DOI:10.1093/jnci/djt438
Advance Access publication January 15, 2014
© The Author 2014. Published by Oxford University Press. All rights reserved.
For Permissions, please e-mail: [email protected].
Editorial
Antihypertension and Colorectal Cancer Prevention: Getting Two
Birds With One Stone?
Mingyang Song, Edward L. Giovannucci
Correspondence to: Edward L. Giovannucci, MD, ScD, Departments of Nutrition and Epidemiology, Harvard School of Public Health, 665 Huntington Avenue,
Boston, MA 02115 (e-mail: [email protected]).
As a classical antihypertensive drug, angiotensin I–converting
enzyme inhibitor (ACEI) has been suggested to protect against
cancer ever since Lever et al.’s landmark report in 1998 of a 28%
reduced cancer incidence among ACEI users compared with general control subjects (1). Despite the ongoing enthusiasm, most
prospective studies failed to find any secondary benefit of ACEI use
on colorectal cancer (CRC) or overall cancer risk (2–4). Similarly,
null associations have been observed for another antihypertensive
drug, angiotension receptor blocker (ARB), with CRC risk in all
but one prospective study (5). In a recent meta-analysis of 70 randomized controlled trials, no evidence of benefit was observed on
any cancer (including CRC) incidence or mortality after ACEI or
ARB therapy, although an increased overall cancer risk was suggested for the combinations of ACEI plus ARB in a fixed-effects
model (6). Several limitations of this secondary analysis of trials
have been noted, including short duration of included trials, cancer
as a nonpredefined outcome, domination of results by one trial, and
incomplete inclusion of studies (7,8).
In the current issue of the Journal, Makar et al. (9) report findings from a nested case–control study within a large primary care
database in the United Kingdom that examined the association
between ACEI/ARB therapy and risk of CRC. The study found a
16% reduction in CRC incidence after 3 or more years of ACEI/
ARB therapy among patients with a diagnosis of hypertension or
hypertension-related complication. The magnitude of this reduction increased with longer exposure and higher dose of therapy.
The large computerized database used by this study offers several important advantages, including a large sample size, a representative study population, and systematic and accurate collection
of medication information. However, some limitations of the study
need to be considered. First, it is not clear how occasional and regular drug users were distinguished for exposure definition, especially
when combined use of multiple antihypertensive drugs is common.
For example, a person using ACEI as an auxiliary therapy sporadically during the past 3 years has a substantially different exposure
history from the person using the drug every day in the same
period. Second, as in other observational studies, confounding is a
concern of this study, especially when only a moderate association
is observed. A previous study using the same database reported that
users of ACEIs and ARBs were more likely to be obese, to have ever
smoked, to have diabetes, and to use aspirin, statins, and nonsteroidal anti-inflammatory drugs than users of other antihypertensive
drugs (4). Whereas obesity, smoking, and diabetes are important
1 of 2 Editorial | JNCI
risk factors for CRC, long-term use of aspirin and nonsteroidal
anti-inflammatory drugs can substantially reduce CRC risk. Given
the complex lifestyle and medication pattern, it is difficult to determine the direction of aggregate confounding. Although the investigators considered several potential confounders, according to the
criterion of “including only factors leading to change of at least
10% in the association,” only the average number of physician contacts after the incident diagnosis of hypertension was adjusted in
addition to the matching factors (ie, sex, age, calendar period and
duration of follow-up after diagnosis of hypertension before the
index date). Although physician visits might reflect disease severity
to a certain extent, many factors can influence patients’ contacts
with doctors, and thus adjustment for this single composite factor
may not address the complicated confounding pattern in this study.
On the other hand, adjustment for physician visits could induce
selection bias because both drug use and subclinical symptoms or
important risk factors (eg, family history) for CRC can lead to more
frequent physician visits (10). This is more notable when a statistically significant association with 3 or more years of ACEI/ARB
therapy emerges only after such adjustment (crude relative risk
[RR] = 0.92, 95% confidence interval [CI] = 0.79 to 1.08; adjusted
RR = 0.84, 95% CI = 0.72 to 0.98), whereas statistically significant
associations with some other agents disappear after the adjustment
(for ≥5 years of diuretics use: crude RR = 0.80, 95% CI = 0.65 to
0.98; adjusted RR = 0.85, 95% CI = 0.70 to 1.05; for ≥5 years of
beta-blocker use: crude RR = 0.80, 95% CI = 0.67 to 0.95; adjusted
RR = 0.85, 95% CI = 0.71 to 1.01). Again, aggregation of small
biases can induce a spuriously moderate association.
Given the findings by Makar et al. (9), an obvious question arises: What is the mechanism if ACEI/ARB therapy really
reduces CRC risk among hypertension patients? Accumulating
data indicate that the renin-angiotensin system (RAS), the target
of ACEI and ARB action, is frequently dysregulated in malignancy.
Abnormalities in the RAS signaling influence proliferation and
apoptosis of cancer cells and promote angiogenesis and formation of cancer inflammatory microenvironment (11). Some experimental data suggest an anticarcinogenic role of RAS inhibitors in
colorectal tumors, although the exact mechanisms remain largely
unclear. Recent in vivo evidence demonstrates that ACEI or ARB
administration substantially reduces the total number of colonic
premalignant lesions and decreases oxidative stress and expression
of inflammatory cytokines in metabolically disordered mice, suggesting that suppression of inflammation might mediate the effect
Vol. 106, Issue 2 | djt438 | February 5, 2014
of RAS inhibitors against colorectal carcinogenesis (12). Because
obesity-related metabolic abnormalities may increase the risk of
CRC through augmented inflammation, it would be interesting to
examine whether obesity modifies the association of ACEI/ARB
use with CRC risk. In addition, treatment with ACEI/ARB and
cyclooxygenase-2 inhibitors has been shown to suppress expression
of insulin-like growth factor I receptor in colon cancer cells and
reduce tumor growth in vivo (13).
Despite the encouraging findings, more research needs to be
done before recommending clinical use of ACEI/ARB for CRC
chemoprevention. Cancer incidence should be monitored as a secondary outcome in the ongoing and future clinical trials of antihypertensive drugs. At the same time, further observational studies
should be conducted by using the existing cohorts or administrative health databases. When doing this, rigorous study design and
analysis are needed to minimize the influence of confounding and
biases, as proposed in the recent guidelines for good pharmacoepidemiology practices (14). Furthermore, the potential interaction
with lifestyle factors (eg, obesity) and genetic susceptibility warrants more investigation. A recent study suggests that a polymorphism in the ACE gene modifies the association between ACEI/
ARB therapy and CRC risk (15). In addition, the stage at which
these medications might impact the adenoma-carcinoma sequence
is speculative. ACEI use has been shown to be associated with
decreased risk of recurrence of advanced adenomatous colon polyps (16), and a recent cross-sectional study in a CRC screening
program found that ACEI therapy was associated with lower prevalence of advanced neoplasia at colonoscopy (CRC or intermediateor high-risk dysplastic polyps) (17). Given the sparse data, more
prospective studies are needed. Regarding mechanistic research
of RAS inhibitors, current focus is still on cancer progression and
metastasis, thus highlighting the need for more translational studies to examine CRC initiation-related change upon ACEI/ARB
therapy and decipher the involved mechanisms, if such effect exists.
In summary, given that 40% of adults aged 25 years and older
in the world have raised blood pressure, any protection of antihypertensive drugs against CRC, even small in magnitude, can
produce substantial public health benefits. The promising findings by Makar et al. (9) should provide a stimulus for further work.
However, at this moment clinicians should not recommend that
their patients take ACEI/ARB for prevention of CRC until further
conclusive evidence is obtained.
jnci.oxfordjournals.org
References
1. Lever AF, Hole DJ, Gillis CR, et al. Do inhibitors of angiotensin-I-converting enzyme protect against risk of cancer? Lancet. 1998;352(9123):179–184.
2.Boudreau DM, Koehler E, Rulyak SJ, et al. Cardiovascular medication use and risk for colorectal cancer. Cancer Epidemiol Biomarker Prev
2008;17(11):3076–3080.
3.Friis S, Sorensen HT, Mellemkjaer L, et al. Angiotensin-converting
enzyme inhibitors and the risk of cancer: a population-based cohort study
in Denmark. Cancer. 2001;92(9):2462–2470.
4. Azoulay L, Assimes TL, Yin H, et al. Long-term use of angiotensin receptor blockers and the risk of cancer. PloS One. 2012;7(12):e50893.
5. Huang CC, Chan WL, Chen YC, et al. Angiotensin II receptor blockers
and risk of cancer in patients with systemic hypertension. Am J Cardiol.
2011;107(7):1028–1033.
6. Bangalore S, Kumar S, Kjeldsen SE, et al. Antihypertensive drugs and risk
of cancer: network meta-analyses and trial sequential analyses of 324,168
participants from randomised trials. Lancet Oncol. 2011;12(1):65–82.
7.Barrios V, Escobar C. Angiotensin-receptor blockade, cancer, and concerns. Lancet Oncol. 2010;11(9):819–820; author reply 821–822.
8.Meredith PA, McInnes GT. Angiotensin-receptor blockade, cancer, and
concerns. Lancet Oncol. 2010;11(9):819; author reply 821–822.
9. Makar GA, Holmes JH, Yang Y-X. Angiotensin-converting enzyme inhibitor therapy and colorectal cancer risk. JNCI J Natl Cancer Inst (2014)
106 (2):djt374 doi:10.1093/jnci/djt374
10.Hernan MA, Hernandez-Diaz S, Robins JM. A structural approach to
selection bias. Epidemiology. 2004;15(5):615–625.
11. George AJ, Thomas WG, Hannan RD. The renin-angiotensin system and
cancer: old dog, new tricks. Nat Rev Cancer. 2010;10(11):745–759.
12. Kubota M, Shimizu M, Sakai H, et al. Renin-angiotensin system inhibitors
suppress azoxymethane-induced colonic preneoplastic lesions in C57BL/
KsJ-db/db obese mice. Biochem Biophys Res Comm. 2011;410(1):108–113.
13. Yasumaru M, Tsuji S, Tsujii M, et al. Inhibition of angiotensin II activity
enhanced the antitumor effect of cyclooxygenase-2 inhibitors via insulinlike growth factor I receptor pathway. Cancer Res. 2003;63(20):6726–6734.
14.International Society of Pharmacoepidemiology (ISPE). Guidelines for
good pharmacoepidemiology practices (GPP). Pharmacoepidemiol Drug
Safety. 2008;17(2):200–208.
15. van der Knaap R, Siemes C, Coebergh JW, et al. Renin-angiotensin system
inhibitors, angiotensin I-converting enzyme gene insertion/deletion polymorphism, and cancer: the Rotterdam Study. Cancer. 2008;112(4):748–757.
16.Kedika R, Patel M, Pena Sahdala HN, et al. Long-term use of angiotensin converting enzyme inhibitors is associated with decreased incidence of advanced adenomatous colon polyps. J Clin Gastroenterol.
2011;45(2):e12–e16.
17.Mansouri D, McMillan DC, Roxburgh CS, et al. The impact of aspirin,
statins and ACE-inhibitors on the presentation of colorectal neoplasia in a
colorectal cancer screening programme. Br J Cancer. 2013;109(1):249–256.
Affiliations of authors: Department of Nutrition (MS, ELG) and Department
of Epidemiology (ELG), Harvard School of Public Health, Boston MA.
JNCI | Editorial 2 of 2