Blogger’s note: This blog, which is rough going in places, will be presented in either 2 or 3 parts (I won’t know which until next week). I’ll post a part each week until it is complete, but due to overwhelming popular demand I promise to maintain the every-other-week posting of the far more amusing Weekly Waluation of the Weasel Words of Woo/2.


On Feb. 25, 2008, the federal Office for Human Research Protections (OHRP) cited Columbia University Medical Center (CUMC) for violating Title 45, Part 46 of the Code of Federal Regulations: Protection of Human Subjects (45CFR§46). The violations involved Columbia’s administration of the NIH-sponsored trial of the bizarre “Gonzalez Regimen” for treating cancer of the pancreas.† The OHRP’s determination letter to Steven Shea, MD, the Director of the Division of General Medicine and Senior Vice-Dean at CUMC, cited ethical problems of a serious kind:

We determine that the informed consent for the 40 of 62 subjects referenced by CUMC was not documented prior to the start of research activities, nor was the requirement for documentation waived by the CUMC IRB for subjects in this study.

It was the second time that the OHRP had cited Columbia for its dubious management of the “Gonzalez” trial. The first occurred in Dec. 2002, after investigators had determined that the trial’s consent form “did not list the risk of death from coffee enemas.” The OHRP listed several other violations at that time, but “redacted” them from the letter that it made available to the public.

The “Gonzalez Regimen”

The full name of the trial, which began in 1999, is:

Prospective Cohort Study of Gemcitabine Versus Intensive Pancreatic Proteolytic Enzyme Therapy With Ancillary Nutritional Support (Gonzalez Regimen) in Patients With Stage II, III, or IV Adenocarcinoma of the Pancreas

The trial ostensibly investigates a “detoxification” regimen advocated by Nicholas Gonzalez, a physician who has been found “guilty of negligence and incompetence on more than one occasion” and has been ordered to pay large, justified malpractice settlements on others. The regimen includes pancreatic enzymes taken by mouth—which, according to the National Cancer Institute (NCI) description, “may help kill cancer cells”—and the following:

Coffee enemas are performed twice a day, along with skin brushing daily, skin cleansing once a week with castor oil during the first 6 months of therapy, and a salt and soda bath each week. Patients also undergo a complete liver flush and a clean sweep and purge on a rotating basis each month during the 5 days of rest.

Writer Michael Specter described the regimen’s oral medications for an article in the New Yorker several years ago:

The list of supplements that Gonzalez hands out to most patients with solid tumors runs to four single-spaced pages. It includes, in part, sixty freeze-dried, porcine-pancreatic enzyme pills (swallowed in six batches, all of which must be taken with water, and none of which may be taken with food or within an hour of a meal). During breakfast and dinner, each patient must swallow capsules of adrenal medulla, amino acids, bone marrow, selenium 50, thyroid, Vitamin A 10,000, and Vitamin E succinate.

There are separate pills for the lymph, liver, and kidney–all to help balance deficiencies caused by lack of enzymes, or because the body is overwhelmed fighting cancer. During lunch, each patient must take a pill with twenty-five thousand units of beta carotene, as well as pills with copper gluconate, manganese glycerophosphate, potassium citrate, and Vitamin D. Twice each day, whenever it’s convenient, patients must dilute a mixture of black-walnut formula in water and drink it. Patients who suffer from metal toxicity also need to take nine pills of sodium alginate.

In all there are about 150 pills per day. The regimen is “alternative,” not “integrative.”

An “Eccentric” Dentist

Gonzalez inherited most of this regimen and his ideas about cancer from a dentist in Texas. He frequently recounts his experience, beginning as a medical student at Cornell, interviewing

…my mentor, Dr. William Kelley, the eccentric and controversial dentist who, over a 20-year period, developed a very intensive way of treating cancer using diet, supplements, large doses of pancreatic enzymes, the controversial detoxification routines, such as coffee enemas.


I learned about Dr. Kelley in 1981, after my second year of medical school at Cornell. I was fortunate at the time that I was already kind of being taken under the wing of Robert Good, who, at that time, was president of Sloan-Kettering Cancer Institute.


I had met Kelley in the summer of 1981. Again, the details of which aren’t important for this session. At that time, he was very controversial. He was just coming off of that Steve McQueen fiasco. Steve McQueen, the actor who did Kelley’s program.

Kelley was desperate to have his work tested. This is 1981, before the NCI had an Office of Alternative Medicine, before there was a National Center for Alternative Medicine, before any of those things were up and running.

It was a very controversial time to be treating cancer patients with nutrition, particularly if you were a dentist and not a physician, so there were all kinds of legal issues involved.

When I met Kelley, he expressed one interest and one intention. Whatever’s said about him or isn’t said about him, whatever people may think or not think about him today in 2000, in 1981 he had one intention. He wanted his work properly evaluated by the orthodox medical community.

He believed he was doing something of value. He also said in our first session that if he was doing something of value it needed to be tested properly so it could be mainstreamed into orthodox medicine where any big therapy belongs. I thought that was a very noble and honorable intention, and he maintained that intent through the next 5 years.

Under Dr. Good’s direction, initially, as a medical student, but subsequently as an immunology fellow in his research group which, at that point, eventually moved to the University of South Florida, I completed an intensive investigation of Kelley’s work.


We were confronted with an alternative practitioner in 1981, who had treated literally over 10,000 patients using this nutritional program over a 20-year period. It was a mass of information.

One of the first things Kelley did for which I respect him to this day enormously, is he opened all his records to me. There were no secrets. His successes, his failures, the people that loved him, the people that wrote him hate letters, all that stuff was opened to me from the time I first started investigating his work. There were no secrets.

We were confronted with a mass of records. Half of them were in Dallas, where we had an office, half of them were in Washington state; some of them went back 25 years. At one point, he had had a fire in his house, some of his records were destroyed. And I was confronted with this enormous project that Dr. Good was encouraging me to do without the slightest idea how to do it.


It took eventually 5 years to do that. Of course, I began the study while I was a medical student, continued it extensively while I was a fourth-year medical student. I had a 4-month block of time to really get involved with this. Then during my 2-year immunology fellowship this is primarily what I did.

To the day I die, will be grateful to Dr. Good that he allowed me to do this while I was doing my immunology training. I eventually went through 10,000 of Kelley’s records, evaluated over 1,000 very intensively. By intensive evaluation, I mean, we got complete medical records. I interview all his patients, over 1,000. Of this group, we were able to select 455 that we felt at least met the initial criteria for the second part of the research study. Terrible prognosis, biopsy-proven, radiographic studies either ──── survival or tumor regression. We eventually selected that down to 50 patients that we evaluated — that I evaluated extensively. Some of these patients I interviewed 6, 8, 10 times. Some of them I actually went to their home. We got complete medical records.

I apologize for that voluminous and tedious quotation, but it amounts to a small fraction of a recurrent litany of name-dropping, euphemism, false modesty, other acts of pseudo-self-deprecation, and unfalsifiable claims.

According to Dr. Gonzalez, the result of his 5-year investigation of Kelley’s practice was a 500-page manuscript that he never published: “One Man Alone: An Investigation of Nutrition, Cancer, and William Donald Kelley.” It consisted of detailed case reports of the above-mentioned 50 patients. That manuscript, however, was examined in detail by six physician reviewers for the Congressional Office of Technology Assessment (OTA), as reported in 1990. The OTA reviewers found that

In all cases,…documentation presented in the manuscript was inadequate to confirm critical details of the narrative, and in many cases, it appeared that critical pieces of information did not exist in the medical record at all (e.g., confirmation of metastatic disease), mainly because the patients had not been followed up with tests and scans to determine the status of their disease.

The OTA’s assessment of both Kelley’s regimen and the similar regimen of Max Gerson did not suggest real evidence of efficacy.

Dr. Gonzalez’s CV reveals that he spent much of his 4th year of medical school pursuing his interest in Kelley, and that his postgraduate clinical training consisted solely of an internship (1 year) in internal medicine at Vanderbilt. Above, he admits to having spent most of his “2-year immunology fellowship” poring over Kelley’s records and writing his manuscript. Thus Dr. Gonzalez’s implicit claims to clinical expertise in general, to specific expertise in oncology, and to expertise in conducting clinical trials are unwarranted.

Or maybe just a Garden-Variety Quack

It is instructive to compare Dr. Gonzalez’s statements about Kelley to statements written by two veteran and savvy health fraud investigators, Stephen Barrett, MD, the acerbic founder and editor of Quackwatch, and his frequent co-author, the late hematologist and nutritionist Victor Herbert, MD, JD:

In the 1960s, William Donald Kelley, D.D.S., developed a program for cancer patients that involved dietary measures, vitamin and enzyme supplements, and computerized “metabolic typing.” Kelley classified people as “sympathetic dominant,” “parasympathetic dominant,” or metabolically “balanced” and made dietary recommendations for each type. He claimed that his “Protein Metabolism Evaluation Index” could diagnose cancer before it was clinically apparent and that his “Kelley Malignancy Index could detect “the presence or absence of cancer, the growth rate of the tumor, the location of the tumor mass, prognosis of the treatment, age of the tumor and the regulation of medication for treatment.”

In 1970, Kelley was convicted of practicing medicine without a license after witnesses testified that he had diagnosed lung cancer on the basis of blood from a patient’s finger and prescribed dietary supplements, enzymes, and a diet as treatment. In 1976, following court appeals, his dental license was suspended for five years. However, he continued to promote his methods until the mid-1980s through his Dallas-based International Health Institute. Under the institute’s umbrella, licensed professionals and “certified metabolic technicians” throughout the United States would administer a 3,200-item questionnaire and send the answers to Dallas. The resultant computer printout provided a lengthy report on “metabolic status” plus detailed instructions covering foods, supplements (typically 100 to 200 pills per day), “detoxification” techniques, and lifestyle changes.

Basic Science: Implausible

As is true for Gonzalez, Kelley also did not concoct the ideas that were the bases for his treatments, but borrowed them from fanciful notions that had begun early in the 20th century. The history is recounted in Mr. Specter’s New Yorker article and in the NCI’s “PDQ Summary,” but is best explained in an article by the late Sloan-Kettering biochemist Saul Green. Green’s essay dismantles each component of the “Kelley/Gonzalez” claim and exposes its most fundamental defect: it fails to acknowledge that cancer is a genetic disease; thus the “daughters” of cancer cells are themselves cancer cells.

What that means is that the Kelley/Gonzalez “detoxification” regimen, even if it were what its name implies (it isn’t), even if it were capable of removing carcinogens (it is not), and even if its several other tenets were valid (they are not), would already be too late once cancer has begun. From the standpoint of basic science, several unanimously unlikely sub-claims add up to a vanishingly unlikely whole.

Empirical Evidence: Implausible

But what about empirical evidence? It is the claim of clinical successes, after all, that constitutes the formal rationale for the study. In public, it seems, Gonzalez is careful not to offer quantitative rates of cure or survival times for patients on his regimen. In numerous places, however, he implies miracle cures of patients who had been on death’s door when he (or Kelly, before him) met them. Such statements can be found on his website, in the New Yorker article, and in lectures. According to a 1998 article by Victor Herbert, moreover, information discovered in the course of a successful $2.65 million malpractice case against Gonzalez suggests that what he tells patients is different from what he tells the press.¹ Dr. Herbert had been one of the plaintiff’s expert witnesses, and thus had access to the evidence:

Gonzalez’s own tape and his subpoenaed office records showed that he stated to [the plaintiff] that he cured 75% of all cancers with his ‘detoxification’ therapy.

He said that the only people he could not cure were those whose ability to be detoxified was destroyed by chemotherapy and/or radiation. He talked her out of the regime, which was scheduled pelvic irradiation to destroy residual endometrial cancer following hysterectomy. Gonzalez represented to her that hair analysis showed that she was full of cancer and cancer toxins, and he would detoxify her by prescribing daily about 150 supplements plus 6 coffee enemas and by doing monthly hair analysis to show her progression to cure.

Subpoena revealed that Gonzalez’s hair analyses for non-existent cancer toxins were carried out on a word processor, not on a hair analyzer. He said her ‘cancer indicators’ were falling, without naming any toxin or indicator…¹

There has been one previous “study,” reported by Dr. Gonzalez himself, purporting to demonstrate a favorable effect of the “Gonzalez regimen” on cancer of the pancreas. According to the NCI,

…a prospective nonconsecutive case series conducted by the developer and an associate, included 11 patients diagnosed with adenocarcinoma of the pancreas (stage II or stage IV). None of the patients had received chemotherapy or radiation therapy, and none had undergone surgical resection with curative intent. All the patients had pancreatic tumors that were either unresected or partially resected. Survival from the time of diagnosis was the only study endpoint, and all 11 patients (including one who left the study) were included in this survival analysis.

The investigators reported a median survival time of 17 months and a mean survival time of 25.2 months for these patients. Nine patients (82%) survived 1 year, five patients (45%) survived 2 years, and four patients (36%) survived 3 years. At the time the study was reported, two patients were alive: one who had survived 3 years, and one who had survived 4 years. The researchers concluded that the 1-year and 2-year survival percentages for this group of patients were superior to those observed for other U.S. patients diagnosed with adenocarcinoma of the pancreas (1-year survival, all stages = 25%; 2-year survival, all stages = 10%).

The key phrase in that summary is “nonconsecutive case series,” which means that Dr. Gonzalez and his associate did not enroll every patient with pancreatic cancer who came along, but selected among them. According to the abstract of his case series, the 11 subjects were accrued over 3 years, between January 1993 and April 1996. Although we don’t know how many potential subjects there were during that time, Gonzalez reported in 1999:

I have followed thousands of patients over the years who have done coffee enemas in some cases for decades: virtually all patients report an increase sense of well being. I have done them myself daily since first learning about them in 1981.

Since he didn’t begin to follow others “doing coffee enemas” until 1987, he must have averaged nearly 200 or more “coffee enema” patients/year over 12 years. They may not all have had cancer of the pancreas, but there may have been many more than 200/yr, and it seems likely that patients with untreatable cancers would have been preferentially attracted by Gonzalez’s claims. Of the four most common malignant tumors—those of colon, lung, breast, and pancreas—that of the pancreas is the least likely to be treatable at the time of its discovery. By these criteria, Dr. Gonzalez may have seen as many as several hundred patients per year with pancreatic cancer, but probably no fewer than 50/yr.

Elsewhere, however, Dr. Gonzalez disputes this, and seems to contradict his previous assertion about “coffee enemas”:

Over the years, I have repeatedly heard the claim that Dr. Isaacs and I must be processing and treating thousands and thousands of new cancer patients each year to obtain the results illustrated by these case reports. In fact, a good friend of mine recently remarked that I must be seeing “350-450” new cases of pancreatic cancer yearly, because we are well known for our success with this particular illness. This is simply not the case. In reality, we see no more than 3-5 new cases a year.

Maybe yes, maybe no. It matters, of course, because the 11-subject case series is meaningless without a denominator, i.e., the total number of patients from which the 11 were culled. I’ve known plenty of patients who lived with cancer of the pancreas for more than one year, a few who lived 2 years, and one who lived 4 years, and I’m not even a cancer specialist. None of those patients “did” the “Gonzalez regimen,” but if they had it would have been a simple matter to compile their case histories in such a way as to cast a favorable light on it. We’ve mentioned such “selective reporting” previously on SBM.

In an article in 1995, philosopher Douglas Stalker argued that such reporting of “best” cancer cases is inevitably misleading:

It is almost axiomatic that if a therapy has been in use for a period of time, some patients will have received the therapy and will also have experienced a positive outcome of their medical problem. The therapy may have caused the positive outcome, or it may not have. A positive case does not tell us about cause and does not distinguish the true from the false therapy hypothesis; positive cases alone do not change the prior probability of a hypothesis.

Stalker showed that the NCI had no idea how to respond to positive case reports, including those involving Laetrile and the first 3 that Gonzalez himself had offered in the early 1990s. Stalker demonstrated the fallacy of the NCI’s “Best Case Series Program” to evaluate “alternative” cancer claims, which began in 1991. It is still in force and does not require that the cases be “consecutive” or that a denominator be reported. He predicted, correctly, that the program was likely to result in the NIH awarding grants for trials of methods for which the evidence was less favorable than that for many plausible chemotherapy regimens that had been automatically dismissed after failing the standard uncontrolled screening test of the day, the Gehan two-stage trial.

How could the NIH have ever allocated $1.5 million of taxpayers’ money for a trial of the “Gonzalez regimen”?

Next Week: The Politics of “Alternative Cancer Cures” and More

[1] Herbert V. The Continuing Case of Nicholas Gonzalez. Scientific Review of Alternative Medicine 1998;2(2):43-44


†The “Gonzalez Regimen” Series:

1. The Ethics of “CAM” Trials: Gonzo (Part I)

2. The Ethics of “CAM” Trials: Gonzo (Part II)

3. The Ethics of “CAM” Trials: Gonzo (Part III)

4. The Ethics of “CAM” Trials: Gonzo (Part IV)

5. The Ethics of “CAM” Trials: Gonzo (Part V)

6. The Ethics of “CAM” Trials: Gonzo (Part VI)

7. The “Gonzalez Trial” for Pancreatic Cancer: Outcome Revealed

8. “Gonzalez Regimen” for Cancer of the Pancreas: Even Worse than We Thought (Part I: Results)

9. “Gonzalez Regimen” for Cancer of the Pancreas: Even Worse than We Thought (Part II: Loose Ends)

10. Evidence-Based Medicine, Human Studies Ethics, and the ‘Gonzalez Regimen’: a Disappointing Editorial in the Journal of Clinical Oncology Part 1

11. Evidence-Based Medicine, Human Studies Ethics, and the ‘Gonzalez Regimen’: a Disappointing Editorial in the Journal of Clinical Oncology Part 2



Posted by Kimball Atwood