Comments on Proposed U.S. Biogenerics Legislation

On December 22, Essential Action submitted comments to the Federal Trade Commission (FTC) regarding proposals for U.S. biogenerics legislation.

The comments argue that “providing timely access to affordable, safe and effective products should be the central purpose” of biogenerics legislation. “Provisions that extend the monopoly protection period of brand-name companies, or otherwise make it unreasonably difficult to sell affordable biogenerics to patients as soon as possible after patent expiration, would therefore defeat the purpose of the new rules.” To ensure that this purpose is met, new rules should:

1. Avoid Inappropriate Marketing Monopolies (Data Exclusivity); and
2. Ensure Timely Patent Dispute Resolution and Patent Disclosure.

A .pdf version of the comments is available here:

The text of the comments follows the continuation of this post.

Essential Action
Access to Medicines Project
P.O. Box 19405, Washington, DC, USA 20036
(202) 387-8030

December 22, 2008

Federal Trade Commission
Office of the Secretary
Room H-135 (Annex F)
600 Pennsylvania Ave., NW
Washington, DC 20580

Re: Emerging Health Care Competition and Consumer Issues – Comment, Project No. P083901

Dear Sir or Madam,

Thank-you for providing Essential Action the opportunity to submit comments on issues raised during the November 21, 2008 “FTC Roundtable on Follow-on Biologic Drugs: Framework for Competition and Continued Innovation.”

Essential Action is a non-profit organization involved in global access to medicines issues for more than a decade. We are strongly supportive of an efficient biogenerics approval process — one that would ensure timely access to safe, effective and interchangeable products, yielding significant savings for individual consumers, government agencies and other healthcare payers, and making important medicines more widely available. Our Access to Medicines Project has been funded by the Ford Foundation and the Open Society Institute.

Comments on Key Issues Raised during the FTC Roundtable

Providing timely access to affordable, safe and effective products should be the central purpose of U.S. biogenerics legislation. Provisions that extend the monopoly protection period of brand-name companies, or otherwise make it unreasonably difficult to sell affordable biogenerics to patients as soon as possible after patent expiration, would therefore defeat the purpose of the new rules. To ensure that this purpose is met, new rules should:

1. Avoid Inappropriate Marketing Monopolies (Data Exclusivity)
2. Ensure Timely Patent Dispute Resolution and Patent Disclosure

1. Avoiding Inappropriate Marketing Monopolies (Data Exclusivity)

The primary rationale for data exclusivity is that drug development is expensive and risky. Data exclusivity, according to its proponents, is necessary both to provide an incentive for brand-name companies to undertake research and development (R&D) and to ensure that they are not placed at unfair disadvantage as against ‘free-riding’ generic firms.

We believe that data exclusivity provisions should not be included in U.S. biogenerics legislation that amends the Public Health Services Act (PHSA), for the following reasons:

A. Data exclusivity is not needed for brand-name biotech companies to re-coup their R&D Costs;
B. Data exclusivity is a major barrier to generic competition;
C. Data exclusivity overcompensates data originators; and,
D. The cost-sharing approach to registration data is a more efficient, pro-competitive approach to rewarding innovation than data exclusivity.

A. Data Exclusivity is not needed for brand-name biotech companies to re-coup their R&D Costs
There is reason to challenge whether data exclusivity or any other form of compensation for registration data is needed to provide adequate incentive for biotech R&D. Brand-name biotech companies already have large incentives under the patent system to conduct R&D – patents are 20-year marketing monopolies that allow innovators to set a price that will allow them to recoup all of their costs plus earn substantial profits.

A related but distinct argument is whether data protection, including data exclusivity, is necessary to provide a fair return to the companies that conducted the clinical trials, which are but the final stage of the research and development process. There is an actual cost to conducting clinical trials, this argument runs; if generic competitors are able to rely on marketing approvals based on those trials in order to obtain their own marketing approval, they will unfairly free ride on the investment of the originator firm. This argument has less force for products for which companies benefit from patent protection, because free from competitive pressures they can set a price that allows them to earn profits, and not just re-coup their R&D costs.

It must be noted that the brand-name biotech industry is in fact asserting that the patent system does not adequately incentivize R&D, primarily in support of its claim that very extensive data exclusivity — as much as fourteen years — should be granted to biologic drugs approved under the PHSA. This exclusivity period is substantially longer than that five to eight years available to conventional drugs and biologics approved under the Food, Drug and Cosmetic Act (FDCA). BIO (the brand-name Biotechnology Industry Association) has argued that there is the “very real potential” patent protection will not provide the incentives needed for continued biologics innovation – and data exclusivity must step in to take patents’ place. In particular, BIO asserts that an extended data exclusivity period is necessary because patent protection might be narrower for many of these products. This might allow generics manufacturers to design around patent claims, producing products that are similar, but nevertheless, not infringing the innovator’s patents.

But that is not what BIO told U.S. Congress last year during debate over the patent reform bill. Then, BIO lobbied for strong patent protection, citing accomplishments in biotechnology and stating, “All of this innovation is possible because of the certainty and predictability provided by the U.S. patent system.”

It is clear, therefore, that the brand-name biotechnology industry believes patents do play a critical role in incentivizing innovation, while at the same time it seeks unprecedented levels of data exclusivity. But the only possible public justification for an extended period of data exclusivity is that the patent system will not work to reward innovators. The industry cannot have it both ways: either patents do or do not provide adequate incentive for research and development. Because the limits of patentability and patent claims are not being considered in the context of U.S. biogenerics legislation, no data exclusivity should be granted to brand-name companies, to ensure the industry does not receive a windfall at the public’s expense.

Additionally, even in the absence of data exclusivity or patent protection, brand-name firms enjoy both the same benefits as first entrants in conventional pharmaceutical markets (including building up brand-name identity and allegiance) but advantages unique to the biologics market. Biologics may enjoy a period of de facto exclusivity resulting from the difficulties inherent in producing them. As with brand-name companies, it will also take generic firms several years to develop FDA-approved manufacturing processes for biologics, and even with a streamlined regulatory pathway — something which certainly remains politically contested — this approval process will likely be considerably longer and more expensive bin many cases than for conventional drugs. In some cases, biogeneric production may push the limits of cost-efficiency.

B. Data Exclusivity is a Major Barrier to Generic Competition

Providing data exclusivity in addition to patents to reward innovators is also not desirable because data exclusivity poses major barriers to generic entry. In general, if generic firms are unable to use or rely on originators’ data, they will not enter the market until they are able to rely on the data. Redoing the tests conducted by brand-name companies is not only wasteful, it is frequently too time-consuming and expensive for the relatively low-capitalized generic industry to manage, not to mention unethical in the case of testing that involves humans. Thus data exclusivity confers an effective marketing monopoly for the term of the exclusivity period, potentially delaying the onset of generic competition, keeping medicine prices high for a longer period of time.

Where patent monopolies extend beyond the period of data exclusivity provided, data exclusivity may have little practical effect. But frequently data exclusivity will be of consequence. The provision of exclusive rights to registration data can provide patent-like protections in cases where the patent is found to be invalid, or for which the patent term has expired.

If its patents are invalid, it is not clear that the brand name product contributes significantly to innovation, and it is not clear that the public should insulate the company from competition. Indeed, when a product lacks the innovative properties to qualify for patent protection, it is generally in the interest of innovation to promote competition, so that the next innovations – like another company’s improvement on an existing drug – are not held up by exclusive rights.

In the case of U.S. biogenerics legislation, the likelihood of extensive delays to price-lowering generic competition resulting from data exclusivity is a major concern because brand-name companies are advocating for an unprecedented exclusivity period of 14 years, which significantly longer that five to eight years granted to innovators of conventional and biologic drugs under Hatch-Waxman. The Hatch-Waxman history is replete with abuses of the exclusivity process, including elaborate and sophisticated evergreening strategies that game the system in order to extend monopoly protections beyond those envisioned by the statute. There is every reason to suspect similar strategies will be employed with a parallel system for biologics, so that exclusivity may in practice extend even beyond 14 years.

C. Data exclusivity overcompensates data originators
Data exclusivity is also economically inefficient, because an effective marketing monopoly is likely to provide overcompensation for data originators, enabling them to earn many times the cost of the clinical tests that produced the data.

There is also no evidence to support brand-name company arguments that a substantially lengthier exclusivity period is warranted for biologic products approved under the PHSA than is available for products approved under the FDCA. For example, the brand-name industry frequently cites published studies that report the cost of producing brand-name conventional drugs and biologic drugs is almost identical ($1.2 billion versus 1.3 billion). If it doesn’t cost significantly more to develop biotech products, there is no objective justification for the brand-name industry’s proposal to increase the amount of data exclusivity granted to these products by almost 300 percent, from a minimum of five years guaranteed by Hatch-Waxman to fourteen years.

D. The Cost-sharing approach to registration data is a more efficient, pro-competitive approach than data exclusivity
If policymakers wish to create an additional incentive specifically for the cost of clinical trials, there are approaches — such as sharing the cost of clinical trials — that satisfy the public policy rationale for providing data exclusivity to innovators, at a much lower cost and while avoiding data exclusivity-conferred marketing monopolies that undermine access to medicines and other public health objectives. If a policy decision is made that innovator companies need investment protections beyond those afforded by the patent system, the cost-sharing approach would be an efficient and pro-public health alternative to the data exclusivity approach. This approach gives generic firms an automatic right to use originators’ data, but requires them to pay a share of the documented costs of generating the data, proportionate to the size of the markets in which they are selling their product.

The cost-sharing approach gives generic competitors an automatic right to rely on the registration data generated by originator companies, or marketing approval authorizations based on that data. But it requires the generic entrants to pay for use of the data (or for relying on marketing approvals based on the data). Under the cost-sharing approach, the amount generic competitors pay for using or relying on the data is based on the actual costs of generating the data and the proportionate global market share obtained by the generic competitor.

The key features of such a system are:
• The originator of the data must disclose and document the actual costs incurred in generating the data
• The generic competitor pays a proportionate share of cost.

To avoid overcompensation for data, or double compensation for originators of data, caps and limits on payment should be applied under this approach:
• If a product is covered by a patent, no registration data compensation is paid;
• When the company that originated the data earns from sales a certain multiple (we propose 20 times) of its cost in generating registration data, it loses its right to data compensation from generic competitors;
• The right to compensation expires five years after marketing approval has been granted to data originators.

Here’s how the system would work in practice: Innovator A receives marketing approval for biologic pharmaceutical P in 2010. Generic company G1 gains marketing approval in 2012 and immediately gains a 50 percent market share. Stipulate that P is sold only in the United States. A documents clinical testing costs of $100 million. G1 must pay $10 million to A at the end of 2012, at the end of 2013 and the end of 2014.

Now assume another generic entrant, G2 enters the market in 2013. A, G1 and G2 each have a 33 percent market share. G1 and G2 must each pay $6.67 million to A in 2013 and 2014.

This system would be administratively manageable. A version is already in effect for U.S. approval of agricultural chemicals, although the agrichemical cost-sharing scheme follows only after grant of an initial marketing monopoly.

The cost-sharing approach acknowledges that there are genuine and significant costs associated with conducting clinical trials to obtain marketing approval for biologics. By providing compensation to companies that originate registration data, the cost-sharing approach deals directly with the claim by brand-name companies that “free-riding” by generic entrants will undermine R&D incentives or unfairly situate the originators of registration data.

The cost-sharing approach also narrowly tailors the reward offered to data originators. It provides direct compensation based on the actual cost of data used to obtain marketing approval, ensuring that data originators obtain proportionate compensation for others’ use of the results of the originators’ investment.

This approach contrasts sharply to the data exclusivity approach, which rewards data originators with effective marketing monopolies. The cost-sharing approach considers an effective marketing monopoly as likely to provide overcompensation for data originators, enabling them to earn many times the cost of their investments, with monopoly rewards unrelated to the size of their investment. The cost-sharing approach also rejects the idea of marketing monopoly as appropriate for an investment-based compensatory scheme – one that is trying to avoid uses of the fruits of originators’ investment that may be considered “unfair,” but is not trying to reward creative genius in the fashion of patents.

2. Timely Patent Dispute Resolution and Patent Disclosure

The patent resolution system in U.S. biogenerics legislation should aim for clear disclosure of patents claimed by first registrants, and rapid resolution of potential patent claims. We believe that key features should be:

A. Initial registrants should be required to disclose claimed patents as a condition of enforcement;
B. Second entrants should be able to seek resolution of patent claims at any point after the reference product registration;
C. There should be no requirement for second entrants to share confidential information in advance of any litigation process; this objective can be achieved by placing responsibility for initiating a patent resolution process on the second entrant; and
D. Second entrants should have the option to bypass the early patent resolution system

A. Patent Disclosure Should be Mandatory

The patent system is premised on public disclosure. Not only is the basic fact of a patent claim supposed to be public knowledge, but the very provision of a patent is supposed to embody a trade-off whereby the means to make the underlying invention is publicized in exchange for grant of the patent monopoly. Moreover, to perform their property-delineating function effectively, patents must provide effective notice to the public and potential industry competitors.

Given the essential public component and notice functions of the patent system, there is no legitimate public policy rationale in patent claims on medicines being treated as proprietary or subjected to industry gamesmanship.

Routine patent disclosure should therefore be the norm for medicines. For conventional drugs registered under the Food, Drug and Cosmetic Act, this routine disclosure is achieved through Orange Book listings. This is a problematic approach because of the linkage system, but it does at least achieve the disclosure objective. We believe a sound public policy approach would require disclosure of claimed patents as a condition of enforcement, and believe this regime should be adopted for biologics registered under the Public Health Service Act.

Thus, initial registrants should be required at the time of application to indicate any granted or filed patents that they believe apply to the biologic for which they seek marketing approval. This should include both patents granted to the registrant or which have been licensed to them. They should be required to update this list for any new patent filings, within a statutorily defined period, perhaps 30 days. Failure to disclose should forfeit the right to enforce.

B. The Patent Resolution Process Should Be Available at Any Point After Initial Registration

Given the centrality of patents to pharmaceutical manufacture, and the considerable up-front costs of undertaking tests to determine generic substitutability (or comparability, or therapeutic equivalence, or similarity), it is often impractical for generic manufacturers to introduce a product onto the market without ascertaining that they can do so without infringing the patents held or licensed by the registrant of the reference product. For biologics, the expected greater cost of achieving and demonstrating substitutability, comparability, equivalence, or similarity will likely deter in many cases pre-marketing investments unless there is certainty about the patent landscape. It is thus vital that there be a system for pre-marketing resolution of the validity and applicability of reference product patents to a subsequent generic or similar product.

The objective of such a system should be to clear patent claims so that a) invalid patents do not delay investment in, or introduction of, generic or similar products; b) non-applicable patents do not delay investment in, or introduction of, generic or similar products; and c) all potential patent claims are resolved in advance of any applicable marketing exclusivities.

The originators have a legitimate interest in protecting and enforcing their patents. They do not have a legitimate interest in enforcing invalid patents, however, or delaying second entrant entry by brandishing patents that do not apply to the second entrant’s product.

Delays in starting the process of pre-marketing patent resolution serve only to enable invalid or non-applicable patents to delay second entrant investment or marketing. If a pre-marketing patent resolution process leads to a finding that a patent is valid and/or applicable to a second entrant, then the originator will be able to obtain full protection for that patent, no matter when the process is originated.

We thus believe that potential second entrants should be free to initiate patent resolution processes at any point following approval of an originator product.

With such a system, there may be cases in which a second entrant initiating a patent resolution process does so before developing its process to make its version of the reference product. In such a case, it might not be able to obtain clarity on process patents. This would be a risk borne by the second entrant. It would retain the right to initiate a patent resolution process for potentially applicable process patents at a later date.

C. The Second Entrant Should Not Be Required to Share Confidential Information During the Administrative Process

Some legislative proposals for early patent resolution require the second entrant to share confidential information with the maker of the reference product. Statutory promises of protection notwithstanding, it is hard to imagine such information remaining confidential and not being shared with scientists employed by the originator company. Such a requirement to share confidential information is notably discordant with the confidentiality protections afforded to originators.

Second entrants should not be required to share confidential information with reference product makers, at least until a court proceeding is underway.

This problem can be avoided by placing responsibility for initiating a patent resolution process on to the second entrant. If the second entrant identifies claimed patents that it believes to be invalid or not to cover its product, then those disputes can be litigated or resolved through an appropriate process, without any pre-screening of second entrant confidential information by the originator.

The originator company would reserve the right to enforce at a later date any patent not addressed through the pre-marketing patent resolution process.

D. Second Entrants Should Have the Right to Opt Out of the Early Patent Resolution System

Second entrants should reserve the right to bypass the early patent resolution system. It is especially important to preserve this right if the early patent resolution system requires the second entrant to share confidential information.

There is no diminution of the patent holders’ rights if a second entrant chooses to bypass a pre-marketing patent resolution process.

Because there are significant business risks in doing so, it is unlikely that most second entrants would exercise this option. But it should remain open. It may be the preferred choice for second entrants in particular cases, or because the pre-marketing patent resolution process evolves in such a fashion as to constitute a barrier to investment and marketing

Essential Action appreciates the opportunity to comment on issues raised in the FTC roundtable on developing a regulatory pathway for biogenerics. We would be happy to provide further input or clarification on our comments, as needed.


Robert Weissman,

Sarah Rimmington,

Cost of Living: Who Gets New Drugs?

By Sarah Rimmington
The New York Times
Letter to the Editor

To the Editor
Re “British Balance Benefit vs. Cost of Latest Drugs” (“The Evidence Gap” series, front page, Dec. 3):

It is not surprising that several of the drugs found by a British government agency to be not worth the cost for prolonging lives, or whose use is limited, are biotech drugs (known as “biologics”), like Avonex, Biogen’s blockbuster treatment for multiple sclerosis.

Biologics, drugs made from living proteins, are extraordinarily expensive, thanks to brand-name company pricing decisions. In some cases, the cost is more than $100,000 annually.

The first biologics are now losing the patent protection that permits companies to charge high prices, but there is no process for the United States Food and Drug Administration to approve generic versions. Generic competition typically results in price reductions as steep as 80 percent.

An efficient system for approval of generic biologics could save taxpayers and consumers tens of billions of dollars. But the terms of such a process — like whether brand-name companies are given new monopoly protections — are critical. Let’s hope Congress gets it right.

Sarah Rimmington
Washington, Dec. 3, 2008

The writer is a lawyer for the Access to Medicines Project at Essential Action

This letter responds to the following story.

The Evidence Gap: British Balance Benefit vs. Cost of Latest Drugs
By Gardiner Harris
The New York Times
December 3, 2008

RUISLIP, England — When Bruce Hardy’s kidney cancer spread to his lung, his doctor recommended an expensive new pill from Pfizer. But Mr. Hardy is British, and the British health authorities refused to buy the medicine. His wife has been distraught.

“Everybody should be allowed to have as much life as they can,” Joy Hardy said in the couple’s modest home outside London.

If the Hardys lived in the United States or just about any European country other than Britain, Mr. Hardy would most likely get the drug, although he might have to pay part of the cost. A clinical trial showed that the pill, called Sutent, delays cancer progression for six months at an estimated treatment cost of $54,000.

But at that price, Mr. Hardy’s life is not worth prolonging, according to a British government agency, the National Institute for Health and Clinical Excellence. The institute, known as NICE, has decided that Britain, except in rare cases, can afford only £15,000, or about $22,750, to save six months of a citizen’s life.

British authorities, after a storm of protest, are reconsidering their decision on the cancer drug and others.

For years, Britain was almost alone in using evidence of cost-effectiveness to decide what to pay for. But skyrocketing prices for drugs and medical devices have led a growing number of countries to ask the hardest of questions: How much is life worth? For many, NICE has the answer.

Top health officials in Austria, Brazil, Colombia and Thailand said in interviews that NICE now strongly influences their policies.

“All the middle-income countries — in Eastern Europe, Central and South America, the Middle East and all over Asia — are aware of NICE and are thinking about setting up something similar,” said Dr. Andreas Seiter, a senior health specialist at the World Bank.

Even in the United States, rising costs have led some in Congress to propose an institute that would compare the effectiveness of new medical technologies, although the proposals so far would not allow for price considerations. At the present rate of growth, medical costs will increase to 25 percent of the nation’s gross domestic product in 2025 from 16 percent, with half of the increase coming from new drugs and devices, according to the Congressional Budget Office.

To arrest this trend, the United States needs to adopt at least some of NICE’s methods, said Dr. Mark McClellan and Dr. Sean Tunis, who served earlier in the Bush administration as, respectively, administrator and chief medical officer of the Center for Medicare and Medicaid Services. Dr. Tunis said he spent a lot of time in government “learning about NICE and trying to adopt the processes and mechanisms they used, and we just couldn’t.”

That’s because the idea of using price to determine which drugs or devices Medicare or Medicaid provides has provoked fierce protests. But Dr. McClellan said the American government would soon have no choice.

Drug and device makers, which once routinely denounced the British for questioning product prices, have begun quietly slashing prices in Britain to gain NICE’s coveted approval, especially because other nations are following the institute’s lead. Companies have said that they will consult with NICE to help determine which experimental compounds enter the final stage of clinical trials, so the British agency’s officials will soon influence which drugs enter the market in the United States.

The British government created NICE a decade ago to ensure that every pound spent buys as many years of good-quality life as possible, but the agency is increasingly rejecting expensive treatments. The denials have led to debate over what is to blame: company prices or the health institute’s math.

Dr. Michael Rawlins, chairman of NICE, blames the industry, saying that some companies raise prices “to get profits up so their executives can get better bonuses.” Dr. Karol Sikora, a prominent London oncologist, said that the institute’s math was flawed and that Dr. Rawlins had a “personal vendetta” against cancer treatments.

Drug company executives who were interviewed uniformly promised to cooperate with NICE, but industry advocates were not so kind. Robert Goldberg, vice president of the Center for Medicine in the Public Interest, an advocacy group financed by drug makers, likened Dr. Rawlins and his institute to terrorists and said their decisions were morally indefensible.

Developing a Method

It all started with Viagra.

Pfizer’s introduction of the drug in 1998 panicked British health officials, who feared it would wreck the government’s health budget. So they placed restrictions on its use. Pfizer sued, claiming the government’s decision was arbitrary. To defend itself against similar claims, the government needed a standard method of rationing. The following year, NICE opened.

Asked whether he thought the institute would succeed, Frank Dobson, the Labor health minister at the time, famously said, “Probably not, but it’s worth a bloody good try.”

Britain’s National Health Service provides 95 percent of the nation’s care from an annual budget, so paying for costly treatments means less money for, say, sick children. Before NICE, hospitals and clinics often came to different decisions about which drugs to buy, creating geographic disparities in care that led to outrage. (Such disparities are common in the United States, even for federal Medicare patients.)

Now, any drug or device approved by the institute must be offered to patients. The institute has also written hundreds of treatment guidelines in hopes of improving, and making more consistent, basic medical care.

The institute has analyzed the cost-effectiveness of surgical operations, cancer screening tests and medical devices. For example, it found that drug-coated cardiac stents were worth only $450 more than bare-metal ones. In the United States, stent price differences are often far wider.

Five years ago, the British health institute recommended more emergency room CT scans of patients suffering from head trauma — forcing hospitals to buy more machines.

But the decisions that get the most attention are those involving new drugs. Any drug that provides an extra six months of good-quality life for £10,000 — about $15,150 — or less is automatically approved, while those that give six months for $22,750 or less might get approved. More expensive medicines have been approved only rarely. The spending limits represent the health institute’s best guess for how much the nation can afford.

After consulting a citizens group, the institute decided that the nation should spend the same amount saving or improving the life of a 75-year-old smoker as it would a 5-year-old.

‘Muddling Through’

The institute’s decision-making process involves a series of independent assessments, consultations with manufacturers, committee meetings, comment periods for outsiders and appeals that, taken together, Dr. Rawlins described as “procedural justice,” or “muddling through elegantly.” While the institute provides advice, decisions are made by one of three committees made up of doctors, nurses and economists from outside the government.

Transparency recently became a high priority, but gaps in the idea of openness remain. At the institute’s first public decision-making appraisal meeting in September, staff members handed a reporter a stack of documents, only to snatch them back moments later. The committee’s chairman, Dr. David Barnett, was so intent on keeping the meeting brief that he told a committee member: “This must be the last question. It must be relevant. Otherwise, you will feel my wrath.”

To analyze the value of the drug that Mr. Hardy, the kidney cancer patient, wanted, and the value of three other kidney cancer medicines, the British institute hired a university group that considered how many months the drugs delayed cancer’s progress.

Firestorm of Protest

The academics got drug prices and calculated the costs of administering them and treating their side effects. Not one of the drugs came close to being worth their expense, the group suggested. In a preliminary ruling in August, a committee from NICE agreed.

The decision caused a firestorm. Twenty-six prominent British oncologists wrote a letter to The Sunday Times saying that the institute assessed cancer treatments poorly and that patients were remortgaging their homes to buy drugs freely available in other countries.

Given that fewer than 6,000 people per year in England and Wales are diagnosed with kidney cancer, “Why put ourselves through so much heartache for very little money?” Andrew Dillon, the institute’s chief executive, asked in a September interview. “The answer is that if we don’t apply the same criteria even to small groups of patients, there’s little value to what we do at all.”

Dr. Sikora, who helped organize the August protest, predicted in a September interview that the institute would buckle under political pressure.

Flooded with anguished comments, the institute beat a hasty retreat. A preliminary consultation posted Nov. 5 said that the institute would instruct its appraisal committees to consider approving highly expensive life-saving drugs for terminal illnesses affecting fewer than 7,000 patients per year — a policy that seems tailor-made for Sutent and the three other kidney cancer drugs.

Negotiations with companies on possible discounts are continuing, and a committee is scheduled on Jan. 14 to make public this nascent compromise.

NICE has stood fast in other areas, though, rejecting Kineret for rheumatoid arthritis and Avonex for multiple sclerosis. In 2001, NICE ruled that Aricept and two other drugs used to treat Alzheimer’s disease were worth their costs only if patients’ conditions had increased from mild to moderate severity.

The analysis put a value on patients’ improved thinking skills, and possible savings from delayed entry into nursing homes. Instead of pills, the institute suggested more counseling.

Advocates for patients with Alzheimer’s disease called the decision heartless.

Dr. Rawlins said he was frustrated that his institute had been censured instead of the drug company executives who set sky-high prices. Take the case of Celgene, the maker of Revlimid, a drug for multiple myeloma, a bone-marrow cancer, that in a preliminary ruling on Oct. 28 the institute said was too costly.

Celgene’s first big seller was thalidomide, a decades-old medicine now used as a cancer treatment, which is so cheap to manufacture that a company in Brazil sells it for pennies a pill.

Celgene initially spent very little on research and priced each pill in 1998 at $6. As the drug’s popularity against cancer grew, the company raised the price 30-fold to about $180 per pill, or $66,000 per year. The price increases reflected the medicine’s value, company executives said.

In 2005, the company introduced Revlimid, a derivative of thalidomide that is supposed to be less toxic, but may be no more effective. Celgene priced it at about $260 per pill, or $94,000 per year.

Offering Discounts

Private and public insurers in the United States must pay whatever Celgene and other makers of unique cancer medicines decide to charge, so prices are soaring. Spending on cancer drugs and other such specialty medicines rose 9 percent last year and now represents 24 percent of the nation’s drug bill, according to Health Strategies Group, a New Jersey consulting company. Drug expenses in 2006 grew faster than any other part of the nation’s health bill except home care.

But because of the institute, Britain’s National Health Service has been among the first to balk at paying such prices, which has led many companies to offer the British discounts unavailable almost anywhere else.

Johnson & Johnson, for instance, agreed to charge for Velcade, another drug for multiple myeloma, only if tests showed it was effective in a particular patient. Novartis agreed to give free injections of Lucentis, a drug for age-related macular degeneration, if patients needed more than 14 shots. Dr. Rawlins said these deals were constructed by drug makers to hide from other countries the discounts offered in Britain.

“It’s a good deal for us, but I can’t see that it will work in the long run because I can’t see that others countries will be so dim as to not notice it,” Dr. Rawlins said.

A more prudent bureaucrat would never make such a remark. Dr. Rawlins said that he delighted in controversy, “although I’ll admit that it doesn’t always work out.” He wears thick glasses and fine suits whose pockets are stuffed with nicotine gum packages that rattle as he walks. He laughs easily, plays the piano and viola, and moves effortlessly between politics and medicine.

His criticisms of the pharmaceutical industry have sharpened.

“I want them to produce new drugs for conditions we really need treatments for, but I loathe their marketing practices, which corrupt doctors in a dreadful way,” said Dr. Rawlins, who until recently practiced general medicine and for years was chairman of the British version of the Food and Drug Administration. “And I’m very conscious that the prices the pharmaceutical industry charges are what they think the market will bear.”

In 10 years, the health institute’s budget has grown to $50 million from $13 million, and it is scheduled to rise to $142 million in four years. NICE has 270 employees, who include doctors, economists and pharmacists.

Worldwide Impact

Agencies like NICE are popping up across the globe. Dr. Leonardo Cubillos, Colombia’s national director of insurance, said that Colombia was using British methods to choose drugs for a national health insurance package.

Membership in an international group of drug and device assessment agencies grew to 45 last year from 8 in 1992. The British institute has created a consulting group to advise foreign governments.

Much of the reason for this proliferation of agencies is that, while prescription drugs represent just 10.3 percent of overall medical spending in the United States, that share is 17 percent on average in industrialized countries.

As spending on drugs soared in many nations — often haphazardly — overall health often showed little improvement. So international aid agencies are advising governments to adopt British assessments and deliberations to improve their public’s health while lowering costs, and officials are listening — a trend that is likely to accelerate during the present global economic slowdown.

The health institutes in both Britain and Germany may soon suggest prices for drugs, a strategy intended to deflect political pressure back on the companies and shorten negotiations that now often take months.

“We have been told that the price is the price, but the worm is turning now,” Dr. Barnett said.

Company executives acknowledge that they are increasingly acceding to British demands to slash prices.

But the most pressing question for the industry is what influence the British institute will have in the United States. The United States already spends more than twice as much per capita on health care as the average of other industrialized nations, while getting generally poorer health outcomes.

Michael O. Leavitt, the Bush administration’s secretary of health and human services, said in a September speech that, at its present growth rate, health care spending “could potentially drag our nation into a financial crisis that makes our major subprime mortgage crisis look like a warm summer rain.”

And while there is fierce disagreement about how and whether to control drug and device expenses as part of a broader reform of the health system, many say some cost controls are inevitable. At a September device industry conference in Washington, a seminar on the issue was standing-room only and half of the questioners mentioned NICE.

John R. Dwyer Jr., a Washington lawyer who represents device makers, said that many in the industry have believed that major changes to control costs in the federal Medicare program were inevitable, and “people see NICE as the only workable paradigm.”

Meanwhile, Mr. Hardy waits. In recent weeks his growing tumor has pressed on a nerve that governs his voice. He can barely speak and is increasingly out of breath. The Hardys are hoping that in January NICE will approve the use of Sutent, allowing Mr. Hardy further treatment.

“It’s hard to know that there is something out there that could help but they’re saying you can’t have it because of cost,” said Ms. Hardy, who now speaks for her husband of 45 years. “What price is life?”