Table 11. Concordance of F1CDx and an externally validated NGS assay for platform-wide variants and CDx biomarkers.
| Variant Type | F1CDx+/evNGS+ | F1CDx-/evNGS+ | F1CDx+/evNGS- | F1CDx-/evNGS- | PPA (95% CI)a | NPA (95% CI)a |
|---|---|---|---|---|---|---|
| All SVs | 1282 | 73 | 375 | 284218 | 94.6% (93.3%, 95.8%) | 99.9% (99.9%, 99.9%) |
| SUBs | 1111 | 39 | 334 | 242540 | 96.6% (95.4%, 97.6%) | 99.9% (99.8%, 99.9%) |
| INDELs | 171 | 34 | 41 | 41678 | 83.4% (77.6%, 88.2%) | 99.9% (99.9%, 99.9%) |
| All CNAs | 51 | 10 | 38 | 13845 | 83.6% (71.9%, 91.8%) | 99.7% (99.6%, 99.8%) |
| CNA: Amplification | 36 | 8 | 22 | 13878 | 81.8% (67.3%, 91.8%) | 99.8% (99.8%, 99.9%) |
| CNA: Homozygous deletions | 15 | 2 | 16 | 13911 | 88.2% (63.6%, 98.5%) | 99.9% (99.8%, 99.9%) |
| Rearrangements | 14 | 3 | 6 | 8713 | 82.4% (56.6%, 96.2%) | 99.9% (99.9%, 100%) |
| Total (CNAs and rearrangements) | 65 | 13 | 44 | 22,558 | 83.3% (73.2%, 90.8%) | 99.8% (99.7%, 99.9%) |
| PIK3CA substitutions in breast cancer | 53 | 0 | 0 | 48 | 100.00% (93.3%, 100.0%) | 100.00% (92.6%, 100.0%) |
| FGFR2 fusionsb | 25 | 2 | 1 | 130 | 87.08% (61.40%, 98.30%) | 99.59% (92.87%, 100.00%) |
| MET exon 14 SNVs and INDELs | 49 | 0 | 1 | 118 | 100.0% (92.8%, 100.0%) | 99.2% (95.4%, 100.0%) |
| HRR gene substitutions | 35 | 1 | 1 | 8243 | 97.22% (85.47%, 99.93%) | 99.99% (99.93%, 100.00%) |
| HRR gene INDELs | 75 | 6 | 2 | 17627 | 92.59% (84.57%, 97.23%) | 99.99% (99.96%, 100.00%) |
| HRR gene rearrangements | 10 | 1 | 5 | 1824 | 90.91% (58.72%, 99.77%) | 99.73% (99.36%, 99.91%) |
| HRR gene CNAs | 20 | 1 | 3 | 1356 | 95.24% (76.18%, 99.88%) | 99.78% (99.36%, 99.95%) |
| NTRK1, NTRK2, NTRK3 fusions | 78c,d | 0c,d | 18c,d | 492c,d | 90.00% (75.00%, 100%)c,f | 99.94% (99.92%, 99.97%)c,f |
| 16c,e | 2c,e | 0c,e | 20c,e | |||
| 64g,h | 10g,h | 4g,h | 510g,h | 54.08% (37.94%, 71.37%)g,j | 99.98% (99.96%, 100.00%)g,j | |
| 15g,i | 3g,i | 0g,i | 20g,i |
CI = confidence interval; CNA = copy number alteration; CTA = clinical trial assay; evNGS = externally validated next-generation sequencing; F1CDx = FoundationOne®CDx; F1 LDT = FoundationOne® Laboratory Developed Test; FGFR2 = fibroblast growth factor receptor 2; HRR = homologous recombination repair; INDEL = insertion/deletion; NGS = next-generation sequencing; NPA = negative percent agreement; NTRK = neurotrophic receptor tyrosine kinase; PIK3CA = phosphatidylinositol-4,5-bisphosphate 3-kinase catalytic subunit alpha; PPA = positive percent agreement; SV = short variants; SUB = substitution.
a The PPA and NPA were calculated without adjusting for the distribution of samples enrolled using the F1 LDT; therefore, these estimates may be biased upward.
b PPA and NPA were adjusted using a prevalence of 9.6% to account for sampling differential.
c Primary analysis: a sample was considered as positive if an NTRK1/2/3 rearrangement was detected; otherwise, it was considered as negative.
d Subset 1: samples where F1CDx served as the selection assay. Adjusted PPA and NPA based on an estimated prevalence of 0.32% in the intended use population to account for sampling differences were 100.00% (95% CI: 95.31%, 100.00%) and 99.94% (95% CI: 99.91%, 99.96%), respectively, based on the primary analysis.
e Subset 2: clinical trial samples where local clinical trial assays (LCTAs) served as the selection assay. PPA and NPA were 88.89% (95% CI: 67.20%, 96.90%) and 100.00% (95% CI: 83.89%, 100%), respectively, based on the primary analysis.
f The weighted PPA and NPA based on the bootstrapping of the combined dataset 10,000 times are shown for the primary analysis.
g Secondary analysis: a sample was considered F1CDx positive only if it met the NTRK1/2/3 biomarker rule; otherwise, it was considered as F1CDx negative.
h Subset 1: samples where F1CDx served as the selection assay. Adjusted PPA and NPA based on an estimated prevalence of 0.32% in the intended use population to account for sampling differences were 13.58% (95% CI: 8.66%, 25.25%) and 99.98% (95% CI: 99.96%, 100.00%), respectively, based on the secondary analysis.
i Subset 2: clinical trial samples where local CTAs served as the selection assay. PPA and NPA were 83.33% (95% CI: 60.78%, 94.16%) and 100.00% (95% CI: 83.89%, 100%), respectively, based on the secondary analysis.
j The weighted PPA and NPA based on the bootstrapping of the combined dataset 10,000 times are shown for the secondary analysis.