Plasma cell-free metagenomic next generation sequencing in the clinical setting for the diagnosis of infectious diseases: a systematic review and meta-analysis

Deciphering the potential of plasma cell-free metagenomic next-generation sequencing using the Karius test

“…Results varied between the participating institutions (PPA 46–86%, NPA 57–79%), as did the underlying disease states and process of patient selection [2–5,9,10 ▪▪ ]. Overall, these results have not matched performance of initial studies, likely because plasma cf-mNGS is often considered as a test of last resort as currently used [3,10 ▪▪ ]. With low positive agreement, it is difficult to be confident that infection has been ruled out when the assay is negative.…”

“…Initial promising results catalyzed broader implementation, and a retrospective review has now documented performance in varied, actual practice settings [2–5,7,8 ▪ ,9]. In the first published meta-analysis, Niles et al [10 ▪▪ ] examined five retrospective studies involving a total of 552 patients in which the Karius Test was ordered. In total, their results showed a Positive Percentage Agreement (PPA) of 67% and a Negative Percentage Agreement (NPA) of 70% when comparing plasma cf-mNGS to a composite reference standard of conventional testing and expert case review that determined whether the result was consistent with the patients’ clinical status [10 ▪▪ ].…”

“…In the first published meta-analysis, Niles et al [10 ▪▪ ] examined five retrospective studies involving a total of 552 patients in which the Karius Test was ordered. In total, their results showed a Positive Percentage Agreement (PPA) of 67% and a Negative Percentage Agreement (NPA) of 70% when comparing plasma cf-mNGS to a composite reference standard of conventional testing and expert case review that determined whether the result was consistent with the patients’ clinical status [10 ▪▪ ]. Results varied between the participating institutions (PPA 46–86%, NPA 57–79%), as did the underlying disease states and process of patient selection [2–5,9,10 ▪▪ ].…”

“…In total, their results showed a Positive Percentage Agreement (PPA) of 67% and a Negative Percentage Agreement (NPA) of 70% when comparing plasma cf-mNGS to a composite reference standard of conventional testing and expert case review that determined whether the result was consistent with the patients’ clinical status [10 ▪▪ ]. Results varied between the participating institutions (PPA 46–86%, NPA 57–79%), as did the underlying disease states and process of patient selection [2–5,9,10 ▪▪ ]. Overall, these results have not matched performance of initial studies, likely because plasma cf-mNGS is often considered as a test of last resort as currently used [3,10 ▪▪ ].…”

“…[4] and 89.6% by Wilke et al . [9], but separate clinical adjudication in the meta-analysis found PPA to be 52 and 78%, respectively [10 ▪▪ ]. There may also be intangible impacts not reflected in the medical record; one analysis found 52.1% of physicians reported that cf-mNGS testing was helpful despite changing management only 32.4% of the time [9].…”