Virtual Library
Start Your Search
X.Q. Koh
Author of
-
+
P1.02 - Poster Session with Presenters Present (ID 454)
- Event: WCLC 2016
- Type: Poster Presenters Present
- Track: Biology/Pathology
- Presentations: 1
- Moderators:
- Coordinates: 12/05/2016, 14:30 - 15:45, Hall B (Poster Area)
-
+
P1.02-059 - Evaluation of Plasma DNA Extraction, Droplet PCR and Droplet next Generation Sequencing Methods for Liquid Biopsy Analysis (ID 6407)
14:30 - 14:30 | Author(s): X.Q. Koh
- Abstract
Background:
The ability to detect tumour mutations from blood and other bodily fluids promises many sample access, convenience, and monitoring benefits. However, the extremely rare levels at which mutations are present in these fluids obliges the use of optimal extraction and detection methods. Here, the performance of two extraction, two droplet PCR (dPCR) and a droplet next generation sequencing (dNGS) method for blood plasma analysis was systematically evaluated.
Methods:
Limits of detection were assessed using 15 blinded healthy donor blood samples spiked with equivolume mixtures of H1975 (containing EGFR L858R and T790M mutations) and H1650 (EGFR exon 19 deletion, e19del) cells at 10%, 1%, 0.1%, 0.01%, 0.001% H1650 cells in triplicate. A series of 32 blinded blood plasma samples from non-small cell lung cancer (NSCLC) patients with known tumour EGFR mutation status was also tested. Samples were processed for plasma, and 1ml plasma each underwent Qiagen Circulating Nucleic Acid and Promega Maxwell Circulation Cell Free DNA extraction processing. The plasma DNA samples were analysed using the Biorad dPCR and Raindance Raindrop dPCR method for L858R and exon 19 deletion mutations, and the 50-gene Raindance Thunderbolts dNGS protocol.
Results:
No significant difference in DNA yield and detection patterns was observed between the two extraction methods. L858R mutations were detected by both dPCR methods at 0.001% in 1/3 replicates and 0.01% in 3/3 replicates. Of 4 cases with L858R tumour mutations, mutations were detected in the same 3 plasma samples by both Biorad and Raindance dPCR (sensitivity 75%). “False positive” L858R mutations were identified in 2 (specificity 92%) and 7 (75%) cases respectively. For 8 tumour e19del mutations, the sensitivities were 38% and 25%, and specificities were 96% and 75% respectively. Of 16 clinical samples analysed by dNGS, an average of 20 mutations per sample were identified after filtering for quality, non-synomyous, and non-germline variant status. The sensitivity and specificity for detecting 2 L858R tumour mutation was 100% and 50%, and 1 e19del tumour mutation was 100% and 55% respectively. The allele frequencies for the majority of “false positives” for dPCR and dNGS were less than 5%, although some “true positives” were also detected at that level.
Conclusion:
dPCR and dNGS methods can enable detection of tumour mutations in blood, albeit imperfectly. Future work to determine optimal detection thresholds will help to maximize sensitivity and specificity.
