Chapter 151
Defeating Infectious Agents — 15
In 2014, Sharon Peacock, a professor of microbiology at the University of Cambridge, wrote, “Microbial sequencing should be done as close to the patient as possible.” More than a decade later, despite remarkable advances in low-cost, rapid sequencing, this is only done in a few hospitals around the world. This approach is known as metagenomics, a comprehensive sequencing analysis of everything within a sample. It is an unbiased, direct way of determining the pathogen in a patient, be it bacteria, virus, fungus, or parasite. Not only can the microbe be identified within hours but the sequence can tell us whether resistance to therapy is likely and provide guidance for appropriate antibiotics. Surprisingly, since 2014, with the diagnosis of neuroleptospirosis using metagenomics to save a fourteen-year-old boy’s life, this technology has been known but collecting dust. A report in the Wall Street Journal in 2023 captured the virtue of this diagnostic approach with the catchy headline “even brain-eating amoeba can’t hide from this cutting-edge diagnosis technology.”
The traditional approach, what I’ve called a “culture of blood cultures” entails getting multiple blood samples and other relevant body fluids (such as sputum, stool, urine, and cerebrospinal fluid) sent for culture, which takes a few days to read out, and even longer to determine antibiotic resistance. In the meantime, the patient gets “empiric” antibiotic therapy, which typically consists of multiple potent, broad-spectrum antibiotics, which carry toxicities, and may not even provide coverage against the offending pathogen. For example, it could be a fungus or virus, not a bacteria culprit, or a bacterium that’s resistant to the empiric antibiotics selected. Some pathogens are very slow growing in culture, taking several days to show up. Yet we are stuck in this culture of doing cultures.
Metagenomics is also known as “shotgun” sequencing, and you may be wondering a bit on how it’s done. As you can see from figure 10.4, the sample undergoes extraction for short DNA fragments and rapid sequencing (at a cost of $150–$250), which leads to a bunch (about one hundred million snippets) of unclassified DNA sequences. Of these, about 99 percent are human; the nonhuman snippets are the ones of interest. Separating out contaminants and bystanders can be a challenge, but automated interpretation tools help get that done. With the support of cloud-based algorithms of libraries for the gamut of pathogens, and AI, identification of the culprit is made.
In 2020, the University of California, San Francisco, published data on diagnosing unknown infections in 160 patients. With a turnaround time of less than six hours, using nanopore sequencing, compared with traditional culture methodology, there was highly accurate diagnosis of bacteria and fungi pathogens (sensitivity of 79% and 91%, for bacteria and fungi, respectively; specificity 91% and 89%, respectively). In London, among thirty-four patients with a breathing tube in the intensive care unit, within eight hours, metagenomics of sputum samples made the diagnosis in all but one patient (92%). Many cases that were culture negative (no growth in culture) were diagnosed with metagenomics. In Shanghai, in ninety-nine patients, compared with blood cultures, metagenomics made the diagnosis in sixty-five patients versus only thirteen by culture; the other patients had no proof of an infection. A more recent report from multiple centers in France with over two hundred patients also had a very high yield with rapid sequencing. These studies point to the complementarity of sequencing and culture approaches, but the speed and accuracy of metagenomics gets to the pathogen diagnosis and right treatment days faster. Sending out the patient sample to companies that can do this results in multiple-day delays that approach the time it takes for results from blood cultures to come back. Beyond metagenomics, there are also ultra-rapid peptide tests to diagnose sepsis and assess antimicrobial resistance, also preempting the need for blood cultures. What can’t be emphasized enough is that time to the proper treatment is the critical determinant of outcomes for patients with sepsis.