NIH discusses IgeneX unreliability for Lyme testing

Below is a statement from the NIH’s National Institute of Allergy and Infectious Diseases (NIAID) about the unreliability of IgeneX testing.

IgeneX testing has also been cited as problematic by the FDA.

NIH- National Institute of Allergy and Infectious Diseases

February 14, 2001

Reliability of Blood and Urine Tests for Lyme Disease

In 1996, the National Institute of Allergy and Infectious Diseases (NIAID) awarded a five-year contract to the New England Medical Center (NEMC), Boston, MA, to study how chronic Lyme disease develops and how to improve its treatment. The proposed study included an evaluation of old and new tests used to diagnose and manage Lyme disease and other tick-borne diseases. The investigators included this evaluation because diagnostic tests for Lyme disease have been controversial due to problems with their ability to detect evidence of the Lyme bacterium and to distinguish it from closely related bacteria; the lack of standardized materials used to perform the tests; and variation within and among laboratories performing the tests.

In the February 15 issue of the American Journal of Medicine, Mark S. Klempner, M.D., of NEMC and his scientific collaborators in New York, Connecticut and Washington, D.C., report their findings about the reliability of two Lyme disease tests: an IgG Western blot blood test and the Lyme urine antigen test, or LUAT. The IgG Western blot is a licensed test used to screen blood samples for antibodies to the Lyme bacterium, Borrelia burgdorferi. The LUAT detects proteins derived from the bacterium in urine samples. Although the LUAT has not been approved by the Food and Drug Administration as a valid diagnostic test for Lyme disease, it is widely used, and the NIAID Lyme Disease Advisory Panel asked that it be further evaluated.

To perform the IgG Western blot, the investigators collected blood samples from 21 patients who met the case definition of Lyme disease established by the Centers for Disease Control and Prevention (CDC). All patients had been treated for acute Lyme disease symptoms with recommended antibiotics, but they had continued to suffer symptoms of fatigue, muscle and joint pains, or neurocognitive problems for at least six months thereafter. These 21 blood samples were analyzed and compared with those from 10 healthy volunteers who had no history or symptoms of Lyme disease.

Testing was performed at NEMC, a laboratory recognized by CDC as qualified to do so. In accordance with current recommendations, the IgG test was considered positive if it revealed 5 or more of the 10 bands (indicative of antibodies) deemed significant for Lyme disease.

Duplicate fractions of blood from the 21 patients with chronic Lyme disease symptoms were tested at separate times. The laboratory performing the test was not aware of the results from the previous test. On initial testing, two-thirds (14/21) of the patient samples tested positive for Lyme disease by standard IgG Western blot criteria; the remainder tested negative. The same results were obtained when the duplicate fractions of the same 21 samples were tested. All samples from the same 10 healthy volunteers tested negative.

The LUAT was performed by the test’s manufacturer (IgeneX, Palo Alto, CA) on urine samples taken from 10 healthy volunteers who had no history or symptoms of Lyme disease. Each urine sample was divided into five equal fractions, or replicates, so that a total of 50 individual LUATs were performed. The LUAT results, unlike the highly reproducible IgG Western blot results, varied markedly. At least one urine fraction from each of the 10 samples examined tested false-positive. Two urine samples consistently showed false-positive results. Replicates for the eight remaining samples examined were a mixture of positive and negative values, making it impossible to conclude that they were positive or negative.

The study results indicate that the IgG Western blot is a highly reproducible test for detecting antibodies to the Lyme disease bacterium. These antibodies indicate prior exposure to B. burgdorferi. In contrast, the variable results obtained with the LUAT indicate that it is not a reliable test for evaluating patients with active or suspected Lyme disease.

Although not part of the published report, the investigators also examined the reliability of standard tests for three other tick-borne diseases — tick-borne encephalitis, babesiosis and ehrlichiosis — and found all three to be highly reproducible when conducted in experienced laboratories.

Klempner MS, et al. Intralaboratory reliability of serologic and urine testing for Lyme disease. American Journal of Medicine 110(3): 217-19 (2001).

Additional resources

Dr. Jesse Goodman, MD, MPH: The Diagnosis of Lyme Disease: Good News, Bad News (commentary on the above-discussed Klempner study)

American Journal of Medicine: Correspondence regarding the above-discussed Klempner study

CDC: Laboratory tests that are not recommended

CDC: Notice to Readers: Caution Regarding Testing for Lyme Disease

Nova Scotia Infectious Diseases Expert Group: Guidance for Primary Care and Emergency Medicine Providers in the Management of Lyme Disease in Nova Scotia

Fallon BA, et al. A comparison of lyme disease serologic test results from 4 laboratories in patients with persistent symptoms after antibiotic treatment. Clinical Infectious Diseases. 2014;59(12):1705-1710.

Dattwyler RJ, Arnaboldi PM. Comparison of lyme disease serologic assays and lyme specialty laboratories. Clinical Infectious Diseases. 2014;59(12):1711-1713.

LymeScience: How chronic Lyme recruits followers

Patrick DM, et al. Lyme Disease Diagnosed by Alternative Methods: A Phenotype Similar to That of Chronic Fatigue Syndrome. Clin Infect Dis. 2015;61(7):1084-1091.

Dr. Rakel Kling, MD, Dr. Eleni Galanis, MD, Dr. Muhammad Morshed, PhD, Dr. David M. Patrick, MD: Diagnostic testing for Lyme disease: Beware of false positives

San Francisco Chronicle: Agency could shut lab tied to S.F. medical examiner

San Francisco Gate: Peninsula lab gets a clean bill of health

San Francisco Gate: S.F. coroner’s aide sues over his ouster

New York Times: Unproved Lyme Disease Tests Prompt Warnings

Dr. Mark Crislip: Lyme Testimony

Forbes: Lyme Inc

Forum discussion: Study finds Igenex has a 57.5% FALSE positive rate – I’m horrified!

ABC Australia: Correspondence from CDC and IgeneX

Public Health Wales Observatory: Is there any evidence that testing for Lyme disease using the IGENEX or ARMINLABS facilities is superior to what is available in the UK?

FDA: The Public Health Evidence for FDA Oversight of Laboratory Developed Tests: 20 Case Studies

Beth Daley and colleagues writing for the New England Center for Investigative Reporting and Boston Globe, resulting in Daley winning a journalism award:

US Department of Labor Employees’ Compensation Appeals Board:  In the Matter of BEVERLY H. STOUT and DEPARTMENT OF THE ARMY, CORPS OF ENGINEERS, Mobile, AL

US Department of Labor Employees’ Compensation Appeals Board: In the Matter of CAROL L. DAVIS and DEPARTMENT OF AGRICULTURE, FOREST SERVICE, CLEARWATER NATIONAL FOREST, Orofino, ID

Table 2 from Klempner’s evaluation of IgeneX testing:

Results of Lyme Urine Antigen Testing in 10 Control Subjects (A-J) without Lyme Disease

SampleTest InterpretationAntigen Level*
A-1Highly positive208
A-2Highly positive142
A-3Highly positive228
A-5Highly positive>400
B-1Highly positive259
B-3Highly positive371
B-4Highly positive59
B-5Highly positive151
C-5Highly positive92
D-3Highly positive202
D-4Highly positive78
D-5Highly positive188
E-1Highly positive166
E-2Highly positive248
E-3Highly positive248
E-4Highly positive309
E-5Highly positive242
G-1Highly positive122
G-3Highly positive76
G-5Highly positive73
H-3Highly positive48
I-2Highly positive126
I-3Highly positive110
I-5Highly positive144
J-1Highly positive64
J-3Highly positive133
J-5Highly positive58

The SDs of the antigen levels in the 10 subjects were as follows: subject A = 136 ng/mL, subject B = 150 ng/mL, subject C = 39 ng/mL, subject D = 91 ng/mL, subject E = 51 ng/mL, subject F = 18 ng/mL, subject G = 44 ng/mL, subject H = 21 ng/mL, subject I = 62 ng/mL, subject J = 39 ng/mL.