University of Southern California Dissertations and Theses

The association of prediagnostic metformin use with prostate cancer in the multiethnic cohort study

(USC Thesis Other)

The association of prediagnostic metformin use with prostate cancer in the multiethnic cohort study

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content THE ASSOCIATION OF PREDIAGNOSTIC METFORMIN USE WITH PROSTATE
CANCER IN THE MULTIETHNIC COHORT STUDY
by
Kaixin Chu
A Thesis Presented to the
FACULTY OF THE USC KECK SCHOOL OF MEDICINE
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment ofthe
Requirements for the Degree
MASTER OF SCIENCE
(BIOSTATISTICS)
May 2024
Copyright 2024 Kaixin Chu

ii
TABLE OF CONTENTS
List of Tables............................................................................................................................iii
Abstract .................................................................................................................................... iv
Introduction ............................................................................................................................... 1
Chapter 1: Materials and Methods.............................................................................................. 3
Study Population ................................................................................................ 3
Statistical Analysis ............................................................................................. 4
Chapter 2: Results...................................................................................................................... 6
Study Population Characteristics by Metformin Use Status................................. 6
Association ofPre-diagnostic Metformin Use with Prostate Cancer Incidence .... 6
Effect Modification by PSA, BMI, and Smoking................................................. 7
Chapter 3: Discussion……………………………………………………………………………..8
Chapter 4: Conclusion ...............................................................................................................11
References ............................................................................................................................... 12

iii
List of Tables
Table 1: Demographic and Lifestyle Factors by Pre-diagnostic Metformin Use………………...vi
Table 2: Association of pre-diagnostic metformin use with prostate cancer incidence
by race/ethnicity…………………...……………………………………………………………..vii
Table 3: Association of pre-diagnostic metformin use with prostate cancer incidence
stratified by PSA screening status prior to cancer diagnosis in the overall study………………viii
Table 4: Association of pre-diagnostic metformin use with prostate cancer incidence
stratified by BMI at baseline in the overall study ………………………………………………..ix
Table 5: Association of pre-diagnostic metformin use with prostate cancer incidence
stratified by smoking status at baseline in the overall study……………………………………....x

iv
ABSTRACT
Background: Past research has generated mixed findings regarding the association between prediagnostic metformin use and risk of prostate cancer. These studies were conducted
predominantly in White populations and had limited focus on aggressive forms of prostate
cancer.
Methods: In the Multiethnic Cohort (MEC), we conducted a cohort analysis to investigate the
association between pre-diagnostic metformin use and incidence of prostate cancer (total,
localized, advanced, low-grade, high-grade, aggressive, and non-aggressive) in 31,062 male
participants from five major racial/ethnic groups (White, African American, Japanese American,
Latino, and Native Hawaiian). We also evaluated potential effect modification by history of
prostate-specific antigen (PSA) screening, body mass index (BMI), and smoking status at
baseline.
Results: Among 31,062 men included in this analysis, 2,216 (7. 1%) ofthem reported metformin
use, and 2,748 (8.8%) developed prostate cancer. After adjusting for potential confounders, a
non-significant negative association was observed between pre-diagnostic metformin use and all
prostate cancer phenotypes with the hazard ratio (HR) ranging from 0.77 to 0.92. In Japanese
Americans (36% of overall study), pre-diagnostic metformin use was significantly associated
with a substantially lower risk of total (HR = 0.56 [0.39-0.82], p = 0.003), low-grade (HR = 0.49
[0.31-0.79], p = 0.003), localized (HR = 0.51 [0.33-0.78], p = 0.002), and non-aggressive
prostate cancer (HR = 0.48 [0.29-0.81], p = 0.006). A significantly lower risk of aggressive
prostate cancer was also observed among men with no history of PSA screening (HR = 0.53
[0.27, 1.01], p = 0.054) or men who were obese at baseline (HR = 0.47 [0.24, 0.95], p = 0.04).

v
Conclusion: These results provide further support that the association of metformin use with risk
of prostate cancer may be heterogeneous across racial/ethnic populations. Adding to the existing
evidence, pre-diagnostic metformin was also found to be associated with lower risks of
aggressive forms of prostate cancer in populations with shared specific characteristics.

1
INTRODUCTION
Prostate cancer (PCa) is the most commonly diagnosed non-skin cancer and the second
leading cause of cancer death among males in the United States1,2
. African Americans have the
highest rates of prostate cancer of any racial or ethnic group and are more likely to be diagnosed
at an advanced stage or with a more aggressive tumor. Aging, being African American, and
positive family history are known to be associated with an increased risk of prostate cancer.
There is increasing evidence supporting excess body weight and abdominal adiposity as potential
risk factors for high-risk prostate cancer3
. Despite extensive epidemiological research, few
lifestyle-modifiable risk factors have been identified for prostate cancer.
Metformin is the first-line treatment for type 2 diabetes mellitus (T2DM), a condition
characterized by high blood sugar levels due to insulin resistance or insufficient insulin
production by the body. Metformin works by helping restore the body’s response to insulin. It
reduces the amount of blood sugar produced by the liver and absorbed by the intestines or
stomach4
. Metformin has been in clinical use for decades because of its good safety record and
limited toxicity and the benefits of reducing the risk of myocardial infarction and deaths due to
cardiovascular disease5,6
. The continued growth ofthe diabetic population and its use not only to
treat diabetes but also conditions such as polycystic ovary syndrome keep its use at a high level.
In the U.S, the number of metformin prescriptions rose from 40.89 million in 2004 to 91.15
million in 20217
.
In addition to lower blood sugar, over the past decade, metformin has since been shown
to have preventive effects on multiple diseases, including obesity, cardiovascular disease, agingrelated diseases, and several cancers8,9
. A comprehensive review of multiple observational
studies concluded that metformin use was significantly associated with a 27% reduction risk of

2
all malignancies, a 66% reduction risk of liver cancer, a 17% reduction risk of colorectal cancer,
a 44% reduction risk of pancreatic cancer, a 17% reduction risk of gastric cancer, and a 10%
reduction risk of esophageal cancer10
. However, findings have been mixed regarding the effect of
metformin on the development of prostate cancer. A recent systematic review of 47 studies (44
were in men of European descent), including two randomized control trials (RCTs), 18 cohort
studies, and three case-control studies, found no significant association between metformin and
risk of prostate cancer in the overall meta-analysis or by study design11
. Metformin use was,
however, associated with a 27% reduced risk of prostate cancer in the meta-analysis ofthree East
Asian studies11
. In the few studies investigating metformin use and prostate cancer risk in
racially/ethnically diverse populations, metformin use was found to be associated with a 38%
reduction in risk of prostate cancer among Hispanic men, but no association was detected among
non-Hispanic Whites or African Americans12
.
Previous research on metformin use and risk of prostate cancer have been conducted
predominantly in White populations and had limited focus on aggressive forms of disease. To
provide further insight, we conducted this study to investigate the association between prediagnosis metformin use and incidence of total prostate cancer and its subtypes in a large diverse
prospective cohort.

3
MATERIALS AND METHODS
Study Population
We conducted a cohort analysis in the Multiethnic Cohort Study (MEC) to investigate the
association between pre-diagnostic metformin use and prostate cancer incidence. The MEC is an
ongoing prospective cohort study established to study risk factors for cancer. Over 215,000
residents of the State of Hawaii and Los Angeles (104,000 in HI and 112,000 in CA) aged 45 to
75 years were recruited into the MEC in 1993-1996, primarily from five racial/ethnic groups:
White, African American, Japanese American, Latino, and Native Hawaiian. At baseline, all
participants completed a 26-page questionnaire including basic demographic information, family
history of cancer, medication history, lifestyle factors, and dietary intake. Follow-up
questionnaires were sent every five years to update this information and to collect additional
information such as the history of prostate specific antigen (PSA) testing from male participants.
From 2001 to 2006, approximately 70,000 men and women in the MEC provided blood and
urine samples to the MEC biorepository. The biorepository participants also completed a
medication questionnaire providing detailed information on all the medications they had taken in
the two weeks prior to the questionnaire completion date. Of the 36,860 men who completed the
medication questionnaire, a total of 31,062 were included in the study. Participants were
excluded from the study if they (1) had a prostate cancer diagnosis prior to the medication
questionnaire, (2) had missing information on metformin use, date of cancer diagnosis, or date of
medication questionnaire, or (3) self-reported as “Other” ethnicity. This study was approved by
the Institutional Review Boards overseeing research on human subjects at the University of
Hawaii (Honolulu, HI) and the University of Southern California (Los Angeles, CA). All
individuals included in this analysis provided informed consent.

4
Prostate cancer cases in the MEC were identified through linkage of the cohort to the
Surveillance Epidemiology and End Results (SEER) cancer registries in the state of Hawaii and
California. Clinical characteristics of the tumors such as stage and grade were also obtained to
determine the disease aggressiveness in this study. The cohort was also linked to death certificate
files in Hawaii and California and the National Death Index. Through December 31, 2017, a total
of 2,748 cases of prostate cancer were identified in the MEC, including 1,945 low-grade
(Gleason score ≤ 7), 639 high-grade (Gleason score ≥8), 2,035 localized, 424 advanced (regional
or distant), 1,599 non-aggressive (low-grade and localized), and 880 aggressive (advanced or
high-grade) prostate cancer.
Statistical Analysis
The cohort analysis was conducted using Cox proportional hazard models with age as the
time metric. All eligible participants contributed person-time at risk from the date of completion
of the medication questionnaire to date of prostate cancer diagnosis, death, or censored through
December 31, 2017. The pre-diagnostic metformin use (yes or no) was determined based on
information from the medication questionnaire. Proportional hazards assumptions were evaluated
by correlating Schoenfeld residuals over analysis time and found to be met. In this analysis,
multivariable Cox regression models were used to evaluate the associations between prediagnostic metformin use and incidence of total prostate cancer and the subtypes (localized,
advanced, low-grade, high-grade, aggressive, and non-aggressive), adjusting for age at MEC
cohort entry, history of diabetes prior to cancer diagnosis (yes, no), first-degree family history of
prostate cancer (positive, negative), body mass index (BMI; <25 kg/m2
, 25-30 kg/m2
, ≥30 kg/m2
),
smoking status (never, former, current), education ( ≤ 8th grade, 9th-12th grade, vocational
school/some college, graduated college or higher), daily moderate or vigorous physical activity

5
(MET-hour/day in quartiles), and PSA screening status prior to cancer diagnosis (yes, no). These
variables were adjusted in the model because they are established risk factors or have been
reported to be associated with prostate cancer in the MEC13,14
. All the analyses were performed
in the overall study and in each racial/ethnic group separately. History of diabetes, family
history of prostate cancer, and PSA screening status were derived from all questionnaires prior to
the event of interest while the remaining variables were from the baseline cohort questionnaire.
Missing values of all variables were coded as an “unknown” category in the multivariate
analyses.
We further evaluated the potential effect modification by race/ethnicity, history of PSA
testing (yes, no), BMI (<25 kg/m2
, 25-30 kg/m2
, ≥ 30 kg/m2
), and smoking status (never, current,
former) on the association between pre-diagnostic metformin use and prostate cancer incidence.
The likelihood ratio test was used to evaluate the interaction between metformin use and each
covariate by comparing models with and without the interaction terms. All statistical tests were
two-sided, and p-values <0.05 were considered statistically significant. Data analysis was
performed using Stata 17.

6
RESULTS
Study Population Characteristics by Metformin Use Status
Among 31,062 men included in this analysis, 2,216 (7. 13%) of them reported metformin
use in the medication questionnaire, with the prevalence being highest among Native Americans
(10.4%), followed by Latinos (10.0%), Japanese Americans (7.2%), African Americans (6.1%),
and Whites (3.4%; Table 1). Compared with non-users, metformin users were more likely to
have a history of diabetes, being a former or current smoker, and being obese at baseline.
Metformin users were less likely to have a higher educational level or engage in moderate or
vigorous physical activities compared to their non-user counterparts.
Association of Pre-diagnostic Metformin Use with Prostate Cancer Incidence
In our study, a non-significant negative association was observed between pre-diagnostic
metformin use and all prostate cancer phenotypes, with the hazard ratio (HR) ranging from 0.77
to 0.92 (Table 2). More specifically, pre-diagnostic metformin users had an approximately 13%
lower risk oftotal prostate cancer compared to non-users (HR = 0.87 [0.72-1.05], p = 0. 16), and
the non-significant inverse association was similar for low-grade (HR = 0.85 [0.68-1.07], p =
0.17), high-grade (HR = 0.83 [0.56-1.22], p = 0.35), and localized (HR = 0.86 [0.69-1.07], p =
0. 18) prostate cancer but appeared to be more evident for aggressive prostate cancer (HR = 0.77
[0.55-1.07], p = 0.12).
In Japanese Americans, which accounts for 36% of the study population, pre-diagnostic
metformin use was significantly associated with a substantially lower risk of total (HR = 0.56
[0.39-0.82], p = 0.003), low-grade (HR = 0.49 [0.31-0.79], p = 0.003), localized (HR = 0.51
[0.33-0.78], p = 0.002), and non-aggressive prostate cancer (HR = 0.48 [0.29-0.81], p = 0.006;
Table 2). This inverse association was weaker and not statistically significant for high-grade

7
(HR = 0.76 [0.39-1.48], p = 0.42) and aggressive prostate cancer (HR = 0.75 [0.42-1.34], p =
0.34), and was null for advanced prostate cancer (HR = 1.05 [0.46-2.41], p = 0.91; Table 2).
It was suggested that the associations of pre-diagnostic metformin use with total (PLRT =
0.053), localized (PLRT = 0.051) and non-aggressive prostate cancer (PLRT = 0.054) were
heterogeneous across racial/ethnic groups (Table 2), with the inverse associations being
strongest among Japanese Americans (HR ranging from 0.48 to 0.56) but appeared to be in the
opposite direction among Latino men (HR ranging from 1.13 to 1.36).
Effect Modification by PSA, BMI, and Smoking
We found no evidence that the association between pre-diagnostic metformin use and
prostate cancer incidence was modified by PSA, BMI, or smoking status (all PLRT > 0.15; Table
3, Table 4, and Table 5). Although not significant in the overall population, metformin use was
suggestively associated with a 47% lower risk of aggressive prostate cancer (95% CI = 0.27-1.01,
P = 0.054) among men reported no history of PSA screening and this association was mostly null
among men with a positive history of PSA screening (OR=0.90, 95% CI = 0.60-1.35, p = 0.60;
Table 3). Similarly, among men who were obese at baseline, metformin users had a 53% lower
risk of aggressive prostate cancer than non-users (95% CI = 0.24-0.95, p = 0.04), and this
association was weaker among overweight men (OR = 0.81, 95% CI = 0.51-1.29, p = 0.38) and
not observed among men with normal weight at baseline (OR = 1.25, 95% CI = 0.61-2.57, p =
0.54; Table 4).

8
DISCUSSION
In this large multiethnic cohort analysis, we did not find statistically significant
associations between metformin use and risk of prostate cancer in the overall study. When
stratified by history of PSA screening and BMI, metformin users had a 47% and 53% lower risk
of aggressive prostate cancer than non-uses among men with no past PSA screening and those
that were obese at baseline, respectively. Within each racial/ethnic population, metformin use
was significantly associated with a 44% to 52% reduction in risk of total and non-aggressive
forms of prostate cancer (localized and low-grade) in Japanese American men.
In our study, metformin use was not associated with risk of prostate cancer in Whites,
which is aligned with the most recent meta-analysis including 18 observational studies in men of
European descent15
. The strong protective effects of metformin on risk of prostate cancer
observed in Japanese Americans are consistent with previous reports in other East Asian
populations. In a large retrospective cohort analysis of 395,481 men from the Taiwan National
Health Insurance reimbursement database, metformin use was associated with a 53% reduced
risk of prostate cancer and the reduction in prostate cancer risk was even greater (77%) among
users with longer duration and/or higher cumulative dose16
. When compared with users of
sulfonylurea, another anti-diabetic medication, metformin users have a 20% lower risk of
prostate cancer in 51,390 male patients with T2DM from public hospitals/clinics in Hong Kong17
.
Given the limited research in non-White populations, further investigations are warranted to
better understand whether the heterogeneous effect of metformin on risk of prostate cancer
across populations contributes to racial/ethnic disparities in prostate cancer incidence.
There is increasing evidence supporting a protective effect of metformin on risk of
aggressive forms of prostate cancer (advanced, high-grade). In our study, metformin use was

9
suggestively associated with a 23% reduction in risk of aggressive prostate cancer and the
reduction was greater and significant (> 45%) among men with no history of PSA screening or
among obese men. Although not significant, metformin use was consistently associated with a
25% to 44% lower risk of aggressive prostate cancer across racial/ethnic groups except in
African Americans where the association was null. In alignment with our findings, in a crosssectional study of aggressive versus low/intermediate risk prostate cancer, metformin use was
associated with a 20% lower risk of aggressive prostate cancer in Whites but a two-fold risk of
aggressive prostate cancer in African American men18
. The protective effect of metformin on
aggressive forms of prostate cancer in White men was further supported by a SEER-Medicare
cohort study with a predominantly white population (72.2%) where metformin use was
associated with a 31% reduction in the incidence of advanced prostate cancer in diabetic men19
.
A more recent study using SEER-Medicare data (77.2% were Whites) also reported a 39%
reduction in risk of high-grade prostate cancer and a 27% reduction in risk of advanced prostate
cancer among metformin users20
.
Several limitations of this study need to be noted. Our study lacks sufficient statistical
power to detect the association between metformin use and risk of prostate cancer, especially for
the subtypes. However, the direction of our findings is consistent with the previous studies
suggesting an inverse association between pre-diagnostic metformin use and total or aggressive
(advanced, high-grade) prostate cancer. Furthermore, detailed information on duration and
dosage of metformin use was not available in our study, which prevented us from assessing the
dose-response relationship between metformin and prostate cancer. Despite these weaknesses,
our study represents an important effort to better understand the effect of metformin on the
development of prostate cancer, especially the more aggressive forms of disease, in a

10
racial/ethnically diverse population including minority groups that were often under-represented
in past studies.

11
CONCLUSION
In conclusion, findings from this study provide further support to the racial/ethnic
differences in the effect of pre-diagnostic metformin use on risk of prostate cancer. Our results
also add to the increasing body of evidence suggesting a protective effect of metformin on
aggressive forms of prostate cancer. However, due to the limited power in population-specific
and subtype-specific analyses, these results should be interpreted with caution and further
investigations are warranted to validate our findings in larger studies.

12
REFERENCES
1. Zaidi, S., Gandhi, J., Joshi, G., Smith, N. L., & Khan, S. A. (2019). The anticancer potential of
metformin on prostate cancer. Prostate Cancer and Prostatic Diseases, 22(3), 351–361.
https://doi.org/10.1038/s41391-018-0085-2
2. Libby, G., Donnelly, L. A., Donnan, P. T., Alessi, D. R., Morris, A. D., & Evans, J. M. M.
(2009). New Users of Metformin Are at Low Risk ofIncident Cancer: A cohort study among
people with type 2 diabetes. Diabetes Care, 32(9), 1620–1625. https://doi.org/10.2337/dc08-2175
3. Zhu, D., Toker, M., Shyr, W., Fram, E., Watts, K. L., & Agalliu, I. (2022). Association of
Obesity and Diabetes With Prostate Cancer Risk Groups in a Multiethnic Population. Clinical
Genitourinary Cancer, 20(3), 299-299.e10. https://doi.org/10. 1016/j.clgc.2022.01.016
4. Nasri, H., & Rafieian-Kopaei, M. (2014). Metformin: Current knowledge. Journal ofResearch
in Medical Sciences, 19(7), 658–664.
5. Pharmacologic Approaches to Glycemic Treatment: Standards of Medical Care in Diabetes2018. (2018). Diabetes Care, 41(Suppl 1), S73–S85. https://doi.org/10.2337/dc18-s008
6. Holman, R. R., Paul, S. K., Bethel, M. A., Matthews, D. R., & Neil, H. A. W. (2008). 10-Year
Follow-up ofIntensive Glucose Control in Type 2 Diabetes. The New England Journal of
Medicine, 359(15), 1577–1589. https://doi.org/10.1056/NEJMoa0806470
7. Statista. (2024). Number of metformin prescriptions in the U.S. from 2004 to 2021 [Graph].
Statista. Retrieved April 27, 2024, from https://www.statista.com/statistics/780332/metforminhydrochloride-prescriptions-number-in-the-us/
8. Lv, Z., & Guo, Y. (2020). Metformin and Its Benefits for Various Diseases. Frontiers in
Endocrinology (Lausanne), 11, 191–191. https://doi.org/10.3389/fendo.2020.00191
9. Hua, Y., Zheng, Y., Yao, Y., Jia, R., Ge, S., & Zhuang, A. (2023). Metformin and cancer
hallmarks: shedding new lights on therapeutic repurposing. Journal ofTranslational Medicine,
21(1), 403–403. https://doi.org/10.1186/s12967-023-04263-8
10. Franciosi, M., Lucisano, G., Lapice, E., Strippoli, G. F. M., Pellegrini, F., & Nicolucci, A.
(2013). Metformin Therapy and Risk of Cancer in Patients with Type 2 Diabetes: Systematic
Review. PloS One, 8(8), e71583–e71583. https://doi.org/10. 1371/journal.pone.0071583
11. Cui, H., Wang, Y., Yang, S., He, G., Jiang, Z., Gang, X., & Wang, G. (2022). Antidiabetic
medications and the risk of prostate cancer in patients with diabetes mellitus: A systematic
review and meta-analysis. Pharmacological Research, 177, 106094–106094.
https://doi.org/10.1016/j.phrs.2022.106094

13
12. Wang, C. P., Lehman, D. M., Lam, Y. F., Kuhn, J. G., Mahalingam, D., Weitman, S.,
Lorenzo, C., Downs, J. R., Stuart, E. A., Hernandez, J., Thompson, I. M., & Ramirez, A. G.
(2016). Metformin for Reducing Racial/Ethnic Difference in Prostate Cancer Incidence for Men
with Type II Diabetes. Cancer prevention research (Philadelphia, Pa.), 9(10), 779–787.
https://doi.org/10.1158/1940-6207.CAPR-15-0425
13. Hernandez, B. Y., Park, S. Y., Wilkens, L. R., Henderson, B. E., & Kolonel, L. N. (2009).
Relationship of Body Mass, Height, and Weight Gain to Prostate Cancer Risk in the Multiethnic
Cohort. Cancer Epidemiology, Biomarkers & Prevention, 18(9), 2413–2421.
https://doi.org/10.1158/1055-9965.EPI-09-0293
14. Park, S.-Y., Haiman, C. A., Cheng, I., Park, S. L., Wilkens, L. R., Kolonel, L. N., Le
Marchand, L., & Henderson, B. E. (2015). Racial/ethnic differences in lifestyle-related factors
and prostate cancer risk: the Multiethnic Cohort Study. Cancer Causes & Control, 26(10), 1507–
1515. https://doi.org/10. 1007/s10552-015-0644-y
15. Cui, H., Wang, Y., Yang, S., He, G., Jiang, Z., Gang, X., & Wang, G. (2022). Antidiabetic
medications and the risk of prostate cancer in patients with diabetes mellitus: A systematic
review and meta-analysis. Pharmacological Research, 177, 106094–106094.
https://doi.org/10.1016/j.phrs.2022.106094
16. Tseng, C.-H. (2014). Metformin significantly reduces incident prostate cancer risk in
Taiwanese men with type 2 diabetes mellitus. European Journal of Cancer (1990), 50(16), 2831–
2837. https://doi.org/10. 1016/j.ejca.2014.08.007
17. Lee, Y. H. A., Zhou, J., Hui, J. M. H., Liu, X., Lee, T. T. L., Hui, K., Chan, J. S. K., Wai, A.
K. C., Wong, W. T., Liu, T., Ng, K., Lee, S., Dee, E. C., Zhang, Q., & Tse, G. (2022). Risk of
New-Onset Prostate Cancer for Metformin Versus Sulfonylurea Use in Type 2 Diabetes Mellitus:
A Propensity Score-Matched Study. Journal ofthe National Comprehensive Cancer Network,
20(6), 674–682. https://doi.org/10.6004/jnccn.2022.7010
18. Khan, S., Cai, J., Nielsen, M. E., Troester, M. A., Mohler, J. L., Fontham, E. T. H., Farnan,
L., Drake, B. F., Olshan, A. F., & Bensen, J. T. (2018). The association ofmetformin use with
prostate cancer aggressiveness among Black Americans and White Americans in a populationbased study. Cancer Causes & Control, 29(11), 1143–1150. https://doi.org/10.1007/s10552-018-
1087-z
19. Raval, A. D., Mattes, M. D., Madhavan, S., Pan, X., Wei, W., & Sambamoorthi, U. (2016).
Association between Metformin Use and Cancer Stage at Diagnosis among Elderly Medicare
Beneficiaries with Preexisting Type 2 Diabetes Mellitus and Incident Prostate Cancer. Journal of
Diabetes Research, 2016, 2656814–12. https://doi.org/10.1155/2016/2656814
20. Lopez, D. S., Malagaris, I., Polychronopoulou, E., Tsilidis, K. K., Milani, S. A., Kristen Peek,
M., Villasante‐Tezanos, A., Alzweri, L., Baillargeon, J., Kuo, Y., & Canfield, S. (2022).
Metformin and testosterone replacement therapy inversely associated with hormone‐associated

14
cancers (prostate, colorectal and male breast cancers) among older White and Black men.
Clinical Endocrinology (Oxford), 97(6), 792–803. https://doi.org/10.1111/cen.14803

vi
TABLES AND FIGURES
Table 1 Demographic and Lifestyle Factors by Pre-diagnostic Metformin Use
Characteristics Metformin Nonusers
(N = 28,846)
Metformin Users
(N = 2,216)
Race/Ethnicity, N (%)
Japanese American
White
Latino
African American
Native Hawaiian
Age at baseline, mean (SD)
First-degree family history of prostate cancer, N (%)
Negative
Positive
Missing
BMI, N (%)
Normal
Overweight
Obese
Missing
Education, N (%)
≥8th grade
9th-12th grade
Vocational school/some college
Graduated college or higher
Missing
History of diabetes, N (%)
No
Yes
Moderate or vigorous activity (MET-hours/day), N (%)
Quartile 1
Quartile 2
Quartile 3
Quartile 4
Missing
Smoking status, N (%)
Never
Former
Current
Missing
PSA testing, N (%)
No
Yes
Missing
10,375 (36.0)
6,514 (22.6)
6,343 (22.0)
3,437 (11.9)
2,177 (7.5)
58.81 (8.5)
25,068 (86.9)
3,667 (12.7)
111 (0.4)
10,532 (36.5)
13,745 (47.6)
4,462 (15.5)
107 (0.4)
2,334 (8.1)
7,213 (25.0)
8,697 (30. 1)
10,373 (36.0)
229 (0.8)
25,654 (88.9)
3,192 (11.1)
7,478 (25.9)
6,941 (24.1)
6,458 (22.4)
6,889 (23.9)
1,080 (3.7)
9,536 (33.1)
14,760 (51.2)
4,293 (14.9)
257 (0.9)
9,663 (33.5)
18,383 (63.7)
800 (2.8)
804 (36.3)
231 (10.4)
705 (31.8)
223 (10.1)
253 (11.4)
58.58 (7.7)
1,971 (88.9)
234 (10.6)
11 (0.5)
351 (15.8)
1,043 (47.1)
809 (36.5)
13 (0.6)
247 (11.1)
645 (29.1)
688 (31.0)
610 (27.5)
26 (1.2)
509 (23.0)
1,707 (77.0)
744 (33.6)
509 (23.0)
441 (19.9)
440 (19.9)
82 (3.7)
595 (26.9)
1,238 (55.9)
353 (15.9)
30 (1.4)
819 (37.0)
1,340 (60.5)
57 (2.6)

vii
Table 2 Association of pre-diagnostic metformin use with prostate cancer incidence by race/ethnicity.
Phenotype
Overall
(N = 31,062)
White
(N = 6,745)
African
American
(N = 3,660)
Japanese
American
(N = 11,179)
Latino
(N = 7,048)
Native American
(N = 2,430) PLRTa
Total PCa
No. of events 2,748 515 572 899 588 174
HR (95% CI) 0.87 (0.72-1.05) 0.82 (0.43-1.57) 1. 10 (0.73-1.67) 0.56 (0.39-0.82) 1.13 (0.81-1.59) 0.77 (0.44-1.35) 0.053
P 0.16 0.54 0.64 0.003 0.47 0.37
High-grade PCa
No. of events 639 147 103 239 104 46
HR (95% CI) 0.83 (0.56-1.22) 0.77 (0.22-2.69) 0.94 (0.35-2.57) 0.76 (0.39-1.48) 0.86 (0.36-2.07) 0.64 (0.27-1.53) 0.34
P 0.35 0.69 0.91 0.42 0.74 0.32
Low-grade PCa
No. of events 1945 346 407 634 438 120
HR (95% CI) 0.85 (0.68-1.07) 0.97 (0.45-2.09) 0.95 (0.56-1.61) 0.49 (0.31-0.79) 1. 16 (0.79-1.71) 0.91 (0.43-1.92) 0.13
P 0.17 0.93 0.85 0.003 0.46 0.80
Advanced PCa
No. of events 424 103 71 122 91 37
HR (95% CI) 0.91 (0.56-1.46) 0.34 (0.04-2.64) 1.80 (0.62-5.23) 1.05 (0.46-2.41) 0.67 (0.25-1.83) 0.66 (0.20-2.17) 0.54
P 0.68 0.30 0.28 0.91 0.44 0.50
Localized PCa
No. of events 2035 376 419 717 390 133
HR (95% CI) 0.86 (0.69-1.07) 0.91 (0.42-1.94) 1.05 (0.63-1.74) 0.51 (0.33-0.78) 1.22 (0.81-1.84) 0.89 (0.46-1.69) 0.051
P 0.18 0.80 0.85 0.002 0.34 0.71
Aggressive PCa
No. of events 880 198 147 308 158 69
HR (95% CI) 0.77 (0.55-1.07) 0.62 (0.18-2.09) 1.06 (0.47-2.36) 0.75 (0.42-1.34) 0.69 (0.34-1.42) 0.56 (0.24-1.29) 0.63
P 0.12 0.44 0.89 0.34 0.32 0.17
Non-aggressive PCa
No. of events 1,599 289 340 536 333 101
HR (95% CI) 0.92 (0.72-1.18) 0.94 (0.41-2.14) 1.08 (0.61-1.89) 0.48 (0.29-0.81) 1.36 (0.89-2. 10) 1.07 (0.49-2.30) 0.054
P 0.53 0.88 0.80 0.006 0.16 0.87
aPLRT from the LRT on the effect modification by ethnicity.

viii
Table 3 Association of pre-diagnostic metformin use with prostate cancer incidence
stratified by PSA screening status prior to cancer diagnosis in the overall study.
Phenotype
PSA
(N = 19,723)
No PSA
(N = 10,482) PLRT
Total PCa
No. of events 1,875 807
HR (95% CI) 0.89 (0.71-1.13) 0.80 (0.57-1. 12) 0.76
P 0.35 0.20
High-grade PCa
No. of events 421 207
HR (95% CI) 1.01 (0.63-1.60) 0.59 (0.28-1.22) 0.19
P 0.97 0.15
Low-grade PCa
No. of events 1,337 560
HR (95% CI) 0.79 (0.59-1.07) 0.87 (0.59-1.29) 0.28
P 0.13 0.49
Advanced PCa
No. of events 283 131
HR (95% CI) 0.98 (0.55-1.76) 0.67 (0.27-1.64) 0.60
P 0.95 0.38
Localized PCa
No. of events 1,380 611
HR (95% CI) 0.98 (0.55-1.76) 0.67 (0.27-1.64) 0.49
P 0.95 0.38
Aggressive PCa
No. of events 581 281
HR (95% CI) 0.90 (0.60-1.35) 0.53 (0.27-1.01) 0.27
P 0.60 0.054
Non-aggressive PCa
No. of events 1,096 467
HR (95% CI) 0.88 (0.64-1.21) 0.92 (0.61-1.41) 0.31
P 0.44 0.72

ix
Table 4 Association of pre-diagnostic metformin use with prostate cancer incidence
stratified by BMI at baseline in the overall study.
Phenotype <25 kg/m2
(N = 10,883)
25-30 kg/m2
(N = 14,788)
>= 30 kg/m2
(N = 5,271) PLRT
Total PCa
No. of events 987 1,337 412
HR (95% CI) 1.00 (0.67-1.50) 0.90 (0.69-1. 18) 0.79 (0.55-1.12) 0.21
P 1.00 0.44 0.19
High-grade PCa
No. of events 214 329 91
HR (95% CI) 1.59 (0.76-3.32) 0.76 (0.44-1.30) 0.60 (0.26-1.38) 0.15
P 0.22 0.31 0.23
Low-grade PCa
No. of events 701 941 297
HR (95% CI) 0.92 (0.56-1.51) 0.84 (0.59-1. 18) 0.88 (0.59-1.31) 0.24
P 0.73 0.31 0.53
Advanced PCa
No. of events 123 229 71
HR (95% CI) 1.34 (0.36-4.96) 1.20 (0.63-2.28) 0.49 (0.21-1.14) 0.88
P 0.67 0.58 0.10
Localized PCa
No. of events 749 976 301
HR (95% CI) 1.03 (0.66-1.62) 0.81 (0.58-1. 13) 0.90 (0.60-1.36) 0.15
P 0.89 0.21 0.63
Aggressive PCa
No. of events 290 457 128
HR (95% CI) 1.25 (0.61-2.57) 0.81 (0.51-1.29) 0.47 (0.24-0.95) 0.28
P 0.54 0.38 0.04
Non-aggressive PCa
No. of events 586 763 245
HR (95% CI) 1.03 (0.61-1.72) 0.90 (0.62-1.31) 0.96 (0.61-1.50) 0.32
P 0.92 0.60 0.86

x
Table 5 Association of pre-diagnostic metformin use with prostate cancer incidence
stratified by smoking status at baseline in the overall study.
Phenotype Never
(N= 10,131)
Former
(N= 15,998)
Current
(N= 4,646) PLRT
Total PCa
No. of events 960 1406 348
HR (95%CI) 0.84 (0.59-1.20) 0.89 (0.69-1.14) 0.73 (0.43-1.24) 0.51
P 0.33 0.34 0.25
High-grade PCa
No. of events 208 359 66
HR (95%CI) 0.94 (0.47-1.88) 0.75 (0.44-1.26) 0.82 (0.26-2.52) 0.76
P 0.87 0.28 0.72
Low-grade PCa
No. of events 702 969 248
HR (95%CI) 0.82 (0.54-1.26) 0.86 (0.63-1.18) 0.69 (0.36-1.30) 0.52
P 0.37 0.36 0.25
Advanced PCa
No. of events 144 210 63
HR (95%CI) 1.45 (0.60-3.47) 0.77 (0.40-1.49) 0.55 (0.12-2.50) 0.32
P 0.41 0.44 0.44
Localized PCa
No. of events 725 1,045 243
HR (95%CI) 0.75 (0.49-1.14) 0.91 (0.68-1.23) 0.73 (0.39-1.35) 0.72
P 0.18 0.55 0.32
Aggressive PCa
No. of events 289 474 108
HR (95%CI) 0.77 (0.40-1.46) 0.74 (0.47-1.16) 0.74 (0.30-1.83) 0.55
P 0.42 0.19 0.51
Non-aggressive PCa
No. of events 586 794 199
HR (95%CI) 0.90 (0.57-1.41) 0.94 (0.67-1.31) 0.73 (0.36-1.47) 0.70
P 0.64 0.71 0.38

Abstract (if available)

Abstract Background: Past research has generated mixed findings regarding the association between pre- diagnostic metformin use and risk of prostate cancer. These studies were conducted predominantly in White populations and had limited focus on aggressive forms of prostate cancer.

Methods: In the Multiethnic Cohort (MEC), we conducted a cohort analysis to investigate the association between pre-diagnostic metformin use and incidence of prostate cancer (total, localized, advanced, low-grade, high-grade, aggressive, and non-aggressive) in 31,062 male participants from five major racial/ethnic groups (White, African American, Japanese American, Latino, and Native Hawaiian). We also evaluated potential effect modification by history of prostate-specific antigen (PSA) screening, body mass index (BMI), and smoking status at baseline.

Results: Among 31,062 men included in this analysis, 2,216 (7. 1%) of them reported metformin use, and 2,748 (8.8%) developed prostate cancer. After adjusting for potential confounders, a non significant negative association was observed between pre-diagnostic metformin use and all prostate cancer phenotypes with the hazard ratio (HR) ranging from 0.77 to 0.92. In Japanese Americans (36% of overall study), pre-diagnostic metformin use was significantly associated with a substantially lower risk of total (HR = 0.56 [0.39-0.82], p = 0.003), low-grade (HR = 0.49 [0.31-0.79], p = 0.003), localized (HR = 0.51 [0.33-0.78], p = 0.002), and non-aggressive prostate cancer (HR = 0.48 [0.29-0.81], p = 0.006). A significantly lower risk of aggressive prostate cancer was also observed among men with no history of PSA screening (HR = 0.53 [0.27, 1.01], p = 0.054) or men who were obese at baseline (HR = 0.47 [0.24, 0.95], p = 0.04).

Conclusion: These results provide further support that the association of metformin use with risk of prostate cancer may be heterogeneous across racial/ethnic populations. Adding to the existing evidence, pre-diagnostic metformin was also found to be associated with lower risks of aggressive forms of prostate cancer in populations with shared specific characteristics.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Pathogenic variants in cancer predisposition genes and risk of non-breast multiple primary cancers in breast cancer patients

PDF

Body size and the risk of prostate cancer in the multiethnic cohort

PDF

Diet quality and pancreatic cancer incidence in the multiethnic cohort

PDF

Machine learning-based breast cancer survival prediction

PDF

Understanding prostate cancer genetic susceptibility and chromatin regulation

PDF

Racial/ethnic differences in patient experiences with health care in association with the Gleason’s score level at prostate cancer diagnosis: findings from the SEER-CAHPS data

PDF

Fish consumption and risk of colorectal cancer

PDF

Comparison of Cox regression and machine learning methods for survival analysis of prostate cancer

PDF

Association of comorbidity with prostate cancer tumor characteristics in African American men

PDF

The ADRB3 TRP64ARG variant and obesity in African American breast cancer cases

PDF

Prostate cancer: genetic susceptibility and lifestyle risk factors

PDF

A comparison of three different sources of data in assessing the adolescent and young adults cancer survivors

PDF

Identification and fine-mapping of genetic susceptibility loci for prostate cancer and statistical methodology for multiethnic fine-mapping

PDF

Extremity primary tumors in non-rhabdomyosarcoma soft tissue sarcoma: survival analysis

PDF

Air pollution and breast cancer survival in California teachers: using address histories and individual-level data

PDF

Functional analysis of a prostate cancer risk enhancer at 7p15.2

PDF

Analysis of factors associated with breast cancer using machine learning techniques

PDF

An assessment of necrosis grading in childhood osteosarcoma: the effect of initial treatment on prognostic significance

PDF

Comparative study of the POG and FNCLCC grading systems in non-rhabdomyosarcoma soft tissue sarcomas

PDF

Prostate cancer disparities among Californian Latinos by country of origin: clinical characteristics, incidence, treatment received and survival

Asset Metadata

Creator Chu, Kaixin Krystal (author)

Core Title The association of prediagnostic metformin use with prostate cancer in the multiethnic cohort study

School Keck School of Medicine

Degree Master of Science

Degree Program Biostatistics

Degree Conferral Date 2024-05

Publication Date 05/22/2024

Defense Date 05/21/2024

Publisher Los Angeles, California (original), University of Southern California (original), University of Southern California. Libraries (digital)

Tag metformin,OAI-PMH Harvest,prostate cancer

Format theses (aat)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Chen, Fei (committee chair), Barkauskas, Donald (committee member), Li, Ming (committee member)

Creator Email kaixinch@usc.edu,krystalchu1220@gmail.com

Permanent Link (DOI) https://doi.org/10.25549/usctheses-oUC113954490

Unique identifier UC113954490

Identifier etd-ChuKaixinK-13005.pdf (filename)

Legacy Identifier etd-ChuKaixinK-13005

Document Type Thesis

Format theses (aat)

Rights Chu, Kaixin Krystal

Internet Media Type application/pdf

Type texts

Source 20240522-usctheses-batch-1160 (batch), University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the author, as the original true and official version of the work, but does not grant the reader permission to use the work if the desired use is covered by copyright. It is the author, as rights holder, who must provide use permission if such use is covered by copyright.

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA

Repository Email cisadmin@lib.usc.edu