|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
Journal of Clinical Oncology, Vol 22, No 20 (October 15), 2004: pp. 4157-4164 © 2004 American Society of Clinical Oncology. DOI: 10.1200/JCO.2004.11.123 Circulating Deoxyribonucleic Acid As Prognostic Marker in Non–Small-Cell Lung Cancer Patients Undergoing ChemotherapyFrom the Institute of Medical Oncology, University Hospital; and Division of Epidemiology and Biostatistics, Department of Social and Preventive Medicine, University of Berne; and Clinic and Policlinic of Oncology, and Division of Thoracic Surgery, Zurich University Hospital, Zurich, Switzerland Address reprint requests to A. Ziegler, PhD, Zurich University Hospital, Clinic and Policlinic of Oncology, Laboratory of Molecular Oncology, Haeldeliweg 4, 8044 Zurich, Switzerland; e-mail: annemarie.ziegler{at}usz.ch
PURPOSE: Circulating cell-free DNA is present in increased amounts in the blood of cancer patients, but the clinical relevance of this phenomenon remains unclear. We conducted a clinical study to assess the value of circulating DNA as a prognostic marker in patients with non–small-cell lung cancer (NSCLC). PATIENTS AND METHODS: A standard protocol for the quantification of circulating DNA by real-time polymerase chain reaction was set up and validated at two oncology units. One hundred eighty-five informed patients with NSCLC and 46 healthy controls were included in the study. DNA concentrations were determined in paired plasma and serum samples and analyzed for a relationship with leukocyte counts and lactate dehydrogenase (LDH) levels. DNA concentrations in healthy controls and in patients were compared, and cutoff levels for plasma and serum DNA were determined. Patient survival was analyzed relative to baseline DNA concentrations, and the relationship between tumor responses and changes in DNA concentrations was assessed in patients receiving chemotherapy. RESULTS: We found a significant correlation between increased plasma DNA concentrations and elevated LDH levels (P = .009), advanced tumor stage (P < .003), and poor survival (P < .001). Tumor progression after chemotherapy was significantly (P = .006) associated with increasing plasma DNA concentrations. Serum DNA concentrations strongly correlated (P < .001) with leukocyte counts. CONCLUSION: Our data demonstrate that quantification of plasma DNA is an accurate technique amenable to standardization, which might complement current methods for the prediction of patient survival. This approach might be considered for evaluation in large prospective studies.
Lung cancer is the leading cause of cancer death in the world, and its incidence is still rising in women.1,2 Determination of the prognosis for an individual patient with lung cancer remains difficult, despite the existence of several established prognostic indicators.3,4 Although numerous molecular and biologic markers are currently under investigation in lung cancer tissue,5 the availability of readily accessible tumor markers from the blood of patients might facilitate clinical decision making at the time of diagnosis and in the later course of the disease. The finding that tumors are capable of shedding nucleic acids into the bloodstream has opened new areas in translational cancer research (reviewed in6,7). Although considerably degraded, DNA can be recovered from a patient's serum or plasma and used as a surrogate source of tumor DNA. Accordingly, the number of studies evaluating the potential use of serum or plasma DNA in cancer diagnosis and prognosis has increased steadily in the past decade. In the case of non–small-cell lung cancer (NSCLC), both quantitative8-10 and qualitative11-18 studies suggest potential applications in disease management. However, translation of these findings into the clinic is a matter of current debate because of the difficulty of comparing and normalizing the existing data. This stems mainly from the lack of technical standardization and the relatively small collectives examined in several studies. In addition, NSCLC and small-cell lung cancer (SCLC) have often been evaluated together. The current study was designed to address several unresolved issues that are essential for the broad application of plasma and/or serum DNA methodology. Our goals were as follows: (1) to establish an accurate and reproducible technique to allow the quantification of circulating DNA in two independent laboratories; (2) to evaluate, in the context of a multicenter study, a sufficiently large number of NSCLC patients to attain statistically significant data; (3) to compare the prognostic utility of serum versus plasma DNA quantification in NSCLC patients undergoing chemotherapy; and (4) to analyze DNA levels for associations with clinical parameters and patient survival. We demonstrate that serum and plasma DNA are not equivalent in terms of their prognostic potential, although both were significantly associated with patient survival. Our data suggest that quantification of circulating DNA can be standardized and has the potential to complement current prognostic procedures for the management of NSCLC.
Patient Enrollment and Study Protocol A total of 185 patients were recruited between April 2001 and January 2003 at the Medical Oncology Units of the University Hospitals of Zurich and Berne (Switzerland). The study was approved by the corresponding ethics committees, and written informed consent was obtained from all patients. Eligibility criterion was the presence of a newly diagnosed or relapsed, histologically confirmed NSCLC, whereas patients with severe infection, an Eastern Cooperative Oncology Group performance score more than 2, active comorbidity, or an expected survival time of less than 3 months were excluded. Baseline blood samples were drawn before the initiation of a new treatment modality, and for patients undergoing chemotherapy, a second sample was obtained at the time of medical evaluation 1 to 3 months after the start of treatment. Survival data was obtained until May 2003 and was available for 163 patients. One hundred twenty-seven patients underwent chemotherapy, and for 91 of the patients, follow-up data consisting of a blood drawing within 3 weeks of evaluation by chest radiograph or computed tomography scan was available. Blinded assessment of tumor response was performed by radiologists according to WHO criteria.19 Leukocyte counts and lactate dehydrogenase (LDH) levels obtained the same day as the study blood samples were available for 119 and 78 patients, respectively. Control blood samples from 46 healthy individuals were obtained from the Swiss Red Cross (Berne, Switzerland) and from volunteers.
DNA Purification
Quantitative Real-Time PCR
Statistical Analysis
Quantification of DNA Because laboratory procedures were to be performed independently in two centers, we addressed various sources likely to introduce experimental variability (data not shown). The brand of tubes used for blood collection (Vacutainer; BD, Plymouth, UK; and S-Monovette; Sarstedt, Sevelen, Switzerland) did not influence DNA concentrations obtained from serum and plasma samples (variation range, 0% to 7%; median, 4%). Furthermore, DNA concentrations could be reproducibly determined in both centers as established by the exchange of blinded samples (variation range, 0% to 55%; median, 17%) and of whole prepipetted Taqman plates (variation range, 0% to 551%; median, 25%). Higher variation occasionally occurred near the detection limit of the technique but corresponded to minute absolute amounts of DNA. Comparable results were obtained for cell-line DNAs that were quantitated by spectrophotometry and by real-time PCR (variation range, 8% to 48%; median, 20%). In this case, variability is also influenced by the sensitivity of the spectrophotometer. The interassay variability range for repeat experiments was 0% to 192%, with a median variation of 24%. Finally, we modified the DNA purification protocol provided by the manufacturer of the QIAamp Blood Mini kit (see Patients and Methods) to remove inhibitors of the PCR reaction, which were evidenced in dilution experiments under insufficient washing conditions. We concluded that this methodology was suitable for reproducible quantification of serum and plasma DNA in independent centers.
Patient Baseline Characteristics
Circulating DNA Levels in Healthy Individuals and in Patients With NSCLC Baseline DNA concentrations were measured in paired serum and plasma from 173 of 185 patients. In agreement with published data,22-24 the mean DNA concentration was 10-fold higher in serum compared with paired plasma (Table 1, Fig 1). The same trend was observed among healthy controls (median serum DNA concentration, 12.6 ng/mL; median plasma DNA concentration, 1.8 ng/mL). Among the latter group, 45 (98%) of 46 healthy individuals presented DNA concentrations less than 50 ng/mL in serum and less than 10 ng/mL in plasma. Comparable ranges have been obtained using other accurate techniques for DNA quantification.24,25 Therefore, we defined these values as cutoff levels to differentiate between normal and elevated DNA. As determined by the Mann-Whitney rank sum test, serum and plasma DNA concentrations were significantly higher (P < .001 for each) in NSCLC patients than in healthy controls. In serum, this difference was already apparent at early disease stages. In plasma, Cox proportional hazards regression and the Kruskal-Wallis test revealed a significant trend (P = .003) towards higher DNA concentrations at advanced tumor stages. The trend was less obvious in serum (P = .028), mainly because of the notoriously high DNA levels in early-stage samples (Fig 1).
Relationship of Circulating DNA to Leukocyte Counts and LDH Levels Existing evidence suggests that serum DNA levels are higher than those in paired plasma samples because of in vitro lysis of blood cells during the clotting process.22,23 To examine this possibility, available leukocyte counts from NSCLC patients were plotted against paired baseline DNA concentrations (Fig 2). In support of this notion, we found a strong correlation between leukocyte counts and paired DNA concentrations in serum (Spearman rank correlation test, r = 0.529; P < .001) but not in plasma (r = 0.193, P = .036) samples. We cannot conclude the precise identity of cells undergoing lysis because leukocytes comprise distinct cell types. However, when serum DNA concentrations were plotted against either lymphocyte or neutrophil subpopulations, the association persisted for neutrophils but not for lymphocytes (data not shown). Because elevated serum LDH is a prognostic factor in advanced NSCLC,19 DNA concentrations were further plotted against available paired LDH levels (Fig 3). A highly significant correlation was found between plasma DNA concentrations and paired LDH levels (r = 0.261, P = .009) but not between serum DNA concentrations and paired LDH levels (r = 0.126, P = .273). Elevated serum and plasma DNA levels were not associated with any particular histologic subtype (data not shown).
Circulating DNA and Survival According to their baseline DNA concentration, patients were assigned to a group with either high or low levels of circulating DNA using the cutoff values described earlier. As determined by the Kaplan-Meier method and log-rank test, high DNA concentrations at baseline were significantly associated with decreased survival (Fig 4). The impact on survival was greater for elevated plasma DNA levels (P < .001) than for elevated serum DNA (P = .046). Similarly, survival analysis using continuous values showed significance for plasma DNA (Cox multivariate, P < .001) but not for serum DNA (Cox multivariate, P = .669). Multivariate Cox analysis for survival showed that age  70 years, a tumor stage of IV, elevated serum DNA concentration, and elevated plasma DNA concentration were significant independent prognostic parameters (Table 2). Because of limited sample size (n = 73), LDH was analyzed using univariate analysis, which showed that elevated (> 480 U/L) LDH also was significantly (P = .008) associated with decreased survival.
Monitoring of Circulating DNA During the Course of Chemotherapy After completion of one to three cycles of chemotherapy, clinical follow-up evaluation was performed as described in Patients and Methods. According to tumor response, patients were either assigned to a group with stable disease or remission or to a group with progression. DNA concentrations measured at baseline and at follow-up were compared using the Wilcoxon matched-pairs signed rank test (Fig 5). For patients amenable to analysis, the group with stable disease or remission showed a significant (P = .002) decrease in serum DNA concentrations compared with baseline values, whereas plasma DNA concentrations remained unchanged. Patients with tumor progression showed a significant (P = .006) increase in plasma DNA concentrations but no change in serum DNA concentrations.
The possibility of recovering tumor-derived DNA from the patient's blood has provoked great expectation. This method offers a noninvasive means to obtain tumor surrogate material, which could represent a unique source for diagnostic and prognostic applications. In lung cancer patients, both quantitative8-10 and qualitative11-18 studies have addressed the utility of serum and plasma DNA for disease management. In terms of DNA quantification, the existing studies are marked by the diversity of the techniques used, which prevents the comparison of data and the estimation of the real prognostic potential of this approach. Also, the numbers of patients analyzed have been rather small, and one study consisted of a mixed population of NSCLC and SCLC patients.8 Despite this, some congruent results have been obtained. All studies found significantly elevated DNA levels in the plasma of lung cancer patients compared with healthy controls, although no explicit cutoff values were defined in any of these publications. Furthermore, no correlation was found between plasma DNA levels and several histopathologic features. In contrast, disparities exist with regard to parameters such as clinical staging. In the first work by Fournié et al,8 plasma DNA was highest in patients with stage IV disease, whereas two later studies by Sozzi et al9,10 found no such association. An association with age was evidenced in the second, but not in the first, study by Sozzi et al. Important differences concern the predictive value for patient survival. No correlation was found between plasma DNA levels and relapse-free or overall survival by one study,9 whereas an association of plasma DNA with survival, LDH, and neuron-specific enolase could be established for a mixed group of SCLC and NSCLC patients.8 These data provide an example of the current standing of plasma and serum DNA research and of the need for larger controlled studies with standardized procedures. In the present work, we used real-time PCR for DNA quantification, which can be regarded as the standard method currently available.26 We also addressed the issue of reproducibility by performing the complete methodology in two independent clinical centers. Adhering to an internal protocol with periodical cross-validation tests, we showed that the method is feasible, accurate, and suitable for routine clinical application. To our knowledge, this constitutes the first attempt to standardize techniques for serum and plasma DNA analyses. Because this is the quantitative study with the largest group of NSCLC patients, our findings also attained statistical significance. Despite the association of plasma DNA with disease stage and LDH levels, the follow-up of patients for up to 2 years revealed that values above normal are significant and independent predictors of poor survival. This is in agreement with findings by Fournié et al,8 although in their study, the cutoff was high (100 ng/mL), and all cases above this level corresponded to stage IV disease. This suggests that survival might indirectly reflect disease stage rather than DNA levels in their evaluation. In the case of serum, we found a weaker association with patient survival, and serum DNA levels did not increase with advancing disease stage but were rather related to the patient's leukocyte counts. A similar observation was made in patients with leukemia.27 Our data support the current notion that excess DNA recovered from serum is liberated in vitro during the clotting process.22,23 The nature of cells undergoing lysis remains unclear. Male WBCs underwent some degree of lysis when spiked into female blood,22 and the release of nucleosomes from apoptotic lymphocytes has been demonstrated in cell culture.28 However, our results reveal a correlation between serum DNA levels and total neutrophil counts but not lymphocyte counts. Furthermore, the clotting process is rather rapid compared with the progressive lysis of lymphocytes observed in culture,28 suggesting that the release of DNA into serum might involve additional events, perhaps of a mechanical nature. It is still not possible to draw a conclusion on this issue. However, our work and the work of others suggest that plasma DNA is better suited for quantitative studies because it is more representative of tumor status and is enriched for tumor DNA. Regarding the predictive value of DNA concentrations for patient's response to treatment, we again found differences between serum and plasma. Furthermore, the existing data are rather discordant. For progressive disease, we found significant increases in plasma DNA levels but not in serum DNA levels. Increased median plasma DNA concentrations have also been noted in five patients with postoperative relapse.10 In contrast, serum DNA showed a trend to increase in a small group of lung cancer patients with progressive disease after radiation therapy.25 Because both previous studies used reliable techniques for DNA quantification, the differences observed are probably not related to the methodology in this case. For patients showing a response or stable disease, we found significant decreases in serum DNA but not plasma DNA, but median plasma DNA levels were reduced in patients with no postoperative relapse in a recent study.10 In our opinion, changes in serum and plasma DNA are not accurate or reliable enough for response prediction, despite the fact that we and others have attained statistically significant results. Decreases in DNA levels can occur despite disease progression. Furthermore, in our study, plasma DNA levels increased but remained below the cutoff value in most patients with progression. In clinical routine, any reduction or increase within a normal DNA range would be hard to interpret. In the work by Sozzi et al,10 median plasma DNA concentrations were significantly higher in patients with relapse, but decreases were also frequent in this group. Therefore, neither the reduction in DNA levels nor the setting of a cutoff value to separate responders from nonresponders appear to show the necessary attributes to compete with current methods to assess clinical response. Our data suggest that the quantification of circulating DNA is a noninvasive technique amenable to standardization. In NSCLC patients, elevated plasma DNA concentration is a prognostic factor and indicates advanced disease stage, decreased survival, and tumor progression under therapy. The quantification of plasma DNA might provide a means to discriminate individuals with more aggressive disease, which could aid in the choice of appropriate therapies. The existing data suggest that the prognostic utility of plasma DNA should be assessed in large prospective trials. The same effort should be directed towards the analysis of tumor-specific genetic alterations in circulating DNA, to address the issues of early disease detection, response prediction, and the choice of personalized therapies.
Authors' Disclosures of Potential Conflicts of Interest
We are grateful to the nursing staffs of the Medical Oncology facilities for the collection of blood samples and to U.E. Nydegger, the Central Laboratory of the Swiss Red Cross, and volunteers for the control blood samples.
Supported in part by grants from the Swiss Cancer League (KFS-703-8-1998), the Bernese Cancer League, the Cancer League of Zurich, the Terry Fox Foundation, and by Eli Lilly (Suisse) S.A. Authors' disclosures of potential conflicts of interest are found at the end of this article.
1. Alberg AJ, Samet JM: Epidemiology of lung cancer. Chest 123: 21-49, 2003 (suppl 1) 2. Tyczynski JE, Bray F, Parkin DM: Lung cancer in Europe in 2000: Epidemiology, prevention, and early detection. Lancet Oncol 4: 45-55, 2003[CrossRef][Medline]
3. Brundage MD, Davies D, Mackillop WJ: Prognostic factors in non-small cell lung cancer: A decade of progress. Chest 122: 1037-1057, 2002 4. Feld R, Borges M, Giner V, et al: Prognostic factors in non-small cell lung cancer. Lung Cancer 11: 19-23, 1994 (suppl 3)[Medline] 5. Iyengar P, Tsao MS: Clinical relevance of molecular markers in lung cancer. Surg Oncol 11: 167-179, 2002[CrossRef][Medline] 6. Anker P, Mulcahy H, Stroun M: Circulating nucleic acids in plasma and serum as a noninvasive investigation for cancer: Time for large-scale clinical studies? Int J Cancer 103: 149-152, 2003[CrossRef][Medline] 7. Ziegler A, Zangemeister-Wittke U, Stahel RA: Circulating DNA: A new diagnostic gold mine? Cancer Treat Rev 28: 255-271, 2002[CrossRef][Medline] 8. Fournié GJ, Courtin JP, Laval F, et al: Plasma DNA as a marker of cancerous cell death: Investigations in patients suffering from lung cancer and in nude mice bearing human tumours. Cancer Lett 91: 221-227, 1995[CrossRef][Medline]
9. Sozzi G, Conte D, Mariani L, et al: Analysis of circulating tumor DNA in plasma at diagnosis and during follow-up of lung cancer patients. Cancer Res 61: 4675-4678, 2001
10. Sozzi G, Conte D, Leon ME, et al: Quantification of free circulating DNA as a diagnostic marker in lung cancer. J Clin Oncol 21: 3902-3908, 2003
11. Esteller M, Sanchez-Cespedes M, Rosell R, et al: Detection of aberrant promoter hypermethylation of tumor suppressor genes in serum DNA from non-small cell lung cancer patients. Cancer Res 59: 67-70, 1999
12. Sozzi G, Musso K, Ratcliffe C, et al: Detection of microsatellite alterations in plasma DNA of non-small cell lung cancer patients: A prospect for early diagnosis. Clin Cancer Res 5: 2689-2692, 1999
13. Bruhn N, Beinert T, Oehm C, et al: Detection of microsatellite alterations in the DNA isolated from tumor cells and from plasma DNA of patients with lung cancer. Ann NY Acad Sci 906: 72-82, 2000
14. Bearzatto A, Conte D, Frattini M, et al: P16(INK4A) hypermethylation detected by fluorescent methylation-specific PCR in plasmas from non-small cell lung cancer. Clin Cancer Res 8: 3782-3787, 2002 15. An Q, Liu Y, Gao Y, et al: Detection of hypermethylation in circulating plasma DNA of non-small cell lung cancer patients. Cancer Lett 188: 109-114, 2002[CrossRef][Medline]
16. Usadel H, Brabender J, Danenberg KD, et al: Quantitative adenomatous polyposis coli promoter methylation analysis in tumor tissue, serum and plasma DNA of patients with lung cancer. Cancer Res 62: 371-375, 2002 17. Ramirez JL, Sarries C, de Castro PL, et al: Methylation patterns and K-ras mutations in tumor an paired serum of resected non-small-cell lung cancer patients. Cancer Lett 193: 207-216, 2003[CrossRef][Medline] 18. Beau-Faller M, Gaub MP, Schneider A, et al: Plasma DNA microsatellite panel as sensitive and tumor-specific marker in lung cancer patients. Int J Cancer 105: 361-370, 2003[CrossRef][Medline] 19. Miller AB, Hoogstraten B, Staquet M, et al: Reporting results in cancer treatment. Cancer 47: 207-214, 1981[CrossRef][Medline]
20. Donovan JW, Ladetto M, Zou G, et al: Immunoglobulin heavy-chain consensus probes for real-time PCR quantification of residual disease in acute lymphoblast leukemia. Blood 95: 2651-2658, 2000 21. Lo YM, Tein MS, Lau TK, et al: Quantitative analysis of fetal DNA in maternal plasma and serum: Implications for noninvasive prenatal diagnosis. Am J Hum Genet 1998 62: 768-775, 1998[CrossRef][Medline] 22. Lee TH, Montalvo L, Chrebtow V, et al: Quantitation of genomic DNA in plasma and serum samples: Higher concentrations of genomic DNA found in serum than in plasma. Transfusion 41: 276-282, 2001[CrossRef][Medline]
23. Lui YYN, Chik K-W, Chiu RWK, et al: Predominant hematopoietic origin of cell-free DNA in plasma and serum after sex-mismatched bone marrow transplantation. Clin Chem 48: 421-427, 2002 24. Thijssen MAMA, Swinkels DW, Ruers TJM, et al: Difference between free circulating plasma and serum DNA in patients with colorectal liver metastases. Anticancer Res 22: 421-426, 2002[Medline]
25. Leon SA, Shapiro B, Sklaroff DM, et al: Free DNA in the serum of cancer patients and the effect of therapy. Cancer Res 37: 646-650, 1977
26. De Kok JB, Hendriks JC, van Solinge WW, et al: Use of real-time quantitative PCR to compare DNA isolation methods. Clin Chem 44: 2201-2204, 1998 27. Carpentier NA, Izui S, Rose LM, et al: The presence of circulating deoxyribonucleic acid (DNA) in patients with acute or chronic leukaemia: Relation to serum anti-DNA antibodies and Clq binding activity. Human Lymphocyte Diff 1: 93-104, 1981 28. Emlen W, Niebur JA, Kadera R: Accelerated in vitro apoptosis of lymphocytes from patients with systemic lupus erythematosus. J Immunol 152: 3685-3692, 1994[Abstract] Submitted November 20, 2003; accepted June 30, 2004. This article has been cited by other articles:
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||
|
|||||||||||
|
|
Copyright © 2004 by the American Society of Clinical Oncology, Online ISSN: 1527-7755. Print ISSN: 0732-183X
|
TOP
ABSTRACT
INTRODUCTION
.05 was considered statistically significant. 
