A multianalyte algorithm PCR-based blood test outperforms single analyte ELISA-based blood tests for neuroendocrine tumor detection
A key issue in management of neuroendocrine tumors (NETs) is specific and sensitive biomarkers. Measurements of single analytes in blood are widely utilized but have significant limitations. We developed a 51 transcript blood NET signature and compared it with standard approaches.
Methods: The multigene signature was evaluated in prospectively collected NETs (n=41, 61% small intestinal, 50% metastatic, 44% under treatment). These were age (NETs: mean 56.9 years, range: 31-76; controls: mean 56.4, range: 33-75) and sex-matched (M:F 10:31) with controls (1:1). Samples were analyzed by 2-step PCR protocol and ELISAs: (DAKO-CgA), pancreastatin (CusaBio-PST) and neurokinin A (RayBiotech-NKA). Sensitivity comparisons included chi-square, non-parametric measurements and ROC analyses.
Results: The NETest identified thirty eight of 41 NETs with equivalent performance metrics: sensitivity/specificity 93% and an AUC of 0.96. For the single analyte ELISA assays, metrics ranged from 31-93% and AUCs from 0.55-0.67. The multigene transcript NETest significantly outperformed single analyte tests (Z-statistic=4.85-6.58, p<0.0001).
Conclusions: A 51 panel multigene blood transcript analysis is significantly more sensitive and efficient (>93%) than any single analyte assay (CgA, PST or NKA) for NET detection. Our data indicate that a blood-based multigene analytic measurement will provide increased sensitivity and specificity in minimally invasive disease detection.
Methods: The multigene signature was evaluated in prospectively collected NETs (n=41, 61% small intestinal, 50% metastatic, 44% under treatment). These were age (NETs: mean 56.9 years, range: 31-76; controls: mean 56.4, range: 33-75) and sex-matched (M:F 10:31) with controls (1:1). Samples were analyzed by 2-step PCR protocol and ELISAs: (DAKO-CgA), pancreastatin (CusaBio-PST) and neurokinin A (RayBiotech-NKA). Sensitivity comparisons included chi-square, non-parametric measurements and ROC analyses.
Results: The NETest identified thirty eight of 41 NETs with equivalent performance metrics: sensitivity/specificity 93% and an AUC of 0.96. For the single analyte ELISA assays, metrics ranged from 31-93% and AUCs from 0.55-0.67. The multigene transcript NETest significantly outperformed single analyte tests (Z-statistic=4.85-6.58, p<0.0001).
Conclusions: A 51 panel multigene blood transcript analysis is significantly more sensitive and efficient (>93%) than any single analyte assay (CgA, PST or NKA) for NET detection. Our data indicate that a blood-based multigene analytic measurement will provide increased sensitivity and specificity in minimally invasive disease detection.
