Three-Gene Signature ID’s Active TB – MedPage Today
8 months ago Comments Off on Three-Gene Signature ID’s Active TB – MedPage Today
A gene expression signature that includes just three genes was highly sensitive and specific for detecting active TB infection in multiple datasets, with little confounding by HIV infection, bacterial drug resistance, or vaccination against the disease, researchers reported.
The three-gene diagnostic set — identified through analysis of 266 genes found to over- or under-express in active, compared to latent, tuberculosis — was found to be highly diagnostic for active tuberculosis in cohorts that included children and adults, and healthy controls and people with latent TB infection and other diseases.
A test based on the three-gene set has the potential to both advance the detection of active tuberculosis in areas of the world where the disease is endemic and better monitor patients being treated for TB, researcher Purvesh Khatri, PhD, and colleagues from Sanford University School of Medicine wrote in Lancet Respiratory Medicine, published online Feb. 19.
In an interview with MedPage Today, Khatri explained that while a highly sensitive and specific test for active tuberculosis already exists, the test involves sputum sampling.
“It is difficult to get small children to cough up enough sputum to get a usable sample,” he said. “That is one problem. Another problem is that once treatment starts and patients become less symptomatic, they may not generate enough sputum.”
The World Health Organization has called for a new nonsputum-based diagnostic test for active TB that is simple to run with a target sensitivity of more than 80% in patients with HIV co-infection and greater than 66% in children with culture positive tuberculosis.
Khatri noted that earlier attempts to develop a blood test to distinguish active TB from latent TB and other diseases have involved gene sets that turned out not to be generalizable or were complicated and cost prohibitive.
“We were looking for something that would be cost effective that could work in (low resource) areas,” he said.
Khatri and colleagues hypothesized that integrating publicly available gene expression data from heterogeneous patient populations with active tuberculosis across various ages, countries, and inclusion criteria “would yield a set of conserved genes that are indicative of active tuberculosis” that could be generalized across cohorts.
They searched two public gene expression microarray repositories and retained datasets that examined clinical cohorts of active tuberculosis infection in whole blood. Using their own validated multi cohort analysis framework, they compared gene expression in patients with either latent tuberculosis or other diseases versus patients with active TB.
Three datasets were used as discovery datasets, meta analysis was used to examining gene effects, and the diagnostic capacity of the three-gene set was validated in 11 other datasets.
The 14 total datasets contained 2,572 samples from 10 countries, and they included both adult and pediatric subjects. Three datasets (n=1,023) were used to discover the set of three genes — GBP5, DUSP3, and KLF2 — that were highly specific for active tuberculosis.
The researchers then validated the diagnostic power of the three-gene set to separate active tuberculosis from healthy controls (global area under the ROC curve [AUC] 0.90, 95% CI 0.85–0.95),latent tuberculosis (0.88, 95% CI 0.84-0.92),and other diseases (0.84, 95% CI 0·80-0.95) in eight independent datasets that included both children and adults from 10 countries.
Expression of the three-gene set was not confounded by HIV infection status, bacterial drug resistance, or tuberculosis vaccination. Analysis of four additional datasets showed that the tuberculosis scores declined during treatment of patients with active tuberculosis.
Although HIV-positive patients had a lower AUC for latent tuberculosis versus active tuberculosis, it was still high (0.97 HIV-negative, 0.89 HIV positive).
The researchers concluded that the three-gene set could be developed as a cost effective, easy to read, clinical test with a single cut-off for diagnosing active TB. But they also conceded that a different gene set with similar diagnostic power could still be identified. They also noted that the global ROCs required a re-centering of mean to accommodate for changes in baseline gene expression measurement by different technologies, which was a study limitation.
“However, such a centering is justified because in a real-world application of the three-gene set, the same technology with a global mean will be used across all cohorts,” they wrote.
The researchers noted that the cost of a test based on the three-gene set would probably be comparable to that of the sputum-based GeneXpert MTB/RIF (Cepheid) TB test — about $10 to $20 per cartridge.
Lead researcher Timothy E. Sweeney is a scientific adviser for Multerra Biosciences.
Sweeney and Khatri have patents pending on the technology developed through this research, filed by the Stanford University Office of Technology Licensing.
F. Perry Wilson, MD, MSCE Assistant Professor, Section of Nephrology, Yale School of Medicine
Three-Gene Signature ID’s Active TB – MedPage Today