RESUMO
A problematic aspect of massive parallel sequencing is that somatic mutations and viral loads are typically quantified as a fraction relative to wild-type human DNA, yet wild-type levels vary with diverse biologic and preanalytic interferences. A novel strategy was devised to quantify target analytes in copies per mL of plasma after normalizing for read counts of spiked DNAs. Five synthetic DNAs (called EndoGenus spikes) were added to plasma before library preparation (modified ArcherDX LiquidPlex 28). By normalizing to the fractional recovery of EndoGenus spike reads, numerical values for each disease marker were reportable in units of copies per mL. To show how well this system operates, replicate assays were performed on 40 mock plasmas having 23 engineered mutations and on 21 natural plasmas. Reads for all five EndoGenus spikes were recovered (means, 313 and 376 copies/mL in mock and natural plasmas, respectively). Normalizing read counts for the proportional recovery of spikes helped control for variables in the multistep protocol, reducing the CV in replicate tests from 34% to 22% for mutations and from 25% to 7% for viral loads. In conclusion, the EndoGenus system is useful for evaluating efficiency of the total test system and for precisely quantifying target molecules. This system may benefit patients being monitored for disease burden while also tracking emerging subclones.
