George D. Lundberg M.D. Chief Medical Officer CollabRx
Clifford Baron Ph.D. Chief Operating Officer CollabRx

Various Methods Allow Clinical Laboratories to Maximize Their Efficiency and Usefulness

The delineation of a brain-to-brain loop in clinical laboratory testing first published in 1981 has never been more pertinent. Its subsequent development and current application in clinical molecular oncology in 2016 can make all the difference.

This discussion focuses on the factors that drive the ordering of a lab test and the many components thereof, itemizes pre- and postanalytic causes of diagnostic error, and recommends how a laboratory can help ensure the usefulness of the entire process.

Just as a chain is no stronger than its weakest link, a loop that isn’t closed is (obviously) still open.

Technical and laboratory workers tend naturally to define their work by their technical products and procedures, as well they should. In clinical laboratory testing, that tends to be the step called “analysis.”

The success or failure of an “analysis” may well depend upon the pre- and postanalytic phases at least as much as the analysis itself.

In response to a tragic death of a young man in Los Angeles in 1969 because of a failure to recognize and act promptly on a critical laboratory value at the far end of the laboratory testing process, we created the concept of a complete brain-to-brain loop as essential components of any patient-focused clinical laboratory test. In 1981, one of us (GDL) described a brain-to-brain loop for all clinical laboratory testing in an editorial in the Journal of the American Medical Association entitled “Acting Upon Significant Laboratory Results” (JAMA 1981;245:1762–1763, doi:10.1001/jama.1981.03310420052033).

In 2011, we published a 40-year follow-up paper in the American Journal of Clinical Pathology (“The Brain-to-Brain Loop Concept for Laboratory Testing 40 Years after its Introduction,” AJCP 2011;136:829–833).

Good and Not-So-Good Reasons for Ordering Laboratory Tests

There are seven principal established reasons—diagnosis, screening; monitoring, previous abnormal result, prognosis, education, and medical/legal.

But there are many more actual reasons in practice:

To confirm a clinical opinion, establish a baseline, complete a database, curiosity, insecurity, public relations, documentation, peer pressure, patient pressure, pressure from recent literature, question of accuracy of prior result, unavailability of prior result, personal education, research, personal reassurance, to show to an attending physician, hospital policy, institutional marketing, state legal requirement, concern for liability, CYA, personal or hospital profits, fraud and kickbacks, hunting or fishing expeditions, frustration at nothing better to do (don’t know what’s wrong with this patient, better get some lab tests), to buy time (maybe by the time the lab tests come back I will have some better ideas what is wrong with this patient), simple availability, and ease of doing.

The nine brain-to-brain loop steps include ordering, specimen collection, patient and specimen positive identification, transportation, and separation (or preparation) followed by analysis, reporting, interpretation, and action. Unless the action taken benefits the patient, the entire exercise is likely to be at best a waste, at worst, a tragedy.

With molecular oncology tests, there are particularly vexing issues in the pre- and postanalytic loop. In addition to the “why” should this test be done, the “where” includes specimen and laboratory choices. Should it be a blood or tissue sample? If blood, should circulating tumor cells or cell-free DNA be sought for study? If tissue, should it be the primary tumor or a metastasis, or a sample from each metastatic location? Should the sample be fresh or frozen or formalin fixed or from a paraffin block? When should the sample be taken? How large must it be?

Should the sample and testing begin at the original diagnosis of a potentially lethal cancer or only after metastasis has occurred? Or should a blood biopsy or urine, saliva, or feces be the sample considered to either screen for (any) cancer in asymptomatic patients or those thought to be at greatest risk? Or to monitor the effects of therapy and any evidence of posttreatment recurrence? Tumor heterogeneity and inevitable low-frequency subclonal cell populations are crucial considerations.

How should a physician choose the best laboratory to which to submit the specimen? Should it be a commercial reference lab, an academic medical center research lab, a cancer center lab, an integrated delivery health system lab, or a community hospital lab?

How should a medical professional choose where the cancer tumor testing should be done? Should the number of genes that should be included in a “cancer gene panel” drive that decision? One, 2–5, 6–10, 11–20, >30, >50, >200?

Should price and reimbursement for the test be an upfront consideration? Must there be preauthorization from the likely payer?

At the far end, must there be pre-authorization approval for use of a therapeutic drug?

In addition to the standard “wet lab” execution, the ordering physician (and patient) must be concerned about base-calling, variant calling, and variant annotation. Is there a “gold standard”? Therapeutic guidance, prescribing information for on-label or off-label (approved or investigational) drugs or enrollment in most suitable clinical trial is critical.

Is there a role for Molecular Tumor Boards? Who should convene such? Participate in such?

If a clinical trial seems to be the best action to be taken for a given patient, who should make that happen? The pathologist, the clinical oncologist, the health system, the patient or family or some third-party “cancer navigator”?

Correct and Not-So-Correct Responses to Laboratory Tests

The most “correct” action would be for the physician to:

  1. Inform the patient of the results, describe the context, discuss treatment options, frame costs and benefits, and jointly choose the best option.

Unfortunately, the other 12 common physician actions include:

  1. Receive but take no action.
  2. Fail to receive/recognize/comprehend the result.
  3. Receive, fail to understand, disbelieve, doubt, and ignore.
  4. Receive, disbelieve, doubt, and repeat the test.
  5. Compare to prior test results.
  6. Seek a formal consultation from another physician.
  7. Inquire of the lab what the result means.
  8. Ask another physician what the test result means and what to do with it.
  9. Order another different, but related test.
  10. Go to a book, journal, guideline, internet site, or smart phone search engine for interpretive help.
  11. Read and consider any interpretive information received with the test result.
  12. Consult with the insurance company about financial coverage for any further tests or treatments.

Proper implementation of the full brain-to-brain loop concept includes that the laboratory routinely provides laboratory request and reporting formats that “guide and teach.”

Finally, at the farthest end of the brain-to-brain loop, how is compliance with the therapeutic regimen to be assured and what is the ultimate outcome realized from the initial decision to obtain a molecular laboratory test? Outcome is the 10th step in the brain-to-brain loop, foretold in the JAMA article in 1998 but first published in AJCP in 2014.

The goal is to perform the right test on the right patient at the right time with results that benefit the patient.

George D. Lundberg, M.D. ([email protected]), is Chief Medical Officer at CollabRx, and Clifford Baron, Ph.D., is Chief Operating Officer at CollabRx.  

This article was originally published in the April 2016 issue of Clinical OMICs. For more content like this and details on how to get a free subscription to this digital publication, go to

Previous articleMedImmune, Regeneron Launch Cancer Collaboration
Next articleLipids Found Critical to Breast Cancer Cell Proliferation