Factors affecting the success of next-generation sequencing in cytology specimens.
Author(s): Roy-Chowdhuri S, Goswami RS, Chen H, Patel KP, Routbort MJ, Singh RR, Broaddus RR, Barkoh BA, Manekia J, Yao H, Medeiros LJ, Staerkel G, Luthra R, Stewart J
Publication: Cancer Cytopathol, 2015, Vol. 123, Page 659-68
PubMed ID: 26230354 PubMed Review Paper? No
Purpose of Paper
The purpose of this paper was to retrospectively compare NGS success rates between cell block and smear specimens, between smears on fully-frosted and frosted tip slides, and among specimens with different estimated tumor percentages. The authors also investigated the relationship between DNA input and NGS success.
Conclusion of Paper
NGS was successful (>300,000 reads with a quality score of <1 misaligned base per 100 bases) for 164 of the 207 specimens (79%); and NGS success was positively correlated with DNA yield. A higher percentage of cell block than smear specimens were successfully sequenced (88% versus 72%). The percentage of tumor in the specimen samples did not affect NG success, and smear specimens on fully-frosted and frosted tip slides generated comparable NGS success rates. There was considerable variation in the percentage of specimens that met molecular adequacy criteria among the eight reviewing cytopathologists as well as NGS success rates.
Studies
-
Study Purpose
This study retrospectively compared NGS success rates between cell block and smear tumor specimens, between smears on fully-frosted and frosted tip slides, and among specimens with different estimated tumor percentages. The authors also investigated the relationship between DNA input and NGS success. A total of 208 specimens (116 smear and 92 cell blocks) were obtained by fine needle aspiration (FNA) from lung, soft tissue, bone, lymph node, thyroid, liver, adrenal, kidney and eye tissue and pleural fluid; and patient diagnoses included sarcoma, adenocarcinoma, and melanoma. To prepare the 92 cell blocks, aspirates were centrifuged at 1500 rpm for 10 min, resuspended in a solution with a final concentration of 47.5% ethanol and 5% formalin, pelleted by centrifugation at 1500 rpm for another 10 minutes, placed into a tissue cassette, and formalin-fixed and paraffin-embedded. Cell blocks sections were cut at a thickness of 4-5 µM. Tumor fraction was estimated based on hematoxylin and eosin (H&E) staining of the first and last cell block section. The 116 smears were on fully frosted slides (23), frosted-tip (65) or other slides and either methanol-fixed and Diff-Quik-stained and/or ethanol-fixed and Papanicolaou stained; no other details were provided. DNA was extracted from microdissected areas of 2-23 unstained cell-block sections (median: 6 sections) by incubation in PicoPure at 65°Covernight, followed by incubation at 95°C for 10 minutes and purification using the Agencourt AMPure XP kit. DNA yield was quantified using the Qubit DNA high sensitivity assay kit. Sequencing libraries were prepared using IT Ampliseq Cancer Panel primers and sequenced on a Ion Torrent personal genome machine Sequencing success was defined as a minimum of 300,000 reads with a quality score of <1 misaligned base per 100 bases (AQ20).
Summary of Findings:
NGS was successful for 164 of the 207 specimens evaluated (79%). Mean and median DNA yield were higher in specimens that were successful analyzed by NGS (5.2 versus 2.5 ng/µL and 2.5 versus 0.2 ng/µL, respectively) than those that were not; and, NGS success was positively correlated with DNA yield (P<0.0001). When broken down further by specimen type, median DNA yields were higher among successfully sequenced cell block and smear specimens than those which failed sequencing (P<0.05, both). A higher percentage of cell block than smear specimens were successfully sequenced (88% versus 72% (P<0.009), although DNA yield did not differ between specimen types. There was no difference in the tumor fraction between specimens successfully sequenced and those that failed sequencing. NGS success rates were comparable for smears on fully-frosted and frosted tip slides, but DNA yield was significantly higher from cell smear specimens on frosted-tip slides (P=0.018). There was considerable variation in NGS success rates as well as among the percentage of specimens that met molecular adequacy criteria among the eight cytopathologists.
Biospecimens
- Tissue - Eye
- Tissue - Kidney
- Tissue - Liver
- Tissue - Adrenal Gland
- Tissue - Lung
- Tissue - Lymph Node
- Tissue - Bone
- Bodily Fluid - Pleural Fluid
Preservative Types
- Formalin
- Other Preservative
- Ethanol
Diagnoses:
- Neoplastic - Carcinoma
- Neoplastic - Melanoma
- Neoplastic - Sarcoma
Platform:
Analyte Technology Platform DNA Fluorometry Cell count/volume H-and-E microscopy DNA Next generation sequencing Pre-analytical Factors:
Classification Pre-analytical Factor Value(s) Biospecimen Aliquots and Components Type of slide Frosted-tip slide
Fully frosted slide
Biospecimen Aliquots and Components Biospecimen heterogeneity A range of tumor cell contents examined
Biospecimen Acquisition Method of cell acquisition Cell block
Smear
Biospecimen Aliquots and Components Biospecimen components A range of tumor cell contents examined