Potential diagnostic value of methylation profile in pleural fluid and serum from cancer patients with pleural effusion

Abstract

BACKGROUND.

The objective of this study was to investigate the diagnostic value of methylation profiles for discrimination between malignant and benign pleural effusions. A secondary objective was to examine the concordance of methylation in samples of serum and pleural fluid.

METHODS.

The authors used methylation-specific polymerase chain reaction (MSP) analysis to examine the promoter methylation status of 4 genes in patients with pleural effusion: death-associated protein kinase (DAPK), Ras association domain family 1A (RASSF1A), retinoic acid receptor β (RARβ), and p16/INK4a. Pleural effusions were collected from 87 patients who had their diagnoses confirmed on cytologic and/or histologic examinations and clinical evolution. Pleural effusions were classified as malignant (n = 53 patients) or benign (n = 34 patients).

RESULTS.

Methylation was detected in serum from 45.3% of patients with malignant pleural effusions and from 0% of patients with benign pleural effusions, and it was detected in pleural fluid samples from 58.5% of patients with malignant pleural effusions and from 0% of patients with benign pleural effusions (P = .001). The sensitivity of MSP was greater than that of cytologic examination alone (39.1%; P = .001). When MSP was used together with cytologic examination, sensitivity increased to 69.8% (P = .001).

CONCLUSIONS.

Cell-free methylated DNA in pleural fluid can be detected in patients with neoplastic malignancy in a single extraction by thoracocentesis. Adequate management of the extracted pleural fluid can provide a rapid and reliable diagnosis in patients with pleural effusions who have suspected malignancy. MSP, used together with cytologic examination, may obviate the need for other invasive diagnostic tests. Cancer 2006. © 2006 American Cancer Society.

The pleura is the serous membrane lining the thoracic cavity. The pleural cavity normally contains approximately 1 mL of fluid, which is the balance between hydrostatic and oncotic forces in the visceral and parietal pleural vessels and lymphatic drainage. Pleural effusions result from disruption of this balance. The clinical importance of pleural effusions ranges from incidental manifestations of cardiopulmonary diseases to symptomatic inflammatory or malignant diseases requiring evaluation and treatment.1 Exact diagnosis of pleural effusions is difficult; in 40% of malignant effusions, cytologic examination of pleural fluid does not detect tumor cells,2, 3 and the diagnosis of malignant pleural effusions often requires combined procedures.4

Changes in DNA methylation are among the most common molecular alterations in human neoplasia.5, 6 Aberrant hypermethylation of the promoter regions of genes is a major mechanism for silencing tumor suppressor genes or other cancer-associated genes in many kinds of human cancer.7 The reciprocal relation between the density of methylated cytosine residues and the transcriptional activity of a gene has been widely documented.8

Retinoic acid receptor β (RARβ), Ras association domain family 1A (RASSF1A), p16/INK4a, and death-associated protein kinase (DAPK) are strong candidate biomarkers for early detection of cancer9 and are involved in important cellular regulatory pathways, apoptosis, and ras signal transduction.10 It has been observed that RARβ,11RASSF1A,12p16/INK4a,13 and DAPK14 harbor hypermethylated promoters in 15% to 45% of solid tumors,15 and hypermethylation changes may be excellent tumor markers.16 Recent publications also have demonstrated the presence of promoter hypermethylation of various genes in bodily fluids, including serum,17, 18 bronchoalveolar lavage from patients with lung cancer,19 sputum,20 and pleural fluid.21 The methylation-specific polymerase chain reaction (PCR) technique (MSP) is highly sensitive and can detect 0.1% tumor DNA from a heterogeneous cell population.22

For the current study, we analyzed the promoter hypermethylation pattern of RARβ, RASSF1A, p16/INK4a, and DAPK in cell-free DNA that was obtained from simultaneous samples of serum and pleural fluid from 87 patients with pleural effusions. Our objectives were to confirm the presence of cell-free DNA in serum and pleural fluid, to investigate the nature of that DNA by detection of promoter hypermethylation, to examine the concordance between methylation of selected genes in serum and pleural fluid, and to examine the role of hypermethylation as a marker of malignant pleural effusions.

MATERIALS AND METHODS

Patients

We evaluated 87 patients (55 males and 32 females) with pleural effusions from the Hospital General de Alicante, (Alicante, Spain) between February 2, 2002 and May 14, 2004. The mean age (± standard deviation) was 66 ± 15.6 years. All patients gave their signed informed consent, and the study was approved by the hospital ethics committee. Effusions were considered malignant if 1 of the following criteria was met: demonstration of malignant cells by cytologic examination or in biopsy specimen or histologically proven primary malignancy with the exclusion of any other cause known to be associated with pleural effusion. Effusions were malignant in 53 patients and benign in 34 patients (Table 1). All pleural effusions were studied by cytologic examination, pleural biopsy, and/or thoracoscopy with biopsy of visually identified abnormal areas of the pleura.

Table 1. Diagnosis of Patients with Pleural Effusions

Diagnosis	No. of patients
* Lung malignancies included adenocarcinoma (11 patients), squamous cell carcinoma (7 patients), and oat cell (5 patients). † Digestive adenocarcinomas included colon adenocarcinoma (1 patient), gastric adenocarcinoma (7 patients), rectal adenocarcinoma (1 patient), and pancreatic adenocarcinoma (1 patient). ‡ Other malignancies included endometrial carcinoma (3 patients), myeloma (1 patient), sarcoma (1 patient), and unknown primary malignancies (2 patients).
Malignant pleural effusion	53
Lung*	23
Breast carcinoma	10
Digestive adenocarcinoma†	10
Ovarian carcinoma	3
Other‡	7
Nonmalignant pleural effusion	34
Emphysema	12
Cardiac failure	9
Tuberculous pleurisy	8
Pulmonary embolism	1
Liver disease-cirrhosis	3
Hypoproteinemia	1
Total	87

Collection and Processing of Blood and Pleural Fluid Samples and DNA Extraction

We collected 10 mL of pleural fluid and blood samples with a hollow needle in sterile Vacutainer tubes that contained SST gel and clot activator (Becton Dickinson, Oxford, UK) from patients at the time of thoracocentesis or thoracoscopy. The pleural fluid and serum were collected after 15 minutes of centrifugation at × 1600 g and were stored in 1-mL aliquots at −20°C until DNA extraction. DNA was purified from 400 μL of serum or pleural fluid by using DNA QIAmp Blood Mini Kit (Qiagen, Valencia, CA) according to the manufacturer's instructions.

MSP

DNA methylation patterns were tested in the CpG islands of p16/INK4a, RASSF1A, DAPK, and RARβ with MSP after sodium bisulphate treatment (Fig. 1). The primers used are shown in Table 2. At least 100 ng of sample DNA mixed with 1 μg of salmon sperm (Sigma Chemical Company, St. Louis, MO) were submitted to chemical modification according to the method described by Herman et al.22 Briefly, DNA was denatured with 2 M NaOH, followed by treatment with 10 mM hydroquinone and 3 M sodium bisulphate (Sigma Chemical Company) at 55°C for 16 hours. After purification in a Wizard SV Plus kit column (Promega, Madison, WI), the DNA was treated with 3 M NaOH and precipitated with 3 volumes of 100% ethanol, a 33% volume of 10 M NH4OAc, and 20 μg glycogen (Roche Molecular Biochemicals, Mannheim, Germany) at −20°C. The precipitated DNA was washed with 70% ethanol and dissolved in distilled water. PCR was conducted with primers specific for either the methylated or the unmethylated versions of the p16/INK4a, DAPK, RASSF1A, and RARβ promoter regions (Table 2). The 25-μL total reaction volume contained modified DNA, all 4 deoxynucleotide triphosphates (each at 300 μM), 3 mM MgCl₂, 0,75 μM PCR primers, and 1 unit of Hot Start DNA polymerase (Qiagen). DNA was substituted for water as a negative control. DNA from peripheral blood lymphocytes and genomic DNA (Roche Molecular Biochemicals) treated with SssI methyltransferase (New England Biolabs Inc., Beverly, MA) were used as positive controls for methylated reaction. DNA from peripheral blood lymphocytes and genomic DNA (Roche Molecular Biochemicals) were used as positive controls for unmethylated reaction. DNA was amplified by an initial cycle at 95°C for 15 minutes, as required for enzyme activation, followed by 40 cycles at 95°C for 30 seconds, annealing for 30 seconds, and 72°C for 30 seconds, and ending with a 10-minute extension at 72°C in a 2700 Thermocycler (Applied Biosystems, Foster City, CA). PCR products were separated onto 4% Nusieve agarose gels and visualized after staining with ethidium bromide. Results were confirmed by repeating MSP analysis twice for all samples.

Figure 1. (A) Retinoic acid receptor β, (B) Ras association domain family 1A, (C) p16/INK4a, and (D) death-associated protein kinase CpG island methylation analysis by methylation-specific polymerase chain reaction (PCR) in serum and pleural fluid (PF) samples. A PCR band under lanes marked M indicates methylated genes, and a PCR band under lanes marked U indicates unmethylated genes. Arrows indicate expected bands. Methylated genes were identified in (A) serum from Patient (Pt) 194, (B) serum and PF from Patient 140 and in serum from Patient 198, (C) PF from Patient 105 and in serum and PF from Patient 118, and (D) serum from Patient 105.

Table 2. Primer Sequences for Methylation-Specific Polymerase Chain Reaction Analysis

Gene	Sense	Antisense	Temperature (°C)	Size (base pairs)
M indicates methylated; U, unmethylated; RASSF1A, Ras association domain family 1A; DAPK, death-associated protein kinase; RARβ, retinoic acid receptor β.
p16/INK4A	M: TTATTAGAGGGTGGGGCGGATCGCGTGC; U: TTATTAGAGGGTGGGGTGGATTGT	M: ACCCGACCCCGAACCGCGACCGTAA; U: CAACCCCAAACCACAACCATAA	M, 65; U, 63	M, 149; U, 151
RASSF1A	M: GGGTTTTGCGAGAGCGCG; U: GGTTTTGTGAGAGTGTGTTTAG	M: GCTAACAAACGCGAACCG; U: CACTAACAAACACAAACCAAAC	M, 65; U, 60	M, 155; U, 155
DAPK	M: GGATAGTCGGATCGAGTTAACGTC; U: GGAGGATAGTTGGATTGAGTTAATGTT	M: CCCTCCCAAACGCCGA; U: CAAATCCCTCCCAAACACCAA	M, 65; U, 62	M, 102; U, 98
RARβ	M: TCGAGAACGCGAGCGATTCG; U: TTGAGAATGTGAGTGATTTGA	M: GACCAATCCAACCGAAACGA; U: AACCAATCCAACCAAAACAA	M, 55; U, 53	M, 146; U, 146

Statistical Analysis

The diagnostic accuracy of MSP-based detection of hypermethylation in serum and pleural fluid was assessed by calculating sensitivity, specificity, positive predictive value, and negative predictive value. The chi-square test or the Fisher exact test was used to determine the association between categoric variables. The Pearson correlation coefficient was used to evaluate the association between the number of hypermethylated genes in serum and pleural fluid. Statistical significance was set at P < .05.

RESULTS

We tested promoter methylation patterns of the p16/INK4a, RASSF1A, DAPK, and RARβ genes in 87 paired samples of serum and pleural fluid from 2 groups of patients who had pleural effusions with and without malignancy. To investigate the tumor-specific methylation of the 4 genes, we compared the frequency of hypermethylation at each promoter region between 53 cancer patients and 34 cancer-free patients. DNA was extracted from each sample, and all the samples were useful for molecular analysis. Conventional cytology of pleural effusion was performed in 48 of 53 cancer patients. A mean of 1.4 ± 1 cytologic examination was performed for each patient and was observed by the cytologists in charge of the hospital's cytology laboratory.

Frequency of Methylation in Serum Samples

Among the patients with cancer, methylated DNA was detected in serum samples from 13.2% of patients for DAPK, 18.9% of patients for p16/INK4a, 17% of patients for RASSF1A, and 18.9% of patients for RARβ (Table 3). Overall, 24 of 53 patients (45.3%) were methylation-positive for at least 1 of the 4 genes tested (Table 4). No aberrant methylation was observed in serum DNA samples from the group of 34 cancer-free patients. Aberrant methylation detected in serum had no correlation with patient data, including age, gender, diagnosis, or length of survival (data not shown).

Table 3. Frequency of Methylation in Four Genes

Gene	No. of patients (%)
	Malignant pleural effusions (n = 53)		Nonmalignant pleural effusions (n = 34)
	Serum	Pleural fluid	Serum	Pleural fluid
DAPK indicates death-associated protein kinase; RASSF1A, Ras association domain family 1A; RARβ, retinoic acid receptor β.
DAPK	7 (13.2)	10 (18.9)	0 (0)	0 (0)
p16/INK4a	10 (18.9)	20 (37.7)	0 (0)	0 (0)
RASSF1A	9 (17)	8 (15.1)	0 (0)	0 (0)
RARβ	10 (18.9)	11 (20.8)	0 (0)	0 (0)

Table 4. Methylation Status in Serum and Pleural Effusion Samples from Patients with Malignant Pleural Effusion

No. of methylated genes	Patients with malignant pleural effusions (n = 53): No. (%)
	Serum	Pleural fluid
No genes	30 (54.7)	22 (41.5)
1 gene	14 (26.4)	18 (34)
2 genes	8 (15.1)	9 (17)
3 genes	2 (3.8)	3 (5.7)
4 genes	0 (0)	1 (1.9)

Frequency of Methylation in Pleural Fluid Samples

Among the patients with cancer, methylated DNA was detected in pleural fluid from 18.9% of patients for DAPK, 37.7% of patients for p16/INK4a, 15.1% of patients for RASSF1A, and 20.8% of patients for RARβ (Table 3). Overall, 31 of 53 patients (58.5%) were methylation-positive for at least 1 of the 4 genes tested (Table 4). No aberrant methylation was observed in pleural fluid DNA samples from the group of 34 cancer-free patients. Aberrant methylation detected in pleural fluid had no correlation with patient data, including age, gender, diagnosis, or length of survival (data not shown).

Concordance between Methylation in Serum and in Pleural Fluid

Aberrant methylation was identified in paired samples (1 or 2 genes in each sample) in 14 patients who showed concordance between the methylation of selected genes in serum and in pleural fluid (r = 0.6; P = .0001) (Table 5). In 10 patients, methylation was observed only in pleural fluid and not in serum; conversely, in 3 patients, it was observed in serum and not in pleural fluid. Hypermethylation was detected 1.3 times more frequently in pleural fluid than in serum.

Table 5. Summary of Methylation Status of Death-Associated Protein Kinase, p16/INK4a, Ras Association Domain Family 1A, and Retinoic Acid Receptor β in 53 Serum and Pleural Fluid Samples from Patients with Malignant Disease

Patient no.	Age, years	Gender	Tumor type	Cytology	DAPK		p16/INK4a		RASSF1A		RARß
					S	PF	S	PF	S	PF	S	PF
DAPK indicates death-associated protein kinase; RASSF1A, Ras association domain family 1A; RARß, retinoic acid receptor β; S, serum; PF, pleural fluid; +, positive; U, unmethylated genes; M, methylated genes; −, negative; ND, not done.
62	53	Female	Breast	+	U	U	U	U	U	U	U	U
93	47	Female	Breast	+	U	M	M	M	U	U	U	M
140	71	Female	Breast	−	U	U	U	U	U	U	M	M
163	60	Female	Breast	−	U	U	U	U	U	U	U	U
172	81	Female	Breast	+	U	U	U	U	U	U	U	M
178	64	Female	Breast	+	U	M	M	M	M	M	U	M
188	33	Female	Breast	ND	U	M	U	M	U	U	U	M
194	64	Female	Breast	−	U	U	U	U	M	U	M	U
201	62	Female	Breast	+	U	U	U	U	U	U	U	U
25	48	Male	Breast	−	U	U	U	U	U	U	U	U
74	74	Male	Colon	−	M	M	M	M	M	U	U	U
92	69	Male	Gastric	−	U	U	U	U	U	U	U	U
99	57	Male	Rectal	−	U	U	M	M	U	U	M	M
109	55	Male	Gastric	−	U	U	U	M	U	U	U	U
143	83	Male	Gastric	−	U	M	U	U	U	U	M	U
144	76	Female	Gastric	−	U	U	U	U	U	U	U	U
156	81	Male	Gastric	+	M	U	U	U	U	U	U	U
182	40	Male	Gastric	−	U	U	U	U	U	U	U	U
197	66	Male	Pancreas	−	U	U	U	U	U	U	U	U
198	61	Male	Gastric	−	U	U	U	U	U	U	M	U
124	49	Female	Endometrial	+	U	M	U	M	U	U	M	U
203	54	Female	Endometrial	−	U	U	U	U	U	M	M	U
202	55	Female	Endometrial	−	U	U	U	U	U	U	U	U
39	52	Male	Lung, adenocarcinoma	+	U	U	U	U	U	U	U	U
43	49	Male	Lung, adenocarcinoma	+	U	U	U	M	U	U	M	M
53	69	Male	Lung, squamous	−	U	U	U	U	U	U	U	U
56	49	Male	Lung, oat cell	−	U	U	U	U	M	M	U	U
57	77	Male	Lung, adenocarcinoma	−	U	M	U	U	U	U	U	U
65	53	Male	Lung, squamous	−	U	U	U	U	U	U	U	U
79	66	Male	Lung, oat cell	−	U	U	U	U	U	U	U	U
86	79	Male	Lung, oat cell	ND	M	U	M	M	M	M	U	U
105	74	Male	Lung, squamous	−	M	U	U	M	U	U	U	U
118	60	Female	Lung, adenocarcinoma	−	U	U	M	M	U	U	U	U
121	91	Male	Lung, adenocarcinoma	+	U	U	U	U	M	U	U	M
132	74	Male	Lung, oat cell	+	U	U	M	M	U	M	M	M
147	38	Male	Lung, adenocarcinoma	−	U	U	M	M	U	U	M	U
150	67	Female	Lung, adenocarcinoma	+	U	U	U	U	U	U	U	U
154	81	Male	Lung, squamous	−	U	M	U	M	U	U	U	U
158	61	Male	Lung, squamous	−	U	U	U	M	U	U	U	U
165	44	Male	Lung, adenocarcinoma	+	M	M	U	U	U	U	U	U
183	64	Male	Lung, adenocarcinoma	−	M	U	U	M	U	M	U	U
185	62	Male	Lung, adenocarcinoma	−	U	U	U	U	U	U	U	U
186	73	Male	Lung, oat cell	+	U	U	U	M	U	U	U	U
189	69	Male	Lung, squamous	ND	U	U	U	U	U	U	U	U
200	76	Male	Lung, adenocarcinoma	−	U	U	U	U	U	U	U	M
199	74	Male	Lung, adenocarcinoma	−	U	U	U	U	U	U	U	U
120	46	Female	Myeloma	ND	U	U	M	M	M	U	U	M
102	67	Female	Ovary	+	U	U	M	M	M	U	U	U
130	27	Female	Ovary	−	U	U	U	U	U	M	U	U
167	77	Female	Ovary	ND	M	U	U	M	M	U	U	U
12	30	Female	Sarcoma	+	U	U	U	U	U	U	U	U
47	77	Male	Unknown primary	+	U	M	U	U	U	M	U	U
152	44	Male	Unknown primary	−	U	U	U	U	U	U	U	U

Accuracy of MSP-Based Detection of Promoter Hypermethylation in Serum and Pleural Fluid

The accuracy of MSP-based detection of promoter hypermethylation in serum and pleural fluid was assessed by calculating sensitivity, specificity, positive predictive value, and negative predictive value (Table 6). Forty-three percent of serum samples and 58.5% of pleural fluid samples had methylation in at least 1 of the 4 genes with 100% specificity, whereas all 34 control samples were negative for methylation in all 4 genes. Conventional cytology of pleural effusion samples revealed neoplastic cells in 17 of 48 cancer patients (39.1% sensitivity) with 100% specificity (Table 6). When both conventional cytology of pleural effusion and detection of hypermethylation in pleural fluid were combined, sensitivity increased to 69.8% with 100% specificity (Table 6). When the detection of hypermethylation in both bodily fluids was combined, sensitivity was 64.2% with 100% specificity.

Table 6. Accuracy of Methylation-Specific Polymerase Chain Reaction-Based Detection of Promoter Hypermethylation in Serum and Pleural Fluid, as Assessed by Calculation of Sensitivity, Specificity, Positive Predictive Value, and Negative Predictive Value

Result	Sensitivity (%)	Specificity (%)	PPV	NPV
PPV indicates positive predictive value; NPV, negative predictive value.
Cytology (n = 79)	39.1	100	100	54.1
At least 1 gene methylated in serum (n = 87)	45.3	100	100	55.4
At least 1 gene methylated in pleural fluid (n = 87)	58.5	100	100	62.1
Positive cytology or at least 1 gene methylated in pleural fluid (n = 87)	69.8	100	100	68.0
At least 1 gene methylated in pleural fluid or in serum (n = 87)	64.2	100	100	64.2

DISCUSSION

The current study included patients with suspected malignant pleural effusions. The average life expectancy for these patients is low. A major factor in the high mortality of cancer patients is the presence of metastatic tumors in approximately 66% of patients at the time of diagnosis.23 Therefore, the ability to determine the malignancy or benignancy of pleural effusions is very useful both for prognosis and for clinical management.

Several methods are used currently for the diagnosis of pleural effusions. The most informative laboratory procedure is thoracocentesis and cytology. Several factors influence the accuracy of cytologic diagnosis, and the main factor is the primary tumor.2 If cytologic results are negative, then the next step is a new thoracocentesis, pleural biopsy, and thoracoscopy. Thoracoscopy provides a tissue diagnosis in 80.3% of patients who have recurrent effusions that are not diagnosed after thoracocentesis, pleural biopsy, and bronchoscopy.1

Changes in the status of DNA methylation are among the most common molecular alterations in human neoplasia.5 Several genes are involved in the pathogenesis of cancer and are inactivated frequently by aberrant promoter methylation.24 The potential prognostic and therapeutic values of DNA methylation in cancer have been studied in various types of neoplasia, including nasopharyngeal carcinomas,25 nonsmall cell lung carcinomas,26 and gallbladder carcinomas.27 In the current study, tumor tissue was not available for every patient. However, it has been demonstrated that the presence of promoter methylation of various genes in bronchoalveolar lavage19, 24 and serum17, 18 is a surrogate for methylation of the same genes in tumor tissue. Moreover, circulating DNA found in serum harbors the same genetic characteristics that are observed in paired tumor DNA, as reported by several authors.17, 18, 28 We detected aberrant methylation of some of the 4 genes studied in patients with malignant tumors but detected 0% frequency in both fluids tested in the control group. These findings are along the same lines as the findings reported by other authors.19 Hypermethylation of several genes in the bronchial epithelium and sputum of cancer-free individuals24 may be related to direct exposure of epithelial cells to the carcinogenic factor cells; however, it appears that hypermethylation in pleural fluid may be limited to patients who have metastases in the pleural cavity. In our study, the gene panel was successful in detecting neoplastic DNA with a sensitivity of 58.5% in pleural fluid, whereas the sensitivity of cytology was 39.1%. In general, when detecting tumor cells in the pleural fluid from patients with squamous cell carcinoma, positive results are uncommon, because the pleural effusions usually are caused by bronchial obstruction or lymphatic blockade.2 In our hospital, sensitivities up to 50% have been reached,29 and the low frequency observed in the current study may have been because 13% of patients had squamous cell carcinoma. We agree with other authors21 that the prevalence of methylation helps to increase the sensitivity of conventional cytology. In our study, methylation increased the sensitivity in a single sample to 69.8%, obviating the need for surgical procedures except in specific situations. In addition to the diagnosis of malignant pleural effusions, methylation may be beneficial in the monitoring of disease progress after treatment.

In the current study, we demonstrated the role of hypermethylation as a marker of malignant pleural effusions, detecting neoplastic methylated DNA in pleural effusions, even in patients who had negative results on cytologic analysis. Our gene panel successfully detected almost all types of solid tumors that were included in the study, although 35.8% of our tumors (Table 6) remained undetected. Possible explanations for this result include the specific characteristics in the progression of each tumor and/or the quantity and quality of DNA template extracted from bodily fluids, which are likely to differ from the quality of the original tumor tissue based on time of collection, the content of DNase, and the DNA integrity, among other factors. It is possible that none of the genes selected in our panel are methylated in the primary tumor; thus, no methylation would be detected in bodily fluids. Conversely, it is possible that the primary tumors harbored some methylation of the gene panel, but we were not able to detect it in the fluids. Therefore, further studies are warranted to improve the sensitivity of the assay by verifying the minimum number of markers required to identify solid tumors that can metastasize to the pleural cavity and by including the detection of PCR fragments by fluorescence-based techniques (e.g., real-time PCR)30 or nested PCR21. In conclusion, the detection of cell-free methylated DNA in pleural fluid can be performed in patients with neoplastic malignancy by means of a single extraction by thoracocentesis and adequate management, allowing for a rapid and reliable diagnosis in patients with pleural effusions suspected of malignancy without the need for other invasive procedures. Larger studies are warranted to validate these findings.

Acknowledgements

We thank Elena Pena, Rosana Montoyo, Jorgina Brana, Rosa Cabrera, Elena Bernabe, and Catherine Pantaleon for assistance. We also thank Renee O'Brate for assistance with the article.