Blood Transcript Profiling for the Detection of Neuroendocrine Tumors: Results of a Large Independent Validation Study.
Front Endocrinol (Lausanne)
; 9: 740, 2018.
Article
em En
| MEDLINE
| ID: mdl-30564197
ABSTRACT
Background:
Available neuroendocrine biomarkers are considered to have insufficient accuracy to discriminate patients with gastro-entero-pancreatic neuroendocrine tumors (GEP-NETs) from healthy controls. Recent studies have demonstrated a potential role for circulating neuroendocrine specific transcripts analysis-the NETest-as a more accurate biomarker for NETs compared to available biomarkers. This study was initiated to independently validate the discriminative value of the NETest as well as the association between tumor characteristics and NETest score.Methods:
Whole blood samples from 140 consecutive GEP-NET patients and 113 healthy volunteers were collected. Laboratory investigators were blinded to the origin of the samples. NETest results and chromogranin A (CgA) levels were compared with clinical information including radiological imaging to evaluate the association with tumor characteristics.Results:
The median NETest score in NET patients was 33 vs. 13% in controls (p < 0.0001). The NETest did not correlate with age, gender, tumor location, grade, load, or stage. Using the cut-off of 14% NETest sensitivity and specificity were 93 and 56%, respectively, with an AUC of 0.87. The optimal cut-off for the NETest in our population was 20%, with sensitivity 89% and specificity 72%. The upper limit of normal for CgA was established as 100 µg/l. Sensitivity and specificity of CgA were 56 and 83% with an AUC of 0.76. CgA correlated with age (rs = 0.388, p < 0.001) and tumor load (rs = 0.458, p < 0.001).Conclusions:
The low specificity of the NETest precludes its use as a screening test for GEP-NETs. The superior sensitivity of the NETest over CgA (93 vs. 56%; p < 0.001), irrespective of the stage of the disease, emphasize its potential as a marker of disease presence in follow up as well as an indicator for residual disease after surgery.
Texto completo:
1
Base de dados:
MEDLINE
Tipo de estudo:
Diagnostic_studies
Idioma:
En
Ano de publicação:
2018
Tipo de documento:
Article