Clinical Trials of Laboratory Tests

 

The standard for judging laboratory tests has always been accuracy, reproducibility, and utility.  Historically, laboratory tests have never been judged by their effect upon treatment outcomes.  For example, estrogen and progesterone receptor testing (ER and PR) to guide the use of hormonal therapy drugs such as tamoxifen has never been shown in a prospective randomized trial to improve response or survival in breast cancer.  It is not that the studies were performed and the tests failed to show treatment benefit – it is only that prospective, randomized trials are never carried out for the purpose of demonstrating improved outcomes on the basis of having performed certain laboratory tests .  On the other hand, no physician would be surprised to learn that ER expression has been shown in retrospective studies to correlate positively with response to hormonal treatments. 

 

Despite the fact that no laboratory test ever has been shown in a prospective, randomized clinical trial to improve patient outcomes, many such tests are considered invaluable.  This is because the tests are accurate and reproducible and – based upon retrospective correlation studies - the tests are known to correlate reliably with the clinical or biological phenomena which they are intended to detect, measure, or predict. 

 

Functional Tumor Cell Profiling has been subjected to greater scrutiny in clinical trials than most other tests used routinely in oncology – not only in terms of accuracy and reproducibility but also with respect to correlation with clinical outcomes.  Data from more than 40 studies uniformly demonstrate close correlation between prospective predictions of drug activity and patient chemotherapy response - and also, where such follow-up data are available, with overall survival.  In meta-analysis of cell death-based Functional Tumor Cell Profiling studies, an assay “positive” drug was 7 – 9 times more likely to produce objectively-measured patient response than an assay “negative” drug (click here). 

 

Functional Tumor Cell Profiling, therefore, can identify for each patient drugs which have greater than average likelihood of producing clinical benefit and also drugs which have less than average likelihood of producing clinical benefit.  The claim is modest and unambiguous.  Study after study has borne out the claim and no person who is familiar with the large body of published literature which supports functional tumor cell profiling has ever disputed it.  Many physicians today order the tests routinely for their patients because they continually see concrete benefits.  The burden of proof, therefore, now falls upon those who would withhold the use of such useful and accurate testing to explain why it is that slight suggestions of efficacy derived from scant and largely non-reproducible retrospective studies of molecular tests - in which response alone and not survival is the endpoint - are vaunted as sufficient justification for immediate clinical adoption while inappropriate and never-before-applied standards of proof for Functional Tumor Cell Profiling, which already is far more well-studied than molecular tests, are manufactured and endlessly debated. 

 

In February 2007, National Heritage Insurance, the Medicare administrator for the state of California, concluded a year-long, comprehensive review of data relating to Functional Tumor Cell Profiling and ruled in favor of complete coverage of the service for Medicare patients (click here to see the ruling), even while acknowledging an earlier and much-publicized ASCO review which had selectively ignored precisely those data which are most relevant - those which demonstrate the accuracy, reliability, and utility of cell death-based Functional Tumor Cell Profiling.  (For an analysis of ASCO's limited review, please click here.)