Diagnostic Accuracy: A Point-of-Care Tuberculosis Test for Those With HIV

Compared with Alere Determine TB LAM Ag (AlereLAM), the only commercially available tuberculosis point-of-care test, the novel Fujifilm SILVAMP TB LAM (FujiLAM) assay offers superior diagnostic sensitivity while also maintaining specificity, according to a study published in The Lancet: Infectious Diseases. The applicability of FujiLAM in various settings requires prospective assessment, but the test could be very beneficial for hospital inpatients with HIV.

Most cases of tuberculosis-related deaths among patients with HIV could be prevented with early diagnosis and treatment, but the commercially available AlereLAM has suboptimal sensitivity, restricting its use in clinical practice. The FujiLAM assay was developed to improve AlereLAM sensitivity. This diagnostic accuracy study was designed to compare the 2 tests using biobanked urine samples from the Foundation for Innovative New Diagnostics (FIND) Specimen Bank and the University of Cape Town Biobank. These samples were collected from adult hospital inpatients with HIV from 3 independent cohort studies done at 2 South African hospitals. The FujiLAM and AlereLAM assays were conducted and reported blind to other test results. Diagnostic accuracy was determined against composite and microbiologic reference standards.

Urine samples from 968 hospital patients with HIV were tested, and 62% (n=600) showed microbiologically confirmed tuberculosis, with a median CD4 count of 86 cells per μL. When the microbiologic reference standard was used, FujiLAM sensitivity was estimated at 70.4% (95% CI, 53.0-83.1) compared with 42.3% for AlereLAM (95% CI, 31.7-51.8), for a difference of 28.1%. The estimated specificity of FujiLAM was 90.8% (95% CI, 86.0-94.4) compared with 95.0% for AlereLAM (95% CI, 87.7-98.8), for a difference of -4.2%.

When the composite reference standard was used, specificity was higher for both assays: 95.7% for FujiLAM (95% CI, 92.0-98.0) vs 98.2% for AlereLAM (95% CI, 95.7-99.6), for a difference of -2.5%. However, the sensitivity was lower at 64.9% for FujiLAM (95% CI, 50.1-76.7) and 38.2% for AlereLAM (95% CI, 28.1-47.3), for a difference of 26.7%. In a comparison of diagnostic yield comparing the 2 assays with other rapid diagnostic tests performed within the first 24 hours of hospital admission, 65% (n=91) of 141 cases of tuberculosis could have been diagnosed with FujiLAM within a few hours of presentation compared with 43% (n=61) of cases with AlereLAM. A combination of FujiLam and sputum Xpert within the first 24 hours would have diagnosed 72% (n=102) of microbiologically confirmed cases, and a combination of FujiLAM and sputum smear microscopy would have diagnosed 70% (n=98) of cases.

Study investigators noted that studies of tuberculosis diagnostic assessments frequently exclude patients unable to produce sputum, which biases the study and excludes the population that would most benefit from nonsputum-based tests. Researchers concluded that “considering the higher sensitivity and rapid, point-of-care design of FujiLAM, this assay has the potential to transform the diagnosis of tuberculosis in hospital inpatients with HIV and, potentially, for people with HIV in the general population.”

Multiple authors are employed by the nonprofit FIND, 2 authors report patents in the field of lipoarabinomannan detection, and multiple authors report receiving grants from a variety of sources.

Reference

Broger T, Sossen B, du Toit E, et al. Novel lipoarabinomannan point-of-care tuberculosis test for people with HIV: a diagnostic accuracy study [published online May 30, 2019]. Lancet Infect Dis. doi: 10.1016/S1473-3099(19)30001-5