Sage Journals: Discover world-class research

Abstract

BACKGROUND:

Carcinoembryonic antigen (CEA) is the only established serum biomarker for colorectal cancer (CRC). To facilitate therapy decisions and improve the overall survival of CRC patients, prognostic biomarkers are required.

OBJECTIVE:

We studied the prognostic value of five different cell free circulating DNA (fcDNA) fragments. The potential markers were ALU115, ALU247, LINE1-79, LINE1-300 and ND1-mt.

METHODS:

The copy numbers of the DNA fragments were measured in the peripheral blood serum of 268 CRC patients using qPCR, the results were compared to common and previously described markers.

RESULTS:

We found that ALU115 and ALU247 fcDNA levels correlate significantly with several clinicopathological parameters. An increased amount of ALU115 and ALU247 fcDNA fragments coincides with methylation of HPP1 ( $P<$ 0.001; $P<$ 0.01), which proved to be a prognostic marker itself in former studies and also with increased CEA level (both $P<$ 0.001). ALU115 and ALU247 can define patients with poor survival in UICC stage IV (ALU115: HR $=$ 2.9; 95% Cl 1.8–4.8, $P<$ 0.001; ALU247: HR $=$ 2.2; 95% Cl 1.3–3.6; $P=$ 0.001). Combining ALU115 and HPP1, the prognostic value in UICC stage IV is highly significant ( $P<$ 0.001).

CONCLUSIONS:

This study shows that an increased level of ALU fcDNA is an independent prognostic biomarker for advanced colorectal cancer disease.

Keywords

Colorectal cancer biomarkers fcDNA prognosis ALU repeats

Abbreviations

CEA	Carcinoembryonic antigen
CRC	Colorectal cancer
fcDNA	cell free circulating DNA
UICC	Union for International Cancer Control
TNM	Tumor site and size, Lymph node involve-
	ment, Metastatic spread

APC	Adenomatous polyposis coli gene
CIN	chromosomal instability
EGFR	epidermal growth factor receptor
TGFB	transforming growth factor beta
qPCR	real-time quantitative polymerase chain
	reaction
HLTF	Helicase-like transcription factor
HPP1	Hyperplastic polyposis 1
ALU	Arthrobacter luteus
LINE	long interspersed elements

1. Introduction

Colorectal cancer (CRC) is the second most common cancer in Europe with almost 500,000 new cases in 2018 and is responsible for the second most cases of cancer-related deaths in men and women [13]. This high incidence, particularly in western Europe, may reflect changes in lifestyle such as an increased prevalence of obesity and great variability in diet and physical activity across all European countries [3]. Although great efforts have been undertaken to improve cancer treatment and new therapy options find their way into clinical practice, survival of patients is still poor, especially in metastatic cases [22].

CRC carcinogenesis is characterized by genetic and epigenetic changes, which can be attributed to different signaling pathways. Basically, development of colorectal cancer through mutation in or loss of the adenomatous polyposis coli gene (APC) follows the adenoma-carcinoma sequence. Subsequently, different pathways lead to cancer development. Amongst others, the most frequent pathway in 65–70% of the cases is the Chromosomal instability (CIN) – pathway, where mutations of P53 and KRAS activate different pathways in turn, such as Wnt, MYC, EGFR (epidermal growth factor receptor), TGFB (transforming growth factor beta) and Akt signaling [1, 37].

Currently, the most valuable prognostic tool for cancer patients is the UICC (Union for International Cancer Control) TNM (Tumor site and size, Lymph node involvement, Metastatic spread) staging classification [6]. Although it is the only valid marker in predicting patient outcome [4], the TNM system is surgically oriented and neither integrates tumor biology nor differentiates the metastatic stage [32]. For this reason, TNM system is not able to specify the individual risk of a given patient [38]. Biopsies taken from the tumor also cannot reflect tumor biology properly due to genetic heterogeneity of the tumor tissue [55]. Nomograms based on TNM classification, combined with different clinicopathological parameters such as age and race, were created to predict survival of CRC patients [30].

A prognostic marker for CRC, widely used in clinical practice, is carcinoembryonic antigen (CEA). CEA proved to be an independent prognostic marker in various studies and guidelines recommend measuring preoperative CEA serum levels to determine prognosis, keep patients under surveillance after curative surgery and monitor therapy in advanced diseases [11]. CEA levels vary between individuals. Therefore, CEA baseline levels must be established for each patient. Although CEA determination is also recommended in the metastatic disease [10], Wanebo et al. reported that there was no correlation between CEA level and survival in metastatic disease [56].

Cell-free DNA is circulating in the blood stream in different concentrations. Compared to normal individuals, the amount of cell free circulating DNA (fcDNA) is higher in the blood of cancer patients [2] and is also increased in metastasized disease compared to nonmetastatic patients [21, 28]. Tumor-related nucleic acids as potential biomarkers for malignancies have been intensively studied recently [16, 24]. Circulatory microRNAs are also the subject of current research and have even been demonstrated to distinguish between HCV associated fibrosis, cirrhosis, and hepatocellular carcinoma [15, 58]. Sequencing of plasma DNA revealed that fcDNA represents the complete tumor genome as it can be released from either the primary tumor, metastasis, or apoptotic circulating tumor cells. Apoptosis, necrosis and even active secretion have been discussed as possible sources of tumor DNA, but the mechanism of release of fcDNA is still incompletely understood [54].

fcDNA fragments can be measured using real-time quantitative polymerase chain reaction (qPCR), which offers a quick, cost-effective, and robust method for quantification of the fcDNA fragments of interest. Alternative approaches compared are microarray-based linear amplification or Illumina sequencing of fcDNA [14, 47].

Notably, epigenetic alterations proved to be a promising predictive and prognostic tool. We previously described the prognostic value of methylation of various genes in free circulating DNA, namely helicase-like transcription factor (HLTF) and hyperplastic polyposis 1 (HPP1) [40]. It could be demonstrated that HLTF/HPP1 methylation status corresponds with shorter survival and therefore turned out to be a prognostic marker in UICC stage IV.

In addition to DNA methylation, DNA integrity has already been discussed as diagnostic and prognostic marker for colorectal cancer [57] based on the idea that the length of nucleic acids released from cancer cells differs from the DNA of non-malignant apoptotic cells [33].

Non-LTR (long terminal repeats) retrotransposons, which include short interspersed elements (SINEs) like the ALU repetitive sequence and long interspersed elements (LINEs), represent almost one third of the whole genome [8]. ALU elements are characterized by having a cleavage site for the Arthrobacter luteus (ALU) restriction endonuclease [19]. SINEs are shorter than LINEs and contain about 300bp. These non-coding DNA elements have function in regulating transcription and gene expression [51].

European population is yet not well considered in studies concerning fcDNA as a biomarker for cancer. The prognostic and diagnostic value of ALU repeats as biomarkers has been shown for different types of cancer, such as breast cancer, non-small-cell lung cancer and colorectal cancer, and across different populations, such as American and Egyptian [12, 46, 52, 53].

LINE1 has lately been discussed not only as a biomarker but also therapeutic drug target as it seems to have a regulatory impact on gene expression in cancer cells using different pathways of carcinogenesis. Furthermore, promoter methylation, transcription, translation, and retro transposition are different activities LINE1 is said to perform in cancer cells [26, 43]. LINE1 as a prognostic and diagnostic biomarker has been investigated in different types of cancer such as colorectal cancer, gastric cancer, and breast cancer, in addition to the Japanese mostly in the American population [9, 25, 49].

Human cells contain about 200 mitochondria, which possess their own genomic material. Due to its molecular characteristics such as an inefficient DNA repair system, mitochondrial DNA (mtDNA) is more susceptible to oxidative damage than nuclear DNA. Thus, DNA mutations accumulate easily and lead to mitochondrial dysfunction, which is associated with tumorigenesis. Through activation of the Toll-like 9 receptor (TLR9) pathway, inflammatory processes are initiated, which provide a condition for cancer development [29, 36].

Furthermore, copy numbers of circulating mtDNA have been discussed as a biomarker for different types of cancer such as hepatocellular carcinoma in an Egyptian population [18], lung cancer in a Chinese population [7] or head and neck squamous cell cancer in Northeast India [27].

Using mtDNA with its simple structure and a higher copy number than nuclear DNA allows a sensitive detection and exact quantification even of very low concentrations of circulating DNA in serum and other body fluids. Thus, mtDNA is ideally suited as biomarker.

The aim of this study was to evaluate the prognostic value of cell free circulating DNA fragments in colorectal cancer patients with metastasized and non-metastasized disease. We selected five potential markers, namely ALU115, ALU247, LINE1-70, LINE1-300 and ND1-mt, which were able to discriminate between CRC patients with benign and malignant tumors and healthy individuals in blood samples of a British patient population [34] and compared the results with common tumor markers as CEA and other predictive tools, e.g. the HLTF and HPP1 methylation status [40].

2. Methods

2.1 Patients and serum samples

Male and female patients over the age of 18 years were recruited from the Medical Department 2, Ludwig-Maximilians-University Munich and were diagnosed by histopathological examination of biopsies taken during colonoscopy or sigmoidoscopy using the valid WHO-classification at that time. Blood serum samples from 268 patients were taken before any therapeutic interventions and processed blinded to patient data in the central laboratory using routine procedures. All subjects underwent medical treatment at the Ludwig-Maximilians-University of Munich. Exclusion criteria were pretreatment of the current cancer disease, clinical history of any other cancer and simultaneous tumors. Staging was based on postoperative pathology findings or, for inoperable patients, on diagnostic imaging, namely CT of thorax and abdomen.

All samples were centrifuged at 2,000 g at 4 ${}^{\circ}$ C for 10 minutes. Aliquots were stored at $-$ 80 ${}^{\circ}$ C. We previously described the prognostic value of the HLTF/HPP1 methylation status in an almost identical sample set [40]. The study was approved by the ethical committee of the Medical Faculty of the University of Munich (project number 22-0873).

2.2 DNA isolation

After thawing the serum aliquots at room temperature, genomic DNA was isolated from 100 $\mu$ l of each serum sample using the High Pure Viral Nucleic Acid Kit (Roche Applied Science, Product No. 11858874001, Mannheim, Germany) according to the manufacturer’s instructions. The isolated DNA was eluted in 50 $\mu$ l of elution buffer, diluted 1:40 and stored at 4 ${}^{\circ}$ C until quantification of DNA. Peripheral blood leucocytes DNA from the blood of a healthy volunteer was isolated [35]. DNA concentration was determined by measuring absorbance at 260 nm and 280 nm using a NanoDrop 2000c spectrophotometer (Thermo Scientific).

2.3 Quantitative polymerase chain reaction

Two different fluorescence-based real-time qPCR assays were performed. The primer sequences for all markers are shown in Table S1[9, 29, 49, 53]. We confirmed in silico that these primers form neither unwanted secondary structures nor primer dimers using the software Primer Express v2.0 (Applied Biosystems). Focusing on the amplification of ALU115 and L1-79 DNA fragments, we determined an intraassay coefficient of variation (CoV) of 3.2% and the interassay CoV was 3.0% for ALU115. The corresponding values for L1-79 were 5.3% (intraassay CoV) and 2.8% (interassay CoV).

To quantify the ALU repeats, the following reaction mixture was used: 1 $\times$ Phusion buffer, 1 $\times$ Q-Solution, 4 $\mu$ l genomic DNA, 200 $\mu$ M deoxynucleotide triphosphate mixture, 2 mM MgCl ${}_{2}$ , 0.5 $\mu$ M forward Primer, 0.5 $\mu$ M reverse Primer and 0.02 U/ $\mu$ l Phusion ${}^{\text{\@setsize{\scriptsize}{8pt}{\viipt}{\@viipt}\textregistered}}$ High-Fidelity DNA Polymerase. Sybr Green was used as DNA binding dye. Quantitative PCR was done in a Mastercycler ep realplex ${}^{4}$ (Eppendorf, Hamburg, Germany) with precycling heat activation of the polymerase at 98 ${}^{\circ}$ C for 30 s, followed by 40 cycles of 98 ${}^{\circ}$ C for 30 s, 62 ${}^{\circ}$ C for 10s and 72 ${}^{\circ}$ C for 15 s. The reaction mixture for LINE1-79, LINE1-300 and ND1-mt consisted of 1 $\times$ PCR buffer, 1 $\times$ Q-solution, 4 $\mu$ l genomic DNA, 200 $\mu$ M deoxynucleotide triphosphate mixture, 0.5 $\mu$ M forward Primer, 0.5 $\mu$ M reverse Primer, 0.02 U/ $\mu$ l Paq5000 DNA Polymerase.

Real-time PCR was performed using the following conditions: 95 ${}^{\circ}$ C for 120 s, followed by 40 cycles of 95 ${}^{\circ}$ C for 20 s, 62 ${}^{\circ}$ C for 20 s and 72 ${}^{\circ}$ C for 15 s.

For each fcDNA of interest, a standard curve was generated using leucocyte DNA. A single standardised solution of leucocyte DNA was used as a standard curve reference in each qPCR run. Mean values across triplicates were used for further analysis. For each serum sample, the number of fcDNA copies was normalized to the sample volume.

2.4 Statistical analysis

Comparisons of marker values between different clinico-pathological groups were done using the Wilcoxon-Mann-Whitney test in case of two groups and the Jonckheere-Terpstra test in case of three or more groups.

Survival analysis was done with Kaplan-Meier survival curves with UICC stage 1 to 3 and UICC stage 4 being analyzed separately. For calculation of overall survival, the period between the date of diagnosis of the primary tumor and the date of death respectively end of follow-up was included. Hazard ratios (HRs) for univariate and multivariate analysis of overall survival were estimated using Cox’s regression model. All statistical analyses were performed with SAS 9.3 (SAS Institute, Cary, NC, USA).

3. Results

3.1 Correlation of DNA markers with clinicopathological parameters and DNA methylation status

Table 1
Correlation between copy numbers/CEA concentration/methylation status and clinical features of our study population

	Alu115			Alu247			L1-79			L1-300			ND1-mt
Clinicopathological variables	Median no. of copies	IQR	$p$ -values	Median no. of copies	IQR	$p$ -values	Median no. of copies	IQR	$p$ -values	Median no. of copies	IQR	$p$ -values	Median no. of copies	IQR	$p$ -values
Sex
Male ( $n=$ 142, 53%)	640	1,588		635	1,267		324	602		165	278		480	693
Female ( $n=$ 126, 47%)	729	1,693	n.s. ${}^{\text{a}}$	722	1,687	n.s. ${}^{\text{a}}$	359	616	n.s. ${}^{\text{a}}$	194	379	n.s. ${}^{\text{a}}$	366	617	n.s. ${}^{\text{a}}$
Age
$<$ 65 years ( $n=$ 135, 50%)	624	1,270		505	1,238		280	578		162	290		497	732
$\geqslant$ 65 years ( $n=$ 133, 50%)	771	1,896	n.s. ${}^{\text{a}}$	839	1,635	n.s. ${}^{\text{a}}$	437	650	0.036	199	315	n.s. ${}^{\text{a}}$	440	504	0.023
Localisations
Colon ( $n=$ 124, 46%)	618	1,695		604	1,373		295	573		180	337		492	669
Sigmond ( $n=$ 52, 20%)	689	1,321		895	1,687		353	735		150	252		321	559
Rectum ( $n=$ 92, 34%)	712	1,612	n.s. ${}^{\text{a}}$	581	1,168	n.s. ${}^{\text{a}}$	374	553	n.s. ${}^{\text{a}}$	181	269	n.s. ${}^{\text{a}}$	409	667	n.s. ${}^{\text{a}}$
Tumor size
T1 ( $n=$ 15, 5%)	685	1,001		562	1,183		228	650		192	372		467	736
T2 ( $n=$ 59, 22%)	432	657		505	942		295	426		141	276		388	721
T3 ( $n=$ 160, 60%)	761	1,936		716	1,684		334	583		166	311		442	711
T4 ( $n=$ 34, 13%)	958	2,527	0.050	774	3,269	0.038	574	807	n.s. ${}^{\text{a}}$	249	496	n.s. ${}^{\text{a}}$	487	467	n.s. ${}^{\text{a}}$
Nodal status
N0 ( $n=$ 146, 55%)	491	845		522	1,045		288	650		164	276		478	727
N1 ( $n=$ 69, 26%)	1,078	2,729		847	1,772		417	604		212	381		383	652
N2 ( $n=$ 49, 19%)	878	2,624	$<$ 0.001	568	3,732	0.008	333	389	n.s. ${}^{\text{a}}$	165	296	n.s. ${}^{\text{a}}$	386	534	n.s. ${}^{\text{a}}$
Metastasis
M0 ( $n=$ 185, 69%)	613	942		515	983		296	592		167	302		477	727
M1 ( $n=$ 83, 31%)	1,199	1,199	0.005	1,294	1,294	$<$ 0.001	363	637	n.s. ${}^{\text{a}}$	196	334	n.s. ${}^{\text{a}}$	325	494	n.s. ${}^{\text{a}}$
UICC stage
I ( $n=$ 57, 21%)	478	639		562	996		281	485		185	282		414	772
II ( $n=$ 69, 26%)	382	819		462	916		241	723		163	301		494	736
III ( $n=$ 59, 22%)	917	2,531		665	1,288		394	527		203	311		450	675
IV ( $n=$ 83, 31%)	1,199	2,796	$<$ 0.001	1,294	3,945	$<$ 0.001	363	637	n.s. ${}^{\text{a}}$	196	334	n.s. ${}^{\text{a}}$	325	494	n.s. ${}^{\text{a}}$
Tumor grade
G1 and G2 ( $n=$ 142, 55%)	626	1,277		581	1,230		293	546		171	291		446	665
G3 and G4 ( $n=$ 115, 45%)	752	1,665	n.s. ${}^{\text{a}}$	716	1,918	n.s. ${}^{\text{a}}$	398	605	0.022	163	310	n.s. ${}^{\text{a}}$	405	686	n.s. ${}^{\text{a}}$
CEA
$<$ 2.5 ng/ml ( $n=$ 116, 43%)	508	980		482	1,086		298	543		154	288		447	802
$\geqslant$ 2.5 ng/ml ( $n=$ 152, 57%)	836	2,313	$<$ 0.001	884	2,157	$<$ 0.001	371	655	n.s. ${}^{\text{a}}$	198	324	n.s. ${}^{\text{a}}$	429	617	n.s. ${}^{\text{a}}$
$<$ 27 ng/ml ( $n=$ 217, 81%)	613	1,181		562	1,192		312	590		165	288		460	666
$\geqslant$ 27 ng/ml ( $n=$ 51, 19%)	1,349	2,897	0.003	1,401	3,823	$<$ 0.001	538	725	0.008	234	544	n.s. ${}^{\text{a}}$	325	642	n.s. ${}^{\text{a}}$
HLTF
Negative ( $n=$ 227, 85%)	628	1,435		597	1,311		316	576		162	292		423	677
Positive ( $n=$ 41, 15%)	935	2,688	n.s. ${}^{\text{a}}$	1,122	3,125	0.014	637	1,087	0.032	237	435	0.027	490	620	n.s. ${}^{\text{a}}$
HPP1
Negative ( $n=$ 218, 81%)	613	1,057		585	1,186		323	576		165	290		442	681
Positive ( $n=$ 50, 19%)	2,261	4,241	$<$ 0.001	1,413	3,755	0.002	477	895	n.s. ${}^{\text{a}}$	229	457	n.s. ${}^{\text{a}}$	400	686	n.s. ${}^{\text{a}}$

${}^{\text{a}}p$ -value not significant.

Figure 1.

Correlation of copy numbers with clinicopathologigal parameters.

The characterization of our study population is shown in Table 1. The population consists of 142 female and 126 male patients with cancer in cecum, ascending, transverse and descending colon (46%), sigmoid colon (20%) and rectum (34%). 45% of the patients had positive regional lymph nodes. Distant metastases were present in 83 patients. UICC stage I tumors was 21%, UICC stage II 26%, UICC stage II 22% and UICC stage IV 31% of our patient population.

First, we quantified the copy numbers of each marker using qPCR. The median concentrations are shown in Table S2. Common clinicopathological parameters were analyzed for a relationship with the copy numbers of the candidate biomarkers (Table 1). The amount of ALU115 and ALU247 fcDNA showed a significant correlation with tumor size ( $P=$ 0.050 and $P=$ 0.038; respectively Fig. 1a–b) and also with nodal status ( $P<$ 0.001 and $P=$ 0.008; respectively Fig. 1c–d). In metastasized colorectal cancer, the copy numbers of ALU115 and ALU247 were significantly higher than in non-metastasized disease ( $P=$ 0.005 and $P<$ 0.001; respectively Fig. 1e–f). Furthermore, UICC stage correlated highly significantly with the copy number of both ALU fragments ( $P<$ 0.001; respectively Fig. 1g–h). Related to tumor grade, the amount of L1-79 was significantly higher in grade 3 and 4 than in grade 1 and 2 ( $P=$ 0.022). L1-79 and ND1-mt fragments significantly increased in patients above age 65 years ( $P=$ 0.036 and $P=$ 0.023).

Figure 2.

Overall survival in association with UICC stage

Besides the association with clinical parameters, we also analyzed the correlation of our markers with carcinoembryonic antigen (CEA) as a widely used tumor marker and the methylation status of HLTF and HPP1, which have already been demonstrated to be prognostic markers for CRC [40]. Carcinoembryonic antigen is more efficient in discriminating between good and poor prognosis when individualized cutoff values are used [23]. In this study, two different CEA levels served as cutoff values: The lower concentration (2.5 ng/ml) represents the 95 ${}^{\text{th}}$ percentile of healthy persons in the study population, the higher cutoff value (27 ng/ml) is based on the median CEA-concentration of UICC stage IV cases. Copy numbers of ALU115 and ALU247 showed a highly significant correlation with CEA levels above 2.5 ng/dl (both $P<$ 0.001). The amount of ALU115, ALU247 and L1-79 fragments correlated significantly with CEA levels above 27 ng/ml. Three markers correlated with the methylation status of HLTF: ALU247, L1-79 and L1-300 showed significantly higher copy numbers when HLTF was methylated ( $P=$ 0.014, $P=$ 0.032 and $P=$ 0.027). We also found a significant correlation of ALU115 and ALU247 with methylation of HPP1 ( $P<$ 0.001 and $P=$ 0.002).

3.2 Prognostic role of fcDNA biomarkers in overall survival

Next, we were interested in the prognostic value of raised copy numbers of ALU115, ALU247, L1-79, L1-300 and ND1-mt.

Figure 3.

Survival proportions with combinations of ALU115 and HPP1.

Table 2

Overall survival in UICC stages I–IV

	Cutoff value (copy no./ $\mu$ l)	$n$	Median survival (years)	95%-CI	$p$ -value
Alu115
$<$ median	668	134	9.6	5.4–13.5	0.004
$\geqslant$ median	668	134	5.2	3.0–6.4
1. quartile	$<$ 262	67	10.6	4.8–n.d.	0.001
2. quartile	262–667	67	9.6	4.1–n.d.
3. quartile	668–1,899	67	7.6	4.3–9.9
4. quartile	$\geqslant$ 1,900	67	2.8	1.2–5.6
Alu247
$<$ median	675	134	9.6	6.4–12.1	0.003
$\geqslant$ median	675	134	4.3	2.8–6.3
1. quartile	$<$ 234	66	11.7	6.7–n.d.	$<$ 0.001
2. quartile	234–674	68	7.8	4.1–10.6
3. quartile	675–1,656	67	9.0	4.6–n.d.
4. quartile	$\geqslant$ 1,687	67	2.1	1.2–3.6
LINE1-79
$<$ median	342	134	8.0	5.6–11.7	0.041
$\geqslant$ median	342	134	4.9	3.2–7.2
1. quartile	$<$ 112	67	8.0	4.8–n.d.	n.s.
2. quartile	112–341	67	8.7	3.3–11.7
3. quartile	342–715	67	4.9	3.1–7.8
4. quartile	$\geqslant$ 716	67	4.7	1.6–9.0
LINE1-300
$<$ median	168	134	8.0	5.4–11.7	0.041
$\geqslant$ median	168	134	5.2	3.0–7.5
1. quartile	$<$ 75	67	5.4	3.4–9.6	0.047
2. quartile	75–167	67	11.4	6.0–n.d.
3. quartile	168–385	67	4.3	2.5–8.0
4. quartile	$\geqslant$ 386	67	5.9	2.5–9.9
ND1-mt
$<$ median	438	134	5.6	4.1–8.7	n.s.
$\geqslant$ median	438	134	6.7	4.4–9.6
1. quartile	$<$ 167	67	5.4	3.7–8.7	n.s.
2. quartile	167–437	67	6.0	2.8–12.7
3. quartile	438–832	67	6.4	2.6–9.7
4. quartile	$\geqslant$ 833	67	7.6	5.2-15.1

n.d.: upper limit of 95% CI could not be calculated due to insufficient number of events in this group.

Overall survival in association with our potential markers was analyzed for all patients in all UICC stages (Fig. 2). Kaplan-Meier survival statistics were carried out with the median copy number as cutoff-value. Additionally, quartiles of the marker levels were used as cutoff-values for survival analysis. The results are shown in Table 2. The amount of ALU115 fcDNA fragments in serum of cancer patients showed a highly significant correlation with overall survival with both cutoff-values ( $P=$ 0.004; $P=$ 0.001). Thus, median surv4ival declined with increasing copy numbers of ALU115. The same relationship was observable for ALU247 fcDNA. Overall survival declined from 9.6 years to 4.3 years when ALU247 copy number was above the median ( $P=$ 0.003). Using quartiles for survival statistics, there was even a highly significant decrease in survival ( $P<$ 0.001). Furthermore, an increase of L1-79 fcDNA fragments correlated with shorter survival ( $P=$ 0.041). Patients with L1-300 fcDNA above the cutoff values died earlier ( $P=$ 0.041; $P=$ 0.047). Kaplan-Meier survival analysis did not reveal prognostic information of ND1-mt fcDNA.

Table 3

Overall survival of patients in stage UICC IV; multivariate analysis with biomarker combinations (including HLTF and HPP1)

	Univariate analysis						Multivariate analysis
	$n$	Median survival (years)	95%-CI	$p$ -value	Hazard ratio	$p$ -value	Hazard ratio	$p$ -value
Alu115					2.9 (1.8–4.8)	$<$ 0.001	2.8 (1.6–4.6)	$<$ 0.001
$<$ 1000	41	2.2	1.5–3.4
$\geqslant$ 1000	42	1.0	0.7–1.2	$<$ 0.001			–	–
Alu247					2.2 (1.3–3.6)	0.002
$<$ 1290	41	2.0	1.2–3.4				–	–
$\geqslant$ 1290	42	1.1	0.8–1.3	0.001
LINE1-79					2.0 (1.2–3.2)	0.006	–	–
$<$ 360	41	1.7	1.2–2.5
$\geqslant$ 360	42	1.0	0.8–1.3	0.005			–	–
LINE1-300					2.3 (1.4–3.9)	0.001
$<$ 170	41	1.9	1.1–6.6				–	–
$\geqslant$ 170	42	1.1	0.8–1.3	0.001
ND1-mt					1.3 (0.8–2.1)	0.290	–	–
$<$ 325	41	1.5	0.9–2.1
$\geqslant$ 325	42	1.2	0.9–1.9	0.289			–	–
CEA					1.7 (1.1–2.8)	0.028
$<$ 27 ng/ml	43	1.8	1.1–2.5				–	–
$\geqslant$ 27 ng/ml	40	1.1	0.8–1.5	0.027
HLTF					2.4 (1.4–4.3)	0.003	–	–
Neg.	65	1.6	1.2–2.1
Pos.	18	0.8	0.5–1.0	0.002			–	–
HPP1					2.1 (1.3–3.3)	0.003	1.9 (1.2–3.1)	0.008
Neg.	43	1.7	1.2–2.4
Pos.	40	0.9	0.6–1.2	0.003

Second, overall survival of patients in stage UICC IV according to the five potential biomarkers was determined separately (Table 3). Determination of survival in UICC stage IV subset was done using the median copy number of each marker as cutoff value. The amount of ALU115 fragments showed a highly significant correlation with survival in UICC stage IV ( $P<$ 0.001, Fig. 1a). Median survival was 2.2 years for copy numbers below and only 1.0 years for copy numbers above the cutoff value. ALU247 also turned out to be able to discriminate between poor and good prognosis in metastasized disease ( $P=$ 0.001, Fig. 1b). Likewise, increased copy numbers of L1-79 and L1-300 fragments above the cutoff values (360/ $\mu$ l; 170/ $\mu$ l) significantly correlated with survival ( $P=$ 0.005, Fig. 1c; $P=$ 0.001, Fig. 1d). There was no statistically significant correlation observable for the amount of ND1-mt fragments and patient survival in UICC stage IV.

3.3 Multivariate survival analysis

The five fcDNA sequences, the methylation markers HLTF and HPP1, which already have proved their prognostic value for CRC in former studies of our group [39], and CEA were included in a multivariate Cox model. Only ALU115 (HR $=$ 2.754; 95% Cl 1.662–4.565; $P<$ 0.001) and HPP1 (HR $=$ 1.925; 95% Cl 1.187–3.121; $P=$ 0.008) turned out to be independent prognostic markers for CRC in UICC stage IV (Table 3).

Survival statistics with combinations of both markers showed highly significant differences in survival ( $P<$ 0.001). When both markers were negative (meaning the copy numbers were below the cutoff-value), median survival in UICC stage IV was 2.4 years. With ALU115 negative and HPP1 positive, patients survived 1.5 years. When ALU115 was positive and HPP1 negative, median survival was 1.2 years. With both markers being positive, patients survived only 0.7 years (Fig. 3).

4. Discussion

It is estimated that incidences of the 14 most common cancer types including CRC will increase in the future [42]. Furthermore, according to these projections concerning cancer incidences and cancer-related deaths in Germany, CRC will be the third most common cause of cancer-related deaths in men and women by 2030. Even though overall survival of colorectal cancer patients is improving continuously [5], among patients with metastatic disease no progress in survival can be observed [31]. Therefore, new therapeutic strategies would be needed, individually adapted to a patient’s medical conditions, his or her medical condition considering his comorbidities and commensurate to the risks of any medical treatment. To improve the overall survival of UICC stage IV patients and evaluate the individual risk of each patient avoiding under- or over-treatment, it is important to establish new prognostic tools.

In this study, we were able to identify ALU115 fcDNA as an independent prognostic marker for the survival of CRC patients. The prognostic value of ALU115 could be demonstrated in the overall survival analysis for all tumor stages. Notably, survival of patients with UICC stage IV was significantly shorter when ALU115 was above the cutoff value, which was determined by the median copy number of ALU115 fragments in the blood serum of UICC stage IV patients. Compared to CEA as a common clinical marker, ALU115 correlated stronger with clinicopathological parameters as well as survival in UICC stage IV.

In order to maximize the prognostic value, combination of markers could be a useful approach. In a previous study, we demonstrated the prognostic value of HPP1 methylation for UICC stage IV in the identical patient population [40]. Combining HPP1 and ALU115 in Kaplan Meier survival analysis, we were able to define three different risk groups. With both markers being positive, the overall survival was substantially shorter than with only one marker respectively none of both markers being positive. In summary, both markers have a similar prognostic value, but in combination, the prognostic value is even higher.

The analysis of fcDNA in serum as a non-invasive, low-risk “liquid profiling” [44] provides various information about the primary tumor characteristics. Therefore, it is a helpful instrument for understanding metastatic progression and hence a potential prognostic device for cancer patients. Here, sampling of the specimens can easily be repeated to, for example, monitor the amount of fcDNA over a longer period of time.

However, it hast to be taken into consideration that concentrations of cell-free DNA are not only higher in the blood of CRC patients. Shaban et al. evaluated the diagnostic and prognostic value of ALU fcDNA in different cancer types and found that ALU levels in different cancers (type or stage) are strongly varying in the included studies due probe processing, analysis and calculation of concentrations [45]. Levels of fcDNA can also be increased due to chronic inflammations, e.g. systemic lupus erythematosus [50], ischemic heart disease, diabetes mellitus [48] and is discussed to be influenced by psychosocial and physical stress conditions [20]. Furthermore, significant differences in ALU concentrations were found in psychiatric disorders like schizophrenia, major depressive disorder and alcohol-induced psychotic disorder [41] due to neuroinflammatory processes.

In our study population, besides sex and age, only clinicopathological data concerning CRC was recorded. Comorbidities, notably inflammatory conditions, or other types of cancer, were not captured in our study in contrast to Mead et al. [34], where patients were screened by history, patient letters and recent blood tests. However, we had no evidence that ALU concentrations were confounded by other diseases as our data were intrinsically conclusive and we had strong correlations with other markers such as CEA and HLTF/HPP1.

Further studies studying the suitability of fcDNA as a biomarker should consider the general health conditions of the study patients to exclude confounding comorbidities.

Considering the fact that our ALU markers significantly correlate with UICC stage, tumor size, nodal status and metastasis, we expect a decrease of fcDNA concentration after surgery due to a loss of tumor burden. Former studies showed inconsistent data: Bhangu et al. could detect a slight increase of fcDNA concentration after surgery, whereas ALU115 and ALU247/115 significantly and progressively decreased after surgical treatment in a study by Hao et al. [17] An increase of fcDNA markers shortly after surgery may be caused by mechanical and thermic manipulation as well as destruction of both healthy and tumor cells. So, the period of time between surgery and measurement of fcDNA concentrations seems to play an important role and should be chosen thoroughly. The changes of fcDNA concentrations in blood after different types of medical treatment, i.e., surgery, chemo- and immunotherapy, should be subject of further investigations as this question was not addressed before and new therapies are developing quickly.

In order to introduce more biomarkers to clinical practice, future studies need to consider the combination of markers and avoid selection bias. Sample proceeding should be standardized to reduce costs, accelerate analysis and generate a comparable standard. Most importantly, results need to be validated in larger studies in various populations, followed by prospective trials.

In conclusion, according to our findings, the main advantage of ALU115 fcDNA over common markers is to improve the prognostic assessment for CRC patients with metastatic disease.

Author contributions

Conception: Anzinger Isabel, Kolligs Frank T.

Interpretation or analysis of data: Anzinger Isabel, Nagel Dorothea, Ofner Andrea.

Preparation of the manuscript: Anzinger Isabel, Herbst Andreas.

Revision for important intellectual content: Philipp Alexander B., De Toni Enrico N., Holdt Lesca M., Teupser Daniel.

Supervision: Herbst Andreas, Kolligs Frank.

Footnotes

Supplementary data

Supplementary Table S1

Primer sequences

fcDNA marker	Forward primer	Reverse primer
ALU115	5’-CCTGAGGTCAGGAGTTCGAG-3’	5’-CCCGAGTAGCTGGGATTACA-3’
ALU247	5’-GTGGCTCACGCCTGTAATC-3’	5’-CAGGCTGGAGTGCAGTGG-3
L1-79	5’-AGGGACATGGATGAAATTGG-3’	5’-TGAGAATATGCGGTGTTTGG-3
L1-300	5’-ACACCTATTCCAAAATTGACCAC-3’	5’-TTCCCTCTACACACTGCTTTGA-3
ND1-mt	5’-CACCCAAGAACAGGGTTTGT-3’	5’-TGGCCATGGGATTGTTGTTAA-3

Supplementary Table S2

Median concentrations of fcDNA biomarkers/CEA; HLTF/HPP1: no. of patients with positive methylation status

	ALU115	ALU247	L1-79	L1-300	ND1-mt	CEA	HLTF	HPP1
Median concentration	668/ $\mu$ l	675/ $\mu$ l	342/ $\mu$ l	168/ $\mu$ l	438/ $\mu$ l	3.2 ng/ml	–	–
(range)	(13–18,016)	(1.7–26,096)	(1.6–3,925)	(4.4–3,980)	(8.5–6,644)	(1.0–3,945)	–	–
No./% positive	–	–	–	–	–	–	41/15	50/19
IQR	1636	1423	605	311	666	11.1	–	–

References

Abo-Salem

H.M.

Gibriel

A.A.

El Awady

M.E.

and Mandour

A.H.

, Synthesis, Molecular Docking and Biological Evaluation of Novel Flavone Derivatives as Potential Anticancer Agents Targeting Akt, Med Chem 17 (2021), 158–170.

Alix-Panabières

Schwarzenbach

and Pantel

, Circulating tumor cells and circulating tumor DNA, Annual Review of Medicine 63 (2012), 199–215.

Arnold

Karim-Kos

H.E.

Coebergh

J.W.

Byrnes

Antilla

Ferlay

Renehan

A.G.

Forman

and Soerjomataram

, Recent trends in incidence of five common cancers in 26 European countries since 1988: Analysis of the European Cancer Registry database, European Journal of Cancer (Oxford, England: 1990) (2013).

Bianchi

Laghi

Delconte

and Malesci

, Prognostic value of colorectal cancer biomarkers, Cancers 3 (2011), 2080–2105.

Brenner

Bouvier

A.M.

Foschi

Hackl

Larsen

I.K.

Lemmens

Mangone

and Francisci

, Progress in colorectal cancer survival in Europe from the late 1980s to the early 21st century: the EUROCARE study, International Journal of Cancer 131 (2012), 1649–1658.

Brierley

J.D.

Gospodarowicz

M.K.

and Wittekind

, TNM Classification of Malignant Tumours, Wiley, 2016.

Chen

Zhang

Zhou

Luo

Wang

and Wang

, Clinical application of plasma mitochondrial DNA content in patients with lung cancer, Oncol Lett 16 (2018), 7074–7081.

Cordaux

and Batzer

M.A.

, The impact of retrotransposons on human genome evolution, Nature Reviews Genetics 10 (2009), 691–703.

Diehl

Schmidt

Choti

M.A.

Romans

Goodman

Thornton

Agrawal

Sokoll

Szabo

S.A.

Kinzler

K.W.

Vogelstein

and Diaz

L.A.

, Circulating mutant DNA to assess tumor dynamics, Nature Medicine 14 (2008), 985–990.

10.

Duffy

M.J.

, Carcinoembryonic antigen as a marker for colorectal cancer: is it clinically useful?, Clinical Chemistry 47 (2001), 624–630.

11.

Duffy

M.J.

van Dalen

Haglund

Hansson

Holinski-Feder

Klapdor

Lamerz

Peltomaki

Sturgeon

and Topolcan

, Tumour markers in colorectal cancer: European Group on Tumour Markers (EGTM) guidelines for clinical use, European Journal of Cancer (Oxford, England: 1990) 43 (2007), 1348–1360.

12.

El-Gayar

El-Abd

Hassan

and Ali

, Increased Free Circulating DNA Integrity Index as a Serum Biomarker in Patients with Colorectal Carcinoma, Asian Pac J Cancer Prev 17 (2016), 939–44.

13.

Ferlay

Colombet

Soerjomataram

Dyba

Randi

Bettio

Gavin

Visser

and Bray

, Cancer incidence and mortality patterns in Europe: Estimates for 40 countries and 25 major cancers in 2018, Eur J Cancer 103 (2018), 356–387.

14.

Gibriel

A.A.

, Options available for labelling nucleic acid samples in DNA microarray-based detection methods, Brief Funct Genomics 11 (2012), 311–8.

15.

Gibriel

A.A.

Ismail

M.F.

Sleem

Zayed

Yosry

El-Nahaas

S.M.

and Shehata

N.I.

, Diagnosis and staging of HCV associated fibrosis, cirrhosis and hepatocellular carcinoma with target identification for miR-650, 552-3p, 676-3p, 512-5p and 147b, Cancer Biomark 34 (2022), 413–430.

16.

González-Masiá

J.A.

García-Olmo

and García-Olmo

D.C.

, Circulating nucleic acids in plasma and serum (CNAPS): applications in oncology, OncoTargets and Therapy 6 (2013), 819–832.

17.

Hao

T.B.

Shi

Shen

X.J.

X.H.

Tang

Y.Y.

and Ju

S.Q.

, Circulating cell-free DNA in serum as a biomarker for diagnosis and prognostic prediction of colorectal cancer, British Journal of Cancer 111 (2014), 1482–1489.

18.

Hashad

D.I.

Elyamany

A.S.

and Salem

P.E.

, Mitochondrial DNA Copy Number in Egyptian Patients with Hepatitis C Virus-Related Hepatocellular Carcinoma, Genet Test Mol Biomarkers 19 (2015), 604–9.

19.

Houck

C.M.

Rinehart

F.P.

and Schmid

C.W.

, A ubiquitous family of repeated DNA sequences in the human genome, J Mol Biol 132 (1979), 289–306.

20.

Hummel

E.M.

Hessas

Muller

Beiter

Fisch

Eibl

Wolf

O.T.

Giebel

Platen

Kumsta

and Moser

D.A.

, Cell-free DNA release under psychosocial and physical stress conditions, Transl Psychiatry 8 (2018), 236.

21.

Jahr

Hentze

Englisch

Hardt

Fackelmayer

F.O.

Hesch

R.D.

and Knippers

, DNA fragments in the blood plasma of cancer patients: quantitations and evidence for their origin from apoptotic and necrotic cells, Cancer Research 61 (2001), 1659–1665.

22.

Jemal

Bray

Center

M.M.

Ferlay

Ward

and Forman

, Global cancer statistics, CA: A Cancer Journal for Clinicians 61 (2011), 69–90.

23.

Jeon

B.G.

Shin

Chung

J.K.

Jung

I.M.

and Heo

S.C.

, Individualized Cutoff Value of the Preoperative Carcinoembryonic Antigen Level is Necessary for Optimal Use as a Prognostic Marker, Annals of Coloproctology 29 (2013), 106–114.

24.

Kin

Kidess

Poultsides

G.A.

Visser

B.C.

and Jeffrey

S.S.

, Colorectal cancer diagnostics: biomarkers, cell-free DNA, circulating tumor cells and defining heterogeneous populations by single-cell analysis, Expert Review of Molecular Diagnostics 13 (2013), 581–599.

25.

Kananazawa

Yamada

Kakinuma

Matsuno

Ando

Kuriyama

Matsuda

and Yoshida

, Methylation status and long-fragment cell-free DNA are prognostic biomarkers for gastric cancer, Cancer Med 10 (2021), 2003–2012.

26.

Kou

Y.N.

Cen

and Li

X.Y.

, Research and application on LINE-1 in diagnosis and treatment of tumorigenesis, Yi Chuan 43 (2021), 571–579.

27.

Kumar

Srivastava

Singh

S.A.

Das

A.K.

Das

G.C.

Dhar

Ghosh

S.K.

and Mondal

, Cell-free mitochondrial DNA copy number variation in head and neck squamous cell carcinoma: A study of non-invasive biomarker from Northeast India, Tumour Biol 39 (2017), 1010428317736643.

28.

Leon

S.A.

Shapiro

Sklaroff

D.M.

and Yaros

M.J.

, Free DNA in the serum of cancer patients and the effect of therapy, Cancer Research 37 (1977), 646–650.

29.

Lin

C.-S.

Wang

L.-S.

Tsai

C.-M.

and Wei

Y.-H.

, Low copy number and low oxidative damage of mitochondrial DNA are associated with tumor progression in lung cancer tissues after neoadjuvant chemotherapy, Interactive Cardiovascular and Thoracic Surgery 7 (2008), 954–958.

30.

Liu

Baklaushev

V.P.

Chekhonin

V.P.

Peltzer

Wang

and Zhang

, Nomogram for predicting overall survival in colorectal cancer with distant metastasis, BMC Gastroenterol 21 (2021), 103.

31.

Majek

Gondos

Jansen

Emrich

Holleczek

Katalinic

Nennecke

Eberle

and Brenner

, Survival from colorectal cancer in Germany in the early 21st century, British Journal of Cancer 106 (2012), 1875–1880.

32.

Malesci

and Laghi

, Novel prognostic biomarkers in colorectal cancer, Digestive Diseases (Basel, Switzerland) 30 (2012), 296–303.

33.

Marzese

D.M.

Hirose

and Hoon

D.S.B.

, Diagnostic and prognostic value of circulating tumor-related DNA in cancer patients, Expert Review of Molecular Diagnostics 13 (2013), 827–844.

34.

Mead

Duku

Bhandari

and Cree

I.A.

, Circulating tumour markers can define patients with normal colons, benign polyps, and cancers, British Journal of Cancer 105 (2011), 239–245.

35.

Miller

S.A.

Dykes

D.D.

and Polesky

H.F.

, A simple salting out procedure for extracting DNA from human nucleated cells, Nucleic Acids Research 16 (1988), 1215.

36.

Mohd Khair

S.Z.N.

Abd Radzak

S.M.

and Mohamed Yusoff

A.A.

, The Uprising of Mitochondrial DNA Biomarker in Cancer, Dis Markers 2021 (2021), 7675269.

37.

Nguyen

L.H.

Goel

and Chung

D.C.

, Pathways of Colorectal Carcinogenesis, Gastroenterology 158 (2020), 291–302.

38.

Nitsche

Maak

Schuster

Künzli

Langer

Slotta-Huspenina

Janssen

K.-P.

Friess

and Rosenberg

, Prediction of prognosis is not improved by the seventh and latest edition of the TNM classification for colorectal cancer in a single-center collective, Annals of Surgery 254 (2011), 793–800; discussion 800.

39.

Philipp

A.B.

Nagel

Stieber

Lamerz

Thalhammer

Herbst

and Kolligs

F.T.

, Circulating cell-free methylated DNA and lactate dehydrogenase release in colorectal cancer, BMC Cancer 14 (2014), 245.

40.

Philipp

A.B.

Stieber

Nagel

Neumann

Spelsberg

Jung

Lamerz

Herbst

and Kolligs

F.T.

, Prognostic role of methylated free circulating DNA in colorectal cancer, International Journal of Cancer 131 (2012), 2308–2319.

41.

Chen

L.Y.

Shen

X.J.

and Ju

S.Q.

, Analytical Value of Cell-Free DNA Based on Alu in Psychiatric Disorders, Front Psychiatry 10 (2019), 992.

42.

Quante

A.S.

Ming

Rottmann

Engel

Boeck

Heinemann

Westphalen

C.B.

and Strauch

, Projections of cancer incidence and cancer-related deaths in Germany by 2020 and 2030, Cancer Med 5 (2016), 2649–56.

43.

Rodic

, LINE-1 activity and regulation in cancer, Front Biosci (Landmark Ed) 23 (2018), 1680–1686.

44.

Schwarzenbach

Hoon

D.S.B.

and Pantel

, Cell-free nucleic acids as biomarkers in cancer patients, Nature Reviews Cancer 11 (2011), 426–437.

45.

Shaban

S.A.

Al-Rahim

A.M.

and Suleiman

A.A.

, ALU repeat as potential molecular marker in the detection and prognosis of different cancer types: A systematic review, Mol Clin Oncol 16 (2022), 86.

46.

Soliman

S.E.

Alhanafy

A.M.

Habib

M.S.E.

Hagag

and Ibrahem

R.A.L.

, Serum circulating cell free DNA as potential diagnostic and prognostic biomarker in non small cell lung cancer, Biochem Biophys Rep 15 (2018), 45–51.

47.

Souissi

Ben Said

Ben Ayed

Elloumi

Bouzid

Mosrati

M.A.

Hasnaoui

Belcadhi

Idriss

Kamoun

Gharbi

Gibriel

A.A.

Tlili

and Masmoudi

, Novel pathogenic mutations and further evidence for clinical relevance of genes and variants causing hearing impairment in Tunisian population, J Adv Res 31 (2021), 13–24.

48.

Spindler

K.-L.G.

Appelt

A.L.

Pallisgaard

Andersen

R.F.

Brandslund

and Jakobsen

, Cell-free DNA in healthy individuals, noncancerous disease and strong prognostic value in colorectal cancer, International Journal of Cancer (2014).

49.

Sunami

A.-T.

Nguyen

S.L.

Giuliano

A.E.

and Hoon

D.S.B.

, Quantification of LINE1 in circulating DNA as a molecular biomarker of breast cancer, Annals of the New York Academy of Sciences 1137 (2008), 171–174.

50.

Truszewska

Foroncewicz

and Paczek

, The role and diagnostic value of cell-free DNA in systemic lupus erythematosus, Clin Exp Rheumatol 35 (2017), 330–336.

51.

Ule

, Alu elements: at the crossroads between disease and evolution, Biochem Soc Trans 41 (2013), 1532–5.

52.

Umetani

Giuliano

A.E.

Hiramatsu

S.H.

Amersi

Nakagawa

Martino

and Hoon

D.S.B.

, Prediction of breast tumor progression by integrity of free circulating DNA in serum, Journal of Clinical Oncology: Official Journal of the American Society of Clinical Oncology 24 (2006), 4270–4276.

53.

Umetani

Kim

Hiramatsu

Reber

H.A.

Hines

O.J.

Bilchik

A.J.

and Hoon

D.S.B.

, Increased integrity of free circulating DNA in sera of patients with colorectal or periampullary cancer: direct quantitative PCR for ALU repeats, Clinical Chemistry 52 (2006), 1062–1069.

54.

van der Vaart

and Pretorius

P.J.

, Circulating DNA. Its origin and fluctuation, Annals of the New York Academy of Sciences 1137 (2008), 18–26.

55.

Volckmar

A.L.

Sultmann

Riediger

Fioretos

Schirmacher

Endris

Stenzinger

and Dietz

, A field guide for cancer diagnostics using cell-free DNA: From principles to practice and clinical applications, Genes Chromosomes Cancer 57 (2018), 123–139.

56.

Wanebo

H.J.

Rao

Pinsky

C.M.

Hoffman

R.G.

Stearns

Schwartz

M.K.

and Oettgen

H.F.

, Preoperative Carcinoembryonic Antigen Level as a Prognostic Indicator in Colorectal Cancer, New England Journal of Medicine 299 (1978), 448–451.

57.

Wang

B.G.

Huang

H.-Y.

Chen

Y.-C.

Bristow

R.E.

Kassauei

Cheng

C.-C.

Roden

Sokoll

L.J.

Chan

D.W.

and Shih

I.-M.

, Increased plasma DNA integrity in cancer patients, Cancer Research 63 (2003), 3966–3968.

58.

Yasser

M.B.

Abdellatif

Emad

Jafer

Ahmed

Nageb

Abdelshafy

Al-Anany

A.M.

Al-Arab

M.A.E.

and Gibriel

A.A.

, Circulatory miR-221 & miR-542 expression profiles as potential molecular biomarkers in Hepatitis C Virus mediated liver cirrhosis and hepatocellular carcinoma, Virus Res 296 (2021), 198341.

Cell-free circulating ALU repeats in serum have a prognostic value for colorectal cancer patients

Abstract

BACKGROUND:

OBJECTIVE:

METHODS:

RESULTS:

CONCLUSIONS:

Keywords

Abbreviations

1. Introduction

2. Methods

2.1 Patients and serum samples

2.2 DNA isolation

2.3 Quantitative polymerase chain reaction

2.4 Statistical analysis

3. Results

3.1 Correlation of DNA markers with clinicopathological parameters and DNA methylation status

Table 1 Correlation between copy numbers/CEA concentration/methylation status and clinical features of our study population

4. Discussion

Author contributions

Footnotes

Supplementary data

References

Table 1
Correlation between copy numbers/CEA concentration/methylation status and clinical features of our study population