The New York Times, Feb 26, 2004 pC1(L) col 02 (37 col in)

2 Cancer Drugs, No Comparative Data. (Business/Financial Desk) Gardiner Harris.

Nearly two years into the planning of a head-to-head federal test of two expensive widely used cancer drugs, not a single patient has been enrolled in the study. But in a sense there is no hurry: Congress has barred the government from using the results in deciding how much to pay for the drugs.

Medicare officials sought the study, hoping to see if Aranesp, a drug made by Amgen that costs about $1,300 a vial, is superior to Johnson & Johnson's Procrit, an earlier version of the drug that costs $470 a vial. The federal Medicare program spends more than $1 billion a year on the two drugs, which stimulate the bone marrow to produce red blood cells in patients who have become anemic during treatment for kidney disease or cancer.

''If Medicare is going to spend more on a new drug, we need to establish that it is going toward something that provides significantly improved care for seniors,'' explained Dr. Sean Tunis, Medicare's chief medical officer.

But the drug industry, which counts on high-priced new drugs for its growth, persuaded Congress last fall to include a provision in Medicare legislation forbidding the program to use evidence that two drugs have ''functional equivalence'' in its payment decisions. The industry argued that such a balancing act would discourage makers from developing innovative drugs.

Under the circumstances, George Morrow, executive vice president for commercial operations at Amgen -- which lobbied for the provision -- contends that there is no point any longer in conducting the study. The tortured history of the Procrit-Aranesp trial helps explain why comparative testing may not be the cure for rising drug costs that its advocates contend. Not only do such tests become bogged down in politics, but black-and-white conclusions are hard to come by.

With backing from Consumers Union, AARP, managed care groups and many others, Senator Hillary Rodham Clinton, Democrat of New York, successfully pushed for a provision in the fall Medicare bill providing $50 million for head-to-head trials of comparable drugs.

''I felt strongly about the amendment because this was one more way of getting information to decision makers -- physicians and patients,'' Senator Clinton said in an interview.

Yet President Bush eliminated the money entirely from his proposed 2004-5 budget. That upset the Senate's majority leader, Bill Frist of Tennessee, who is pushing to restore some funding. But even $50 million might not be enough to produce meaningful comparisons, according to officials of the government's Agency for Healthcare Research and Quality, which was to get the money under the original Medicare bill.

The hope among its backers is that comparative drug testing will help slow the growth of drug costs. Retail drug spending more than doubled from 1997 to 2002, to $162.4 billion, according to federal Medicare officials.

Among the reasons for the rise, most experts agree, is that makers set the prices of their newest drugs well above those of the older medicines they are intended to replace. The higher prices cover the cost of research, the industry says, and are justified by the drugs' benefits.

But sometimes, the newer, more expensive medications do not work better than older, cheaper ones.

For instance, Celebrex and Bextra, painkillers from Pfizer, cost as much as $4 a pill and have become among the most popular medicines in the world. Pfizer markets the drugs as easier on the stomach than older anti-inflammatory medications. But that benefit has not been demonstrated conclusively for either drug, and Pfizer acknowledges that neither soothes pain any better than ibuprofen, an older medicine that costs about 20 cents a pill.

''The problem with our health system today is that with the help of advertising to consumers, there is often this false understanding that the newest medicine is the greatest,'' said Gail Shearer, director of health policy analysis for Consumers Union. ''We need these studies to find out whether the most expensive and newest drug really is better than the tried-and-true drug.''

Many executives in the managed care industry want the federal government to sponsor head-to-head tests of a broad range of drugs. ''We need to get information about the relative clinical effectiveness of drugs so that when we look at price, we can make real evaluations,'' said Dr. Sharon Levine, associate executive director of Kaiser Permanente, a large managed care group. Where the performance is similar, she said, government programs and managed care companies would be justified in buying only the cheapest.

But reaching such conclusions is not easy.

Rarely are such studies able to answer all the most important questions. The National Cancer Institute has been mulling the appropriate design for the Aranesp-Procrit trial for nearly two years and will probably need another year before starting the test, said Dr. Ellen Feigal, acting director of the institute's division of cancer treatment and diagnosis.

When Medicare asked for the study in 2002, officials suggested a simple design: give varying doses of the two drugs to a group of patients, take samples of their blood, and figure out how much of each drug is needed to achieve an equivalent rise in their red blood cell counts.

''We decided that that was not the best approach to use going forward,'' Dr. Feigal said, in part because measuring red blood cell numbers does not prove convincingly which drug helps patients more. Instead, cancer researchers have retrieved reams of data from Amgen and Johnson & Johnson and will soon develop a computer simulation of the trial to help determine the best trial design. In the end, more than one trial may be needed, Dr. Feigal said.

Dr. Feigal declined to estimate the cost or size of the eventual trial or trials, but similar tests have cost millions of dollars. Indeed, for comparative trials to be the size needed to measure true differences between drugs, they generally need to be large, lengthy and expensive.

Yet even then, the results are not certain to be universally accepted. For instance, whatever the results of the Aranesp-Procrit trial, Mr. Morrow of Amgen suggested that it would not be able to take into account all of Aranesp's virtues. Aranesp is effective for a longer period than Procrit, Mr. Morrow said, so it may allow exhausted cancer patients to make fewer trips to the doctor.

''That's a huge benefit,'' Mr. Morrow said. ''We would be concerned that a straight cost-effectiveness study would not really factor that in.''

Dr. Bob DeLap, vice president for regulatory affairs at Johnson & Johnson, said he thought the trial would probably show both drugs to be about equally effective. ''Both drugs do target the same receptor,'' Dr. DeLap said. Since it's drug is cheaper, Johnson & Johnson would probably benefit if Procrit and Aranesp are found to work about the same.

These kinds of debates accompany even the best-designed comparisons.

In December 2002, the National Institutes of Health announced the results of a 33,000-patient, 8-year, $125 million study comparing the effectiveness of three blood pressure medicines. The trial found that diuretics -- 50-year-old drugs that cost 10 cents a pill -- worked better than Pfizer's huge-selling drug Norvasc, which costs about $2 a pill.

Yet fierce debate continues over the best way to treat high blood pressure. One of the study's problems is that no one was allowed to take more than one kind of pill, even though patients suffering from hypertension routinely do just that. Norvasc's popularity appears to have suffered little. Sales last year grew 13 percent, to $4.34 billion, making it one of the top-selling drugs in the world.

European regulators already require drug makers to compare new drugs with older ones, while the Food and Drug Administration simply asks that drug makers compare new drugs with placebos. Some advocates say that the F.D.A. should also require comparative studies, but Dr. Robert Temple, the agency's associate director of medical policy, said that such trials were rarely useful.

''When you look at the results, there is almost never a difference between active treatments,'' Dr. Temple said.

Indeed, he argued, if the goal is saving money, there is no need to spend tens of millions of dollars on comparative drug trials. Government agencies and insurers should always favor the cheapest among drugs in the same class, according to Dr. Temple, unless a drug maker spends the millions needed to prove convincingly that its more expensive drug is more effective. Merck, for instance, undertook a study that proved convincingly that its anticholesterol drug Zocor saves lives. Dr. Temple noted that Pfizer has yet to prove the same for Lipitor, its popular and similar pill.

''I generally assume these drugs are all the same unless somebody goes out and proves differently,'' Dr. Temple said. ''I don't think you lose much if you just always use the cheapest drug.''

Products: Aranesp (Medication) - Product development
Procrit (Medication) - Product development