An investigation of plasma cell-free RNA for the detection of colorectal cancer: From transcriptome marker selection to targeted validation.
Author(s): Northrop-Albrecht EJ, Wu CW, Berger CK, Taylor WR, Foote PH, Doering KA, Gonser AM, Bhagwate A, Sun Z, Mahoney DW, Burger KN, Boardman LA, Kisiel JB
Publication: PLoS One, 2024, Vol. 19, Page e0308711
PubMed ID: 39146279 PubMed Review Paper? No
Purpose of Paper
This paper compared RNA sequencing metrics of plasma isolated from blood collected in LBgard or EDTA tubes and sought to identify colorectal cancer (CRC) markers in EDTA plasma by next-generation sequencing using two different cohorts and workflows (total RNA versus targeted capture sequencing). Additionally, expression of the markers identified in the first two cohorts and those identified from the literature were evaluated in a third specimen cohort by real-time PCR.
Conclusion of Paper
RNA from LBgard plasma had lower mapping rates and a lower percentage of reads that mapped to annotated genes than RNA from EDTA plasma (7% versus 81%); LBgard plasma samples were subsequently excluded from further analysis. Unsupervised clustering based on the RNA expression profile did not clearly separate specimens from cancer patients and those from healthy volunteers, nor did it separate specimens based on cancer stage. The authors identified exons that were differentially expressed in the plasma from CRC patients and healthy volunteers by next-generation sequencing in two different cohorts, but the identified markers did not overlap between cohorts. In a follow-up study using real-time PCR, six genes (5 identified from the literature and one from this study) were differentially expressed between the plasma of CRC patients and healthy volunteers, and eight genes (six from the literature and two from this study) were differentially expressed between the plasma of stage 4 CRC patients and healthy individuals (six from the literature and two from the present study).
Studies
-
Study Purpose
This study compared RNA sequencing metrics of plasma isolated from blood collected in LBgard or EDTA tubes and sought to identify colorectal cancer (CRC) markers in EDTA plasma by next-generation sequencing using two different cohorts and workflows (total RNA versus targeted capture sequencing). Additionally, expression of the markers identified in the first two cohorts and those identified from the literature were investigated in a third cohort by real-time PCR. For the comparison between tube types and the initial identification of CRC genes, blood was collected from 24 CRC patients (2 patients stage I, 8 patients stage II, 12 patients stage III and 2 patients stage IV) and 25 healthy controls into EDTA tubes and from 8 CRC patients (2 patients stages I, 4 stage III, and 3 patients stage IV) and 27 healthy controls. RNA was isolated from separated plasma (details not provided) using the Qiagen miRNeasy Serum/Plasma Advanced Kit and quantified using the Quant-it RiboGreen assay. Sequencing libraries were prepared using the SMARTer Stranded Total RNA-seq Kit-Pico input kit without the fragmentation step and sequenced on an Illumina Hiseq 4000 instrument. For the marker selection study, EDTA blood was collected from 35 healthy volunteers, 29 patients with CRC (6, 9, 11 and 13 with stage I, II, III or 4, respectively), and 9 patients with adenomas; RNA was extracted from plasma using the Qiagen’s QIAamp Circulating Nucleic Acid Kit with DNase treatment. Isolated RNA was quantified with Quant-it RiboGreen and a bioanalyzer. RNA was depleted of ribosomes using the NEBNext rRNA Depletion kit v2 and reverse transcribed using the NEBNext Ultra II RNA First Strand Synthesis Kit but purified with Zymo Oligo Clean & Concentrator instead of beads. Sequencing libraries were prepared using the Roche KAPA HyperPrep Kit, subjected to a hybrid capture reaction using a custom discovery probe pool (135 genes and 3 controls), followed by 14 cycles of amplification prior to sequencing on an Illumina Novaseq SP flow cell. Sequencing data from the discovery and markers selection studies were analyzed using the Mayo Analysis Pipeline MAPS-seq (v3). Read quality was verified using RSeQC and differential gene expression analysis was conducted using DESeq2 in R (4.1.2) and a threshold of false discovery rate (FDR) ≤ 0.05 and a fold change threshold of >2. For validation, buffy coat was collected from 30 healthy volunteers and 28 patients with CRC (13, 6 and 9 with stage II, III and IV, respectively) and plasma was collected from 30 healthy volunteers and 27 patients with CRC (13, 6 and 8 with stage II, III and IV, respectively). RNA was extracted from plasma using the QIAamp Circulating Nucleic Acid Kit with DNase treatment and RNA cleanup, reverse transcribed with the Applied Biosystems High-Capacity RNA to cDNA Kit (Waltham, MA) and preamplified using the TaqMan Preamp Master Mix before quantification of 99 amplicons using TaqMan real-time PCR. The stability of 15 reference genes was analyzed by real-time PCR using RNA extracted from buffy coat with the miRNeasy Mini Kit with DNase treatment that was cleaned and reverse transcribed with Applied Biosystems High- Capacity RNA to cDNA Kit.
Summary of Findings:
RNA from LBgard plasma had lower mapping rates and a lower percentage of reads mapped to annotated genes than RNA from EDTA plasma (7% versus 81%) and were subsequently excluded from further analysis. Unsupervised clustering based on the RNA expression profile did not clearly separate specimens from cancer patients and those from healthy volunteers, nor did it separate specimens based on cancer stage. A total of 895 exons were differentially expressed between specimens from CRC patients and healthy volunteers. Of these, 570 exons (264 genes) were downregulated and 325 exons (189 genes) were upregulated. Pathway analysis determined that these genes are involved in protein kinase A, molecular mechanisms of cancer, thrombin, integrin link kinase, alpha adrenergic, and IL-8 signaling. The authors report that these pathways have previously been implicated in CRC. In the marker selection cohort, there was a trend toward higher RNA yield from the plasma of CRC patients compared to healthy volunteers (20.06 ng versus 14.15 ng, P=0.06), with an intermediate yield from the plasma of adenoma patients (17.32 ng), and no differences observed between cancer stages. Interestingly, none of the exons identified as differentially expressed in the discovery cohort were validated in the marker selection cohort, but 13 exons were found to be differentially expressed in patients with an adenoma (7 downregulated and 6 upregulated) relative to healthy volunteers, and 224 exons were differentially expressed in patients with CRC (85 downregulated and 138 upregulated) relative to healthy volunteers. The lack of consistency between the cohorts may be attributable to different workflows (total RNA versus targeted capture sequencing). In the validation cohort, many markers showed higher expression in the plasma of cancer patients than healthy controls, but only six (5 from the literature and one from this study) were significant. A total of eight genes were differentially expressed between stage 4 CRC patients and healthy individuals, six of which were selected based on prior literature/other studies and two were identified in the present study.
Biospecimens
Preservative Types
- None (Fresh)
Diagnoses:
- Normal
- Neoplastic - Benign
- Neoplastic - Carcinoma
Platform:
Analyte Technology Platform RNA Automated electrophoresis/Bioanalyzer RNA Fluorometry RNA Next generation sequencing RNA Real-time qRT-PCR Pre-analytical Factors:
Classification Pre-analytical Factor Value(s) Preaquisition Diagnosis/ patient condition CRC
Healthy
Adenoma
Preaquisition Prognostic factor Stage I
Stage II
Stage III
Stage IV
Biospecimen Acquisition Type of collection container/solution LBGard
EDTA
Fluorometry Specific Technology platform Total RNA sequencing
Targeted capture sequencing
TaqMan real-time RT-PCR