Analytic validation of NeXT Dx™, a comprehensive genomic profiling assay

Abstract

We describe the analytic validation of NeXT Dx, a comprehensive genomic profiling assay to aid therapy and clinical trial selection for patients diagnosed with solid tumor cancers. Proprietary methods were utilized to perform whole exome and whole transcriptome sequencing for detection of single nucleotide variants (SNVs), insertions/deletions (indels), copy number alterations (CNAs), and gene fusions, and determination of tumor mutation burden and microsatellite instability. Variant calling is enhanced by sequencing a patient-specific normal sample from, for example, a blood specimen. This provides highly accurate somatic variant calls as well as the incidental reporting of pathogenic and likely pathogenic germline alterations. Fusion detection via RNA sequencing provides more extensive and accurate fusion calling compared to DNA-based tests. NeXT Dx features the proprietary Accuracy and Content Enhanced technology, developed to optimize sequencing and provide more uniform coverage across the exome. The exome was validated at a median sequencing depth of >500x. While variants from 401 cancer-associated genes are currently reported from the assay, the exome/transcriptome assay is broadly validated to enable reporting of additional variants as they become clinically relevant. NeXT Dx demonstrated analytic sensitivities as follows: SNVs (99.4%), indels (98.2%), CNAs (98.0%), and fusions (95.8%). The overall analytic specificity was >99.0%.

Keywords: comprehensive genomic profiling; precision medicine; tumor-normal; whole exome sequencing; whole transcriptome sequencing.

Figure 1. CNA distribution among validation clinical samples.

A description of the CNA distribution in the validation is shown by tumor type, which represents a cross section of clinically relevant copy number alterations that would be seen in clinical samples.

Figure 2

(A) Correlation of NeXT Dx tumor mutation burden with reference method. An XY plot of the TMB values from the two methods shows a linear regression and correlation between the methods reported in mutations per megabase. The Pearson correlation coefficient is 0.985, showing strong correlation between the methods. The NeXT Dx measures TMB by scanning the whole exome for mutations, then eliminating those that are present in the corresponding normal sample so that only tumor-associated mutations are considered. This may explain the regression slope of 0.69, as the reference method does not correct for germline mutations and thus could over-report TMB. (B) TMB distribution of validation samples by cancer type. The box and whiskers plot demonstrate the distribution of TMB in mutations per megabase. The TMB axis is on a logarithmic scale. Boxes denote the range from first to third quartile of TMB scores. The vertical line in each box denotes the median TMB value. The low and high whiskers indicate the lowest and highest TMB values that are within ± 1.5 times the interquartile range. Outliers outside 1.5 times the interquartile range are shown as individual points. The tumor types are ranked by median TMB from highest to lowest.

Figure 3. Fusions detected by NeXT Dx in the validation set.

A total of 121 fusions, representing 13 different tumor types, were detected in the validation set.

Figure 4. Allele frequency reproducibility between operators.

(A) The reproducibility of AF measurements was determined between two different operators. Ten different samples were used, containing a total of approximately 2,400 SNVs. Each sample was run multiple times on different days by the two operators. A plot of the AFs measured by the two operators on the same SNV is shown. In all, 557,391 comparisons are plotted. (B) The data in 4a is shown, but over the AF range of 0 to 10%.

Figure 5. Correlation between observed AF and expected AF for 401 cancer-associated genes in a tumor tissue sample diluted with the corresponding normal specimen.

The sample shown is representative of those tested in the limit of detection study. The input DNA quantity was 100 ng. (A) Data shown over the range 0 to 100% AF. (B) Data shown over the range 0 to 25% AF.

Figure 6. Percent positive agreement of variant calls as a function of allele frequency interval, for 401 cancer-associated genes.

(A) single nucleotide variants (SNVs). (B) insertions and deletions (indels). The dashed lines indicate the 95% PPA threshold. Results are shown for input DNA quantities of 150 ng and 200 ng.