Journal reference: 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 (2010) 10.1016/S1470-2045(10)70260-6 (published online, 30 Nov 2010) ^[1]

Link: http://dx.doi.org./10.1016/S1470-2045(10)70260-6

Published: 30 November 2010

Evidence cookie says...

Antihypertensive drugs do not appear to increase cancer risk or cancer-related deaths. This includes angiotensin receptor antagonists/blockers (ARBs). The exception is the combination of ARBs and angiotensin converting enzyme inhibitors (ACEIs); together they are associated with an increased risk of cancer. This finding is of unclear significance and causation should not be assumed. Following a precautionary principle, it is reasonable to avoid the combination of ARBs and ACEIs unless there is a clear clinical indication.

More details:

Article details

Study design:

systematic review, direct and network meta-analyses and trial sequential analyses of randomised trials

Study aim:

assess the association between antihypertensive drugs and cancer risk in a comprehensive analysis of data from randomised clinical trials

Methods summary:

• Literature search of PubMed, Embasde and the Cochrane Central Register of Controlled Trials
• Search terms were for the classes of antihypertensives, names of individual medications, in humans from 1950 to August 2010 (see “Methods”, page 2) ^[1]
• Eligibility criteria:
  • randomised controlled trials
  • comparing the drug classes listed in the search strategy; note, the classes were:
    • angiotensin receptor antagonists (ARBs)
    • angiotensin converting enzyme inhibitors (ACEIs)
    • β-blockers
    • calcium channel blockers (CCBs)
    • diuretics
  • follow up of at least 1 year
  • enrol at least 100 patients
  • able to report the outcomes of interest:
    • cancer
    • cancer-related death
• Trial quality and risk of bias was formally assessed
• Disagreements were resolved by consensus.
• Statistical analysis:
  • intention-to-treat meta-analysis (direct comparison)
    • Heterogeneity was assessed
  • multiple comparison (network) meta-analyses
    • allows for comparisons of drugs not directly addressed within any individual trials
• Trials grouped into 8 comparison categories:
  • placebo
  • ARBs
  • ACEIs
  • β-blockers
  • CCBs
  • diuretics
  • ACEI and ARBs
  • other controls (non-placebo active treatment)

Results summary:

• 70 randomised controlled trials with 148 comparator groups that met inclusion criteria were identified
• 324 168 participants with a mean follow-up of 3.5 years
• Odds ratio for cancer when compared to placebo, multiple comparisons (fixed effects model):
  • ARBs: 1.01, (95% confidence interval [CI] 0.93-1.12)
  • ACEIs: 1.00 (0.92-1.09)
  • β-blockers: 0.97 (0.88-1.07)
  • CCBs: 1.05 (0.96-1.13)
  • diuretics: 1.00 (0.90-1.11)
  • Controls: 0.97 (0.74-1.24)
  • ACEI + ARB: 1.14 (95% CI, 1.02-1.28)
• The combination of an ACEI and ARB was found to have a statistically significant higher relative risk of cancer when compared to placebo, though only on the fixed effects model.  It was not found statistically significant when analysed with the random effects model, with the confidence interval crossing 1.0.
• There were no findings of statistical significance for cancer-related death and antihypertensives.

Study conclusion:

Our analysis refutes a 5·0–10·0% relative increase in the risk of cancer or cancer-related death with the use of ARBs, ACEIs, β-blockers, diuretics, and CCBs. However, increased risk of cancer with the combination of ACEIs and ARBs cannot be ruled out.

Participants:

• 324 168 participants from 70 randomised controlled trials
• Cohort data from individual trials given in the study but in collated form
  • mean age in most studies are between 50-70 years with
  • most common participant cohorts were those with hypertension, coronary artery disease and heart failure

Methodological weaknesses

• Data on drug compliance and patient drop out was not consistent across the studies.
• Multiple comparison analysis assumes homogeneity of the effect (cancer risk) across treatment groups.  It is probable that the actual patient groups may have levels of heterogeneity and this is a threat to internal validity.
• No analysis of dosage of drugs due to limited data.
• Treatment with antihypertensives are known to improve cardiovascular survival; elevated cancer risk might represent a survival effect.
• Relatively short mean follow-up time (3.5 years) reduces the strength of the conclusions; antihypertensives are typically used for much longer periods.
• Analysis by medication groups cannot rule out the possibility of increased cancer risk from an individual drug.

Methodological strengths

• Extensive review of literature and extraction of data.
• Exhaustive and highly detailed statistical analysis of the data.

Biases and conflicts of interests

• Kjeldsen SE (third author) has received ad-hoc consultant/speaker fees from a number of pharmaceutical companies.
• Grossman E (fifth author) has received ad-hoc consultant/speaker fees from a number of pharmaceutical companies.
• Messerli FH (final author) has received ad-hoc consultant/speaker fees from a number of pharmaceutical companies.
• Gupta AK (seventh author) has received travel assistance to attend conferences from Pfizer and Servier.
• Sever PS (eighth author) has received ad-hoc consultant/speaker fees and grant support from Pfizer and Novartis.

Clinical relevance to primary health care

Antihypertensive drugs are widely used by Australian general practitioners.  An increase in cancer risk from these agents, regardless of drug class, has not traditionally been considered a side-effect.  However in mid-2010, a meta-analysis of randomised controlled trials suggested that angiotensin receptor antagonists/blockers (ARBs) increased the risk of new cancers ^[2].

This study refutes an increased relative risk of cancer of 5-10% for all the common classes of antihypertensives ^[1].  In other words, this study had sufficient power to demonstrate that if antihypertensives actually did increase the risk of cancer, its effect can be no greater than 5-10%.  It is probable that the studied classes of antihypertensives do not have any clinically significant effect on cancer risk.

The one exception was the combination of ARBs and ACEIs; the combination was associated with a higher risk of cancer when compared to placebo in one analytic model (but not the other). There was no association for a higher risk of cancer-related death.  Limitations to this study precludes a causative conclusion; the association should rather be considered hypothesis generating.

From the primary health care perspective, it is unlikely that any of the commonly used antihypertensive agents will lead to an increased risk of cancer, including ARBs.  The evidence is unclear for the combination of ARBs and ACEIs.  A reasonable precautionary practice would be to avoid the combination of ARBs and ACEIs unless there is a clear clinical indication for the combination.

References

1. 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 (2010) 10.1016/S1470-2045(10)70260-6 (published online, 30 Nov 2010)
2. Sipahi I, Debanne SM, Rowland DY, et al. Angiotensin-receptor blockade and risk of cancer: meta-analysis of randomised controlled trials. Lancet Oncol 11 (2010), pp. 627–636