In his Manhattan office, physician Nicholas J. Gonzalez treats cancer patients with a nutritional regimen that relies on coffee enemas, a restricted diet and as many as 160 vitamin and enzyme capsules a day.

He doesn't use standard tests to monitor the treatment's effectiveness. Instead, he sends a sample of a patient's hair to a Louisiana laboratory, where it is analyzed to derive a "total cancer" score--supposedly a measure of the body's tumor burden.

Some cancer specialists believe Gonzalez's treatment is ludicrous, but officials at the National Cancer Institute aren't scoffing. In partnership with the National Center for Complementary and Alternative Medicine, a division of the National Institutes of Health, the institute is sponsoring a five-year, $1.4 million study of the treatment. It is expected to enroll as many as 90 patients with pancreatic cancer, one of the disease's most lethal forms.

Gonzalez's ability to win federal backing is testament to the growing political and economic clout of "alternative medicine" and to the increasing patient demand for treatments that defy traditional medical techniques. Alternative medicine is now a thriving business, despite concerns about the cost, safety and effectiveness of many of the treatments.

Some question why, with federal research funding so hard to come by, officials are willing to test a treatment pushed by Gonzalez--a doctor who lacks traditional training in cancer treatment and has been disciplined by a state medical board.

Congress has been under increasing pressure from alternative medicine advocates to fund research into unorthodox treatments. Several years ago, it set up an office at the NIH with a budget to be specifically used for this purpose.

Yet many practitioners have avoided scientific scrutiny of their work. In that respect, Gonzalez's willingness to have his theories tested by independent researchers is unusual, according to experts in the field.

In his campaign to win federal support, Gonzalez got a boost from Rep. Dan Burton (R-Ind.), the chairman of the House Government Reform and Oversight Committee and a longtime alternative medicine ally. Burton held a June 1998 meeting in his office that brought together Gonzalez and NCI Director Richard Klausner to discuss a study of the regimen.

"Klausner was very aware of the work . . . and had no problem at all saying the NCI would support it," Gonzalez recalled. "That took care of everything." Klausner said he did little more than offer "moral support" to a project for which funding was already underway.

Burton said he has never "dictated, nor would I try to dictate, to the NIH or the FDA [Food and Drug Administration] on how they run their business. But if there's something that's been successful, we have tried to bring that to their attention and ask them if they would look into it."

Gonzalez's theory that cancer should be treated with a dietary regimen is similar to so-called "metabolic therapies" available in Mexico and Europe that attract thousands of American patients each year.

But his care of six advanced cancer patients brought harsh criticism from New York regulators in 1994, who found the doctor negligent and ordered him to undergo retraining. As the patients' conditions worsened, the board found, Gonzalez failed to order tests or X-rays and sometimes disregarded evidence of tumor spread.

A Manhattan jury concluded in 1997 that Gonzalez incompetently treated uterine cancer patient Julianne Charell of New York, who was on his regimen for eight months and eventually went blind and suffered spinal damage. The jury assessed damages of $4.8 million--eventually reduced to about $2.3 million.

Charell's lawyer Jeffrey Bloom called it "outrageous" that "the government is paying for this man's fraudulent medical care. . . ."

But officials at Columbia University's cancer center, where the study is being carried out, and at the NCI say his work merits review. They cite a pilot study conducted by Gonzalez that found 11 pancreatic cancer patients who followed his regimen lived longer, on average, than most people with this type of tumor.

"It's not him, it's the treatment" that's being evaluated, said Karen Antman, the center's director. "He believes this treatment is effective. He is willing to have it tested in the most rigorous way. Why is that bad?"

Enzymes and Enemas

Gonzalez became interested in nutritional cancer treatments--an idea now several decades old--at Cornell Medical School in the early 1980s. He was intrigued by the work of William D. Kelley, a Texas orthodontist who advocated a regimen that included coffee enemas and pancreatic enzyme capsules.

Gonzalez began evaluating Kelley's regimen. He set his research aside to take on a three-year residency in internal medicine at Vanderbilt University but dropped out of the program after a year. He never completed his residency, an in-hospital training program that is considered standard preparation for medical practice.

He says he eventually reviewed the medical records of 1,306 of Kelley's cancer patients and wrote a report, which was never accepted for publication in a medical journal. Unable to secure more research funding but convinced the treatment was effective, he returned to New York and began treating cancer patients with a modified version of Kelley's program.

For each patient, Gonzalez devises a prescribed diet--for instance, plenty of red meat for people with leukemia and lymphomas (cancers of blood cells), and a mostly vegetarian diet with some fish for pancreatic cancer patients. Patients also are asked to swallow as many as 160 pills a day, a grueling routine that includes some middle-of-the-night doses. About half of the pills are vitamins and minerals; the rest contain pancreatic enzymes, chemicals sometimes prescribed as a digestive aid.

Gonzalez believes that the enzymes are absorbed into the bloodstream and help the immune system fight cancer. He acknowledges he has no solid scientific evidence for this theory.

Gonzalez's treatment includes at least two coffee enemas a day to cleanse the body of unspecified "toxins." Patients "almost universally say they feel better when they do the enemas," he said.

His treatment is not cheap and is rarely covered by insurance. Gonzalez said he charges $2,000 for an initial consultation and $70 to $100 for follow-ups. Enzymes and vitamins cost about $6,000 to $7,500 per year.

Researchers familiar with Gonzalez's regimen said that it doesn't appear to be dangerous, since there were no fatalities due to the treatment among the pilot study's patients. Side effects include fevers, chills, rashes, nausea, vomiting or muscle aches--which Gonzalez attributed to "detoxification."

Gonzalez generally doesn't order X-rays or scans, convinced his therapy can produce false readings. Pancreatic enzymes, he said, "can cause an inflammatory reaction around tumors which can actually make tumors swell." A visibly enlarging tumor doesn't mean the therapy isn't working, he maintains.

That notion is disputed by NCI Chief of Surgery Steven A. Rosenberg, who said he knows of no effective cancer therapy that makes tumors look bigger, rather than smaller. Tumor growth is generally a sign of treatment failure.

Gonzalez insists his only goal is helping patients. He said he usually turns away patients who might be cured by surgery, chemotherapy or radiation. His patients, he said, usually have advanced cancer and have exhausted other options.

"We never tell them what to do or what not to do," he said.

Gonzalez said he will accept a patient considered a good prospect for chemotherapy--if the patient assures him that he or she will refuse such treatment.

"I have two patients with testicular cancer who refused chemotherapy," which cures 80 to 90 percent of such tumors, he said. He said he informed them of chemotherapy's success rate, and "made them write letters and say it was their own decision."

One highly controversial feature of Gonzalez's program is his use of a hair analysis test to devise a "total cancer" score. The score, he says, measures how much cancer is present in a patient's body. He tells patients that a score of zero to five is normal; six to 10 signifies a few cells; 11 to 20 is "moderate"; 21 to 40 signals "significant cancer activity" and above 40 means "a lot of cancer." Patients are given a score at the beginning of treatment and the analysis is repeated periodically.

Gonzalez said the test was developed in England and he couldn't explain how it works. "I'm not a physicist," he said. "Basically, the molecular structure of hair changes depending on the disease. . . . I think that looking at any tissue sample, even a hair sample, you can really tell whether a patient has cancer or not and even how much."

Joan McLure, the hair analysis technician, also declined to explain the procedure. "I'm just a little guinea pig in a big operation," she said. "It is FDA-approved, as far as the latest technology is concerned."

But an FDA spokeswoman said the agency has never approved such a test. "That sounds too good to be true and it is," she said.

Gonzalez's novel procedures led to the New York state medical board's 1994 order that he be retrained. Gonzalez, who believes he was the victim of the board's bias against alternative practitioners, said he has completed the required retraining. But state licensing officials said he is still on probation until he completes 200 hours of community service.

Charell's lawsuit against Gonzalez charged that he discouraged her from using conventional cancer treatments and assured her that her cancer score was decreasing on his regimen.

Based on his advice, she said she declined radiation treatments and chemotherapy. Her oncologist, James F. Holland of Mount Sinai School of Medicine, testified that her tumor might have been arrested. Charell declined to be interviewed.

Gonzalez maintains that Charell's cancer had spread to her spine by the time she started his treatment, and that he extended her survival.

Victor Herbert, a professor of medicine at New York's Mount Sinai School of Medicine who testified on Charell's behalf, said, "People want to believe in magic. He's selling magic."

Herbert said he considers federal funding for a study of Gonzalez's regimen "a terrible waste of money."

Trials and Truth

The most rigorous analysis of Gonzalez's treatment was done by Gonzalez himself--along with his partner Linda L. Isaacs. The work was conducted with an advisory committee's supervision, and results were published last spring in a little-known medical journal, Nutrition and Cancer.

Gonzalez's findings were startling, suggesting that patients who used his regimen could extend their life expectancy. It was hopeful news for patients with pancreatic cancer, who usually live only five to eight months after diagnosis. Only 25 percent survive a year. The disease kills 28,600 Americans annually.

Gonzalez reported that nine of his 11 patients, or 81 percent, survived at least one year, and five, or 45 percent, survived at least two years.

Although the pilot study was convincing enough to help Gonzalez win federal attention, a prominent cancer expert who reviewed the published report said it is flawed and reaches questionable conclusions.

It's not clear that all 11 subjects had pancreatic cancer, said Robert Mayer, director of the center for gastrointestinal oncology at Boston's Dana Farber Cancer Institute, former president of the American Society for Clinical Oncology and an associate editor of the New England Journal of Medicine.

With three patients, Mayer said, the descriptions of their conditions made him "strongly question whether all those people had cancer of the pancreas."

Furthermore, he added, Gonzalez failed to closely monitor patients' weights or their adherence to the regimen. His study participants--some of whom traveled long distances to take part--were apparently much healthier than the average pancreatic cancer patient. And in compiling final results, Gonzalez did not count other patients who dropped out of the study after a few weeks. Mayer said this skews the results and violates accepted research practice.

"This is standard Clinical Trials 101," Mayer said.

Gonzalez said he was confident that all patients had pancreatic cancer, and he defended the decision not to count patients who followed his regimen only briefly.

"Pilot studies routinely have a lead-in period to eliminate people who decide not to follow through," he said. "The point was to see whether anybody, no matter how many patients I treated, lived one year. Anybody who lives that long [with pancreatic cancer] is amazing."

George Blackburn, who directs the Center for the Study of Nutrition and Medicine at Boston's Beth Israel/Deaconess Medical Center, served on the pilot study's supervisory committee. He acknowledged that the committee did not independently review patient tissue samples to confirm their diagnosis.

When the journal article appeared, many who read it "got very angry," said Leonard A. Cohen, a cell biologist and the current editor of Nutrition and Cancer. "When I saw it in the journal, I was as shocked as anybody," he said.

But former editor Gio Gori said Gonzalez's article went through the normal "peer review"--a critique by independent researchers. Gori said he could not guarantee that all patients in the study had pancreatic cancer.

"No peer review can guarantee you that," Gori said. "Keep in mind, we have not published this paper here with any pretense of scientific accuracy. It was presented as a way of stimulating interest."

A Federal Boost

Gonzalez's treatment also sparked congressional interest. In 1998, a Burton committee staffer heard Gonzalez speak on alternative cancer treatments and told her boss. Gonzalez at the time was urging Columbia University researchers to conduct a larger trial of his regimen.

"I have been interested in alternative medicine . . . for a long, long time," said Burton, whose wife received alternative treatment several years ago while also undergoing chemotherapy for breast cancer.

Burton held 10 hearings in two years on alternative medicine, questioning the commitment of the National Institutes of Health to studying such therapies. He was impressed with Gonzalez's efforts to cooperate with the research establishment.

A month after Burton convened the June 1998 meeting with Klausner and Gonzalez, Klausner mentioned during testimony at a Burton hearing that the NCI planned to cosponsor a trial of Gonzalez's regimen. The funding, he said, would come from NCCAM, the division of NIH created in 1992 to study alternative medicine.

Klausner said that before Burton arranged the meeting, Columbia researchers were already planning a study and had been assured of NCCAM funding.

"Congressman Burton has been consistently concerned about whether we are supportive of doing clinical trials of alternative medicine," Klausner said. "I didn't do anything other than meet with [Gonzalez] and was asked whether I would be supportive of such a trial."

Two NCI officials helped supervise Columbia's study design, and the NCI continues to provide oversight, said Jeffrey D. White, director of NCI's office of cancer complementary and alternative medicine.

But several researchers suggested that political pressure, not scientific merit, prompted the federal decision.

"We would fall over dead to get a trial like this," said Boston's Blackburn.

Unlike the cancer institute, which gives funding priority to the most promising therapies, NCCAM's budget ($68 million for the current fiscal year) is dedicated to alternative research. Current projects include studying shark cartilage for the treatment of lung cancer, green tea as a cancer preventative and St. John's wort as an antidepressant.

"That budget, by congressional mandate, has to be spent," Mayer said. "The real issue is, what were the pressures that led to that office being formed in the first place?"

Push for Legitimacy

Purveyors of unorthodox cancer therapies have often resisted scrutiny, but Gonzalez's fight to win scientific review has earned him the respect of some alternative medicine experts.

"If you think of him as one of the large cadre of individuals practicing unproven methods. . . . at least the guy is collecting data," said Barrie Cassileth, chief of the integrative medicine service at Memorial Sloan Kettering Cancer Center.

Efforts to recruit patients for the Columbia trial began early last year. But the researchers ran into an unexpected problem: patients' strong aversion to chemotherapy. Under the study's design, between 60 and 90 patients with inoperable pancreatic cancer are to be randomly assigned either to Gonzalez's nutritional regimen or to receive gemcitabine, the standard chemotherapy drug. Then, survival rates will be compared.

Such a study, called a randomized controlled trial, is considered the best way to judge the relative merits of two treatments.

But most patients have balked at even the possibility of chemotherapy. Of more than 200 cancer patients approached, only three have agreed to be randomly assigned to either treatment, said Chabot. The rest declined to enroll.

"They really come to us with a strong bent toward alternative medicine," Chabot said. "When they confront the potential to be randomized to a chemotherapy arm, they back away from the study."

Chabot said the researchers may be forced to redesign the study, enrolling all eligible patients in the nutritional regimen and matching them with comparable pancreatic cancer patients who are undergoing chemotherapy at hospitals.

He acknowledged that such a trial, called a "case-control study," could yield less conclusive results.

"It'll be easier to criticize and easier to poke holes in," Chabot said.

If the Columbia trial is not conclusive, researchers fear that more patients will spend money, time and energy on the Gonzalez regimen and similar treatments without knowing whether they work.

"If we don't do a large enough study that's well controlled," said the NCI's White, "we're going to get a muddy answer and we'll be worse off than we are now."

CAPTION: Nicholas J. Gonzalez's nutritional therapy for cancer, which relies on coffee enemas, a restricted diet and as many as 160 vitamin and enzyme capsules a day, is the subject of a five-year, $1.4 million federal study.

CAPTION: Morton Schneider, whose cancer has been in remission since 1991, when he began nutritional therapy, counts pills with his wife, Evelyn, at home in Casselberry, Fla.