Monday June 14 2010
ARBs are used to lower blood pressure
“Blood pressure pills taken by up to one million Britons have been linked to cancer,” warned the Daily Express. It said that a study has found that patients on the drugs, known as angiotensin receptor blockers (ARBs), were slightly more likely to be diagnosed with the disease than those not taking them.
A well-conducted, well-reported systematic review underpins this report. The study found a modest overall increase in the risk of new cancers in groups of people taking ARBs.
People who are taking ARBs should continue to take them and speak to their GP about any concerns they have. Cancers are rare, and they occurred rarely in these studies. The findings from such research illustrate the difficulty in establishing the balance of benefits and harms for some treatments. ARBs are a well-established and proven treatment for high blood pressure and, therefore, prevent deaths related to cardiovascular disease. This study does not prove that they cause cancer, but the evidence from a well-conducted systematic review suggesting an association with increased risk needs further investigation.
Where did the story come from?
The study was carried out by researchers from University Hospitals Case Medical Center and Case Western Reserve University School of Medicine in Cleveland, USA. The study was published in the peer-reviewed medical journal The Lancet Oncology.
This systematic review and meta-analysis of randomised controlled trials is a robust analysis of the available literature on this issue. The researchers have reported the study well and the newspapers have given a balanced report of the findings. Importantly, all those reviewed reported that the size of the increased risk is modest.
What kind of research was this?
This was a systematic review and meta-analysis. The researchers searched for all trials on the efficacy of any of the seven currently available angiotensin receptor blockers (ARBs). They were specifically interested in the effects of ARBs on new cancer occurrence (all types), on specific types of cancer (lung, breast, prostate and ‘other’ cancers) and on deaths from cancer. ARBs are drugs that block the effects of a hormone called angiotensin II in the body, leading to vasodilation (dilation of the blood vessels) and a number of other processes that reduce blood pressure. They are prescribed as a treatment for high blood pressure and heart failure in the UK, although the first choice drug is usually an angiotensin converting enzyme (ACE) inhibitor.
Systematic reviews provide a powerful way to assess the overall effects of a treatment. However, it is important that researchers take care to find all of the relevant research and combine it in statistically appropriate ways. The similarity of the trials to each other also needs to be considered. The original trials are likely to have slightly different methods and these are limitations that have to be taken into account.
What did the research involve?
The researchers searched several medical databases for relevant clinical trials published before November 2009. To be eligible for inclusion, studies had to be randomised controlled trials in at least 100 people. Patients had to have been given ARBs in at least one of the treatment groups. As cancer is a rare adverse event that takes a long time to develop (long latent period) studies also had to have at least a year of follow-up. Studies meeting these criteria were then assessed for whether they reported cancer as an adverse outcome. The Food and Drug Administration (FDA) website (a US government agency for the manufacture and testing of drugs) was also searched for additional unpublished information from the trials, and for any other relevant unpublished trials.
Five RCTs met the inclusion criteria and were included in the analyses for the occurrence of new cancers, they included a total of 61,590 participants. For the secondary objectives assessing the occurrence of specific cancer types and cancer-related deaths, five studies (68,402 participants) and eight studies (93,515 participants) were included, respectively. The researchers were also able to assess the overall cancer risk and the risk of specific cancer types associated with ARB treatment on its own and in combination with ACE inhibitors.
A statistical method called meta-analysis was used to combine the results from the different trials.
This is a well-conducted review and analysis in which the researchers had clear inclusion criteria. They considered the quality of studies and performed some subgroup and sensitivity analyses.
What were the basic results?
Overall, there was a slightly increased risk of cancer developing in people taking ARBs (with or without ACE inhibitors) compared with people not taking ARBs. During follow-up, a new cancer was diagnosed in 7.2% of people in the ARBs groups, compared with 6% in the non-ARB groups. This represents an increased risk of 8% (RR 1.08, 95% CI 1.01 to 1.15).
A similar increase was seen when the analysis was limited to three RCTs that had pre-specified that they would measure cancer as an adverse event (RR 1.11, 95% CI 1.04 to 1.18).
The various subgroup analyses also showed a similar degree of increase in risk, for example when analysis was limited to the most common ARB (telmisartan), either combined or not combined with ACE inhibitors, and where analyses were limited only to those who were cancer-free at the beginning of the study (this information was available in only two studies).
For specific cancers, there was an overall increase in the risk of lung cancer (0.9% vs 0.7%), though this cancer was very rare in the sample. This was only significant for those also receiving ACE inhibitor therapy in combination with ARBs. There was no significant increase in the risk of prostate cancer, breast or ‘other’ cancers. There was also no difference between the ARB and non-ARB groups in risk of cancer-related deaths.
How did the researchers interpret the results?
The researchers say that this systematic review and meta-analysis suggests that “ARBs are associated with a modestly increased risk of new cancer diagnosis”.
However, they note that, because of the limited data available, it isn’t possible to make conclusions about the exact risk of cancer associated with each of the seven currently available ARBs. They call for further investigation.
This is a well-conducted review with meta-analysis. The researchers did what they could to supplement published data with those given to regulatory bodies (the FDA). They found a modest but significant increased risk in new cancer occurrence. There are several important points to raise in relation to this study and its conclusions, many of which the researchers themselves highlight:
- Cancers were rare in these groups. Although there was an alarming sounding 8% increase in risk, the absolute risk difference was actually only 1.2%. Put another way, 2,510 people out of the 35,015 treated with ARBs got cancer compared with 1,602 of the 26,575 not receiving ARBs.
- The studies pooled in the analyses were not designed to look at the outcome of cancer, and the diagnosis of cancer would have been different between the studies. The researchers note this as a potential limitation, but say that cancer risk was still higher when the analysis was limited to the three studies that had pre-specified that cancer would be an outcome (and would, therefore, have collected the data and diagnoses more rigorously).
- Ideally, meta-analyses use individual patient level data from the studies that can be pooled. However, in this review, such level of detail was not available and the researchers could only pool study-level results. Their study may, therefore, have had reduced power and was not able to investigate subtle effects on cancer, such as time to event.
The findings from such research illustrate the difficulty in establishing the balance of benefits and harms for some treatments. ARBs are a well-established and proven treatment for high blood pressure and, therefore, prevent deaths related to cardiovascular disease. This study does not prove that they cause cancer, but the evidence from a well-conducted systematic review suggesting an association with increased risk needs further investigation.