PMCCPMCCPMCC

Search tips
Search criteria 

Advanced

 
Logo of medicineHomeSearchSubmit a ManuscriptMedicine
 
Medicine (Baltimore). 2017 April; 96(16): e6396.
Published online 2017 April 21. doi:  10.1097/MD.0000000000006396
PMCID: PMC5406047

Association between diabetes mellitus and subsequent ovarian cancer in women

A systematic review and meta-analysis of cohort studies
Monitoring Editor: Jongwha Chang.

Abstract

Epidemiologic studies have suggested that diabetes mellitus (DM) might be associated with risk of ovarian cancer; however, the results have been inconsistent. The aim of this study was to determine the relationship between DM and the incidence of ovarian cancer on the basis of cohort studies.

Relevant studies from PubMed, Embase, and the Cochrane Library until September 2016 were collected. The summary risk ratio (RR) was used as the effect measure in a random effects model. Sensitivity analysis, subgroup analysis, and calculation of publication bias were conducted.

Thirteen articles including 14 cohorts comprising a total of 3708, 313 women and reporting 5534 cases of ovarian cancer were included. The summary RR suggested that patients with DM had a higher risk of ovarian cancer than patients without DM (RR: 1.19; 95% confidence interval: 1.06–1.34; P = .004), and no evidence of publication bias was found. The subgroup analysis indicated a higher incidence of ovarian cancer in patients with DM in studies published after 2010, studies not conducted in Europe or the United States, studies that did not adjust for body mass index or smoking status, and studies with lower Newcastle–Ottawa Scale scores.

The present findings indicated that DM is a risk factor for ovarian cancer, and future large-scale epidemiologic studies should be performed to evaluate this relation in specific populations.

Keywords: diabetes mellitus, meta-analysis, ovarian cancer, systematic review

1. Introduction

Ovarian cancer is the 5th leading cause of death among malignancies and accounting for approximately 240,000 cases and 150,000 deaths in 2012 worldwide.[1] Nearly 2/3 of cases are diagnosed in its advanced stages or unstaged and just 30% for 5-year survival rate for these patients.[2,3] The survival rates are poor, owing to the lack of effective screening strategies.[4] This emphasizes the need to focus on identifying risk factors, in order to reduce the risk of ovarian cancer. The associations among dietary flavonoids, flavonoid subclasses, bilateral salpingectomy, depression and anxiety, dietary fat and fatty acid, overweight, or obesity and the incidence of ovarian cancer have been evaluated in previous meta-analyses.[59] However, several other factors associations with the risk of ovarian cancer are not precisely understood and remain controversial.

Diabetes mellitus (DM) is a growing global pandemic affecting approximately 3.0% to 4.0% of adults worldwide.[10] Systematic reviews and meta-analyses have already evaluated the risk of cancer incidence at different sites.[1115] A previous study indicated that women with DM have a moderately increased risk of developing ovarian cancer.[16] However, traditional case–control studies were included in the previous study, which are less strong than cohort studies, and the findings of stratified analyses were affected by differences in study design. Furthermore, whether this relation differs according to the characteristics of participants remains unclear.

Several prospective cohort studies that explored the relationship between DM and ovarian cancer risk have already been published.[1729] Several studies suggested that DM is associated with an elevated risk of ovarian cancer,[19,28] whereas other studies showed no evidence for this association.[17,18,2027,29] Clarifying any potential correlation between DM and ovarian cancer, which has not been definitively established, is particularly important in the general population. We therefore attempted a comprehensive examination of the available cohort studies to determine the association between DM and the incidence of ovarian cancer.

2. Materials and methods

2.1. Data sources, search strategy, and selection criteria

This study was conducted and reported according to the Meta-analysis of Observational Studies in Epidemiology protocol.[30] Any prospective observational study that examined the relationship between DM and ovarian cancer was eligible for inclusion in this review. The included articles were restricted to those published in English; however, there was no restriction on publication status (published, in press, or in progress). Electronic databases including PubMed, Embase, and the Cochrane Library were systematically searched by using the combination of the search terms (diabetes or DM or glucose) and (cancer, carcinoma, neoplasm, and tumor) and ovarian, with the deadline being September 2016. To identify unpublished studies or collect updated information in some of the included studies that could provide useful data, we contacted the authors, reviewed abstracts and presentations of recent major meetings, and searched for ongoing studies that were registered as completed but not yet published. Manual searches of reference lists from all the relevant articles were conducted to identify additional potential eligible studies. All analyses were based on previous published studies, thus no ethical approval and patient consent are required.

After the literature search, the selected studies were independently reviewed by 2 of the authors, and a group discussion settled any inconsistencies until a consensus was reached. A study was eligible for inclusion if the following criteria were met: the study had a prospective cohort design; the study investigated the association between DM and the risk of ovarian cancer; and the study reported an effect estimate (risk ratio [RR], hazard ratio, or odds ratio) and 95% confidence intervals (CIs) for comparisons of DM and non-DM participants. We excluded all retrospective case–control studies owing to various confounding factors that could bias the results. We included the latest outcomes if there were cases of overlapping reports.

2.2. Data extraction and quality assessment

The data extraction and quality assessment processes were conducted independently by 2 of the authors. Any discrepancy was resolved through a discussion or by consulting the corresponding author. Information was extracted from the included studies and recorded in electronic Excel sheets. Data were extracted on the basis of the first author, publication year, country, sample size, mean age at baseline, body mass index (BMI), number of ovarian cancer cases, follow-up duration, and adjusted factors. For studies that reported several multivariable adjusted effect estimates, we selected the maximally adjusted estimates for potential confounders. The quality of the included studies was appraised by using the Newcastle–Ottawa scale (NOS).[31] This scale was categorized into 3 groups of items: selection (representativeness of the exposed cohort; selection of the nonexposed cohort; ascertainment of DM; demonstration that ovarian cancer was not present at start of study; 4 points), comparability (comparability on the basis of the design or analysis; 2 points), and outcome (assessment of outcome; adequate follow-up duration; adequate follow-up rate; 3 points), with a total of 9 points. In this review, studies with a total of score of 8 or 9 were considered to be of high quality.

2.3. Statistical analysis

We examined the relationship between DM and the risk of ovarian cancer on the basis of the effect estimate (RR, hazard ratio, or odds ratio) and its 95% CI published in each study. The random effects model was used for data pooling of DM versus non-DM participants.[32] The heterogeneity was tested by using the Q statistics and I2 test. I2 values of <25%, 25% to 50%, 50% to 75%, and >75% were considered to indicate no, low, moderate, and high heterogeneity, respectively. P < .05 or I2 > 50% was deemed to denote a significant heterogeneity.[33,34] Subgroup analyses were conducted for ovarian cancer incidence on the basis of publication year, country, mean age, follow-up duration, adjusted BMI, adjusted smoking status, and study quality. The P values for heterogeneity between subgroups were calculated by using the χ2 test.[35] Sensitivity analysis was conducted by removing a single study from the overall analysis to evaluate the impact of an individual study.[36] Visual inspections of funnel plots for ovarian cancer were conducted. The Egger[37] and Begg[38] tests were employed to statistically assess publication bias for ovarian cancer. All reported P values were 2-sided, and P values < .05 were considered statistically significant for all included studies. Statistical analyses were performed by using STATA software (version 10.0; Stata Corporation, College Station, TX).

3. Results

Among the 2042 studies that were identified through searching PubMed, Embase, and the Cochrane library, 1984 duplicates (n = 1152) and irrelevant (n = 832) records were first excluded. Fifty-eight full-text studies were assessed for eligibility. Furthermore, 9 studies with a case–control design, 13 studies showing insufficient data, and 23 epidemiology studies on other risk factors were excluded. Finally, 13 studies including 14 cohorts were included for qualitative synthesis. The selection process is presented in Fig. Fig.11.

Figure 1

Study selection process.

Table Table11 summarizes the general characteristics of the included studies. Of the 13 included studies, 5 were conducted in Europe,[17,2022,27] 2 were conducted in the United States or Canada,[18,24] and the remaining 6 were conducted in Asia.[19,23,25,26,28,29] The number of participants ranged from 11,686 to 1,152,122, with a total of 3,708,313 women. The mean age of participants ranged from 44.0 to 62.6 years, and the follow-up duration ranged from 3.5 to 15.0 years. The NOS score of the included studies ranged from 6 to 8, with 5 studies being of high quality and the remaining 8 studies being of low quality (Table (Table1).1). Overall, 5 studies had a score of 8, 6 studies had a score of 7, and the remaining 2 studies had a score of 6.

Table 1

Baseline characteristic of studies included.

All of the included studies reported the relationship between DM and ovarian cancer incidence. The summary result is shown in Fig. Fig.2,2, and we noted that women with DM had a higher risk of developing ovarian cancer when compared with non-DM participants (RR: 1.19; 95% CI: 1.06–1.34; P = .004; Fig. Fig.2),2), and a significant heterogeneity was observed (I2 = 43.8%; P = .040). As a result, a sensitivity analysis was conducted, and after each study was sequentially excluded from the pooled analysis, the conclusion was not affected by the exclusion of any specific study (Table (Table22).

Figure 2

Association between diabetes mellitus and the risk of ovarian cancer. CI = confidence interval, RR = risk ratio.

Table 2

Sensitivity analysis for ovarian cancer incidence.

Q-test for the analysis showed a value of P < .05 for ovarian cancer. Therefore, we conducted subgroup analyses to minimize the heterogeneity among the included studies and evaluate this relationship in specific subsets (Table (Table3).3). Overall, we noted that women with DM had an elevated risk of ovarian cancer in studies published after 2010 (RR: 1.18; 95% CI: 1.04–1.34; P = .010), studies not conducted in Europe or the United States (RR: 1.26; 95% CI: 1.06–1.50; P = .008), studies that did not adjust for BMI (RR: 1.15; 95% CI: 1.04–1.27; P = .007) or smoking status (RR: 1.20; 95% CI: 1.05–1.35; P = .005), and studies with lower NOS scores (RR: 1.22; 95% CI: 1.07–1.38; P = .003). No other significant difference was observed between the presence and absence of DM for ovarian cancer based on predefined factors.

Table 3

Subgroup analysis.

A review of the funnel plots could not rule out the potential for publication bias for ovarian cancer (Fig. (Fig.3).3). The Egger and Begg test results showed no evidence of publication bias (P value for Egger = .378; P value for Begg = .381).

Figure 3

Funnel plot for ovarian cancer.

4. Discussion

The purpose of the present meta-analysis was to determine the correlates of DM and the incidence of ovarian cancer. Thirteen cohort studies that included 3,708,313 women were identified. The results showed that DM is associated with an increased risk of ovarian cancer. The findings of sensitivity analysis were consistent with the overall analysis. This result will help in better defining the risk factors of ovarian cancer, and could help physicians in evaluating the risk score of the general population.

The impact factor for the incidence of ovarian cancer has already been reported in previous meta-analyses. Hua et al[5] reported that consumption of dietary flavonoids and their subtypes (isoflavones, flavonols) has a protective effect against ovarian cancer (reduces the risk of ovarian cancer); however, flavone consumption had no such effect. Yoon et al[6] found that removal of the fallopian tubes is an effective approach to protect against ovarian cancer risk in the general population. Liu et al[9] indicated that a high body weight might have a harmful impact on the risk of ovarian cancer, especially for premenopausal women with severe obesity. However, the impact of other chronic diseases such as DM was not confirmed. Few prospective cohort studies have confirmed the relationship between DM and ovarian cancer. Chodick et al[19] conducted a retrospective cohort study and found that DM increased the ovarian cancer risk by 139% after an average of 8 years follow-up. Dankner et al[28] indicated that DM posed an increased risk of ovarian cancer, with the RRs being particularly elevated during the first year after the diabetes diagnosis; however, their study had a shorter follow-up duration of between 1 and 2 years. These 2 cohort studies were specifically conducted in Israel. The increased risk of ovarian cancer could be due to perturbations in insulin, insulin-like growth factors, gonadotropin, and steroid hormone metabolism, which could affect cell proliferation.[39,40] Furthermore, DM might play an important role in endometrioid cell histologic subtypes.

The subgroup analyses suggested that DM had a harmful impact on ovarian cancer in studies not conducted in Europe or the United States. The reason for this could be that the prevalence of DM was higher in Asia, and the diagnosis of DM took a longer time than in Western countries. In studies that did not adjust for BMI or smoking status, a significant difference was found between the presence and absence of DM for ovarian cancer, which might have biased the true correlates of DM and ovarian cancer. Progesterone and leptin may be potential endocrine mediators of the risk of ovarian cancer, and obesity was associated with the levels of insulin, androgens, and free insulin-like growth factor-I, which play an important role in the risk of ovarian cancer.[40,41] Furthermore, Faber et al[42] suggested that current smokers had an increased risk of developing invasive mucinous and borderline mucinous ovarian tumors, and former smokers also have an increased risk of developing borderline serous ovarian tumors. In addition, we noted DM was associated with higher incidence of ovarian cancer if the study published after 2010. The reason for this could be mostly included studies published after 2010 (9/13). The relationship between DM and ovarian cancer in study published before 2010 was available in few studies, and the follow-up, number of participants were lower than study published after 2010, which always acquired broad CIs, that is, no statistically significant difference. In this study, we intended to conduct a stratified analysis based on these factors; however, data in each subset were not available. In addition, for several other subpopulations, the findings might be unstable owing to the smaller number of included studies. We therefore provided relative results and a synthetic review.

The present meta-analysis has some limitations. First, a language bias may exist in the selection of studies published only in English. Second, the types of DM and ovarian cancer were not available, which could have introduced confounder biases. Third, the heterogeneity among included studies was relatively high, and the heterogeneity in subgroup analysis was also higher. Fourth, the relationship between DM and ovarian cancer in specific subpopulations was not available. Finally, the adjusted models, which might play an important role in the development of ovarian cancer, differed among the included studies.

5. Conclusions

The present meta-analysis demonstrated that women with DM had an increased risk of developing ovarian cancer, especially among Asians. To lower the risk of ovarian cancer, any potential risk factors need to be investigated, in order to allow early diagnosis and treatment. Additional epidemiologic studies about this relationship in specific populations need to be further conducted.

Footnotes

Abbreviations: BMI = body mass index, CI = confidence interval, DM = diabetes mellitus, RR = risk ratio.

The authors have no funding and conflicts of interest to disclose.

References

[1] Siegel R, Ma J, Zou Z, et al. Cancer statistics, 2014. CA Cancer J Clin 2014;64:9–29. [PubMed]
[2] Ferlay J, Soerjomataram I, Dikshit R, et al. Cancer incidence and mortality worldwide: sources, methods and major patterns in GLOBOCAN 2012. Int J Cancer 2015;136:E359–86. [PubMed]
[3] Hankinson SE, Danforth KN. Schottenfeld D, Fraumeni J, editors. Ovarian cancer. Cancer Epidemiology and Prevention 3rd EdnNew York, NY: Oxford University Press; 2006. 1013–26.
[4] Heintz AP, Odicino F, Maisonneuve P, et al. Carcinoma of the ovary. FIGO 26th Annual Report on the Results of Treatment in Gynecological Cancer. Int J Gynaecol Obstet 2006;95Suppl 1:S161–92. [PubMed]
[5] Hua X, Yu L, You R, et al. Association among dietary flavonoids, flavonoid subclasses and ovarian cancer risk: a meta-analysis. PLoS One 2016;11:e0151134. [PMC free article] [PubMed]
[6] Yoon SH, Kim SN, Shim SH, et al. Bilateral salpingectomy can reduce the risk of ovarian cancer in the general population: A meta-analysis. Eur J Cancer 2016;55:38–46. [PubMed]
[7] Watts S, Prescott P, Mason J, et al. Depression and anxiety in ovarian cancer: a systematic review and meta-analysis of prevalence rates. BMJ Open 2015;5:e007618. [PMC free article] [PubMed]
[8] Hou R, Wu QJ, Gong TT, et al. Dietary fat and fatty acid intake and epithelial ovarian cancer risk: evidence from epidemiological studies. Oncotarget 2015;6:43099–119. [PMC free article] [PubMed]
[9] Liu Z, Zhang TT, Zhao JJ, et al. The association between overweight, obesity and ovarian cancer: a meta-analysis. Jpn J Clin Oncol 2015;45:1107–15. [PubMed]
[10] Palmer SC, Mavridis D, Navarese E, et al. Comparative efficacy and safety of blood pressure-lowering agents in adults with diabetes and kidney disease: a network meta-analysis. Lancet 2015;385:2047–56. [PubMed]
[11] Bao C, Yang X, Xu W, et al. Diabetes mellitus and incidence and mortality of kidney cancer: a meta-analysis. J Diabetes Complications 2013;27:357–64. [PubMed]
[12] McAuliffe JC, Christein JD. Type 2 diabetes mellitus and pancreatic cancer. Surg Clin North Am 2013;93:619–27. [PubMed]
[13] Sekikawa A, Fukui H, Maruo T, et al. Diabetes mellitus increases the risk of early gastric cancer development. Eur J Cancer 2014;50:2065–71. [PubMed]
[14] Xu H, Jiang HW, Ding GX, et al. Diabetes mellitus and prostate cancer risk of different grade or stage: a systematic review and meta-analysis. Diabetes Res Clin Pract 2013;99:241–9. [PubMed]
[15] Kramer HU, Schottker B, Raum E, et al. Type 2 diabetes mellitus and colorectal cancer: meta-analysis on sex-specific differences. Eur J Cancer 2012;48:1269–82. [PubMed]
[16] Lee JY, Jeon I, Kim JW, et al. Diabetes mellitus and ovarian cancer risk: a systematic review and meta-analysis of observational studies. Int J Gynecol Cancer 2013;23:402–12. [PubMed]
[17] Stattin P, Bjor O, Ferrari P, et al. Prospective study of hyperglycemia and cancer risk. Diabetes Care 2007;30:561–7. [PubMed]
[18] Gapstur SM, Patel AV, Diver WR, et al. Type II diabetes mellitus and the incidence of epithelial ovarian cancer in the cancer prevention study-II nutrition cohort. Cancer Epidemiol Biomarkers Prev 2012;21:2000–5. [PubMed]
[19] Chodick G, Heymann AD, Rosenmann L, et al. Diabetes and risk of incident cancer: a large population-based cohort study in Israel. Cancer Causes Control 2010;21:879–87. [PubMed]
[20] Wotton CJ, Yeates DG, Goldacre MJ. Cancer in patients admitted to hospital with diabetes mellitus aged 30 years and over: record linkage studies. Diabetologia 2011;54:527–34. [PubMed]
[21] Lambe M, Wigertz A, Garmo H, et al. Impaired glucose metabolism and diabetes and the risk of breast, endometrial, and ovarian cancer. Cancer Causes Control 2011;22:1163–71. [PubMed]
[22] Rapp K, Schroeder J, Klenk J, et al. Fasting blood glucose and cancer risk in a cohort of more than 140,000 adults in Austria. Diabetologia 2006;49:945–52. [PubMed]
[23] Inoue M, Iwasaki M, Otani T, et al. Diabetes mellitus and the risk of cancer: results from a large-scale population-based cohort study in Japan. Arch Intern Med 2006;166:1871–7. [PubMed]
[24] Johnson JA, Bowker SL, Richardson K, et al. Time-varying incidence of cancer after the onset of type 2 diabetes: evidence of potential detection bias. Diabetologia 2011;54:2263–71. [PubMed]
[25] Lo SF, Chang SN, Muo CH, et al. Modest increase in risk of specific types of cancer types in type 2 diabetes mellitus patients. Int J Cancer 2013;132:182–8. [PubMed]
[26] Khan M, Mori M, Fujino Y, et al. Site-specific cancer risk due to diabetes mellitus history: evidence from the Japan Collaborative Cohort (JACC) Study. Asian Pac J Cancer Prev 2006;7:253–9. [PubMed]
[27] Bjorge T, Lukanova A, Tretli S, et al. Metabolic risk factors and ovarian cancer in the Metabolic Syndrome and Cancer project. Int J Epidemiol 2011;40:1667–77. [PubMed]
[28] Dankner R, Boffetta P, Balicer RD, et al. Time-dependent risk of cancer after a diabetes diagnosis in a cohort of 2.3 million adults. Am J Epidemiol 2016;183:1098–106. [PubMed]
[29] Chen HF, Chang YH, Ko MC, et al. A large scale population-based cohort study on the risk of ovarian neoplasm in patients with type 2 diabetes mellitus. Gynecol Oncol 2014;134:576–80. [PubMed]
[30] Stroup DF, Berlin JA, Morton SC, et al. Meta-analysis of observational studies in epidemiology: a proposal for reporting. Meta-analysis of Observational Studies in Epidemiology (MOOSE) group. JAMA 2000;283:2008–12. [PubMed]
[31] Wells GA, Shea B, O’Connell D. The Newcastle-Ottawa Scale (NOS) for assessing the quality of nonrandomised studies in meta-analyses. http://wwwohrica/programs/clinical_epidemiology/oxfordasp. Accessed October 21, 2016.
[32] Ades AE, Lu G, Higgins JP. The interpretation of random-effects meta-analysis in decision models. Med Decis Making 2005;25:646–54. [PubMed]
[33] Higgins JP, Thompson SG, Deeks JJ, et al. Measuring inconsistency in meta-analyses. BMJ 2003;327:557–60. [PMC free article] [PubMed]
[34] Higgins JP, Thompson SG. Quantifying heterogeneity in a meta-analysis. Stat Med 2002;21:1539–58. [PubMed]
[35] Deeks JJ, Altman DG, Bradburn MJ. Egger M, Davey Smith G, Altman DG, editors. Statistical methods for examining heterogeneity and combining results from several studies in meta-analysis. Systematic Reviews in Health Care: Metaanalysis in Context 2nd edLondon: BMJ Books; 2001. 285–312.
[36] Tobias A. Assessing the influence of a single study in the meta-analysis estimate. Stata Technical Bulletin 1999;8:7526–9.
[37] Egger M, Davey Smith G, Schneider M, et al. Bias in meta-analysis detected by a simple, graphical test. BMJ 1997;315:629–34. [PMC free article] [PubMed]
[38] Begg CB, Mazumdar M. Operating characteristics of a rank correlation test for publication bias. Biometrics 1994;50:1088–101. [PubMed]
[39] Beauchamp MC, Yasmeen A, Knafo A, et al. Targeting insulin and insulin-like growth factor pathways in epithelial ovarian cancer. J Oncol 2010;2010:257058. [PMC free article] [PubMed]
[40] Risch HA. Hormonal etiology of epithelial ovarian cancer, with a hypothesis concerning the role of androgens and progesterone. J Natl Cancer Inst 1998;90:1774–86. [PubMed]
[41] Calle EE, Kaaks R. Overweight, obesity and cancer: epidemiological evidence and proposed mechanisms. Nat Rev Cancer 2004;4:579–91. [PubMed]
[42] Faber MT, Kjær SK, Dehlendorff C, et al. Cigarette smoking and risk of ovarian cancer: a pooled analysis of 21 case-control studies. Cancer Causes Control 2013;24:989–1004. [PMC free article] [PubMed]

Articles from Medicine are provided here courtesy of Wolters Kluwer Health