Erschienen in:
01.02.2016 | Review Article
Gene expression classifier for the diagnosis of indeterminate thyroid nodules: a meta-analysis
verfasst von:
Prasanna Santhanam, Rodhan Khthir, Todd Gress, Ayman Elkadry, Omolola Olajide, Abid Yaqub, Henry Driscoll
Erschienen in:
Medical Oncology
|
Ausgabe 2/2016
Einloggen, um Zugang zu erhalten
Abstract
Prior studies demonstrate that a novel genomic test, the gene expression classifier (GEC), could identify a benign gene expression signature in those nodules with indeterminate cytology with a negative predictive value of greater than 95 %. Examine the performance of the AFIRMA gene expression classifier in predicting benign and malignant nodules in patients with cytologically indeterminate nodules. MEDLINE and EMBASE search for studies meeting eligibility criteria between January 1, 2005, and August 30, 2015. A total of 58 studies identified. After excluding duplicates, case reports, reviews, commentary, insufficient data, a total of seven studies selected for analysis. We combined individual patient data from seven studies that examined the GEC test for indeterminate thyroid nodules. The reference standard for determination of benign or malignant nodules was the histopathology of the thyroidectomy specimen. A QUADAS-2 report for all studies included in the final analysis was tabulated for risk of bias and applicability. The pooled sensitivity of the GEC was 95.7 % (95 % CI 92.2–97.9, I
2 value 45.4 %, p = 0.09), and the pooled specificity was 30.5 % (95 % CI 26.0–35.3, I
2 value 92.1 %, p < 0.01). Overall, the diagnostic odds ratio was 7.9 (95 % CI 4.1–15.1). Patients with benign GEC were not followed long enough to ascertain the actual false-negative rates of the index test. Our meta-analysis revealed a high pooled sensitivity and a low specificity for the AFIRMA-GEC test for indeterminate thyroid nodules. This makes it an excellent tool to rule out malignancy.