Did the FDA miss signals on a troubled heart drug?

Introduction

It has been a rough eight months for the drug maker Actavis, Inc. and its powerful heart drug, Digitek. Following repeated Food and Drug Administration (FDA) criticisms of its manufacturing processes, Actavis Totowa, a subsidiary of Actavis, issued a recall of Digitek on April 25, after discovering that some of the tablets were twice as thick as intended, and thus contained double the dose meant for a patient. Digitek is so potent that even a slight increase in dosage can be toxic or fatal. In August, Actavis closed a plant in Little Falls, New Jersey, that manufactured Digitek, to institute “remediation” efforts; that closure also effectively shut two other Actavis facilities that provide laboratory and packaging services. In October, two members of the House of Representatives launched an inquiry into the FDA’s oversight of Actavis. Then, just last month, the U.S. Justice Department asked a federal judge to permanently enjoin Actavis and Actavis Totowa from manufacturing drugs until the firm could demonstrate compliance with FDA “Good Manufacturing Practice” requirements.

The recall was initiated by Actavis, and was handled in part by the drug’s distributor, Mylan Pharmaceuticals, which contacted wholesalers and other drug distributors. Actavis maintains that none of the improperly manufactured Digitek pills ever reached the public, telling the Center that it recalled the drug simply out of an “abundance of caution.” But public health experts and the FDA’s own data indicate that Digitek could be the source of serious problems. David L. Brown, a prominent cardiologist at the SUNY-Stony Brook School of Medicine, said that the tablets could pose a “major public health problem.” A Center analysis of FDA records shows that between April 1 and June 30, Digitek was cited by doctors, other health professionals, and consumers as the “primary suspect” in the deaths of hundreds of patients worldwide, the overwhelming majority in the United States. In response to the Center’s findings, the FDA has confirmed reports of 667 deaths that health care professionals and consumers attributed to Digitek during that same period — a number that has not previously been released. The FDA had received just one reported death attributed to Digitek in the previous three-month reporting period.

Public health experts and the FDA caution that these reports to the FDA’s Adverse Events Reporting System (AERS) must be interpreted with an abundance of caution; the reports can, after all, be made by anyone. Some outcomes ascribed to a drug may be caused by a patient’s underlying condition — in this case, heart disease. And the reports, say FDA officials, may be affected by “publicity about an event” — a caution that would certainly apply to a reporting period concurrent with a drug recall.

But some public health authorities say the spike in reports of Digitek problems in the spring should at least have served as a ‘signal’ to the FDA that further investigation was needed, and that the agency needed to more broadly sound the alarm to doctors, pharmacists, and the public. They note that AERS reports on a given drug may capture only a fraction of actual problems. And they warn that many patients and health care providers remain unaware of the recall — or of the spike in reported deaths attributed to Digitek — highlighting a broader problem in how the FDA notifies the public and what it does to get dangerous drugs out of consumers’ hands. Agency officials said they believed the recall addressed the major problems identified by FDA, and asserted that the recall was noted on the FDA’s MedWatch Alerts webpage. But former FDA advisor Curt Furberg, a well-known cardiovascular epidemiologist at the Wake Forest University School of Medicine, says it’s still not clear how many potentially “poisonous pills” are in patients’ medicine cabinets. And he says the Digitek episode represents a serious “systems failure” at the agency.

A History of Problems

Actavis Inc. of Morristown, New Jersey, is the U.S. manufacturing and marketing division of the Icelandic Actavis Group, which operates in 40 countries and has 11,000 employees worldwide. The company is one of the five largest generic drug manufacturers in the world. Actavis Totowa, LLC, a subsidiary of Actavis headquartered in Little Falls, New Jersey, sold Digitek, a generic form of digoxin, until it was withdrawn on April 25. The drug was distributed by Mylan Pharmaceuticals Inc., which is located in Morgantown, West Virginia, under a “Bertek” label and by UDL Laboratories Inc. of Rockford, Illinois, under a “UDL” label. Digitek accounted for roughly one-half of all the digoxin on the market in the United States.

Barbara Guthrie of Denver, Colorado, whose husband, James Guthrie died on August 18, 2007, after taking Digitek, filed a wrongful death suit in U.S. District Court for the District of Colorado on August 1st. She claims that Actavis should have known of the dangers of selling the drug because of well-known problems at the manufacturing plant and that it did so to “maximize sales and profits at the expense of the public health and safety and to maintain the Digitek brand integrity.” Teresa Toriseva, an attorney in Wheeling, West Virginia, and chairperson of a judge-appointed plaintiff’s steering committee for Digitek lawsuits filed nationwide, says attorneys are currently screening cases, and that hundreds or even thousands of suits may eventually be filed. Actavis said that it cannot comment on pending litigation.

The FDA has had concerns about Actavis that go well beyond Digitek. Five FDA inspections in 2006 and 2007 of three Actavis Totowa plants in New Jersey — two in Totowa and one in Little Falls — uncovered shoddy manufacturing practices and numerous “longstanding” quality control and reporting violations dating back to 1999 in regard to various drugs produced by the firm. During an August 2006 inspection, the FDA said it was impossible for Actavis to reliably ensure that their pills even contained the drug they said they contained. In a warning letter dated February 1, 2007, the FDA wrote that there was no assurance that many of the company’s drugs “have the identity, strength, quality and purity that they purport to possess.” Record keeping by the company was equally poor. Company records had been altered or “overwritten” and undesirable or “out of specification” test results were simply discarded and replaced with newer test reports “without any justification” for discarding the earlier results, according to the inspection report.

Asked why the FDA didn’t act when serious violations were found in August 2006, a spokesperson for the agency said in an email that “when questions regarding these deficiencies were raised by the FDA, the firm elected to voluntarily recall all the products manufactured at this facility.” The agency said Digitek “was one of the product lines to be included in the recall,” a reference to Actavis’s own recall this past April.

The FDA’s failure to ask Actavis to withdraw Digitek is “business as usual,” according to Vera Hassner Sharav, president of the Alliance for Human Research Protection, who said that although the FDA has the power to pull a manufacturer’s license to market a drug, it considers pulling a license equivalent to “the atom bomb — so they never do it.”

A Class I Recall

The recall of Digitek by Actavis was designated by the FDA as a Class I recall — the most serious type of recall, reserved for drugs that can cause serious or deadly reactions. Despite the gravity of the designation, neither the company nor the FDA did much to ensure that the public and health care providers were notified, critics charge. What the FDA posted on its MedWatch website was not a statement from the agency, but a news release written by Actavis. Regardless of who writes them, such postings are easily missed by the public and journalists, health experts say. For serious problems, they say, the FDA usually sends e-mails to journalists and holds a news conference. In 2004, when the painkiller Vioxx was found to cause heart attacks, the agency not only issued its own news releases, but also held hearings that resulted in an avalanche of news coverage.

But in the case of Digitek, the FDA never issued its own news release, nor did it hold a news conference. The result was predictable: only a smattering of short articles appeared, largely in local newspapers. No articles appeared in The New York Times or a number of other major outlets. Some, but not all, pharmacies called or mailed notices to patients taking Digitek. Many patients posting messages on consumer-focused websites say they received the notices weeks after the recall or not at all. Nor did the company or the FDA send warning alerts to physicians. Dr. Clyde Yancy, chief of cardiothoracic transplantation at Baylor University Medical Center in Dallas and President-Elect of the American Heart Association, says he learned of the recall from “one of several daily [cardiovascular] news sources” that he accesses. “I suspect that the system is not very well refined for broad based notification of physicians and likely needs revision to insure patient safety.”

Furberg, a former member of the FDA’s Data Safety and Risk Management Advisory Committee, said problems with Actavis indicate a “total failure of the system to oversee the production of medications and follow up on recalls.” He called the agency’s posting of the Actavis news release on its MedWatch website “inadequate,” and said that the FDA “should have issued an alert or asked the manufacturer to issue an alert to be sent to all physicians.” No one, he added, has tracked the millions of pills he estimates are affected by the recall in order to account for how many have been returned and how many potentially “poisonous pills” might be sitting in medicine cabinets or on the shelves of doctors’ offices today. Mylan Pharmaceuticals, which had at least partial responsibility for the recall, did not respond to numerous requests for comment.

Larry Sasich, a pharmacist and chairman of the department of pharmacy practice at the Lake Erie College of Osteopathic Medicine in Erie, Pennsylvania, agrees with Furberg that current systems are inadequate to conduct recalls fully and promptly. “When someone donates a [contaminated] unit of blood in Bangor, Maine, we can track that unit to any one of the eleven people who might receive a product from that unit of bad blood, whether they are in Bangor, Maine, or Santa Cruz, California,” Sasich said. But, he added, there is no similar “look-back mechanism” for drugs when they are recalled. “Pharmacies aren’t even required to keep track of lot numbers of the drugs they dispense,” he said. The problem was first raised by Sasich’s former employer, the watchdog group Public Citizen, in 1995, according to Sasich, when substandard lots of the anti-seizure drug Dilantin were released on the market. The group has pushed for improved mechanisms to recall drugs since then.

The Precautionary Principle

Several sources said that the danger posed by the double-dose Digitek pills could have been averted had the FDA acted when the agency first uncovered serious deficiencies in early 2006. Instead, for three years, the FDA simply sent several warning letters to the company. The FDA’s record prompted an inquiry by Congressmen John Dingell, U.S. House Committee on Energy and Commerce, and Bart Stupak, chairman of its oversight subcommittee, both Democrats of Michigan. On October 8, Dingell and Stupak wrote to FDA commissioner Andrew C. von Eschenbach, asking for records relating to the agency’s history of Actavis oversight in an effort to determine “whether the FDA permitted additional products from this firm onto the market while the agency knew or should have known about the breakdown of manufacturing practices at Actavis or its subsidiaries.”

On November 14, then-U.S. Attorney Christopher J. Christie of New Jersey filed a complaint in U.S. District Court in Newark seeking a permanent injunction barring Actavis Totowa and Actavis from manufacturing or distributing generic drug products until the firms demonstrated compliance with FDA requirements. The complaint states that the “FDA’s five inspections of Actavis Totowa’s facilities over the last three years have revealed numerous and recurring violations of the current Good Manufacturing Practice (CGMP) requirements for drugs.” As a result, says the complaint, the company’s drugs were deemed “adulterated” and “misbranded” — the latter term indicating that the company “lack[s] scientific evidence to demonstrate that they are safe and effective in their directions for use.” No hearings have yet been scheduled in the case.

Despite the recent actions, and the AERS reports, both Actavis and the FDA say that reports of injuries or deaths sent to the FDA are not proof that patients were harmed by Digitek. Actavis acknowledged receiving “approximately 2,200 adverse reaction calls related to Digitek” but said in an e-mail that “the overwhelming majority of those calls were in May – June – July, after the recall was announced.” As for the reported deaths, the company stated, “We do not know what caused those deaths, and there is no evidence that Digitek was in any way to blame.”

The FDA echoed the company’s claim, stating that the uptick in AERS reports “took place exclusively after the recall was announced publicly in the second quarter of 2008 and not within the interval period since 2006 when earlier releases of the same batch believed to also contain the double-thick pills occurred.” When asked whether the FDA believes that any of the deaths are actually due to Digitek, the agency would only say, “The recent adverse events reported for Digitek are similar to side effects that can occur when using any formulation of digoxin, even under prescribed conditions to treat heart failure and abnormal heart rhythms.” In response to a question about whether FDA has conducted an investigation to determine causality, the agency stated that “forming a causal link…would be difficult.”

But one organization is performing an analysis. The Institute for Safe Medication Practices (ISMP), a public interest group, spotted the “signal” in the FDA database. The group issued a notice in its ISMP Medication Safety Alert! newsletter on December 8, telling readers of a “marked signal linking reported deaths and serious injuries” to recalled Digitek. The group is currently analyzing the data and, according to Thomas J. Moore, senior scientist with ISMP.

Other outside experts remain concerned. Jerome Hoffman, a professor of medicine at UCLA, noted that not all of the reported AERS deaths occurred after the recall. “And, since the patients being treated tend to be at high risk in general,” said Hoffman, “it is common for doctors to assume that any deterioration that occurs is due simply to their underlying disease; this is well-known to lead to vast underreporting of drug-related adverse effects, for many medications — especially in the absence of systematic vigilance on the part of public agencies.” Hoffman argues that the high number of reports actually suggests that there may be many more, both before and after the recall, that went unrecognized.

Still, there is a larger question that concerns Hoffman — what he sees as the FDA’s sluggish response. “There couldn’t have been any scientific quandary in this case,” he said of Digitek. “Any medical student can tell you about the narrow margin of safety with digoxin.” As soon as officials knew that some patients could have been given double-doses of digoxin, he stressed, “the FDA should act to protect the public, unless and until it is substantially shown that there is no danger — rather than taking no action until someone can ‘prove’ that there is danger.”

David Donald and Caitlin Ginley contributed to this story.