Five days a week, and many nights, Margaret Akpan weaves her way through the sprawling D.C. General Hospital complex toward Unit 52, the main cancer treatment area in the District's only public hospital. A small woman with a huge smile, soft voice and beepers weighing down her white doctor's coat, Akpan passes an odd combination of one unfinished renovation attempt tumbling into the next. Every floor features a different eye-jarring fluorescent colored theme. Shackled prisoners with armed escorts from the nearby jail shuffle by, their ankle chains clanking loudly against the well-worn brown-and-tan tiled floor. Unit 52 is up on the fifth floor. Its nurses station is a blinding combination of electric yellow and orange that seems designed to overcome the gray, sagging spirits of most who must visit here. Just beyond is a cramped room that, on most days, serves as part storage space, part office for two nurses. On chemotherapy clinic days, it serves as D.C. General's primary treatment room, with two huge Barcalounger-like reclining beds surrounded by intravenous drip equipment and two televisions suspended from the ceiling. For Akpan's patients, sick with breast cancer or lung cancer or throat cancer, this room and a scheduled dose of toxins are their only hope. Settled into the upright recliners, with an IV in one arm and the TV remote control in the other, they gird themselves for what is to come -- the burning in the veins and nausea in the stomach -- and pray for remission or cure. Once the infusion starts, the battle inside their bodies begins. The drugs kill cells indiscriminately, attacking not just the cancerous ones but some healthy cells, too. The hope is that the cancer will die before they do. Some of Akpan's patients get so sick that for the next several days all they can do is sleep and vomit. Others lose their hair. The drugs leave a wretched taste in their mouths and a horrid smell in their bodies that won't subside for days. A few refuse to come back for more. Sometimes the medicine seems more deadly, more painful than the disease itself. Akpan knows that, for most cancers, this cocktail of liquid chemicals with names that sound as toxic as they feel usually doesn't represent a cure. At best, she hopes the cancer will stop spreading, maybe even disappear for a time. These chemotherapy drugs are among the few weapons she's got. Only there's one weapon missing from her arsenal, a cancer drug that has been hailed as close to a miracle drug, if such a thing can exist in the fight against cancer: Taxol. Discovered more than 30 years ago in the bark of the slow-growing Pacific yew tree, Taxol has been shown to aggressively attack a wide range of cancers in a way that other drugs have not. Leading oncologists now consider its use standard for breast and ovarian cancers, and increasingly for other types as well. In the case of ovarian cancer, which will kill about 50 percent of the women diagnosed with it, Taxol has provided the first real possibility of extending their lives. Akpan would like to give her patients Taxol whenever they need it, but she can't. For D.C. General, a cash-strapped public hospital, the No. 1-selling cancer drug in the world is just too expensive. It doesn't even stock Taxol in its pharmacy. Taxol can cost a patient on average anywhere from $10,000 for a complete treatment of ovarian cancer to $20,000 for breast. For that amount, Akpan can treat many more patients with other, though possibly less effective, therapies. "When you look at cost-effectiveness, you want to go with the less expensive drugs," she says, sounding frustrated but resigned. "I will use what I know that is still effective, even though the overall response might not be as good as Taxol . . . I will try those first." D.C. General is not the only public hospital where Taxol's price is a barrier to its use. Half a dozen such hospitals contacted by The Washington Post stock the drug in their pharmacies, but reserve Taxol for use after other therapies have failed, even for ovarian cancer. A couple other public hospitals were like D.C. General and do not even stock Taxol in their pharmacies. Inequities are not new to American medicine. What is remarkable about Taxol is this: It was discovered and developed by the federal government at substantial taxpayer expense. Yet it is now produced and priced exclusively by a private pharmaceutical company, Bristol-Myers Squibb, which has made millions on the drug. The government has been working with private-sector partners to bring its discoveries to market for decades. Historically, this arrangement was seen as a way to manage the enormous risks of developing new medicines. But the question of access remains. In the case of Taxol, one of the priciest cancer drugs on the market, some taxpayers are simply priced out. Finding lifesaving drugs is not easy or cheap. The path a promising compound will take from discovery to patient can cost hundreds of millions of dollars in research and testing, requiring decades of work and a wealth of scientific expertise. Even then, the risks are enormous. Each new stage of development is another chance for a drug to fail -- and most do. For every 10,000 compounds screened, only one or two will make it through the gantlet of testing to approval by the Food and Drug Administration. For decades, no drug company was willing to take those risks for cancer. The simple reason was that there was no market for cancer drugs. No one thought cancer could be treated. Medical science didn't even understand what cancer was, how it damaged cells and why it killed. The few cancer drugs that did exist were so toxic that the patients usually died anyway. Most doctors felt treating the disease was useless. "When I came {to Yale-New Haven Medical Center} as a resident in the mid-1960s, people used to pull me aside and say, You don't want to give these drugs to patients. It just makes them sick,' " recalls Vincent DeVita, who went on to head the National Cancer Institute in the 1980s. "So, it was a self-fulfilling prophecy. You can't cure it, don't treat them. You don't treat them, you don't cure it." Because of this mentality, it fell to the government to take on cancer. In 1955, with few therapies available, Congress pushed NCI into the drug business, and told it to find a cure for cancer. The intention behind the move was simple. The government would take the risks that the industry feared and put up the tax dollars to do so. Ultimately, it was hoped, after the public subsidized the discovery and development, these drugs would become profitable enough that companies one day might take the risk of developing the next generation of cancer drugs alone. Fueled by hundreds of millions of dollars that Congress pumped into its budget every year, NCI began a systematic attack on cancer. It marched thousands of investigators across the globe to find plants that might have anti-cancer properties. It created a massive network of testing laboratories that screened and followed thousands of potential leads. It conducted the large-scale human and animal studies necessary to bring promising drugs to market. Recalls Samuel Broder, who ran NCI from 1989 to 1995, "The NCI had to, in effect, become what amounted to a government-run pharmaceutical company." With one important exception. As a government agency, NCI was not equipped or inclined to manufacture and sell the products of its research. When it came across a great new drug, it had to find a private company to pair with in order to bring the drug to market. It was usually a hard sell. Despite NCI's efforts, the pharmaceutical industry continued into the 1960s and 1970s to view cancer drugs as money losers. Cancer was not like having high blood pressure or ulcers, where the market was huge and ongoing. For these conditions, people often take medicine every day for decades. Fewer people get cancer, and those who do take the drugs for a relatively short period of time and get better for a time or die. In some cases, though, deals were struck -- but only after NCI helped increase the chances of profitability by assuming much of the risk of development, and handing the companies drugs that were ready for FDA approval. All NCI asked in return was that the company take the drug to market as quickly as possible. NCI got no financial gain -- and sought none -- from the arrangements. The government's payback was the improvement of the public health. Two-thirds of all cancer drugs on the market were developed this way. At the time, this arrangement seemed a model of government-private sector cooperation. It was revolutionizing the war on cancer: By exploiting a company's desire for profit, the public good was served and the government got lifesaving treatments to dying patients. In hindsight, critics would wonder about whether this arrangement was too beneficial to the drug companies. Virtually everything known about cancer, from causes to diagnosis to treatment, was born out of research that began at NCI. Its researchers have found cures for some cancers, such as some lymphomas and leukemias, and life-prolonging treatments for others. Cancer is no longer the death sentence it seemed when Congress sent NCI out to attack the disease. Taxol is the one of the latest, and arguably the greatest, of NCI's accomplishments. It goes after cancer cells like no other drug before it. It demonstrates breathtaking responses in a wide range of cancers. Taxol has been a blockbuster hit for Bristol-Myers Squibb, posting sales of $2.8 billion since being approved in December 1992. Last year, Taxol accounted for nearly 40 percent of Bristol's cancer drug sales. By 2004, stock analysts predict, Taxol's slice of those sales will soar to 55 percent. The government, through its various health-benefit programs, is the largest purchaser of Taxol, with Medicare alone accounting for more than $160 million in sales in fiscal year 1997. The price charged for Taxol, according to Bristol officials, is a fair one and it will allow the company to recoup the millions of dollars it spent bringing the drug to market and generate further profit to offset the high cost and risk of developing newer and better drugs in the future. Any effort by the government to force lower prices would jeopardize that effort and the public health, they say. The company and its stockholders took a multimillion-dollar risk, and they deserve the payoff, if there is one, at the end. "Substantial financial incentives are necessary to justify an enormous investment, sometimes measured in hundreds of millions of dollars required for the rapid development of new pharmaceutical products," Zola Horovitz, then Bristol's vice president for business development and planning, told a 1993 congressional hearing looking into Taxol's price. "Few companies are willing or able to make investments of such magnitude. They want and deserve some assurance that they will have an opportunity to recover these investments and earn a reasonable return." But what about the American public? If a company and its stockholders demand a fair return, what can the public expect in return for its investment? Is a "reasonable price" unreasonable? "I had the worst hangover in my life," Kurt Blum sheepishly admits, remembering the day he stumbled across a Pacific yew tree in a Washington state forest. The night before, he and three fellow government botanists had set up camp on a riverbank near a down-on-his-luck logger. One of the botanists, a brash Texan, was certain some "cheap wine" could help the poor guy's woes. It seemed like a good idea at the time. But not that morning. As he stood among the towering firs, Blum's head pounded. Streaks of yellow rays peaked through the verdant canopy above. As he slowly scanned the wall of trees surrounding him, scarlet berry-like seeds dangling from a branch caught his eye. The bright red was a vivid aberration against the greens and browns. Blum was intrigued. He picked the berries, pulled the twigs, stripped pieces of bark, dug up some roots, carefully placed each specimen into individual plastic bags and put them all into his burlap shoulder sack. It was the summer of 1962. Blum was only 22. He had enlisted as a part-time foot soldier in the U.S. government's escalating war on cancer. Around the world, technicians like Blum were randomly gathering exotic plants and sending them back to NCI labs for testing of their cancer-killing abilities. Blum's mission for the summer was to tramp through forests up and down the Pacific Northwest, collecting "anything that looked like we could put a name on it": plants, leaves, bark, twigs, flowers. If it could be picked, plucked or pulled, it was bagged, marked and shipped off. Now a biology professor at Middle Tennessee State University, Blum recalls sitting around a campfire each night with his colleagues wondering whether their carefully tagged bags would ever amount to anything. Indeed, it took years before anyone realized his chance encounter in the Pacific Northwest would lead to such an important cancer drug. For two years Blum's packets went through testing at NCI, where extracts of yew bark were deemed to possess some cancer-killing properties. To pursue it further, NCI sent the samples to Monroe Wall and M.C. Wani of the Research Triangle Institute in North Carolina with the request that they figure out what molecule within the bark was responsible. At the time, the bark was just one of a handful of substances that Wall and Wani were evaluating. They soon realized it was unlike anything they had ever studied. "We became increasingly more and more interested," recalls Wall, now 81 years old and still working, along with Wani, at RTI. Soon, Wall sent a letter to NCI reporting "the broadest spectrum of {anti-cancer} activity that we have ever noted in our samples." It took until 1967 for the researchers to finally pinpoint the active compound, which they christened Taxol, after the tree's Latin name, Taxus brevifolia. There was little enthusiasm at the time from NCI. The Taxol molecule was incredibly complex, making it hard to work with. In addition, developing a working supply from the bark of a rare, slow-growing tree -- one that was somewhat inaccessible and that died when its bark was stripped -- was a serious problem. Other tests of the compound at NCI showed only modest anti-cancer activity. Taxol was shelved. But Wall remained convinced. For years, he tirelessly but unsuccessfully campaigned to get NCI to reopen the investigation into Taxol. In 1976, he finally found a supporter in the new head of NCI's plant and animal products section. A year later, Taxol got its second chance. It was sent out for further study to Susan Horwitz, a molecular pharmacologist at the Albert Einstein College of Medicine in New York. She and her graduate student assistant Peter Schiff ran some tests and couldn't believe their initial results. "We were worried that maybe there was something else funny going on . . . contamination . . . some other compound mixed in with it that was responsible, not the Taxol," says Schiff, now chairman of the department of radiation oncology at Columbia-Presbyterian Medical Center in New York. Further tests were run, and it was clear: Taxol attacked cancer cells in a way never seen before. Unlike other drugs, Taxol attached itself to microtubules in the tested cancer cells. Microtubules are an important structural element of the cell essential to cell division. They are like tiny rods that form the cell's skeleton. For a cell to divide, these microtubules must disintegrate so the cell can collapse into two. In cancer, cells divide uncontrollably. Other cancer drugs simply prevented these microtubules from forming or stunted their development. Taxol did the opposite, it allowed microtubules to form, but not break down. Instead, Taxol bound to them, preventing their disintegration and halting cell division. By clogging cells with microtubules, Taxol paralyzed the division process, and the cells died. The discovery ignited intense interest. But it wasn't until 1983 that Taxol could be tested in humans. When the human tests began, though, the drug nearly died again. Patients experienced shortness of breath, irregular heartbeats and sudden drops in blood pressure. Others had to be resuscitated and there was one death. Alarmed, NCI halted the testing and called a conference of clinical investigators. "The question was, did anybody think it was worth continuing to try and develop the drug? Nobody wanted to do it," recalls Peter Wiernik, one of the doctors who had been running some of the Taxol testing at Albert Einstein. "NCI was about ready to stop the studies . . . I basically begged them to allow us to go on." Convinced of Taxol's promise despite its complexity and its supply problem, Wiernik devised a new regimen for giving the drug that involved premedication and eliminated many of the problems. NCI allowed testing of Taxol to continue. Amid these trials, William McGuire, then a physician at Johns Hopkins Hospital in Baltimore, was approached by a woman with advanced ovarian cancer. Every drug had failed. Her cancer had grown so large that her abdomen was grossly swollen. He gave her no more than a month to live. He was running tests on a possible anti-cancer drug, he told the woman. There could be no guarantees. No one knows if Taxol works, he told her. It's probably not a cure. Did she want to participate? "Two weeks after her first course of therapy, one of her sons called and said his mom was much better," McGuire recalls. "I saw her shortly thereafter and could not detect any tumor. You don't have to be a rocket scientist to figure out that's significant. It basically totally disappeared." Taxol gave her an extra year of life before her cancer killed her. Similar responses were coming back to NCI from other investigators. McGuire's hunch was that Taxol would be far more important than anyone realized, "one of those broadly active drugs. We get one of these once every 10 years." Indeed, in 1988, McGuire found that Taxol had a 30 percent response rate (cases in which tumors shrank by at least half) in patients with refractory ovarian cancer, which had resisted previous treatments. Other drugs had response rates of less than 15 percent, at best. There was no longer any doubt about the importance of Taxol. NCI was besieged by telephone calls and emotional letters from desperate cancer patients, their doctors, members of Congress and others, all pleading for Taxol. Now, the issue was not whether this drug worked, but how to get more of it. NCI estimated that 12,500 women with refractory ovarian cancer would need the drug each year; it only had enough of the drug to treat 500. The supply problem had now become a supply crisis. Sam Broder had just been sworn in as NCI director in January 1989 when the Taxol crisis hit. As a pioneer AIDS researcher at NCI, Broder did work that led to the some of the very first AIDS treatments. He understood all too well the desperate pleas of patients facing a deadly disease. "You cannot have a central government agency in effect forcing people to beg for their lives," he says. "It's okay to beg for a driver's license, but you should not have to beg for your life." Taxol quickly became his number one priority. The agency mobilized a full-bore effort that insiders dubbed their "Manhattan Project," after the intense scramble to build the atom bomb. They were going to tackle the one thing that stood between dying cancer patients and this life-prolonging drug, the critically short supply of it. NCI signed deals with the U.S. Forest Service and the Bureau of Land Management, which controlled the federal lands where most Pacific yew trees grew. It contracted with Hauser Inc., a Colorado company, to coordinate collection of the tree bark and process it into Taxol for the clinical trials. NCI held a conference of researchers -- chemists, botanists, foresters, oncologists and others -- to discuss, brainstorm and share information. NCI made an emergency request to find alternative supply options for Taxol -- from new natural sources to synthetic ones -- because the yew bark supply problem was significantly slowing the drug's development. And on August 1, 1989, NCI advertised for an industry partner to assist it with the further development of Taxol. The request produced a decidedly unenthusiastic response. The private sector was leery of the risk -- especially the supply problems, the unlikelihood of a patent in the future (since Taxol had been in the public domain for so long), and the small market of ovarian cancer sufferers. Of 20 companies that inquired, only four offered proposals. Bristol-Myers Squibb, an international conglomerate that dominated the cancer drug market, was selected. A Fortune 500 mainstay, Bristol-Myers Squibb sells everything from Enfamil baby formula to Clairol hair coloring. Last year, its total company sales were $16.7 billion. Cancer drugs sales were $2.4 billion, of which Taxol accounted for $941 million. In many ways, Bristol was the only logical choice to be the Taxol developer. The company had a stellar reputation within the oncology community, enormous resources, development expertise and a long history of partnerships with NCI. As NCI pioneered other new cancer treatments, Bristol was right behind, manufacturing and selling whatever NCI discovered and developed. In fact, this nexus with NCI was crucial to Bristol. Vincent DeVita, now at Yale Cancer Center, recalls asking Bristol executives early on whether they thought NCI and the federal government should get out of the cancer drug development business. "They would turn pale because they were just trying to start up . . . Without the NCI's screening programs, they would have had to set up their own screening program. Without NCI's toxicology programs, the investment would have been so enormous that they couldn't possibly convince their accountants that they should do something in it." Of the 15 cancer drugs sold today by Bristol, none were discovered by the company. Fourteen were developed by NCI or clinicians working on contract for it. Bruce Ross, former head of Bristol's U.S. Division, says the reasons for the company's great success in the cancer market are "very simple: Back in the early days, the late 1950s-early 1960s, they recognized a large unmet need and decided to consciously create a business. They created this business by developing other people's discoveries." While "the field was born at the NCI," DeVita says, Bristol made it work. The Taxol partnership was simply a continuation of the many successful collaborations NCI and Bristol had behind them. Bristol had been watching Taxol's development from the beginning. But, like other companies, it was put off by the obstacles. When NCI trials showed the drug's unexpected ability to fight ovarian cancer, though, Bristol's interest was piqued. Dianne DeFuria, then Bristol's liaison with NCI, had been watching Taxol with keen interest and recalls that the "awareness grew and the desire to have the product grew." When NCI announced that it was seeking an industry partner, there was a long debate inside Bristol. "The internal debate came down to the fact should we or shouldn't we?" DeFuria says. "There were good arguments on both sides: This drug appears to have promise; NCI wants to take it forward. They need a partner. If the NCI and BMS can't do it, then it can't be done. The oncology community . . . was looking to Bristol to be a leader . . . It came down to the moral question of, Do we do it?' Ultimately, that was the question that was answered rather than, Can we do it?' and, What are the patient population response numbers?' " The negotiations with NCI took more than a year as Bristol worried about whether it would be able to find a renewable source for Taxol or come up with an alternative manufacturing method to end reliance on the yew tree altogether. In the meantime, newer studies showed that the remarkable responses seen in ovarian cancer also occurred with breast cancer. Initial results showed a 50 percent response rate for advanced breast cancer. For Bristol, this meant the potential market for Taxol had grown from a little more than a hundred thousand people to more than a million. (Just this month, Taxol was found to dramatically increase the survivability of women with early breast cancer.) In January 1991, a deal finally was struck. NCI would continue development of the drug by sponsoring additional clinical trials, supporting research on alternative manufacturing methods and helping secure access to the yew bark. It would also provide Bristol exclusive access to all NCI research necessary to win FDA approval for Taxol and right of first refusal for all future NCI Taxol-related inventions -- giving Bristol something close to patent protection for the drug. In return, Bristol agreed to handle manufacturing, provide increased supplies of Taxol for more clinical trials, assist NCI in planning additional research and actively seek synthetic and alternative sources of the drug. Bristol would give NCI free supplies of Taxol. The company agreed to spend $65 million to $114 million on these efforts. Finally, it promised to acknowledge the public's investment when setting a "reasonable" price. Six months later, Bristol received virtually exclusive access to all Pacific yew trees on federal land. The company would not pay directly for the yew bark, but would reimburse the Agriculture and Interior departments for costs associated with allowing the collection and inventorying of the Pacific yew tree. Bristol would work with Hauser, which had been extracting and purifying Taxol for NCI since 1989, paying it $100 million to meet the increased Taxol demands, which required Hauser to build a new facility. Hauser would collect the bark from the forests, and manufacture the white Taxol powder that Bristol would then make into liquid Taxol. The company moved fast, mustering its massive resources and widespread expertise, and beating every time line set out in the agreement with NCI. On July 22, 1992, 30 years after Blum first stripped bark off the yew tree, Bristol asked the FDA to approve Taxol for the treatment of refractory ovarian cancer. Five months later, it was a done deal, the fastest regulatory approval up to that point for a cancer treatment. That very morning, representatives of NCI and Bristol gathered in a third-floor conference room of NCI's Bethesda headquarters. The meeting had begun in a celebratory mood, but it quickly vanished once the topic at hand was broached: What price? The group had been meeting all summer. Now, as its members sat around an oversize table, Bristol representatives began their final presentation. The company had already invested more than $100 million at this point, they said. Taxol would have gone nowhere if Bristol hadn't gotten involved, a point with which the NCI officials readily agreed. The drug company deserved to not only make back its investment, but also to profit from its gamble. "Taxol was developed in the early 1960s and languished for almost 30 years because nobody could make it" until Bristol came along, says Bruce Ross, who was the company's lead negotiator that day. NCI officials were unhappy to be put in a position of trying to set a price for Taxol. They felt pricing was not NCI's job: It lacked both expertise to know what a fair price was and power to enforce it. The agreement NCI and Bristol had signed in 1991 stipulated only that Bristol would acknowledge the public's investment and NCI would be sensitive to the difficulties, expense and lack of straightforward patent protection that could make Taxol vulnerable to competition. The most important thing, the NCI officials felt, was that Bristol would get the drug out to people who otherwise might die. But, now, Congress was pressuring NCI to do more. Lawmakers already had suggested that Bristol had gotten too much, and NCI -- and the public -- too little. With Taxol set to come to market, they wanted to remedy that. They wanted a "reasonable price" -- the buzzword on the Hill that summer -- that all Americans could afford. In their previous meetings, NCI had asked Bristol for documentation to justify its pricing but had gotten no response. NCI had given Bristol a list of 15 recently approved cancer therapies and their costs, and asked that Taxol be priced somewhere in the middle -- more specifically, at the mid-range of other ovarian cancer drugs. It also requested that Bristol provide the drug free to those who needed it but lacked insurance and could not afford it. Bristol came back with a price of $4.87 per milligram, which company officials said would mean that a full treatment would cost a patient from $3,000 to $6,000. But that was the high end, company officials said. When this price was averaged out with various discounts, and with the free drugs Bristol promised to give to the poor through its indigent care program and to NCI for continued testing, company officials said, then the "net effective weighted average cost" of Taxol would be much less, about $3.43 per milligram, which would mean about $2,000 to $4,000 for a full treatment. NCI officials were elated. According to their calculations, this would place Taxol just below the cost of carboplatin and just above cisplatin, the two standard ovarian cancer therapies at the time that were also produced by Bristol (and initially developed by NCI). The company had come up with a price NCI felt it could justify to Congress and one that would get Taxol out to virtually all of those in need. As the meeting broke up, everyone seemed to breathe a great sigh of relief. Drug pricing, however, is a complicated business. There are hundreds of different prices paid for the same drug, depending on who is buying and what can be bargained. By law, certain government agencies get price breaks, but those breaks differ from one agency to the next. Medicare, for instance, the largest government purchaser of Taxol, until recently got no special discount. At the hospital level, a large institution or a chain might pay less for a drug than a smaller competitor across town. There are also two ways to purchase pharmaceuticals, through a wholesaler or directly from the pharmaceutical company, which can be cheaper. Then there's the average wholesale price (AWP), which, like the sticker price on a new car, provides the opening bargaining point between wholesalers and hospitals or doctors. Bristol, like other pharmaceutical companies, will not discuss the specifics of how it prices its drugs, including Taxol, saying such matters are proprietary and confidential. But, in general, company officials cite the extremely high risk, and high costs, associated with discovering and developing new drugs and say that without a payoff at the end, these drugs would simply never happen. For an individual patient, the cost of a course of therapy will depend on the price the hospital paid for the drug, the markup taken by the hospital or a physician, and the amount of drug required for that patient's weight and particular cancer. In determining for NCI its price for Taxol, Bristol used a woman of about 105 pounds for its calculations, according to various hospital pharmacists consulted for this article and testimony presented to Congress by the Taxpayers Asset Project, part of the public advocacy Center for Study of Responsive Law. It also calculated its price based on a smaller-than-average dosage for ovarian cancer, according to these same sources. Finally, it based its price on what the drug would cost if purchased directly from Bristol -- with no wholesale markup, no hospital markup and no physician markup. In truth, there are very few patients who would fit this profile. The price an average woman -- weighing 140 pounds and receiving not a small, but an average dosage -- was likely to pay was as much as three times higher. For this woman, the price for a complete treatment could easily cost more than $10,000, according to the hospital pharmacists contacted. A similar woman, receiving a cisplatin or carboplatin treatment would pay several thousand dollars less, these pharmacists said. Taxol, it turns out, broke a threshold; it was the single most expensive cancer treatment on the market at that time. The high patient price for Taxol came at a time when the cost of making the drug was, in fact, falling. Through continued production improvements, Hauser had increased dramatically the Taxol yield from the yew bark and was now making it for Bristol at about 25 cents a milligram. This was less than half what it had cost to produce the drug for NCI's earlier development program, according to Hauser's president, Dean Stull. At 25 cents a milligram, a complete ovarian cancer treatment for the average woman, direct from the manufacturer, would be as little as $560. So how did Taxol go from 25 cents a milligram (the price Hauser charged Bristol) to $4.87 a milligram (the price Bristol told NCI it would charge), a price that in any case was much lower than the AWP of $6.08 a milligram (closer to the price a patient would actually have to pay)? Says Andrew Bodnar, Bristol's vice president of medical and legal affairs, "We're not going to talk about our manufacturing costs . . . not gross margins and their relationships to our pricing on any of our products." But, he adds, "we don't pick a number out of the air. The price was fair, is fair and in all circumstances is appropriate." Manufacturing costs are typically the smallest part of the pricing equation, according to a 1993 study of the pharmaceutical industry by the now-defunct federal Office of Technology Assessment. Research and development -- finding a possible compound and honing it into a testable drug -- constitutes a large share, but it is also the most mysterious. During the 1993 congressional hearings on the price of Taxol, now-retired Bristol vice president Zola Horovitz said the company had spent "far in excess of the $114 million and committed more than 205 man-years" to the development of Taxol, even after all of NCI's work. Bristol officials declined to provide any details about the spending on Taxol. Today, company officials assert that Bristol has spent $1 billion on Taxol since the 1991 agreement between Bristol and NCI (included in that amount is as much as $30 million in "royalties" Bristol agreed to provide NCI -- primarily in the form of free Taxol and help with ongoing human testing -- when the 1991 agreement was extended in 1996). But again, Bristol declined to provide any details, including profit figures for the drug. Bodnar says it's "one of the myths" that the company "was handed a sure thing by the government . . . It was a huge effort not predicated on the knowledge that we were going to succeed and ultimately recoup our costs. It was a total commitment by the company to do something that was -- by any measure, I think, in the annals of pharmaceutical development -- unique." Others have argued that the largest risks were the government's. By the time Bristol became involved, the government had already spent close to 30 years on Taxol -- finding it, isolating its activity, manufacturing it on a small scale, and testing it in the lab, in animals and in humans. Countless man-years and about $32 million had been spent through 1992, according to NCI. (More has been spent since then, though NCI did not provide specifics.) But those figures don't begin to describe the real cost -- and risk -- assumed by the government. "A break needed to be afforded to the people who sort of leveraged this, because it isn't all Bristol-Myers risk. We put money into it. It's our risk, too," says Graydon Forrer, who was one of the congressional investigators during the 1993 Taxol hearings. "Bristol-Myers is doing a lot of research, spending a lot of time and their own money, but so are we -- at the taxpayers' expense," Forrer says. "Those scientists at NCI are being paid for with federal dollars. Their laboratories are being paid for with federal dollars. And, to the point that they {NCI} support anything else that Bristol-Myers does in a cooperative venture, there's the upfront money that goes into the grants. There's the upfront money that goes into making the cooperative agreement work, but then there's also all the federal structure behind it that the taxpayer, the American public pays for. There's a substantial investment here." Similarly, once the 1991 agreement with Bristol was signed, the government played a major role -- providing unfettered access to Pacific yew bark, conducting large-scale clinical trials, helping to speed the FDA approval process and aiding in a search for alternative sources of Taxol. Although NCI officials uniformly agree Taxol couldn't have made it to market without Bristol, they just as strongly believe that Bristol, or any other company, would never have taken the risk and spent the money in the first place to pursue a difficult compound like Taxol. "The companies would have dropped it like a hot potato," says Saul Schepartz, a retired NCI official who was involved in cancer drug discovery work for decades. "They may say something different, {but} they're constantly looking at their overall priorities, and the technical difficulties with Taxol are such that, in my judgment, they would not do it." In any case, the supply issue that so bedeviled the development of Taxol, and was the main reason it was considered such a daunting risk, became moot in 1993, when Bristol announced it would be making Taxol semi-synthetically. It had licensed a manufacturing method to do so from a scientist long funded by NCI. Soon, Hauser was no longer involved. From this point on, the drug would be made through semi-synthesis; within two years, the company was no longer reliant on Pacific yew bark at all.

Last December 29, Bristol lost its five-year right to be the exclusive seller of Taxol, a limited monopoly granted to all new drugs under the Waxman-Hatch Act. It was a moment much anticipated on all sides. NCI officials had counted on this being the beginning of a market-based solution -- in the jargon of economists -- to the problem of Taxol's high price. Indeed, Bristol's competitors were lining up to grab a part of the nearly billion-dollar world market for Taxol. Bristol officials, meanwhile, were fearful of losing the U.S. monopoly on their blockbuster -- having recently lost it in a number of other countries, including Canada and China. In those countries, competition had brought a noticeable drop in the price of Taxol and of new, generic versions of the drug that had sprouted. But before anyone could get a foot into the U.S. market, Bristol threw up legal roadblocks at every market entrance. Although Taxol itself cannot be patented, the way it is used -- infused into a patient over a three-hour period, for example -- can be. So this is what Bristol did. Each variation in usage requires extensive testing for FDA approval. Bristol claimed that all of the most common and proven ways of treating cancer patients with Taxol were Bristol's exclusively. Just before its exclusivity was to expire, Bristol received 17- and 20-year patent protections for use of Taxol. NCI's years of research on Taxol became part of Bristol's patents. That meant that competitors would have to conduct their own testing from scratch, a process that would take years. Immunex Corp., a Seattle company already selling a generic version of Taxol in Canada, had been gearing up to enter the U.S. market. Last August, Immunex asked for FDA approval of its Taxol equivalent, despite the patents already given Bristol. Bristol responded by suing Immunex for patent infringement, and the matter is now in court. The FDA, following standard procedure, put a 30-month hold on its decision on Immunex's request. Another competitor, Florida-based IVAX, also failed. A small company, it hoped to gain FDA approval for its version of Taxol, called Paxene, as a treatment for Kaposi's sarcoma, a cancer common to AIDS patients. It already had tested the drug on AIDS sufferers and was conducting tests on breast and ovarian cancer patients. IVAX was hoping that once Paxene had received approval as an orphan drug -- the designation given to drugs used for illnesses affecting small populations -- it would then be prescribed by doctors for the much larger breast and ovarian cancer market. But Bristol was not to be outsmarted. It filed its own orphan drug application for Taxol more than a month before IVAX, and got FDA approval for treating Kaposi's sarcoma. Because orphan drug approval gives a company a seven-year monopoly, IVAX was locked out. Much of this is classic business maneuvering to defend market share. Any smart company would do it. As Bodnar says, "We would be remiss if we didn't actively explore, at all times, life-cycle management of all our products. Every pharmaceutical company that is not doing that is not only making big mistakes, but is exposing itself to trouble. We are determined to find ways where we can . . . extend the exclusivity. When a product loses exclusivity, it is difficult to maintain the research support behind it." What makes this more than a simple display of corporate might is this: Although Taxol's price has not gone up since the drug came onto the market in 1992, neither has it come down. Among those who worked on Taxol at NCI, there is little second-guessing about the drug's price. "If there's going to be a problem, this is a good problem," says Broder, the former NCI director. He is now working for IVAX, overseeing the development of its version of Taxol. "I don't mean to be misunderstood. The real problem we had was: How are we going to deal with a drug that the government has produced and can produce in small quantities that lots and losts of people want and lots and lots of people are asking for? Those who criticize whatever steps were taken at the time have to recognize that was what we faced." So what about D.C. General and Margaret Akpan's patients? The public hospital has always had the toughest job of any hospital in the area: treating the poorest of the District's poor. Its mandate is to care for all who seek it and, so, no one is ever denied. More than 50 percent of care given by D.C. General is unreimbursed -- no one pays those bills. Money is always tight there and high-priced pharmaceuticals are an enormous strain. As a result, Taxol, the drug of salvation and promise for so many, is not on D.C. General's formulary, a list of approved drugs the hospital's doctors turn to when they prescribe drugs for their patients. In every hospital, the formulary is king. A formulary is based not just on what's best for patients, but on what a hospital can afford -- and D.C. General cannot afford Taxol. When other chemotherapy treatments have failed, a special order of it can be called up, but at that point the cancer may have spread and become more resistant. Robin Newton, D.C. General's head of medicine, says the hospital must weigh the benefits of treating one person with Taxol against the cost of not having that money to treat others who face equally life-threatening conditions -- diabetics or heart patients or asthmatics. "We are forced to make choices . . . Would you put an expensive drug on the formulary, or would you put a lot of inexpensive drugs that may have a greater impact? Take Taxol: It's expensive. Lifesaving? Yes. But a lot of the other drugs are very inexpensive . . . These are the choices that you end up with. They're not pretty. Those people who think we don't ration care really have no sense of reality. We do it every day here. We ration every day." Bristol officials say the company's compassionate, or indigent, care program supplies millions of dollars' worth of Taxol to needy patients at places like D.C. General. But as far as Newton is concerned, these programs are too cumbersome and time-consuming for an understaffed hospital like hers to use. "Understand this: They are very labor intensive," she says. "I don't think {companies with indigent care programs} have in mind entire institutions that serve hundreds upon thousands of people who have nothing." Akpan, pen and patient folders in hand, is doing what she can. Before her on a recent spring day sits a middle-aged woman who has breast cancer. She has just had a radical mastectomy. If this were another, richer hospital, the woman probably would get Taxol now. But at Unit 52 she is getting something different. Akpan hopes for the best.. Li Fellers is freelance journalist who lives in Washington. CAPTION: PRICED OUT: Margaret Akpan, a doctor at D.C. General Hospital, would like to give her patients Taxol. But, for a cash-strapped public hospital, it's simply too expensive. CAPTION: TAXOL'S BIRTHPLACE: The drug was developed at the NCI lab at Fort Detrick in Frederick, which is now using a similar process, opposite page, on a host of newer cancer drugs. Upper left: extracting a new compound. Lower left, upper right: raw materials and extract in cold storage. Lower right: adding extract to cancer cells to test it. CAPTION: TAXOL FOOT SOLDIER: In 1962, when he was a 22-year-old federal government technician, Kurt Blum stumbled upon the tree bark that would eventually yield the No. 1-selling cancer drug in the world. He is now a biology professor at Middle Tennessee State University. CAPTION: THE SOURCE: Taxol was discovered in the bark of the Pacific yew tree.