Next Article in Journal
Updates in Management of SARS-CoV-2 Infection
Previous Article in Journal
Prospective Association between Multimorbidity and Falls and Its Mediators: Findings from the Irish Longitudinal Study on Ageing
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
JCM
Volume 11
Issue 15
10.3390/jcm11154469
jcm-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles thumb_up
...
Endorse textsms
...
Comment
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Metformin, Statin Use, and Female Colorectal Cancer: A Population-Based Cohort Study in Taiwan

by
Shu-Hua Hsu
1,
De-Kai Syu
2,
Yu-Chiao Wang
3,
Chih-Kuang Liu
3,4,* and
Ming-Chih Chen
5,*
1
Department of Family Medicine, Fu Jen Catholic University Hospital, Fu Jen Catholic University, New Taipei City 243, Taiwan
2
Department of Orthopedics, Fu Jen Catholic University Hospital, Fu Jen Catholic University, New Taipei City 243, Taiwan
3
Graduate Institute of Business Administration, College of Management, Fu Jen Catholic University, New Taipei City 242, Taiwan
4
Department of Urology, Fu Jen Catholic University Hospital, Fu Jen Catholic University, New Taipei City 243, Taiwan
5
Artificial Intelligence Development Center, Fu Jen Catholic University, New Taipei City 242, Taiwan
*
Authors to whom correspondence should be addressed.
J. Clin. Med. 2022, 11(15), 4469; https://doi.org/10.3390/jcm11154469
Submission received: 17 June 2022 / Revised: 15 July 2022 / Accepted: 26 July 2022 / Published: 31 July 2022
(This article belongs to the Section Gastroenterology & Hepatopancreatobiliary Medicine)
Browse Figures
Review Reports Versions Notes

Abstract

:
In the last few years, the incidence of colorectal cancer (CRC) in women has gradually increased. However, epidemiological studies on the relationship between type II diabetes mellitus (T2DM) and female CRC and the effect of metformin or statins on female CRC are insufficient. To determine their association, we conducted a population-based cohort study on women in Taiwan. We collected data on a total of 396,521 women aged 40 to 64 years old from 1 January 2007 to 31 December 2009 from the National Health Insurance Research Database. We followed up on all participants in the cohort until the occurrence of CRC, the date for all death, or 31 December 2015. Full development of CRC was identified using the International Classification of Disease (ICD), 9th Revision, code 153. We estimated hazard ratios (HRs) and corresponding 95% confidence intervals (CIs) using the Cox proportional hazards model. Both metformin (adjusted hazard ratio, aHR = 1.12; 95% CI: 0.934–1.335, p = 0.227) and statin (aHR = 1.03; 95% CI: 0.906–1.172, p = 0.645) use showed no association with female CRC in a multivariate analysis. The findings indicate that metformin and statin use showed no protective effect against female colorectal cancer (CRC). An additional randomized trial is necessary to investigate the effect of metformin and statin use in CRC prevention.

1. Introduction

Colorectal cancer (CRC) is the third most common malignancy and the second most common cause of cancer death worldwide [1]. The epidemiological characteristics and risk factors of major types of cancers vary in each country [2]. In Taiwan, malignant neoplasms have been the leading cause of death since 1982 [3]. As the Taiwanese lifestyle has become more westernized, the mortality rate from diseases such as CRC has gradually increased for males and females since 1971 [2]. CRC is primarily considered a “lifestyle” disease. Demographic factors including age, gender, genetics, the consumption of high-calorie foods and diets high in animal fat, alcohol consumption, and obesity are considered to be potential risk factors [4]. Vulnerability differences in CRC may be divergent due to gender and sex hormones [5]. CRC is the second most common cancer in females in Taiwan [6]. For women, the lifetime risk of developing colorectal cancer is 6% and the five-year survival rate of colorectal cancer among women is lower than that among men [7]. The above epidemiological evidence implies that colorectal cancer is a major health threat for women, although it has often been underestimated.
Type 2 diabetes mellitus (T2DM) is a risk factor for solid malignancies such as liver, pancreatic, colon, breast, bladder, and endometrial cancer [8,9,10,11]. The cancer-causing mechanisms in diabetes are complex, including excessive reactive oxygen species (ROS) formation, chronic inflammation, and impaired healing phenomena, collectively leading to carcinogenesis under diabetic conditions [12,13]. Metformin is a first-line anti-diabetic drug, and study results on the relationship between metformin use and CRC are controversial and conflicting [14]. A case-control study showed that there was no association between metformin use and the incidence of CRC in T2DM patients (hazard ratio (HR) 0.96; 95% confidence intervals (CI) 0.88–1.04) [15]. However, a meta-analysis suggested that metformin users had a significantly lower incidence of CRC (relative risk (RR) 0.76, 95% CI: 0.69–0.84, p < 0.001) compared to non-metformin users [16].
Epidemiological studies have reported that increased cholesterol levels are associated with higher cancer incidence, including CRC [17,18,19,20]. Statins remain the first-line treatment for managing dyslipidemia as they can inhibit the cell cycle and induce apoptosis; however, there has also been debate as to the effect of statins on CRC. [21]. Lee et al. found that compared to non-users, the adjusted hazard ratio (aHR) for risk of CRC in high statin users was 0.56 (CI: 0.42–0.75) in men and 0.64 (CI: 0.46–0.90) in women [22]. To our knowledge, there are controversial results and a lack of conclusive evidence in previous literature about the influence of metformin and statin on CRC. Therefore, this study aimed to investigate the effect of metformin and statin use on female CRC.

2. Materials and Methods

2.1. Data Source and Study Population

This was a propensity score-matched, retrospective cohort study based on a population of women aged 40–64 years in Taiwan. The National Health Insurance (NHI) data. base, the Taiwan Cancer Registry (TCR), and the National Cause of Death Registry (NCDR) database, which are all available from the Health and Welfare Data Science Center (HWDC), were used in this study. The NHI database is an electronic, de-identified, administrative healthcare database containing the medical records of all outpatients and inpatients [23]. The national population-based TCR incorporates the diagnosis date of pathologically confirmed cancer cases into the registry and has high-quality data [24]. Following the center’s ethical guidelines, patients’ personal information was anonymized before we accessed it; consequently, the Research Ethics Committee waived the requirement for informed consent. This study was approved by the Institutional Review Board (IRB) and Ethics Committee of Fu Jen Catholic University (IRB approval no.: C107099).

2.2. Participant Selection

Figure 1 shows a flowchart of the selection procedure used in this study. A total of 4,018,833 Taiwanese women aged 40 to 64 (as CRC is rare in young people) who sought medical treatment from 2007 to 2009 were initially enrolled in the cohort and were divided into diabetic and non-diabetic groups based on disease code and medication prescription. The diabetic group was defined by records of at least three outpatient or inpatient diabetes diagnoses and prescribed antidiabetic medications within one year. The definition was according to the previous study from the Taiwan NHI Research Database [25], and the first diagnosis date was regarded as the index date. The other patients were classified into the non-diabetic group, and the date of their first visit to the doctor was regarded as the index date. Patients who had had type 1 DM and CRC prior to the index date were excluded from our study. Definitions related to disease diagnosis and drugs are listed in Appendix A Table A1, Table A2 and Table A3 based on a previously published study in Taiwan [26].

2.3. Propensity Score Matching and Covariates

To reduce the effects of potential confounders and sampling bias in the diabetic and non-diabetic groups, this research used a 1:4 ratio in propensity score matching (PSM). This is a common technique for selecting controls with identical background covariates to minimize differences among groups of study participants. The matching variables used were age, comorbidities (hypertension, chronic obstructive pulmonary disease (COPD), asthma, stroke, nephropathy, ischemic heart disease (IHD), peripheral arterial disease (PAD), eye disease, dyslipidemia, and obesity), concomitant medications (fibrates, angiotensin-converting enzyme inhibitor (ACEI), angiotensin receptor blocker (ARB), calcium channel blockers (CCB), aspirin, dipyridamole, clopidogrel, ticlopidine, and other non-steroidal anti-inflammatory drugs (NSAIDs)), and potential colon cancer detection examinations (definitions are listed in Appendix A Table A1, Table A2 and Table A3 and see Table 1). After PSM, the final cohort of patients with T2DM comprised 86,992 patients, whereas the comparison cohort contained 309,529 patients.

2.4. Statistical Analyses

Basic demographic characteristics were defined as categorical (presented as N (%)) and continuous variables (presented as mean ± standard deviation (SD)) in this study. The primary outcome of this study was the end point of CRC (ICD, 9th Revision, code 153). The person-years of follow-up (censoring time) for each participant were calculated from the index date until the occurrence of CRC, the date for all death, or the last date of linked data available from the TCR and NCDR (31 December 2015), whichever occurred first.
We used a Cox proportional hazards regression to calculate hazard ratios (HRs) with 95% confidence intervals (CIs) between each risk factor and CRC. Variables were considered for more than three records in one year during the five years before the index date, including hypertension, chronic obstructive pulmonary disease (COPD), asthma, stroke, nephropathy, ischemic heart disease (IHD), peripheral arterial disease (PAD), eye disease, dyslipidemia, obesity, statin, fibrates, angiotensin-converting enzyme inhibitor (ACEI), angiotensin receptor blocker (ARB), calcium channel blockers (CCB), aspirin, dipyridamole, clopidogrel, ticlopidine, non-steroidal anti-inflammatory drugs (NSAIDs), sulfonylurea, insulin, acarbose, thiazolidinedione (TZD), potential colon cancer detection examinations, and metformin use (see Appendix A Table A1, Table A2 and Table A3).
Even if PSM is applied, residual imbalance might still exist in a population [27], and a multivariate Cox regression analysis should still be performed. Therefore, we performed a multivariate analysis for variables that were statistically significant in univariate analysis. The results of the 95% CI of HR were also presented visually as forest plots. The life table method was used to estimate the cumulative incidence of CRC per year of follow-up among subjects who had used metformin and statins prior to enrollment. All p values < 0.05 were considered statistically significant. The SAS statistical package (version 9.4; SAS Institute, Inc., Cary, NC, USA) and STATA (version 16.0; StataCorp, College Station, TX, USA) were used for all data analyses.

3. Results

A total of 396,521 women were included in this study. The number of CRCs developed during the observation period was 665 in the diabetes group and 2192 in the non-diabetes group (Figure 1). Table 1 displays the baseline characteristics of the female patients aged 40 to 64 with and without DM. After propensity score matching with index year, age, multiple chronic diseases, and drugs, there was little difference between the non-DM and DM groups. The prevalence of nephropathy, PAD, and eye disease was higher in the DM group compared to the non-DM group. These factors were comorbidities of DM, which may cause a confounding effect in this study.
Table 2 shows the risk factors associated with CRC incidence. The univariate analysis results showed that age (every one-year increment, crude HR = 1.07; 95% CI: 1.050–1.083), diabetes status (crude HR = 1.22; 95% CI: 1.112–1.334), nephropathy status (crude HR = 1.29; 95% CI: 1.033–1.621), statin use (crude HR = 1.12; 95% CI: 1.015–1.245), calcium channel blocker use (crude HR = 1.18; 95% CI: 1.000–1.401), other NSAIDs use (crude HR = 0.71; 95% CI: 0.565–0.894), sulfonylurea use (crude HR = 1.23; 95% CI: 1.121–1.345), and metformin use (crude HR = 1.26; 95% CI: 1.147–1.379) were statistically significantly associated with CRC.
Figure 2 shows the cumulative incidence of female colorectal cancer based on metformin and statin use. For the non-metformin and non-statin use groups, the median follow-up period was 8.76 years. During the observation period, the Nelson–Aalen estimate of the cumulative incidence of female CRC was 0.0004% in one year, 0.0015% in three years, 0.0029% in five years, and 0.0056% in eight years. For both the metformin and statin use groups, the median follow-up period was 8.18 years. The cumulative incidence of female CRC was 0.0006% in one year, 0.0023% in three years, 0.0047% in five years, and 0.0088% in eight years. The cumulative incidence in the metformin and statin use groups was significantly higher than that in the non-metformin and non-statin use groups (log-rank test p < 0.001; Figure 2).
Table 3 shows the combined effects of statins and metformin. When adjusted for multiple comparisons, age (aHR = 1.07; 95% CI: 1.052–1.085, p < 0.001), nephropathy status (aHR = 1.31; 95% CI: 1.040–1.641, p = 0.022), CCB use (aHR = 1.19; 95% CI: 1.007–1.416, p = 0.042), and other NSAIDs use (aHR = 0.73; 95% CI: 0.575–0.915, p = 0.007) were statistically significantly associated with CRC. Additionally, we assessed the combination effect of metformin and statin use on CRC incidence. Compared to the non-metformin use and non-statin use groups, the risk of CRC was positively associated with the metformin and statin use groups (aHR = 1.25; 95% CI: 1.043–1.499, p = 0.016).

4. Discussion

This is a nationwide retrospective cohort study. The accuracy of the NHI database guaranteed appropriate statistical power for detecting the association between two diseases and completeness in identifying incident cases of female CRC. We determined the characteristics of CRC among the Taiwanese female population and the fact that metformin and statin use showed no protective effect on the risk of CRC in women.
In our study, the risk of colon cancer in patients with T2DM remained higher than that in patients without T2DM in the Cox regression analyses. This result is consistent with the results of previous studies, which showed that T2DM was a precipitating factor of CRC. Soltani et al. demonstrated a significant association between suffering from T2DM and having colon adenoma, which is a pre-cancerous condition [28]. A prospective study of two US cohorts showed that T2DM was statistically significantly associated with an increased risk of CRC [29]. A meta-analysis analyzing 151 cohort studies comprising 32 million people showed that T2DM was associated with an increased risk of CRC [8].
In the multivariate analysis, diabetes status showed no statistically significant difference with regard to CRC (Table 3). The incidence of nephropathy, PAD, and eye diseases was higher in the DM group, and these variables were cofactors of DM. Diabetes’ effect on CRC may be diluted in multivariate analysis. Comorbidity, such as nephropathy, was also associated with a higher risk of CRC both in univariate and multivariate analysis (Table 2 and Table 3). Kidney function is damaged by poor long-term glycemic control, and so diabetic nephropathy is a comorbidity of T2DM. It is currently the leading cause of chronic kidney disease and end-stage renal disease [30]. The relationship between PAD and DM neuropathy also implied a complex link between diabetes and CRC, which may be influenced by the severity of diabetes. Thus, we included these chronic comorbidities in this study even if they may cause confounding effects. The result would have been more conservative if we had excluded these variables.
The effect of metformin on CRC is inconsistent between different studies. Knapen et al. revealed that metformin users had a 1.2-fold increased risk of colorectal cancer compared with a non-user [31]. Bodmer et al. showed that metformin was associated with a slightly increased risk of colorectal cancer compared with non-users [32]. Some studies have reported that there is no association between metformin use and CRC [33,34]. However, other studies have shown a protective effect. In a meta-analysis report, which includes eight cohort studies and three case-control studies, metformin was associated with a 25% reduction in CRC incidence among T2DM patients [35]. Another meta-analysis showed that metformin intake was associated with a 25% reduction in colorectal adenoma incidence and a 22% decrease in CRC risk in T2DM metformin users when compared with T2DM non-metformin patients [36].
In our study, metformin showed no protective effect on female CRC. From a pathophysiological perspective, since the primary actions of metformin significantly reduce circulating glucose and plasma insulin, hence improving insulin resistance, it may be beneficial for reducing the risk of diabetes-related cancer incidents [37]. However, elevated glucose levels lead to the proliferation of various solid tumor cell lines and play a role in the development of cancer [8,38]. Some studies have revealed that there is a positive association between serum glucose and the risk of cancer. Stocks et al. examined six European cohorts, analyzing 274,126 men and 275,818 women, which showed that for one mmol/l increase in blood glucose, the relative risk of cancer in men was 1.05 and in women was 1.11 [39]. A study in Austria, which included more than 140,000 adults with an average of 8.4 years’ follow-up, showed that high fasting blood glucose was associated with several cancers, including CRC in women [40]. In addition, a meta-analysis demonstrated that high levels of glucose and hemoglobin A1C (HbA1c) were related to incidences of CRC [41].
NSAIDs use showed a protective effect on CRC in our study due to its anti-inflammatory effect [42]. We found that dyslipidemia was not statistically significantly correlated with female CRC, and the age range was 40 to 64 years. According to previous literature, the risk of CRC is also related to estrogen exposure and the severity of dyslipidemia. The effect of statins on CRC has also shown conflicting results in previous studies. There is an increasing body of data supporting an inverse association between the use of statins and the mortality rate in CRC [43]. However, a meta-analysis revealed five publications on statins and summary RRs which reported that there was no association between statin use and CRC incidence compared to non-use [44]. Yanqiong Liu et al. suggest that statin use is associated with a modestly reduced risk of CRC. However, long-term statin use did not appear to significantly affect the risk of CRC [45]. In our study, statins had a slightly positive association with CRC in the univariate analysis and showed no association with CRC in the multivariate analysis. The potential protective effect of statins may also be weakened by poor T2DM control.
Our study has several limitations. This cohort included only Taiwanese women; some studies showed that there may be ethnic differences in the etiology and biology of CRC between Asians and non-Asians. Therefore, the generalization of our study findings to other ethnicities should be reconfirmed. We did not have biochemical data such as glucose, HbA1c, or lipid profiles to evaluate their potential effects; however, DM severity, dyslipidemia severity, or glycemic control status is related to the risk of CRC and can be further analyzed. Moreover, the duration of insulin use was related to T2DM severity, and it would be worth exploring the effect of insulin on CRC in future studies. It is difficult to evaluate newly detected T2DM and dyslipidemia, as their status could change during the follow-up period. Therefore, we might have underestimated the relative risk of CRC associated with T2DM and dyslipidemia. Some potential confounders such as dietary factors, physical activity, and family history were not measured. As this is an observational study, an experimental design is necessary to confirm the effect of metformin and statin use on the development of CRC.

5. Conclusions

This study found that metformin and statin use revealed no protective effect on female CRC. Poor glycemic control may increase the risk of CRC and may weaken the possible effects of metformin or statin use on CRC, and an additional randomized trial is necessary to investigate the effect of metformin and statin use on CRC prevention. It is important to prevent metabolic syndrome and aggressively control blood glucose and HbA1C to fall within adequate levels in T2DM patients for the first-degree prevention of CRC. Our study supports public health initiatives to combat the increased prevalence of T2DM and CRC among the Taiwanese population.

Author Contributions

Conceptualization, S.-H.H.; methodology, S.-H.H. and D.-K.S.; software, S.-H.H. and Y.-C.W.; validation, S.-H.H., Y.-C.W. and D.-K.S.; formal analysis, S.-H.H., D.-K.S., Y.-C.W., C.-K.L. and M.-C.C.; investigation, S.-H.H. and Y.-C.W.; resources, S.-H.H. and Y.-C.W.; data curation, S.-H.H. and Y.-C.W.; writing—original draft preparation, D.-K.S., Y.-C.W., C.-K.L. and M.-C.C.; writing—review and editing, S.-H.H., D.-K.S. and Y.-C.W.; visualization, S.-H.H., D.-K.S. and Y.-C.W.; supervision, C.-K.L. and M.-C.C.; project administration, C.-K.L. and M.-C.C.; funding acquisition, not applicable. All authors have read and agreed to the published version of the manuscript.

Funding

This research was funded by Fu Jen Catholic University (A0110183).

Institutional Review Board Statement

This study was conducted according to the guidelines of the Declaration of Helsinki and was approved by the Institutional Review Board of the Fu Jen Catholic University Institutional Review Board (IRB no.: C107099).

Informed Consent Statement

Patient consent was waived due to this study using a claims database with encrypted personal identities, and the IRB exempted the study from review.

Data Availability Statement

The datasets generated and/or analyzed in this study are not publicly available in accordance with the policy of the Health and Welfare Data Science Center, Ministry of Health and Welfare, Taiwan, but are available from the corresponding author upon reasonable request.

Acknowledgments

The authors would like to thank You San-Lin for their help with the project.

Conflicts of Interest

The authors declare no conflict of interest.

Appendix A

Table
Table A1. International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) codes used for disease diagnosis in our study.
Table A1. International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) codes used for disease diagnosis in our study.
Disease DiagnosisICD-9-CMDatabase Sources
Diabetes250.1–250.9NHI database
Type 1 diabetes mellitus250.01, 250.03, 250.11, 250.13,
250.21, 250.23, 250.31, 250.33,
250.41, 250.43, 250.51, 250.53,
250.61, 250.63, 250.71, 250.73,
250.81, 250.83, 250.91, 250.93		NHI’s Registry for Catastrophic Illness database
CRC153TCR Annual Report database
Dyslipidemia272.0–272.4NHI database
Obesity278.00NHI database
Hypertension401–405NHI database
COPD490–496NHI database
Asthma493.00–493.99NHI database
Stroke430–438NHI database
Nephropathy580–589NHI database
IHD410–414NHI database
PAD250.7, 785.4, 443.81, 440–448NHI database
Eye Disease250.5, 362.0, 369, 366.41, 365.44NHI database
Notes: Abbreviations: NHI, National Health Insurance; CRC, colorectal cancer; TCR, Taiwan Cancer Registry; COPD, chronic obstructive pulmonary disease; IHD, ischemic heart disease; PAD, peripheral arterial disease.
Table
Table A2. Anatomical Therapeutic Chemical (ATC) Classification codes used for medications in our study.
Table A2. Anatomical Therapeutic Chemical (ATC) Classification codes used for medications in our study.
MedicationATC Code
Diabetes drugsA10
StatinC10AA, C10BA, C10BX
FibratesC10AB
ACEI/ARBC09
CCBC08
AspirinB01AC06′, ′N02BA01
DipyridamoleB01AC07
Clopidogrel/TiclopidineB01AC04/B01AC05
Other NSAIDsM01A
SulfonylureaA10BB
MetforminA10BA02
InsulinA10A
AcarboseA10BF01
TZDA10BG
Notes: Abbreviations: ACEI, angiotensin-converting enzyme inhibitor; ARB, angiotensin receptor blocker; CCB, calcium channel blocker; NSAIDs, non-steroidal anti-inflammatory drugs; TZD, thiazolidinedione.
Table
Table A3. Current procedural terminology codes in Taiwan.
Table A3. Current procedural terminology codes in Taiwan.
ProcedureOrder Code
Potential colon cancer detection examinations
Colonoscopy28017C
Rectoscopy28011C
Sigmoidoscopy28013C
Lower G-I series33010B, 33011B

References

  1. Sung, H.; Ferlay, J.; Siegel, R.L.; Laversanne, M.; Soerjomataram, I.; Jemal, A.; Bray, F. Global cancer statistics 2020: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J. Clin. 2021, 71, 209–249. [Google Scholar] [CrossRef] [PubMed]
  2. Chen, C.J.; You, S.L.; Lin, L.H.; Hsu, W.L.; Yang, Y.W. Cancer epidemiology and control in Taiwan: A brief review. Jpn. J. Clin. Oncol. 2002, 32, S66–S81. [Google Scholar] [CrossRef] [Green Version]
  3. Department of Health. Health Statistics. Vol. II. Vital Statistics. Taipei: Department of Health, Executive Yuan 1998. (in Chinese). Available online: https://dep.mohw.gov.tw/DOS/cp-5225-62361-113.html (accessed on 23 July 2021).
  4. Chao-Hsien, L.E.E.; Cheng, S.C.; Hong-Yi, T.U.N.G.; Chang, S.C.; Ching, C.Y.; Shu-Fen, W.U. The risk factors affecting survival in colorectal cancer in Taiwan. Iran. J. Public Health 2018, 47, 519. [Google Scholar]
  5. Li, S.; Chen, Y.; Xie, L.; Meng, Y.; Zhu, L.; Chu, H.; Gu, D.; Zhang, Z.; Du, M.; Wang, M. Sex hormones and genetic variants in hormone metabolic pathways associated with the risk of colorectal cancer. Environ. Int. 2020, 137, 105543. [Google Scholar] [CrossRef] [PubMed]
  6. Huang, Y.C.; Chen, Y.H. Cancer incidence characteristic evolution based on the National Cancer Registry in Taiwan. J. Oncol. 2020, 2020, 1408793. [Google Scholar] [CrossRef]
  7. Park, H.C.; Shin, A.; Kim, B.W.; Jung, K.W.; Won, Y.J.; Oh, J.H.; Jeong, H.-C.; Yu, C.K.; Lee, B.H. Data on the characteristics and the survival of korean patients with colorectal cancer from the Korea central cancer registry. Ann. Coloproctol. 2013, 29, 144. [Google Scholar] [CrossRef] [Green Version]
  8. Ling, S.; Brown, K.; Miksza, J.K.; Howells, L.; Morrison, A.; Issa, E.; Yates, T.; Khunti, K.; Davies, M.J.; Zaccardi, F. Association of Type 2 Diabetes with Cancer: A Meta-analysis with Bias Analysis for Unmeasured Confounding in 151 Cohorts Comprising 32 Million People. Diabetes Care 2020, 43, 2313–2322. [Google Scholar] [CrossRef]
  9. Bjornsdottir, H.H.; Rawshani, A.; Franzén, S.; Svensson, A.M.; Sattar, N.; Gudbjörnsdottir, S. A national observation study of cancer incidence and mortality risks in type 2 diabetes compared to the background population over time. Sci. Rep. 2020, 10, 1–12. [Google Scholar] [CrossRef]
  10. Lu, Y.; Tao, J. Diabetes Mellitus and Obesity as Risk Factors for Bladder Cancer Prognosis: A Systematic Review and Meta-Analysis. Front. Endocrinol. (Lausanne) 2021, 12, 699732. [Google Scholar] [CrossRef]
  11. Srivastava, S.P.; Goodwin, J.E. Cancer Biology and Prevention in Diabetes. Cells 2020, 9, 1380. [Google Scholar] [CrossRef]
  12. Abudawood, M. Diabetes and cancer: A comprehensive review. J. Res. Med. Sci. 2019, 24, 94. [Google Scholar] [CrossRef] [PubMed]
  13. Harding, J.L.; Andes, L.J.; Gregg, E.W.; Cheng, Y.J.; Weir, H.K.; Bullard, K.M.; Burrows, N.R.; Imperatore, G. Trends in cancer mortality among people with vs. without diabetes in the USA, 1988–2015. Diabetologia 2020, 63, 75–84. [Google Scholar] [CrossRef] [PubMed]
  14. Kobiela, J.; Dobrzycka, M.; Jędrusik, P.; Kobiela, P.; Spychalski, P.; Śledziński, Z.; Zdrojewski, T. Metformin and Colorectal Cancer—A Systematic Review. Exp. Clin. Endocrinol. Diabetes 2019, 127, 445–454. [Google Scholar] [CrossRef] [PubMed]
  15. Demb, J.; Yaseyyedi, A.; Liu, L.; Bustamante, R.; Earles, A.; Ghosh, P.; Gutkind, J.S.; Gawron, A.J.; Kaltenbach, T.R.; Matinez, M.E.; et al. Metformin is associated with reduced odds for colorectal cancer among persons with diabetes. Clin. Transl. Gastroenterol. 2019, 10, 11. [Google Scholar] [CrossRef] [PubMed]
  16. Ng, C.A.W.; Jiang, A.A.; Toh, E.M.S.; Ng, C.H.; Ong, Z.H.; Peng, S.; Tham, H.Y.; Sundar, R.; Chong, C.S.; Khoo, C.M. Metformin and colorectal cancer: A systematic review, meta-analysis and meta-regression. Int. J. Colorectal Dis. 2020, 35, 1501–1512. [Google Scholar] [CrossRef]
  17. Radisauskas, R.; Kuzmickiene, I.; Milinavičienė, E.; Everatt, R. Hypertension, serum lipids and cancer risk: A review of epidemiological evidence. Medicina (Kaunas) 2016, 52, 89–98. [Google Scholar] [CrossRef]
  18. Murai, T. Cholesterol lowering: Role in cancer prevention and treatment. Biol. Chem. 2015, 396, 1–11. [Google Scholar] [CrossRef] [Green Version]
  19. Törnberg, S.A.; Holm, L.E.; Carstensen, J.M.; Eklund, G.A. Risks of cancer of the colon and rectum in relation to serum cholesterol and beta-lipoprotein. N. Engl. J. Med. 1986, 315, 1629–1633. [Google Scholar] [CrossRef]
  20. Mamtani, R.; Lewis, J.D.; Scott, F.; Ahmad, T.; Goldberg, D.S.; Datta, J.; Yang, Y.-X.; Boursi, B. Disentangling the association between statins, cholesterol, and colorectal cancer: A nested case–control study. PLoS Med. 2016, 13, e1002007. [Google Scholar] [CrossRef]
  21. Jacobs, E.J.; Rodriguez, C.; Brady, K.A.; Connell, C.J.; Thun, M.J.; Calle, E.E. Cholesterol-lowering drugs and colorectal cancer incidence in a large United States cohort. J. Nat. Cancer Inst. 2006, 98, 69–72. [Google Scholar] [CrossRef] [Green Version]
  22. Lee, J.W.; You, N.Y.; Kim, Y.; Kim, Y.; Kim, J.; Kang, H.T. Statin use and site-specific risk of colorectal cancer in individuals with hypercholesterolemia from the National Health Insurance Service-National Health Screening Cohort (NHIS-HEALS) Nutr. Metab. Cardiovasc. Dis. 2019, 29, 701–709. [Google Scholar] [CrossRef]
  23. Hsieh, C.Y.; Su, C.C.; Shao, S.C.; Sung, S.F.; Lin, S.J.; Yang, Y.H.K.; Lai, E.C.C. Taiwan’s National Health Insurance Research Database: Past and future. Clin. Epidemiol. 2019, 11, 349–358. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  24. Chiang, C.J.; You, S.L.; Chen, C.J.; Yang, Y.W.; Lo, W.C.; Lai, M.S. Quality assessment and improvement of nationwide cancer registration system in Taiwan: A review. Jpn. J. Clin. Oncol. 2015, 45, 291–296. [Google Scholar] [CrossRef]
  25. Yen, F.S.; Lo, Y.R.; Hwu, C.M.; Hsu, C.C. Early-onset type 2 diabetes < 60 years and risk of vascular complications. Diabetes Res. Clin. Pract. 2021, 182, 109129. [Google Scholar] [CrossRef] [PubMed]
  26. Tseng, C.H. Diabetes, metformin use, and colon cancer: A population-based cohort study in Taiwan. Eur. J. Endocrinol. 2012, 167, 409–416. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  27. Nguyen, T.L.; Collins, G.S.; Spence, J.; Daurès, J.P.; Devereaux, P.J.; Landais, P.; Le Manach, Y. Double-adjustment in propensity score matching analysis: Choosing a threshold for considering residual imbalance. BMC Med. Res. Methodol. 2017, 17, 78. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  28. Soltani, G.; Poursheikhani, A.; Yassi, M.; Hayatbakhsh, A.; Kerachian, M.; Kerachian, M.A. Obesity, diabetes and the risk of colorectal adenoma and cancer. BMC Endocr. Disord. 2019, 19, 113. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  29. Hu, Y.; Zhang, X.; Ma, Y.; Yuan, C.; Wang, M.; Wu, K.; Tabung, F.K.; Tobias, D.; Hu, F.B.; Giovannucci, E.; et al. Incident Type 2 Diabetes Duration and Cancer Risk: A Prospective Study in Two US Cohorts. J. Natl. Cancer Inst. 2021, 113, 381–389. [Google Scholar] [CrossRef]
  30. Samsu, N. Diabetic Nephropathy: Challenges in Pathogenesis, Diagnosis, and Treatment. Biomed Res. Int. 2021, 2021, 1497449. [Google Scholar] [CrossRef]
  31. M Knapen, L.; TAM Dittrich, S.; de Vries, F.; Starup-Linde, J.; Vestergaard, P.; MA Henry, R.; Leo, M.L.S.; Cees, N.; Marloes, T.B. Use of biguanides and the risk of colorectal cancer: A register-based cohort study. Curr. Drug Saf. 2013, 8, 349–356. [Google Scholar] [CrossRef]
  32. Bodmer, M.; Becker, C.; Meier, C.; Jick, S.S.; Meier, C.R. Use of metformin is not associated with a decreased risk of colorectal cancer: A case–control analysis. Cancer Epidemiol Biomarkers Prev. 2012, 21, 280–286. [Google Scholar] [CrossRef] [Green Version]
  33. Lin, C.M.; Huang, H.L.; Chu, F.Y.; Fan, H.C.; Chen, H.A.; Chu, D.M.; Wu, L.W.; Wang, C.C.; Chen, W.L.; Lin, S.L.; et al. Association between gastroenterological malignancy and diabetes mellitus and anti-diabetic therapy: A nationwide, population—Based cohort study. PLoS ONE 2015, 10, e0125421. [Google Scholar] [CrossRef]
  34. Kowall, B.; Stang, A.; Rathmann, W.; Kostev, K. No reduced risk of overall, colorectal, lung, breast, and prostate cancer with metformin therapy in diabetic patients: Database analyses from Germany and the UK. Pharmacoepidemiol. Drug Saf. 2015, 24, 865–874. [Google Scholar] [CrossRef]
  35. Nie, Z.; Zhu, H.; Gu, M. Reduced colorectal cancer incidence in type 2 diabetic patients treated with metformin: A meta-analysis. Pharm Biol. 2016, 54, 2636–2642. [Google Scholar] [CrossRef] [Green Version]
  36. Liu, F.; Yan, L.; Wang, Z.; Lu, Y.; Chu, Y.; Li, X.; Liu, Y.; Rui, D.; Nie, S.; Xiang, H. Metformin therapy and risk of colorectal adenomas and colorectal cancer in type 2 diabetes mellitus patients: A systematic review and meta-analysis. Oncotarget 2017, 8, 16017. [Google Scholar] [CrossRef] [Green Version]
  37. Salani, B.; Del Rio, A.; Marini, C.; Sambuceti, G.; Cordera, R.; Maggi, D. Metformin, cancer and glucose metabolism. Endocr. Relat. Cancer 2014, 21, R461–R471. [Google Scholar] [CrossRef] [Green Version]
  38. Wang, M.; Yang, Y.; Liao, Z. Diabetes and cancer: Epidemiological and biological links. World J. Diabetes 2020, 11, 227. [Google Scholar] [CrossRef]
  39. Stocks, T.; Rapp, K.; Bjørge, T.; Manjer, J.; Ulmer, H.; Selmer, R.; Lukanova, A.; Johansen, D.; Concin, H.; Tretli, S.; et al. Blood glucose and risk of incident and fatal cancer in the metabolic syndrome and cancer project (Me-Can): Analysis of six prospective cohorts. PLoS Med. 2009, 6, e1000201. [Google Scholar] [CrossRef]
  40. Rapp, K.; Schroeder, J.; Klenk, J.; Ulmer, H.; Concin, H.; Diem, G.; Oberaigner, W.; Weiland, S.K. Fasting blood glucose and cancer risk in a cohort of more than 140,000 adults in Austria. Diabetologia 2006, 49, 945–952. [Google Scholar] [CrossRef]
  41. Pisani, P. Hyper-insulinaemia and cancer, meta-analyses of epidemiological studies. Arch. Physiol. Biochem. 2008, 114, 63–70. [Google Scholar] [CrossRef]
  42. Harewood, R.; Disney, R.; Kinross, J.; von Wagner, C.; Cross, A.J. Medication use and risk of proximal colon cancer: A systematic review of prospective studies with narrative synthesis and meta-analysis. Cancer Causes Control 2021, 32, 1047–1061. [Google Scholar] [CrossRef] [PubMed]
  43. Guerra, B.; Recio, C.; Aranda-Tavío, H.; Guerra-Rodríguez, M.; García-Castellano, J.M.; Fernández-Pérez, L. The Mevalonate Pathway, a Metabolic Target in Cancer Therapy. Front. Oncol. 2021, 11, 626971. [Google Scholar] [CrossRef] [PubMed]
  44. Jung, Y.S.; Park, C.H.; Eun, C.S.; Park, D.I.; Han, D.S. Statin use and the risk of colorectal adenoma: A meta-analysis. J. Gastroenterol. Hepatol. 2016, 31, 1823–1830. [Google Scholar] [CrossRef]
  45. Liu, Y.; Tang, W.; Wang, J.; Xie, L.; Li, T.; He, Y.; Deng, Y.; Peng, Q.; Li, S.; Qin, X. Association between statin use and colorectal cancer risk: A meta-analysis of 42 studies. Cancer Causes Control 2014, 25, 237–249. [Google Scholar] [CrossRef] [PubMed]
Jcm 11 04469 g001 550
Figure 1. Flow chart for the selection of study subjects.
Figure 1. Flow chart for the selection of study subjects.
Jcm 11 04469 g001
Jcm 11 04469 g002 550
Figure 2. Cumulative incidence of female colorectal cancer based on different scenarios of metformin and statin use.
Figure 2. Cumulative incidence of female colorectal cancer based on different scenarios of metformin and statin use.
Jcm 11 04469 g002
Table
Table 1. Baseline characteristics of female patients aged 40–64 years with and without diabetes mellitus.
Table 1. Baseline characteristics of female patients aged 40–64 years with and without diabetes mellitus.
VariablesUnmatchedMatched to Propensity Scores
non-DM
(n = 3,925,317)		DM
(n = 87,229)		Standardized Differencenon-DM
(n = 348,916)		DM
(n = 87,229)		Standardized Difference
Age48.96 ± 7.0154.36 ± 6.20.81654.47 ± 6.3654.35 ± 6.20−0.018
Hypertension581,079 (14.8%)51,811 (59.4%)−1.041185,152 (59.8%)51,585 (59%)0.012
COPD215,797 (5.5%)10,439 (12%)−0.23138,735 (12.5%)10,411 (12%)0.019
Asthma113,219 (2.9%)5041 (5.8%)−0.14319,331 (6.2%)5021 (5.8%)0.023
Stroke73,235 (1.9%)8647 (9.9%)−0.34728,734 (9.3%)8617 (9.9%)−0.027
Nephropathy46,121 (1.2%)6302 (7.2%)−0.30516,702 (5.4%)6268 (7.2%)−0.091
IHD163,156 (4.2%)15,857 (18.2%)−0.45756,336 (18.2%)15,803 (18%)0.001
PAD46,858 (1.2%)10,344 (11.9%)−0.44226,764 (8.6%)10,281 (12%)−0.132
Eye Disease13,145 (0.3%)13,940 (16%)−0.59612,961 (4.2%)13,707 (16%)−0.441
Dyslipidemia305,862 (7.8%)44,911 (51.5%)−1.090157,819 (51%)44,694 (51%)−0.010
Obesity16,269 (0.4%)1829 (2.1%)−0.1526553 (2.1%)1824 (2.1%)0.002
Fibrates76,635 (2%)17,323 (19.9%)−0.60052,656 (17%)17,271 (20%)−0.095
ACEI/ARB310,000 (7.9%)43,437 (49.8%)−1.043143,766 (46.4%)43,212 (50%)−0.080
CCB403,407 (10.3%)37,991 (43.6%)−0.809132,139 (42.7%)37,839 (44%)−0.020
Aspirin299,482 (7.6%)31,040 (35.6%)−0.722104,535 (33.8%)30,960 (36%)−0.047
Dipyridamole95,552 (2.4%)12,136 (13.9%)−0.42837,880 (12.2%)12,104 (14%)−0.063
Clopidogrel/Ticlopidine15,380 (0.4%)3274 (3.8%)−0.2388422 (2.7%)3251 (3.7%)−0.072
Other NSAIDs2,797,482 (71.3%)73,193 (83.9%)−0.307260,994 (84.3%)72,976 (84%)0.010
Potential colon cancer detection examinations2179 (0.1%)88 (0.1%)−0.016306 (0.1%)88 (0.1%)−0.001
Abbreviations: COPD, chronic obstructive pulmonary disease; IHD, ischemic heart disease; PAD, peripheral arterial disease; ACEI, angiotensin-converting enzyme inhibitor; ARB, angiotensin receptor blocker; CCB, calcium channel blocker; NSAIDs, non-steroidal anti-inflammatory drugs.
Table
Table 2. Risk factors associated with colon cancer incidence for women aged 40 to 64 years.
Table 2. Risk factors associated with colon cancer incidence for women aged 40 to 64 years.
VariablesInterpretationColon Cancer Cases (n)Univariate
HR a95% CIp Value
AgeEvery 1-year increment28571.07(1.050, 1.083)<0.001
DiabetesYes vs. no 665/21921.22(1.112, 1.334)<0.001
HypertensionYes vs. no 1930/9271.18(0.951, 1.474)0.131
COPDYes vs. no 374/24830.93(0.774, 1.122)0.456
AsthmaYes vs. no 206/26511.02(0.809, 1.295)0.845
StrokeYes vs. no 290/25671.04(0.847, 1.277)0.708
NephropathyYes vs. no 227/26301.29(1.033, 1.621)0.025
IHDYes vs. no 607/22501.11(0.942, 1.305)0.213
PADYes vs. no 265/25920.82(0.667, 1.014)0.068
Eye DiseaseYes vs. no 237/26201.13(0.852, 1.498)0.396
DyslipidemiaYes vs. no 1608/12490.96(0.780, 1.184)0.709
ObesityYes vs. no 58/27991.05(0.722, 1.528)0.798
StatinYes vs. no 1127/17301.12(1.015, 1.245)0.024
FibratesYes vs. no 554/23031.13(0.953, 1.327)0.165
ACEI/ARBYes vs. no 1544/13131.15(0.936, 1.403)0.186
CCBYes vs. no 1470/13871.18(1.000, 1.401)0.0498
AspirinYes vs. no 1068/17890.91(0.775, 1.070)0.254
DipyridamoleYes vs. no 434/24231.02(0.849, 1.228)0.824
Clopidogrel/TiclopidineYes vs. no 115/27421.32(0.995, 1.746)0.054
Other NSAIDsYes vs. no 2363/4940.71(0.565, 0.894)0.004
SulfonylureaYes vs. no 1054/18031.23(1.121, 1.345)<0.001
MetforminYes vs. no 1027/18301.26(1.147, 1.379)<0.001
InsulinYes vs. no 192/26651.14(0.941, 1.373)0.183
AcarboseYes vs. no 282/25751.12(0.966, 1.309)0.130
TZDYes vs. no 333/25241.02(0.882, 1.175)0.809
Potential colon cancer detection examinationsYes vs. no2/28570.68(0.124, 3.734)0.657
Abbreviations: COPD, chronic obstructive pulmonary disease; IHD, ischemic heart disease; PAD, peripheral arterial disease; ACEI, angiotensin-converting enzyme inhibitor; ARB, angiotensin receptor blocker; CCB, calcium channel blocker; NSAIDs, non-steroidal anti-inflammatory drugs; TZD, thiazolidinedione; HR, hazard ratio; CI, confidence interval. a: Adjusted for age via a Cox proportional hazards regression.
Table
Table 3. Multivariate analysis of the incidence of colon cancer in women aged 40–64 years with the combination of statins and metformin.
Table 3. Multivariate analysis of the incidence of colon cancer in women aged 40–64 years with the combination of statins and metformin.
VariablesItemsHR a (95% CI)95% CI of HRp Value
Age 1.07(1.052, 1.085) Jcm 11 04469 i001<0.001
DiabetesNo1.00Referent
Yes1.09(0.977, 1.223)0.119
NephropathyNo1.00Referent
Yes1.31(1.040, 1.641)0.022
CCBNo1.00Referent
Yes1.19(1.007, 1.416)0.042
Other NSAIDsNo1.00Referent
Yes0.73(0.575, 0.915)0.007
SulfonylureaNo1.00Referent
Yes1.04(0.892, 1.214)0.614
Metformin/ StatinNo/No1.00Referent
No/Yes1.03(0.906, 1.172)0.645
Yes/No1.12(0.934, 1.335)0.227
Yes/Yes1.25(1.043, 1.499)0.016
a Multiple Cox regression. Covariates: age, diabetes, nephropathy, CCB, other NSAIDs, sulfonylurea, metformin, and statin.
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Share and Cite

MDPI and ACS Style

Hsu, S.-H.; Syu, D.-K.; Wang, Y.-C.; Liu, C.-K.; Chen, M.-C. Metformin, Statin Use, and Female Colorectal Cancer: A Population-Based Cohort Study in Taiwan. J. Clin. Med. 2022, 11, 4469. https://doi.org/10.3390/jcm11154469

AMA Style

Hsu S-H, Syu D-K, Wang Y-C, Liu C-K, Chen M-C. Metformin, Statin Use, and Female Colorectal Cancer: A Population-Based Cohort Study in Taiwan. Journal of Clinical Medicine. 2022; 11(15):4469. https://doi.org/10.3390/jcm11154469

Chicago/Turabian Style

Hsu, Shu-Hua, De-Kai Syu, Yu-Chiao Wang, Chih-Kuang Liu, and Ming-Chih Chen. 2022. "Metformin, Statin Use, and Female Colorectal Cancer: A Population-Based Cohort Study in Taiwan" Journal of Clinical Medicine 11, no. 15: 4469. https://doi.org/10.3390/jcm11154469

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop