Xpert Ultra assay shows superior sensitivity for TB in patients with HIV

The Xpert Ultra assay had superior sensitivity compared with Xpert for detecting tuberculosis in patients with paucibacillary disease and patients with HIV, study data showed.

“WHO recommended the Xpert MTB/RIF assay (Cepheid, Sunnyvale, CA), an automated, integrated, catridge-based molecular assay, as the initial test for tuberculosis to increase case detection and improve identification of rifampicin resistance directly from sputum,” Claudia M. Denkinger, MD, head of TB at FIND, Geneva, Switzerland, and colleagues wrote. “However, Xpert’s sensitivity for tuberculosis detection is inadequate when few bacilli are present in a clinical specimen. … The Xpert MTB/RIF Ultra assay (Xpert Ultra) was developed to overcome the limitations of the Xpert assay.”

The researchers conducted a prospective, multicenter diagnostic accuracy study of adults with symptoms of pulmonary TB from South Africa, Uganda, Kenya, India, China, Georgia, Belarus and Brazil. Patients who had not taken any TB medications for the past 6 months were put in the case detection group, and those who had taken TB medications were put in the multidrug-resistance risk group. The researchers recorded medical history, chest imaging results, HIV test results and demographic information at enrollment. Each patient provided at least three sputum samples on 2 different days. Denkinger and colleagues compared Xpert and Xpert Ultra diagnostic performance with culture and drug susceptibility testing. The main outcomes were to compare the sensitivity of Xpert with Xpert Ultra for detecting rifampicin resistance and smear-negative TB, and to compare the specificity of each assay for detecting rifampicin resistance. Patients in the multidrug-resistance group (n = 314) were included only in rifampicin-resistance detection analyses, whereas those in the case detection group (n = 1,439) were included in all analyses.

For patients with smear-negative a culture-positive sputum (n = 137), Xpert Ultra demonstrated a sensitivity of 63%, compared with 46% in Xpert, the researchers reported (difference, 17%; 95% CI, 10%-24%). The sensitivities were 90% and 77%, respectively, for patients with HIV who had culture-positive sputum (n = 115; 13%; 95% CI, 6.4%-21%), and 88% and 83% in all patients with culture-positive sputum (n = 462; 5.4%; 3.3%-8%).

However, Xpert UItra demonstrated a specificity of 96% overall, compared with 98% in Xpert (difference, –2.7%; –3.9% to –1.7%), the researchers reported. It was also less specific in patients who had a history of tuberculosis (93% vs. 98%). The two assays performed similarly in the detection of rifampicin resistance.

“Metaphorically speaking, if Xpert is a knife, Xpert Ultra is a sharper knife,” Sandra M. Arend, MD, PhD, professor at Leiden University Medical Center, Leiden, Netherlands, and Dick van Soolingen, PhD, of the National Institute for Public Health and the Environment, Bilthoven, Netherlands, wrote in an accompanying editorial. “However, the trade-off of the higher sensitivity is a lower specificity, as is customary for diagnostic tests. We eagerly await additional studies and algorithms that assess Xpert Ultra’s improved sensitivity and solutions to avoid the trap of treating patients for dead bacilli.” – by Andy Polhamus

Disclosures: Arend and van Soolingen report no relevant financial disclosures. Denkinger is an employee of FIND, a nonprofit organization that supports the evaluation of publicly prioritized TB assays and the implementation of WHO-approved assays using donor grants. The foundation has agreements with several private companies to review evaluate diagnostic products independently and neutrally.