24 June 2010
A meta-analysis of RCTs has found that diagnosis of new cancer was increased in patients randomised to receive angiotensin-II receptor antagonists (A2RAs), compared with control. The absolute increase in the risk of cancer was small (1.2%) but statistically significant. However, because of the large number of people taking A2RAs, even a small absolute increase would, if true, produce a large number of additional cancers.
Level of evidence:
Level 2 (limited quality patient-oriented evidence) according to the SORT criteria.
This meta-analysis provides a safety signal about a possible increased risk of cancer in people who are taking A2RAs. However, due to the inherent limitations of the data, this cannot be regarded as definitive. The authors point out that the clinical significance of the excess cancer risk they found is unclear, and the finding of a 1.2% increase in the absolute risk of new cancer diagnosis over an average of four years needs to be interpreted in view of the estimated 41% lifetime risk of cancer.
We anticipate that regulatory authorities will be examining this data. In the meantime, this safety concern adds weight to the argument that ACE inhibitors, not A2RAs, are the first-line choice when a renin-angiotensin system drug is indicated. ACE inhibitors have a more robust evidence base than A2RAs for all indications in terms of evidence for efficacy, safety and most patient factors. The major benefit of A2RAs over ACE inhibitors is a lower rate of cough. Hence, A2RAs are an alternative where a renin-angiotensin system drug is indicated, but an ACE inhibitor has to be discontinued because of an intolerable ACE inhibitor-induced cough.
What is the background to this?
Clinical studies of A2RAs have generally assessed their effects on cardiovascular and renal endpoints, rather than the incidence of cancers. The first indication of a possible increased risk of cancer with A2RAs was seen in the CHARM study, which unexpectedly found that the risk of fatal cancer was significantly higher in the candesartan group compared with the placebo group (2.3% vs. 1.6%, P=0.038). In addition, experimental studies have suggested that angiotensin-II receptors are involved in regulation of cell proliferation, angiogenesis, and tumour progression. This meta-analysis, which included all publicly available data for the development of cancers from RCTs of A2RAs, aimed to determine whether A2RAs are associated with the development of cancer.
What does this study claim?
The meta-analysis found that A2RAs were associated with a modest increase in the risk of new cancer occurrence compared with controls (7.2% vs. 6.0%, relative risk [RR] 1.08, 95% confidence interval [CI] 1.01 to 1.15, P=0.016). When analysis was limited to the three clinical trials where cancer was a prespecified endpoint and cancer data was rigorously collected, again, an increased risk of cancer was seen with A2RAs compared with controls (RR 1.11, 95%CI 1.04 to 1.18, P=0.001).
The number needed to harm (NNH) was calculated using the background cancer incidence for people aged 65–69 years, which corresponds to the mean age of patients in the meta-analysis (2240.5 cancers per 100,000 people per year). Using the relative risk for new cancer occurrence from all trials, 143 people (95%CI 76 to 793) would need to receive treatment with an A2RA for about 4 years (weighted average duration of follow-up) for one excess cancer diagnosis to be made. In the three trials where cancer was a prespecified outcome, the NNH was 105 (95%CI 63 to 271) using the same method.
This study shows an association between A2RAs and cancer but, due to the limitations of the study, does not prove that A2RAs cause cancer. Although the meta-analysis includes data from RCTs, which reduces the possibility of confounding and bias compared with observational studies, the effect of confounding factors such as sex, age and smoking could not be considered because patient level data was not available. In addition, the pooled RCTs were not designed to look at cancer outcomes as the primary outcome measure and the adjudication of cancer diagnoses was not uniform among the included studies. Nevertheless, when analysis was confined to the three studies in which cancer was a prespecifed endpoint and cancer data were collected rigorously, there was a significant increase in the risk of new cancer occurrence with A2RAs. As the authors state, the findings warrant further investigation.
The increased risk of new cancer diagnosis with A2RAs from this meta-analysis was modest but statistically significant. Although the risks to an individual may be low, when used in the population, A2RAs could potentially constitute a significant risk to public health if this increased cancer risk is confirmed.
From the study, it is not possible to draw conclusions about the exact risk of cancer associated with individual A2RAs. In the studies with data on new cancer occurrence, telmisartan was the study drug in 85.7% patients who received A2RAs. However, when analysis was limited to telmisartan▼, the increased risk of new cancer occurrence was of borderline statistical significance (P=0.05). Other A2RAs with new cancer occurrence data included losartan▼ and candesartan, but no data was available for valsartan▼, irbesartan, olmesartan or eprosartan. It is also unclear how the duration of exposure to an A2RA will affect the risk.
In spite of its limitations, this meta-analysis does raise concerns over the safety of A2RAs and regulators will no doubt be considering the data. In the meantime, ACE inhibitors remain the first-line choice when a renin-angiotensin system drug is indicated. They have a more robust evidence base than A2RAs for all indications in terms of evidence for efficacy, safety and most patient factors. A2RAs are an alternative to ACE inhibitors if a renin-angiotensin system drug is indicated but an ACE inhibitor cannot be used because of an intolerable ACE inhibitor-induced cough. The major benefit of A2RAs over ACE inhibitors is their lower rate of cough. However, cough may not be as common with ACE inhibitors as many health professionals perceive. The percentage of people reporting cough with ACE inhibitors in RCTs is about 10% and as low as 2% in observational, real world studies. Discontinuation rates due to cough are lower than this. In the very large ONTARGET study, only 4.2% of patients in the ACE inhibitor group stopped treatment due to cough compared with 1.1% in the A2RA group.
As we discuss in a recent MeReC Bulletin, ACE inhibitors should comprise a higher proportion of renin-angiotensin system drug prescribing than is currently the case. Across primary care in England, only about 70% of renin-angiotensin system drug prescribing is for ACE inhibitors at the current time, with A2RAs accounting for 70% of the total spend on renin-angiotensin system drugs. A review of prescribing in this area could improve quality of care and reduce prescribing costs. However, implementing changes to prescribing in this area is not straightforward, and a simple switch programme is not appropriate.
This study adds to the reasons for reviewing A2RA prescribing on an individual patient basis at the next routine appointment.
Sipahi I, et al. Angiotensin-receptor blockade and risk of cancer: meta-analysis of randomised controlled trials. The Lancet Oncology, Early Online Publication, 14 June 2010 doi:10.1016/S1470-2045(10)70106-6
Design: Meta-analysis of nine RCTs examining the effect of A2RAs on the occurrence of new cancers. Studies were included if an A2RA was given in at least one group, at least 100 patients were enrolled and follow-up was at least 1 year.
Patients and outcomes: For the primary outcome, occurrence of new cancers, five trials were included (n=61,590). Cancer was a prespecified endpoint in three of these trials (n=40,739). For the secondary outcome of assessing occurrence of specific solid organ cancers, five trials were included (n=68,402). For the secondary outcome of assessing cancer-related deaths, eight trials were included (n=93,515).
For the primary outcome of new cancer occurrence, telmisartan was the study drug in 85.7% of patients who received A2RAs. Patients randomly assigned to receive A2RAs had a significantly increased risk of new cancer occurrence compared with patients in control groups (7.2% vs. 6.0%, RR 1.08, 95%CI 1.01 to 1.15, P=0·016). When analysis was limited to trials where cancer was a prespecified endpoint, the RR was 1.11 (95% CI 1.04 to 1.18, P=0.001). Among specific solid organ cancers examined, only new lung-cancer occurrence was significantly higher in patients randomly assigned to receive A2RAs than in those assigned to receive control (0.9% vs. 0.7%, RR 1.25, 1.05 to 1.49, P=0.01). No statistically significant difference in cancer deaths was observed (1.8% vs. 1.6%, RR 1.07, 0.97 to 1.18, P=0.18).
Sponsorship: There was no funding source for this study.
Make sure you are signed up to NPC Email updates — the free email alerting system that keeps you up to date with the NPC news and outputs relevant to you.