Ovarian cancer is the most common malignancy in the female genital system and the fourth most common cause of death in women from cancer.1 The need for specific and sensitive markers of epithelial ovarian cancer is critical because, in two thirds of all patients, it is diagnosed only in advanced stages.2 The majority of women with epithelial ovarian cancer have vague or nonspecific symptoms. The most important sign of epithelial ovarian cancer is the presence of a pelvic mass on physical examination. The diagnosis of ovarian cancer requires a exploratory laparotomy or a diagnostic laparoscopy even though serum CA 125 levels have been shown to be useful in distinguishing malignant from benign pelvic masses.3 For a postmenopausal patient, an adnexal mass combined with a very high serum CA 125 level (higher than 200 U/mL) shows a 96% positive predictive value for malignancy. However, for premenopausal patients, the specificity of the test is low.4 An attractive alternative seems to be the use of a combination of markers, such as imaging markers and blood biomarkers, in ovarian screening.5
A new diagnostic opportunity is the measurement of total circulating cell-free DNA levels in cancer. Most studies report that elevated levels of circulating cell-free nuclear DNA were observed in various cancers, suggesting that the circulating molecules might be a potential biomarker for clinical applications.6–9 The presence of circulating cell-free mitochondrial DNA in circulation first was demonstrated in 2000 for diabetes mellitus.10 Compared with the nuclear genome, the high copy numbers of the mitochondrial genome allow the suggestion that the amount of circulating cell-free mitochondrial DNA should be higher than that of circulating cell-free nuclear DNA.11 Mehra et al demonstrated elevated levels of circulating cell-free mitochondrial DNA in the plasma of prostate cancer patients using quantitative polymerase chain reaction (PCR).12 Recently, Elliger et al (2008) observed that circulating cell-free mitochondrial DNA in the serum of patients with prostate cancer has a predictive value for biochemical recurrence after prostatectomy.13
In this study, using a multiplex real-time PCR, we examined the circulating cell-free mitochondrial DNA and circulating cell-free nuclear DNA in the serum and plasma of patients with ovarian tumors and compared them with those of a healthy control group to investigate whether circulating cell-free nuclear DNA and circulating cell-free mitochondrial DNA also have a value in the management of cases of benign or malignant ovarian lesions.
MATERIALS AND METHODS
The study was undertaken at the Department of Obstetrics and Gynaecology, Department of BioMedicine, University of Basel, and was approved by the local institutional review board (Ethics commission beider Basel). Written consent forms were collected from all patients involved in this study.
Using a common study protocol, from March 2006 until May 2007, 104 women were recruited from Women’s Hospital of University Basel and the Women’s Hospital of Liestal. Most individuals were European whites. The patients were identified with clinical examinations and ultrasonography. The blood samples were taken before any invasive procedures and before any treatments. Clinical information including demographic information, CA 125 level, tumor stage, histology, and ultrasonographic finding was obtained from the patients’ medical records and pathology reports. Patients with a history of cancer or a cancer from other systems or inflammatory diseases as a side diagnosis were excluded. All endometriosis patients had affected ovaries. Women in for an annual checkup in the hospitals who agreed to have blood sampling but who had no history of cancer or endometriosis, no suspected lesions, no inflammatory diseases or diabetes mellitus, and no drugs taken comprised the control group. No further invasive diagnostic procedures were performed for the control group.
The study cohort (Nequals;104) (Table 1) was divided into four main groups: endometriosis (nequals;23), benign epithelial ovarian tumor (nequals;24), epithelial ovarian cancer (nequals;21), and a healthy control group (nequals;36). The benign epithelial ovarian tumor group contained 17 patients with serous cystadenoma, six with mucinous cystadenoma, and one with serous-mucinous cystadenoma. All diagnoses of epithelial ovarian cancers (serous or mucinous type), benign epithelial ovarian tumors, and endometriosis were confirmed by histopathology after surgery.
The mean age of the endometriosis group was 34 years (25–43 years), the mean age of the benign epithelial ovarian tumor group was 58 years (22–93 years), the mean age of the epithelial ovarian cancer group was 61 years (32–90 years), and the control group had a mean age of 53 years (21–78 years). Whereas the other three groups were age matched (Kruskal-Wallis test, P=.537), the endometriosis group shows a significantly younger age (Kruskal-Wallis test, P<.001) because of the pathogenesis of the disease.
The tumor volume in epithelial ovarian cancers was obtained before any tumor reduction by ultrasonography. CA 125 was measured before any treatment or surgery. The common Swiss cutoff value of 35 U/mL was used to distinguish between normal and pathologic cases.
Blood samples were processed strictly according to our common experimental protocol described in previous publications.14 All samples were stored at −80°C until further use.
DNA was extracted from 500 microliters of serum and plasma using the automated method with the MagNA Pure LC DNA Isolation Kit—Large Volume and the MagNA Pure LC Instrument (Roche Applied Science, Indianapolis, IN). DNA was eluted into 100-microliter elution buffers.
The amounts of nuclear DNA and mitochondrial DNA were quantified by multiplex real-time PCR for both the glyceraldehype-3-phosphodehydrogenase (GAPDH) gene and the mitochondrial DNA encoded ATPase (MTATP) 8 gene.
The primer and probe sequences for GAPDH and MTATP 8 gene are shown as follows: GAPDH (forward): 5′ CCC CAC ACA CAT GCA CTT ACC 3′; (reverse): 5′CCT AGT CCC AGG GCT TTG ATT 3′; probe 5′ (MGB) TAG GAA GGA CAG GCA AC (VIC) 3′. Mitochondrial DNA (forward): 5′ AAT ATT AAA CAC AAA CTA CCA CCT ACC 3′; (reverse): 5′ TGG TTC TCA GGG TTT GTT ATA 3′; probe: 5′ (MGB) CCT CAC CAA AGC CCA TA (FAM) 3′.
Quantification of circulating cell-free DNA using multiplex real-time PCR was performed using the ABI Prism 7000 Sequence Detector (Applied Biosystems, Foster City, CA). Theoretical and practical aspects of real-time PCR have been described in detail previously.15–17
The amplification reactions and the thermal profile were set up as an optimized standard in our laboratory.18,19 All samples were analyzed in triplicate, and results with delta Ct (ΔCt, difference of Ct value) greater than 0.8 were repeated (Ct, a threshold cycle, reflects the cycle number at which a fluorescence signal within a reaction crosses a threshold). Multiple water blanks were included in each run, and the standard calibrator curve with known concentration for genomic DNA ranging from 3.125×104 to 10 pg/microliter with a dilution factor of 5 (including 31,250, 6,250, 1,250, 250, 50, and 10 pg/microliter) was used. The results were expressed as genome equivalents by use of the conversion factor of 6.6 pg of DNA per cell.20,21 The plasma and serum circulating cell-free nuclear DNA concentration per milliliter was calculated using an equation published previously.17
We examined the amplification efficiency for both GAPDH and MTATP 8 on experimental serial dilutions. The amplifications of mitochondrial DNA and nuclear DNA on serial dilutions showed a good correlation, with an efficiency of close 100%. Based on the comparable efficiency, mitochondrial DNA concentration was calculated using the following formula: Cnmitochondrial DNA = 2ΔCT(mitochondrial DNA-nuclear DNA) × Cnnuclear DNA. (Cn, concentration: genome equivalents per mL).
All statistical analyses were performed using SPSS 15.0 (SPSS Inc., Chicago, IL). The Shapiro-Wilk test showed that our data were not normally distributed. The Kruskal-Wallis test and Mann-Whitney test were used in this study. The Spearman rank test was applied to analyze the relationship between levels of circulating cell-free nuclear DNA and quantities of circulating cell-free mitochondrial DNA. A P≤.05 was considered significant. Receiver operating characteristic (ROC) curves were used to determine cutoff points for distinguishing between affected and nonaffected individuals.22
Levels of circulating cell-free nuclear DNA and circulating cell-free mitochondrial DNA in the four study groups (healthy controls, benign epithelial ovarian tumor, epithelial ovarian cancer, and endometriosis) are given in Table 2 as the mean, median, and range.
In the plasma samples, circulating cell-free mitochondrial DNA levels were significantly higher than circulating cell-free nuclear DNA levels (107,892–21,945,446 compared with 354–46,137, 305–476 folds, P<.001). No correlation was found between plasma circulating cell-free mitochondrial DNA levels and plasma circulating cell-free nuclear DNA levels (Spearman rank test, correlation coefficient 0.216, P=.138).
Plasma circulating cell-free nuclear DNA was significantly increased in the epithelial ovarian cancer group compared with two of the groups—the benign epithelial ovarian tumor group and the control group (circulating cell-free nuclear DNA 3,690 compared with 2,000 and 3,690 compared with 1,708, Mann-Whitney test P=.027 and .009) (Fig. 1A). The level of plasma circulating cell-free nuclear DNA was lower in the endometriosis group when compared with the level in the epithelial ovarian cancer group (2,185 compared with 3,690). However, no statistically significant difference was found between the two groups (Mann-Whitney test P=.105).
An elevated level of plasma circulating cell-free mitochondrial DNA was found in the epithelial ovarian cancer group compared with the other three groups (benign epithelial ovarian tumor, endometriosis, and the control group) (circulating cell-free mitochondrial DNA 1,781,096 compared with 1,109,503, 1,781,096 compared with 1,124,673, 1,781,096 compared with 1,206,300, Mann-Whitney test P=.002, .013, and .022, respectively) (Fig. 1B).
The circulating cell-free nuclear DNA and circulating cell-free mitochondrial DNA levels in serum samples were significantly higher than those in plasma samples in all four study groups (Table 2); however, no significance was found between any groups. The release of cellular DNA artifact during blood-clotting procedures may cause the results. No correlation was observed between serum circulating cell-free nuclear DNA levels and serum circulating cell-free mitochondrial DNA levels (Spearman rank test correlation coefficient 0.180, P=.323).
Using an ROC curve, the results of the two markers (plasma circulating cell-free mitochondrial DNA and circulating cell-free nuclear DNA) for discriminating between individuals with malignancies and healthy individuals as well as between malignant and benign cases were evaluated. Figure 2A and Figure 3A, for circulating cell-free nuclear DNA results, and Figure 2B and Figure 3B, for circulating cell-free mitochondrial DNA results, show graphics with the sensitivity plotted on the y coordinate compared with 1-specificity or the false positive rate plotted on the x coordinate. The optimal discrimination threshold as a cutoff point has been selected according to the binary classifier system. The cutoff values of a mean of 2,653 circulating cell-free nuclear DNA genome equivalent/mL and a mean of 1,686,154 mitochondrial DNA genome equivalent/mL were chosen by ROC curve analysis, which provided a sensitivity of 74% and a specificity of 69% for the circulating cell-free nuclear DNA assay (area under the curve [AUC] 0.72, 95% confidence interval [CI] 0.567–0.874) and a sensitivity of 63% and a specificity of 67% for the circulating cell-free mitochondrial DNA assay (AUC 0.71, 95% CI 0.569–0.852), to discriminate between the epithelial ovarian cancer group and healthy control group (Fig. 2A and B).
The cutoff values of a mean of 2,587 circulating cell-free nuclear DNA genome equivalent/mL and a mean of 1,335,407 were chosen by ROC curve analysis, which provided a sensitivity of 74% and a specificity of 67% for the circulating cell-free nuclear DNA assay (AUC 0.704, 95% CI 0.535–0.874) and a sensitivity of 79% and a specificity of 67% for the circulating cell-free mitochondrial DNA assay (AUC 0.789, 95% CI 0.65–0.929) to discriminate between epithelial ovarian cancer and benign epithelial ovarian tumors (Fig. 3A and B).
The cutoff value of a mean of 1,335,407 circulating cell-free mitochondrial DNA genome equivalent/mL was chosen by ROC analysis, which provided a sensitivity of 79% and a specificity of 63% (AUC 0.717, 95% CI 0.553–0.882) to discriminate between epithelial ovarian cancer and endometriosis.
We compared the plasma circulating cell-free DNA levels between the borderline tumor group (nequals;8) and the invasive cancer group (nequals;13). The levels of plasma circulating cell-free DNA were higher overall in the invasive cancer subgroup than in the borderline tumor subgroup, but no statistically significant difference was found between the two groups (Mann-Whitney test, P=.176 and 0.9 for circulating cell-free nuclear DNA and circulating cell-free mitochondrial DNA, respectively).
We also did not find any significance in the circulating cell-free DNA levels between the following subgroups: CA 125 <35 U/mL (nequals;8) compared with CA 125 >35 U/mL (nequals;9) (Mann-Whitney test, P=.48 and .906 for circulating cell-free nuclear DNA and circulating cell-free mitochondrial DNA, respectively), tumor volume less than 500 mL (nequals;5) compared with tumor volume greater than 500 mL (nequals;5) (Mann-Whitney test, P=.456 for both), and early stage (nequals;4) compared with advanced stage (nequals;9) (Mann-Whitney test, P=.643 and .857). No further subgroup analyses were performed because of the small sample size.
We determined both circulating cell-free nuclear DNA and circulating cell-free mitochondrial DNA in paired plasma and serum samples from patients with epithelial ovarian cancer, benign epithelial ovarian tumors, and endometriosis as well as from healthy individuals by real-time PCR, which is considered a gold standard for genetic analysis with high sensitivity and specificity. Elevated levels of both circulating cell-free nuclear DNA and circulating cell-free mitochondrial DNA were found in the plasma samples from patients with epithelial ovarian cancer when compared with the plasma samples from healthy individuals and from patients with benign ovarian tumors. The possible diagnostic value of using the plasma circulating cell-free DNA as markers was evaluated by ROC curve analysis. A sensitivity of 63–79% and a specificity of 62–69% when using the circulating cell-free DNA assay was achieved to discriminate between participants with epithelial ovarian cancer and healthy individuals, as well as between those with epithelial ovarian cancer and benign epithelial ovarian tumors. A significant difference in circulating cell-free mitochondrial DNA, but not in circulating cell-free nuclear DNA, was found between the endometriosis group and the epithelial ovarian cancer group. A possible diagnostic value was given by ROC curve, showing a sensitivity of 79% and a specificity of 63%. The results show importance and the necessity of testing two markers simultaneously.
No correlations between the quantities of plasma circulating cell-free DNA and the traditional predictive markers, such as tumor stage, tumor volume, and CA 125 detection, were found, suggesting that plasma circulating cell-free nuclear DNA and circulating cell-free mitochondrial DNA may be independent markers in the circulation of patients with ovarian cancer.
However, although a clear trend is visible in the circulating cell-free DNA levels between the invasive cancer subgroup and the borderline tumor subgroup, no statistically significant difference was found between the two groups. That may be because of the limited sample size. Further research into better understanding the clinical relevance on a large-scale sample size is warranted.
Circulating cell-free DNA is overall dramatically higher in serum samples than in plasma samples. Similar results have been reported.14,23,24 Release of cellular DNA during blood clotting caused higher levels of DNA molecules in serum samples. The high levels of cellular DNA in the serum samples could be interpreted as artifact without diagnostic relevance. Our study confirmed the interpretation that although elevated levels of plasma circulating cell-free DNA were observed in ovarian cancer and showed possible diagnostic values to discriminate between epithelial ovarian cancer cases and healthy individuals, the levels of serum circulating cell-free DNA reveal the results without clinical relevance. It also can be supported by our previous studies of circulating cell-free DNA in breast cancer that although malignant and benign breast tumors could be distinguished using plasma circulating cell-free DNA as a marker, the patients from the malignant and benign groups showed comparable levels of circulating cell-free DNA in serum samples.14,23 Similar phenomena also have been observed by other studies. Although Mehra et al (2007) showed an elevated plasma mitochondrial DNA level in patients with advanced prostate cancer,12 Elliger et al (2008) could not find any significance between benign and malignant prostate lesions in terms of serum circulating cell-free mitochondrial DNA.13 However, levels of serum circulating cell-free DNA can be used to examine cell fragilities during clotting procedures.24
The origin of circulating cell-free DNA is cell turnover. In our study, both circulating cell-free nuclear DNA and circulating cell-free mitochondrial DNA in plasma samples are increased in epithelial ovarian cancer, with probable diagnostic value. However, no correlation between circulating cell-free nuclear DNA and circulating cell-free mitochondrial DNA quantities was found in our study, implying that both circulating cell-free nuclear DNA and circulating cell-free mitochondrial DNA might be released from different parts by different mechanisms.
In conclusion, we have shown that levels of circulating cell-free mitochondrial and nuclear DNA in plasma are significantly elevated in patients with epithelial ovarian cancer, with probable diagnostic value provided by ROC curves, compared with a healthy control group and patients with benign ovarian lesions. Circulating cell-free DNA in serum samples seems to be cellular DNA artifacts predominated during sample preparations, which limits the use of the species in a diagnostic aspect. No correlation between circulating cell-free nuclear DNA and circulating cell-free mitochondrial DNA was found, suggesting a different release mechanism of nuclear genomes (nuclear DNA) and cytoplasmic genomes (mitochondrial DNA) in the events of cell turnover in circulation. From our study, we can conclude that both circulating cell-free nuclear DNA and circulating cell-free mitochondrial DNA in plasma can be potentially useful biomarkers in ovarian cancer.
1. Jemal A, Siegel R, Ward E, Hao Y, Xu J, Murray T, et al. Cancer statistics, 2008. CA Cancer J Clin 2008;58:71–96.
2. Jacobs IJ, Menon U. Progress and challenges in screening for early detection of ovarian cancer. Mol Cell Proteomics 2004;3:355–66.
3. Malkasian GD Jr, Knapp RC, Lavin PT, Zurawski VR Jr, Podratz KC, Stanhope CR, et al. Preoperative evaluation of serum CA 125 levels in premenopausal and postmenopausal patients with pelvic masses: discrimination of benign from malignant disease. Am J Obstet Gynecol 1988;159:341–6.
4. Benjapibal M, Neungton C. Pre-operative prediction of serum CA125 level in women with ovarian masses. J Med Assoc Thai 2007;90:1986–91.
5. Rosenthal AN, Menon U, Jacobs IJ. Screening for ovarian cancer. Clin Obstet Gynecol 2006;49:433–47.
6. Sozzi G, Conte D, Leon M, Ciricione R, Roz L, Ratcliffe C, et al. Quantification of free circulating DNA as a diagnostic marker in lung cancer. J Clin Oncol 2003;21:3902–8.
7. Allen D, Butt A, Cahill D, Wheeler M, Popert R, Swaminathan R. Role of cell-free plasma DNA as a diagnostic marker for prostate cancer. Ann N Y Acad Sci 2004;1022:76–80.
8. Taback B, O’Day SJ, Hoon DS. Quantification of circulating DNA in the plasma and serum of cancer patients. Ann N Y Acad Sci 2004;1022:17–24.
9. Gormally E, Hainaut P, Caboux E, Airoldi L, Autrup H, Malaveille C, et al. Amount of DNA in plasma and cancer risk: a prospective study. Int J Cancer 2004;111:746–9.
10. Zhong S, Ng MC, Lo YM, Chan JC, Johnson PJ. Presence of mitochondrial tRNA(Leu(UUR)) A to G 3243 mutation in DNA extracted from serum and plasma of patients with type 2 diabetes mellitus. J Clin Pathol 2000;53:466–9.
11. Chiu RW, Chan LY, Lam NY, Tsui NB, Ng EK, Rainer TH, et al. Quantitative analysis of circulating mitochondrial DNA in plasma. Clin Chem 2003;49:719–26.
12. Mehra N, Penning M, Maas J, van Daal N, Giles RH, Voest EE. Circulating mitochondrial nucleic acids have prognostic value for survival in patients with advanced prostate cancer. Clin Cancer Res 2007;13:421–6.
13. Ellinger J, Muller SC, Wernert N, von Ruecker A, Bastian PJ. Mitochondrial DNA in serum of patients with prostate cancer: a predictor of biochemical recurrence after prostatectomy. BJU Int 2008; [Epub ahead of print].
14. Zhong XY, Ladewig A, Schmid S, Wight E, Hahn S, Holzgreve W. Elevated level of cell-free plasma DNA is associated with breast cancer. Arch Gynecol Obstet 2007;276:327–31.
15. Heid CA, Stevens J, Livak KJ, Williams PM. Real time quantitative PCR. Genome Res 1996;6:986–94.
16. Luthra R, McBride JA, Cabanillas F, Sarris A. Novel 5′ exonuclease-based real-time PCR assay for the detection of t(14;18)(q32;q21) in patients with follicular lymphoma. Am J Pathol 1998;153:63–8.
17. Lo YM, Tein MS, Lau TK, Haines CJ, Leung TN, Poon PM, et al. Quantitative analysis of fetal DNA in maternal plasma and serum: implications for noninvasive prenatal diagnosis. Am J Hum Genet 1998;62:768–75.
18. Zhong XY, Laivuori H, Livingston JC, Ylikorkala O, Sibai BM, Holzgreve W, et al. Elevation of both maternal and fetal extracellular circulating deoxyribonucleic acid concentrations in the plasma of pregnant women with preeclampsia. Am J Obstet Gynecol 2001;184:414–9.
19. Zhong XY, Burk MR, Troeger C, Jackson LR, Holzgreve W, Hahn S. Fetal DNA in maternal plasma is elevated in pregnancies with aneuploid fetuses. Prenat Diagn 2000;20:795–8.
20. Lo YM, Zhang J, Leung TN, Lau TK, Chang AM, Hjelm NM. Rapid clearance of fetal DNA from maternal plasma. Am J Hum Genet 1999;64:218–24.
21. Saiki RK, Gelfand DH, Stoffel S, Scharf SJ, Higuchi R, Horn GT, et al. Primer-directed enzymatic amplification of DNA with a thermostable DNA polymerase. Science 1988;239:487–91.
22. Clark RD, Webster-Clark DJ. Managing bias in ROC curves. J Comput Aided Mol Des 2008;22:141–6.
23. Zanetti-Dallenbach RA, Schmid S, Wight E, Holzgreve W, Ladewing A, Hahn S, et al. Levels of circulating cell-free serum DNA in benign and malignant breast lesions. Int J Biol Markers 2007;22:95–9.
© 2008 The American College of Obstetricians and Gynecologists
24. Zhong XY, Hahn S, Kiefer V, Holzgreve W. Is the quantity of circulatory cell-free DNA in human plasma and serum samples associated with gender, age and frequency of blood donations? Ann Hematol 2007;86:139–43.