Gender as a Risk Factor for Advanced Neoplasia and Colorectal Cancer: A Systematic Review and Meta-analysis

Share Embed


Descrição do Produto

CLINICAL GASTROENTEROLOGY AND HEPATOLOGY 2009;7:676 – 681

Gender as a Risk Factor for Advanced Neoplasia and Colorectal Cancer: A Systematic Review and Meta-analysis STEPHEN P. NGUYEN,*,‡ STEPHEN BENT,‡,§ YEA-HUNG CHEN,‡ and JONATHAN P. TERDIMAN*,‡ *Division of Gastroenterology, ‡Department of Medicine, and §Department of Epidemiology and Biostatistics, University of California, San Francisco, San Francisco, California

See related article, Chan AT et al, on page 2127 in Gastroenterology. Background & Aims:: Studies have reported higher rates of advanced colorectal neoplasia in men than in women. We performed a meta-analysis to provide a quantitative pooled risk estimate of the association between gender and advanced colorectal neoplasia. Methods:: We conducted a systematic review to identify studies of averagerisk and asymptomatic individuals undergoing screening colonoscopy. We also included studies of subjects with a family history of colorectal neoplasia. We used random effects models to evaluate pooled relative risk estimates and performed heterogeneity and publication bias analyses. The primary outcome measure was relative risk of advanced neoplasia in men compared with women. A secondary outcome measure was relative risk for colorectal cancer. Results:: Seventeen studies consisting of 18 different populations were included, comprising 924,932 men and women. The pooled relative risk estimate for advanced neoplasia for men compared with women was 1.83 (95% confidence interval [CI], 1.69 –1.97). This positive association between gender and advanced neoplasia was significant across all age groups from 40 to older than 70 years. In 5 studies, the relative risk estimate for cancer for men compared with women was 2.02 (95% CI, 1.53–2.66). Significant heterogeneity was found for the overall analysis and for studies reporting on cancer but not for studies that excluded subjects with a family history or for those analyses grouped by age. Conclusions:: This meta-analysis provides strong evidence that men are at greater risk for advanced colorectal neoplasia across all age groups. This might inform decisions to create sex-specific colorectal cancer screening recommendations.

C

olorectal cancer (CRC) is the second leading cause of death from cancer in the United States.1 Screening is recommended to begin at 50 years of age in average-risk populations. Some authorities have questioned screening strategies based solely on patient age and family history, without additional consideration of gender, race, or other factors that might significantly impact the risk of advanced adenoma and colorectal cancer.2 These concerns are driven in part by the inherent risks of endoscopy and because colonoscopy remains a limited resource. A recent joint committee from the American Cancer Society and the United States Multi-Society Task Force on Colorectal Cancer has postponed alteration to current guide-

lines, pending further study of factors that might affect risk for advanced neoplasia.3 Population-based cancer registry data from the United States and Europe indicate that men have a higher age-adjusted incidence and death rate from CRC than women.4 – 6 One study with cancer registry data from the United States and national mortality statistics from different countries found that women reached equivalent levels of CRC incidence and mortality 4 – 8 years later than men.7 The gender differences were remarkably constant across populations and over time. The reasons for the differences were unclear, but they might be related to biologic variation, differential access to care, or behaviors regarding screening. The registry-based studies, however, do not have data on screening rates or differential rates of advanced adenomas between men and women. A determination of whether the rates of advanced adenomas differ among men and women might best inform our decision about whether gender-based screening recommendations make sense. We evaluated the association between gender and the risk for advanced colorectal neoplasia by analyzing studies of asymptomatic, average-risk individuals undergoing screening colonoscopy from North America, Europe, and Asia. We conducted a comprehensive systematic review and meta-analysis of the data from these studies to provide a summary risk estimate and to determine the consistency of the studies and results.

Methods Definition of Exposure and Outcome Gender and its association with advanced colorectal neoplasia are the focus of the pooled analysis. We included studies that defined advanced neoplasia as any adenoma equal to or greater than 10 mm, with any villous histology or highgrade dysplasia, or invasive adenocarcinoma. In 2 studies, advanced neoplasia also included polyps with “moderate” dysplasia.8,9

Search Strategy A systematic search for published manuscripts that evaluated the association between gender and advanced neoplasia during screening colonoscopies was undertaken. PubMed was searched by using the keywords “sex OR gender OR advanced Abbreviations used in this paper: CI, confidence interval; CRC, colorectal cancer; NNS, number needed to screen. © 2009 by the AGA Institute 1542-3565/09/$36.00 doi:10.1016/j.cgh.2009.01.008

June 2009

neoplasia AND colorectal cancer AND screening colonoscopy.” This resulted in 942 abstracts (for dates 1950 through September 2008). Non–English language articles and reviews were excluded. The remaining 774 abstracts were reviewed by one investigator (S.P.N.), who applied a priori criteria for inclusion and exclusion. The majority of articles were excluded for the following reasons: case reports; basic science articles; screening by methods other than colonoscopy; studies involving symptomatic subjects or subjects with a history of polyps, colon cancer, inflammatory bowel disease, or who had undergone a colonoscopy within the previous 5 years; and studies that did not use the established definition of advanced neoplasia. The same search strategy was implemented for the EMBASE database, but no new studies were identified. In addition, a manual review of the reference list from all included articles and relevant published reviews and guidelines was performed, although no new studies were found. In the end, 30 publications were selected for a detailed and independent review by 2 investigators (S.P.N. and J.P.T.). Disagreements were resolved by consensus.

Study Selection and Data Abstraction Studies were included if they met the following inclusion criteria: (1) subjects were asymptomatic, and either at average risk for CRC or with family history of CRC, undergoing (2) screening colonoscopy with (3) data available on advanced colorectal neoplasia among both men and women, and (4) reported a relative risk with confidence intervals (CIs) or provided sufficient data to permit their calculation. In case of multiple reports on the same population, we considered the estimates from the most recent or most informative report. Of the 30 publications reviewed in detail, 15 were initially excluded for the following reasons: relevant data not available in published form (7); inclusion of symptomatic patients (4); pre-established definition of advanced neoplasia not used or not defined (2); duplicate study (1); and study included only male subjects (1). We attempted to contact the authors of 7 studies that appeared to meet our inclusion criteria but in which the published data were insufficient for inclusion. This resulted in the inclusion of 2 additional studies,10,11 resulting in 17 studies, comprising 18 distinct study populations, being included in our meta-analysis.

Data Analysis and Statistical Methods For each cohort selected for analysis, except when estimates were already provided, we computed (1) an overall estimate for the relative risk of neoplasia (comparing men with women) and (2) a standard error. We then conducted a metaanalysis of these rates by using all cohorts that had been selected for inclusion (n ⫽ 18). A number needed to screen (NNS) to detect one advanced neoplasm was calculated for each study from the individual rates of advanced neoplasm provided for both men and women. However, it was not possible to statistically combine the NNS values across studies. We provided separate summary estimates with the studies by Brenner et al12 and Regula et al13 excluded. This was performed because these studies included large numbers of subjects when compared with the other cohorts, and thus they might obscure the effects of the smaller studies. We also performed subgroup analyses for cohorts that did not include subjects with a family history of colorectal cancer (n ⫽ 10) and for cohorts that did (n ⫽ 9). The cohort in the study by Regula et al allowed for a

GENDER AS A RISK FOR ADVANCED COLON NEOPLASIA

677

separate analysis of subjects with and without a family history of CRC. Many studies reported age group–stratified counts and rates of advanced neoplasia. For each of these, we computed age group–stratified estimates by 10-year age groups from age 40 to older than 70. The age group cutoffs were not consistently reported across studies; it was necessary at times to combine age groups across studies that did not exactly match. Specifically, the 60 – 69 age group we report for the cohort in the study by Rex et al14 was actually reported as a 60 – 64 age group in that study, the 70⫹ age group for the cohort in the study by Rex et al we report was a 65⫹ age group in that study, and the 50 –59 age group we report for the study by Brenner et al12 was a 55–59 age group. Finally, for each cohort for which counts or rates were available (n ⫽ 5), we computed an estimate for the relative risk for CRC (and a corresponding standard error). Of note, 2 studies reported no cases of cancer in women, which would complicate computation of relative risks.10,15 Thus, we assumed 0.5 cases among women to facilitate calculation of relative risks. As recommended in the statistical literature, 0.5 was also added to all cells (numerator and denominator) in the 2 studies.16 The data on cancer for each study are provided in Supplementary Table 1. We used random effects models and assessed heterogeneity with the Q statistic.17 To test for publication bias, we used the Begg adjusted rank correlation test and the Egger regression asymmetry test.18,19 A critical level of 0.05 was used for all hypothesis tests; all CIs are at the 95% level. For the test of heterogeneity, a P value of ⬍.1 was considered significant. Meta-analysis was performed via the rmeta package in R.20 There are no established markers of study quality for studies of the type included in this analysis; therefore, we did not distinguish among the studies on the basis of quality. Among plausible markers of quality, 13 of the 18 studies specified the method by which polyp size was determined; most used open forceps. Nine studies used the World Health Organization criteria with respect to grading the pathology of polyps. Five used a single, central pathologist. These details are provided in Supplementary Table 2.

Results Seventeen studies, published between 1993 and 2008 and reporting on 18 distinct study populations, were included in the analysis (Table 1). One study reported on rates of advanced neoplasia in 2 distinct populations from Seattle and Taiwan.21 With one exception, all studies included men and women from the same population. One study matched averagerisk women to male veterans to compare rates of advanced neoplasia.22,23 Table 1 provides details of the 18 cohorts included in this analysis, comprising 924,932 subjects (387,431 men and 537,501 women). Six of the cohort studies were conducted in North America, 4 in Europe, and 8 in Asia. Supplementary Table 3 lists the data available from each individual study, including number of participants and rates of advanced neoplasia for both men and women of different age groups.

Data Synthesis Overall, rates of advanced neoplasia among men ranged from 2.62%–16.5%, whereas rates among women varied from 1.80%–9.30%. All of the studies but one10 showed a higher rate

678

NGUYEN ET AL

CLINICAL GASTROENTEROLOGY AND HEPATOLOGY Vol. 7, No. 6

Table 1. Description of the Included Studies First author

Year

Design

Location

Men/women

Age range (y)

Family Hx CRC

Rex14 Betes24 Sung9 Imperiale11 Chiu8 Liu46 Schoenfeld22 Soon(a)21 Soon(b)21 Regula13 Anderson47 Brenner12 Byeon48 Choe49 Kim39 Liou50 Rundle10 Marbet15

1993 2003 2003 2003 2005 2005 2005 2005 2005 2006 2007 2007 2007 2007 2007 2007 2008 2008

Retrospective Prospective Prospective Retrospective Prospective Prospective Prospectivea Prospective Prospective Retrospective Retrospective Retrospective Prospective Retrospective Retrospective Prospective Retrospective Prospective

USA Spain China USA Taiwan Taiwan USA Taiwan USA Poland USA Germany Asia Korea Korea Taiwan USA Switzerland

310/186 1649/561 224/281 1753/1272 1034/674 3125/2848 2206/1198 903/553 1666/1737 18,012/32,136 1356/1137 349,249/490,900 471/389 1623/812 946/798 1193/913 688/217 1023/889

50–75 ⱖ40 50–79 ⬎50 ⬍50 to ⱖ70 40 to ⱖ70 50–79 ⱖ40 ⱖ40 40–66 ⱖ40 ⱖ55 20–90 50–84 ⱖ30 ⱖ50 40–59 50–80

Excluded Excluded Excluded Included Excluded Excluded Excluded Included Included Included Included Included Included Excluded Included Excluded Excluded Included

Hx, history. aMatched men and women from CONCERN and VA 380 trials.22,23

of advanced neoplasia among men than women, resulting in a higher NNS to detect an advanced neoplasm in women as compared with men (Table 2). All but two of the studies8,10 showed a significant positive association between male sex and advanced neoplasia. The summary relative risk estimate of rates of advanced neoplasia for men versus women for all cohorts was 1.83 (95% CI, 1.69 –1.97), indicating an 83% higher risk of advanced neoplasia being detected during screening colonoscopy in men (Figure 1). The summary estimates for the association between gender and advanced neoplasia revealed evidence of heterogeneity (P ⫽ .03). Statistical tests did not show evidence of publication bias (Begg test P ⫽ .38 and Egger test P ⫽ .16). The cohorts of Regula et al13 and Brenner et al12 included large numbers of subjects and could potentially obscure the effects of the smaller studies. Separate summary estimates were

therefore calculated for all studies excluding these 2 cohorts. The results were nearly identical (Table 3). We also calculated separate summary estimates for studies that did or did not include subjects with a family history of CRC (Table 3). The summary relative risk estimate for advanced neoplasia for 10 cohorts without a family history was 1.83 (95% CI, 1.70 –1.96), whereas that for 9 cohorts with a family history was 1.87 (95% CI, 1.69 –2.06). The test for heterogeneity was significant for the subset of studies that included patients with a family history (P ⫽ .01); there was no heterogeneity in the studies of subjects without a family history (P ⫽ .51).

Table 2. Individual Study Statistics First author

% Advanced neoplasia (M/F)

Relative risk (95% CI)

NNS (M/F)

Rex14 Betes24 Sung9 Imperiale11 Chiu8 Liu46 Schoenfeld22 Soon(a)21 Soon(b)21 Regula13 Anderson47 Brenner12 Byeon48 Choe49 Kim39 Liou50 Rundle10 Marbet15

12/4 13.5/6.6 16.5/9.3 7.99/3.22 3.4/2.4 4.0/2.6 8.6/4.5 5.3/1.8 6.2/3.6 7.9/4.3 10.6/8.2 7.9/4.6 5.9/2.8 5.1/2.0 6.0/2.6 6.7/3.7 2.62/2.76 11.7/4.72

3.00 (1.39–6.45) 2.05 (1.46–2.86) 1.77 (1.11–2.84) 2.48 (1.77–3.49) 1.42 (0.79–2.53) 1.51 (1.13–2.00) 1.91 (1.42–2.56) 2.94 (1.50–5.78) 1.72 (1.27–2.34) 1.87 (1.74–2.00) 1.29 (1.01–1.66) 1.74 (1.71–1.77) 2.11 (1.06–4.19) 2.55 (1.51–4.31) 2.31 (1.41–3.78) 1.81 (1.22–2.68) 0.95 (0.38–2.36) 3.46 (1.75–6.84)

8.3/25 7.4/15 6.1/11 13/31 29/42 25/38 12/22 19/56 16/28 13/23 9.4/12 13/22 17/36 20/50 17/38 15/27 38/36 8.5/21

Figure 1. Relative risk for advanced neoplasia in men vs women: individual study estimates.

June 2009

GENDER AS A RISK FOR ADVANCED COLON NEOPLASIA

679

Table 3. Summary Risk Estimates of Advanced Neoplasia for Men Versus Women

Population All studies All studies excluding Brenner et al12 All studies excluding Brenner et al,12 Regula et al13 10 studies without family history CRC 9 studies with family history CRC

Relative risk (95% CI)

Test for heterogeneity (P value)

1.83 (1.69–1.97) 1.88 (1.69–2.08)

.03 .07

1.90 (1.66–2.16)

.05

1.83 (1.70–1.96)

.51

1.87 (1.69–2.06)

.01

Twelve studies with 13 cohorts reported age-specific data. To examine the influence of age, we calculated summary estimates for 10-year age groups from 40 to older than 70 years (Table 4). There was a significant positive association between male sex and risk of advanced neoplasia across all age groups. Moreover, there was no heterogeneity within the age subgroup analyses (Table 4). Analyses of the studies grouped by geographic region around the world (North America, Europe, and Asia) were undertaken as well, and once again a significant association between male gender and advanced neoplasia was found (data not shown). The rates of CRC for men and women were available from 5 studies, and meta-analysis of these studies was performed (Figure 2). The summary relative risk estimate of CRC for men versus women was 2.02 (95% CI, 1.53–2.66). Two of the 5 studies included only screening colonoscopy subjects without a family history of CRC.10,24 The test for heterogeneity for this analysis was significant (P ⫽ .06). However, the individual studies all showed a strong positive association between male gender and the risk of CRC. Analysis did not show evidence of publication bias (Begg test P ⫽ 1.0 and Egger test P ⫽ .31).

Discussion Although the data are scarce, it is generally believed that the lifetime incidence for CRC is similar among men and women, possibly because women live longer than men.13 This systematic review and meta-analysis, however, demonstrated that men are at greater age-specific risk for advanced colorectal neoplasia than women. This finding was consistent across different populations around the world. The higher risk of advanced neoplasia in men compared with women resulted in a larger NNS to detect one patient with advanced neoplasia in women in any particular age group. The NNS in younger patients might be of interest to policy makers who must grapple with whether it makes

Table 4. Relative Risk of Advanced Neoplasia in Men Versus Women by Age Groups

Age group (y)

Number of cohorts

Relative risk (95% CI)

Test for heterogeneity (P value)

40–49 50–59 60–69 ⱖ70

6 13 11 11

1.53 (1.23–1.91) 1.86 (1.77–1.95) 1.78 (1.74–1.82) 1.53 (1.26–1.86)

.67 .40 .55 .26

Figure 2. Relative risk for CRC in men vs women: individual study estimates.

sense for women to commence screening at the same age as men. Three studies provided the data to calculate the NNS for men and women aged 50 –54.11,13,14 The NNS were 21, 17, and 10 for men in this age group, as compared with 66, 28, and 50 for women, respectively. The age groups at which women reach a similar NNS as men aged 50 –54 in these 3 studies were 60 – 64, 60 – 66, and 65–75, respectively. Similar results are seen when comparing the NNS in men and women between the ages of 50 and 60 (Supplementary Table 3). The findings of a positive association for gender and advanced colorectal neoplasia are consistent with other large colonoscopy-based studies that ultimately were excluded from this analysis because they included symptomatic patients or did not include sufficient detail to meet our selection criteria.25–27 One recent study with the Clinical Outcomes Research Initiative database of 85,525 asymptomatic white and black men and women undergoing screening colonoscopies found that white women had a much lower risk of polyps sized more than 9 mm in diameter when compared with white men (odds ratio, 0.59; 95% CI, 0.55– 0.63).28 Black women also had a lower risk compared with black men (odds ratio, 0.82; 95% CI, 0.67–1.00). Another large study of 83,646 colonoscopies in Wisconsin revealed a cancer rate of 6.8% in men as opposed to 4.6% in women.29 The etiology for the gender difference remains uncertain, but it might be related to hormonal differences. McMichael and Potter30 first suggested a role for estrogens and progestins in preventing colon cancer. Since that time, large-scale studies have found evidence for the protective effect of postmenopausal hormone replacement therapy against CRC, but only with combination estrogen and progestin preparations.31,32 Smaller studies have also found that hormone replacement therapy was associated with reduced risk of advanced neoplasia.33 Other factors might explain the gender differences. Exposure to dietary and lifestyle risk factors for CRC might differ between

680

NGUYEN ET AL

men and women. For example, smoking and alcohol use have been shown to be associated with advanced colorectal neoplasia.34 –36 The differences in rates of advanced neoplasia might reflect the sex differences in smoking and alcohol consumption. The association between body mass index and CRC also has been evaluated. Earlier studies found similar relative risks among men and women on the basis of body mass index, whereas more recent investigations suggest the relationship between obesity and CRC is stronger for men than for women.37 Genetic differences between men and women might account for some of the differences in the rate of advanced neoplasia. There are no data yet that specifically address this issue, but a recent study did find that specific polymorphisms or gene variants linked to the development of colon cancer resulted in opposite survival outcomes for men and women with metastatic CRC, suggesting that sex-specific genetic differences might be important.38 Strengths of the current study include the large number of studies analyzed, comprising a very large number of subjects from around the world. The available data permit an analysis of the differential risk for advanced adenomas as well as CRC. There are potential limitations of this analysis. There is heterogeneity in the summary relative risk estimates of advanced neoplasia as well as the subgroup analysis for CRC. Thus, these summary estimates must be viewed with some caution. It is worth noting that the heterogeneity is not due to variation between studies in the finding that men had more neoplasia than women but around the estimate in the magnitude of that increased risk. We thus believe that reporting these summary risk estimates remains valid and important. The observation of heterogeneity should not reduce the confidence in the finding of increased risk but just add some uncertainty about the magnitude of that increased risk. It should be remembered that the analyses of the relative risk by age groups as well as the analysis of the studies including only average-risk subjects without a family history of CRC showed no evidence of heterogeneity. It is possible that the varying age ranges among the studies contributed to the heterogeneity of the overall analysis, and it appears that the studies that included subjects with a family history certainly did as well. Importantly, all but two8,10 of the studies included in this meta-analysis demonstrated a significantly higher rate of advanced neoplasia in men, and that elimination of these studies from the analysis did not eliminate the heterogeneity. The lack of significance in the study by Chiu et al8 is likely a result of a higher number of average-risk men younger than age 50 included in the study as compared with women (23% of men versus 15% of women in the study were younger than 50). Another limitation of this study is that it could not address many of the confounding variables such as diet, lifestyle, and other factors that might contribute to the difference in rates of advanced neoplasia among men and women. For example, only one study adjusted for the rates of smoking and alcohol consumption among men and women. In this study when controlled for smoking and alcohol, the prevalence of adenomas was still significantly higher in men than in women.39 Finally, although all of the subjects were enrolled in screening colonoscopy studies, thereby in great part eliminating differential access to care or different behaviors regarding screening as a cause for the observed differences in risk for advanced colorectal neoplasia, subtle sex-based differences in study enrollment could theoretically confound our results. For example, many of the Asian studies included patients who were self-referred for screening colonoscopy. This might lead to selection bias that might impact

CLINICAL GASTROENTEROLOGY AND HEPATOLOGY Vol. 7, No. 6

the results. Moreover, women might be more health conscious and likely to enroll in a screening colonoscopy study, whereas men might be more inclined to enroll only if they had subtle underlying symptoms not captured by study investigators, thereby putting them at increased risk for advanced neoplasia. It seems extremely unlikely, however, that such behavioral differences between men and women would have a major impact on our results. If anything, average-risk women seem less inclined to undergo screening colonoscopy than men.40,41 In conclusion, this systematic review and meta-analysis provide strong evidence for an association between gender and the risk of advanced colorectal neoplasia. Currently, guidelines across countries do not customize screening on the basis of gender. It can be argued that customizing screening recommendations on the basis of gender might add another layer of complexity for primary care doctors, perhaps diluting a message that is relatively simple now: screen at age 50 for everyone. Nevertheless, we believe that the current data are robust enough to support a critical reassessment of current screening guidelines. The magnitude of the sex-specific differences in the risk for advanced colorectal neoplasia we found, for example, is very similar to that seen between patients with and without a family history of colorectal neoplasia.42,43 Unlike gender, family history has been embraced as a risk factor worthy of screening customization.44,45 Supplementary Table 3.

Supplementary Data Note: to access the supplementary material accompanying this article, visit the online version of Clinical Gastroenterology and Hepatology at www.cghjournal.org. References 1. Jemal A, Siegel R, Ward E, et al. Cancer statistics, 2007. CA Cancer J Clin 2007;57:43– 66. 2. Lieberman D. Race, gender, and colorectal cancer screening. Am J Gastroenterol 2005;100:2756 –2758. 3. Levin B, Lieberman D, McFarland B, et al. Screening and surveillance for the early detection of colorectal cancer and adenomatous polyps, 2008: a joint guideline from the American Cancer Society, the US Multi-Society Task Force on Colorectal Cancer, and the American College of Radiology. CA Cancer J Clin 2008; 58:130–160. 4. Matanoski G, Tao XG, Almon L, et al. Demographics and tumor characteristics of colorectal cancers in the United States, 1998 – 2001. Cancer 2006;107(Suppl):1112–1120. 5. Ferlay J, Autier P, Boniol M, et al. Estimates of the cancer incidence and mortality in Europe in 2006. Ann Oncol 2007;18:581–592. 6. Brenner H, Hoffmeister M, Haug U. Family history and age at initiation of colorectal cancer screening. Am J Gastroenterol 2008;103:2326 –2331. 7. Brenner H, Hoffmeister M, Arndt V, et al. Gender differences in colorectal cancer: implications for age at initiation of screening. Br J Cancer 2007;96:828 – 831. 8. Chiu HM, Wang HP, Lee YC, et al. A prospective study of the frequency and the topographical distribution of colon neoplasia in asymptomatic average-risk Chinese adults as determined by colonoscopic screening. Gastrointest Endosc 2005;61:547–553. 9. Sung JJ, Chan FK, Leung WK, et al. Screening for colorectal cancer in Chinese: comparison of fecal occult blood test, flexible sigmoidoscopy, and colonoscopy. Gastroenterology 2003;124:608 – 614. 10. Rundle AG, Lebwohl B, Vogel R, et al. Colonoscopic screening in average-risk individuals ages 40 to 49 vs 50 to 59 years. Gastroenterology 2009;134:1311–1315. 11. Imperiale TF, Wagner DR, Lin CY, et al. Using risk for advanced

June 2009

12.

13.

14.

15.

16. 17. 18. 19. 20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

proximal colonic neoplasia to tailor endoscopic screening for colorectal cancer. Ann Intern Med 2003;139:959 –965. Brenner H, Hoffmeister M, Stegmaier C, et al. Risk of progression of advanced adenomas to colorectal cancer by age and sex: estimates based on 840,149 screening colonoscopies. Gut 2007;56:1585–1589. Regula J, Rupinski M, Kraszewska E, et al. Colonoscopy in colorectal-cancer screening for detection of advanced neoplasia. N Engl J Med 2006;355:1863–1872. Rex DK, Lehman GA, Ulbright TM, et al. Colonic neoplasia in asymptomatic persons with negative fecal occult blood tests: influence of age, gender, and family history. Am J Gastroenterol 1993;88:825– 831. Marbet UA, Bauerfeind P, Brunner J, et al. Colonoscopy is the preferred colorectal cancer screening method in a population based program. Endoscopy 2008;40:650 – 655. Fleiss JL. Statistical methods for rates and proportions. New York: Wiley & Sons, 1981. Egger M, Davey-Smith G, Altman DG. Systematic reviews in health care. London: BMJ Publishing Group, 2001. Begg CB, Mazumdar M. Operating characteristics of a rank correlation test for publication bias. Biometrics 1994;50:1088 –1101. Egger M, Smith GD, Schneider M, et al. Bias in meta-analysis detected by a simple, graphical test. Br Med J 1997;315:629 – 634. R Development Core Team. R: a language and environment for statistical computing. Vienna, Austria: R Foundation for Statistical Computing, 2007. Soon MS, Kozarek RA, Ayub K, et al. Screening colonoscopy in Chinese and Western patients: a comparative study. Am J Gastroenterol 2005;100:2749 –2755. Schoenfeld P, Cash B, Flood A, et al. Colonoscopic screening of average-risk women for colorectal neoplasia. N Engl J Med 2005; 352:2061–2068. Lieberman DA, Weiss DG, Bond JH, et al. Use of colonoscopy to screen asymptomatic adults for colorectal cancer: Veterans Affairs Cooperative Study Group 380. N Engl J Med 2000;343: 162–168. Erratum in N Engl J Med 2000;343:1204. Betés M, Muñoz-Navas MA, Duque JM, et al. Use of colonoscopy as a primary screening test for colorectal cancer in average risk people. Am J Gastroenterol 2003;98:2648 –2654. Leung WK, Ho KY, Kim WH, et al. Colorectal neoplasia in Asia: a multicenter colonoscopy survey in symptomatic patients. Gastrointest Endosc 2006;64:751–759. Park DI, Kim YH, Kim HS, et al. Diagnostic yield of advanced colorectal neoplasia at colonoscopy, according to indications: an investigation from the Korean Association for the Study of Intestinal Diseases (KASID). Endoscopy 2006;38:449 – 455. Chen SC, Rex DK. Endoscopist can be more powerful than age and male gender in predicting adenoma detection at colonoscopy. Am J Gastroenterol 2007;102:856 – 861. Lieberman DA, Holub JL, Moravec MD, et al. Prevalence of colon polyps detected by colonoscopy screening in asymptomatic black and white patients. JAMA 2008;300:1417–1422. Callcut RA, Kaufman S, Stone-Newsom R, et al. Gender disparities in colorectal cancer screening: true or false? J Gastrointest Surg 2006;10:1409 –1417. McMichael AJ, Potter JD. Reproduction, endogenous and exogenous sex hormones, and colon cancer: a review and hypothesis. J Natl Cancer Inst 1980;65:1201–1207. Newcomb PA, Zheng Y, Chia VM, et al. Estrogen plus progestin use, microsatellite instability, and the risk of colorectal cancer in women. Cancer Res 2007;67:7534 –7539. Rossouw JE, Anderson GL, Prentice RL, et al. Risks and benefits of estrogen plus progestin in healthy postmenopausal women: principal results from the Women’s Health Initiative randomized controlled trial. JAMA 2002;288:321–333.

GENDER AS A RISK FOR ADVANCED COLON NEOPLASIA

681

33. Terry MB, Neugut AI, Bostick RM, et al. Risk factors for advanced colorectal adenomas: a pooled analysis. Cancer Epidemiol Biomarkers Prev 2002;11:622– 629. 34. Cho E, Smith-Warner SA, Ritz J, et al. Alcohol intake and colorectal cancer: a pooled analysis of 8 cohort studies. Ann Intern Med 2004;140:603– 613. 35. Tsong WH, Koh WP, Yuan JM, et al. Cigarettes and alcohol in relation to colorectal cancer: the Singapore Chinese Health Study. Br J Cancer 2007;96:821– 827. 36. Botteri E, Iodice S, Raimondi S, et al. Cigarette smoking and adenomatous polyps: a meta-analysis. Gastroenterology 2008; 134:388 –395. 37. Jacobs ET, Thompson PA, Martínez ME. Diet, gender, and colorectal neoplasia. J Clin Gastroenterol 2007;41:731–746. 38. Press OA, Zhang W, Gordon MA, et al. Gender-related survival differences associated with EGFR polymorphisms in metastatic colon cancer. Cancer Res 2008;68:3037–3042. 39. Kim SE, Shim KN, Jung SA, et al. An association between obesity and the prevalence of colonic adenoma according to age and gender. J Gastroenterol 2007;42:616 – 623. 40. McGregor SE, Hilsden RJ, Li FX, et al. Low uptake of colorectal cancer screening 3 yr after release of national recommendations for screening. Am J Gastroenterol 2007;102:1727–1735. 41. Ananthakrishnan AN, Schellhase KG, Sparapani RA, et al. Disparities in colon cancer screening in the Medicare population. Arch Intern Med 2007;167:258 –264. 42. Johns LE, Houlston RS. A systematic review and meta-analysis of familial colorectal cancer risk. Am J Gastroenterol 2001;96:2992. 43. Butterworth AS, Higgins JP, Pharoah P. Relative and absolute risk of colorectal cancer for individuals with a family history: a metaanalysis. Eur J Cancer 2006;42:216. 44. Winawer SJ, Fletcher RH, Miller L, et al. Colorectal cancer screening: clinical guidelines and rationale. Gastroenterology 1997; 112:594. 45. Winawer S, Fletcher R, Rex D, et al. Colorectal cancer screening and surveillance: clinical guidelines and rationale—update based on new evidence. Gastroenterology 2003;124:544. 46. Liu HH, Wu MC, Peng Y, et al. Prevalence of advanced colonic polyps in asymptomatic Chinese. World J Gastroenterol 2005; 11:4731– 4734. 47. Anderson JC, Messina CR, Dakhllalah F, et al. Body mass index: a marker for significant colorectal neoplasia in a screening population. J Clin Gastroenterol 2007;41:285–290. 48. Byeon JS, Yang SK, Kim TI, et al. Colorectal neoplasm in asymptomatic Asians: a prospective multinational multicenter colonoscopy survey. Gastrointest Endosc 2007;65:1015–1022. 49. Choe JW, Chang HS, Yang SK, et al. Screening colonoscopy in asymptomatic average-risk Koreans: analysis in relation to age and sex. J Gastroenterol Hepatol 2007;22:1003–1008. 50. Liou JM, Lin JT, Huang SP, et al. Screening for colorectal cancer in average-risk Chinese population using a mixed strategy with sigmoidoscopy and colonoscopy. Dis Colon Rectum 2007;50: 630 – 640.

Reprint requests Address requests for reprints to: Jonathan P. Terdiman, MD, Department of Gastroenterology Box 1623, University of California, San Francisco, San Francisco, California 94143. e-mail: [email protected] ucsf.edu; fax: (415) 502-2249. Conflicts of Interest The authors disclose no conflicts.

June 2009

GENDER AS A RISK FOR ADVANCED COLON NEOPLASIA

681.e1

Supplementary Table 1. Rates of Cancer Study Betes et al Brenner et al

Regula et al

Rundle et al Marbet et al

Age and number of participants Age ⬎40 M: 1649 W:561 Age ⬎55 M: 34,9249 W: 490,900 Age 55–59 M: 66,330 W: 112,399 Age 60–64 M: 110,597 W: 160,880 Age 65–69 M: 97,411 W: 125,133 Age 70–74 M: 47,018 W: 55,582 Age 75–79 M: 21,220 W: 27,414 Age ⱖ80 M: 6673 W: 9492 Age 40–49 M: 889 W: 1503 Age 50–66 M: 13,512 W: 23,801 Age 40–59 M: 688 W: 217 Age 50–80 M: 1023 W: 889

Rates cancer men (Women) 10 (1)

132.4 (82.1) per 100,000 210.5 (113.0) 313.6 (180.1) 441.5 (261.0) 456.9 (351.5) 551.4 (481.2) 9 (7) 187 (143) 1 (0) 11 (0)

Supplementary Table 2. Methods of Individual Studies Study

Assessment of polyp size

Pathology criteria

Number of pathologists

Rex Betes Sung Imperiale Chiu Liu Schoenfeld Soon (a) Soon (b) Regula Anderson Brenner Byeon Choe Kim Liou Rundle Marbet

Not specified After removal Forceps Method chosen by each endoscopist (unavailable) Forceps Forceps or after removal Olympus guidewire Forceps Forceps Forceps or after removal After removal Not specified Forceps Forceps Forceps and after removal Not specified Forceps or after removal Forceps

Standard criteria WHO WHO WHO WHO Not specified Not specified WHO WHO WHO WHO Not specified WHO Not specified Not specified Not specified Not specified Not specified

Single GI pathologist GI pathologists Single pathologist Three pathologists Single pathologist Not specified Single GI pathologist Not specified Not specified Local pathologists each city in Poland Single pathologist Local pathologists Local pathologists each Asian city Not specified Not specified Not specified Not specified Not specified

GI, Gastroenterologist; WHO, World Health Organization.

681.e2

NGUYEN ET AL

CLINICAL GASTROENTEROLOGY AND HEPATOLOGY Vol. 7, No. 6

Supplementary Table 3. Individual Study Statistics

Study Rex

Betes Sung Imperiale

Chiu Liu

Schoenfeld

Soon (a)

Soon (b)

Regula

Anderson Brenner

Byeon Choe

Kim

Number participants by age Age 50–75 M: 310 W: 186 Age 50–54 M: 63 W: 44 Age 55–59 M: 81 W: 53 Age 60–64 M: 77 W: 50 Age 65–75 M: 89 W: 39 Age ⱖ40 M: 1649 W: 561 Age 50–79 M: 224 W: 281 Age ⱖ50 M: 1753 F: 1272 Age 50–54 M: 662 F: 459 Age 55–59 M: 454 F: 332 Age 60–64 M: 274 F: 188 Age ⬎64 M: 363 F: 293 Age ⬍50 to ⱖ70 M: 1034 W: 674 Age ⱖ40 M: 3125 W: 2848 Age 40–49 M: 1348 W: 1308 Age 50–59 M: 1067 W: 836 Age 60–69 M: 480 W: 462 Age ⬎69 M: 230 W: 242 Age 50–79: M: 2206 W: 1198 Age 50–59: N/A Age 60–69: N/A Age 70–79: N/A Age ⱖ40 M: 903 W: 553 Age 40–49 M: 366 W: 199 Age 50–59 M: 315 W: 204 Age 60–69 M: 172 W: 120 Age ⱖ70 M: 50 W: 30 Age ⱖ40 M: 1666 W: 1737 Age 40–49 M: 101 W: 113 Age 50–59 M: 912 W: 956 Age 60–69 M: 425 W: 435 Age ⱖ70 M: 228 W: 233 All subjects (includes family hx) Age 40–66 M: 18,012 W: 32,136 Age 40–49 M: 2646 W: 4460 Age 50–66 M: 15,366 W: 27,676 Subjects without family hx: Age 40–66 M: 14,401 W: 25,304 Age 40–49 M: 889 W: 1503 Age 50–66 M: 13,512 W: 23,801 Age ⱖ40 M: 1356 W: 1137 Age ⱖ55 M: 349,249 W: 490,900 Age 55–59 M: 663,30 W: 112,399 Age 60–64 M: 110,597 W: 160,880 Age 65–69 M: 97,411 W: 125,133 Age 70–74 M: 47,018 W: 55,582 Age 75–79 M: 21,220 W: 27,414 Age ⱖ80 M: 6673 W: 9492 Age 20–90 M: 471 W: 389 Age: 50–84 M: 1623 W: 812 Age 50–59 M: 1064 W: 523 Age 60–69 M: 479 W: 254 Age ⱖ70 M: 80 W: 35 Age ⱖ30 M: 946 W: 798 Age 30–39 M: 149 W: 127 Age 40–49 M: 275 W: 219 Age 50–59 M: 228 W: 177 Age 60–69 M: 202 W: 172 Age ⬎69 M: 92 W: 103

Relative risk men/women (95% Confidence Interval)

Number needed to screen age 50–59 to detect one advanced neoplasm men (women)

3.17 (1.44–6.97)

12 (50)

2.05 (1.47–2.86) 1.79 (1.11–2.86) 2.48 (1.77–3.49)

N/A N/A 18 (47)

1.43 (0.80–2.56) 1.51 (1.14–2.00)

N/A 28 (36)

1.91 (1.42–2.56) 1.62 (0.90–2.78) 2.10 (1.30–3.42) 0.88 (0.46–1.90) 2.94 (1.50–5.76)

21 (34)

1.74 (1.25–2.31)

21 (38)

1.87 (1.74–2.00)

N/A

10.6 (8.2) 7.93 (4.55) 7.5 (4.2) 8.4 (4.8) 9.2 (5.8) 9.7 (6.5) 9.5 (7.3)

1.30 (1.01–1.66) 1.74 (1.71–1.77)

N/A N/A

5.9 (2.8) 5.1 (2.0) 3.8 (1.7) 7.1 (2.4) 11.3 (2.9) 6.0 (2.6) 2.0 (0) 2.9 (1.4) 8.3 (2.8) 11.8 (3.5) 3.3 (6.8)

2.11 (1.06–4.19) 2.60 (1.53–4.40)

N/A 26 (59)

2.29 (1.40–3.74)

12 (36)

% Rates advanced neoplasia men (women) 12 (4) 10 (2) 7 (2) 16 (4) 15 (10) 13.5 (6.6) 16.5 (9.3) 7.99 (3.22) 4.83 (1.52) 6.82 (3.01) 12.0 (3.72) 12.1 (5.80) 3.4 (2.4) 3.9 (2.6) 1.3 (1.2) 3.6 (2.8) 8.3 (4.1) 12.2 (7.0) 8.6 (4.5) 4.7 (2.9) 10.6 (5.0) 10.6 (11.8) 5.3 (1.8) 3.6 (1.5) 5.7 (2.0) 5.2 (2.5) 16.0 (0) 6.2 (3.6) 3.0 (3.5) 4.8 (2.6) 7.5 (4.8) 10.5 (5.6) 7.93 (4.25) 4.50 (2.78) 8.53 (4.49) 7.7 (4.2)

18 (50)

3.4 (1.9) 8.0 (4.3)

June 2009

GENDER AS A RISK FOR ADVANCED COLON NEOPLASIA

681.e3

Supplementary Table 3. Continued

Study Liou

Rundle

Marbet

Number participants by age Age Age Age Age Age Age Age Age Age Age Age

ⱖ50 M: 1193 W: 913 50–59: 1179 60–69: 686 ⬎69: 241 40–59 M: 688 W: 217 40–49 M: 417 W: 136 50–59 M: 271 W: 81 50–80 M: 1023 W: 889 50–59 M: 495 W: 412 60–69 M: 375 W: 345 70–80 M: 153 W: 132

N/A, Not Available; M, Men; W, Women.

% Rates advanced neoplasia men (women) 6.7 (3.7) 5.4 (2.3) 7.0 (3.8) 12 (11) 2.62 (2.76) 1.92 (2.21) 3.69 (3.70) 11.7 (4.72) 8.3 (2.4) 13.9 (6.1) 17.6 (8.3)

Relative risk men/women (95% Confidence Interval)

Number needed to screen age 50–59 to detect one advanced neoplasm men (women)

1.86 (1.23–2.80) 2.44 (1.26–4.75) 1.88 (0.92–3.85) .837 (0.49–2.44) 0.9 (0.4–2.4)

19 (43)

3.46 (1.75–6.84)

12 (42)

27 (27)

Lihat lebih banyak...

Comentários

Copyright © 2017 DADOSPDF Inc.