer.
Conclusion. We found no evidence for a decreased ovarian cancer risk for women with regular acetaminophen or NSAID exposure.

Introduction

Acetaminophen (paracetamol), a common analgesic, was associated with a decreased risk of developing ovarian cancer in a recent epidemiologic study.[1] In this case-control analysis of 563 women with incident ovarian cancer and 523 age- and area-matched control subjects, the odds ratio (OR) for subjects taking acetaminophen at least once/week for a continuous period of at least 6 months (based on interview information) was 0.52 (95% confidence interval [CI] 0.31-0.88), compared with those who did not take acetaminophen.[1] A United States-based cohort study yielded additional evidence for a possible association between longer term exposure to acetaminophen and a reduced risk of developing ovarian cancer (relative risk 0.55, 95% CI 0.27-1.09).[2] However, the results of a recent large case-control analysis of 780 ovarian cancer cases did not support these findings; no evidence was found for a negative association between regular acetaminophen intake and ovarian cancer (OR 1.0, 95% CI 0.6-1.5).[3]

Breast cancer predominantly affects women, and as in ovarian cancer, a certain etiologic role of female hormones (particularly estrogens) in breast cancer pathophysiology has been shown.[4] Although the association between acetaminophen exposure and the risk of developing breast cancer to our knowledge has not been investigated, studies in animals and humans reported controversial results about possible effects of nonsteroidal antiinflammatory drugs (NSAIDs) on the risk of breast cancer.[5-11]

Evidence documents a negative association between long-term exposure to aspirin or nonaspirin NSAIDs and the risk of developing colorectal cancer.[12-22] The proposed mechanism for the preventive effect of NSAIDs is the inhibition of cyclooxygenase-2, an enzyme that is upregulated in colorectal cancer.[23] Acetaminophen, which does not or only marginally inhibits the cyclooxygenase enzyme complex, has not been associated with a reduced risk of developing colon cancer.[20,21]

We explored a possible association between acetaminophen or NSAID exposure and the risk of developing ovarian, breast, or colon cancer.

Methods

Data Source

We conducted three separate case-control analyses of data for women with incident ovarian, breast, or colon cancer. Data were derived from the large United Kingdom-based General Practice Research Database (GPRD). General practitioners from 350 general practices record medical information on over 3 million patients in a standard manner and supply it anonymously to provide data for research purposes. The computer records contain patient demographics, characteristics (i.e., height, weight, smoking status), symptoms and diagnoses (by Oxford Medical Information System [OXMIS] codes, which are mapped onto International Classification of Diseases [ICD] codes), referrals, hospitalizations, and drug prescriptions in chronologic order. The GPRD is described in detail in the literature.[24-26] The accuracy and comprehensiveness of diagnoses and drug prescriptions are validated and docu-mented,[27,28] and there is particular experience with studies on the GPRD involving patients with cancer diagnoses.[29,30] The United Kingdom Medicines Control Agency administers the database.

Case Definition

We identified all women with an incident diagnosis of ovarian cancer (ICD code 183.0), breast cancer (ICD code 174.0), or colon cancer (ICD code 153.8) from January 1, 1992-September 30, 1997. We restricted the study to women aged 50-89 years at the date of the first-time cancer diagnosis (subsequently referred to as index date) who had a recorded medical history in the GPRD at least 3 years before the index date. We excluded women who already had any other type of malignancy (except nonmelanomatous skin cancer) before the index date.

Case Validation

Previous experience from two large GPRD-based studies documented the high reliability of cancer diagnoses in the GPRD: over 95% of computer-recorded incident cancer diagnoses were validated by review of original medical records.[29,30] Based on this experience, we did not validate the diagnoses for these outcomes by review of original medical documents in our analysis. However, we reviewed the computer records of all potential cases, while all exposure information of interest (i.e., acetaminophen, NSAIDs) was concealed from the reviewer. We included only those cases with substantial evidence for a valid first-time cancer diagnosis, that is, women who were referred and hospitalized, who had general practitioner-recorded comments in their medical record such as "malignant" or "positive histology," or women whose cancer diagnosis was followed by radiation, chemotherapy, or surgery.

Control Subjects

For each case woman, we randomly selected up to four control women from the base population, matched for age (same year of birth), years of medical history in the computer record, general practice attended, and calendar time (by using the "index date" of the case). We applied the same exclusion criteria to control subjects as to case subjects. All controls of ovarian cancer cases had to have at least one ovary and thus did not have a diagnosis of bilateral oophorectomy before the index date in the computer record.

Exposure Assessment

We assessed exposure to acetaminophen, aspirin, and NSAIDs before the index date for all case and control subjects from the computer records. We assessed acetaminophen in combi-nation with metoclopramide or propoxyphene (combinations that are available only by prescription) or acetaminophen alone or in combination with codeine (also available over-the-counter). Exposure to NSAIDs encompassed ibuprofen, mefenamic acid, indomethacin, tenoxicam, nabumetone, tiaprofenic acid, etodolac, acemetacin, naproxen, diclofenac, ketoprofen, diflunisal, fenoprofen, piroxicam, fenbufen, flurbiprofen, or sulindac.

Women were classified as being nonexposed if they had no prescription for a study drug before the index date or as exposed according to the number of prescriptions recorded. We further classified those exposed by recency of exposure; exposure in women who took acetaminophen or NSAIDs in the year preceding the index date was defined as current exposure, whereas exposure in those who did not was categorized as past exposure. The authors of one study[1] used various exposure categories to distinguish between those taking acetaminophen on a regular long-term basis and those taking acetaminophen only irregularly. In an attempt to similarly characterize acetaminophen exposure in our study population, we identified all women who had three or more prescriptions in any given year before the index date and classified them according to the number of exposed (i.e., years with >/= three prescriptions) or unexposed (i.e., years with < three prescriptions) years. To address possible confounding by indication, we further conducted an analysis in which we did not take into account any exposure of interest in the year before the index date.

Since some acetaminophen, aspirin, or ibuprofen preparations can be purchased over-the-counter in the United Kingdom, general practitioner-recorded exposure in a patient's medical record may not fully represent total exposure to these drugs. To explore to what extent alleged nonexposed women in the GPRD took such preparations over-the-counter, we identified at random 10 large group practices and asked these general practitioners to interview a random sample of 60 female patients (30 aged < 60 yrs, 30 aged >/= 60 yrs) about over-the-counter self-treatment with these drugs. These women had no exposure to acetaminophen, aspirin, or ibuprofen recorded in their computerized patient record during the previous 5 years. According to their self-reported average intake of study drugs of interest during the previous 5 years, exposure was classified as none, rare (< 10 tablets/yr), occasional (10-60 tablets/yr), regular (61-120 tablets/yr), or heavy (>/= 121 tablets/yr).

Statistical Analyses

We conducted matched case-control analyses (conditional logistic regression) with the software program SAS (SAS Institute, Cary, NC, release 6.12), adjusting for the potential confounders smoking status (nonsmoker, current smoker, ex-smoker, unknown) and body mass index (BMI < 25, 25-29.9, 30+ kg/m2).

Results

Over-the-Counter Analgesic Exposure

From the random sample of 60 women, we received exposure information for 40 (16 women aged < 60 yrs and 24 women aged >/= 60 yrs). Overall, 32 (80%) of 40 women reported taking acetaminophen, aspirin, or ibuprofen at least once during the previous 5 years. Acetaminophen was taken most often (27 women), but 24 women reported only rare or occasional exposure, 2 women reported regular exposure, and 1 woman reported heavy exposure. Self-reported ibuprofen exposure was almost entirely (11 of 12) rare or occasional, and the same was found for self-reported aspirin exposure (10 of 12; Table 1).

Acetaminophen and Cancer

We included 483 women with incident ovarian cancer and 1877 matched controls, 3706 women with incident breast cancer and 14,155 matched controls, and 635 women with incident colon cancer and 2434 matched controls in our analyses. Distribution of age, BMI, smoking status, and years of medical history in the GPRD before the index date of case and control subjects is displayed in Table 2.

Relative risk estimates, expressed as ORs, for exposure to acetaminophen and NSAIDs, adjusted for each other as well as for smoking status and BMI, are displayed in Table 3. For ovarian, breast, or colon cancer, the adjusted ORs for subjects with 30 or more acetaminophen prescriptions, compared with subjects who did not take acetaminophen, were 1.0 (95% CI 0.6-1.5), 0.8 (95% CI 0.7-1.0), and 1.0 (95% CI 0.7-1.4), respectively.

When compared with nonexposed women, the adjusted ORs for those with current exposure to acetaminophen 30 or more prescriptions were 1.0 (95% CI 0.6-1.5) for ovarian cancer, 0.8 (95% CI 0.7-1.0) for breast cancer, and 1.0 (95% CI 0.7-1.4) for colon cancer, and the ORs were close to 1 for those with past exposure in the three cancer groups.

When compared with nonexposed women, adjusted ORs for women with 1-2, 3-4, or 5 or more exposed years (i.e., years with >/= 3 acetaminophen prescriptions) before the index date were 1.2 (95% CI 0.9-1.7), 1.5 (95% CI 1.0-2.2), and 0.9 (95% CI 0.5-1.4), respectively, for ovarian cancer. For breast cancer, relative risk estimates were 0.9 (95% CI 0.8-1.1), 0.9 (95% CI 0.7-1.0), and 0.8 (95% CI 0.7-0.9), respectively, and for colon cancer 1.4 (95% CI 1.1-1.8), 0.9 (95% CI 0.7-1.3), and 0.8 (95% CI 0.5-1.2), respectively. We further stratified the analyses by age (50-64 yrs, 65+ yrs): there was no suggestion of effect modification by age, with results closely similar across age groups for the three types of cancer.

For ovarian cancer, the OR for 30 or more NSAID prescriptions was 1.1 (95% CI 0.7-1.8) compared with the reference group who did not take NSAIDs, after adjusting for acetaminophen exposure, smoking status, and BMI. For breast cancer, the adjusted OR for 30 or more NSAID prescriptions was 1.0 (95% CI 0.8-1.1), and for colon cancer 0.5 (95% CI 0.3-0.9). For colon cancer, we also analyzed the effect of ibuprofen alone on the risk of developing colon cancer. We did this analysis because there is evidence that ibuprofen -- like other NSAIDs -- has a protective effect on the risk of developing colon cancer,[18,22] and because ibuprofen is also available over-the-counter like acetaminophen. The number of subjects who had exposure to only ibuprofen was rather low; the OR for 10 or more prescriptions for only ibuprofen was 0.6 (95% CI 0.1-2.6).

In addition, we censored exposure to analgesics 1 year before the index date for ovarian cancer cases because cancer-related abdominal pain may have led to increased intake of analgesics before the index date. The ORs for 30 or more prescriptions for acetaminophen or NSAIDs were 0.9 (95% CI 0.6-1.5) and 1.1 (95% CI 0.6-1.8), respectively, adjusted for smoking status and BMI.

We further stratified ovarian cancer cases and their controls into mutually exclusive categories of exposure to acetaminophen alone, aspirin alone, or other NSAIDs alone, or mixed exposure (i.e., subjects who took various analgesics of these groups). There was a suggestion of an association between heavy acetaminophen exposure (i.e., >/= 30 prescriptions) and the risk of developing ovarian cancer (OR 0.5, 95% CI 0.2-1.8), and there was only 1 case subject and 31 control subjects who had any prescriptions recorded for aspirin, leading to an OR of 0.1 (95% CI 0.02-1.0).

We further adjusted the breast cancer analysis for the following potential confounders: prior hysterectomy (yes/no), prior oophorectomy (yes/no), prior history of benign breast lumps (yes/no), and longer term exposure to postmenopausal estrogens (1-9, 10-19, >/= 20 prescriptions). The ovarian cancer analysis was adjusted for hysterectomy (yes/no) and unilateral oophorectomy (yes/no). None of these parameters substantially affected the relationship of acetaminophen exposure to the risk of developing cancer.

Discussion

The three case-control analyses in women with incident cancer of the ovaries, breast, or colon provided little evidence for an association between acetaminophen exposure and cancer risk. There was a suggestion of a possible negative association between longer term acetaminophen exposure and the risk of developing breast cancer, and some evidence of a reduced risk of ovarian cancer in association with acetaminophen or aspirin in one subanalysis of the data set.

A previous case-control analysis[1] differed from our study with regard to the exposure assessment; that analysis was interview-based, whereas ours was based on prescription data from general practitioners. Interview-based exposure assessment can be erroneous due to recall bias, whereas the major limitation of our study was potential underassessment of acetaminophen exposure owing to possible over-the-counter drugs. Our general practitioner- conducted interview of a sample of female patients provided evidence that substantial exposure misclassification is rare, that is, women who never receive a prescription for acet-aminophen do not seem to take substantial amounts of the over-the-counter drug on a regular basis. Rare or occasional exposure was quite common, but we do not consider this to be of material importance for our analysis. We decided to interview only women who had no exposure to acetaminophen recorded in the medical record because we considered such misclassification to be potentially most important, whereas occasional intake of over-the-counter acetaminophen by women who already had regularly taken this drug by prescription would not cause important misclassification.

The authors of one study[31] analyzed the effect of NSAIDs on the risk of developing colorectal cancer. They compared two sources of exposure information, data from U.S.-based health maintenance organization prescription claims and self-reported, interview-based NSAID exposure information including intake of certain over-the-counter NSAIDs. Conclusions about the effect of NSAIDs on colorectal cancer did not differ materially when exposure was classified as self-reported (i.e., both over-the-counter and prescription) or when classified as prescription only. They concluded that prescription data can be sufficient for epidemiologic research under certain circumstances despite some misclassi-fication due to additional intake of these drugs on an over-the-counter basis.[31]

We compared women with heavy exposure (i.e., >/= 30 prescriptions before the index date) with nonexposed women. Since some of the alleged nonexposed women may have taken acetaminophen occasionally, this could have led to a dilution of a possible acetaminophen effect and to some distortion of the relative risk estimate toward the null. However, we still would have expected to find at least a suggestion of a risk difference between heavily exposed and nonexposed women in our analysis, if an acetaminophen effect on the cancer risk indeed existed. In addition, certain acetaminophen preparations (i.e., in combination with metoclopramide or propoxyphene) are available only by prescription, and this kind of exposure is not affected by possible misclassification due to over-the-counter exposure.

We also studied women with incident colon cancer to assess a possible ibuprofen effect on the cancer risk. Since ibuprofen is available over-the-counter and widely taken in that form in the United Kingdom, we would expect the same kind of exposure misclassification in an analysis based only on prescriptions for this drug. We found evidence for a reduced colon cancer risk related to ibuprofen (OR 0.6, 95% CI 0.1-2.6) based on small numbers that did not allow conclusive evidence. This finding is closely similar to a recent GPRD-based analysis (OR 0.7, 95% CI 0.4-1.2).[22] Furthermore, in a previous analysis[32] of the effect of NSAIDs on the risk of upper gastrointestinal bleeding, the authors found an increased risk for those taking ibuprofen (OR 2.9, 95% CI 1.7-5.0), with a particularly elevated risk for those taking high daily doses of 1500 mg (OR 6.5, 95% CI 2.6-16.4). These findings for ibuprofen also indicate that sufficient exposure to this drug by prescription is recorded in the GPRD, allowing credible findings despite possible exposure misclassification due to additional over-the-counter exposure.

In the case-control study mentioned earlier,[1] the authors adjusted their analysis for several potential confounders, such as religion, education, marital status, parity, previous intake of oral contraceptives, and indication for an analgesic. We were not able to control the analysis for most of these parameters because they are not routinely recorded in the GPRD. However, in their analysis,[1] none of these parameters confounded the association between study drug exposure and the risk of ovarian cancer. There is a theoretic possibility that socioeconomic status may be related both to the likelihood of getting acetaminophen or NSAIDs by prescription and to the risk of developing ovarian cancer. Such possible confounding could not be evaluated in our analysis since socioeconomic status is not coded or recorded in the GPRD.

In a subanalysis exploring the effects of acetaminophen only, aspirin only, or NSAIDs only in mutually exclusive groups, any exposure to aspirin by prescription was associated with a substantially decreased risk of developing ovarian cancer (OR 0.1, 95% CI 0.02-1.0). A recent study found a similar finding (OR 0.36, 95% CI 0.11-1.18) for women who reported regular aspirin intake in the years preceding the index date.[33]

The large case-control analysis on breast cancer yielded some evidence of a possible acetaminophen effect on the risk of developing breast cancer. Long-term acetaminophen exposure (i.e., >/= 30 prescriptions) was associated with a suggestion of slightly decreased breast cancer risk (OR 0.8, 95% CI 0.7-1.0). However, we did not find a decreased breast cancer risk in those with long-term exposure to NSAIDs, as suggested in some recent epidemiologic studies. The authors of several studies[6-9] reported negative associations between longer term NSAID exposure and breast cancer risk, whereas the authors of studies on NSAIDS[5] and aspirin only[10,11] did not find evidence for such an effect of NSAID exposure.

In summary, we found only little evidence for an altered risk of developing ovarian cancer in association with longer term exposure to acetaminophen or NSAIDs. There was a suggestion of a decreased breast cancer risk associated with longer term exposure to acetaminophen, but not with NSAIDs. The colon cancer risk was decreased in subjects with regular exposure to NSAIDs, but not affected by acetaminophen exposure.

Tables

Table 1. Interview-Based Self-Assessment of Over-the-Counter Exposure to Acetaminophen, Ibuprofen, and Aspirin in a Random Sample of 40 Women in the GPRD


Exposure Age < 60 years (n=16) Age >/= 60 years (n=24) Total (N=40)
APAP Ibu ASA APAP Ibu ASA APAP Ibu ASA
None 5 8 11 8 20 17 13 28 28
Any 11 8 5 16 4 7 27 12 12
   Rare (< 10 tablets/yr) 1 1 1 13 2 3 14 3 4
   Occasional (10-60 tablets/yr) 8 6 4 2 2 2 10 8 6
   Regular (61-120 tablets/yr) 1 0 0 1 0 1 2 0 1
   Heavy (>/= 121 tablets/yr) 1 1 0 0 0 1 1 1 1


GPRD = United Kingdom-based General Practice Research Database; APAP = acetaminophen; Ibu = ibuprofen; ASA = aspirin.

Table 2. Distribution of Age, Body Mass Index, Smoking Status, and Medical History in the GPRD in Case and Control Subjects


Parameter Ovarian Cancer Breast Cancer Colon Cancer
Cases
(n=483)		 Controls
(n=1877)		 OR (95% CI) Cases
(n=3706)		 Controls
(n=14,155)		 OR (95% CI) Cases
(n=635)		 Controls
(n=2434)		 OR (95% CI)
Age (yrs)
   50-59 170 663 -- 1182 4614 -- 94 368 --
   60-69 129 506 -- 952 3693 -- 126 484 --
   70-79 122 468 -- 929 3515 -- 226 875 --
   80-89 62 240 -- 643 2333 -- 189 707 --
BMI (kg/m2)
   < 25 161 595 1.0 (ref) 1124 4385 1.0 (ref) 212 699 1.0 (ref)
   25-29.9 114 455 0.9 (0.7-1.2) 883 3319 1.0 (0.9-1.2) 133 514 0.9 (0.7-1.1)
   30+ 63 243 1.0 (0.7-1.3) 500 1779 1.1 (1.0-1.3) 56 219 0.8 (0.6-1.1)
   Unknown 145 584 0.9 (0.7-1.2) 1199 4672 1.0 (0.9-1.1) 234 1002 0.7 (0.6-0.9)
Smoking status
   Nonsmoker 294 1075 1.0 (ref) 2188 8008 1.0 (ref) 378 1401 1.0 (ref)
   Current smoker 76 280 1.0 (0.7-1.3) 532 2224 0.9 (0.8-1.9) 79 285 1.0 (0.8-1.4)
   Ex-smoker 41 150 1.0 (0.7-1.5) 300 1090 1.0 (0.9-1.2) 40 171 0.9 (0.6-1.3)
   Unknown 72 372 0.7 (0.5-0.9) 686 11,712 0.9 (0.8-1.0) 138 577 0.8 (0.7-1.1)
Mean years of history in GPRD 5.3 5.3 -- 5.3 5.3 -- 5.4 5.4 --


BMI = body mass index; GPRD = United Kingdom-based General Practice Research Database; ref = reference group.

Table 3. Distribution of Acetaminophen and NSAID Exposure in Case and Control Subjects and Risk of Developing Cancer


No. of Prescriptions Ovarian Cancer Breast Cancer Colon Cancer
Cases
(n=483)		 Controls
(n=1877)		 OR (95% CI) Cases
(n=3706)		 Controls
(n=14,155)		 OR (95% CI) Cases
(n=635)		 Controls
(n=2434)		 OR (95% CI)
Acetaminophen
   0 210 910 1.0 (ref) 1773 6574 1.0 (ref) 238 952 1.0 (ref)
   1-9 185 645 1.2 (1.0-1.6) 1319 4950 1.0 (0.9-1.1) 261 887 1.3 (1.0-1.5)
   10-19 37 116 1.4 (0.9-2.2) 230 844 1.0 (0.8-1.2) 54 215 1.1 (0.7-1.5)
   20-29 21 59 1.7 (1.0-2.9) 112 567 0.7 (0.6-0.9) 25 108 1.0 (0.6-1.6)
   30+ 30 147 1.0 (0.6-1.5) 272 1220 0.8 (0.7-1.0) 57 272 1.0 (0.7-1.4)
NSAIDs
   0 197 820 1.0 (ref) 1465 5439 1.0 (ref) 305 1067 1.0 (ref)
   1-9 224 748 1.1 (0.9-1.4) 1547 6029 1.0 (0.9-1.1) 248 967 0.9 (0.7-1.1)
   10-19 24 117 0.8 (0.5-1.2) 275 1034 1.0 (0.9-1.2) 39 161 0.8 (0.6-1.2)
   20-29 11 61 0.7 (0.4-1.4) 148 548 1.1 (0.9-1.3) 20 84 0.8 (0.5-1.4)
   30+ 27 95 1.1 (0.7-1.8) 271 1105 1.0 (0.8-1.1) 23 155 0.5 (0.3-0.9)


NSAIDs = nonsteroidal antiinflammatory drugs; ref = reference group.

Exposure variables were adjusted for each other, for smoking status (nonsmoker, current smoker, ex-smoker, unknown) and body mass index (< 25, 25-29.9, 30+ kg/m2, unknown).

References

  1. Cramer DW, Harlow BL, Titus-Ernsthoff L, Bohlke K, Welch WR, Greenberg ER. Over-the-counter analgesics and risk of ovarian cancer. Lancet 1998;351:104-7.
  2. Rodriguez C, Henley SJ, Calle EE, Thun MJ. Paracetamol and risk of ovarian cancer mortality in a prospective study of women in the USA. Lancet 1998;352:1354-5.
  3. Rosenberg L, Palmer JR, Rao RS, et al. A case-control study of analgesic use and ovarian cancer. Cancer Epidemiol Biomarkers Prev 2000;9:933-7.
  4. Collaborative Group on Hormonal Factors in Breast Cancer. Breast cancer and hormonal contraceptives: collaborative reanalysis of individual data on 53297 women with breast cancer and 100,239 women without breast cancer from 54 epidemiological studies. Lancet 1996;347:1713-27.
  5. Langman MJS, Cheng KK, Gilman EA, Lancashire RJ. Effect of anti-inflammatory drugs on overall risk of common cancer: case-control study in general practice research database. BMJ 2000;320:1642-6.
  6. Sharpe CR, Collet JP, McNutt M, Belzile E, Boivin JF, Hanley JA. Nested case-control study of the effects of non-steroidal anti-inflammatory drugs on breast cancer risk and stage. Br J Cancer 2000;83;112-20.
  7. Harris RE, Alshafie GA, Abou-Issa H, Seibert K. Chemoprevention of breast cancer in rats by celecoxib, a cyclooxygenase 2 inhibitor. Cancer Res 2000;60:2101-3.
  8. Coogan PF, Rao RS, Rosenberg L, et al. The relationship of non-steroidal anti-inflammatory drug use to the risk of breast cancer. Prev Med 1999;29:72-6.
  9. Harris RE, Kasbari S, Farrar WB. Prospective study of non-steroidal anti-inflammatory drugs and breast cancer. Oncol Rep 1999;6:71-3.
  10. Egan KM, Stampfer MJ, Giovannucci E, Rosner BA, Colditz GA. Prospective study of regular aspirin use and the risk of breast cancer. J Natl Cancer Inst 1996;88:988-93.
  11. Schreinemachers DM, Everson RB. Aspirin use and lung, colon, and breast cancer incidence in a prospective study. Epidemiology 1994;5:138-46.
  12. Vainio H, Morgan G, Kleihues P. An international evaluation of the cancer-preventive potential of nonsteroidal anti-inflammatory drugs. Cancer Epidemiol Biomarkers Prev 1997;6:749-53.
  13. Rosenberg L, Palmer JR, Auber AG, Warsahauer ME, Stolley PD, Shapiro S. A hypothesis: nonsteroidal anti-inflammatory drugs reduce the incidence of large-bowel cancer. J Natl Cancer Inst 1991;83:355-8.
  14. Suh O, Mettlin C, Petrelli NJ. Aspirin use, cancer, and polyps of the large bowel. Cancer 1993;72:1171-7.
  15. Thun MJ, Namboodiri MM, Calle EE, Flanders WD, Heath CW Jr. Aspirin use and risk of fatal cancer. Cancer Res 1993;53:1322-7.
  16. Giovannucci E, Egan KM, Hunter DJ, et al. Aspirin and the risk of colorectal cancer in women. N Engl J Med 1995;333: 609-14.
  17. Friedman GD, Coates AO, Potter JD, Slatteri ML. Drugs and colon cancer. Pharmacoepidemiol Drug Saf 1998;7:99-106.
  18. Smalley W, Ray WA, Daugherty J, Griffin MR. Use of nonsteroidal anti-inflammatory drugs and incidence of colorectal cancer. Arch Intern Med 1999;159:161-6.
  19. Stürmer T, Glynn RJ, Min Lee I, Manson JE, Buring JE, Hennekens CH. Aspirin use and colorectal cancer: post-trial follow-up data from the physicians health study. Ann Intern Med 1998;128:713-20.
  20. Muscat JE, Stellman SD, Wynder EL. Nonsteroidal antiinflammatory drugs and colorectal cancer. Cancer 1994;74:1847-54.
  21. Peleg II, Lubin MF, Cotsonis GA, Clark WS, Wilcox CM. Long-term use of nonsteroidal antiinflammatory drugs and other chemopreventors and risk of subsequent colorectal neoplasia. Dig Dis Sci 1996;41:1319-26.
  22. Garcia Rodriguez LA, Huerta-Alvarez C. Reduced risk of colorectal cancer among long-term users of aspirin and nonaspirin antiinflammatory drugs. Epidemiology 2001;12: 88-93.
  23. Eberhardt CE, Coffey RJ, Radhlka A, Giardiello FM, Ferrenbach S, DuBois RN. Upregulation of cyclooxygenase 2 gene expression in human colorectal adenomas and adenocarcinomas. Gastroenterology 1994;107:1183-8.
  24. Walley T, Mantgani A. The UK General Practice Research Database. Lancet 1997;350:1097-9.
  25. Jick H. A database worth saving. Lancet 1997;350:1045-6.
  26. Garcia Rodriguez LA, Perez Gutthann S. Use of the UK General Practice Research Database for pharmacoepidemiology. Br J Clin Pharmacol 1998;45:419-25.
  27. Jick H, Jick SS, Derby LE. Validation of information recorded on general practitioner-based computerised data resource in the United Kingdom. BMJ 1991;302:766-8.
  28. Jick H, Terris BZ, Derby LE, Jick SS. Further validation of information recorded on a general practitioner-based computerized data resource in the United Kingdom. Pharmacoepidemiol Drug Saf 1992;1:347-9.
  29. Jick H, Jick SS, Derby LE, Vasilakis C, Wald Myers M, Meier CR. Calcium-channel blockers and risk of cancer. Lancet 1997;349:525-8.
  30. Meier CR, Derby LE, Jick SS, Jick H. ACE-inhibitors, calcium channel blockers and breast cancer. Arch Intern Med 2000;160:349-53.
  31. Ulcickas Yood M, Rothman KJ, Johnson CC, et al. Using prescription claims data for drugs available over-the-counter (OTC) [abstr]. Pharmacoepidemiol Drug Saf 2000;9(suppl 1):S37.
  32. Garcia Rodriguez LA, Jick H. Risk of upper gastrointestinal bleeding and perforation associated with individual non-steroidal anti-inflammatory drugs. Lancet 1994;343:769-72.
  33. Akhmedkhanov A, Toniolo P, Zeleniuch-Jacquotte A, Kato I, Koenig KL, Shore RE. Aspirin and epithelial ovarian cancer. Prev Med 2001;33:682-7.
Christoph R. Meier, Ph.D., M.Sc., Stephen Schmitz, M.D., M.P.H., Hershel Jick, M.D.; Basel Pharmacoepidemiology Unit, Division of Clinical Pharmacology and Toxicology, University Hospital, Basel, Switzerland (Dr. Meier); and the Boston Collaborative Drug Surveillance Program, Boston University School of Medicine, Lexington, Massachusetts.

Back To Top © 2001-2009 RxRama ---- All rights reserved.