Cancer Diagnosis Program | Biorepositories and Biospecimen Research Branch

Search BRD Papers

Refined Procedure to Purify and Sequence Circulating Cell-Free DNA in Prostate Cancer.

Author(s): Rahimirad S, Derderian S, Hamel L, Scarlata E, McKercher G, Brimo F, Rajan R, Rompre-Brodeur A, Kassouf W, Sanchez-Salas R, Aprikian A, Chevalier S

Publication: Int J Mol Sci, 2025, Vol. 26, Page

PubMed ID: 40565311 PubMed Review Paper? No

Purpose of Paper

Conclusion of Paper

Studies

1. Study Purpose

   This study described  DNA extraction optimization (addition of proteinase K step, elution solution and number of elutions), sequencing library construction (input amount and number of amplification steps) and sequencing depth for analysis of cell-free DNA (cfDNA) from the plasma of prostate cancer patients. cfDNA yield and fragment size were compared among plasma specimens obtained from patients at RP, and from patients that were disease-free for >6 years post-RP and from patients with metastatic castration-resistant prostate cancer. Additionally, relationships between cfDNA concentration and patient age, blood PSA, tumor grade or stage, progression on Abiraterone or Docetaxel treatment and patient survival were investigated. Plasma and buffy coat were separated from EDTA blood by centrifugation at 2500 g for 15 min within 2 h of blood collection and stored at -80°C. DNA was extracted from buffy coat (germline) using the QIAamp DNA Kit.  DNA was extracted from plasma (cfDNA) using the QIAamp Circulating Nucleic acid Kit with and without (during optimization only) the addition of a proteinase K incubation step. During optimization, the effect of elution with 56°C water versus AVE buffer (85 µL) was investigated as well as the effect of serial elution with 85 μL, 40 μL, 20 μL and 20 µL, after which four elution steps were included.  cfDNA was stored at -20°C. cfDNA was quantified using the Qubit dsDNA HS Assay Kit and the high-sensitivity Bioanalyzer DNA Kit. For specimens with genomic DNA contamination, genomic DNA was removed with the AMPure XP Magnetic Beads. Libraries were prepared with the TruSeq Nano DNA Low-throughput Library Prep Kit using a modified protocol that omitted the fragmentation step. During optimization, the effect of using 8-10 versus 13 amplification cycles and 30 to 50 ng cfDNA was compared before settling on 50 ng cfDNA.  Libraies were sequenced on a Novaseq 6000-S4 using 150-bp paired-end sequencing reads (at low depth (10x) and high depth (170x).  Sequence data was analyzed using the Genpipe pipelines, and CNV were called with IchorCNA. SNVs were called using Mutect2, Strelka2, Vardict, and Varscan2 and called only if: (I) detected by ≥2 of the tools; (II) had a depth ≥10x ; (III) had a VAF of ≥ 5%; (IV) had ≥ 8 s reads in cfDNA and 0 reads in germline DNA; and (V) they had a population allele frequency below 0.01%.

   Summary of Findings:

   Genomic DNA contamination decreased when adding proteinase K to plasma before thawing and centrifuging the plasma.  Most of the DNA was obtained in the first elution, regardless of whether elution was with AVE buffer (77-100%) or water (80-100%).  When eluted with 85 µL water, extraction efficiency depended on the DNA concentration of the sample: 100% of low concentration cfDNA (≤ 1 ng/µL) was collected in the first elution (85µL), but for high concentration cfDNA (>1 ng/µL) 82% was collected in the first elution. The difference in quantified DNA levels between Qubit and Bioanalyzer was significant among specimens with low (4.3 ng/mL) and high (10.7 ng/mL) genomic DNA contamination (P<0.0002). For specimens with low genomic DNA contamination, 80-100% of the DNA was collected in the first elution, but for specimens with high genomic DNA contamination, only 60-80% was recovered in the first elution. Adding a wash with magnetic beads completely removed the genomic DNA contamination. Based on their findings, the authors recommend eluting samples four times with water, followed by a wash step with magnetic beads to remove genomic DNA contamination. Using 8 PCR cycles rather than 13 for library construction led to the lowest off-target amplification peak without affecting the library concentration. When 30 ng rather than 50 ng cfDNA was used for library preparation, there were more false subclonal events and lower tumor fraction in low-pass sequencing and in deep sequencing. After downsampling the deep-sequencing data to 10x, the alteration patterns (amplifications and losses) remained the same, but there were more false subclonal gains. During gradual downsampling, all clonal and subclonal events were noted at 117x and 94x, but a subclonal event was missed at 75x and while it was found at 60x there were many other false subclonal events that were detected. The VAF of SNVs fluctuated with the depth of downsampling, with 117x required to achieve >8 reads for low VAF mutations.  

   cfDNA concentrations as measured by Qubit were higher in the plasma of patients with metastatic prostate cancer than in the plasma of healthy controls (P<0.05) and patients at the time of RP (P<0.01) but were lower in the plasma from patients at the time of RP than after 6 years disease-free survival or in healthy controls (P<0.01, both). cfDNA concentrations as measured by bioanalyzer were only higher in the plasma of patients with metastatic prostate cancer than in the plasma from patients at the time of RP (P<0.05); cfDNA concentrations were lower in plasma from patients at the time of RP than after 6 years disease-free survival (P<0.05) or in healthy controls (P<0.01). The average fragment length of cfDNA was lower in the plasma of patients with metastatic prostate cancer compared to the plasma of healthy controls (137–163 bp versus 138–184 bp, P<0.05) and patients at the time of RP (137–163 bp versus 148–173 bp, P<0.01).  In plasma from patients with metastatic prostate cancer, fragment size was modestly and negatively correlated with cfDNA concentration (R=-0.56, P=0.018), but no such correlation was observed in plasma of the other groups. cfDNA concentration was not correlated with patient age, blood PSA prior to RP, tumor grade or stage, progression while on Abiraterone or Docetaxel treatment or survival of metastatic prostate cancer patients. cfDNA concentration was modestly correlated with blood PSA for metastatic prostate cancer patients at study inclusion (R=0.5, P=0.04) and not pre-RP. In a single patient, plasma that was obtained 6 months prior to death had 3-4 fold higher cfDNA concentrations than plasma obtained from the same patient prior to RP. While no CNVs were detected in the specimen obtained prior to RP, several CNVs (a loss at chromosome 8 and gains at 10, 12 and X), and SNVs were detected.

   Biospecimens

   • Bodily Fluid - Plasma

   Preservative Types

   • Frozen

   Diagnoses:

   • Neoplastic - Carcinoma
   • Normal

   Platform:

   Analyte	Technology Platform
   DNA	Automated electrophoresis/Bioanalyzer
   DNA	Next generation sequencing
   DNA	Fluorometry

   Pre-analytical Factors:

   Classification	Pre-analytical Factor	Value(s)
   Biospecimen Acquisition	Time of biospecimen collection	At radical prostatectomy >6 years post- prostatectomy-disease free
   Preaquisition	Patient age	40-98 years
   Preaquisition	Diagnosis/ patient condition	Metastatic castration-resistant prostate cancer Prostate cancer Healthy
   Analyte Extraction and Purification	Analyte purification	Genomic DNA removed with magnetic beads Genomic DNA not removed
   Next generation sequencing Specific	Nucleic acid amplification	8 cycles 10 cycles
   Analyte Extraction and Purification	Protein digestion	Proteinase K step included No proteinase K step
   Next generation sequencing Specific	Technology platform	Low pass (10x) Deep (170x)
   Preaquisition	Prognostic factor	A variety of PSA levels Gleason grade 1 Gleason grade 2-3 Gleason grade 4-5 Progression on Abiraterone Progression on Docetaxel Progression on Enzalutamide Survival at 0-40 months
   Next generation sequencing Specific	Template/input amount	30 ng 50 ng
   Next generation sequencing Specific	Data handling	Down sampled to 117x Down sampled to 94x Down sampled to 75x Down sampled to 60x Down sampled to 10x
   Analyte Extraction and Purification	Elution Parameters	Eluted in AVE Eluted in water 1 elution step 2 elution steps 3 elution steps 4 elution steps

   View more