Antidepressant Use and Risk of Colorectal Cancer in The Women’s Health Initiative
Background
Some prior studies have reported reduced colorectal cancer risk among individuals using antidepressant (AD) medications, especially selective serotonin reuptake inhibitors (SSRIs). Yet, most studies have not considered the potential role of depression or other confounders in their analyses.
Methods
We utilized prospectively collected data from 145,190 participants in the Women’s Health Initiative, among whom 2,580 incident colorectal cancer cases were diagnosed. AD use and depressive symptoms were assessed at baseline and follow-up study visits. Cox proportional hazards regression models with adjustment for depressive symptoms and other covariates were utilized to estimate hazard ratios (HR) and 95% confidence intervals (CIs) for associations between AD use and colorectal cancer.
Results
AD use was reported by 6.9% of participants at baseline, with SSRIs the most common class of AD used. In multivariable analyses, including adjustment for depressive symptomology, we observed no statistically significant association between AD use overall (HR 0.90, 95% CI 0.75–1.09) or with SSRIs specifically (HR 1.08, 95% CI 0.85–1.37) and colorectal cancer risk. A borderline significant reduction in colorectal cancer risk was observed for use of tricyclic antidepressants (HR 0.76, 95% CI 0.56–1.04). Severe depressive symptoms were independently associated with a 20% increased risk of colorectal cancer (HR 1.21, 95% CI 1.09–1.48). Results were similar for separate evaluations of colon and rectal cancer.
Conclusions
We observed no evidence of an association between AD use, overall or by therapeutic class, and colorectal cancer risk.
Impact
These results suggest that ADs may not be useful as chemopreventive agents for colorectal cancer.
Introduction
Inflammation plays a key role in promoting colorectal cancer (CRC); for example, individuals with inflammatory conditions such as inflammatory bowel disease or Crohn’s disease have a higher CRC risk (1). Conversely, use of non-steroidal anti-inflammatory drugs (NSAIDs) is related to lower CRC risk(2). Antidepressant (AD) medications also have documented anti-inflammatory actions (3, 4).
Serotonin may promote angiogenesis of colon cancer cells, which requires cellular uptake of serotonin (5). Selective serotonin reuptake inhibitors (SSRIs) reduce colorectal tumor cell growth in both mouse xenograft models (6–9) and in in vitro studies (6, 10), perhaps through preventing angiogenesis and promoting apoptosis. Laboratory data suggest tricyclic antidepressants (TCAs) also increase apoptosis of colorectal cancer cells (11, 12), although some animal studies report that TCAs increase colorectal cancer cell growth (13, 14). Because of their potential anti-inflammatory and serotonergic reuptake inhibitory actions, ADs, especially SSRIs, have been hypothesized to be associated with reduced CRC risk.
AD use is increasingly common: 24.4% of U.S. women ages ≥60 reported using ADs in the past month(15). ADs are primarily used to treat depression, although some additional indications exist (e.g. treatment of menopausal vasomotor symptoms). SSRIs are the most commonly used AD class, followed by TCAs. Several studies report decreased CRC risk among regular AD users compared to non-users (16–18), with especially strong risk reductions associated with SSRI use, specifically (17, 18). For example, a case-control study reported a significant 45% reduction in CRC risk among regular SSRI users (OR 0.55, 95% CI: 0.35–0.88) compared with non-users (17). TCA use was not significantly associated with CRC risk (17, 18). Additionally, two recent record-linkage studies reported no association of AD use overall or by class with CRC risk (19, 20). Given the high prevalence of AD use in the population, it is critical to clarify the relationship, if any, between AD use and risk of CRC.
An important limitation of prior studies is the lack of consideration of depression. Depression is itself an inflammatory condition and has been associated with an increased risk of CRC (21). Because the vast majority of ADs are prescribed to treat depression, failure to adjust for depression or depressive symptoms may result in important confounding or interaction effects, obscuring the true association between ADs and CRC. Additional limitations of prior work include use of historical cohorts based on record linkages from large databases that lack information on additional relevant confounders. While reports of significant decreases in CRC risk associated with AD use are provocative, findings are not consistent across studies and the true effects of ADs on CRC risk remain unclear.
We sought to clarify whether ADs, overall and by class, affect CRC risk using data from the large, prospective Women’s Health Initiative (WHI) cohort.
Materials and Methods
Study Population
WHI participants were recruited from 40 clinical centers nationwide between October 1, 1993 and December 21, 1998. Details of WHI recruitment have been reported previously (22). Briefly, the WHI consisted of three clinical trials (CT): postmenopausal hormone therapy (HT; N=27,347), dietary modification (DM; N=48,835), and calcium/vitamin D supplementation (CaD; N=36,282); participants were able to take part in more than one trial. Women who were either not interested or not eligible for the CT were enrolled in an observational study (OS; N=93,676). All participants were between the ages of 50 and 79 at the time of enrollment. Participants provided data through annual in-person clinic visits (CT) or through annual mailed questionnaires and in-person clinic visits every three years (OS). Participants were followed for up to 10 years within the WHI, and many continue follow-up in the WHI Extension Study.
For the present analysis we included eligible women enrolled in the CT and the OS. We excluded women missing information on AD use at baseline (N=2), missing information on follow-up time (N=691), or reporting a history of cancer other than non-melanoma skin cancer at baseline (N=15,925). Our primary analysis included 145,190 women with an average length of follow up time of 14.3 years.
Measurement of Antidepressant Use and Depression
At the enrollment visit and subsequent clinic visits, participants brought in all current prescription medications used regularly, defined as use for at least two weeks. Information on AD medication name and duration of use was recorded by clinical interviewers, and Medi-Span software (First Databank Inc., San Bruno, CA) was used to assign therapeutic class codes to each medication.
We categorized women as AD users or non-users at baseline, and created separate indicator variables for each class of AD used (SSRI, TCA, and other AD). Duration of use was dichotomized at the median (<2 years, ≥2 years). Additionally, we incorporated AD use data collected at the year 3 clinic visit, when medication data were available for both OS and CT participants, to evaluate consistency of use (i.e. never AD use, AD use at baseline only, AD use at year 3 only, AD use at baseline and year 3).
Information on depressive symptoms was self-reported at baseline using the Burnam eight item scale (23). Consistent with prior studies in the WHI (24, 25), Burnam scores at or above 0.06 were used to classify women with severe depressive symptoms. We relied on depressive symptoms at baseline and did not incorporate subsequent measurements of depressive symptoms in the present analysis.
Measurement of Colorectal Cancer Incidence
Participants initially self-reported diagnoses of CRC on the semi-annual (CT) or annual (OS) questionnaires. Self-reports were then centrally adjudicated as previously described (26). Briefly, medical records were collected and CRC diagnoses confirmed through review of pathology reports. Tumor registry coders determined grade and stage of the tumor. We defined CRC cases as those confirmed by adjudication in these analyses.
Measurement of Covariates
Additional information on potentially important covariates was derived from data provided at the baseline clinic visit. We considered the following covariates, based on prior knowledge of relationships with AD use and/or CRC: age (continuous), race/ethnicity (Caucasian, Asian Pacific Islander, African American, Hispanic), smoking status (never, <5 pack-years, 5–20 pack-years, >20 pack-years, smokers with unknown pack-years), diabetes (no, yes), diverticulitis (no, yes), hypertension (no, yes), primary care provider (no, yes), hormone replacement therapy (HT) use at baseline (never, past estrogen-alone, past estrogen and progesterone, current estrogen, current estrogen and progesterone), WHI HT trial study arm (not in HT trial, estrogen-alone intervention, estrogen-alone control, estrogen+progesterone intervention, estrogen+progesterone control), physical activity (MET-hours/week (quartiles)), family history of CRC (no, yes), ever had CRC screening (no, yes), history of colorectal polyp removal (no, yes), alcohol use (none, past <1 drink per month, <1 drink per week, 1–7 drinks per week, 7+ drinks per week), current duration of non-steroidal- anti-inflammatory drugs (NSAIDs) use (never, <1 year, 1–3 years, 3–8 years, and 8–20 years), BMI (underweight, <18.5 kg/m; normal, 18.5–24.9 kg/m; overweight, 24.9–<30 kg/m; obese, ≥30kg/m), and healthy eating index (HEI) score (quartiles).
Statistical Analysis
We evaluated the distribution of covariates between AD users and non-users using t-tests or chi-square tests as appropriate.
We used multivariable Cox proportional hazards regression to estimate hazard ratios (HR) and 95% confidence intervals (CI) of the association between AD use and time to CRC diagnosis while adjusting for potential confounders. Our initial multivariable model included all variables with p<0.25 in univariate analyses, and backward selection was used based on likelihood ratio tests to generate a parsimonious model including all variables with p<0.05; age, WHI study arm, and BMI were included in the final model regardless of statistical significance. Women contributed person-time to the analysis until diagnosis of CRC, death, loss-to follow-up, or September 30, 2015, whichever happened first. We also evaluated associations between AD class (SSRI, TCA, other AD) and CRC, and also by the duration of AD use. In secondary analyses we separately examined the risk of colon and rectal cancers. Additionally, we examined the joint distribution of AD use and depression by testing for the significance of an interaction term between AD use and depression as well as modeling the effect of a jointly defined exposure variable: no AD use/no depression, AD use only, depression only, AD use and depression.
We incorporated AD use information from the year 3 clinical visit to evaluate associations between groups defined by consistency of AD use at baseline and year 3 (never, AD use at baseline only, AD use at year 3 only, and AD use at both baseline and year 3) on CRC risk. For these analyses we restricted our sample to women alive and at risk for incident CRC at their year 3 visit. Follow-up time began on the date of the year 3 visit and continued until a censoring event as defined above.
Two sided p-values (p≤ 0.05) were considered statistically significant. All analyses were performed with STATA version 14.0 (Stata Corporation, College Station, TX).
Study Population
WHI participants were recruited from 40 clinical centers nationwide between October 1, 1993 and December 21, 1998. Details of WHI recruitment have been reported previously (22). Briefly, the WHI consisted of three clinical trials (CT): postmenopausal hormone therapy (HT; N=27,347), dietary modification (DM; N=48,835), and calcium/vitamin D supplementation (CaD; N=36,282); participants were able to take part in more than one trial. Women who were either not interested or not eligible for the CT were enrolled in an observational study (OS; N=93,676). All participants were between the ages of 50 and 79 at the time of enrollment. Participants provided data through annual in-person clinic visits (CT) or through annual mailed questionnaires and in-person clinic visits every three years (OS). Participants were followed for up to 10 years within the WHI, and many continue follow-up in the WHI Extension Study.
For the present analysis we included eligible women enrolled in the CT and the OS. We excluded women missing information on AD use at baseline (N=2), missing information on follow-up time (N=691), or reporting a history of cancer other than non-melanoma skin cancer at baseline (N=15,925). Our primary analysis included 145,190 women with an average length of follow up time of 14.3 years.
Measurement of Antidepressant Use and Depression
At the enrollment visit and subsequent clinic visits, participants brought in all current prescription medications used regularly, defined as use for at least two weeks. Information on AD medication name and duration of use was recorded by clinical interviewers, and Medi-Span software (First Databank Inc., San Bruno, CA) was used to assign therapeutic class codes to each medication.
We categorized women as AD users or non-users at baseline, and created separate indicator variables for each class of AD used (SSRI, TCA, and other AD). Duration of use was dichotomized at the median (<2 years, ≥2 years). Additionally, we incorporated AD use data collected at the year 3 clinic visit, when medication data were available for both OS and CT participants, to evaluate consistency of use (i.e. never AD use, AD use at baseline only, AD use at year 3 only, AD use at baseline and year 3).
Information on depressive symptoms was self-reported at baseline using the Burnam eight item scale (23). Consistent with prior studies in the WHI (24, 25), Burnam scores at or above 0.06 were used to classify women with severe depressive symptoms. We relied on depressive symptoms at baseline and did not incorporate subsequent measurements of depressive symptoms in the present analysis.
Measurement of Colorectal Cancer Incidence
Participants initially self-reported diagnoses of CRC on the semi-annual (CT) or annual (OS) questionnaires. Self-reports were then centrally adjudicated as previously described (26). Briefly, medical records were collected and CRC diagnoses confirmed through review of pathology reports. Tumor registry coders determined grade and stage of the tumor. We defined CRC cases as those confirmed by adjudication in these analyses.
Measurement of Covariates
Additional information on potentially important covariates was derived from data provided at the baseline clinic visit. We considered the following covariates, based on prior knowledge of relationships with AD use and/or CRC: age (continuous), race/ethnicity (Caucasian, Asian Pacific Islander, African American, Hispanic), smoking status (never, <5 pack-years, 5–20 pack-years, >20 pack-years, smokers with unknown pack-years), diabetes (no, yes), diverticulitis (no, yes), hypertension (no, yes), primary care provider (no, yes), hormone replacement therapy (HT) use at baseline (never, past estrogen-alone, past estrogen and progesterone, current estrogen, current estrogen and progesterone), WHI HT trial study arm (not in HT trial, estrogen-alone intervention, estrogen-alone control, estrogen+progesterone intervention, estrogen+progesterone control), physical activity (MET-hours/week (quartiles)), family history of CRC (no, yes), ever had CRC screening (no, yes), history of colorectal polyp removal (no, yes), alcohol use (none, past <1 drink per month, <1 drink per week, 1–7 drinks per week, 7+ drinks per week), current duration of non-steroidal- anti-inflammatory drugs (NSAIDs) use (never, <1 year, 1–3 years, 3–8 years, and 8–20 years), BMI (underweight, <18.5 kg/m; normal, 18.5–24.9 kg/m; overweight, 24.9–<30 kg/m; obese, ≥30kg/m), and healthy eating index (HEI) score (quartiles).
Statistical Analysis
We evaluated the distribution of covariates between AD users and non-users using t-tests or chi-square tests as appropriate.
We used multivariable Cox proportional hazards regression to estimate hazard ratios (HR) and 95% confidence intervals (CI) of the association between AD use and time to CRC diagnosis while adjusting for potential confounders. Our initial multivariable model included all variables with p<0.25 in univariate analyses, and backward selection was used based on likelihood ratio tests to generate a parsimonious model including all variables with p<0.05; age, WHI study arm, and BMI were included in the final model regardless of statistical significance. Women contributed person-time to the analysis until diagnosis of CRC, death, loss-to follow-up, or September 30, 2015, whichever happened first. We also evaluated associations between AD class (SSRI, TCA, other AD) and CRC, and also by the duration of AD use. In secondary analyses we separately examined the risk of colon and rectal cancers. Additionally, we examined the joint distribution of AD use and depression by testing for the significance of an interaction term between AD use and depression as well as modeling the effect of a jointly defined exposure variable: no AD use/no depression, AD use only, depression only, AD use and depression.
We incorporated AD use information from the year 3 clinical visit to evaluate associations between groups defined by consistency of AD use at baseline and year 3 (never, AD use at baseline only, AD use at year 3 only, and AD use at both baseline and year 3) on CRC risk. For these analyses we restricted our sample to women alive and at risk for incident CRC at their year 3 visit. Follow-up time began on the date of the year 3 visit and continued until a censoring event as defined above.
Two sided p-values (p≤ 0.05) were considered statistically significant. All analyses were performed with STATA version 14.0 (Stata Corporation, College Station, TX).
Results
At baseline, 6.9% of the study population reported current AD use. AD users were more likely to be obese, smokers, diabetic, hypertensive, and depressed (Table 1), and they also were more likely to have a primary care provider and to have had CRC screening and a history of colorectal polyp removal. Current use of postmenopausal hormone therapy and NSAIDs was higher among women using ADs compared to non-users, while physical activity levels were lower among AD users (Table 1). Of the 7% of participants who used ADs, SSRIs were the most common, with 51.1% of AD users taking SSRIs, followed by 40.7% taking TCAs and 15.1% using others. Additionally, some women reported concurrent use of multiple AD classes (303 TCAs and SSRIs; 312 SSRIs and other AD; 81 TCAs and other AD; 7 all three classes). At baseline, the majority of women using SSRIs (56.3%) reported durations of use <2 years, while the majority of TCA users (62.1%) had used these medications for ≥2 years.
Table 1
Baseline characteristics of study population, by antidepressant (AD) use
| Characteristic | No AD Use N= 135,154 | Current AD Use N=10,036 |
|---|---|---|
| Age; Mean (SD) | 63.1 (7.2) | 62.0 (7.2) |
| Race/ethnicity; N (%) | ||
| Asian or Pacific Islander | 3,812 (2.8) | 77 (0.8) |
| Black or African American | 12,756 (9.5) | 538 (5.4) |
| Hispanic/Latino | 5,548 (4.1) | 339 (3.4) |
| White (not Hispanic origin) | 110,560 (82.0) | 8,913 (89.0) |
| Other | 2,150 (1.6) | 144 (1.4) |
| Body mass index, kg/m; N (%) | ||
| Underweight (<18.5) | 1,155 (0.9) | 75 (0.8) |
| Normal (18.5–<24.9) | 46,425 (34.6) | 2,771 (27.9) |
| Overweight (25.0–<30) | 46,658 (34.8) | 3,414 (34.3) |
| Obese (≥30) | 39,766 (29.7) | 3,683 (37.0) |
| Healthy Eating Index Score; N (%) | ||
| < 60.05 | 33,666 (25.0) | 2,762 (27.6) |
| 60.05–68.63 | 33,868 (25.1) | 2,493 (24.9) |
| 68.63–75.66 | 33,759 (25.0) | 2,468 (24.6) |
| ≥75.66 | 33,609 (24.9) | 2,296 (22.9) |
| Physical Activity, MET-hours/week; N (%) | ||
| <2.25 | 31,853 (24.8) | 3,143 (32.7) |
| 2.25–<8.34 | 32,220 (25.1) | 2,465 (25.6) |
| 8.34–<17.76 | 31,905 (24.9) | 2,168 (22.6) |
| ≥17.76 | 32,422 (25.3) | 1,838 (19.1) |
| Alcohol consumption; N (%) | ||
| None | 14,956 (11.2) | 963 (9.7) |
| Past | 24,022 (17.9) | 2,622 (26.3) |
| <1 drink/month | 16,732 (12.5) | 1,276 (12.8) |
| <1 drink/week | 27,638 (20.6) | 2,004 (20.1) |
| 1 to <7 drinks/week | 34,965 (26.1) | 2,158 (21.7) |
| >7 drinks/week | 15,821 (11.8) | 939 (9.4) |
| Smoking, Pack-years; N (%) | ||
| Never | 69,069 (51.8) | 4,477 (45.3) |
| <5 | 18,872 (14.1) | 1,406 (14.2) |
| 5–20 | 18,766 (14.1) | 1,369 (13.8) |
| >20 | 23,628 (17.7) | 2,418 (24.4) |
| Don’t know | 3,087 (2.3) | 222 (2.2) |
| Has a Primary Care Provider; N (%) | 124,894 (93.3) | 9,739 (97.8) |
| Has Had a Medical Visit within past year; N (%) | 105,461 (80.8) | 8,958 (91.5) |
| History of Diabetes; N (%) | 7,515 (5.6) | 854 (8.5) |
| History of Diverticulitis; N (%) | 9,982 (7.9) | 1,119 (11.8) |
| History of Hypertension; N (%) | 44,181 (32.7) | 4,047 (40.3) |
| Colorectal cancer screening (colonoscopy, sigmoidoscopy or flexible sigmoidoscopy); N (%) | ||
| Never | 64,091 (50.1) | 3,627 (37.9) |
| <5 years ago | 39,614 (31.0) | 3,935 (41.1) |
| >5 years ago | 23,940 (18.7) | 1,995 (20.8) |
| Yes, unsure of date | 235 (0.2) | 19 (0.2) |
| History of Colon Polyp Removal; N (%)1 | 10,623 (15.0) | 1,117 (17.4) |
| Depressive Symptoms; N (%) | ||
| Not Depressed | 119,126 (90.6) | 6,882 (70.9) |
| Depressed | 12,389 (9.4) | 2,832 (29.2) |
| Postmenopausal Hormone Therapy History; N (%) | ||
| Never | 60,307 (44.6) | 2,807 (28.0) |
| Past, Estrogen-alone | 12,682 (9.4) | 988 (9.8) |
| Past, Estrogen and Progesterone | 8,002 (5.9) | 669 (6.7) |
| Current Estrogen alone | 29,541 (21.9) | 3,318 (33.1) |
| Current Estrogen and Progesterone | 24,621 (18.2) | 2,254 (22.5) |
| Nonsteroidal anti-inflammatory drug use; N (%) | ||
| Never | 80,716 (59.7) | 4,653 (46.4) |
| Use for up to 1 Year | 16,962 (12.6) | 1,618 (16.1) |
| 1–3 years | 12,377 (9.2) | 1,193 (11.9) |
| 3–8 years | 11,652 (8.6) | 1,235 (12.3) |
| 8–20 years | 9,578 (7.1) | 1,020 (10.2) |
| >20 Years | 3,869 (2.9) | 317 (3.2) |
| First degree relative with colorectal cancer; N (%) | 20,267 (15.7) | 1,453 (15.2) |
| Hormone Trial (HT) study arm; N (%) | ||
| Not randomized to HT | 110,223 (81.6) | 8,586 (85.6) |
| Estrogen alone intervention | 4,719 (3.5) | 328 (3.3) |
| Estrogen alone control | 4,771 (3.5) | 330 (3.3) |
| Estrogen and progesterone intervention | 7,916 (5.9) | 396 (4.0) |
| Estrogen and progesterone control | 7,525 (5.6) | 396 (4.0) |
The median length of follow-up was 16.9 years (range 0.003 – 21.7 years). During the follow-up period, 2,580 women were diagnosed with CRC, including 2,150 colon cancers and 490 rectal cancers; 60 women had cancer in both the colon and rectal regions. We included participants with complete data on needed covariates in our multivariable analysis, including 2,197 cases of CRC. Among the CRC cases, 123 cases were diagnosed among women reporting AD use. In multivariable analyses (Table 2), we observed no statistically significant association between any AD use at baseline and CRC risk (HR 0.90, 95% CI 0.75–1.09). When examining by class, we found no association with SSRI use specifically (HR 1.08, 95% CI 0.85–1.37), though we did observe a borderline significant reduced risk associated with TCA use (HR 0.76, 95% CI 0.56–1.04), which was stronger and statistically significant among women using TCAs for <2 years (HR 0.42, 95% CI 0.22–0.81). We likewise observed borderline significant inverse associations with use of other ADs and CRC (HR 0.63, 95% CI 0.36–1.11), though the number of CRC cases among users of other ADs was very small (N=12 cases).
Table 2
Age- and multivariable adjusted hazard ratios (HR) and 95% confidence intervals (CI) of baseline antidepressant (AD) use and colorectal cancer risk1
| AD use category at baseline | Cases N=2,197 | Person-Years | Crude Incidence Rate (per 100,000 person-years) | Age-adjusted | Multivariable adjusted2 | ||
|---|---|---|---|---|---|---|---|
| HR | 95% CI | HR | 95% CI | ||||
| No AD use | 2074 | 1,646,416.70 | 125.97 | -- | -- | -- | -- |
| Any AD use | 123 | 117,083.73 | 105.05 | 0.90 | 0.75–1.08 | 0.90 | 0.75–1.09 |
| < 2 years of use | 56 | 56,058.89 | 99.89 | 0.87 | 0.66–1.13 | 0.85 | 0.65–1.11 |
| ≥ 2 years of use | 67 | 61,024.85 | 109.79 | 0.93 | 0.73–1.19 | 0.96 | 0.75–1.22 |
| No SSRI use | 2124 | 1,702,963.30 | 124.72 | -- | -- | -- | -- |
| SSRI use | 73 | 60,537.14 | 120.59 | 1.10 | 0.87–1.38 | 1.08 | 0.85–1.37 |
| < 2 years of use | 44 | 33,691.94 | 130.60 | 1.18 | 0.87–1.59 | 1.14 | 0.85–1.55 |
| ≥ 2 years of use | 29 | 26,845.20 | 108.03 | 0.99 | 0.69–1.43 | 0.99 | 0.68–1.43 |
| No TCA use | 2154 | 1,716,517.90 | 125.49 | -- | -- | -- | -- |
| TCA use | 43 | 46,982.57 | 91.52 | 0.74 | 0.55–1.00 | 0.76 | 0.56–1.04 |
| < 2 years of use | 9 | 17,955.97 | 50.12 | 0.41 | 0.21–0.79 | 0.42 | 0.22–0.81 |
| ≥ 2 years of use | 34 | 29,026.60 | 117.14 | 0.94 | 0.67–1.31 | 0.97 | 0.69–1.37 |
| No other AD use | 2185 | 1,746,072.80 | 125.14 | -- | -- | -- | -- |
| Other AD use | 12 | 17,427.66 | 68.86 | 0.62 | 0.35–1.10 | 0.63 | 0.36–1.11 |
| < 2 Years of Use | 5 | 9,253.94 | 54.03 | 0.50 | 0.21–1.20 | 0.47 | 0.19–1.13 |
| ≥ 2 Years of Use | 7 | 8,173.72 | 85.64 | 0.76 | 0.36–1.59 | 0.82 | 0.39–1.73 |
Overall, similar patterns were observed for colon cancer when examined separately in multivariable analyses. Of note, we observed a statistically significant inverse association between TCA use and colon cancer (N=32 cases; HR 0.68, 95% CI 0.48–0.96), which was statistically significant among women using these drugs for <2 years (N=7 cases; HR 0.39, 95% CI 0.19–0.82) but not among those using them for ≥2 years (N=25 cases; HR 0.85, 95% CI 0.57–1.26). There were too few rectal cancer cases to provide reliable estimates in multivariable analyses.
In our multivariable models we consistently observed a statistically significant approximate 20% increased risk of CRC associated with severe depressive symptoms (e.g. HR 1.21, 95% CI 1.09–1.48 in the model for AD use). Because of the high concordance between AD use and depression, we jointly evaluated their effect on CRC risk; we observed no statistically significant interaction between depressive symptoms and any of the AD use variables (all p>0.3).
At year 3, 4.4% of participants continued to use an AD, 4.4% started using an AD, 2.3% stopped using an AD, and 88.9% continued to be non-AD users. We further evaluated whether consistency of AD use between baseline and year 3 affected CRC risk (Table 3). We observed slight, non-statistically significant, inverse associations among users of any ADs or specific classes of ADs at both baseline and year 3 and among users at year 3 only.
Table 3
Age- and multivariable-adjusted hazard ratios (HR) and 95% confidence intervals (CI) of antidepressant (AD) use status between baseline and Year 3 with colorectal cancer risk.
| AD use at baseline and Year 3 | Cases N=1,544 | Person-Years | Crude Incidence Rate (per 100,000 person-years) | Age-adjusted | Multivariable-adjusted1 | ||
|---|---|---|---|---|---|---|---|
| HR | 95% CI | HR | 95% CI | ||||
| Any AD | |||||||
| Non-AD users | 1,407 | 1,110,703.9 | 126.68 | 1.00 | -- | 1.00 | -- |
| Baseline AD use only | 32 | 26,611.296 | 120.25 | 0.87 | 0.73–1.47 | 1.02 | 0.72–1.46 |
| Year 3 AD use only | 51 | 50,771.707 | 100.45 | 0.84 | 0.63–1.11 | 0.84 | 0.63–1.12 |
| Baseline and Year 3 AD use | 54 | 53,275.781 | 101.36 | 0.87 | 0.67–1.15 | 0.89 | 0.67–1.17 |
| SSRI | |||||||
| Non-SSRI users | 1,461 | 1,167,200.3 | 125.17 | 1.00 | -- | 1.00 | -- |
| Baseline SSRI use only | 25 | 15,843.157 | 157.80 | 1.46 | 0.98–2.16 | 1.43 | 0.96–2.13 |
| Year 3 SSRI use only | 32 | 33,121.128 | 96.62 | 0.84 | 0.59–1.19 | 0.84 | 0.59–1.20 |
| Baseline and Year 3 SSRI use | 26 | 25,198.174 | 103.18 | 0.94 | 0.63–1.38 | 0.93 | 0.63–1.38 |
| TCA | |||||||
| Non-TCA users | 1,496 | 1,193,388 | 125.36 | 1.00 | -- | 1.00 | -- |
| Baseline TCA use only | 13 | 14,668.704 | 88.62 | 0.73 | 0.42–1.26 | 0.74 | 0.43–1.28 |
| Year 3 TCA use only | 17 | 15,741.12 | 108.00 | 0.85 | 0.53–1.37 | 0.85 | 0.53–1.37 |
| Baseline and Year 3 TCA use | 18 | 17,564.942 | 102.48 | 0.82 | 0.52–1.31 | 0.87 | 0.55–1.39 |
Discussion
In this large prospective study of postmenopausal women, including 2,197 incident cases, we found no evidence of an association between total AD use or SSRI use specifically on the risk of CRC. CRC risk was 58% lower among short-term TCA users (<2 years) but not among longer term users; however, this result was not consistent with prior work and also was based on only nine cases. Additionally, AD use did not affect CRC risk even among women using these medications at both baseline and a follow-up visit three years later. Women with severe depressive symptoms had a 20% increase in CRC risk compared to those without, though no effect modification by AD use was apparent. While two record-linkage studies included more cases (19, 20), our study is the largest, prospective epidemiological cohort study to include participant-provided data on important potential confounders, including depression. Further, the careful and thorough collection of current medication use and the adjudication of CRC cases are important strengths and provide confidence in our results. Additionally, prior studies included both males and females, without a separate examination of associations in these groups. Our study population included only women, which may contribute to differences between our findings and those of prior work.
Prior evidence of the relationship between TCAs and CRC is contradictory. While some studies observed no association (17–20), another reported a non-significant 30% reduced CRC risk among TCA users (16). Interestingly, only short-term use of TCAs was associated with statistically significant reductions in CRC risk in our study. The failure to observe a decrease with longer duration of use may indicate acute effects in reducing risk that are not maintained. Alternatively, our findings could be related to the method of exposure assessment used in the WHI. In our study participants self-reported duration of use of each current medication only; thus AD use prior to the baseline visit or taken after baseline but discontinued before the year 3 visit were not captured. Given the episodic nature of depression as well as the tendency for patients to need to switch between and within classes of ADs to try to find the most effective treatment for their depression (27), we may have substantial misclassification in duration of use. Such misclassification of exposure could explain our non-significant results, especially for longer term use of specific classes that would require use of the exact same medication for an extended period of time (at least 2 years). Further, we were not able to incorporate AD use beyond year 3, which also might induce misclassification and attenuate our results. While our findings are suggestive of reduced CRC risk associated with TCA use, further prospective studies would need to confirm this finding, especially given the small numbers of cases contributing to the TCA use duration analyses.
We observed no association between SSRI use and CRC, in contrast to some prior studies reporting reduced CRC risk among SSRI users (16–18). One case-control study of SSRIs and CRC risk reported a significant 45% decreased risk among SSRI users (17), while another reported a 30% decreased risk among SSRI users taking a high daily dose within the previous 5 years (18). Another nested case-control study reported a 30% reduced CRC risk with any AD use, which was similar but not statistically significant when restricting to SSRI users only (16). However, our results are in agreement with two studies, one of which was prospective, that reported no association between SSRI use and CRC (17, 20).
When we jointly evaluated AD use and depression, we observed that women with depressive symptoms but without current AD use had a significant increase in CRC risk compared to women with neither AD use nor depression. Compared to healthy individuals, those with depression have higher levels of inflammatory markers, such as C-reactive protein, pro-inflammatory cytokines, and raised serum levels of interleukin-6, interleukin-1 beta, and tumor necrosis factor alpha (28, 29), which may be involved in development of CRC (30). Our findings are consistent with those reported in the Nurses’ Health Study, in which depressive symptomatology was associated with an increased risk of CRC (21). Similarly, our measure of depression only included depressive symptoms as opposed to a clinical diagnosis, but it has acceptable validity in identifying individuals with severe depressive symptoms (31).
Our results must be interpreted in the context of some limitations. AD use was fairly low (7%) within our population, and small numbers of CRC cases were observed in some AD use categories. Second, SSRIs were new to the market at the time of WHI baseline. SSRI prescriptions quickly outpaced TCAs following their introduction in 1987, but long-term use of SSRIs was not common at WHI baseline in 1993–1997. Third, AD use was assessed by visual inspection of pill bottles participants brought with them to clinic visits; therefore, prior use of AD medications within the same class or in a different class was not captured, possibly causing non-differential misclassification and attenuating results. Our ability to estimate cumulative, long-term usage of ADs is limited as a result. However, this approach to assessing medication use has the advantage of capturing actual, current usage, as opposed to relying on pharmacy databases that identify filled prescriptions but offer no insight on whether the medications were actually taken. Additionally, current AD use was assessed only at baseline and year 3; therefore, we were unable to evaluate effects of cumulative duration of AD use on our outcomes.
Previous studies evaluating relationships between AD use and CRC are mainly derived from record linkages within large health insurance systems or population databases. While such studies are important and have many advantages, they typically lack the ability to fully adjust for potential confounders (e.g. diet, physical activity). Further, the reliance on database information on AD use may capture prescriptions written and/or filled, yet may overestimate actual usage if the medications were not taken. The differences in approach to classifying AD use may account for the different results observed across studies. Given these limitations and the inconsistent findings of prior studies, we sought to evaluate whether ADs were associated with CRC risk in a large, population-based cohort. Particular strengths of our study are its prospective design, large sample size, and comprehensive data on potential confounders, including depressive symptoms, utilization of CRC screening, history of colorectal polyp removal, family history of CRC, diet, and BMI. Importantly, all medication use was verified by a research nurse examining participants’ pill bottles, and all CRC cases included in the analysis were adjudicated, incident cancers.
In summary, we did not find compelling evidence that ADs reduced risk of CRC. Importantly, we observed no increase in CRC risk associated with use of any class of AD. Our findings agree with those of two large registry-based studies; the consistency of these findings indicate that there is likely no significant association between AD use and later risk of CRC. While it does not appear that ADs, especially SSRIs, would be useful as chemopreventive agents for CRC, it is reassuring that use of ADs, which can effectively treat depression in most individuals, does not result in elevated risk of CRC.
Acknowledgments
Funding: The Women’s Health Initiative program is supported by the National Heart, Lung, and Blood Institute, National Institutes of Health, U.S. Department of Health and Human Services through contracts HHSN268201600018C, HHSN268201600001C, HHSN268201600002C, HHSN268201600003C, and HHSN268201600004C. No additional funding was obtained for this particular analysis.
Program Office: (National Heart, Lung, and Blood Institute, Bethesda, Maryland) Jacques Rossouw, Shari Ludlam, Joan McGowan, Leslie Ford, and Nancy Geller Clinical Coordinating Center: (Fred Hutchinson Cancer Research Center, Seattle, WA) Garnet Anderson, Ross Prentice, Andrea LaCroix, and Charles Kooperberg Investigators and Academic Centers: (Brigham and Women’s Hospital, Harvard Medical School, Boston, MA) JoAnn E. Manson; (MedStar Health Research Institute/Howard University, Washington, DC) Barbara V. Howard; (Stanford Prevention Research Center, Stanford, CA) Marcia L. Stefanick; (The Ohio State University, Columbus, OH) Rebecca Jackson; (University of Arizona, Tucson/Phoenix, AZ) Cynthia A. Thomson; (University at Buffalo, Buffalo, NY) Jean Wactawski-Wende; (University of Florida, Gainesville/Jacksonville, FL) Marian Limacher; (University of Iowa, Iowa City/Davenport, IA) Jennifer Robinson; (University of Pittsburgh, Pittsburgh, PA) Lewis Kuller; (Wake Forest University School of Medicine, Winston-Salem, NC) Sally Shumaker; (University of Nevada, Reno, NV) Robert Brunner; (University of Minnesota, Minneapolis, MN) Karen L. Margolis
Abstract
Background
Some prior studies have reported reduced colorectal cancer risk among individuals using antidepressant (AD) medications, especially selective serotonin reuptake inhibitors (SSRIs). Yet, most studies have not considered the potential role of depression or other confounders in their analyses.
Methods
We utilized prospectively collected data from 145,190 participants in the Women’s Health Initiative, among whom 2,580 incident colorectal cancer cases were diagnosed. AD use and depressive symptoms were assessed at baseline and follow-up study visits. Cox proportional hazards regression models with adjustment for depressive symptoms and other covariates were utilized to estimate hazard ratios (HR) and 95% confidence intervals (CIs) for associations between AD use and colorectal cancer.
Results
AD use was reported by 6.9% of participants at baseline, with SSRIs the most common class of AD used. In multivariable analyses, including adjustment for depressive symptomology, we observed no statistically significant association between AD use overall (HR 0.90, 95% CI 0.75–1.09) or with SSRIs specifically (HR 1.08, 95% CI 0.85–1.37) and colorectal cancer risk. A borderline significant reduction in colorectal cancer risk was observed for use of tricyclic antidepressants (HR 0.76, 95% CI 0.56–1.04). Severe depressive symptoms were independently associated with a 20% increased risk of colorectal cancer (HR 1.21, 95% CI 1.09–1.48). Results were similar for separate evaluations of colon and rectal cancer.
Conclusions
We observed no evidence of an association between AD use, overall or by therapeutic class, and colorectal cancer risk.
Impact
These results suggest that ADs may not be useful as chemopreventive agents for colorectal cancer.
Footnotes
The authors declare that they have no conflicts of interest.
References
- 1. Lasry A, Zinger A, Ben-Neriah YInflammatory networks underlying colorectal cancer. Nat Immunol. 2016 Mar;17:230–40.[PubMed][Google Scholar]
- 2. Chubak J, Kamineni A, Buist DSM, Anderson ML, Whitlock EP Aspirin Use for the Prevention of Colorectal Cancer: An Updated Systematic Evidence Review for the US Preventive Services Task Force. Rockville (MD): 2015. [[PubMed][Google Scholar]
- 3. Kenis G, Maes MEffects of antidepressants on the production of cytokines. Int J Neuropsychopharmacol. 2002 Dec;5:401–12.[PubMed][Google Scholar]
- 4. Pizzi C, Mancini S, Angeloni L, Fontana F, Manzoli L, Costa GMEffects of selective serotonin reuptake inhibitor therapy on endothelial function and inflammatory markers in patients with coronary heart disease. Clin Pharmacol Ther. 2009 Nov;86:527–32.[PubMed][Google Scholar]
- 5. Nocito A, Dahm F, Jochum W, Jang JH, Georgiev P, Bader M, et al Serotonin regulates macrophage-mediated angiogenesis in a mouse model of colon cancer allografts. Cancer Res. 2008 Jul 1;68:5152–8.[PubMed][Google Scholar]
- 6. Gil-Ad I, Zolokov A, Lomnitski L, Taler M, Bar M, Luria D, et al Evaluation of the potential anti-cancer activity of the antidepressant sertraline in human colon cancer cell lines and in colorectal cancer-xenografted mice. Int J Oncol. 2008 Aug;33:277–86.[PubMed][Google Scholar]
- 7. Barkla DH, Tutton PJInfluence of histamine and serotonin antagonists on the growth of xenografted human colorectal tumors. J Natl Cancer Inst. 1981 Dec;67:1207–11.[PubMed][Google Scholar]
- 8. Tutton PJ, Barkla DHInfluence of inhibitors of serotonin uptake on intestinal epithelium and colorectal carcinomas. Br J Cancer. 1982 Aug;46:260–5.[Google Scholar]
- 9. Tutton PJ, Steel GGInfluence of biogenic amines on the growth of xenografted human colorectal carcinomas. Br J Cancer. 1979 Nov;40:743–9.[Google Scholar]
- 10. Yue CT, Liu YLFluoxetine increases extracellular levels of 3-methoxy-4-hydroxyphenylglycol in cultured COLO320 DM cells. Cell Biochem Funct. 2005 Mar-Apr;23:109–14.[PubMed][Google Scholar]
- 11. Arimochi H, Morita KDesipramine induces apoptotic cell death through nonmitochondrial and mitochondrial pathways in different types of human colon carcinoma cells. Pharmacology. 2008;81:164–72.[PubMed][Google Scholar]
- 12. Arimochi H, Morita KCharacterization of cytotoxic actions of tricyclic antidepressants on human HT29 colon carcinoma cells. Eur J Pharmacol. 2006 Jul 10;541:17–23.[PubMed][Google Scholar]
- 13. Tutton PJ, Barkla DHEffect of an inhibitor of noradrenaline uptake, desipramine, on cell proliferation in the intestinal crypt epithelium. Virchows Arch B Cell Pathol Incl Mol Pathol. 1989;57:349–52.[PubMed][Google Scholar]
- 14. Iishi H, Tatsuta M, Baba M, Taniguchi HEnhancement by the tricyclic antidepressant, desipramine, of experimental carcinogenesis in rat colon induced by azoxymethane. Carcinogenesis. 1993 Sep;14:1837–40.[PubMed][Google Scholar]
- 15. Pratt LA, Brody DJ, Gu Q NCHS Data Brief. 2017. Aug, Antidepressant Use Among Persons Aged 12 and Over:United States,2011–2014; pp. 1–8. [[PubMed][Google Scholar]
- 16. Chubak J, Boudreau DM, Rulyak SJ, Mandelson MTColorectal cancer risk in relation to antidepressant medication use. Int J Cancer. 2011 Jan 1;128:227–32.[Google Scholar]
- 17. Coogan PF, Strom BL, Rosenberg LAntidepressant use and colorectal cancer risk. Pharmacoepidemiol Drug Saf. 2009 Nov;18:1111–4.[Google Scholar]
- 18. Xu W, Tamim H, Shapiro S, Stang MR, Collet JPUse of antidepressants and risk of colorectal cancer: a nested case-control study. Lancet Oncol. 2006 Apr;7:301–8.[PubMed][Google Scholar]
- 19. Cronin-Fenton DP, Riis AH, Lash TL, Dalton SO, Friis S, Robertson D, et al Antidepressant use and colorectal cancer risk: a Danish population-based case-control study. Br J Cancer. 2011 Jan 4;104:188–92.[Google Scholar]
- 20. Haukka J, Sankila R, Klaukka T, Lonnqvist J, Niskanen L, Tanskanen A, et al Incidence of cancer and antidepressant medication: record linkage study. Int J Cancer. 2010 Jan 1;126:285–96.[PubMed][Google Scholar]
- 21. Kroenke CH, Bennett GG, Fuchs C, Giovannucci E, Kawachi I, Schernhammer E, et al Depressive symptoms and prospective incidence of colorectal cancer in women. Am J Epidemiol. 2005 Nov 1;162:839–48.[PubMed][Google Scholar]
- 22. Design of the Women’s Health Initiative clinical trial and observational studyThe Women’s Health Initiative Study Group. Control Clin Trials. 1998 Feb;19:61–109.[PubMed][Google Scholar]
- 23. Burnam MA, Wells KB, Leake B, Landsverk JDevelopment of a brief screening instrument for detecting depressive disorders. Med Care. 1988 Aug;26:775–89.[PubMed][Google Scholar]
- 24. Bertone-Johnson ER, Powers SI, Spangler L, Brunner RL, Michael YL, Larson JC, et al Vitamin D intake from foods and supplements and depressive symptoms in a diverse population of older women. Am J Clin Nutr. 2011 Oct;94:1104–12.[Google Scholar]
- 25. Spangler L, Scholes D, Brunner RL, Robbins J, Reed SD, Newton KM, et al Depressive symptoms, bone loss, and fractures in postmenopausal women. J Gen Intern Med. 2008 May;23:567–74.[Google Scholar]
- 26. Curb JD, McTiernan A, Heckbert SR, Kooperberg C, Stanford J, Nevitt M, et al Outcomes ascertainment and adjudication methods in the Women’s Health Initiative. Ann Epidemiol. 2003 Oct;13:S122–8.[PubMed][Google Scholar]
- 27. Managing chronic depression [Accessed: October 6, 2016];Harvard Mental Health Letter. 2009 Dec; Available: .[PubMed][Google Scholar]
- 28. Dowlati Y, Herrmann N, Swardfager W, Liu H, Sham L, Reim EK, et al A meta-analysis of cytokines in major depression. Biol Psychiatry. 2010 Mar 1;67:446–57.[PubMed][Google Scholar]
- 29. Howren MB, Lamkin DM, Suls JAssociations of depression with C-reactive protein, IL-1, and IL-6: a meta-analysis. Psychosom Med. 2009 Feb;71:171–86.[PubMed][Google Scholar]
- 30. Terzic J, Grivennikov S, Karin E, Karin MInflammation and colon cancer. Gastroenterology. 2010 Jun;138:2101–14e5.[PubMed][Google Scholar]
- 31. Tuunainen A, Langer RD, Klauber MR, Kripke DFShort version of the CES-D (Burnam screen) for depression in reference to the structured psychiatric interview. Psychiatry Res. 2001 Sep 20;103:261–70.[PubMed][Google Scholar]