Cancer Diagnosis Program | Biorepositories and Biospecimen Research Branch

Search BRD Papers

Impact of DNA integrity on the success rate of tissue-based next-generation sequencing: Lessons from nationwide cancer genome screening project SCRUM-Japan GI-SCREEN.

Author(s): Kuwata T, Wakabayashi M, Hatanaka Y, Morii E, Oda Y, Taguchi K, Noguchi M, Ishikawa Y, Nakajima T, Sekine S, Nomura S, Okamoto W, Fujii S, Yoshino T

Publication: Pathol Int, 2020, Vol. 70, Page 932-942

PubMed ID: 33030786 PubMed Review Paper? No

Purpose of Paper

Conclusion of Paper

Studies

1. Study Purpose

   This paper determined the next-generation sequencing (NGS) success rates with two different targeted sequencing panels using 2,573 FFPE advanced gastrointestinal tumor specimens and assessed the potential influence of DNA integrity (determined by qPCR), specimen acquisition method (biopsy, surgical resection), FFPE block storage duration, patient gender, patient age, tumor type (primary, metastatic; adenocarcinoma, non-adenocarcinoma), site of the primary tumor (colorectal, non-colorectal), and previous chemotherapy or radiography on NGS success.

   A total of 2,573 samples primary or metastatic tumors were collected by surgical resection or biopsy from the gastrointestinal tract (1,001 colorectal, 743 stomach, 232 esophagus, 247 pancreas, 163 biliary tract, 49 liver, 71 other types) of patients diagnosed with advanced gastrointestinal cancer at one of 19 institutions. Patients included males (1,682) and females (912), those that had (552) and had not (2,042) received chemotherapy prior to tumor collection, and those that had (69) and had not (2,525) received radiotherapy prior to tumor collection. Tumors were formalin-fixed, paraffin-embedded and stored for <1 (1,448), 1-2 (528), 2-3 (257), 3-4 (136), 4-5 (84), 5-6 (43), 6-7 (35), 7-8 (14), 8-9 (12), and >9 (15) years (no additional information on fixation or storage were provided).  DNA and RNA were extracted using the RecoverAll Total Nucleic Acid Isolation Kit and the PureLink RNA Micro Kit, respectively, from nine 7 µm-thick FFPE sections that were macrodissected to ensure tumor content was >50%.  The concentration of DNA and RNA were determined with the Qubit Assay Kit. DNA integrity was evaluated by real-time (q)PCR using 3 ng DNA as input; the qPCR-based cycle threshold (Ct) value of a long (157 bp) amplicon was subtracted from the CT value of a short amplificon (93 bp)(ΔCt). DNA and RNA from FFPE tumor specimens were sequenced by NGS using the Oncomine Cancer Research Panel (OCP, 143 genes) and the CE-IVD panel (25 genes). NGS analysis was considered to be successful if samples met the pre-defined quality control (QC)-metrics (85% uniformity, >1000 mean depth) and produced validated results.

   Summary of Findings:

   Using data from the pilot phase (66 colorectal carcinoma FFPE tumor specimens), the authors determined that NGS success rates were higher when both OCP and CE-IVD sequencing panels were considered together than when either was considered alone. NGS with the OCP panel was successful in 49 of these samples (74.2%), although OCP results were only confirmed by successful analysis with CE-IVD for 17 samples; when OCP and CE-IVD were considered together (combined NGS success with either assay), the NGS success rate increased to 86.4% (57 samples). Using the qPCR-based DNA integrity data that was available for 47 of the 66 specimens, the authors identified thresholds that were predictive of NGS success: ΔCt <4.4 (High integrity), ≥4.4 and ≤6.3 (Intermediate integrity), >6.3 (Low integrity).  Specimens with a high integrity had an OCP/NGS success rate of 94% compared to the 23.5% observed among specimens with a ΔCt >4.4 and all specimens with a ΔCt <6.3 were validated with the CE-IVD panel.  

   Similar findings were observed when the pilot phase was expanded to include the complete sample set of 2,590 patients, as the NGS success rate with the OCP was 68.3% (1769/2590), with 45.9% of those validated by CE-IVD; the combined NGS success rate was 82.9% (2146/2590). Additional analysis was limited to 2,573 samples (ΔCt values were not obtained for 17 samples and tumor was absent from 4 samples). Based on an univariate logistic regression model, there was a significant association between DNA integrity and OCP NGS success, with an odds ratio of 0.210 for high to intermediate DNA integrity (p<0.0001).  The OCP success rate was 90.2% in samples with high DNA integrity but 5.6% in those with low DNA integrity.  Multivariable analyses also revealed a significant association between DNA integrity and combined NGS success, with an odds ratio of 0.215 for intermediate to high DNA integrity (p<0.0001) but only 0.008 for low to high DNA integrity. Univariable analysis also found that the following were significantly associated with higher OCP and combined NGS success rates: female versus male patients (p=0.0441), patients that received chemotherapy prior to specimen collection versus those who did not (p=0.0338), surgically resected specimens versus biopsies (p<0.0001), adenocarcinomas versus non-adenocarcinomas (p=0.0003), colorectal specimens versus other tissue types (p<0.0001), and FFPE blocks stored for < 4 years versus those stored for ≥4 y (p<0.0001). OCP and combined NGS success rates decreased with FFPE block storage duration by as much as 50% over 4 years of storage.  The percentage of samples with high and intermediate DNA integrity also declined with FFPE block storage, with the percentage of samples with high DNA integrity declining by 50% in 2 years. Surgically resected specimens displayed higher OCP NGS success rates than biopsies (72.4% versus 63.0%) and a higher percentage of specimens with DNA of high integrity (87.7% versus 79.8%, respectively). Notably, inter-institutional discordance was both substantial and significant for OCP NGS success (range: 14.3-83.1%, p<0.05), combined NGS success (range: 38.5-94.9%; p<0.05), and the percentage of samples with DNA of high integrity (45.5-95.5%, p<0.05).  Multivariable analysis revealed DNA integrity and FFPE storage duration as independent factors that were strongly associated with OCP and combined NGS success rates and contributed to the discordance between institutions where specimens were collected.

   Biospecimens

   • Tissue - Colorectal
   • Tissue - Stomach
   • Tissue - Esophagus
   • Tissue - Other
   • Tissue - Liver
   • Tissue - Gall Bladder
   • Tissue - Pancreas

   Preservative Types

   • Formalin

   Diagnoses:

   • Neoplastic - Carcinoma

   Platform:

   Analyte	Technology Platform
   DNA	Fluorometry
   DNA	Next generation sequencing
   DNA	Real-time qPCR

   Pre-analytical Factors:

   Classification	Pre-analytical Factor	Value(s)
   Storage	Storage duration	<1 y 1-2 y 2-3 y 3-4 y 4-5 y 5-6 y 6-7 y 7-8 y 8-9 y >9 y
   Next generation sequencing Specific	Technology platform	Oncomine Cancer Research Panel CE-IVD
   Real-time qPCR Specific	Quality metrics	ΔCt <4.4 (High integrity) ΔCt ≥4.4 and ≤6.3 (Intermediate integrity) ΔCt >6.3 (Low integrity)
   Biospecimen Acquisition	Method of tissue acquisition	Biopsy Surgical resection
   Preaquisition	Diagnosis/ patient condition	Primary tumor Metastatic tumor Adenocarcinoma Non-adenocarcinoma
   Biospecimen Acquisition	Biospecimen location	Colorectal Non-colorectal
   Preaquisition	Patient age	>50 y ≤50 y
   Preaquisition	Prior patient medical condition	Previous chemotherapy received Chemotherapy not received Previous radiography received Radiography not received
   Preaquisition	Patient gender	Female Male

   View more