Anemia drugs made billions, but at what cost?

http://www.washingtonpost.com/business/economy/anemia-drug-made-billions-but-at-what-cost/2012/07/19/gJQAX5yqwW_story.html

Anemia drugs made billions, but at what cost?
By Peter Whoriskey, Published: July 19

On the day Jim Lenox got his last injection, the frail 54-year-old cancer patient was waiting to be discharged from the Baltimore Washington Medical Center. He’d put on his black leather coat. Then a nurse said he needed another dose of anemia drugs.

His wife, Sherry, thought that seemed odd, because his blood readings had been close to normal, but Lenox trusted the doctors. After the nurse pumped the drug into his left shoulder, the former repairman for Washington Gas said he felt good enough to play basketball.

The shots, which his cancer clinic had been billing at $2,500 a pop, were expensive.

Hours later, Lenox was dead.

For years, a trio of anemia drugs known as Epogen, Procrit and Aranesp ranked among the best-selling prescription drugs in the United States, generating more than $8 billion a year for two companies, Amgen and Johnson & Johnson. Even compared with other pharmaceutical successes, they were superstars. For several years, Epogen ranked as the single costliest medicine under Medicare: U.S. taxpayers put up as much as $3 billion a year for the drugs.

The trouble, as a growing body of research has shown, is that for about two decades, the benefits of the drug — including “life satisfaction and happiness” according to the FDA-approved label — were wildly overstated, and potentially lethal side effects, such as cancer and strokes, were overlooked.

Last year, Medicare researchers issued an 84-page study declaring that among most kidney patients, the original and largest market for the drugs, there was no solid evidence that they made people feel better, improved their survival or had any “clinical benefit” besides elevating a statistic for red blood cell count.

It was a remarkable finding of futility: While drugmakers had seen billions in profits over 22 years, much of it from taxpayers, millions of patients had been subjected to dangerous doses that might have had little advantage.

How did this happen?

To answer the question, The Washington Post obtained the agreements between the drugmakers and the Food and Drug Administration, reviewed thousands of pages of transcripts and company reports, and relied on new academic research, some by doctors who once administered the drugs but now look askance at the drugmakers’ original claims.

The multibillion-dollar rise and fall of the anemia drugs illustrates how the economic incentives embedded in the U.S. health-care system can make it not only inefficient but also potentially deadly.

Through a well-funded research and lobbying campaign, Amgen won far-reaching approvals from the FDA. Both pharmaceutical companies conducted trials that missed the dangers and touted benefits that years later would be deemed unproven. The companies took more than a decade to fulfill their research commitments. And when bureaucrats tried to rein in the largest doses, a high-powered lobbying effort occurred until Congress forced the regulators to let the drugs flow.

But at the center of any explanation of the popularity of these drugs are the nation’s doctors, clinics and hospitals, and the choices they made for patients.

Americans might like to think that doctors focus on only their health. But physicians and hospitals have to pay the bills, too, and, in some cases, the more they treat a patient, the more they earn. This was especially true in the case of the anemia drugs: The bigger the dose, the more they made.

Unlike medications that a patient picks up at the store, drugs administered by a physician, as these were, can yield a profit for doctors if there is a “spread” — a difference between the price they pay for the drug and the price they charge patients.

In this case, drugmakers worked diligently to make sure that doctors had an incentive to give large doses — that the spread was large. They offered discounts to practices that dispensed the drug in big volumes. They overfilled vials, adding as much as 25 percent extra, allowing doctors to further widen profit margins. Most critical, however, was the company’s lobbying pressure, under which Congress and Medicare bureaucrats forged a system in which doctors and hospitals would be reimbursed more for the drug than they were paying for it.

The markup that doctors, clinics and hospitals received on the drugs given to Medicare patients reached as high as 30 percent, according to the Medicare Payment Advisory Commission, a group that advises Congress. And the markup on doses given to patients covered by private insurance was even larger.

The incentives worked.

At the peak of the boom in 2007, more than 80 percent of 175,000 dialysis patients on Medicare were receiving the drug at levels beyond what the FDA now considers safe, according to federal statistics. Although other patients were receiving the drugs, the United States keeps closer records on dialysis patients.

“It was just so easy to do — you put this stuff in the patient’s arm, and you made thousands of dollars,” said Charles Bennett, endowed chair at the Medication Safety and Efficacy Center of Economic Excellence at the University of South Carolina and one of the critics of the use of the drug in cancer patients. “An oncologist could make anywhere from $100,000 to $300,000 a year from this alone. And all the while they were told that it was good for the patient.”

Take, for example, the doses of Aranesp that Jim Lenox was given several times during his cancer treatment, though not the injection he received at the hospital, which was Procrit.

The insurance company reimbursed the clinic about $900 for each, according to his patient records. The clinic would have paid about $600 for a dose of that size, at average prices from that time, meaning a profit of roughly $300 per administration.

“Jim trusted the doctors,” Sherry Lenox said.

The incentives drove remarkably high profits at Amgen — enough to elevate the small California firm into a Fortune 500 company. Its profit margin reached over 30 percent of sales, far higher than the industry average. As much as a third of that was coming from reimbursements funded by U.S. taxpayers.

Daniel Coyne, a professor of medicine at Washington University who had been a paid speaker for Amgen, promoting the drug, called the case “a paradigm for the pharmaceutical industry.”

Coyne said he became a critic after the drug’s danger surfaced and the company continued to promote higher doses.

“Amgen can say they are very contrite,” Coyne said. “They can say, ‘Isn’t it a shame that we didn’t know.’ But this isn’t a shame for Amgen. They won. They made billions.”

Amgen declined requests for interviews but responded to questions via e-mail. In a statement, the company said that as the understanding of the drugs has evolved, the company “quickly and responsibly communicated these new findings” and updated the product labeling 15 times since its approval.

“Any assertion that Amgen misled the public about the risks and benefits . . . is a gross misstatement of the facts,” the company said. “On the contrary, Amgen’s primary concern is for patients.”

As for the high reimbursement rates for the drugs, the company said it “has consistently advocated for appropriate access to vital therapies, and we routinely engage with policy-makers to share our views on key issues.”

Johnson & Johnson similarly declined interview requests but said in a statement, “As our understanding of the risk-benefit profile of [the drugs] has evolved over time, we have worked closely with the FDA to ensure new and relevant information is included in labeling.”

Emerging in the late ’80s, the new anemia drugs were among the earliest blockbusters from the nascent biotech industry.

Anemia arises when the body produces too few red blood cells, which carry oxygen from the lungs to the rest of the body.

The drugs consisted of man-made versions of a natural hormone called erythropoietin, which stimulates the body to produce red blood cells.

The discovery, which grew out of research funded in part by the National Institutes of Health, gave doctors an entirely new “natural” way of treating anemia. The previous method consisted of giving patients transfusions of red blood cells, a cumbersome process that can take as long as four hours.

To get the new man-made hormone approved, Amgen submitted the results of a key clinical trial of patients on dialysis. Because they, and other patients with kidney disease, frequently suffer from anemia, they would become the core market for the drugs.

That trial established that, indeed, the drug stimulated the production of red blood cells. Patients given the drug showed fewer signs of anemia — their hematocrit, or percentage of red blood cells as a percentage of blood, rose significantly.

The researchers also examined the evidence in the trials for signs of harmful side effects.

“The risks associated with [the] therapy are minimal,” Amgen wrote to the FDA.

The first two drugs of the trio, Epogen and Procrit, were approved by the agency in June 1989 for patients with kidney disease. Amgen made both; Procrit was licensed by Johnson & Johnson. Amgen’s Aranesp would be approved in 2001.

For a narrow portion of those patients — dialysis patients with anemia so severe they needed occasional blood transfusions — the drugs, if used in limited amounts, did offer a critical benefit, one that doctors say amounted to a revolution in treatment. Patients with severe anemia said it could restore their vitality. The new drugs allowed them to avoid the risks of transfusions, which can carry diseases and raise future complications for transplant patients.

The trouble would arise as the drugmakers won FDA approval for vastly expanded uses, pushing it in larger doses, for milder anemia and for patients with a wider array of illnesses. Very quickly, the market included nearly all dialysis patients, not just the roughly 16 percent who required blood transfusions. The size of average doses would more than triple. And over the next five years, the FDA would approve it to treat anemia in patients with cancer and AIDS, as well as those getting hip and knee surgery.

The key to their marketing was the claim that the drugs at higher doses could make patients feel better. By 1994, the drug’s label, approved by the FDA, advertised a range of benefits: “statistically significant improvements for . . . health, sex life, well-being, psychological effect, life satisfaction, and happiness.”

Those claims, withdrawn 13 years later because they did not meet new FDA standards for proof, would be the basis of television and print advertising campaigns, pitched to people with potentially fatal illnesses.

The drugs, according to one, offered “Strength for Life.”

But while ads touted the drugs’ virtues, some at the FDA had raised safety concerns. To address them, the drugmakers agreed to conduct two key safety studies.

The first was supposed to evaluate the drug’s “safety profile” and enrolled 2,100 patients. Scientists affiliated with Amgen published “interim” results in 1991 and 1993.

But the full safety results of the study were never published, and in later lists of safety investigations by the FDA and Amgen, there appears to be no reference to this study.

It’s not clear from the agency’s records how seriously anyone was taking the results, anyway. Amgen filed a “clinical study report” with the agency in 1995, and the company says its research commitment was fulfilled then. But the FDA did not deem the study completed until March 2004, almost 15 years after the company agreed to conduct it.

Neither Amgen nor the FDA would release a copy of the study report.

An FDA spokesman explained that the original report had been “misfiled” and that the agency has instituted tracking procedures to prevent that from happening.

A company spokesman said the report’s findings were consistent with what was known of the risks at the time.

The second promise of a safety report would arise as the drugs were being approved for cancer patients. In 1993, the companies agreed to conduct a study of whether the drugs might have “stimulatory effects” on tumor growth.

That year, Johnson & Johnson started a study of patients with small-cell lung cancer.

It was supposed to have 400 patients.

Eleven years later, the company said it was having difficulty recruiting them, having enrolled only 224. That meant it would be harder to reach statistically significant conclusions. Moreover, the FDA noted that in about 17 percent of the cases, data were missing.

With FDA approval, Johnson & Johnson halted the study, never finding evidence of clear dangers. But as Medicare researchers would later remark, the patients taking the drugs appeared more likely to die than those taking the placebo.

Amgen scientists agreed that the trial results favored the survival of the placebo group but wrote to the FDA that there still wasn’t enough information.

“The data are sparse,” they wrote to the FDA.

With the abandonment of that trial in 2004, the drugmakers committed to doing another study, which was supposed to be completed by 2008, according to an FDA letter to Amgen.

It still has not been completed.

It has been enlarged, Amgen officials said, and is not expected to be finished until 2017 — nearly 25 years after the drug was approved for use in cancer patients.

Amgen noted that the company “has sponsored and conducted many of the studies that have enhanced the understanding of [the drugs], including the discovery of new adverse events.”

One of the first questions facing regulators and doctors was a simple one: How much of this drug should be used?

Huge profits rested on the answer.

Doctors measure anemia by gauging a person’s hematocrit, the red blood cell percentage. In a healthy person, the hematocrit is about 40 percent or higher.

Before the new drugs, doctors might give a person a blood transfusion if their hematocrit dropped below 25 percent. The transfusion would raise them up to about 30 percent hematocrit but not all the way to “normal” levels.

With the advent of the drugs, which were easier than transfusions, doctors and drugmakers began to propose boosting a patient’s hematocrit all the way up to “normal” levels. Should patients who were just slightly anemic — and not just the sliver of transfusion-dependent patients — be given the drug to make them “normal”?

Funded by Amgen, the Normal Hematocrit Trial sought to explore the possibilities of raising the treatment target. It drew in more than 1,200 patients who were on dialysis and had a history of heart trouble. It was one of the largest trials to have been done on the drugs at the time.

About half the patients received enough Epogen to boost their hematocrit up to normal levels of about 42 percent; the other half received only enough to get them to 30 percent, the level typically achieved with transfusion.

Three years after the study began, the trial was halted. Patients in the “normal” higher-dose group were dying or having heart attacks at a higher rate than those in the lower-dose, lower-hematocrit group.

It was an indication that the drugs could be deadly. But what could have been a clear warning turned murky as it was presented to the public.

After the termination of the trial, the FDA added a summary of the results to the Epogen label, but it didn’t limit the recommended dosing levels, which the agency had recently expanded.

The reason for the “increased mortality” at the higher doses “is unknown,” the label said.

Then, when the results were reported in the New England Journal of Medicine in 1998, key information was glossed over or omitted.

Of the eight authors of the journal article, four were employees of Amgen; two others had served as consultants to the firm. This was disclosed in the article.

The authors did recommend that, when using the drugs in patients with heart problems, hematocrit levels should not be boosted all the way to normal. But the researchers sounded a skeptical note about the danger.

The rate of death was higher in the higher-dose group, the paper acknowledged, but “not significantly.”

“A higher . . . dose was not associated with higher mortality,” it said.

But, as it turns out, there was a statistically significant sign of danger — at least as the FDA now reports the study’s results.

Moreover, while the article suggested there were quality-of-life benefits to higher hematocrit levels, it left out the fact that no such difference was detected between patients in the higher-dose and lower-dose groups.

Neither of these issues would have become public if not for Coyne, who filed a Freedom of Information Act request with the FDA asking for the actual results of the trial.

It took the agency 31 / 2 years to respond to his request.

“These are very, very significant discrepancies,” Coyne said.

The four authors of the paper who were not employed by Amgen defended the article and said that “at no time was the intent of the original [article] to mislead anyone,” according to a letter made available to The Post, by doctors Anatole Besarab, W. Kline Bolton, Allen R. Nissenson and Steven Schwab.

The omission of the quality-of-life finding was a “victim of editing,” the letter said. Similarly, the authors said they used a different statistical technique, one they considered more appropriate, to determine that the danger signal fell short of significance following input from the journal editors and a statistician. A New England Journal of Medicine spokesperson said the 1998 article was accurate.

If there was a clear warning sign, however, it appears to have been lost.

Three years after the edited results of the Normal Hematocrit Trial appeared in the New England Journal of Medicine, the first listed author, Besarab, then of the West Virginia University School of Medicine, co-wrote an article recommending high dosage levels. It also said that patients in the trial’s high-dose group had seen “significant improvements” in quality of life.

“Should the Hematocrit Be Normalized . . . ?” it asked. “Yes!”

Similarly, in a 1999 paper, Bolton asked whether raising hematocrit to normal levels with the drug might be the “Viagra” of the field.

Across the country, the dosages continued to rise.

To understand the financial incentives at play as physicians tried to determine how much of the drug to prescribe, consider how much money a small change in FDA policy meant to the drugmakers.

When the agency first approved the drug, it recommended boosting a patient’s hematocrit up to 33 percent but no more; a few years later, after Amgen’s suggestion, it expanded the target range up to 36 percent.

That might not sound like a big difference, but the change had huge financial implications. As hematocrits rise, more of the drug is required to get it to rise again. An average dialysis patient dosed to reach the higher level consumes about 40 percent more of the drug, a jump that would push the amount consumed from $7,000 to $10,000 annually.

Amgen advanced the idea at hospitals around the country that patients should get their hematocrits raised. The sales pitch was that higher hematocrit meant a better quality of life.

“They’d bring lunch in, or they’d have presentations at conferences, and they’d always have the same message: A little was good, more is better,” said Steven Bander, formerly chief medical officer with Gambro, one of the nation’s largest dialysis chains. “They would quote all these studies about quality of life. They never talked about the negative data.”

The pitch worked: The average dose for the drugs more than doubled in the early to mid-’90s.

In a statement, Amgen said it “carefully trains its sales representatives on the importance of communicating the safe use of our medicines and responsible marketing” and alerted doctors directly.

But the company also had another key advocate within many of the nation’s hospitals and clinics.

The most commonly used dosing guidelines that doctors in the field used were issued by a group organized by the National Kidney Foundation, which says it has measures in place to manage conflicts of interest. But Amgen was the “founding and principal sponsor” of the guidelines. Moreover, in 2006, of the 16 members of the foundation’s panel that created the new dosing guidelines, 10 reported receiving consulting fees, speaking fees or research funds from Amgen or Johnson & Johnson’s subsidiary, Ortho Biotech.

It recommended doses at the high end of the FDA target recommendations.

But the enthusiasm for higher doses would go much further than that.

After a sharp rise in spending in the mid-’90s, the bureaucrats instituted a rule: No longer would Medicare cover expenses for the drugs when the patient significantly exceeded the FDA recommended level. Moreover, physicians could no longer apply for exceptions to the limits.

These restrictions provoked a protest — from doctors, dialysis clinics and the drug companies.

Amgen, which already had a sizable in-house lobbying effort, turned to powerful outside help. It spent $2.4 million on lobbyists that year, according to OpenSecrets.org. Among the Amgen representatives were Haley Barbour, the former chairman of the Republican National Committee, and C. Boyden Gray, formerly White House counsel to George H.W. Bush.

But it was then-Sen. Arlen Specter (D-Pa.) who led the charge against the new policy. During a hearing, he angrily questioned Nancy-Ann Min DeParle, the director of the Health Care Financing Administration, which had implemented the policy. As chairman of the subcommittee overseeing the agency’s budget, Specter could command more than the usual deference.

“There is a fury out there in the medical community as to what you are doing,” Specter, then a Republican, told DeParle.

Specter wanted Medicare to cover enough Epogen for patients to reach a hematocrit of 37.5; that change could have raised by 20 percent the amount of Epogen that doctors could freely prescribe in an average patient, adding a cost of $2,000 or more. Moreover, he indicated that if doctors wanted to go higher, they should be able to do so as long as they submitted a written justification.

Specter wanted to know: Who were the bureaucrats to question how doctors prescribe medicine?

“Ms. DeParle, what is your level of expertise in this field? What is your background and training?” Specter asked.

“I am a lawyer, sir,” she replied, according to a transcript.

“So there is no special level of expertise that you have to make this kind of an evaluation?” Specter, a lawyer, wanted to know.

Medical experts on her staff believed that Epogen was being overused, she said. Thousands of patients, after all, showed dosing levels greater than the FDA recommendation.

But within months, the Medicare bureaucrats had not only backed off the restriction but agreed to Specter’s higher limit.

On the day the agency raised the maximum level, Amgen shares spiked 6 percent.

Specter received $7,000 in campaign contributions that election cycle from the Amgen political action committee and $2,000 from the Johnson & Johnson PAC.

Specter did not return phone calls or an e-mail requesting comment.

DeParle declined to comment.

The doses kept rising.

By 2006, about half of all dialysis patients were getting so much of the drugs that their hematocrits were rising beyond the FDA-recommended ceiling of 36 percent. More than 80 percent were getting more than the level now deemed advisable.

Amgen defended the higher doses, saying that it was difficult for doctors to precisely target hematocrit levels.

“Physicians are not necessarily acting inappropriately when patients’ [hematocrit] temporarily exceed the FDA label target range,” the company said in a statement at the time.

The industry’s success at beating back attacks by the Medicare bureaucrats to rein in costs would be repeated again and again.

It wasn’t just the drugmakers who were advocating for the drugs, either. On Capitol Hill, the nation’s dialysis clinics, which were receiving as much as 25 percent of their revenue from using the drugs, were sometimes a key ally of the drugmakers.

One of the nation’s largest dialysis chains, in fact, in 2004 offered bonuses to its chief medical officer if he blocked efforts to reform the payment system. According to a financial filing, Charles J. McAllister, chief medical officer of DaVita, the dialysis company, was to receive a $200,000 bonus if the rules for the drugs’ use being considered by regulators were dropped or delayed. He was to receive an additional $100,000 if the then-new legislation, known as the Medicare Modernization Act, didn’t cut into the company’s revenue.

The Medicare proposal was “deeply flawed,” DaVita spokesman Skip Thurman said in a recent statement, because it limited dosing levels “without regard to the patient.”

At times, the companies would even enlist the patients to lobby on their behalf.

For example, in what may have been the drugmakers’ largest lobbying push, the companies sought to undo a Medicare proposal in May 2007 to restrict the use of the drugs in cancer patients.

The company spent millions trying to turn back this effort, including developing a Web site, Protectcancerpatients.org, that solicited testimonials from patients and instructed them on how to contact officials. Johnson & Johnson set up a similar one called Voiceforcancerpatients.com.

Amgen lobbying expenditures and political efforts jumped that year. The company ranked as the largest contributor to the campaign of House Speaker Nancy Pelosi (D-Calif.), which got $42,050.

Amid the campaign, Reps. Anna G. Eshoo (D-Calif.) and Mike Rogers (R-Mich.) drafted a letter to Medicare, signed by a majority in both houses, warning that the proposed Medicare limits on the drugs could have a “broad range of unintended health consequences.”

Contacted recently, Eshoo and Rogers indicated that the decision to use the drugs should be determined by doctors and patients, not the federal government.

“As a cancer survivor myself, I know every drug has risks,” Rogers said in a statement. “The federal government should not be in the business of dictating the practice of medicine.”

But among the most frequently lobbied issues appears to have been prices and doses, according to lobbyist disclosures.

For years, the profit margin for health-care providers — the “spread” between what they paid for the drugs and what Medicare paid in reimbursement — led the Office of the Inspector General to issue at least seven reports recommending either that the reimbursement price be reduced or the incentives changed. The Government Accountability Office and the Medicare Payment Advisory Commission made similar recommendations. At least a couple of times during the Clinton administration, the president’s proposed budget called for changing the incentives. But the measures didn’t make it through.

Instead, for years, the profit margins remained wide. As late as 2009, dialysis clinics were getting a markup of 9 to 17 percent on the drugs, according to an inspector general’s audit.

Eventually, however, there was no lobby that could overcome the steady drumbeat of health warnings that came from researchers.

For years, a small Bethesda-based nonprofit think tank, the Medical Technology and Practice Patterns Institute, had been publishing studies that challenged the conventional enthusiasm for the drug and the government policies that it said promoted their overuse. Then in November 2006, a study published in the New England Journal of Medicine reported that kidney patients targeted for higher doses were linked to higher risks of hospitalization, strokes and death. In December, a group of Danish researchers said that it had stopped a trial of Aranesp in cancer patients because of an increase in deaths and tumor growths. And that was just the beginning of the bad news, which Amgen didn’t seem particularly eager to share.

After the Danish research, the company waited three months before informing the public and did so only after the Cancer Letter, a newsletter, reported the findings. Analysts at the time asked chief executive Kevin Sharer why there had not been a prompt disclosure.

“Perfection says we should have done that,” he said.

Then the FDA cracked down: The drugs’ use was ruled out in cancer patients considered curable, it was ruled out in patients considered just slightly anemic, maximum recommended doses were lowered, and the agency told doctors in many cases to use the smallest amount possible to avoid a blood transfusion.

The agency also began to look askance at the alleged benefits, for which the evidence, in retrospect, seemed flimsy. There was no solid proof, under revised FDA guidelines for such measures, that use of the drugs leads to “statistically significant” improvements in happiness and other benefits, the agency said. Those quality-of-life claims, once so critical to the drug’s adoption, were removed from the label.

But it wasn’t just safety concerns that cut into sales. Shifting economic incentives depressed them, too.

Last year, nearly two decades after the Office of the Inspector General first suggested it, the economic incentives to use more of the drugs on patients in dialysis disappeared. Medicare added the drugs to a system known as “bundling,” under which a health-care provider is allowed a certain amount of money per dialysis patient, rather than more money for each dose.

The effects were immediate, suggesting again that health is not the only factor that doctors weigh in treating patients. After a quarterly sales plunge in April, Amgen chief operating officer and president Robert Bradway blamed the drop on the new payment scheme.

The lower doses were “driven by changes in the market arising from bundling,” he said.

In the first year of the new policy, sales of Epogen, the drug most affected by the rule, dropped 20 percent.

Although companies that sell dangerous drugs often pay a price in court, no major class-action lawsuits have been mounted, at least in part because the patients taking the drugs were already ill, attorneys said.

Amgen has been hit with whistleblower lawsuits alleging that the company engaged in illegal sales tactics, including the charge from several states that the company overfilled vials to provide an illegal kickback to doctors and hospitals. The company has denied that allegation, saying the overfill is common industry practice to ensure that doctors and nurses can withdraw a full dose from a vial. The company has set aside $780 million to settle the lawsuits, and says it has reached an agreement in principle to resolve the claims.

For those who have lost relatives who had been given the drugs, only doubts remain: What killed their loved ones — the disease or the drugs they took to treat it?

Jim Lenox, who had been fighting cancer off and on for six years, had been given the drugs multiple times, his insurance records show. It is impossible to know with certainty whether these drugs caused or hastened his death, doctors said, but they raised his risks. In a study now cited in a black-box warning on the drug label, the drugs decreased the survival of patients with the same type of cancer.

Sherry Lenox, 58, a waitress at a chain restaurant, still keeps a box of his medical records, wondering if, at some point, what happened will become clear. The night after the hospital visit in January 2008, he came home to a gathering of children and grandchildren. The hospital did not respond to requests for comment.

Shortly afterward, he collapsed, bleeding from his nose and mouth. Sherry Lenox tried to revive him. He was dead within minutes. His death certificate lists his cause of death as cancer.

“These days, when I think about him, I think a lot about the good things — what a great guy he was, our family,” she said.

That’s why Sherry Lenox, who has turned up at FDA meetings to confront physicians and drug company officials, presses them with a question: “Would you give big doses of these drugs to your loved ones?”

She said she has never gotten an answer.