COMPREHENSIVE DIAGNOSTIC EVALUATION OF NEOPLASTIC THYROID LESIONS BY NEXT GENERATION SEQUENCING AND MIRNA GENE EXPRESSION
AACE ePoster Library. Labourier E. 05/15/15; 97747; 1086
Dr. Emmanuel Labourier
REGULAR CONTENT
Login now to access Regular content available to all registered users.
Abstract
Rate & Comment (0)
Objective–Molecular testing for oncogenic gene alterations or gene expression in fine needle aspirations (FNAs) from thyroid nodules with indeterminate cytology increases the diagnostic yield of conventional cytopathology by detecting a subset of malignant or benign nodules with high predictive value. In the present study, we evaluated a comprehensive next generation sequencing (NGS) mutation panel and a novel miRNA-based diagnostic algorithm to further improve the yield of molecular cytology.
Methods–Total nucleic acids from surgical specimens and preoperative FNAs were tested for over 1,500 unique mutations in 68 distinct genomic regions from 20 genes and 46 gene fusions using pooled PCR and targeted NGS, or for 17 well-known alterations in the BRAF, HRAS, KRAS, NRAS, PAX8 and RET genes detected with the miRInform Thyroid test and combined with the qualitative results of a 10-miRNA gene expression classifier. All molecular results were compared against surgical histopathology reference diagnoses.
Results–Among 54 resected thyroid lesions negative for mutation by miRInform, 3 out of 36 malignant cases were positive by NGS at ≥10% variant, 6 by miRNA and 8 by both methods. Four out of 18 benign lesions were also positive by NGS, none by miRNA. At 5% variant threshold, NGS detected additional mutations in 2 malignant cases (1 positive by miRNA) and in 3 benign cases. A similar pattern was observed in 42 preoperative FNAs from thyroid nodules with AUS/FLUS or FN/SFN cytology collected at 5 endocrinology centers. miRInform correctly identified 11 malignant cases, including 6 positive by miRNA. Among the mutation-negative specimens, miRNA testing identified 6 additional malignant cases while NGS detected mutations in 2 malignant cases (both positive by miRNA) and in 1 (≥10% variant) or 4 (≥5% variant) benign cases (all negative by miRNA).
Discussion–Both NGS and the miRNA classifier increased the diagnostic sensitivity of mutation testing which would improve its negative predictive value and the yield of true positive/malignant results. These data suggest that miRNA testing can provide clinically relevant diagnostic results in line with NGS mutational analysis. The advantages of this approach lie in its ability to phenotypically identify malignant nodules that lack oncogenic mutations and to rule out benign nodules with germline or low-level somatic mutations of unknown clinical significance.
Conclusion–When using FNAs truly representative of local clinical practices and blinded surgical histology review, multi-categorical testing for highly specific miRNA and validated oncogenic mutations may provide higher predictive value than broad NGS panels targeting gene mutation hotspots.
Methods–Total nucleic acids from surgical specimens and preoperative FNAs were tested for over 1,500 unique mutations in 68 distinct genomic regions from 20 genes and 46 gene fusions using pooled PCR and targeted NGS, or for 17 well-known alterations in the BRAF, HRAS, KRAS, NRAS, PAX8 and RET genes detected with the miRInform Thyroid test and combined with the qualitative results of a 10-miRNA gene expression classifier. All molecular results were compared against surgical histopathology reference diagnoses.
Results–Among 54 resected thyroid lesions negative for mutation by miRInform, 3 out of 36 malignant cases were positive by NGS at ≥10% variant, 6 by miRNA and 8 by both methods. Four out of 18 benign lesions were also positive by NGS, none by miRNA. At 5% variant threshold, NGS detected additional mutations in 2 malignant cases (1 positive by miRNA) and in 3 benign cases. A similar pattern was observed in 42 preoperative FNAs from thyroid nodules with AUS/FLUS or FN/SFN cytology collected at 5 endocrinology centers. miRInform correctly identified 11 malignant cases, including 6 positive by miRNA. Among the mutation-negative specimens, miRNA testing identified 6 additional malignant cases while NGS detected mutations in 2 malignant cases (both positive by miRNA) and in 1 (≥10% variant) or 4 (≥5% variant) benign cases (all negative by miRNA).
Discussion–Both NGS and the miRNA classifier increased the diagnostic sensitivity of mutation testing which would improve its negative predictive value and the yield of true positive/malignant results. These data suggest that miRNA testing can provide clinically relevant diagnostic results in line with NGS mutational analysis. The advantages of this approach lie in its ability to phenotypically identify malignant nodules that lack oncogenic mutations and to rule out benign nodules with germline or low-level somatic mutations of unknown clinical significance.
Conclusion–When using FNAs truly representative of local clinical practices and blinded surgical histology review, multi-categorical testing for highly specific miRNA and validated oncogenic mutations may provide higher predictive value than broad NGS panels targeting gene mutation hotspots.
Objective–Molecular testing for oncogenic gene alterations or gene expression in fine needle aspirations (FNAs) from thyroid nodules with indeterminate cytology increases the diagnostic yield of conventional cytopathology by detecting a subset of malignant or benign nodules with high predictive value. In the present study, we evaluated a comprehensive next generation sequencing (NGS) mutation panel and a novel miRNA-based diagnostic algorithm to further improve the yield of molecular cytology.
Methods–Total nucleic acids from surgical specimens and preoperative FNAs were tested for over 1,500 unique mutations in 68 distinct genomic regions from 20 genes and 46 gene fusions using pooled PCR and targeted NGS, or for 17 well-known alterations in the BRAF, HRAS, KRAS, NRAS, PAX8 and RET genes detected with the miRInform Thyroid test and combined with the qualitative results of a 10-miRNA gene expression classifier. All molecular results were compared against surgical histopathology reference diagnoses.
Results–Among 54 resected thyroid lesions negative for mutation by miRInform, 3 out of 36 malignant cases were positive by NGS at ≥10% variant, 6 by miRNA and 8 by both methods. Four out of 18 benign lesions were also positive by NGS, none by miRNA. At 5% variant threshold, NGS detected additional mutations in 2 malignant cases (1 positive by miRNA) and in 3 benign cases. A similar pattern was observed in 42 preoperative FNAs from thyroid nodules with AUS/FLUS or FN/SFN cytology collected at 5 endocrinology centers. miRInform correctly identified 11 malignant cases, including 6 positive by miRNA. Among the mutation-negative specimens, miRNA testing identified 6 additional malignant cases while NGS detected mutations in 2 malignant cases (both positive by miRNA) and in 1 (≥10% variant) or 4 (≥5% variant) benign cases (all negative by miRNA).
Discussion–Both NGS and the miRNA classifier increased the diagnostic sensitivity of mutation testing which would improve its negative predictive value and the yield of true positive/malignant results. These data suggest that miRNA testing can provide clinically relevant diagnostic results in line with NGS mutational analysis. The advantages of this approach lie in its ability to phenotypically identify malignant nodules that lack oncogenic mutations and to rule out benign nodules with germline or low-level somatic mutations of unknown clinical significance.
Conclusion–When using FNAs truly representative of local clinical practices and blinded surgical histology review, multi-categorical testing for highly specific miRNA and validated oncogenic mutations may provide higher predictive value than broad NGS panels targeting gene mutation hotspots.
Methods–Total nucleic acids from surgical specimens and preoperative FNAs were tested for over 1,500 unique mutations in 68 distinct genomic regions from 20 genes and 46 gene fusions using pooled PCR and targeted NGS, or for 17 well-known alterations in the BRAF, HRAS, KRAS, NRAS, PAX8 and RET genes detected with the miRInform Thyroid test and combined with the qualitative results of a 10-miRNA gene expression classifier. All molecular results were compared against surgical histopathology reference diagnoses.
Results–Among 54 resected thyroid lesions negative for mutation by miRInform, 3 out of 36 malignant cases were positive by NGS at ≥10% variant, 6 by miRNA and 8 by both methods. Four out of 18 benign lesions were also positive by NGS, none by miRNA. At 5% variant threshold, NGS detected additional mutations in 2 malignant cases (1 positive by miRNA) and in 3 benign cases. A similar pattern was observed in 42 preoperative FNAs from thyroid nodules with AUS/FLUS or FN/SFN cytology collected at 5 endocrinology centers. miRInform correctly identified 11 malignant cases, including 6 positive by miRNA. Among the mutation-negative specimens, miRNA testing identified 6 additional malignant cases while NGS detected mutations in 2 malignant cases (both positive by miRNA) and in 1 (≥10% variant) or 4 (≥5% variant) benign cases (all negative by miRNA).
Discussion–Both NGS and the miRNA classifier increased the diagnostic sensitivity of mutation testing which would improve its negative predictive value and the yield of true positive/malignant results. These data suggest that miRNA testing can provide clinically relevant diagnostic results in line with NGS mutational analysis. The advantages of this approach lie in its ability to phenotypically identify malignant nodules that lack oncogenic mutations and to rule out benign nodules with germline or low-level somatic mutations of unknown clinical significance.
Conclusion–When using FNAs truly representative of local clinical practices and blinded surgical histology review, multi-categorical testing for highly specific miRNA and validated oncogenic mutations may provide higher predictive value than broad NGS panels targeting gene mutation hotspots.
{{ help_message }}
{{filter}}