July 29, 2019
By Will Boggs MD
NEW YORK (Reuters Health) - Reporting of neoplastic adverse events associated with valsartan increased markedly in the months following its recall, according to findings from the Food and Drug Administration Adverse Events Reporting System (FAERS).
"The most interesting finding is that reporting spikes after an FDA recall and media attention, suggesting that adverse reporting is highly subjective and easily influenced by non-medical factors," Dr. Guilherme H. Oliveira from Case Western Reserve University, in Cleveland, Ohio, told Reuters Health by email.
Valsartan was recalled in mid-2018 after trace amounts of the potential carcinogen N-nitrosodimethylamine (NDMA) were found in some formulations of valsartan and other angiotensin-receptor blockers (ARBs). This recall brought wide media and public attention in the U.S.
Dr. Oliveira and Dr. Sadeer G. Al-Kindi evaluated trends in adverse-event reporting associated with valsartan and other ARBs from 2017 through 2018.
During this interval, there were 11,112 adverse events reported (5,151 for valsartan and 5,961 for other ARBs), of which 920 (8.7%) were related to neoplasms (14.7% of valsartan events and 3.6% of other ARB events).
The percentage of all reported ARB events attributed to valsartan increased from 5.3% pre-recall to 23.4% post-recall, the researchers report in Circulation: Cardiovascular Quality and Outcomes, online July 1.
The reporting odds ratio (ROR) for neoplasms - the fraction of all adverse events accounted for by neoplasms for valsartan versus other ARBs - increased abruptly from 1.8 in June 2018 to 15.4 in July 2018, immediately following the recall. This increase persisted through August (18.2) and September (17.2) before decreasing to 2.9 in December 2018.
Among reports submitted only by consumers, the ROR for neoplasms increased from 1.0 in the second quarter of 2018 to 16.5 in the third quarter and decreased to 2.4 in the fourth quarter.
Such an abrupt increase in valsartan-associated neoplasms is "biologically implausible," the authors say.
"Adverse-event reporting to the FDA should be limited to medical professionals and should be followed up by FDA-supervised investigations," Dr. Oliveira said. "Lay person reporting should be done to medical professionals that would then be required to report to FDA, if appropriate."
His advice to physicians: "Don't overreact to FDA recalls and do due diligence. Apply the scientific method to adverse reactions, as to any other scientific report."
Dr. Anton Pottegard of the University of Southern Denmark and Odense University Hospital, whose recent nationwide study did not show a marked increase in the short-term overall risk of cancer in users of valsartan contaminated with NDMA, told Reuters Health by email, "I would characterize the findings as 'unsurprising' and 'nicely highlighting the known limitations of the spontaneous adverse event reporting system.'"
"For detecting rare adverse events, such as cancer, adverse event reporting is and will continue to be a very inefficient tool," he said. "This is mainly due to the very long expected lag time between the drug exposure and the cancer being diagnosed. This makes it very unlikely that patients or health care professionals will think of linking the two and submitting a report."
"To identify such adverse effects, we have to rely on epidemiological analyses of health data, e.g., on drug use and cancer occurrence," said Dr. Pottegard, who was not involved in the new work. "Luckily, the science behind such analysis is evolving rapidly, as is the regulators' understanding and use of such analyses, to the benefit of patients worldwide."
Dr. Rita Banzi from Istituto di Ricerche Farmacologiche Mario Negri IRCCS, in Milan, Italy, who recently commented on the regulatory response to contaminated valsartan, told Reuters Health by email, "Passive pharmacovigilance is not enough or in some cases misleading. In general, clinical studies should proactively collect data on the safety of medical treatment with the same accuracy as for efficacy data (studies always look at benefits and just record adverse events). Harms of treatments are in general neglected."
"In the specific case of drug contamination, it is essential that pharmacovigilance is based on integrating networks of data to highlight possible rare effects that occur in the long term," she said.
"At least in Europe, signaling of adverse events in clinical practice is suboptimal," Dr. Banzi said. "Thus, I would encourage physicians to be more involved in the process, not just because of media reports or patients' concerns."
Circ Cardiovasc Qual Outcomes 2019.
(c) Copyright Thomson Reuters 2019. Click For Restrictions - https://agency.reuters.com/en/copyright.html