Thematic Review Series 2012
Respiration 2012;83:367–376 Published online: May 7, 2012
DOI: 10.1159/000338169
Pleural Effusion in Lung Cancer:
More Questions than Answers
Marios E. Froudarakis
Department of Pneumonology, Medical School, Democritus University of Thrace, Alexandroupolis, Greece
Key Words Introduction
Pleural effusion ⴢ Lung cancer ⴢ Non-small-cell lung
carcinoma ⴢ Target therapy ⴢ Tyrosine kinase inhibitors ⴢ Lung cancer remains the most common fatal malig-
Epidermal growth factor receptor ⴢ Vascular endothelial nancy, despite more aggressive therapies. Patients with
growth factor ⴢ Matrix metalloproteinase ⴢ Cyclooxygenase non-small-cell lung carcinoma (NSCLC) undergoing
radical surgery comprise around 10%, and these are like-
ly to be cured. In advanced-stage disease, first-line che-
Abstract motherapy has shown little improvement in survival, and
Lung cancer remains the most common fatal malignancy, second-line chemotherapy in a few patients in a good
despite more aggressive therapies. Few patients will survive condition has globally shown little effect. In patients with
5 years, as up to 80% of the patients will present with ad- small-cell lung carcinoma (SCLC), chemotherapy showed
vanced-stage disease at diagnosis. Chemotherapy offers lit- important results in terms of response. However, the ef-
tle benefit in terms of median survival and disease-free sur- fect on patients’ survival is globally poor, as a small per-
vival in patients with advanced-stage non-small-cell lung centage of patients with local disease and aggressive ther-
carcinoma (NSCLC). In the last decade, the development of apy combining chemotherapy with mediastinal and cra-
new targeted therapies based on the better understanding nial irradiation hope to survive at 5 years. The development
of different paths of carcinogenesis has given new hope to of novel targeted therapies has given new hope to both
both physicians and patients. Metastatic pleural effusion physicians and patients [1, 2].
from lung cancer has a particularly poor prognosis, and in
NSCLC it is actually reclassified as stage IV disease. A possible
explanation of this observation is differences in the genom-
ics between primary tumors and metastasis, leading to pos- Previous articles in this series: 1. Anevlavis S, Tzouvelekis A, Bouros D:
sible different therapeutic approaches with novel molecular Mechanisms of pleural involvement in orphan diseases. Respiration
therapies in this patient population. The current review aims 2012;83:5–12. 2. Rodriguez-Panadero F, Montes-Worboys A: Mecha-
to summarize the actual situation of research in pleural dis- nisms of pleurodesis. Respiration 2012;83:91–98. 3. Grundy S, Bentley
A, Tschopp JM: Primary spontaneous pneumothorax: a diffuse dis-
ease due to lung carcinoma in relation to novel targeted ease of the pleura. Respiration 2012;83:185–189. 4. Haas AR, Sterman
therapies tested in this patient population. DH: Novel intrapleural therapies for malignant diseases. Respiration
Copyright © 2012 S. Karger AG, Basel 2012;83:277–292.
© 2012 S. Karger AG, Basel Marios E. Froudarakis, MD, PhD
0025–7931/12/0835–0367$38.00/0 Department of Pneumonology, Medical School, Democritus University of Thrace
Fax +41 61 306 12 34 University Hospital of Alexandroupolis
E-Mail karger@karger.ch Accessible online at: GR–68100 Alexandroupolis (Greece)
www.karger.com www.karger.com/res Tel. +30 2551075335, E-Mail mfroud @ med.duth.gr
, In patients with lung cancer, the incidence of pleural Epidermal Growth Factor Receptor Pathway
effusion ranges between 7% (280 out of 4,000 cases) and
23% (5,888 out of 25,464 cases) [3]. Similarly, in a study of The epidermal growth factor receptor (EGFR) gene is
patients presenting with malignant pleural effusion, lung located on chromosome 7p12.1–12.3 in humans. It is a
cancer was the most frequent cause (641 out of 1,783 pa- transmembrane tyrosine kinase receptor that is frequent-
tients; 36%), while the second most common cause was ly expressed in various kinds of epithelial tumors [13]. Its
breast carcinoma (449 cases; 25%) [4]. All histological expression is regulated by one promoter and two enhanc-
types of lung carcinomas are likely to cause pleural effu- er regions [14]. The signaling pathways of activated EGFR
sion [3]. The most frequent histological type seems to be are involved in cancer cell proliferation, apoptosis, angio-
adenocarcinoma (40% of the cases), with SCLC second, in genesis, invasion and metastasis, suggesting that EGFR is
about 25% of the cases [3]. Adenocarcinoma is actually the a target for molecular therapy. NSCLC accumulates mul-
commonest histological type of NSCLC; also, it is more tiple genetic alterations [15]. Within the tyrosine kinase
likely to arise in the lung periphery invading the pleura [5]. domain, mutations of EGFR exons 19 and 21 may play an
Recently, the International Association for the Study of important role in NSCLC carcinogenesis, since it has
Lung Cancer Lung Cancer Staging Project reviewed the T been shown that they are related to prolonged survival,
and M stage of 18,198 NSCLC patients. The number of by increasing sensitivity to tyrosine kinase inhibitors
patients with pleural dissemination – classified as T4 ac- (TKIs). In contrast, insertion mutation of exon 20 is as-
cording to Mountain’s revised classification – was 471. sociated with resistance to TKIs [16, 17].
Five-year and median survival for those patients were 2% Recent studies have defined clinicopathological varia-
and 8 months, respectively, versus 14% and 13 months for tions in terms of response to TKIs, with adenocarcinoma,
clinical T4 disease other than pleural effusion (p ! 0.0001). nonsmokers, females and Asians being more likely to re-
According to these data, the patients were reclassified as spond to therapy [17, 18]. This patient population has
having M1a disease [6], since their outcome was the same been associated with higher rates of EGFR mutations.
as in distant metastasis [7]. In SCLC, median survival of Also, it has been shown that the expression of EGFR mu-
patients with limited disease and ipsilateral malignant tations might be used to predict responsiveness to TKI
pleural effusion (12 months) is worse than in patients with therapy. Identification of clinical and genetic factors un-
limited-stage disease without pleural effusion (18 months) derlying EGFR mutations will be particularly meaning-
but better than those with extensive disease (7 months; p ful not only for understanding lung carcinogenesis in
= 0.0001) [8–10]. The reasons for this intermediate prog- nonsmokers but also for making clinical decisions about
nosis should be clarified by further clinical trials [11]. the use of EGFR TKIs in nonoperative patients. Patients
Overall, different clinical staging systems may indi- with EGFR mutations may benefit from better response
cate different survival in lung cancer patients showing and survival, independently of the treatment received
pleural involvement; however, they cannot explain the [18].
aggressive behavior of pleural disease, despite the disease Although in recent years there has been increased in-
remaining in the same hemithorax. The explanation terest in EGFR mutations and their therapeutic applica-
might be given by the study of novel molecular factors of tions in NSCLC, with more than 3,500 publications to
different pathways of carcinogenesis, and the differences date, few studies have reported data from patients with
in this status between primary tumors and metastasis, pleural metastasis from their carcinoma (table 1). All
leading to application of novel molecular therapies in this these studies involved Asiatic patients (Japan 5, Taiwan
patient population. Many of those factors have already 4, PR China 1 and Singapore 1), with the rate of lung ad-
been assessed in the last decade in patients with meta- enocarcinomas ranging between 85 and 100%. Exons
static pleural effusion, most of them to differentiate exu- studied were 18, 19, 20 and 21, and findings varied ac-
dates from transudates and/or to increase the diagnostic cording to the study. Overall, observed mutation rates
yield of pleural cytology [12]. However, the results are dif- ranged from 24 to 68.7% with a median of 34%. These
ficult to apply in lung carcinoma pleural effusions, since data are comparable to the mutation rates in other EGFR
they concern effusions from metastases of various prima- reports from Asiatic patients studying surgical [19, 20]
ries. The current review aims to summarize the actual and nonsurgical [21, 22] specimens (32–55%). A recent
situation of research in pleural disease due to lung carci- study in Korea [23] comparing the EGFR status between
noma in relation to novel targeted compounds tested in the primary tumor and pleural metastatic disease showed
this patient population. a significant discordance of 16% between the primary tu-
368 Respiration 2012;83:367–376 Froudarakis
