Elsevier

Clinica Chimica Acta

Volume 446, 15 June 2015, Pages 141-146
Clinica Chimica Acta

Controls to validate plasma samples for cell free DNA quantification

https://doi.org/10.1016/j.cca.2015.04.015Get rights and content

Highlights

  • Control for contaminating lymphocyte DNA

  • Progress in the accuracy of cfDNA analysis

  • Suggested approach easily incorporated

Abstract

Recent research has focused on the utility of cell free DNA (cfDNA) in serum and plasma for clinical application, especially in oncology. The literature holds promise of cfDNA as a valuable tumour marker to be used for treatment selection, monitoring and follow-up. The results, however, are diverging due to methodological differences with lack of standardisation and definition of sensitivity. The new biological information has not yet come into routine use.

The present study presents external standardisation by spiking with non-human DNA fragments to control for loss of DNA during sample preparation and measurement. It also suggests a method to control for admixture of DNA from normal lymphocytes by utilizing the unique immunoglobulin gene rearrangement in the B-cells. The results show that this approach improves the quality of the analysis and lowers the risk of falsely increased values. In conclusion we suggest a new method to improve the accuracy of cfDNA measurements easily incorporated in the current technology.

Introduction

Blood and other body fluids contain free circulating DNA, also known as cell-free DNA (cfDNA), as reported more than 60 years ago [1]. The level of cfDNA may be increased in different diseases, including cancer [2], although the exact mechanisms remain to be elucidated [3], [4]. In normal healthy individuals the level of cfDNA can vary and may increase as a result of extreme exercise and subsequently return to normal within hours [5]. The current literature describes several potential applications in clinical oncology [6], [7], [8], [9] but so far cfDNA has not been used in daily routine. One reason is the divergent results, which at least partially can be explained by a number of methodological issues.

The measurement of cfDNA is met with several challenges. One major issue is the lack of consensus concerning pre-analytical protocols and analytical methods. The pre-analytical problems have been discussed in a recent review [10] and it is obvious that standardisation is required to obtain comparable results, which are a prerequisite for clinical application. Several methods have been published for measurement of cfDNA in plasma or serum. BEAMing [11], which may be considered an advanced form of digital PCR, has high sensitivity and seems suitable for analysis of mutated DNA at very low concentrations but it is a rather complicated method demanding many resources, which may impede daily routine use. Digital PCR [12] seems promising as a further development of quantitative real time PCR (qPCR), which remains the method generally accepted in the scientific community.

A somewhat disregarded problem is the loss of DNA during isolation and preparation. Since the concentration of cfDNA in plasma samples may be very low, a substantial fraction of it can be lost during the purification process, e.g. by irreversible binding of the DNA to column material or surface of tubes used in the pre-analytical steps. On automated purification systems a lack of material transfer can occur due to an inappropriately mounted or defect pipette tip. Reproducibility of the purification system and the identification of failed samples due to loss of DNA, DNA degradation due to contaminating DNase in the sample or co-purification of inhibiting components can be addressed by spiking samples with non-human DNA of relevant length prior to purification and measure the amount recovered.

Conversely, falsely increased DNA levels are a potential consequence of contamination with DNA from lysed lymphocytes. This may occur during blood drawing as a result of the pressure drop from vein to collection tube, by prolonged storage time prior to plasma isolation, or accidental inclusion of cells from the interphase or buffy coat by too deep pipetting when isolating the plasma supernatant.

In principle, this problem can be addressed by measuring genetic modifications occurring only in normal blood cells. Normal B-cells have rearranged immunoglobulin genes and a subset of these can be identified by a PCR assay [13]. This approach would allow detection of potential lymphocyte contamination in a plasma sample. Using a quantitative PCR assay will potentially also enable an estimate of the degree of contamination and allow for a correction of the level of DNA.

The inclusion of controls for pre-analytical errors is of utmost importance to qualify plasma samples for analysis. This study describes the development of pre-analytical tests for loss of DNA as well as contamination with DNA from normal lymphocytes enabling reliable quantification of total cell free DNA in plasma samples for clinical evaluation.

Section snippets

Samples

One hundred plasma samples from the control group of the Vejle Biobank for Diabetes, Vejle Hospital (approved by the Danish Data Protection Agency (journal no. 2006-53-1385) and the Scientific Ethical Committee for Southern Denmark (project ID S-20080097)) served as normal donor controls. Residual whole blood DNA from 76 individuals used in routine screening for lactose intolerance was used as normal donor DNA. A pool of approximately 50 ml normal plasma was obtained from three volunteers in the

Purification control

The results of spiking with CPP1 DNA are shown in Fig. 2 based on 120 plasma samples purified for a breast cancer project. Five microliter of the 191 bp CPP1 fragment corresponding to approximately 10,000 molecules was spiked into the plasma samples prior to purification and a constant number of CPP1 alleles close to 100% recovery were found in all samples except for two erroneous ones.

Lymphocyte DNA control

To determine the fraction of cellular DNA detected by the PBC primer set, whole blood DNA from 76 normal donors

Discussion

Cell free DNA holds high hopes as a valuable tumour marker but the utility remains to be proven in the clinical setting. A major stumble stone in that context is the lack of standardisation and validation, not only concerning the investigated clinical material but also with respect to measurements of DNA [10], [15], [16].

Exogenous spike-in controls are used in a number of PCR assays, especially within microbiology, serving as a control for false negative results due to loss of material,

Funding

The study was funded by The Research Council of Vejle Hospital and The Cancer Foundation, none of which had any involvement in any part of the study or the interpretation of the results.

Conflict of interest

The authors declare no conflicts of interest/have no disclosures.

Acknowledgements

We thank Lone Hartmann Hansen, Pia Nielsen and Tina Brandt Christensen for the technical assistance.

References (19)

There are more references available in the full text version of this article.

Cited by (59)

  • Analysis of HOXA9 methylated ctDNA in ovarian cancer using sense-antisense measurement

    2021, Clinica Chimica Acta
    Citation Excerpt :

    The purified ctDNA was eluted in 60 µl M−elution buffer after which 340 µl water was added to all samples. A qPCR for the beta-2-microglobulin gene (B2M) and Glycine max cysteine-rich Polycomb-like Protein (CPP1) was then performed to quantify the amount of DNA in each sample for quality control [24]. Samples were concentrated to 20 µl on Amicon Ultra-0.5 Centrifugal Filter Units (Merck, Darmstadt, Germany)

View all citing articles on Scopus
View full text