Quick assessment of cell-free DNA in seminal fluid and fragment size for early non-invasive prostate cancer diagnosis

BACKGROUND
Liquid biopsy consists in the quantification and qualification of circulating cell-free DNA (cfDNA) and tumor-derived DNA (ctDNA) for cancer recognition. Recently, the characterization of seminal cfDNA (scfDNA) has been reported as a possible biomarker for prostate cancer (PCa) diagnosis.


METHODS
Thirty patients with histologically proven PCa, 33 with benign prostate hyperplasia (BPH) and 21 healthy controls were enrolled. cfDNA was extracted from seminal fluid samples. cfDNA quantification and analysis were performed using Qubit ssDNA Kit and Agilent 2100 Bioanalyzer. Statistical analysis included: Levene's test, Shapiro-Wilk, Kolmogorov-Smirnov and Kruskal Wallis tests.


RESULTS
Median cfDNA was significantly higher in PCa patients 428.45 ng/mL (173.93-1159.62) compared to BPH patients 77.4 ng/mL (18.23-501) and healthy controls 25.4 ng/mL (15.37-76.62). scfDNA fragments longer than 1000 base-pairs were more common in patients with PCa compared to those with BPH and controls.


CONCLUSIONS
scfDNA concentration and fragment size differed significantly in the three groups of PCa, BPH and healthy controls. Both parameters are potential clinical biomarkers for PCa and can be used in both early diagnosis and follow-up. Using automated systems for high-throughput cfDNA quantification could improve the reproducibility of the method and facilitate the implementation of liquid biopsies in the clinical setting.

A C C E P T E D M A N U S C R I P T

Introduction
Liquid biopsy consists in the quantification and qualification of circulating cell-free DNA (cfDNA) and tumor-derived DNA (ctDNA) in bloodstream and biological fluids. Previous studies reported higher plasma concentrations of cfDNA in PCa patients compared to healthy controls [1,2] .
Significant differences in levels and size distribution of cfDNA were also found in seminal fluid of PCa and BPH patients, respectively [3][4][5] .
cfDNA and ctDNA originate through different cellular processes: necrosis, phagocytosis, oncosis and even active cellular secretion. Not only cancer cells, but also healthy cells and the tumour microenvironment are considered the main sources of cfDNA in cancer patients. Each above mentioned compartment releases different forms of cfDNA into the bloodstream or other biological fluids [6,7]. Previous studies associated high cfDNA concentrations with specific size-distribution patterns to cancer diagnosis [8,9]. However, levels of cfDNA/ctDNA in blood samples vary depending on cancer type. ctDNA was detected in over 75% of patients with advanced pancreatic, ovarian, colorectal, bladder, gastroesophageal, breast and hepatocellular cancer, as well as in melanomas and head and neck tumours. In the case of prostate cancer however, significantly lower levels of plasma cfDNA/ctDNA were found [10]; this is probably due to diverse influencing factors, such as tumor clonality and/or the involvement of different compartments in the release of cfDNA, mainly originating from tumour microenvironment [11].
The extraction and characterization of ctDNA from seminal fluid (scfDNA) is probably a major step forward in PCa diagnosis and follow-up compared to traditional diagnostic techniques involving prostate-specific antigen (PSA), digital rectal examination (DRE), transrectal ultrasound (TRUS) and invasive biopsy [12]. Recently, we reported that the quantification and the electrophoretic characterization of cfDNA in seminal plasma can differentiate between PCa patients, BPH patients, and age-matched healthy individuals; indeed, scfDNA concentrations are higher in PCa patients and present a distinct electrophoretic pattern [3,4]. scfDNA can be easily analysed through cost- effective procedures such as fluorometry and electrophoresis. In this scenario, scfDNA may significantly change the diagnosis and management of PCa and other genitourinary tumours [5].
The aim of this study was to assess the levels and fragment size distribution of scfDNA in a cohort of PCa patients, BPH patients and healthy controls, in order to identify a novel biomolecular signature for PCa diagnosis and follow-up.

Ethics statement
The Ethics Committee of the University of Modena and Reggio Emilia approved the study and informed consent was obtained from each recruited subject.
This study was conducted according to the principles of the Helsinki Declaration of 1975, amended in 1996 (http://www.wma.net/e/humanrights/policy_meetings.htm). Patient seminal samples were collected prior to radical prostatectomy and/or chemotherapy or radiotherapy.

ACCEPTED MANUSCRIPT
A C C E P T E D M A N U S C R I P T

Extraction of scfDNA
As the samples ranged from 0.65 to 3.0 mL, a standardized sample volume of 0.65 mL was analysed. Following WHO's guidelines [13], to avoid sperm lysis, we processed samples within 2 hours from collection at low speed centrifugation (400 rcf) for 10 minutes at room temperature.
Then we preserve the supernatant (seminal plasma) from the pellet containing the cells and sperms, and a second step of centrifugation at high speed (16000 rcf) for 5 minutes at room temperature was done to eliminate debris and eventually remaining cells. Seminal plasma samples were then stored at -80 ° C till scfDNA extraction.
scfDNA was extracted with the QIAamp Circulating Nucleic Acid Kit (Qiagen, Hilden, Germany) and eluted in 120 µL buffer AVE according to the manufacturer's instructions. Aliquots of 120 µL scfDNA were cryopreserved at -20°C until quantification.

Quantification of scfDNA.
Following recent evidence stating that scfDNA is both single and double strand [14], the concentration levels were measured using the Qubit ssDNA Assay Kit to assess the total amount of scfDNA.   Figure 1).

PCa is characterized by distinctive scfDNA fragment size distribution pattern
scfDNA fragments were divided into five groups: <100 bp, between 101 and 146 bp, between 147 and 190 bp, between 191 and 999 bp, and >1000 bp.
Longer scfDNA fragments were more common in patients with PCa compared to those with BPH and controls (Figure 2, 3 and 4) Figure 1).
In addition, we suggest that the electrophoresis of seminal plasma cfDNA enables the discrimination between PCa patients and BPH or age-matched healthy individuals, because of a distinct electrophoretic pattern.

A C C E P T E D M A N U S C R I P T
The scfDNA analysis through automated microfluidic electrophoretic systems (i.e. Bioanalyzer) showed distinct DNA fragmentation patterns, characterized by high molecular weight DNA (longer than 1000 bp) compared to BPH patients and healthy individuals. These long DNA fragments in PCa patients could derive from the prominent necrotic processes characterizing PCa growth and differ from the typical apoptotic DNA fragments (147-190 bp and multimeric). From our experience we didn't found over-estimation of the total cfDNA concentration due to fragment length by using Qubit. We had some of PCa' samples with less long fragments but high cfDNA concentration, or some controls' and BPH's samples with long fragments but low cfDNA concentration. Moreover, we analysed some samples using Bioanalyzer for the total amount of cfDNA (data not reported) and there was correlation with the Qubit's measurements. cfDNA fragment size was associated with a more advanced stage at the diagnosis, representing a potential prognostic factor for PCa patients.
The above-mentioned concept has already been discussed in the literature with regard to cfDNA in bloodstream. It is known that higher levels of cfDNA assessed in cancer patients compared to healthy individuals are the result of the release of cfDNA either by malignant or non-malignant cells [6,7]. The assessment of total cfDNA level in the bloodstream can depict tumor dynamics and clonal heterogeneity over time. Thierry AR et al. showed that total cfDNA and mutant cfDNA were strongly correlated with decrease of overall survival as well as high levels of mutation load and fragmentation [6,15]. The fragment size distribution analysis is an important factor, which is associated to tumor load, response to treatment and overall survival of cancer patients.
The determination of cfDNA fragment size and cfDNA concentration is a non-invasive, simple method for PCa diagnosis and for PCa staging. Fluorometric and electrophoretic assessments allow a reliable quantification and qualification of scfDNA, that could be routinely adopted for PCa screening programs.
Seminal fluid can be easily collected from patients, and the quantification of cfDNA in seminal fluid does not add another layer of complexity compared to other substrates as are blood or urine [2,16]. Over and above that, isolating cfDNA from seminal fluid grants more efficiency: all

A C C E P T E D M A N U S C R I P T
conditions being like other isolation methods, it allows the extraction of a greater quantity of cfDNA. The high concentration of scfDNA allow a more accurate and precise primary early diagnosis of localized PCa. This makes scfDNA an attractive clinical marker for non-invasive cancer diagnosis and longitudinal monitoring of response to treatments. The demand on "liquid" biopsies is likely to grow. Automated systems for high-throughput cfDNA quantification and analysis could improve the reproducibility of the process and facilitate its implementation in the clinical setting.
A quick, cost-effective and simple test offers important advantages. cfDNA analytic methods vary in quantitative range, sensitivity, costs and workflow simplicity. Our Agilent 2100 Bioanalyzer technique is much simpler and faster than NGS and qPCR, it provides accuracy and reproducibility, and has the potential to allow any laboratory to perform molecular analysis, without the need for specialized electrophoresis equipment. The Bioanalyzer characterizes DNA fragments for their integrity, purity, concentration, and size distribution; it is faster and more sensitive compared to standard agarose gel electrophoresis in isolating and displaying the differences in genes expression, providing at the same time accurate quantitative information for each DNA sample fragment [17,18].
To sum up, our data suggest that the level and fragment size distribution of seminal cfDNA are potential non-invasive and easily collectable biomarkers for PCa diagnosis and follow-up.

Summary points
Liquid biopsy consists in the quantification and qualification of circulating cell-free DNA (cfDNA) and tumor-derived DNA (ctDNA) for cancer recognition.
The characterization of seminal cfDNA (scfDNA) is a potential biomarker for prostate cancer (PCa) diagnosis. Application of automated systems for high-throughput cfDNA quantification and analysis could facilitate the implementation of liquid biopsy in the uro-oncological clinical setting.

A C C E P T E D M A N U S C R I P T
Conflict of Interest Statement: None declared.

A C C E P T E D M A N U S C R I P T
A C C E P T E D M A N U S C R I P T