Cell Culture Drug Resistance Testing (CCDRT) Cell Death Assays:
Misconceptions Versus Objective Data
- chapter 7 -
Review of Arguments Offered in Opposition to CCDRT-Directed Chemotherapy
The application of CCDRT in the non-investigational management of cancer patients has been criticized or at least not supported in a number of venues. Most prominently, there are:
A. Two recent "negative" editorials written by Dr. Maurie Markman of the Cleveland Clinic, in conjunction with a physician employee of the Aetna Insurance Company.
B. A negative American Medical Association "Tech Review."
C. A negative review by the national Blue Shield "TEC" committee
D. A recent review published in the Journal of Clinical Oncology by Drs. Cortazar and Johnson.
E. The officially stated policy of the National Cancer Institute
F. The "Oncotech Reporter," Issue 5, July, 1999.
Counterbalancing the above are:
A. Two different "postive" reviews by Blue Shield of California, which, as noted in the introduction on page 1, performed the most comprehensive review yet undertaken after uniquely inviting the participation of all of the leading proponents, as well as critics.
B. Postive position statements by both Southern and Northern California Medical Oncology Associations.
C. The administrative law judge ruling fully favorable to the non-investigational application of CCDRT within the
Before examining the arguments against CCDRT, it is important to raise the issue of what standard should be applied in determining the validity of different technologies and the clinical indications for their use. This means what is more important: (1) the opinion of the largest number of "experts?" or (2) the weight of published data? More directly: should opinion trump peer-reviewed scientific data?
In answering this question, it is advisable to look at recent and relevant court decisions.
Of critical importance is a recent U.S. Supreme Court decision. This is the 1993 case of Daubert v Merrel Dow Pharmaceuticals, Inc (113 S Ct 2786, 1993). I believe that this is very relevant to the issues which I now raise.
In evaluating CCDRT claims, insurance companies appropriately go and ask their resident expert, who may say something such as "oh, that stuff is no good and nobody believes in it." So they invoke the "Frye Standard," which basically states that what counts is whether or not something is generally accepted in the particular field. In the past, cell culture testing was not generally accepted, so the companies tried to deny coverage for it. We have, in the Southern California managed care industry, largely moved past this point of view by accumulating a "critical mass" constituency which supports this testing, but, with individual insurance companies in other areas of the country with physicians unfamiliar with our services, the above ("negative") point of few is still prevalent.
But the Daubert decision basically tosses out the Frye standard, and this has tremendous implications for the legal adjudication of coverage disputes.
Let me elaborate.
The "Frye Standard" came from a court case 74 years ago (Frye v United States (293 F2d 1013, DC Cir 1923). In the Frye decision, the court determined that the major factor to be used when determining admissibility of scientific evidence was whether the evidence had gained general acceptance in the particular field in which it belongs.
To broaden the Frye standard to make it less restrictive, Congress in 1974 enacted the Federal Rules of Evidence (FRE), which includes Rule 702: "If scientific, technical, or other specialized knowledge will assist the trier of fact to understand the evidence or to determine a fact in issue, a witness qualified as an expert may testify thereto in the form of an opinion or otherwise."
The new judicial guidelines relating to expert testimony formulated by the Supreme Court in Daubert v Merrell Dow made it clear that the Frye standard NO LONGER APPLIES.
The Daubert case involved Bendectin use and congenital limb deformities. The plaintiffs produced scientific evidence based on in vitro and animal studies, an analysis of the chemical structure of Bendectin and other teratogens, and on unpublished, non-peer reviewed reanalysis of previously published human statistical studies.
The trial and appellate courts rejected the plaintiff's theory, claiming that it failed to meet the standard of "general acceptance." The Supreme Court overruled both lower courts and stated that the correct standard was whether the "reasoning and methodology is scientifically valid," rather than on whether it is generally accepted.
The following parameters are allowed to be considered:
2. Whether the known and potential error rates have been established.
3. Publication or peer review of methods and results.
4. Widespread acceptance in a relevant scientific community (the sole pre-existing standard).
What has happened is that the "widespread acceptance" used to be by far the most important standard. Now, it is only one of the parameters to be considered and is no more important than each of the other three parameters. Where criteria 1-3 make a strong case for the evidence, the lack of 4 cannot be used to dismiss the validity of the arguments and evidence.
It is important to re-focus the thinking on this away from the "popularity contest" point of view onto the evidentiary point of view.
Here's what is relevant:
1. Every single study (more than 40 such studies) ever published shows that patients treated with "assay positive" drugs in cell death (apoptotic) assays do much better than patients treated with "assay negative" drugs and, in fact, do better than the patient population as a whole. Not one critic has ever disputed this unassailable fact. Thus, the methods have been tested. The assays can clearly identify drugs which offer the patient an above-average prognosis, as well as being very good, indeed, at identifying drugs which offer the patient a poor prognosis.
2. The known error rates have been well established.
3. These results have been both published and peer reviewed.
4. The quality and quantity of data establishing the above are comparable or superior to that of many other accepted laboratory tests and the historic standard for approving laboratory tests has always been accuracy of clinical correlation and not improved treatment outcome. I would maintain that the collective strength of the scientific data relating to the technologies and clinical indications which I advocate are beyond serious challenge (one would have to postulate a truly global conspiracy to fabricate and publish data).
5. The above notwithstanding, to the extent that clinical information is available, it may be unanimously concluded that assay-directed therapy is better than empiric therapy, using the best available control group for each individual study and considering limitations such as the limited statistical power of some very small studies to show differences.
6. For almost every type of cancer, there are a number of potential treatments of equal efficacy if chosen empirically.
7. Choice of treatment is a life or death decision.
8. Understanding the above, it is eminently reasonable that patients and physicians would chose to use these tests on a case by case basis.
9. This issue was previously considered by an expert Blue Shield panel in 1994 and unanimously approved, after detailed, pro and con commentary and testimony. New data since that time serve only to strengthen the above conclusions. The issue was again re-examined in detail in October, 1997, and detailed pro and con arguments were again heard. The policy to continue coverage was unanimously re-affirmed.