Summary: Next-generation genetic sequencing of brain tissue samples enabled researchers to identify or rule out potential brain infections in a clinical proof-of-principle study.
Source: Johns Hopkins Medicine.
Next-Generation Sequencing Helps Diagnose Brain Infections
A multidisciplinary team of physicians and bioinformatics specialists at Johns Hopkins University has demonstrated that next-generation sequencing (NGS) of brain biopsy samples can successfully identify infectious agents or help rule out infection in patients with suspected central nervous system (CNS) infections. Published online June 13 in Neurology: Neuroimmunology & Neuroinflammation, the study highlights how large-scale DNA and RNA sequencing combined with rapid computational analysis can become a valuable tool for neuropathology laboratories.
More than half of inflammatory brain disorders and suspected brain infections remain undiagnosed using conventional clinical and microbiological methods, often leading to symptomatic treatments that may be ineffective or even harmful. By applying unbiased sequencing and a computational pipeline to search for microbial genomes in tissue, the Johns Hopkins team aimed to improve diagnostic accuracy, speed, and cost-effectiveness.
“By incorporating modern genetic sequencing techniques into pathology diagnostics, we were able to investigate the potential presence of infection in 10 subjects and found appropriate explanations of clinical problems in eight out of 10 patient cases examined in this study,” says Carlos Pardo-Villamizar, M.D., associate professor of neurology at the Johns Hopkins University School of Medicine. “We hope to develop this technique further as a way to bring the diagnosis rate of inflammatory brain disorders and infections closer to 100 percent so we can treat patients more effectively.”
Study Design and Methods
The prospective pilot study enrolled 10 patients treated at The Johns Hopkins Hospital—six men and four women, ages 16 to 68—who presented with clinical signs suggestive of brain infection, such as fever, sudden neurological decline, focal weakness, numbness, headache, or seizures. Each patient underwent a biopsy of an MRI-identified brain lesion.
Using commercially available next-generation sequencing platforms, researchers generated millions of DNA and RNA reads per sample. They compared sequence data to a comprehensive reference set that included human genomes as well as thousands of bacterial and viral genomes. After filtering out human sequences, the pipeline ranked the most abundant nonhuman species in each biopsy to identify candidate pathogens.
Key Findings
Sequencing clearly identified the causative agent in three patients: Mycobacterium tuberculosis (the bacterium that causes tuberculosis and can infect the brain), JC virus (a virus that typically causes disease in immunocompromised individuals), and Epstein-Barr virus (a common herpesvirus linked to mononucleosis and other conditions). For these cases, standard validation methods confirmed the NGS results.
In two additional patients, NGS pointed to possible infectious organisms. One biopsy contained DNA matching Delftia acidovorans, a rare rod-shaped bacterium that guided antibiotic treatment. In another case, initial analysis detected numerous sequences assigned to Lactococcus, a bacterium not typically associated with brain infection. When the team reanalyzed that sample a year later against an expanded reference database, the sequences best matched Elizabethkingia, a soil-associated bacterium that has been implicated in meningitis in vulnerable populations; the Elizabethkingia genome had not been included in the original database.
In five patients, sequencing did not identify a specific pathogen. Those negative results were nonetheless clinically valuable: they helped clinicians rule out infection and avoid treatments—such as high-dose immunosuppression with corticosteroids—that could worsen an undiagnosed infection. “If someone has an infection, then you don’t necessarily want to give them steroids to reduce inflammation. You’ll make the infection worse by knocking down the immune system,” Pardo-Villamizar notes. “Having a correct diagnosis is the only way to ensure the best, individualized precision treatment.”
Implications, Limitations, and Future Directions
Steven Salzberg, Ph.D., Bloomberg Distinguished Professor and co-author, emphasizes an important limitation of NGS diagnostics: they can only detect pathogens whose genomes are present in reference databases. “As we continue to sequence the genomes of more organisms, the tool will become steadily more powerful,” he says. The study reports that current NGS costs for a tissue sample are under $500 and that sequencing takes a few days, while computational comparison of sequence data to reference genomes can be completed in under two hours.
To move NGS-guided diagnostics into routine laboratory practice, additional research is needed to define which organisms may be present in healthy brain tissue and to validate the method in larger cohorts with known diagnoses to determine sensitivity and specificity. According to global estimates cited by the authors, tens of thousands of deaths worldwide each year are attributed to inflammatory brain conditions, and identifying infectious causes when present enables targeted treatment with antivirals or antibiotics, or immunotherapy when the disorder is autoimmune.
Other authors on the study include Florian Breitwieser, Anupama Kumar, Haiping Hao, Peter Burger, Fausto Rodriguez, Michael Lim, Alfredo Quiñones-Hinojosa, Gary Gallia, Jeffrey Tornheim, Michael Melia and Cynthia Sears of the Johns Hopkins University School of Medicine.
Funding: The study received support from the National Human Genome Research Institute (grant R01 HG006677), the U.S. Army Research Office (grant W911NF-14-1-0490), and the Bart McLean Fund for Neuroimmunology Research—Johns Hopkins Project Restore.
Original research: Next-generation sequencing in neuropathologic diagnosis of infections of the nervous system, published online June 13, 2016, in Neurology: Neuroimmunology & Neuroinflammation.
Abstract
Objective: To assess the feasibility of next-generation sequencing microbiome approaches for diagnosing infectious disorders in brain or spinal cord biopsies from patients with suspected CNS infection.
Methods: In a prospective pilot study, NGS was applied alongside a novel computational analysis pipeline to detect pathogenic microbes in biopsies from 10 patients with neurologic signs of possible infection whose conventional clinical and microbiology tests were negative or inconclusive.
Results: Direct DNA and RNA sequencing produced between 8.3 million and 29.1 million reads per sample. NGS identified the infectious agent with high confidence in three patients, confirmed by validation techniques. Although precise pathogens were not identified in all cases, NGS contributed to the diagnostic understanding in five additional patients. Clinical outcomes aligned with NGS findings in eight of 10 cases.
Conclusions: Metagenomic NGS of brain, spinal cord, or meningeal biopsies offers the potential to improve detection and exclusion of a wide range of CNS pathogens, with benefits in speed, sensitivity, and cost. NGS-based microbiome approaches provide a new opportunity to investigate infectious contributions to neuroinflammatory disease.