Genomics soon to be standard for rapid ID diagnostics
In mid-2014, Mickey Kertesz, PhD, was part of a team researching the genetics of human pathogens at Stanford University. At that time, their work had only what the Israeli-born entrepreneur called “theoretical applications” for next-generation sequencing, but its human relevance would soon become clear.
One morning, Kertesz received a call from a clinician who was collaborating with the Stanford team, asking for his help. The clinician was desperately seeking a diagnosis for an unknown infection causing severe illness in a pediatric patient.
“It was probably in the lungs, they believed, and they couldn’t take the child through another biopsy because he was too weak to survive another one,” Kertesz recalled. “They asked if we could look into the data that we had, which were based on sequencing his cell-free DNA and see if we could diagnose anything.”
Kertesz and colleagues started what he called a 3-day marathon of sifting through the genetic data to find a life-saving diagnosis. But 3 days proved too long, and the child succumbed to the infection. It would take several more days for the researchers to find the offending pathogen, which an autopsy confirmed had indeed caused the infection in the child’s lungs.
“We lost the race for that child,” Kertesz said. “That was the most direct illustration that, if only we had this system in a shorter and more stable turnaround time, we could have made an impact in this case and in many others.”
Kertesz and his research partners sped up the process of next-generation sequencing (NGS) to identify pathogens quickly and precisely by their DNA and to facilitate the appropriate treatment. They soon founded Karius, the Redwood City, California-based diagnostics company that Kertesz heads as CEO.
Karius has developed a genomic pathogen DNA sequencing test that Kertesz says can pinpoint more than 1,000 pathogens, including bacteria, DNA viruses, fungi, molds and protozoa. The test reports a result the next day.
Karius is among the companies exploring infectious disease applications for NGS. Kertesz and other experts argue that the field of infectious diseases is lagging behind other specialties that are already using genomics widely, such as oncology and prenatal care.
More timely diagnosis
The principal advantages that genomics offer ID clinicians include greater sensitivity in identifying pathogens than with culture and with more precision compared with PCR. The days-long process of culture-based diagnostics can force a clinician to use broad-spectrum antibiotics in the hopes that they will control the thus-far unknown cause of infection. That can carry the risk for adverse, and potentially unnecessary, side effects. It can also contribute to antimicrobial resistance and, in cases of severe illness, delay much-needed treatment.
“Frequently, what happens is the cultures are performed and a variety of medications are given — several different antibiotics in some cases, and antifungal drugs as well — until the cultures come back,” said George A. Pankey, MD, director of infectious disease research at the Ochsner Medical Center and an Infectious Disease News Editorial Board member. Pankey spoke specifically about bacterial sepsis diagnostics.
“The cultures might not get back for 2 to 5 days ... Then you have to wait on susceptibility [test results], so you lose more time. If we could have a methodology that would allow us to make a quick diagnosis, theoretically, the patient could get very narrow, directed therapy, and wouldn’t run the risk of multiple treatments that might cause adverse events because of allergies, drug interactions or damage to an organ such as a kidney.”
Pankey said a quick and precise diagnosis might avert antibiotic use altogether if it does not identify a bacterial pathogen.
NGS can also overcome the shortcomings of PCR. Results of a study published in Clinical Infectious Diseases showed that NGS could detect previously unknown clades of the Treponema pallidum bacteria, which cause yaws. The researchers used it to test samples taken in previous studies from patients with suspected yaws in the Solomon Islands and Ghana.
Among 45 samples from the Solomon Islands and 27 from Ghana, 11 tested positive for T. pallidum via PCR. However, the researchers identified six additional undetected subspecies by NGS. The Solomon Islands genome sequences represented distinct clades that had previously gone undetected. The researchers concluded that the PCR tests used in previous studies and by local health officials could not sufficiently support efforts to erdicate yaws.
A study published in the Journal of Bone and Joint Surgery also illustrated the power of genome sequencing. Researchers collected synovial fluid, deep tissue and swab samples from 65 revision and 17 primary arthroplasties. The samples were tested with NGS and culture. Infections were detected in 28 revision surgeries. Of those, 17 cases were detected in culture and 25 by NGS. The two testing methods were concordant in 15 cases.
A third study suggested that sequencing can be used to not only identify pathogens but also to prevent and control nosocomial outbreaks. Researchers investigated an 80-week Acinetobacter baumannii outbreak at a Birmingham, England, hospital in 2011. Using the Vitek 2 system (bioMérieux), they sequenced the genomes of routinely collected isolates and identified 114 multidrug-resistant Acinetobacter isolates. They found that 74 isolates from patients and 10 environmental isolates were genomically similar enough to be considered part of the outbreak.
The researchers concluded that NGS can provide a cost-effective way to track the epidemiology of outbreaks of multidrug-resistant pathogens in a timely manner, thereby helping to stem the spread of disease.
When it comes to specific genomic methods, Charis Eng, MD, PhD, chair of the Cleveland Clinic Genomic Medicine Institute, pointed out two that are currently being explored for ID diagnostics. One is metagenomics, or the study of the collective genome of microorganisms. Eng gave an example of “shotgun sequencing,” by which DNA is extracted from an environmental sample, and small DNA fragments are sequenced to locate certain pathogens from among the sample. An ID clinician may perform shotgun sequencing by extracting bacterial sequences out of a sample to help identify the pathogen, a method that Eng admitted is challenging.
The other method Eng pointed out is microbiomics, the study of the overall microbiome, such as that of the human gut. An ID clinician would enrich the bacterial 16S RNA gene — a component of a ribosome subunit — and sequence that gene to pinpoint the pathogen.
“Microbiomics is currently being used in research and in limited ways clinically,” Eng said. “It is used in very limited fashion at the point of care. For example, when a patient has meningitis that is eluding the doctors, microbiomics has been used, and the bacteria or even amoeba [responsible] has been discovered. This then guides the choice of antibiotics. This is genomics-based precision medicine, and in these instances, life-saving.”
Eng predicts that microbiomics will be used routinely by ID physicians in “perhaps in 2 to 5 years.”
Karius’ NGS infectious disease test takes microbial cell-free DNA from blood samples to identify specific pathogens. The DNA from the offending microbe is harvested as it circulates in the patient’s bloodstream, even when the infection is deep in the body, in locations such as the heart, lung or brain, Kertesz explained.
“By extracting the microbial DNA from patients’ blood, sequencing it and analyzing the heaps of data coming from each sample, we can identify what we call the infectome, that signature of infection that is specific to each patient,” he said. “Our pipeline analyzes the data and produces a clinical-grade report that goes back to the clinician to target treatment to the specific pathogens in that patient’s blood, down to the species level.”
The microorganism that is detected may or may not be the one causing infection, said Karius Medical Director and ID clinician David Hong, MD.
“Results should be interpreted within the context of clinical data, including medical history, physical findings, epidemiologic factors and other diagnostic test results,” he said.
A study presented at ID Week 2017 suggested the Karius test is effective in detecting infectious agents in patients with sepsis. The prospective cohort study included 286 patients, from whom samples were taken and tested with both standard laboratory methods and the Karius test.
The test identified potential pathogens in 60.1% of patients, including DNA viruses, bacteria and fungi. Meanwhile, only 15.7% of patients had an initial blood culture that was positive, and 38.1% had possible infections detected by a composite laboratory standard. In addition, 81.4% of positive NGS tests were considered consistent with the patients’ septic events.
Kertesz said a few dozen institutions and health care systems throughout the United States are successfully using the Karius test.
“We are now ramping up our commercial team toward a nationwide launch to make the Karius test available at every hospital across the country,” he said. “Simply put, our vision of the future is that genomics is going to entirely change the way infectious diseases are diagnosed,” Kertesz added. “In a few years’ time, I have no doubt that the use of genomics for pathogen detection will be the default, rather than a new, development.” – by Joe Green
- Halachev MR, et al. Genome Med. 2014;doi:10.1186/s13073-014-0070-x.
- Marks M, et al. Clin Infect Dis. 2017;doi:10.1093/cid/cix892.
- Tarabichi M, et al. J Bone Joint Surg Am. 2018;doi:10.2106/JGJS.17.00434.
- Thair S, et al. Open Forum Infect Dis. 2018;doi:10.1093/ofid/ofx180.004.
Disclosures: Kertesz is CEO of Karius, and Hong is the company’s medical director. Eng and Pankey report no relevant financial disclosures.