User login
The debate on whether the popular class of antihypertensive drugs, angiotensin receptor blockers (ARBs), may be associated with an increased risk for cancer has been reopened with the publication of a new meta-analysis.
The analysis found an increasing risk for cancer, and specifically lung cancer, with increasing cumulative exposure to these drugs.
The findings are reported in a study published online in PLOS ONE.
The author of this new meta-analysis is Ilke Sipahi, MD, a cardiologist from Acibadem University Medical School, Istanbul, who previously raised this issue in an initial meta-analysis published in 2010.
“The new meta-analysis is important because it is the first study to investigate whether there is a dose response in the association between ARBs and cancer,” Dr. Sipahi told this news organization.
“I found a clear signal of increased risk of cancer as exposure to ARBs increased, and the association started to become significant when the maximum dose was taken for 3 years,” he added.
Dr. Sipahi explained that in the first meta-analysis published in Lancet Oncology, he and his colleagues reported an increased cancer risk with ARBs based on observations from high-exposure trials – those that included higher doses of ARBs with a long duration of follow-up.
Following this publication, an investigation by the U.S. Food and Drug Administration refuted the risk, and a collaboration of ARB trial investigators also performed an analysis published in the Journal of Hypertension (2011. doi: 10.1097/HJH.0b013e328344a7de), which again did not show an increased risk for cancer with use of ARBs.
Dr. Sipahi claims that those analyses by the FDA and the ARB Trialists Collaboration, which were all trial-level meta-analyses, diluted the “high exposure” data (including higher doses taken for longer periods of time) with a large amount of other data on much lower exposures (lower doses and/or shorter time periods).
“The overall risk would then inevitably become nonsignificant. These analyses also did not look at different exposure levels,” he says.
“For cancer, the degree of exposure is obviously very important. The risk associated with smoking 2 or 3 cigarettes a day for a year is very different from that of smoking 2 packs a day for 40 years. The same principle applies to taking a medication,” Dr. Sipahi asserts.
From these latest data, he estimates that 120 patients needed to be treated with the maximal daily dose of an ARB for 4.7 years for one excess cancer diagnosis, and 464 patients needed to be treated for one excess lung cancer.
“Given that at least 200 million individuals are being treated with an ARB globally, approximately 1.7 million excess cancers (and 430,000 lung cancers) in 4.6 years could be potentially caused by this class of drugs,” he suggests.
For the current analysis, Dr. Sipahi used trial-level data taken from the paper by the ARB Trialists Collaboration and investigated the effect of exposure to ARBs – including both the dose taken and the length of treatment – on risk for cancer. He performed metaregression analyses that he says has not been done before.
“I mathematically quantitated the degree of exposure in each trial. And when the degree of exposure was correlated with risk of cancer, there was a significant association.”
The new meta-analysis includes 15 randomized controlled trials. The two coprimary outcomes were the relationship between cumulative exposure to ARBs and risk for all cancers combined and the relationship between cumulative exposure and risk for lung cancer.
In the trials, 74,021 patients were randomly assigned to an ARB, resulting in a total cumulative exposure of 172,389 person-years of exposure to daily high dose (or equivalent), and 61,197 patients were randomly assigned to control.
Results showed a highly significant correlation between the degree of cumulative exposure to ARBs and risk for all cancers combined (slope = 0.07; 95% confidence interval, 0.03-0.11; P < .001) and also lung cancer (slope = 0.16; 95% CI, 0.05-0.27; P = .003).
In trials where the cumulative exposure was greater than 3 years of exposure to daily high dose, there was a statistically significant increase in risk for all cancers combined (risk ratio, 1.11; 95% CI, 1.03-1.19; P = .006).
There was also a statistically significant increase in risk for lung cancers in trials where the cumulative exposure was greater than 2.5 years (RR, 1.21; 95% CI, 1.02-1.44; P = .03).
In trials with lower cumulative exposure to ARBs, there was no increased risk either for all cancers combined or lung cancer.
Dr. Sipahi reports that the cumulative exposure-risk relationship with ARBs was independent of background angiotensin-converting enzyme (ACE) inhibitor treatment or the type of control (placebo or nonplacebo control).
But he acknowledges that since this is a trial-level analysis, the effects of patient characteristics such as age and smoking status could not be examined because of lack of patient-level data.
Dr. Sipahi says he does not know the mechanism behind these findings, but he draws attention to the recent withdrawal of several thousand lots of ARB formulations because of the presence of potentially carcinogenic impurities that have been suggested to be a byproduct of ARB synthesis.
He also claims that unlike some other classes of antihypertensives, ARBs have not been shown to reduce the risk for MI, leading him to conclude that “other classes of antihypertensives with good safety and efficacy data (such as ACE-inhibitors, calcium-channel blockers or others) should become the preferred first-line agents in the treatment of hypertension.”
Dr. Sipahi wants the FDA to reinvestigate the issue of ARBs and cancer risk using individual patient data. “They already have the patient-level data from the trials. They should look at it more carefully and look at exposure levels and how they relate to cancer risk,” he said. “And the fact that there have been studies linking high ARB exposure levels to increased cancer risk should at least get a warning on the drug labels.”
A ‘clear increase’ in risk
Dr. Sipahi also points out that a link between ARBs and cancer has been found in another meta-analysis performed in 2013 by senior FDA analyst Thomas Marciniak, MD.
“Because he worked at the FDA, [Dr.] Marciniak had access to individual patent data. This is the best type of analysis and generally produces more accurate results than a trial-level meta-analysis,” Dr. Sipahi commented.
Dr. Marciniak’s analysis, which is available on the FDA website as part of another document, was not officially published elsewhere, and no further action has been taken on the issue.
Contacted by this news organization, Dr. Marciniak, who has now retired from the FDA, said he not only conducted a patient-level meta-analysis but also followed up adverse effects reported in the trials that could have been a symptom of cancer to establish further whether the patient was later diagnosed with cancer or not.
“I used every scrap of information sent in, including serious adverse event reports. I saw a clear increase in lung cancer risk with the ARBs,” Dr. Marciniak said. He did not, however, perform a dose-response relationship analysis.
Asked why his analysis and those from Dr. Sipahi reach different conclusions to those from the ARB Trialists Collaboration and the official FDA investigations, Dr. Marciniak said: “It may be that there were too many low-exposure trials that just washed out the difference. But trial data generally do not capture adverse events such as cancer, which takes a long time to develop, very well, and if you’re not really looking for it, you’re probably not going to find it.”
Dr. Marciniak said that Dr. Sipahi’s current findings are in line with his results. “Finding a dose response, to me, is extremely compelling, and I think the signal here is real,” he commented. “I think this new paper from Dr. Sipahi verifies what I found. I think the FDA should now release all individual patient data it has.”
Contacted for comment, an FDA spokesperson said, “Generally the FDA does not comment on specific studies but evaluates them as part of the body of evidence to further our understanding about a particular issue and assist in our mission to protect public health.”
They added: “The FDA has ongoing assessment, surveillance, compliance, and pharmaceutical quality efforts across every product area, and we will continue to work with drug manufacturers to ensure safe, effective, and high-quality drugs for the American public. When we identify new and previously unrecognized risks to safety and quality, we react swiftly to resolve the problem, as we have done in responding to the recent findings of nitrosamines in certain medicines.”
Analysis ‘should be taken seriously’
Commenting on this new study, Steve Nissen, MD, a key figure in analyzing such complex data and who has himself uncovered problems with high-profile drugs in the past, says the current analysis should be taken seriously.
Dr. Nissen, who was Dr. Sipahi’s senior during his post-doc position at the Cleveland Clinic, wrote an editorial accompanying Dr. Sipahi’s first paper and calling for urgent regulatory review of the evidence.
He says the new findings add to previous evidence suggesting a possible risk for cancer with ARBs.
“[Dr.] Sipahi is a capable researcher, and this analysis needs to be taken seriously, but it needs to be verified. It is not possible to draw a strong conclusion on this analysis, as it is not based on individual patient data, but I don’t think it should be ignored,” Dr. Nissen stated.
“I will say again what I said 12 years ago – that the regulatory agencies need to carefully review all their data in a very detailed way. The FDA and EMA have access to the individual patient data and are both very capable of doing the required analyses.”
Limitations of trial-level analysis
Asked to evaluate the statistics in the current paper, Andrew Althouse, PhD, an assistant professor of medicine at the University of Pittsburgh, and a clinical trial statistician, explained that the best way to do a thorough analysis of the relationship between ARB exposure and risk for incident cancer would involve the use of patient-level data.
“As such data were not available to Dr. Sipahi, I believe he is doing as well as he can. But without full access to individual patient-level data from the respective trials, it is difficult to support any firm conclusions,” Dr. Althouse said in an interview.
He suggested that the meta-regression analyses used in the paper were unable to properly estimate the relationship between ARB exposure and risk for incident cancer.
“Taken at face value, the current analysis suggests that [in] trials with longer follow-up duration (and therefore greater cumulative exposure to ARB for the treatment group), the risk of developing cancer for patients in the ARB group versus the non-ARB group was progressively higher. But this study doesn’t take into account the actual amount of follow-up time for individual patients or potential differences in the amount of follow-up time between the two groups in each trial,” he noted.
Dr. Althouse says this raises the possibility of “competing risks” or the idea that if ARBs reduce cardiovascular disease and cardiovascular death, then there would be more patients remaining in that arm who could go on to develop cancer. “So a crude count of the number of cancer cases may look as though patients receiving ARBs are ‘more likely’ to develop cancer, but this is a mirage.”
He added: “When there are some patients dying during the study, the only way to tell whether the intervention actually increased the risk of other health-related complications is to have an analysis that properly accounts for each patient’s time-at-risk of the outcome. Unfortunately, properly analyzing this requires the use of patient-level data.”
Cardiologists skeptical?
Cardiology experts asked for thoughts on the new meta-analysis were also cautious to read too much into the findings.
Franz Messerli, MD, professor of medicine at the University of Bern, Switzerland, commented: “Perhaps one would simply ignore this rambling, cherrypicking-based condemnation of ARBs if it were not for the powerful negative connotation of the word cancer. Thus, the meta-analysis of Dr. Sipahi purporting that ARBs could be increasing the development of cancers in a cumulative way is of concern to both physicians and patients.”
But, raising a similar point to Dr. Althouse about competing risks, Dr. Messerli said: “We have to consider that as one gets older, the cardiovascular disease state and cancer state will compete with each other for the outcome of death. The better that therapies protect against cardiovascular death, the more they will increase life expectancy and thus the risk of cancer.”
He also added that “in head-to-head comparisons with ACE inhibitors, ARBs showed similar efficacy in terms of death, CV mortality, MI, stroke, and end-stage kidney disease, so can we agree that the attempt of Dr. Sipahi to disparage ARBs as a class is much ado about nothing?”
Dr. Nissen, however, said he views the idea of competing risk as “a bit of a stretch” in this case. “Although ARBs are effective antihypertensive drugs, I would say there is very little evidence that they would prolong survival versus other antihypertensives.”
Dr. Sipahi also claims that this argument is not relevant to the current analysis. “ARBs did not increase survival in any of the high-exposure trials that showed an excess in cancers. Therefore, competing outcomes, or ‘survival bias’ to be more specific, is not a possibility here,” he says.
George Bakris, MD, professor of medicine at the University of Chicago Medicine, noted that while the current study shows a slight increase in cancer incidence, especially lung cancer, among those taking ARBs for more than 3 years, it “totally ignores the overwhelming cardiovascular risk reduction seen in the trials.”
“Moreover,” he adds, “the author notes that the findings were independent of ACE-inhibitors, but he can’t rule out smoking and age as factors, two major risk factors for cancer and lung cancer, specifically. Thus, as typical of these types of analyses, the associations are probably true/true unrelated or, at best, partially related.”
Dr. Bakris referred to the potentially carcinogenic nitrosamine and azido compounds found in several ARB formulations that have resulted in recalls.
“At any stage of drug synthesis throughout each product’s lifetime, these impurities may evolve if an amine reacts with a nitrosating agent coexisting under appropriate conditions,” he said. “Drug regulatory authorities worldwide have established stringent guidelines on nitrosamine contamination for all drug products. The studies noted in the author’s analysis were done well before these guidelines were implemented. Hence, many of the issues raised by the authors using trials from 10-20 years ago are not of significant concern.”
Still, the cardiology experts all agreed on one thing – that patients should continue to take ARBs as prescribed.
Noting that worldwide authorities are now addressing the issue of possible carcinogen contamination, Dr. Bakris stressed that patients “should not panic and should not stop their meds.”
Dr. Nissen added: “What we don’t want is for patents who are taking ARBs to stop taking these medications – hypertension is a deadly disorder, and these drugs have proven cardiovascular benefits.”
Dr. Sipahi received no specific funding for this work. He reports receiving lecture honoraria from Novartis, Boehringer Ingelheim, Sanofi, Sandoz, Bristol-Myers Squibb, Bayer, Pfizer, Ranbaxy, Servier, and ARIS and served on advisory boards for Novartis, Sanofi, Servier, Bristol-Myers Squibb, Pfizer, Bayer and I.E. Ulagay. The other commenters do not report any relevant disclosures.
A version of this article first appeared on Medscape.com.
The debate on whether the popular class of antihypertensive drugs, angiotensin receptor blockers (ARBs), may be associated with an increased risk for cancer has been reopened with the publication of a new meta-analysis.
The analysis found an increasing risk for cancer, and specifically lung cancer, with increasing cumulative exposure to these drugs.
The findings are reported in a study published online in PLOS ONE.
The author of this new meta-analysis is Ilke Sipahi, MD, a cardiologist from Acibadem University Medical School, Istanbul, who previously raised this issue in an initial meta-analysis published in 2010.
“The new meta-analysis is important because it is the first study to investigate whether there is a dose response in the association between ARBs and cancer,” Dr. Sipahi told this news organization.
“I found a clear signal of increased risk of cancer as exposure to ARBs increased, and the association started to become significant when the maximum dose was taken for 3 years,” he added.
Dr. Sipahi explained that in the first meta-analysis published in Lancet Oncology, he and his colleagues reported an increased cancer risk with ARBs based on observations from high-exposure trials – those that included higher doses of ARBs with a long duration of follow-up.
Following this publication, an investigation by the U.S. Food and Drug Administration refuted the risk, and a collaboration of ARB trial investigators also performed an analysis published in the Journal of Hypertension (2011. doi: 10.1097/HJH.0b013e328344a7de), which again did not show an increased risk for cancer with use of ARBs.
Dr. Sipahi claims that those analyses by the FDA and the ARB Trialists Collaboration, which were all trial-level meta-analyses, diluted the “high exposure” data (including higher doses taken for longer periods of time) with a large amount of other data on much lower exposures (lower doses and/or shorter time periods).
“The overall risk would then inevitably become nonsignificant. These analyses also did not look at different exposure levels,” he says.
“For cancer, the degree of exposure is obviously very important. The risk associated with smoking 2 or 3 cigarettes a day for a year is very different from that of smoking 2 packs a day for 40 years. The same principle applies to taking a medication,” Dr. Sipahi asserts.
From these latest data, he estimates that 120 patients needed to be treated with the maximal daily dose of an ARB for 4.7 years for one excess cancer diagnosis, and 464 patients needed to be treated for one excess lung cancer.
“Given that at least 200 million individuals are being treated with an ARB globally, approximately 1.7 million excess cancers (and 430,000 lung cancers) in 4.6 years could be potentially caused by this class of drugs,” he suggests.
For the current analysis, Dr. Sipahi used trial-level data taken from the paper by the ARB Trialists Collaboration and investigated the effect of exposure to ARBs – including both the dose taken and the length of treatment – on risk for cancer. He performed metaregression analyses that he says has not been done before.
“I mathematically quantitated the degree of exposure in each trial. And when the degree of exposure was correlated with risk of cancer, there was a significant association.”
The new meta-analysis includes 15 randomized controlled trials. The two coprimary outcomes were the relationship between cumulative exposure to ARBs and risk for all cancers combined and the relationship between cumulative exposure and risk for lung cancer.
In the trials, 74,021 patients were randomly assigned to an ARB, resulting in a total cumulative exposure of 172,389 person-years of exposure to daily high dose (or equivalent), and 61,197 patients were randomly assigned to control.
Results showed a highly significant correlation between the degree of cumulative exposure to ARBs and risk for all cancers combined (slope = 0.07; 95% confidence interval, 0.03-0.11; P < .001) and also lung cancer (slope = 0.16; 95% CI, 0.05-0.27; P = .003).
In trials where the cumulative exposure was greater than 3 years of exposure to daily high dose, there was a statistically significant increase in risk for all cancers combined (risk ratio, 1.11; 95% CI, 1.03-1.19; P = .006).
There was also a statistically significant increase in risk for lung cancers in trials where the cumulative exposure was greater than 2.5 years (RR, 1.21; 95% CI, 1.02-1.44; P = .03).
In trials with lower cumulative exposure to ARBs, there was no increased risk either for all cancers combined or lung cancer.
Dr. Sipahi reports that the cumulative exposure-risk relationship with ARBs was independent of background angiotensin-converting enzyme (ACE) inhibitor treatment or the type of control (placebo or nonplacebo control).
But he acknowledges that since this is a trial-level analysis, the effects of patient characteristics such as age and smoking status could not be examined because of lack of patient-level data.
Dr. Sipahi says he does not know the mechanism behind these findings, but he draws attention to the recent withdrawal of several thousand lots of ARB formulations because of the presence of potentially carcinogenic impurities that have been suggested to be a byproduct of ARB synthesis.
He also claims that unlike some other classes of antihypertensives, ARBs have not been shown to reduce the risk for MI, leading him to conclude that “other classes of antihypertensives with good safety and efficacy data (such as ACE-inhibitors, calcium-channel blockers or others) should become the preferred first-line agents in the treatment of hypertension.”
Dr. Sipahi wants the FDA to reinvestigate the issue of ARBs and cancer risk using individual patient data. “They already have the patient-level data from the trials. They should look at it more carefully and look at exposure levels and how they relate to cancer risk,” he said. “And the fact that there have been studies linking high ARB exposure levels to increased cancer risk should at least get a warning on the drug labels.”
A ‘clear increase’ in risk
Dr. Sipahi also points out that a link between ARBs and cancer has been found in another meta-analysis performed in 2013 by senior FDA analyst Thomas Marciniak, MD.
“Because he worked at the FDA, [Dr.] Marciniak had access to individual patent data. This is the best type of analysis and generally produces more accurate results than a trial-level meta-analysis,” Dr. Sipahi commented.
Dr. Marciniak’s analysis, which is available on the FDA website as part of another document, was not officially published elsewhere, and no further action has been taken on the issue.
Contacted by this news organization, Dr. Marciniak, who has now retired from the FDA, said he not only conducted a patient-level meta-analysis but also followed up adverse effects reported in the trials that could have been a symptom of cancer to establish further whether the patient was later diagnosed with cancer or not.
“I used every scrap of information sent in, including serious adverse event reports. I saw a clear increase in lung cancer risk with the ARBs,” Dr. Marciniak said. He did not, however, perform a dose-response relationship analysis.
Asked why his analysis and those from Dr. Sipahi reach different conclusions to those from the ARB Trialists Collaboration and the official FDA investigations, Dr. Marciniak said: “It may be that there were too many low-exposure trials that just washed out the difference. But trial data generally do not capture adverse events such as cancer, which takes a long time to develop, very well, and if you’re not really looking for it, you’re probably not going to find it.”
Dr. Marciniak said that Dr. Sipahi’s current findings are in line with his results. “Finding a dose response, to me, is extremely compelling, and I think the signal here is real,” he commented. “I think this new paper from Dr. Sipahi verifies what I found. I think the FDA should now release all individual patient data it has.”
Contacted for comment, an FDA spokesperson said, “Generally the FDA does not comment on specific studies but evaluates them as part of the body of evidence to further our understanding about a particular issue and assist in our mission to protect public health.”
They added: “The FDA has ongoing assessment, surveillance, compliance, and pharmaceutical quality efforts across every product area, and we will continue to work with drug manufacturers to ensure safe, effective, and high-quality drugs for the American public. When we identify new and previously unrecognized risks to safety and quality, we react swiftly to resolve the problem, as we have done in responding to the recent findings of nitrosamines in certain medicines.”
Analysis ‘should be taken seriously’
Commenting on this new study, Steve Nissen, MD, a key figure in analyzing such complex data and who has himself uncovered problems with high-profile drugs in the past, says the current analysis should be taken seriously.
Dr. Nissen, who was Dr. Sipahi’s senior during his post-doc position at the Cleveland Clinic, wrote an editorial accompanying Dr. Sipahi’s first paper and calling for urgent regulatory review of the evidence.
He says the new findings add to previous evidence suggesting a possible risk for cancer with ARBs.
“[Dr.] Sipahi is a capable researcher, and this analysis needs to be taken seriously, but it needs to be verified. It is not possible to draw a strong conclusion on this analysis, as it is not based on individual patient data, but I don’t think it should be ignored,” Dr. Nissen stated.
“I will say again what I said 12 years ago – that the regulatory agencies need to carefully review all their data in a very detailed way. The FDA and EMA have access to the individual patient data and are both very capable of doing the required analyses.”
Limitations of trial-level analysis
Asked to evaluate the statistics in the current paper, Andrew Althouse, PhD, an assistant professor of medicine at the University of Pittsburgh, and a clinical trial statistician, explained that the best way to do a thorough analysis of the relationship between ARB exposure and risk for incident cancer would involve the use of patient-level data.
“As such data were not available to Dr. Sipahi, I believe he is doing as well as he can. But without full access to individual patient-level data from the respective trials, it is difficult to support any firm conclusions,” Dr. Althouse said in an interview.
He suggested that the meta-regression analyses used in the paper were unable to properly estimate the relationship between ARB exposure and risk for incident cancer.
“Taken at face value, the current analysis suggests that [in] trials with longer follow-up duration (and therefore greater cumulative exposure to ARB for the treatment group), the risk of developing cancer for patients in the ARB group versus the non-ARB group was progressively higher. But this study doesn’t take into account the actual amount of follow-up time for individual patients or potential differences in the amount of follow-up time between the two groups in each trial,” he noted.
Dr. Althouse says this raises the possibility of “competing risks” or the idea that if ARBs reduce cardiovascular disease and cardiovascular death, then there would be more patients remaining in that arm who could go on to develop cancer. “So a crude count of the number of cancer cases may look as though patients receiving ARBs are ‘more likely’ to develop cancer, but this is a mirage.”
He added: “When there are some patients dying during the study, the only way to tell whether the intervention actually increased the risk of other health-related complications is to have an analysis that properly accounts for each patient’s time-at-risk of the outcome. Unfortunately, properly analyzing this requires the use of patient-level data.”
Cardiologists skeptical?
Cardiology experts asked for thoughts on the new meta-analysis were also cautious to read too much into the findings.
Franz Messerli, MD, professor of medicine at the University of Bern, Switzerland, commented: “Perhaps one would simply ignore this rambling, cherrypicking-based condemnation of ARBs if it were not for the powerful negative connotation of the word cancer. Thus, the meta-analysis of Dr. Sipahi purporting that ARBs could be increasing the development of cancers in a cumulative way is of concern to both physicians and patients.”
But, raising a similar point to Dr. Althouse about competing risks, Dr. Messerli said: “We have to consider that as one gets older, the cardiovascular disease state and cancer state will compete with each other for the outcome of death. The better that therapies protect against cardiovascular death, the more they will increase life expectancy and thus the risk of cancer.”
He also added that “in head-to-head comparisons with ACE inhibitors, ARBs showed similar efficacy in terms of death, CV mortality, MI, stroke, and end-stage kidney disease, so can we agree that the attempt of Dr. Sipahi to disparage ARBs as a class is much ado about nothing?”
Dr. Nissen, however, said he views the idea of competing risk as “a bit of a stretch” in this case. “Although ARBs are effective antihypertensive drugs, I would say there is very little evidence that they would prolong survival versus other antihypertensives.”
Dr. Sipahi also claims that this argument is not relevant to the current analysis. “ARBs did not increase survival in any of the high-exposure trials that showed an excess in cancers. Therefore, competing outcomes, or ‘survival bias’ to be more specific, is not a possibility here,” he says.
George Bakris, MD, professor of medicine at the University of Chicago Medicine, noted that while the current study shows a slight increase in cancer incidence, especially lung cancer, among those taking ARBs for more than 3 years, it “totally ignores the overwhelming cardiovascular risk reduction seen in the trials.”
“Moreover,” he adds, “the author notes that the findings were independent of ACE-inhibitors, but he can’t rule out smoking and age as factors, two major risk factors for cancer and lung cancer, specifically. Thus, as typical of these types of analyses, the associations are probably true/true unrelated or, at best, partially related.”
Dr. Bakris referred to the potentially carcinogenic nitrosamine and azido compounds found in several ARB formulations that have resulted in recalls.
“At any stage of drug synthesis throughout each product’s lifetime, these impurities may evolve if an amine reacts with a nitrosating agent coexisting under appropriate conditions,” he said. “Drug regulatory authorities worldwide have established stringent guidelines on nitrosamine contamination for all drug products. The studies noted in the author’s analysis were done well before these guidelines were implemented. Hence, many of the issues raised by the authors using trials from 10-20 years ago are not of significant concern.”
Still, the cardiology experts all agreed on one thing – that patients should continue to take ARBs as prescribed.
Noting that worldwide authorities are now addressing the issue of possible carcinogen contamination, Dr. Bakris stressed that patients “should not panic and should not stop their meds.”
Dr. Nissen added: “What we don’t want is for patents who are taking ARBs to stop taking these medications – hypertension is a deadly disorder, and these drugs have proven cardiovascular benefits.”
Dr. Sipahi received no specific funding for this work. He reports receiving lecture honoraria from Novartis, Boehringer Ingelheim, Sanofi, Sandoz, Bristol-Myers Squibb, Bayer, Pfizer, Ranbaxy, Servier, and ARIS and served on advisory boards for Novartis, Sanofi, Servier, Bristol-Myers Squibb, Pfizer, Bayer and I.E. Ulagay. The other commenters do not report any relevant disclosures.
A version of this article first appeared on Medscape.com.
The debate on whether the popular class of antihypertensive drugs, angiotensin receptor blockers (ARBs), may be associated with an increased risk for cancer has been reopened with the publication of a new meta-analysis.
The analysis found an increasing risk for cancer, and specifically lung cancer, with increasing cumulative exposure to these drugs.
The findings are reported in a study published online in PLOS ONE.
The author of this new meta-analysis is Ilke Sipahi, MD, a cardiologist from Acibadem University Medical School, Istanbul, who previously raised this issue in an initial meta-analysis published in 2010.
“The new meta-analysis is important because it is the first study to investigate whether there is a dose response in the association between ARBs and cancer,” Dr. Sipahi told this news organization.
“I found a clear signal of increased risk of cancer as exposure to ARBs increased, and the association started to become significant when the maximum dose was taken for 3 years,” he added.
Dr. Sipahi explained that in the first meta-analysis published in Lancet Oncology, he and his colleagues reported an increased cancer risk with ARBs based on observations from high-exposure trials – those that included higher doses of ARBs with a long duration of follow-up.
Following this publication, an investigation by the U.S. Food and Drug Administration refuted the risk, and a collaboration of ARB trial investigators also performed an analysis published in the Journal of Hypertension (2011. doi: 10.1097/HJH.0b013e328344a7de), which again did not show an increased risk for cancer with use of ARBs.
Dr. Sipahi claims that those analyses by the FDA and the ARB Trialists Collaboration, which were all trial-level meta-analyses, diluted the “high exposure” data (including higher doses taken for longer periods of time) with a large amount of other data on much lower exposures (lower doses and/or shorter time periods).
“The overall risk would then inevitably become nonsignificant. These analyses also did not look at different exposure levels,” he says.
“For cancer, the degree of exposure is obviously very important. The risk associated with smoking 2 or 3 cigarettes a day for a year is very different from that of smoking 2 packs a day for 40 years. The same principle applies to taking a medication,” Dr. Sipahi asserts.
From these latest data, he estimates that 120 patients needed to be treated with the maximal daily dose of an ARB for 4.7 years for one excess cancer diagnosis, and 464 patients needed to be treated for one excess lung cancer.
“Given that at least 200 million individuals are being treated with an ARB globally, approximately 1.7 million excess cancers (and 430,000 lung cancers) in 4.6 years could be potentially caused by this class of drugs,” he suggests.
For the current analysis, Dr. Sipahi used trial-level data taken from the paper by the ARB Trialists Collaboration and investigated the effect of exposure to ARBs – including both the dose taken and the length of treatment – on risk for cancer. He performed metaregression analyses that he says has not been done before.
“I mathematically quantitated the degree of exposure in each trial. And when the degree of exposure was correlated with risk of cancer, there was a significant association.”
The new meta-analysis includes 15 randomized controlled trials. The two coprimary outcomes were the relationship between cumulative exposure to ARBs and risk for all cancers combined and the relationship between cumulative exposure and risk for lung cancer.
In the trials, 74,021 patients were randomly assigned to an ARB, resulting in a total cumulative exposure of 172,389 person-years of exposure to daily high dose (or equivalent), and 61,197 patients were randomly assigned to control.
Results showed a highly significant correlation between the degree of cumulative exposure to ARBs and risk for all cancers combined (slope = 0.07; 95% confidence interval, 0.03-0.11; P < .001) and also lung cancer (slope = 0.16; 95% CI, 0.05-0.27; P = .003).
In trials where the cumulative exposure was greater than 3 years of exposure to daily high dose, there was a statistically significant increase in risk for all cancers combined (risk ratio, 1.11; 95% CI, 1.03-1.19; P = .006).
There was also a statistically significant increase in risk for lung cancers in trials where the cumulative exposure was greater than 2.5 years (RR, 1.21; 95% CI, 1.02-1.44; P = .03).
In trials with lower cumulative exposure to ARBs, there was no increased risk either for all cancers combined or lung cancer.
Dr. Sipahi reports that the cumulative exposure-risk relationship with ARBs was independent of background angiotensin-converting enzyme (ACE) inhibitor treatment or the type of control (placebo or nonplacebo control).
But he acknowledges that since this is a trial-level analysis, the effects of patient characteristics such as age and smoking status could not be examined because of lack of patient-level data.
Dr. Sipahi says he does not know the mechanism behind these findings, but he draws attention to the recent withdrawal of several thousand lots of ARB formulations because of the presence of potentially carcinogenic impurities that have been suggested to be a byproduct of ARB synthesis.
He also claims that unlike some other classes of antihypertensives, ARBs have not been shown to reduce the risk for MI, leading him to conclude that “other classes of antihypertensives with good safety and efficacy data (such as ACE-inhibitors, calcium-channel blockers or others) should become the preferred first-line agents in the treatment of hypertension.”
Dr. Sipahi wants the FDA to reinvestigate the issue of ARBs and cancer risk using individual patient data. “They already have the patient-level data from the trials. They should look at it more carefully and look at exposure levels and how they relate to cancer risk,” he said. “And the fact that there have been studies linking high ARB exposure levels to increased cancer risk should at least get a warning on the drug labels.”
A ‘clear increase’ in risk
Dr. Sipahi also points out that a link between ARBs and cancer has been found in another meta-analysis performed in 2013 by senior FDA analyst Thomas Marciniak, MD.
“Because he worked at the FDA, [Dr.] Marciniak had access to individual patent data. This is the best type of analysis and generally produces more accurate results than a trial-level meta-analysis,” Dr. Sipahi commented.
Dr. Marciniak’s analysis, which is available on the FDA website as part of another document, was not officially published elsewhere, and no further action has been taken on the issue.
Contacted by this news organization, Dr. Marciniak, who has now retired from the FDA, said he not only conducted a patient-level meta-analysis but also followed up adverse effects reported in the trials that could have been a symptom of cancer to establish further whether the patient was later diagnosed with cancer or not.
“I used every scrap of information sent in, including serious adverse event reports. I saw a clear increase in lung cancer risk with the ARBs,” Dr. Marciniak said. He did not, however, perform a dose-response relationship analysis.
Asked why his analysis and those from Dr. Sipahi reach different conclusions to those from the ARB Trialists Collaboration and the official FDA investigations, Dr. Marciniak said: “It may be that there were too many low-exposure trials that just washed out the difference. But trial data generally do not capture adverse events such as cancer, which takes a long time to develop, very well, and if you’re not really looking for it, you’re probably not going to find it.”
Dr. Marciniak said that Dr. Sipahi’s current findings are in line with his results. “Finding a dose response, to me, is extremely compelling, and I think the signal here is real,” he commented. “I think this new paper from Dr. Sipahi verifies what I found. I think the FDA should now release all individual patient data it has.”
Contacted for comment, an FDA spokesperson said, “Generally the FDA does not comment on specific studies but evaluates them as part of the body of evidence to further our understanding about a particular issue and assist in our mission to protect public health.”
They added: “The FDA has ongoing assessment, surveillance, compliance, and pharmaceutical quality efforts across every product area, and we will continue to work with drug manufacturers to ensure safe, effective, and high-quality drugs for the American public. When we identify new and previously unrecognized risks to safety and quality, we react swiftly to resolve the problem, as we have done in responding to the recent findings of nitrosamines in certain medicines.”
Analysis ‘should be taken seriously’
Commenting on this new study, Steve Nissen, MD, a key figure in analyzing such complex data and who has himself uncovered problems with high-profile drugs in the past, says the current analysis should be taken seriously.
Dr. Nissen, who was Dr. Sipahi’s senior during his post-doc position at the Cleveland Clinic, wrote an editorial accompanying Dr. Sipahi’s first paper and calling for urgent regulatory review of the evidence.
He says the new findings add to previous evidence suggesting a possible risk for cancer with ARBs.
“[Dr.] Sipahi is a capable researcher, and this analysis needs to be taken seriously, but it needs to be verified. It is not possible to draw a strong conclusion on this analysis, as it is not based on individual patient data, but I don’t think it should be ignored,” Dr. Nissen stated.
“I will say again what I said 12 years ago – that the regulatory agencies need to carefully review all their data in a very detailed way. The FDA and EMA have access to the individual patient data and are both very capable of doing the required analyses.”
Limitations of trial-level analysis
Asked to evaluate the statistics in the current paper, Andrew Althouse, PhD, an assistant professor of medicine at the University of Pittsburgh, and a clinical trial statistician, explained that the best way to do a thorough analysis of the relationship between ARB exposure and risk for incident cancer would involve the use of patient-level data.
“As such data were not available to Dr. Sipahi, I believe he is doing as well as he can. But without full access to individual patient-level data from the respective trials, it is difficult to support any firm conclusions,” Dr. Althouse said in an interview.
He suggested that the meta-regression analyses used in the paper were unable to properly estimate the relationship between ARB exposure and risk for incident cancer.
“Taken at face value, the current analysis suggests that [in] trials with longer follow-up duration (and therefore greater cumulative exposure to ARB for the treatment group), the risk of developing cancer for patients in the ARB group versus the non-ARB group was progressively higher. But this study doesn’t take into account the actual amount of follow-up time for individual patients or potential differences in the amount of follow-up time between the two groups in each trial,” he noted.
Dr. Althouse says this raises the possibility of “competing risks” or the idea that if ARBs reduce cardiovascular disease and cardiovascular death, then there would be more patients remaining in that arm who could go on to develop cancer. “So a crude count of the number of cancer cases may look as though patients receiving ARBs are ‘more likely’ to develop cancer, but this is a mirage.”
He added: “When there are some patients dying during the study, the only way to tell whether the intervention actually increased the risk of other health-related complications is to have an analysis that properly accounts for each patient’s time-at-risk of the outcome. Unfortunately, properly analyzing this requires the use of patient-level data.”
Cardiologists skeptical?
Cardiology experts asked for thoughts on the new meta-analysis were also cautious to read too much into the findings.
Franz Messerli, MD, professor of medicine at the University of Bern, Switzerland, commented: “Perhaps one would simply ignore this rambling, cherrypicking-based condemnation of ARBs if it were not for the powerful negative connotation of the word cancer. Thus, the meta-analysis of Dr. Sipahi purporting that ARBs could be increasing the development of cancers in a cumulative way is of concern to both physicians and patients.”
But, raising a similar point to Dr. Althouse about competing risks, Dr. Messerli said: “We have to consider that as one gets older, the cardiovascular disease state and cancer state will compete with each other for the outcome of death. The better that therapies protect against cardiovascular death, the more they will increase life expectancy and thus the risk of cancer.”
He also added that “in head-to-head comparisons with ACE inhibitors, ARBs showed similar efficacy in terms of death, CV mortality, MI, stroke, and end-stage kidney disease, so can we agree that the attempt of Dr. Sipahi to disparage ARBs as a class is much ado about nothing?”
Dr. Nissen, however, said he views the idea of competing risk as “a bit of a stretch” in this case. “Although ARBs are effective antihypertensive drugs, I would say there is very little evidence that they would prolong survival versus other antihypertensives.”
Dr. Sipahi also claims that this argument is not relevant to the current analysis. “ARBs did not increase survival in any of the high-exposure trials that showed an excess in cancers. Therefore, competing outcomes, or ‘survival bias’ to be more specific, is not a possibility here,” he says.
George Bakris, MD, professor of medicine at the University of Chicago Medicine, noted that while the current study shows a slight increase in cancer incidence, especially lung cancer, among those taking ARBs for more than 3 years, it “totally ignores the overwhelming cardiovascular risk reduction seen in the trials.”
“Moreover,” he adds, “the author notes that the findings were independent of ACE-inhibitors, but he can’t rule out smoking and age as factors, two major risk factors for cancer and lung cancer, specifically. Thus, as typical of these types of analyses, the associations are probably true/true unrelated or, at best, partially related.”
Dr. Bakris referred to the potentially carcinogenic nitrosamine and azido compounds found in several ARB formulations that have resulted in recalls.
“At any stage of drug synthesis throughout each product’s lifetime, these impurities may evolve if an amine reacts with a nitrosating agent coexisting under appropriate conditions,” he said. “Drug regulatory authorities worldwide have established stringent guidelines on nitrosamine contamination for all drug products. The studies noted in the author’s analysis were done well before these guidelines were implemented. Hence, many of the issues raised by the authors using trials from 10-20 years ago are not of significant concern.”
Still, the cardiology experts all agreed on one thing – that patients should continue to take ARBs as prescribed.
Noting that worldwide authorities are now addressing the issue of possible carcinogen contamination, Dr. Bakris stressed that patients “should not panic and should not stop their meds.”
Dr. Nissen added: “What we don’t want is for patents who are taking ARBs to stop taking these medications – hypertension is a deadly disorder, and these drugs have proven cardiovascular benefits.”
Dr. Sipahi received no specific funding for this work. He reports receiving lecture honoraria from Novartis, Boehringer Ingelheim, Sanofi, Sandoz, Bristol-Myers Squibb, Bayer, Pfizer, Ranbaxy, Servier, and ARIS and served on advisory boards for Novartis, Sanofi, Servier, Bristol-Myers Squibb, Pfizer, Bayer and I.E. Ulagay. The other commenters do not report any relevant disclosures.
A version of this article first appeared on Medscape.com.