• Home
• Our Treatment Program
• Becoming a Patient
• Current Patients
• Case Reports
• Research Efforts
• Statements of Support
• About Us
• Books and Recordings
• In the News
• Media Inquiries
• Conferences
• E-mail Policy

Subscribe to our announcement list!

Your E-mail Address

More information about this subscription

Summary of Major Problems Undermining the NCI-NCCAM Study

Working together with a group of dedicated scientists, Dr. Isaacs and I organized and brought to completion the pilot study including publication of the data with no great difficulty. And though the animal project proceeded smoothly as well, unfortunately our early optimism about the NCI clinical trial quickly turned to frustration. Numerous problems plagued the effort almost from its inception, and to complicate matters, early on both Dr. Klausner of the NCI and Dr. Harlan of OAM went on to other jobs. Midway, Dr. Antman left Columbia, first to go to the NCI as Associate Director, then to Boston University as Dean of the Medical School. Other staff from the NCI and NIH initially involved also moved on, so by 2002 most of those who had helped nurture the clinical trial in its beginning stages had been replaced, at times by scientists who seemed to me less interested, less motivated and less determined to move the effort toward a successful completion. Now more than 10 years after its enthusiastic beginnings, the project remains effectively uncompleted, its data, in my opinion, largely meaningless, and the controversies surrounding my work still unresolved.

Below, I outline the serious problems in design and management that have in my opinion derailed the clinical trial from its original goal. Note that at times I refer to meetings which Dr. Isaacs and I attended along with the various study supervisors from Columbia, the NCI and NCCAM. These sessions, scheduled at regular three month intervals from the onset of the project in 1997-1998, convened at Columbia in the Milstein Pavilion on 168th Street in Manhattan, home to the offices of both Dr. Antman and Dr. Chabot. Usually, the Washington team flew up, though occasionally they attended via conference phone.

I. Problems with study design

As a start, we believe the original flawed design nearly derailed the study for good. From the beginning of the project, the Columbia-NCI team insisted, against our objections, that the trial be set up as a randomized venture. With this format in place, patients interested in joining the study were to have no say in the treatment they would receive, but instead would be randomly assigned by Dr. Chabot either to the chemotherapy or nutritional arm of the study. In the conventional medical world, researchers have long considered that this type of study represents the ultimate gold standard for clinical trials, based on the assumption that such a design creates two equivalent groups through the unbiased assignment of patients. This approach, the thinking goes, minimizes the chance that investigators in charge of the trial might deliberately or subconsciously enter patients with certain characteristics for a particular treatment to influence the results. In a study evaluating a new cancer drug, for example, a researcher favoring the treatment might assign earlier stage, healthier patients to receive the medication, while deliberately selecting more advanced, sicker subjects for the competing therapy. In this way, the outcome of the trial can be easily influenced.

Dr. Isaacs and I warned from the outset that in our case, such a format invited disaster by creating rather than preventing “bias.” First of all, since our treatment and the study were so unique – comparing my well-known nutritional approach to standard chemotherapy – potential candidates seeking to sign up would do so only because of their interest in our regimen and presumably because of their knowledge of the positive pilot study data. Such patients, we predicted, would never agree to proceed if arbitrarily assigned to receive chemotherapy. Furthermore, Dr. Wynder himself argued that with a disease as virulent as pancreatic cancer, with the known survival rates so dismal, our study hardly required randomization to prove a statistical point about efficacy. Nonetheless, the Columbia-NCI staff dismissed our warnings and Dr. Wynder’s advice, insisting the study proceed with a randomized design.

As we had expected, with this format in place, accrual proved to be a disaster despite the publicity generated by the project. During the first year of the study’s operation, some 260 patients with pancreatic cancer contacted Columbia expressing an initial interest in joining the study, but after learning of its randomized design a grand total of three patients agreed to enter, two of whom promptly quit to pursue other options after being assigned to chemotherapy. With a full year of time wasted, the academic experts supervising the project from Columbia and the NCI finally agreed to revise the trial format, a process that took months during which time no patients could be enrolled. Eventually the various regulatory authorities approved the trial in its new incarnation as a case control study, in which patients could choose the treatment they wanted, either chemotherapy or our nutritional regimen. Despite the changed design, the Columbia team in charge of patient evaluation were to assure, as much as practically possible, that each of the two groups would be more or less evenly matched by various criteria, including extent of disease and overall clinical status.

Unfortunately, the specter of the randomized format continued to haunt the study for years, long after the switch to case control. Repeatedly during the duration of the trial, patients diagnosed with pancreatic cancer called our office, informing my staff they sought our therapy as private patients, wishing no part of the clinical trial because they would not be “forced” to receive chemotherapy. We will never know how many suitable patients never even considered joining the trial because of such thinking. In this way, the randomized format created a significant bias, the very phenomenon it was intended to prevent, by discouraging interest in the project among an entire population of potentially ideal subjects. Ironically, the randomized design would ultimately turn off the very patients who normally prove to be our most dedicated, most compliant – and in turn, most responsive to the treatment - those with a previous interest in alternative approaches and nutritional treatments.

II. Lack of a lead-in period

The randomized issue proved to be only one of many serious obstacles to the study’s successful completion. In our pilot study we included, at Dr. Wynder’s suggestion, a preliminary “lead-in” period, essentially a test run, in which we evaluated suitable patients for compliance with each aspect of the therapy before their formal entry into the trial. If during this time a prospective study candidate could not or would not follow the prescribed regimen, he or she would be discounted from further consideration as a study subject, whereas those that complied appropriately would be then officially entered. Dr. Wynder, one of the world’s preeminent epidemiologists and an expert on clinical trial methodology, particularly for those studies involving lifestyle modification, insisted the pilot study include such a lead-in, to help insure that the patients ultimately entered had the motivation and ability to comply with the prescribed nutritional regimen.

Clinical trials set up to evaluate lifestyle interventions such as dietary change commonly provide for extensive lead-in periods, as well as elaborate pre-screening to assess each patient’s motivation and ability to adhere to the regimen under study. (1) The landmark Diabetes Control and Complications Trial helped establish that tight regulation of blood sugar with strict diet and multiple daily insulin doses drastically decreases complications such as blindness and kidney failure. In this major multi-year study, all patients deemed initially eligible underwent roughly 40 hours of screening, including multiple assessments by a psychologist, before formal enrollment. (2)

Frequently – but not always - clinical studies of some new cancer drug do not include a lead-in. For most chemotherapy treatments, the patient need only show up in the doctor’s office to receive the medication so issues of motivation and compliance become less important than they are with a lifestyle regimen that needs to be applied at home on a daily basis. Nonetheless, studies of new cancer drugs do at times provide for such “test runs”; for example, in 1998, the FDA approved the chemotherapeutic agent gemcitabine for the treatment of pancreatic cancer based on a clinical trial that provided a one week “observation” period. During this time, the investigators sorted out and excluded from the study any patient whose pain could not be controlled. (3) Eventually, 34 subjects were so eliminated from further consideration, and of these, we know absolutely nothing. A total of 126 assigned to either of two chemotherapy regimes being studied eventually qualified as “treated,” so the researchers discounted after the first week approximately 27% of subjects initially considered acceptable, not an insignificant number.

With the somewhat limited resources allocated for our clinical study, we understood extensive pre-screening may not have been possible, but certainly a 1-2 week lead-in period would have helped eliminate subjects not able or not willing to follow through with the intensive dietary/nutritional treatment. Dr. Wynder, Dr. Isaacs and I strongly argued that without such a provision, the entire study might be undermined should too many non-compliant patients be entered.

Despite our repeated arguments and despite precedent in the scientific literature, the chief investigators of the project refused to allow the lead-in period. To make matters worse, after July 2000, at the insistence of the NCI, Dr. Isaacs and I were removed from any involvement with the patient selection process, allegedly to eliminate any possibility of “bias.” Subsequently, Dr. Chabot of Columbia, who had no experience whatsoever with our treatment, its intricacies, its demands, and to our knowledge absolutely no experience evaluating patients for any type of lifestyle regimen, would alone determine which patients qualified for entry. Unfortunately, the majority of patients Dr. Chabot ultimately approved to receive our nutritional therapy either could not, or would not comply with treatment. Had a lead-in been in play, even if only a week long as in the Gemzar study, we could have discounted, out of the total of 39 eventually admitted into the nutrition arm, 11 patients who never started the therapy or dropped out within the first seven days, all of whom died fairly quickly after their initial consultation with us.

III. Intent-to-Treat Provision

Clinical trials lacking a lead-in period often – though not always – adopt an “intent-to-treat” format. With such a provision, researchers agree that all patients qualified and entered into the study for any of the treatments under scrutiny will be considered as having been treated, regardless if they actually proceed with the prescribed therapy or not. Though such an approach on first glance might not make much sense, researchers justify such an “intent-to-treat” rule as necessary to evaluate fully a new drug. For example, if in a study 100 patients receive some new medication but 50 drop out after a week because of serious side effects, certainly it would seem prudent to include these patients as treatment failures rather than discount them, since they quit because of some negative reaction to the drug. On the other hand, such a design can be disastrous for a lifestyle intervention trial such as ours, since patients who might initially be enthusiastic but who can’t or choose not to proceed with the self-administered dietary/nutritional regimen will be counted as having been fully treated.

In our particular study, the supervisors not only denied our request for a lead-in period, but insisted we accept an intent-to-treat format. Unfortunately this requirement allowed Dr. Chabot to count as properly treated the 11 patients assigned to the nutrition arm who did not or could not proceed with treatment beyond several days, not because of any negative reaction to the therapy, but due to choice or their physical inability to follow the program. We believe this provision ultimately helped undermine the value and legitimacy of the data, by requiring that we count multiple patients such as these who never complied with the regimen or who did so only briefly as “Gonzalez treatment failures.”

IV. Attempts to sabotage the clinical study

In standard oncology clinical trials, most of the time everyone involved works together for the benefit of the patient, for knowledge, perhaps even for glory. Having long been the object of tremendous scorn from our colleagues over the years because of the nature of our treatment, Dr. Isaacs and I were not so naïve to have expected enthusiasm from the oncology community about our study. And, as expected, shortly after its inception, the harassment began. As a rather glaring example, the late Dr. Victor Herbert, a long time vocal opponent of any form of “alternative medicine” and my therapy in particular, filed a complaint with the Department of Health and Human Services in Washington in an effort to stop the project. Later, when he attacked my work in the mainstream media and on the Internet, others joined in the fray as well.

Opposition took more subtle forms as well. Despite the backing of such prestigious institutions as Columbia and the NCI, throughout the study oncologists nearly universally refused to refer patients into the trial. Consequently, accrual of patients continued to be a problem even after the change in format to a non-randomized design.

Over the eight or so years of the trial’s duration, not one patient wishing to pursue our nutritional therapy came by way of an oncologist’s referral. Early on, after the change in study design, Dr. Isaacs and I insisted that for the study to succeed at all, the NCI with all its power and authority needed to become more proactively involved to help with patient accrual. At one point, the NCI personnel assigned to the study repeatedly promised they would take out ads in the medical journals, as is often done with government studies, encouraging oncologists to consider our trial for their patients diagnosed with inoperable pancreatic cancer. Though I understand the ads were actually written, none were ever placed anywhere, leaving us at the whims of the word of mouth network to attract patients into the study.

Ultimately, only the oncology team at Columbia cooperated in any way only after much prodding by Dr. Antman and Chabot, and only for the admission of chemotherapy subjects to form the comparison “control” arm as we shall see. Even for this group their referrals proved not helpful.

Oncologists not only refused to refer patients to the trial, but at times actively discouraged their patients who might express an interest from seeking entry. A number of candidates suitable for the study who had learned about our treatment on their own informed our office that their oncologist had strongly argued against considering the project. One well-known Memorial oncologist warned a candidate interested in joining the study that I was a “quack” and the study a “fraud.”

Oncologists also frequently discouraged patients who actually entered the nutritional arm of the study from continuing with the prescribed regimen. Unfortunately, a protocol provision against which we argued and that ultimately caused enormous damage required that each patient assigned to the nutrition arm consult with a physician monthly for an examination and blood work. On the surface, such visits would hardly seem to be the source of potential catastrophe, since, one might think, how can a visit with a doctor be a problem? And trials involving chemotherapy drugs often require frequent physician assessments to monitor closely the toxic side effects of the medications being tested, such as severe anemia or immune suppression.

For those subjects who lived in the New York area, Dr. Isaacs and I could satisfy this rule by meeting with the patient ourselves monthly. We had no problem with such an arrangement, of course. But as it turned out, only three of the patients ultimately entered into the nutrition arm lived in the New York area, with the great majority residing at great distance. Consequently, nearly all subjects assigned to us for treatment were followed by a local doctor, most frequently an oncologist completely unfamiliar with our treatment approach and usually hostile toward it, with only a few exceptions. Repeatedly, we heard from our patients that during the required monthly meetings, the local physicians aggressively discouraged them from continuing their treatment with us, instead urging them to proceed with some standard approach - despite the fact that the conventional therapies for inoperable pancreatic cancer have proven largely worthless.

Even in those cases in which patients clearly seemed to be responding to treatment, often the local physician still tried to persuade them to quit the trial. In several instances, the doctors reminded the patients how sad it was that they chose to spend the last months of their life following a restricted diet which couldn’t possibly help their disease, instead of enjoying themselves with pizza and ice cream.

Should any patient develop a symptom or medical problem – and even our most successful patients with pancreatic cancer will have periodic difficulties even under the best of circumstances – the local physician usually urged them to discontinue our regimen in favor of a more conventional approach. In several instances, the discussions deteriorated into heated arguments, with the words "quackery" branded about. One oncologist told a patient entered for our treatment that the trial was an elaborate publicity stunt on our part meant to lure in patients for financial gain. The oncologist held to this story even when told that the patient had paid us no money and that the study was fully funded by the NIH.

We are unaware of any clinical trial in oncology or in any other branch of medicine that subjected very ill patients facing death to such routine and repeated harassment intended to undermine a prescribed treatment. In conventional drug studies evaluating the latest miracle against cancer, invariably the oncologists and physicians involved act more like cheerleaders, often lauding the potential possible benefits of treatment even in the absence of any proof of efficacy.

The Diabetes Complication and Control Trial demonstrated that patients, to succeed with the prescribed complicated dietary regimen, required frequent and repeated positive reinforcement and encouragement to succeed. In a more recent trial evaluating the effect of dietary modifications in patients diagnosed with prostate cancer, those assigned to the nutrition arm met in weekly support groups. (4) In great contrast, our patients commonly endured repeated and aggressive monthly discouragement.

Not surprisingly the physician harangues and more subtle forms of sabotage repeatedly influenced patient compliance with the regimen. By our tally, five patients dropped out of the study largely because their local physicians insisted they proceed with a more standard, more “acceptable” form of therapy.

As evidence mounted that local physicians rarely supported the project and might be undermining both the accrual of new patients and compliance of those entered, we repeatedly discussed the situation with the study supervisors from Columbia and the NCI. We urged that the NCI send out an official mailing to all US oncologists, asking that for the benefit of science, they cooperate with the trial, to help answer the questions raised about our treatment. Though the NCI investigators assigned to the project agreed such a letter might prove valuable, and though such a letter was actually written after two years of our prodding, to our great frustration it was never sent.

V. Perpetual delays in the study

Over the years, since my fellowship days under Dr. Good in the mid-1980s, I have observed that when the NCI wants to get a study done, it gets done. I have lived through the 1980’s miracle of interleukin II, the cover story in Newsweek Magazine lauding the treatment, the NCI press conferences and press releases, the tens of millions assigned to “fast track” what has turned out to be a very toxic, very expensive and largely ineffective approach. More recently I like so many millions of others lived through the miracle of angiogenesis, the front page articles in major newspapers, the TV specials, the NCI press conferences and press releases, the millions assigned to research this presumed extraordinary therapy that hasn’t proven all that beneficial. Clearly, when the NCI wants something to happen, it happens.

Our study proceeded along a completely differently track. Most if not all NCI clinical studies must be reviewed and re-approved on a yearly basis, usually a fairly routine business. For clinical trials to run smoothly, this process must be quick and simple. And each year, like other government projects, our clinical trial came up for its annual mustering at the NCI, at NCCAM, and at Columbia. In our case, the procedure hardly ran smoothly; for most of the project’s duration, at the onset of each fiscal year we were told that for some reason or another renewal had been delayed, or required some new paperwork, or more commonly, that someone involved with supervising the study had not submitted the appropriate documents in a timely fashion. Not a single delay involved some deficiency or non-compliance on our part.

While we waited for the requested forms to be completed, or located, or submitted, the trial remained suspended so that no new patients could be entered. Due to such bureaucratic delays, during the last three years of its existence, the trial was on hold far more than up and running. For example, during the fiscal year beginning June 2004, over a 12-month period, patient accrual was suspended for 11 months.

No study could possibly run effectively under such conditions, and indeed, the serious and chronic delays became well known in the alternative medicine network, the traditional source of our most determined and compliant patients. To many, these repeated suspensions proved government indifference to our nutritional treatment. We believe these delays, along with the specter of the randomized format, contributed to the widespread distrust among potential study candidates during the last years of the study. By that point, many of the patients diagnosed with pancreatic cancer who called our office anxious to begin our treatment told us they didn’t want to be any part of the project, even when open – as it rarely was - for accrual. We found ourselves in the ironic situation of treating patients diagnosed with pancreatic cancer successfully off study in our private practice as the clinical trial deteriorated into chaos. Whatever the cause, these repeated and lengthy delays helped undermine the integrity of the effort.

VI. Inefficient screening of study candidates at Columbia

Except for the initial 20 months of the study when only 3 patients were actually entered for treatment with us, for nearly its entire duration Dr. Isaacs and I were completely excluded from any aspect of the evaluation and admission of potential trial candidates, a duty relegated completely to Dr. Chabot. Furthermore, even during the period when we could veto patient entry, Dr. Chabot, we were sternly instructed, was still to conduct the initial evaluation of all study candidates. We were told in no uncertain terms that we were to inform all those who might contact our office with a diagnosis of pancreatic cancer, whether expressing an interest in joining the study or not, whether they even knew about the study, that they must go through the screening process at Columbia for the clinical trial. We were even forbidden from discussing the rudiments of our treatment or the trial with any patient diagnosed with pancreatic cancer, at risk of committing a protocol violation. We believe these stark regulations, which even forbade my staff from discussing my own treatment, must have seemed bizarre to those diagnosed with pancreatic cancer seeking information about our approach or the clinical trial. We suspect that these strange rules may have discouraged many potentially suitable patients from following up with Dr. Chabot’s office.

To make matters worse, the evaluation process at Columbia proved to be hardly efficient. After Dr. Chabot assumed sole command of the entry process many potential study candidates we had directed to his office called us back complaining, sometimes angrily, that they had repeatedly left messages for Dr. Chabot expressing their interest in joining the study, but that no one had gotten back to them, days or even weeks later. In one case, a patient who had actually sent his records to Columbia for assessment had not heard back from Dr. Chabot after five weeks, despite repeated phone calls and faxes. For patients whose lifespan might be measured in several months, such delays, during which time the patients pursued no treatment of any kind, we believe put patients’ lives at risk. At times, patients asked us to intervene with Dr. Chabot, but he seemed equally as unresponsive to our calls, faxes and e-mails. We suspect that many patients, aware of their dire situation and with no time to waste, simply chose to seek treatment elsewhere. Had I been in their situation, I would certainly have done the same.

VII. Financial harassment

Initially, at our suggestion the grant for the clinical trial was set up to cover all aspects of the nutrition treatment, including office visits with us, supplement costs, and expenditures for any required equipment such as juicers and water filters. Though our treatment costs a fraction of a standard course of chemotherapy, nonetheless we thought this financial support for patients a good gesture since insurance companies, which usually reimburse for the most expensive drug regimens, rarely if ever cover our alternative approach. Dr. Isaacs and I were not to be paid for our participation in the study, only for the costs of our office visit fees, though at a reduced rate from what we would normally charge. This arrangement seemed fair and simple. In the final version of the protocol, someone, whom we do not know, “forgot” to include coverage for any of our office visits beyond the initial evaluation. Dr. Isaacs and I made no issue of this oversight, since we were grateful for the opportunity to have our work evaluated with colleagues at the NCI. Consequently, we essentially treated each patient for free beyond the first consultation.

After approving the $1.4 million, for the duration of the trial the NCI funneled all grant monies directly into Columbia’s coffers. In turn, our office was to submit to Columbia copies of all bills for patient expenses, such as the cost of the initial consultation and the expenditures for supplements. When paid, we would then reimburse the company providing the supplements to patients on the trial. As an aside, I think it’s important to remark that when an institution such as Columbia agrees to serve as the site for a federally funded study, they receive 10% of the total grant amount as a “gift,” officially termed “oversight,” to use as they wish.

During the last five years of the project, for any variety of reasons that often made no sense to us, payments for the supplements as well as the initial office visits were held up for months. At such times, Dr. Isaacs and I willingly accepted the postponed reimbursement of our initial office fees, but the supplement costs, which generally ran in the neighborhood of $650 per patient per month, represented a much greater problem. Since those entered into the nutrition arm had been informed the grant subsidized the treatment, we did not think it appropriate to announce to them suddenly that they had to start paying for their enzymes and nutrients. Consequently, whenever funds were withheld, Dr. Isaacs, myself and the company providing the products began underwriting the nutrition arm out of our own pockets. At one point, the amount we paid out exceeded $20,000. Only after much effort were the expenditures finally reimbursed, but nonetheless, I have never heard of any NCI study that required the investigator to subsidize for any period of time the costs of the project.

During the fiscal year beginning June 2002, the funding delays had become so pervasive that Dr. Chabot repeatedly warned me the study might be closed down. When I inquired why something so simple as the release of approved monies had become such a constant problem, he explained that those within the NCI hierarchy responsible for overseeing the clinical trial simply did not move the proper paperwork through the appropriate channels as needed. The problem was, he said, bureaucratic.

Though the study didn’t end at that time, when the funding problems persisted, in the spring of 2003 I spoke directly with Dr. Stephen Straus, the Director of NCCAM, the government group directly responsible for underwriting the project. He assured me that both he and NCCAM stood behind the clinical trial – a claim that in retrospect seems now to have been untrue - and he, like Dr. Chabot held the NCI personnel responsible for holding up the release of allocated monies, a charge the NCI later denied. In April of 2003, after Dr. Chabot again advised me that the long suspensions in financial support made it unlikely the clinical trial could continue, I wrote directly to Dr. von Eschenbach, the NCI Director. In this letter, I summed up the various problems we had faced, particularly the untimely transfer and release of funds from the NCI:

When the study has been under threat before, funding is held up, leaving our patients without support to pay for the treatment, which is not covered by insurance. Last year, during a prior hiatus and threat, our office paid out many thousands of dollars to keep the patients on the trial, and fortunately the study continued and the charges were picked up. Now, we simply will not be able to continue to finance the patients on the trial out of our own pocket. If this is not resolved soon, we will have to tell the patients to buy their own supplements. I know in a number of cases, the patients who are currently responding will not themselves be able to pay for the treatment and will die. This prospect has created a nightmare situation for Dr. Isaacs and myself, and I am deeply concerned about this, particularly since as best as I can understand the study may be shut down because paperwork is not moving between offices.

It has been very difficult to conduct a study under the repeated threat that the study is soon to be closed down, and the patients are on their own. For now, accrual is on hold, so just as potentially good patients are volunteering to enter the study, we have to tell them the study is not open.

Is there any way to learn from the NCI what the specific problems are? Is there any way the problems can be dealt with in a productive manner, so that patients accepted into the study in good faith are not put at risk?

Dr. von Eschenbach did not respond directly but instead referred my concerns back to the NCI office of Dr. Wendy Smith – one of the study supervisors identified by Dr. Straus and Dr. Chabot as the source of the problem. Thereafter, the trial continued to struggle along as before, with the various roadblocks, including the endlessly slow release of funds, uncorrected.

Dr. Isaacs and I considered these financial delays - unique as far as anyone else seems to know in clinical trial history - a form of harassment, whether due to deliberate intent, bureaucratic indifference, or ineptness. Regardless of the reason, the effect was the same, enormous financial stress on our office. Despite this burden Dr. Isaacs and I continued onward because we so strongly believed in the importance of clinical trials, and the potential benefit of our approach.

VIII. Epidemic non-compliance in the nutrition arm

For a number of reasons, including physical disability, psychiatric instability, lack of social support, poor motivation and physician harassment, we have calculated that 30 of the 39 patients ultimately entered into the nutrition arm followed the prescribed regimen not at all, for only brief periods of time, or incompletely. Such epidemic poor compliance differs greatly from what we generally observe among those we see in our private practice. In the current study, this widespread poor and non-compliance helped render the data meaningless, since, by the “intent-to-treat” study design, all patients admitted for treatment with us, even those who never took a supplement, were to be considered “Gonzalez patients” as if fully compliant with the prescribed regimen.

IX. Failure to abide by the standards of appropriate clinical trial management

For the results to have any meaning, every legitimate clinical trial must follow certain very precise and well-known rules and regulations. I have chosen to break these down into eight distinct categories as follows: accepted patients must meet the entry criteria of the written protocol; the study must have an adequate number of subjects to make a statistically valid point about the therapy or therapies under scrutiny; the numbers in each arm of the study must be more or less, within certain specified limits, equal; the patients in each group must be more or less equivalent in terms of disease stage and clinical status; patients should be admitted into each arm over time in a steady, balanced manner; the supportive care given subjects in each group must be equivalent; the standards for stopping each treatment must be the same; and finally, the Principal Investigator charged with assessment and acceptance of trial participants must be unbiased toward each treatment being studied, with no financial or intellectual tie to either one.

In my opinion, the NCI-NCCAM investigation of our regimen failed on each and every count, as I will now document.

A. Entry of unsuitable patients

From our initial meetings with Dr. Klausner and Antman, Dr. Isaacs and I were willing to put our reputation on the line, and put our therapy to the test treating patients diagnosed with inoperable pancreatic adenocarcinoma, one of the most deadly of cancers. I think few oncologists would be willing to take on such a challenge, especially if they were as closely watched as we are by the media and by medical colleagues. However, we knew from our private practice and pilot study experience that patients with the disease, if physically able and willing to follow the therapy can do very well. Over the years, we have treated patients diagnosed with inoperable, even stage IV disease who are alive and in good health in excess of five and ten years.

We have limits, of course, to what we can do. Ours is a lifestyle and nutritional treatment that patients must pursue on their own, unlike chemotherapy that only requires patients show up in the doctor’s office. In the case of our regimen, patients must follow a prescribed diet, and ingest some 150 or more supplements to be taken at precise times throughout the day. The proteolytic enzyme component of the therapy, which provides the main anti-cancer effect, is currently unavailable in intravenous or intramuscular forms and must be taken orally. We frequently prescribe up to 100 capsules of these enzymes in divided doses daily so patients must be able to swallow to succeed with the regimen. Since patients administer the therapy at home, they should also be able to care for themselves to some degree; those bedridden, or extremely debilitated, in the final stages of the disease, or those requiring hospitalization, cannot proceed with the therapy. We have no illusions about our approach; we cannot help someone days from death, nor would we ever attempt to treat such a patient in our private practice.

By its very nature, the regimen requires discipline and some determination, as does any lifestyle intervention; patients must be motivated, or else, as we learned long ago, they will not follow through with its day to day application. And patients with obvious mental illness most likely cannot effectively pursue this treatment.

Initially, Dr. Antman, Dr. Wynder, and others supervising the project agreed that in order to achieve our goal, a fair evaluation of my therapy, we needed to enter patients into the nutritional arm physically able, psychologically capable and emotionally willing to adhere to the therapy. Admission of those too ill to proceed with the treatment, mentally ill patients, or subjects with little discipline or motivation could lead easily to epidemic poor compliance, and meaningless results.

Though Dr. Antman disregarded our objections to the randomized format and the “intent to treat” provision, she did incorporate many of our suggestions into the final draft of the written protocol to help assure, at least on paper, that only suitable candidates would be entered. As a start, all patients, to be eligible, had to be admitted into the study within eight weeks of biopsy diagnosis. They had to be able to eat normally or near normally, an absolutely critical requirement for our dietary and nutritional therapy.

Candidates had to be free of any form of mental disability "preventing informed consent or intensive treatment,” to quote the written protocol for the trial, and express the motivation to pursue the home-based treatment plan. Furthermore, the rules of the study required each patient have at least one supportive live-in family member, willing to help with the nuts and bolts of the treatment. We long ago learned that those with advanced cancer need assistance on a daily basis to comply appropriately with the regimen. Finally, each patient seeking entry who met all these basic criteria was to sign an official statement of informed consent as required for all NCI clinical trials, acknowledging the purposes, risks, and possible benefits associated with the nutritional therapy.

As the study geared up, Dr. Isaacs and I thought, because of our 20-year experience with the therapy, that we should be involved in some way with the intake evaluation of patients seeking entry into the nutritional arm, to help reduce the admission of unsuitable subjects. And initially, Dr. Antman and the NCI staff agreed we should have the power to reject those patients we deemed did not meet the explicit entry criteria. But as the project proceeded, Dr. Antman and the others in charge expressed an overriding concern that Dr. Isaacs and I needed to be kept distant from all aspects of the entry process. In this way, we were told, the effort would be protected from any “tainting,” that is, any potential bias with which we might insidiously infect the study. Despite our protests, in July 2000, we were completely removed from patient selection, placing full responsibility under the control of Dr. Chabot at Columbia who had no experience whatsoever with our treatment. If Dr. Chabot accepted a patient into the study, the patient was accepted.

We hoped that the explicit entry criteria would offer us at least some protection from the admission of trial subjects unsuited for our nutritional therapy. Nonetheless, despite the clearly written requirements, over the years of the study Dr. Chabot repeatedly entered patients into the nutritional arm that we believed failed to fulfill one or more of the objective criteria. Three patients assigned for treatment with us despite our objections had been diagnosed by biopsy well over 8 weeks before entry, and should not have been admitted. Many accepted patients were in the final terminal stages of their disease, obviously too debilitated (one arrived in our office in a wheelchair unable to stand) to proceed with the therapy. We have identified 11 patients entered into the nutrition arm whose appetites were so poor they could never possibly adhere to the prescribed regimen.

Dr. Chabot admitted three patients who, because of mental disability, we believe should have been disqualified, and one with no family or social support whatsoever. We estimate that another 10 of the admitted patients lacked the drive, motivation, or faith in the treatment to stick with it for any length of time. And finally, Dr. Chabot sent three patients for treatment with us as if they had been properly consented who, as it turned out lacked any evidence of the required signed informed consent, in obvious violation of the written protocol and Federal regulations.

Discounting overlap – several patients should have been disqualified for more than one reason – we estimate conservatively that 16 individual patients of the 39 admitted into the nutrition arm did not fulfill the written entry criteria.

B. Shortfall in total numbers for the study

As the study first came into being, the Columbia experts, with input from Dr. Wynder, determined that for the final data to achieve statistical legitimacy, we would ultimately need to enter a minimum of 72 patients, counting both the chemotherapy and nutrition groups together. When our study finally closed to accrual in October of 2005, Dr. Chabot had admitted only 58 patients – 80% of the goal. Nonetheless, he and his Columbia colleagues insist the numbers adequate to “make the point,” i.e., that my therapy doesn’t work.

C. Disparity in numbers between the chemotherapy and nutrition groups and Dr. Chabot’s attempts to correct the problem

For any clinical trial to generate truly useful data, the groups being compared must be more or less of equal size in terms of numbers. Otherwise, the potential benefit of a new therapy being studied might be either exaggerated, or obscured.

In our case, when the study finally shut down to accrual, Dr. Chabot had accepted only 23 for chemotherapy, as opposed to 39 for nutrition. Despite the glaring difference, Dr. Chabot and his Columbia colleagues insisted the groups comparable. Subsequently, for reasons never explained to us, he disqualified 4 nutrition patients, so that on paper, the groups seemed somewhat more equivalent in total numbers.

D. Dr. Chabot’s handling of the staging disparity between the chemotherapy and nutrition arms

Oncologists divide pancreatic cancer into four “stages,” reflecting the extent of spread. Stage I, the earliest form and by definition limited only to the pancreas, can be treated surgically and occasionally cured. Stage II indicates the disease has spread only locally, for example into the surrounding fatty tissue or the duodenum, but not into lymph nodes or distant organs. With stage III, the cancer has infiltrated into regional lymph nodes adjacent to the pancreas, and with stage IV, the disease has invaded other organs, such as the liver, the lungs, the bone or distant lymph nodes.

Pancreatic cancer tends to be particularly invidious because the disease can initially be largely asymptomatic: for months, patients might complain of mild indigestion, vague persistent abdominal pain, or reflux after eating. Often, unsuspecting doctors prescribe antacids or proton pump inhibitors like Prilosec or Nexium with no improvement. Inevitably, at some point the patient suffers a rapid decline that prompts an intense diagnostic work-up. By this point, the disease has most often metastasized extensively.

The literature reports only 5% or less of patients will be initially diagnosed with stage I tumors. Of the group categorized as “non-stage I” approximately 70-75% are classified at diagnosis as stage IV, the most advanced level, with 25-30% as stage II and III. (5, p1143) In the Gemzar study from 1997, which excluded operable stage I patients, those entered into each arm fell into this range, with approximately 28% diagnosed as stage II and III, and 72% as stage IV. (3;6) In our pilot study, which also disqualified those with operable disease, 8 of 11 patients, or 73%, presented initially with stage IV, and three as inoperable or locally advanced stage II, the numbers again conforming to the usual distribution of non-stage I subjects. (7)

It is well documented, and not surprising, that earlier stage II and III patients have a far-improved median and mean survival, in the range of 10-14 months, as opposed to those with stage IV, with an average life expectancy in the range of 3-6 months regardless of the conventional therapy offered. (5, p1130,1143)

Our current NCI study excluded patients diagnosed with the earliest stage I disease who might be cured by surgery, but allowed inclusion of those with the more advanced stages II, III and IV. The written protocol for the trial also required that the patients entered into either of the two arms of the study, the chemotherapy and “Gonzalez-enzyme treatment” groups, be evenly matched in terms of stage. Such matching would help prevent an unfair advantage, should, say, one arm consist of primarily early stage patients, the other primarily late stage IV:

The protocol reads as follows:

13.2 Patient Matching and Sample Size: Prognostic factors in survival outcome of patients with unresectable or metastatic pancreatic adenocarcinoma were recently evaluated. Of 34 potential epidemiological, clinical, analytical and tumor related parameters studied in a prospective series of 134 patients, performance status and the presence or absence of distant metastases were independently associated with survival in multivariate analysis….

In this study, patients will be matched on an individual basis for 1) stage of disease, 2) performance status, and 3) nutritional status…”

13.3 Data Analysis
Patient Matching: As previously indicated, patients were matched on an individual basis for disease stage, performance status, and nutritional status upon entry into the trial.

During one of the regularly scheduled group meetings held December 13, 2004 at Dr. Chabot’s Columbia office, at a time the clinical trial had been up and running for a full five years, Dr. Isaacs and I first became aware of a very significant difference in both the total number of patients entered into each arm, and the stage of their disease. By that time, 38 patients had been admitted for nutrition treatment, and of these, approximately 76% by our accounting had been initially diagnosed with the most advanced (stage IV) disease, the other 24% with earlier stage II or III. This pattern approximated, as did our pilot study, the usual distribution of newly diagnosed pancreatic cancer patients as reported in the literature.

Dr. Isaacs and I were very much familiar with these numbers for the nutrition group, since we had been supervising the treatment of these patients ourselves, but we knew very little until that point about those receiving chemotherapy. But during that meeting, Dr. Chabot handed out a “data chart” which included the total numbers of patients admitted into each of the two groups, with the breakdown by stage. First, I was surprised that he had tabulated the numbers incorrectly for our group, which he reported incorrectly consisted of 35% at stage II and III and 64.7% at stage IV. But I was even more surprised to learn that the chemotherapy arm of the study, created under the direction of the Columbia oncologists, consisted of only 14 patients, 61.5% with earlier stage II and III disease, with only some 38% as advanced stage IV – a near reversal of the distribution in the nutrition group, and a reversal of the usual breakdown reported among patients diagnosed with pancreatic cancer.

Though the meager number of chemotherapy subjects enrolled by such a late date in the study’s history helped render any data questionable, the imbalance in the two groups by stage represented an even more difficult problem. Early stage patients tend to live much longer than stage IV patients, even when receiving no treatment. Since the majority of nutrition patients at the time of entry had stage IV cancer, the significantly uneven distribution by stage precluded any legitimate comparison between the two groups.

When I learned of this odd pattern during the December 2004 meeting, I expressed my grave concern that the entry of very advanced patients into the nutrition arm – most of whom weren’t able or chose not to stick with the regimen anyway - and predominantly early stage, good performance status patients into the chemotherapy group had undermined the validity of the study. No one present disagreed with my assessment.

By March 2005, Dr. Chabot had, without explaining why, disqualified two earlier stage chemotherapy patients, reducing the total number in the group to 12, but “improving” the staging balance, so that now 50% were at stage IV, 50% stages II and III. Thereafter, from April 2005 until the study closed permanently to new patient entry in October 2005, Dr. Chabot entered only a single patient for nutrition – who like so many others, followed the therapy only briefly – but 11 for chemotherapy, during a six month period when the study was often on hold. When the trial closed Dr. Chabot had qualified a total of 39 nutrition patients, by our accounting of the medical records 23% at earlier stages, 77% at stage IV. Dr. Chabot did not send us the final data on their staging status until late April of 2006; at that time, he reported that 34.7% of the chemotherapy patients had been diagnosed as stage II and III, 65.5% as stage IV.

Surprisingly, with no explanation, the April 2006 data sheet revealed Dr. Chabot had also disqualified two stage IV nutrition patients he had previously insisted needed to be counted. In addition, he described another five of our group as stage II or III, who - perhaps with one exception – must, by the medical records, be categorized at stage IV. With these unexplained revisions, he reported 40.5% of the nutrition group at stages II and III, 59% at stage IV. Miraculously, six months after the study closed to accrual, six years after its start, and long after all participants and their staging had been reported and documented to the study personnel without a challenge, the nutrition patients as a group appeared less advanced than those entered for chemotherapy. Such last minute juggling, performed by a Principal Investigator (PI) feeling pressured to salvage a failed study, didn’t inspire much confidence in the data.

E. Bunched accrual of patients during the study

Clinical trials can differ in their basic design; some are randomized, some “matched cohort,” some “case control.” Though the details aren’t important here, in any clinical effort the PI, in our case Dr. Chabot must insure that patients enter each group in a gradual, more or less equivalent fashion over time. That is, for each patient admitted into group A of study XYZ, ideally, a patient should be accepted for group B. Though such a perfectly balanced process can be difficult to sustain in a “matched cohort” effort, the PI must work toward some semblance of equitable admission of patients.

In our study, patient entry followed a “bunched” pattern, with large groups being accepted into each arm at very different times. From the time the trial began in its “matched cohort” incarnation in April 2000 until the end of December 2002, a period of nearly three years, Dr. Chabot had admitted 29 patients into the nutrition arm, representing nearly 74% of the 39 that would ultimately be counted. During that same interval, he accepted only two for chemotherapy.

From January 2004 until the project finally closed to accrual in late October 2005, a total of 22 months, Dr. Chabot qualified only three patients for our therapy. Yet, from February 22, 2005 until October 2005, a period of eight months, he managed to admit 11 patients for chemotherapy, 48% of the ultimate total, despite the fact that after July of that year the study remained frequently suspended. These individuals gained admission when the study opened for accrual for a grand total of four months.

F. Differences in supportive care between the nutrition and chemotherapy arms

In any clinical trial, the basic medical care provided all patients should be equivalent, particularly when the patients being studied suffer a deadly and aggressive disease such as pancreatic cancer, associated with all manner of serious symptoms and management problems. In our trial all of the 23 subjects Dr. Chabot ultimately admitted for chemotherapy were treated under the direct supervision of Dr. Robert Fine, director of the Experimental Therapeutics Program at Columbia. Dr. Fine has gained some admiration within the profession for his aggressive, never-give-up approach to patient care, providing the latest advances available at a major medical center. (8)

For those subjects destined to receive chemotherapy, Dr. Fine had at his disposal the resources of Columbia, a team of expert senior physicians, researchers, fellows (oncologists in training), residents, highly skilled oncology nurses and other support personnel, and all the high tech facilities one could wish for. According to his own statements, he and his highly motivated team employ all the benefits of modern academic, hospital based medicine, sparing no intervention, withholding no aid to keep patients alive – even to the point of repeated surgeries, and paracentesis (removal of abdominal fluid) done 2-3 times a week, a procedure rarely in the past pursued so determinedly. (9) Those under Dr. Fine’s charge could not ask for more intense or sophisticated care, from an enthusiastic, supportive staff at a major academic institution. Our patients, on the other hand, faced quite a different and often grim situation at the hands of local doctors at best indifferent and frequently hostile to our therapy.

Many of the nutrition patients Dr. Chabot approved for entry were too sick even to begin the treatment or stay with it for any length of time. Among those who could and did try to follow the regimen, the doctors entrusted with their local care seemed nearly universally determined to convince our patients their situation was hopeless and my treatment worthless. Indeed, these physicians seemed mostly motivated not to cooperate, not help manage patients through treatment crises, but rather get them into hospice or some standard chemotherapy.

Of all the nutrition patients, only one – who ultimately survived 3.5 years - received anywhere near the level of intensive supportive care and encouragement given the chemotherapy patients. In this unique situation, the local doctors coordinated their treatment with me, realizing full well that he was sustaining a most unusual response. In no other case did the local doctors encourage aggressive intervention to keep the patients alive and also on the nutritional therapy. The general physician attitude toward our patients could not be in sharper contrast to the enthusiastic team approach of Dr. Fine, whose talented staff appear to have one ultimate, determined goal with all their intensive care, to keep patients alive as long as possible and on treatment as long as possible.

At this point, I would like to comment on another aspect of the chemotherapy regimen used with most of the patients in our study, GTX, that makes comparison of the two groups in our clinical trial even more fruitless. Two of the three drugs used, Gemzar and Taxotere, must be given intravenously, since no oral form exists. Xeloda was developed for oral ingestion, usually one or two tablets daily for two weeks, followed by a two-week drug vacation, followed by more Xeloda. By the nature of these agents and their route of administration, involving very little intake by mouth, the GTX regimen suits even patients with little or no appetite – unlike our aggressive nutritional/supplement program, which requires patients follow a carefully prescribed diet and consume more than 150 capsules a day. The vastly different natures of the therapies, one primarily intravenous in form, one oral, would give GTX an advantage with very advanced patients. Those too sick to eat, who would be too ill to follow our regimen could still receive GTX. In our clinical trial, even if the patients did ultimately match equally by stage, the battle was still one sided, since so many of the nutrition patients Dr. Chabot entered could not comply with the treatment. For any group of equivalent subjects sent off to Dr. Fine, the lack of appetite, the inability to eat meant little, because the therapy could still be given even to these very advanced patients, even to those too weak to get out of bed.

Furthermore, whenever Dr. Fine needed to hospitalize a patient at Columbia for any reason, he could continue providing the chemotherapy in the institutional setting with everyone’s blessing. In contrast, though every patient assigned to the nutrition arm at some point required hospitalization, not once did the local doctors allow any of them to continue any aspect of the regimen under these circumstances. I believe this additional difference in patient care between the two arms allowed for yet another bias in GTX’s favor.

G. Different standards for stopping each treatment

In Dr. Fine’s worldview, if a patient seems to be worsening, he doesn’t stop therapy and give up, but instead modifies the doses, changes the scheduling of the drugs, becomes more aggressive. (9) In a sense, Dr. Isaacs and I have a similar attitude: if a patient doesn’t appear to be responding, with his or her consent, we change the dose of enzymes, we adjust the daily and monthly scheduling of the supplements, we might modify the diet or any other number of variables. We too, ramp up the treatment.

The written protocol for the study specifically allowed Dr. Isaacs and myself to alter the treatment as needed when scans or other studies indicated possible tumor enlargement or spread. In section 11.0 “REMOVAL OF PATIENTS FROM PROTOCOL THERAPY,” the document clearly states:

11.1 Enzyme-Nutritional Protocol Therapy:

Response or Stable Disease: Continue full treatment with full doses of enzymes and adjunctive nutrients, diet and detoxification routines.

Disease Progression: Modify the doses of enzymes and other nutrients according to the protocol but do not discontinue protocol therapy except at the patient’s request. Describe tumor progression including tumor measurements on flow sheets.

The written protocol criteria could not have been clearer, giving us the right – assuming the patient agreed – to modify our treatment, just as Dr. Fine could do with his chemotherapy regimen.

During the years of our clinical trial, Dr. Chabot seemed quite impressed by the resolve of his Columbia colleague. Yet as we came to learn, it seemed that Dr. Chabot believed our patients needed to be handled quite differently, as if two standards governed the trial - at the first sign of worsening, our nutrition patients were to be sent elsewhere for different treatment.

We learned of this double standard late into the study through a very odd series of events. By 2004, as the project progressed into its 6th year, we realized the NCI would do nothing to counter the ongoing physician hostility to the project that continued to influence both accrual of new patients, and compliance of those entered. We suggested that perhaps to alleviate this ongoing problem, Dr. Chabot himself begin calling the local physicians of patients admitted into the nutrition arm, to encourage their support. He agreed that either he or one of his staff would from that point on make such calls whenever he approved a new subject for the nutrition arm. As the study progressed into 2005, physician opposition continued, so we suspected that such calls, if indeed being made, had not had much effect.

At a meeting of all study personnel convened at Columbia on June 20, 2005, Dr. Isaacs and I asked Dr. Chabot if indeed he called uncooperative local physicians. He said he had, though he was vague about the number, and apparently called only if the patient suggested the local physician might be hostile toward the study.

Beginning early on in the study, after each group meeting Dr. Wendy Smith of the NCI distributed official minutes, summarizing the issues discussed among the various participants. We had the right to suggest corrections, if we noted errors in the document which usually – but not always - arrived for our review days, or at most, a week or so after the meeting. However, the minutes for the June 2005 meeting didn’t arrive until September 12, the day before the next regularly scheduled conference. We noticed a number of significant errors in the notes, including misstatements of Dr. Chabot’s claim about contacting local doctors. In the document, Dr. Smith reported that Dr. Chabot had been contacting physicians assigned to follow nutritional patients “from early on,” a complete fallacy.

In an e-mail sent to Dr. Smith dated September 13, 2005 I wrote:

In the second paragraph of this section, you state that Dr. Chabot telephones physicians to confirm that they are indeed supportive and has made these calls from early on. “From early on” is vague and most readers would assume based on your report that these telephone calls were made to the physicians of the majority of patients. Our recollection is that this policy was only put into place in the last year, and that only after frequent discussions of the hostility of oncologists toward the study were such calls instituted. Frankly, even if calls were made, they had no effect. Patient ----, for example, one of the later patients in the trial, reported constant negativity from his oncologist and private physician. If his physicians received a phone call, it certainly didn’t make much difference in their attitude….

It wasn’t until the March 20, 2006 group meeting that, according to the official minutes, Dr. Smith addressed the issue of Dr. Chabot’s phone calls to local physicians. She reported in the meeting notes, in a section entitled “The PI (Principal Investigator) phone calls to physicians and hostility to the trial,” Dr. Chabot’s discussion of his calls:

If there was any concern about the supportiveness, that is if the patient mentioned anything to him or if it didn’t appear that there was a cooperative physician, he would give them a call to assess whether they were interested in having their patients on this trial, whether they have a problem, and whether they are willing to care for the patient on the trial. In some situations, Dr. Chabot found that the patient had a physician supportive of their patients’ participating in the enzyme arm but upon recognition of increased disease, the physician then acted to facilitate the patient moving to other therapy. (Italics mine).

In the cases we’ve been able to document, physician opposition began before the alleged disease progression, so the explanation offered by Dr. Chabot seems simply a rationalization for bad behavior. Regardless, in terms of the issue at hand, the written protocol criteria could not be more clear, giving us the right – assuming the patient agreed – to modify our treatment, just as Dr. Fine could do with his chemotherapy regimen. Consequently, it was neither acceptable nor appropriate for an outside physician who knew nothing about our therapy or the trial requirements to make decisions about stopping the regimen or pushing any of our patients into a different treatment plan – unless again, the patient so requested. But in his comments, Dr. Chabot isn’t talking here about a change of heart on the part of the patient, he’s specifically permitting the monitoring physician to make the decision. Such outside interference constituted a violation of the written criteria for the trial, and as Principal Investigator charged with upholding the requirements of the protocol Dr. Chabot should not have justified with such ease an obvious breach of study regulations – particularly since he seemed perfectly happy with Dr. Fine’s never-give-up attitude. Unfortunately, Dr. Chabot’s two sets of standards in this case, one for Dr. Fine and one for us, not only reflected a bias – and disregard for the written protocol – but also influenced, as his own statements demonstrate, the treatment course of nutrition patients.

H. Dr. Chabot’s Conflict of Interest

In any legitimate clinical trial, the Principal Investigator must remain objective, be always neutral, and have no tie, either financial or intellectual, to any of the treatments being evaluated. For the first five years of our project, although the chemotherapy patients were being treated at Dr. Chabot’s institution, Columbia, and by his colleagues, I did not suspect any conflict of interest on his part.

I first became suspicious of a potential problem during the December 13, 2004 group meeting, when Dr. Chabot seemed overly pleased with what appeared to be unusually positive survival data in the chemotherapy group. Dr. Isaacs and I attributed these odd results to the advanced stage and poor compliance among our patients contrasted to the early stage of most in the chemotherapy group, not any great benefit of the drug treatment. As the study continued on its bizarre journey throughout 2005, I became only more troubled about Dr. Chabot’s apparent growing enthusiasm with the chemotherapy data, particularly in view of all the management issues which made the results questionable. He hardly appeared, by his manner and by his words, neutral in his position. Subsequently, I was to learn of a significant conflict of interest on Dr. Chabot’s part that should have disqualified him from serving as Principal Investigator.

To understand the nature of the conflict, I thought it might be helpful to review just how the drug regimen GTX (Gemzar, Taxotere and Xeloda) became the treatment of choice for the chemotherapy patients in our clinical trial. In 1998, as our study first came into being, the FDA had just approved Gemzar (gemcitabine) specifically for use with patients diagnosed with pancreatic cancer. The drug, developed by the pharmaceutical giant Eli Lilly, rapidly became the primary tool used by oncologists around the world for treatment of the disease, the new light on the horizon for pancreatic cancer. Gemzar hadn’t been developed at Columbia, so no one there as far as I knew had any special tie to it, either financially or intellectually.

In the original versions of the protocol, we all agreed that the chemotherapy patients should be treated with Gemzar alone, as was at the time the standard of care for the disease. Eventually, as we finalized the document, Dr. Antman suggested we permit the addition of other chemotherapeutic drugs, since, she and Dr. Chabot argued, oncologists and their patients already recognized that Gemzar by itself had proven only marginally effective. Many oncologists, they reported, already had added other drugs to their treatment regimens.

Dr. Antman and her Columbia team then modified the protocol at about the time we changed the overall format to a matched cohort design in early 2000. By that point, Dr. Fine at Columbia had already formalized a triple drug regimen, GTX, adding Taxotere and Xeloda to Gemzar for the treatment of pancreatic cancer. Eventually, since Dr. Fine supervised the treatment of all chemotherapy patients entered into the study, the majority ultimately received the GTX regimen. I want to emphasize that the combination GTX approach was a purely Columbia invention.

This gradual evolution of the regimen administered to the chemotherapy patients, from simple Gemzar to complex GTX, altered the landscape of our clinical trial enormously. We had started with a single drug used everywhere for treatment of pancreatic adenocarcinoma, then changed to a very specific triple agent combination developed and tested specifically at Columbia, by a team of Columbia physicians and scientists. To me, the shifting sands of the chemotherapy administered in our trial signaled a potential danger to a fair study, even a conflict of interest within the Columbia team assigned to our project. Now, my controversial nutritional approach, developed far removed from and often despised within the hallowed halls of academia, was pitted against a drug combination developed at the very institution chosen to be home base for the trial.

Since Dr. Chabot, the Principal Investigator, was a surgeon, not an oncologist, I didn’t initially think he would have any direct involvement with Dr. Fine’s research and GTX. I also trusted Dr. Antman, whom I believed wished for a fair and honest effort, so I didn’t object to Dr. Chabot serving as Principal Investigator. Even after Dr. Antman left Columbia for the NCI, leaving Dr. Chabot completely in charge, I didn’t think any more about GTX and its origins as a purely Columbia regimen. I believed that if a conflict existed, Dr. Chabot would have let us know.

It now appears that Dr. Fine, the supervisor of the patients admitted into the study for chemotherapy, and Dr. Chabot are more than just Columbia colleagues, two professors at a big medical center who might occasionally pass each other in the hall. As it turns out, unbeknownst to me for the years of the trial, they had worked together very closely on the GTX therapy being given the chemotherapy patients in our trial! A lengthy ONCOLOGY NEWS INTERNATIONAL article dated March 4, 2004 makes that point loud and clear. In the last paragraph, the author writes:

“Activation of specific components of the MAP kinase pathway may be a novel target for induction of apoptosis in pancreatic cancer,” Dr. Fine said. “Further study in phase II/III is warranted to ascertain the true clinical benefit of GTX and T-GX.” His colleagues for the studies were Drs. David Fogelman, John Chabot, Ronald Ennis, Stephen Schreibman, James Strauss and Yin Li. (9)

Though the interview came out in early 2004, I actually didn’t see it until 2005, when I began to sort out what had really happened during the many years of this clinical trial. Had I been aware of this interesting connection earlier, I would have asked that Dr. Chabot disqualify himself from the project.

Another article linking Dr. Chabot closely to GTX appeared in the October 2004 issue of Surgical Oncology Clinics of North America. (8) Entitled “The evolution of adjuvant and neoadjuvant chemotherapy and radiation for advanced pancreatic cancer: from 5-fluorouracil to GTX,” this lengthy and technical report puts Dr. Chabot right in the middle of the GTX action. The authors, according to the order in which they appear in the journal, include Fogelman DR, Chen J, Chabot JA, Allendorf JD, Schrope BA, Ennis RD, Schreibman SM, Fine RL. Invariably, publishing etiquette requires that the various authors of a study appear in a specified order of importance. Dr. Chabot’s position as third among eight meant he wasn’t just a casual bystander, but a major player. Furthermore, it’s important to keep in mind that in recent years, medical and scientific journals nearly universally have required that each author sign a statement acknowledging a substantive contribution to the research effort.

The very same month the Surgical Oncology Clinics of North America report appeared, the Columbia team published an in-depth description of their laboratory and early clinical finding in Current Treatment Options in Gastroenterology. (10) Once again, Dr. Chabot appeared as an author along with Dr. Fine.

I thought it would be helpful to quote from the abstract:

Familiarity with the updated results in genetic screening and work-up presented here is essential to early diagnosis and possible cure. In the metastatic setting, we most frequently begin with the GTX regimen, consisting of Gemcitabine, Taxotere, and Xeloda. The regimen is based on our laboratory data demonstrating a synergistic increase in cell killing of pancreatic cancer cell lines…In our initial experience, we have seen a response rate of 40% at metastatic sites and 31% at the primary site after nine cycles of GTX…In those patients who do not tolerate GTX or progress despite the regimen, we have found that a regimen of the same three drugs, administered on a different schedule, can produce responses... (10)

The last sentence above, written with Dr. Chabot as a co-author, not an indifferent bystander, confirms that when patients worsen on GTX, the heroic Columbia team does not refer to hospice but modifies the treatment plan.

After seeing these publications in major journals, I could conceive that it would have been in Dr. Chabot’s best professional interest to prove the chemotherapy regimen he himself helped develop worked better than my scorned and derided nutritional therapy. I am sure his Columbia colleagues, with whom he must work every day, would have been happier with this result. Did not anyone - at the NCI, NCCAM, the NIH - see the obvious conflict of interest? The Principal Investigator of this study should have been a scientist with no personal involvement with either of the regimens being evaluated. In fact, we thought that Dr. Chabot had been chosen to head the study not only because of his expertise, but precisely because he had no tie to the GTX research. But, since he turned out to have been hardly a disinterested third party, he should not have been the man in charge of this clinical trial. I am sure that had Dr. Isaacs, for example, been made Principal Investigator, the objections would have been loud and persistent.

I also now understand why Dr. Chabot seemed so excited during the December 2004 meeting, when he announced the extraordinary survival benefit among the chemotherapy patients – he was part of Dr. Fine’s team, of course he was excited! At the time, though I did think his enthusiasm for GTX excessive, I didn’t put the pieces together until I had reviewed the published literature, including the ONCOLOGY NEWS INTERNATIONAL and Surgical Oncology Clinics reports.

I believe that had the NCI been on top of this study – instead of buried in that peculiar Washingtonian bureaucratic malaise - they would have jumped on this issue years ago, insisting that Dr. Chabot be removed as PI and the trial be moved to some other institution not so closely tied to one of the competing treatments. Such intervention would have required not only some level of concern for true science and legitimate research, but actual thought and work.

Return to JCO Rebuttal

---

---

Copyright © Nicholas J. Gonzalez, M.D.
All Rights Reserved

Nicholas J. Gonzalez, MD PC
Linda L. Isaacs, MD
36A East 36th Street
Suite 204
New York, N.Y. 10016
Phone: 212-213-3337
Fax: 212-213-3414
E-mail policy

Last Modified: