April 13, 2011 (Copenhagen, Denmark)— At odds with a controversial 2010 meta-analysis [1], two new reports--one on an observational cohort study [2] and the other a meta-analysis [3]--have concluded that treatment with an angiotensin-receptor blocker (ARB) does not increase the risk of cancer compared with non-ARB therapy.
The new studies now make at least three that contest a June 2010 meta-analysis that shook the world of hypertension management by finding a "moderate" but significant 8% increased overall risk of incident cancers among ARB recipients vs comparators, which were usually placebo. The study from Dr Ilke Sipahi (University Hospitals Case Medical Center, Cleveland, OH) and colleagues, reported by heartwire , found in particular a significant 25% increase in risk of lung cancer. The report triggered an ARB safety review by the European Medicines Agency [4] and by the FDA [5], both of which remain ongoing.
However, "We believe there is now strong evidence for an absence of association between ARB exposure and cancer," Dr Björn Pasternak (Statens Serum Institute, Copenhagen, Denmark), lead author of the new cohort study, wrote to heartwire in an email. "Collectively, the available data provide support for physicians to continue to prescribe ARBs without concern about an excess risk of cancer."
His group's findings, published online April 11, 2011 in Circulation, were based on the public-health records of >300 000 people in Denmark. They showed no significant rise in risk of incident cancer or overall cancer mortality among those taking ARBs compared with others taking angiotensin-converting enzyme (ACE) inhibitors--independent of sex, type of ARB, and duration of ARB exposure.
In a subgroup analysis, there was no ARB link to risk of lung cancer or any of a long list of other malignancies, except for male genital cancers, primarily of the prostate. But that finding "must be interpreted with caution, considering the possibility of a chance finding resulting from multiple comparisons," Pasternak et al write.
The Danish-cohort findings were similar to those in the new meta-analysis, from the ARB Trialists Collaboration and published in the April 2011 issue of the Journal of Hypertension. It found no increased incidence of cancer for ARB recipients compared with control groups across 15 randomized trials encompassing almost 140 000 patients. Nor was the risk increased for specific types of cancer, including lung cancer, or for individual ARBs or for the combination of an ARB and ACE inhibitor.
A Happy End to the Story?
The new reports corroborate a November 2010 meta-analysis that saw no increased cancer risk from ARB therapy across 70 clinical trials and almost 325 000 patients. They "basically reassure us again that the risk of cancer cannot be documented in such a short period of time," said the November report's senior author, Dr Franz Messerli (St Luke Presbyterian Hospital, New York), to heartwire .
"The Sipahi paper was a very weak paper to start with. The cancer risk [in it] was minimal. We do not have any distinct pathophysiologic mechanism that would explain a cancer risk. So I would say the Sipahi paper has been completely refuted by the new evidence."
An editorial accompanying the ARB Trialists' report called it a "happy end of the angiotensin-receptor antagonists cancer story" [6]. In it, Dr Giuseppe Mancia (University of Milano-Bicocca, Monza, Italy) writes, "We can thus safely conclude that current evidence by no means indicates any risk of cancer associated with use of angiotensin-receptor antagonists, which patients can continue to take for their documented antihypertensive efficacy, cardiovascular protection, and good tolerability profile."
But in an email to heartwire , the lead author of the meta-analysis that sparked the debate said, "There are major methodological problems with both of these studies." Pasternak et al looked at data that are "unreliable to study the cancer risk with ARBs," said Sipahi, given an abundance of potentially confounding differences between the ARB and non-ARB groups.
And the analysis from the ARB Trialists Collaboration "mixes data from trials with the highest exposure to ARBs for the longest durations, which show an excess in cancers, with several low-exposure trials using lower drug doses, with shorter follow-ups or low compliance rates, that do not show such excess. Dilution of the data in this manner makes the excess cancer risk disappear," he said. "When it comes to cancer, the degree of exposure is of utmost importance, whether it is radiation or chemicals."
Incident Cancer Risk Flat, Cancer Mortality Down
Pasternak et al roped together several government databases in Denmark to identify new users of either ARBs (n=107 466) or ACE inhibitors (n=209 692) who had no history of cancer and were aged at least 35 from 1998 through 2006 and compared them for the primary end point of incident cancer overall, as well as by subtype of malignancy, and for overall cancer mortality.
Over mean follow-ups of 2.9 years for ARBs and 2.1 years for ACE inhibitors, the adjusted relative risk (RR) for incident cancer was 0.99 (95% CI 0.95–1.03) for the former vs the latter. Covariates included age, sex, socioeconomic status, degree of urbanization, use of other antihypertensive drugs, comorbidities, and hospitalizations during the previous three years; "we did not have data on smoking."
A history of prior ARB use was not significantly associated with cancer risk compared with past use of ACE inhibitors. The cancer mortality RR for ARBs was significant at 0.77 (95% CI 0.72–0.82).
In subgroup analyses, there was no significant association between ARB use and cancers of the mouth/pharynx, gastrointestinal tract, lungs, bone and joints, skin, breast, female genital organs, urinary tract, nervous system, endocrine glands, mesothelial tissue, hematopoietic tissue, or lymph nodes compared with ACE-inhibitor use. On the other hand, the adjusted RR for male genital cancers was 1.15 (95% CI 1.02–1.28).
Expanding on his criticisms of the analysis, Sipahi specified what he sees as potential confounding factors, including that ARB recipients were "younger, richer, and had fewer comorbidities, including history of heart attack, heart failure, [chronic obstructive pulmonary disease], and diabetes. So ARB patients were overall healthier. This actually led to a significantly lower risk of cancer in patients taking ARBs, which became nonsignificant only after correcting for some confounders. However, there are [also] unrecorded confounders (eg, data on smoking was not recorded in this study), which falsely make the ARBs look good.
"Even after correction for the recorded confounders, there is still a 23% reduction in cancer deaths with ARBs, with a highly significant p value of less than one in a billion. This is not possible and shows how erratic the methodology of this study is, because both our and others' meta-analyses of randomized controlled trials with longer follow-up rule out even a 5% risk reduction in cancer deaths with ARB therapy."
Response to "Undecided Questions"
The new meta-analysis combined 15 trials, with follow-ups of up to 60 months, that randomized patients to ARB-containing therapy or a comparator group. Allocations included ARB vs ACE inhibitor in five trials, ARB vs treatment not containing ACE inhibitors in 11 trials, ARB plus ACE inhibitor vs ACE inhibitor alone in seven trials, and ARB plus ACE inhibitor vs ARB alone in two trials.
Overall for ARBs vs comparator therapies, the odds ratio (OR) for incident cancer was 1.00 (95% CI 0.95–1.04). There was no excess cancer risk regardless of which ARB was used; type of cancer, including lung cancer; or for any of the individual treatment pair-ups, notably including ARB vs placebo without an ACE inhibitor and ARB plus ACE inhibitor vs either drug class alone.
The latter finding is particularly important, according to the Mancia editorial, due to questions about the safety of combined ARB and ACE-inhibitor therapy that "had been left somewhat undecided in a previous meta-analysis in which the combined blockade of the [renin-angiotensin system] RAS . . . was reported to modestly but significantly increase the risk of cancer compared with use of either RAS blocker in monotherapy."
That refers to the November meta-analysis coauthored by Messerli, which found, in addition to no overall increased ARB-cancer risk, a significantly elevated cancer risk among patients taking both an ARB and an ACE inhibitor.
So, in heartwire 's original coverage of that meta-analysis, drug-safety advocate Dr Steven E Nissen (Cleveland Clinic, OH) said it "hasn't completely put to rest the question about these drugs. We are in an area where the evidence isn't strong enough to warrant regulatory action but where it does warrant looking very carefully." Nissen, who had written an accompanying editorial on the Sipahi meta-analysis, also called on regulators to complete their ARB safety reviews promptly
The new studies now make at least three that contest a June 2010 meta-analysis that shook the world of hypertension management by finding a "moderate" but significant 8% increased overall risk of incident cancers among ARB recipients vs comparators, which were usually placebo. The study from Dr Ilke Sipahi (University Hospitals Case Medical Center, Cleveland, OH) and colleagues, reported by heartwire , found in particular a significant 25% increase in risk of lung cancer. The report triggered an ARB safety review by the European Medicines Agency [4] and by the FDA [5], both of which remain ongoing.
However, "We believe there is now strong evidence for an absence of association between ARB exposure and cancer," Dr Björn Pasternak (Statens Serum Institute, Copenhagen, Denmark), lead author of the new cohort study, wrote to heartwire in an email. "Collectively, the available data provide support for physicians to continue to prescribe ARBs without concern about an excess risk of cancer."
His group's findings, published online April 11, 2011 in Circulation, were based on the public-health records of >300 000 people in Denmark. They showed no significant rise in risk of incident cancer or overall cancer mortality among those taking ARBs compared with others taking angiotensin-converting enzyme (ACE) inhibitors--independent of sex, type of ARB, and duration of ARB exposure.
In a subgroup analysis, there was no ARB link to risk of lung cancer or any of a long list of other malignancies, except for male genital cancers, primarily of the prostate. But that finding "must be interpreted with caution, considering the possibility of a chance finding resulting from multiple comparisons," Pasternak et al write.
The Danish-cohort findings were similar to those in the new meta-analysis, from the ARB Trialists Collaboration and published in the April 2011 issue of the Journal of Hypertension. It found no increased incidence of cancer for ARB recipients compared with control groups across 15 randomized trials encompassing almost 140 000 patients. Nor was the risk increased for specific types of cancer, including lung cancer, or for individual ARBs or for the combination of an ARB and ACE inhibitor.
A Happy End to the Story?
The new reports corroborate a November 2010 meta-analysis that saw no increased cancer risk from ARB therapy across 70 clinical trials and almost 325 000 patients. They "basically reassure us again that the risk of cancer cannot be documented in such a short period of time," said the November report's senior author, Dr Franz Messerli (St Luke Presbyterian Hospital, New York), to heartwire .
"The Sipahi paper was a very weak paper to start with. The cancer risk [in it] was minimal. We do not have any distinct pathophysiologic mechanism that would explain a cancer risk. So I would say the Sipahi paper has been completely refuted by the new evidence."
An editorial accompanying the ARB Trialists' report called it a "happy end of the angiotensin-receptor antagonists cancer story" [6]. In it, Dr Giuseppe Mancia (University of Milano-Bicocca, Monza, Italy) writes, "We can thus safely conclude that current evidence by no means indicates any risk of cancer associated with use of angiotensin-receptor antagonists, which patients can continue to take for their documented antihypertensive efficacy, cardiovascular protection, and good tolerability profile."
But in an email to heartwire , the lead author of the meta-analysis that sparked the debate said, "There are major methodological problems with both of these studies." Pasternak et al looked at data that are "unreliable to study the cancer risk with ARBs," said Sipahi, given an abundance of potentially confounding differences between the ARB and non-ARB groups.
And the analysis from the ARB Trialists Collaboration "mixes data from trials with the highest exposure to ARBs for the longest durations, which show an excess in cancers, with several low-exposure trials using lower drug doses, with shorter follow-ups or low compliance rates, that do not show such excess. Dilution of the data in this manner makes the excess cancer risk disappear," he said. "When it comes to cancer, the degree of exposure is of utmost importance, whether it is radiation or chemicals."
Incident Cancer Risk Flat, Cancer Mortality Down
Pasternak et al roped together several government databases in Denmark to identify new users of either ARBs (n=107 466) or ACE inhibitors (n=209 692) who had no history of cancer and were aged at least 35 from 1998 through 2006 and compared them for the primary end point of incident cancer overall, as well as by subtype of malignancy, and for overall cancer mortality.
Over mean follow-ups of 2.9 years for ARBs and 2.1 years for ACE inhibitors, the adjusted relative risk (RR) for incident cancer was 0.99 (95% CI 0.95–1.03) for the former vs the latter. Covariates included age, sex, socioeconomic status, degree of urbanization, use of other antihypertensive drugs, comorbidities, and hospitalizations during the previous three years; "we did not have data on smoking."
A history of prior ARB use was not significantly associated with cancer risk compared with past use of ACE inhibitors. The cancer mortality RR for ARBs was significant at 0.77 (95% CI 0.72–0.82).
In subgroup analyses, there was no significant association between ARB use and cancers of the mouth/pharynx, gastrointestinal tract, lungs, bone and joints, skin, breast, female genital organs, urinary tract, nervous system, endocrine glands, mesothelial tissue, hematopoietic tissue, or lymph nodes compared with ACE-inhibitor use. On the other hand, the adjusted RR for male genital cancers was 1.15 (95% CI 1.02–1.28).
Expanding on his criticisms of the analysis, Sipahi specified what he sees as potential confounding factors, including that ARB recipients were "younger, richer, and had fewer comorbidities, including history of heart attack, heart failure, [chronic obstructive pulmonary disease], and diabetes. So ARB patients were overall healthier. This actually led to a significantly lower risk of cancer in patients taking ARBs, which became nonsignificant only after correcting for some confounders. However, there are [also] unrecorded confounders (eg, data on smoking was not recorded in this study), which falsely make the ARBs look good.
"Even after correction for the recorded confounders, there is still a 23% reduction in cancer deaths with ARBs, with a highly significant p value of less than one in a billion. This is not possible and shows how erratic the methodology of this study is, because both our and others' meta-analyses of randomized controlled trials with longer follow-up rule out even a 5% risk reduction in cancer deaths with ARB therapy."
Response to "Undecided Questions"
The new meta-analysis combined 15 trials, with follow-ups of up to 60 months, that randomized patients to ARB-containing therapy or a comparator group. Allocations included ARB vs ACE inhibitor in five trials, ARB vs treatment not containing ACE inhibitors in 11 trials, ARB plus ACE inhibitor vs ACE inhibitor alone in seven trials, and ARB plus ACE inhibitor vs ARB alone in two trials.
Overall for ARBs vs comparator therapies, the odds ratio (OR) for incident cancer was 1.00 (95% CI 0.95–1.04). There was no excess cancer risk regardless of which ARB was used; type of cancer, including lung cancer; or for any of the individual treatment pair-ups, notably including ARB vs placebo without an ACE inhibitor and ARB plus ACE inhibitor vs either drug class alone.
The latter finding is particularly important, according to the Mancia editorial, due to questions about the safety of combined ARB and ACE-inhibitor therapy that "had been left somewhat undecided in a previous meta-analysis in which the combined blockade of the [renin-angiotensin system] RAS . . . was reported to modestly but significantly increase the risk of cancer compared with use of either RAS blocker in monotherapy."
That refers to the November meta-analysis coauthored by Messerli, which found, in addition to no overall increased ARB-cancer risk, a significantly elevated cancer risk among patients taking both an ARB and an ACE inhibitor.
So, in heartwire 's original coverage of that meta-analysis, drug-safety advocate Dr Steven E Nissen (Cleveland Clinic, OH) said it "hasn't completely put to rest the question about these drugs. We are in an area where the evidence isn't strong enough to warrant regulatory action but where it does warrant looking very carefully." Nissen, who had written an accompanying editorial on the Sipahi meta-analysis, also called on regulators to complete their ARB safety reviews promptly
Δεν υπάρχουν σχόλια:
Δημοσίευση σχολίου