Abstract
Purpose
Most patients suffering from advanced lung cancer die within a few months. To exploit new therapy regimens we need better methods for the assessment of a therapy response.
Material and Methods
In a pilot study we prospectively enrolled 36 patients with advanced NSCLC and SCLC (34 stage IV, 2 stage IIIB) of whom 34 received standard platinum-based chemo/radiotherapy and two were treated with a tyrosine kinase inhibitor. We measured the levels of extracellular methylated SHOX2 DNA (mSHOX2) in plasma before and during therapy until re-staging. The mSHOX2 analysis was blinded with respect to the clinical data making it an observational study.
Results
According to the re-staging of 31 first-line patients, 19 patients were classified as non-responders while 12 patients were in the responder group. We observed a tight correlation between radiological data and the change of plasma mSHOX2 level as the equivalent for a therapy response. A ROC analysis showed a high discriminatory power for both patient groups already one week after therapy start (AUC 0.844). Additionally, a Kaplan-Meier and Cox Proportional Hazards analyses revealed a strong relationship between survival and plasma mSHOX2 value p≤0.001 (hazard ratio 11.08) providing some evidence for mSHOX2 also being a predictive marker.
Conclusion
The longitudinal measurement of extracellular plasma mSHOX2 DNA yields information about the response to cytotoxic treatment and allows an early assessment of treatment response for lung cancer patients. If confirmed in a larger study this would be a valuable tool for selecting and guiding a cytotoxic treatment.
Introduction
Lung cancer is still a health problem and in 2012 there were more than 409,000 new lung cancer cases in Europe [1]. The five-year survival rates for lung cancer at all stages is 16% and only slightly better than it was 30 years ago [2]. In recent years several new therapy regimens were introduced including a variety of different multimodal treatments for patients with locally advanced, late stage and metastatic disease [3]. Advances in the systemic therapies not only lead to an improved survival but also to a reduction of cancer-related symptoms and a higher quality of life [4]. Nevertheless, the therapeutic window is still small, and it is important to have a method for an early response evaluation to choose the optimal therapy. The method of choice for the assessment of treatment response is a re-staging after two to four cycles of systemic therapy (i.e. after 6 to 12 weeks) using an imaging technique like CT, MRI, or PET. Apart from the high costs, these techniques are not very sensitive [5][6]. An alternative would be the use of biomarkers like CYFRA-21, SCCA, CEA and CA-125 for NSCLC patients and ProGRP and NSE for SCLC patients to correlate them with therapy response [7]. Unfortunately there is no universal marker that useful for all different lung cancer histologies and there is not enough evidence for any of them to be routinely used in the clinic.
Mandel and Metais were the first to describe their observation of the presence of extracellular nucleic acids in humans [8]. Tumor-associated genetic alterations can be found in cell-free nucleic acids isolated from all different body fluids [9,10][11]. According to our current knowledge all tumor-associated alterations found in tumor cells can also be detected in extracellular nucleic acids, including epigenetic alterations associated with the development of malignant tumors. DNA methylation and cytosine methylation are a hallmark of mammalian chromatin, play a role in the regulation of development and are important in basic biological processes like embryogenesis and cell differentiation [12] [13]. As such, DNA methylation regulates gene transcription and epigenetic alterations in oncogenes and tumor suppressor genes and are of key importance to cancer development [14]. Recently, the methylation of the SHOX2 gene (mSHOX2) has been described as a novel and powerful marker for an early detection of patients with lung cancer based on the analysis of bronchial aspirates and plasma [15] [16] [17], the evaluation of paramalignant and malignant pleural effusions [18], the examination of needle aspirates for lung cancer staging [19] and as a predictor for outcome in NSCLC patients [20]. This study was performed to evaluate i) whether the quantitative analysis of mSHOX2 plasma DNA correlates with treatment response in lung cancer patients and ii) to determine the best time for performing the analysis of this biomarker.
Material and Methods
Patients
We prospectively enrolled 36 patients which were consecutively referred to our outpatient clinic for diagnosis and treatment of lung cancer. We included patients with a late stage/advanced histologically proven lung tumor (independent of the typ of lung cancer) who were eligible for a chemo/radio-chemotherapy and had signed a written consent to participate in this study. When the clinical data wase combined with the mSHOX2 measurements we realized that five patients had received a treatment before enrollment in our study. All other 31 patients received a first-line therapy. The details of the clinical data of all patients are summarized in Tables 1–3. The specimens for the histopathological diagnosis were obtained by bronchoscopy and/or computed tomography (CT). All but one patient received a standard platinum-based combination chemotherapy and if necessary an additional radiotherapy according to existing guidelines. [21]. As part of the diagnostic workup all lung cancer patients are screened for EGFR mutations. Patients UKH10 and UKH 031 demonstrated an activating EGFR mutation and were treated with Erlotinib. After three therapy cycles the patients were re-staged by physicians of the local tumor board based on repeat-CT. The response evaluation and the assignment of the patient as responders and non-responders, respectively were carried out according to RECIST v1.1 criteria. The study has been approved by the Institutional Review Board (IRB) at the University Hospital of Halle/Saale. Informed consent (written) was obtained from all donors.
Table 1. Clinical data of patients not responding to the therapy.
Patient ID | Smoking status / pack years | Histology | Stage (UICC/AJCC) |
---|---|---|---|
UKH 003 | never | Adenocarcinoma | IV |
UKH 005 | ex-smoker / 35 | SCLC | IV |
UKH 009 | ex-smoker / 22 | SCLC | IV |
UKH 011 | yes / 37 | NSCLC, large cell | IV |
UKH 014 | ex-smoker / 35 | undifferentiated carcinoma | IV |
UKH 016 | ex-smoker / 45 | undifferentiated carcinoma | IV |
UKH 017 | ex-smoker / 6 | Adenocarcinoma | IV |
UKH 018 | ex-smoker /50 | Squamous cell carcinoma | IIIB |
UKH 019 | ex-smoker / 60 | Adenocarcinoma | IV |
UKH 023 | ex-smoker / 36 | undifferentiated carcinoma | IV |
UKH 024 | ex-smoker / 30 | Adenocarcinoma | IV |
UKH 027 | ex-smoker / 24 | SCLC | IV |
UKH 028 | ex-smoker / 50 | Sqamous cell carcinoma | IV |
UKH 033 | yes / 23 | Sqamous cell carcinoma | IV |
UKH 035 | ex-smoker / 27 | Adenocarcinoma | IV |
UKH 036 | yes / 30 | Adenocarcinoma | IV |
UKH 038 | yes / 26 | Adenocarcinoma | IV |
UKH 039 | yes / 20 | Adenocarcinoma | IV |
UKH 041 | yes / 20 | Adenocarcinoma | IV |
Clinical data for first line patients who did not respond to the therapy. Fifteen of the patients were male and four were female. The median age of this patient group is 63 years.
Table 3. Clinical data of second-line patients.
Pat Id | Smoking status / pack years | Histology | Stage (UICC/AJCC) | Responder/ non-responder |
---|---|---|---|---|
UKH 001 | never | Adenocarcinoma | IV | Responder |
UKH 002 | yes / 6 | Squamous cell carcinoma | IV | Responder |
UKH 010 | ex-smoker / 30 | Adenocarcinoma | IV | non-responder |
UKH 020 | ex-smoker / 35 | Squamous cell carcinoma | IV | non-responder |
UKH 037 | yes / 34 | Adenocarcinoma | IV | non-responder |
In this group were four male and one female patients. The only female patient in this group (UKH 001) did respond to therapy. The median age of this patient group is 64 years. Patient UKH 010 demonstrated an EGFR mutation and was treated with TKI.
Preparation of plasma samples
We obtained 2 × 8.5 mL EDTA blood from all patients at the time of diagnosis (pre-therapy = baseline) and every time the patients were checked for their blood counts or when they received a chemotherapy treatment (usually at intervals of 7 to 10 days). The patients were followed until the end of three therapy cycles, i.e. the time of re-staging. The plasma was prepared by spinning the blood samples (within 1 to 2 hrs after blood drawing) for 15 min at 500x g. After careful transfer of the plasma supernatant into a new tube the sample was spun for a second time for 15 min at 2500x g. All samples were stored in 3–4 mL aliquots at -80°C until use.
Real-time quantification of mSHOX2 plasma DNA
Free-circulating DNA from 3.5 mL plasma samples was isolated and bisulfite converted using the Epi proColon Plasma Quick Kit (Epigenomics AG, Berlin, Germany). DNA isolation and bisulfite conversion was carried out following the instruction with minor modifications. The DNA was finally eluted from the beads with 68 μL elution buffer. Together with the patient samples we measured a calibrator sample (i.e. 5 ng artificially methylated bisulfite converted DNA). The sensitive and quantitative qPCR analysis of mSHOX2 was carried out as previously described [16])[18]. Each sample was measured in six PCR replicates and a relative methylation value (= PMR, percent methylation reference) for mSHOX2 was calculated using the adapted ΔΔCT method [16]. The mSHOX2 DNA quantification was performed after all prospectively collected plasma samples were complete, i.e. making this analysis an observational study.
Statistics
Differences of methylation levels (PMR) in blood plasma of reponders and non-responders at base line and follow-up time points 1–8 were tested using unpaired two-sample Wilcox tests (Mann Whitney) given that the PMR data was not normally distributed. The p-values were Bonferroni corrected. Other descriptive data characteristics used were median and median absolute deviation (MAD). Responder Operator Characteristics (ROC) curves were used to visualize the capability of the SHOX2 marker to discriminate between responders and non-responders at different time points. Overall survival was calculated using the Kaplan-Meier and univariate Cox Proportional Hazards regression models. The analyses were carried out with the SPSS 21 software package (IBM, Armonk, NY) and R, respectively [22].
Results
The assignment of the 36 prospectively enrolled patients into responders and non-responders, respectively was completely independent of the mSHOX2 analysis. Thirty one patients received a first-line therapy (Tables 1 and 2) while five patients had been treated before enrollment into the study (Table 3). All but two patients demonstrated a wild-type EGFR gene and received a standard platinum-based chemotherapy, while the two patients with an activating EGFR mutation were treated with TKI. Seven of these 31 patients demonstrated a baseline PMR value ≤ 1% which is assumed to be the level of technical/biological variance. The clinical data from the 31 patients are summarized in Tables 1 and 2. All patients who clinically responded to the therapy demonstrated a decrease of their mSHOX2 plasma DNA (Fig. 1). In this group of responders a decrease of mSHOX2 DNA was seen in most of the patients already at the time of first blood draw (i.e. day 7–10 after therapy start). The median PMR values of the patients responding to the therapy were 4.06% at baseline and dropped to 0.62%, 0.12%, 0.05% at blood draws 1, 2 and 3 after therapy start. In contrast, 8/19 of the non-responding patients demonstrated a reduction of mSHOX2 after start of therapy but the mSHOX2 levels did not change as much as in the responding patients (Fig. 1). In fact the median mSHOX2 values in this group of non-responders were 26.45% at baseline and dropped to 6.86%, 7.78%, 7.96% at blood draws 1, 2 and 3 after therapy start. None of the non-responding patients who demonstrated a mSHOX2 value of ≥ 1% pre-therapy (ranging from 1.1% to 362%) showed a sustained reduction below 1% PMR. (Fig. 1).This observation holds true also for patient UKH 010 who demonstrated an activating EGFR mutation but did not respond to the TKI therapy (Table 3).
Table 2. Clinical data of first-line patients responding to the therapy.
Pat ID | Smoking status / pack years | Histology | Stage (UICC/AJCC) |
---|---|---|---|
UKH 007 | ex-smoker / 35 | SCLC | IV |
UKH 012 | yes / 42 | SCLC | IV |
UKH 015 | ex-smoker / 17 | Adenocarcinoma | IV |
UKH 022 | ex-smoker / 35 | undifferentiated carcinoma | IV |
UKH 025 | ex-smoker / 25 | SCLC | IV |
UKH 026 | yes / 40 | Adenocarcinoma | IIIB |
UKH 029 | ex-smoker / 29 | Adenocarcinoma | IV |
UKH 030 | yes / 33 | Adenocarcinoma | IV |
UKH 031 | never | Adenocarcinoma | IV |
UKH 034 | ex-smoker / 15 | Adenocarcinoma | IV |
UKH 040 | never | Adenocarcinoma | IV |
UKH 042 | ex-smoker / 33 | Sqamous cell carcinoma | IV |
Ten of the patients were male and two were female. The median age of this patient group is 61.5 years. Patient UKH 031 demonstrated an EGFR mutation and was treated with TKI.
The PMR values are calculated as relative amounts of the mSHOX2 gene compared to the ß-actin reference gene (ACTB). The SHOX2 locus is frequently amplified in lung tumors (Schneider et al BMC Cancer 2011) and this copy number variation leads to mSHOX2 PMRs over 100% in patients with very high levels of free circulating tumor DNA.
The median PMR for mSHOX2 at baseline was 4.06% for the 12 responders and 26.45% for the 19 non-responders but this difference is not significant. Interestingly, the ROC curve analyses for the 24 patients with a PMR ≥ 1% showed that the baseline values mSHOX2 did not discriminate between responders and non-responders (area under the curve 0.593), while a discrimination of these two patient populations based on values obtained after start of the therapy demonstrated a high sensitivity and specificity already starting at blood draw one with an AUC of 0.844 which increased to 1.000 at blood draw 7 (Fig. 2). The classification of the patients as responders and non-responders is based on CT scans as the gold standard which was performed by the local tumor board and was completely independent from the mSHOX2 measurements.
A Kaplan-Meier analysis including the 31 patients (i.e. 24 patients with a PMR ≥ 1% plus 7 patients with a PMR ≤ 1%) demonstrated a strong relationship between the survival time and a CT-based assignment of patients into responder and non-responder. The hazard ratio in this calculation is 18 and the P-value ≤ 0.00003 (data not shown). Interestingly, there is a similarly strong relationship between the survival time and the PMR of mSHOX2. When the PMR values from all 36 patients were used for a Kaplan-Meier analysis we could demonstrate that even the plasma mSHOX2 baseline levels showed a trend toward significance (Fig. 3). At blood draw one this difference reached statistical significance which might be interpreted as evidence that this marker also has a predictive value.
Discussion
Due to more powerful drugs [23] and the introduction of a targeted therapy for molecularly selected patient subgroups like patients with an EGFR activation mutation, impressive improvements in the treatment of lung cancer patients were achieved [24]. Additionally, more effective maintenance regimens have shown beneficial effects for patients with advanced stage NSCLC. There is also a survival advantage for patients who are treated with second-line chemotherapy as compared to best supportive care and clinical trials testing new combinations in the second line setting for refractory disease were initiated [25][26]. In order to select the best treatment options, a rapid, specific and sensitive method for the assessment of a therapy response is of crucial importance. The standard procedure for advanced stage lung cancer patients after induction therapy is a CT scan to evaluate the tumor response [21]. Apart from the costs, the sensitivity of this imaging technique is not very high and the inter-observer variability in the measurement of the tumor size is prone to misinterpretation of tumor response [27].
In the last few years several biomarkers have been tested for their usefulness as an indicator for therapy monitoring. Amongst them were eight immunohistochemical biomarkers, none of which could predict chemotherapy response and survival rate and there was only a weak correlation between marker level and treatment response [28]. Neuron specific enolase (NSE) was used for monitoring SCLC patients and found to be only useful in patients with an increased pre-treatment level [29]. The levels of lactate dehydrogenase and chromogranin A did not correlate with treatment response [30]. A longitudinal measurement of soluble interleukin 2 receptor demonstrated a reduction in serum concentration during a therapy but was not a sign for disease remission [31] and thymidine kinase was unable to discriminate between the various response groups of lung cancer patients [32]. There are some additional biomarkers for therapy monitoring in lung cancer patients like CYFRA 21–1 and nucleosome levels which might be better suited but none of them is routinely used in the clinic [33] [34] [35] [36]. In addition there is a growing number of genes which are shown to be hyper- or hypomethylated in lung cancer patients which might be useful as biomarkers [37]. The most frequently analyzed genes like p16, DAPK, APC, RASSF1A, MGMT, FHIT, RARß and GSTP1 were applied as a means of detecting lung cancer or as prognostic factor but none of them had been used for therapy monitoring. Interstingly, all of these methylation markers could not only be detected in tissue but were also found in extracellular nucleic acids isolated from bronchial lavage supernatants, sputum, plasma or serum [37]. The aim of this analysis was to answer the question whether a quantitative measurement of mSHOX2 plasma DNA using a real-time PCR is a useful tool to follow advanced stage lung cancer patients receiving chemo/radio-chemotherapy. Therefore we enrolled all eligible patients who were consecutively admitted to our outpatient department. We considered this approach a proof-of-principle study and for this reason we were more interested in the inclusion of as many patients as possible rather than the establishment of a uniform patient cohort. As a consequence the histologic distribution of the patients included in this analysis does not match exactly the figures given in the literature [38].
Our results demonstrate that a quantitative determination of plasma mSHOX2 DNA appears to be useful for the monitoring of a treatment response for advanced stage lung cancer patients. The turn-over rate of cell-free DNA is rather high as the half-life of extracellular nucleic acids was determined to be less than six hours in an animal model [39]. This correlates well with our observation of a fast and strong decline of plasma mSHOX2 DNA in patients responding to the therapy which holds true for the monitoring of NSCLC and SCLC patients alike. An additional advantage of this method is its applicability for patients with a very low pre-therapeutic mSHOX2 value. We demonstrated that a mSHOX2 measurement taken one week after the start of a therapy is able to divide between responders and non-responders with a very high specificity. If this result can be verified in a large study, physicians have the possibility to switch therapies or spare patients from an invalid therapy. Additionally, we used the data from all 36 patients, i.e. including the patients with a mSHOX2 baseline value of zero and the second-line patients for a Kaplan-Meier survival curve analysis. Interestingly, we demonstrated that the patients with a baseline plasma mSHOX2 level below the median had a slightly longer survival time. When this analysis was performed at blood draw one after therapy start this difference between patients responding to the therapy and non-responders was statistically significant with a p ≤ 0.001. The question whether patients with a very low mSHOX2 baseline value behave differently from patients with a high(er) mSHOX2 level and whether it might be possible to define a pre-therapeutic cut-off value which has a predictive meaning has to be answered in a large study which will be conducted in a multicentric approach.
From our results we conclude that the measurements of the plasma mSHOX2 level are a reflection of the clinical course of late stage lung cancer patients receiving a systemic treatment. Such an analysis allows a more rapid and sensitive determination of tumor response and therapy monitoring than a CT scan. Whether the measurement of extracellular mSHOX2 DNA in plasma might also have a predictive value needs to be demonstrated in future trials.
Acknowledgments
We greatly acknowledge the participation of the patients in this study and the help of Dana Reinicke and Ute Völker. The Epi proLung BL Reflex Assays were kindly provided by Epigenomics.
Data Availability
All relevant data are within the paper.
Funding Statement
This work was supported by the Kirstin Diehl Stiftung. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Metanomics Health GmbH provided support in the form of salaries for authors [Volker Liebenberg], but did not have any additional role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript. The specific roles of these authors are articulated in the ‘author contributions’ section.
References
- 1. Ferlay J, Steliarova-Foucher E, Lortet-Tieulent J, Rosso S, Coebergh JW, et al. (2013) Cancer incidence and mortality patterns in Europe: estimates for 40 countries in 2012. Eur J Cancer 49: 1374–1403. 10.1016/j.ejca.2012.12.027 [DOI] [PubMed] [Google Scholar]
- 2. Siegel R, Naishadham D, Jemal A (2013) Cancer statistics, 2013. CA Cancer J Clin 63: 11–30. 10.3322/caac.21166 [DOI] [PubMed] [Google Scholar]
- 3. Reck M, Heigener DF, Mok T, Soria JC, Rabe KF (2013) Management of non-small-cell lung cancer: recent developments. Lancet 382: 709–719. 10.1016/S0140-6736(13)61502-0 [DOI] [PubMed] [Google Scholar]
- 4. Scheff RJ, Schneider BJ (2013) Non-Small-Cell Lung Cancer: Treatment of Late Stage Disease: Chemotherapeutics and New Frontiers. Semin Intervent Radiol 30: 191–198. 10.1055/s-0033-1342961 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Kratochwil C, Haberkorn U, Giesel FL (2010) [PET/CT for diagnostics and therapy stratification of lung cancer]. Radiologe 50: 684–691. 10.1007/s00117-009-1960-6 [DOI] [PubMed] [Google Scholar]
- 6. William WN Jr, Pataer A, Kalhor N, Correa AM, Rice DC, et al. (2013) Computed tomography RECIST assessment of histopathologic response and prediction of survival in patients with resectable non-small-cell lung cancer after neoadjuvant chemotherapy. J Thorac Oncol 8: 222–228. 10.1097/JTO.0b013e3182774108 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Cho WC (2007) Potentially useful biomarkers for the diagnosis, treatment and prognosis of lung cancer. Biomed Pharmacother 61: 515–519. [DOI] [PubMed] [Google Scholar]
- 8. Mandel P, Metais P (n.d.) Les acides nucleiques du plasma sanguin chez l’homme. C.R.Acad.Sci.Paris 142, 241–243. 1948. [PubMed] [Google Scholar]
- 9. Fleischhacker M, Schmidt B (2007) Circulating nucleic acids (CNAs) and cancer—a survey. Biochim Biophys Acta 1775: 181–232. [DOI] [PubMed] [Google Scholar]
- 10. Jung K, Fleischhacker M, Rabien A (2010) Cell-free DNA in the blood as a solid tumor biomarker—a critical appraisal of the literature. Clin Chim Acta 411: 1611–1624. 10.1016/j.cca.2010.07.032 [DOI] [PubMed] [Google Scholar]
- 11. Alix-Panabieres C, Schwarzenbach H, Pantel K (2012) Circulating tumor cells and circulating tumor DNA. Annu Rev Med 63: 199–215. 10.1146/annurev-med-062310-094219 [DOI] [PubMed] [Google Scholar]
- 12. Smith ZD, Meissner A (2013) DNA methylation: roles in mammalian development. Nat Rev Genet 14: 204–220. 10.1038/nrg3354 [DOI] [PubMed] [Google Scholar]
- 13. Gibney ER, Nolan CM (2010) Epigenetics and gene expression. Heredity (Edinb) 105: 4–13. [DOI] [PubMed] [Google Scholar]
- 14. Suva ML, Riggi N, Bernstein BE (2013) Epigenetic reprogramming in cancer. Science 339: 1567–1570. 10.1126/science.1230184 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Schmidt B, Liebenberg V, Dietrich D, Schlegel T, Kneip C, et al. (2010) SHOX2 DNA methylation is a biomarker for the diagnosis of lung cancer based on bronchial aspirates. BMC Cancer 10: 600 10.1186/1471-2407-10-600 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Kneip C, Schmidt B, Seegebarth A, Weickmann S, Fleischhacker M, et al. (2011) SHOX2 DNA methylation is a biomarker for the diagnosis of lung cancer in plasma. J Thorac Oncol 6: 1632–1638. 10.1097/JTO.0b013e318220ef9a [DOI] [PubMed] [Google Scholar]
- 17. Dietrich D, Kneip C, Raji O, Liloglou T, Seegebarth A, et al. (2012) Performance evaluation of the DNA methylation biomarker SHOX2 for the aid in diagnosis of lung cancer based on the analysis of bronchial aspirates. Int J Oncol 40: 825–832. 10.3892/ijo.2011.1264 [DOI] [PubMed] [Google Scholar]
- 18. Dietrich D, Jung M, Puetzer S, Leisse A, Holmes EE, et al. (2013) Diagnostic and Prognostic Value of SHOX2 and SEPT9 DNA Methylation and Cytology in Benign, Paramalignant and Malignant Pleural Effusions. PLoS One 8: e84225 10.1371/journal.pone.0084225 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Darwiche K, Zarogoulidis P, Baehner K, Welter S, Tetzner R, et al. (2013) Assessment of SHOX2 methylation in EBUS-TBNA specimen improves accuracy in lung cancer staging. Ann Oncol. 10.1093/annonc/mdt547 [DOI] [PubMed] [Google Scholar]
- 20. Dietrich D, Hasinger O, Liebenberg V, Field JK, Kristiansen G, et al. (2012) DNA methylation of the homeobox genes PITX2 and SHOX2 predicts outcome in non-small-cell lung cancer patients. Diagn Mol Pathol 21: 93–104. 10.1097/PDM.0b013e318240503b [DOI] [PubMed] [Google Scholar]
- 21. Goeckenjan G, Sitter H, Thomas M, Branscheid D, Flentje M, et al. (2010) [Prevention, diagnosis, therapy, and follow-up of lung cancer]. Pneumologie 64 Suppl 2: e1–164. 10.1055/s-0029-1243837 [DOI] [PubMed] [Google Scholar]
- 22.R- project for Statistical Computing website. Available: http://www.r-project.org/. Accessed 2014 Nov 17.
- 23. Soria JC, Mauguen A, Reck M, Sandler AB, Saijo N, et al. (2013) Systematic review and meta-analysis of randomised, phase II/III trials adding bevacizumab to platinum-based chemotherapy as first-line treatment in patients with advanced non-small-cell lung cancer. Ann Oncol 24: 20–30. 10.1093/annonc/mds590 [DOI] [PubMed] [Google Scholar]
- 24. Thatcher N, Heighway J (2010) Maintenance and consolidation therapy in patients with unresectable stage III/IV non-small cell lung cancer. Oncologist 15: 1034–1042. 10.1634/theoncologist.2009-0292 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Fathi AT, Brahmer JR (2008) Chemotherapy for advanced stage non-small cell lung cancer. Semin Thorac Cardiovasc Surg 20: 210–216. 10.1053/j.semtcvs.2008.09.002 [DOI] [PubMed] [Google Scholar]
- 26. Varughese S, Jahangir KS, Simpson CE, Boulmay BC (2012) A paradigm shift in the treatment of advanced non-small cell lung cancer. Am J Med Sci 344: 147–150. 10.1097/MAJ.0b013e318246e1b8 [DOI] [PubMed] [Google Scholar]
- 27. Erasmus JJ, Gladish GW, Broemeling L, Sabloff BS, Truong MT, et al. (2003) Interobserver and intraobserver variability in measurement of non-small-cell carcinoma lung lesions: implications for assessment of tumor response. J Clin Oncol 21: 2574–2582. [DOI] [PubMed] [Google Scholar]
- 28. Toffart AC, Timsit JF, Couraud S, Merle P, Moro-Sibilot D, et al. (2013) Immunohistochemistry evaluation of biomarker expression in non-small cell lung cancer (Pharmacogenoscan study). Lung Cancer. 10.1016/j.lungcan.2013.08.029 [DOI] [PubMed] [Google Scholar]
- 29. Splinter TA, Carney DN, Teeling M, Peake MD, Kho GS, et al. (1989) Neuron-specific enolase can be used as the sole guide to treat small-cell lung cancer patients in common clinical practice. J Cancer Res Clin Oncol 115: 400–401. [DOI] [PubMed] [Google Scholar]
- 30. Johnson PW, Joel SP, Love S, Butcher M, Pandian MR, et al. (1993) Tumour markers for prediction of survival and monitoring of remission in small cell lung cancer. Br J Cancer 67: 760–766. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31. Brunetti G, Bossi A, Baiardi P, Jedrychowska I, Pozzi U, et al. (1999) Soluble interleukin 2 receptor (sIL2R) in monitoring advanced lung cancer during chemotherapy. Lung Cancer 23: 1–9. [DOI] [PubMed] [Google Scholar]
- 32. Holdenrieder S, von PJ, Duell T, Feldmann K, Raith H, et al. (2010) Clinical relevance of thymidine kinase for the diagnosis, therapy monitoring and prognosis of non-operable lung cancer. Anticancer Res 30: 1855–1862. [PubMed] [Google Scholar]
- 33. Holdenrieder S, Stieber P, von PJ, Raith H, Nagel D, et al. (2004) Circulating nucleosomes predict the response to chemotherapy in patients with advanced non-small cell lung cancer. Clin Cancer Res 10: 5981–5987. [DOI] [PubMed] [Google Scholar]
- 34. Holdenrieder S, von PJ, Dankelmann E, Duell T, Faderl B, et al. (2009) Nucleosomes and CYFRA 21–1 indicate tumor response after one cycle of chemotherapy in recurrent non-small cell lung cancer. Lung Cancer 63: 128–135. 10.1016/j.lungcan.2008.05.001 [DOI] [PubMed] [Google Scholar]
- 35. Holdenrieder S, von PJ, Dankelmann E, Duell T, Faderl B, et al. (2008) Nucleosomes, ProGRP, NSE, CYFRA 21–1, and CEA in monitoring first-line chemotherapy of small cell lung cancer. Clin Cancer Res 14: 7813–7821. 10.1158/1078-0432.CCR-08-0678 [DOI] [PubMed] [Google Scholar]
- 36. Holdenrieder S, Stieber P, von PJ, Raith H, Nagel D, et al. (2006) Early and specific prediction of the therapeutic efficacy in non-small cell lung cancer patients by nucleosomal DNA and cytokeratin-19 fragments. Ann N Y Acad Sci 1075: 244–257. [DOI] [PubMed] [Google Scholar]
- 37. Fleischhacker M, Dietrich D, Liebenberg V, Field JK, Schmidt B (2013) The role of DNA methylation as biomarkers in the clinical management of lung cancer. Expert Rev Respir Med 7: 363–383. 10.1586/17476348.2013.814397 [DOI] [PubMed] [Google Scholar]
- 38. Hammerschmidt S, Wirtz H (2009) Lung cancer: current diagnosis and treatment. Dtsch Arztebl Int 106: 809–818. 10.3238/arztebl.2009.0809 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39. Rago C, Huso DL, Diehl F, Karim B, Liu G, et al. (2007) Serial assessment of human tumor burdens in mice by the analysis of circulating DNA. Cancer Res 67: 9364–9370. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
All relevant data are within the paper.