Background
Background
Non-small cell carcinoma of the lung (NSCLC) is the most common cause of cancer death
in the United States [1], with more than 160,000 annual deaths [2]. The American Joint Committee on Cancer TNM system is the current staging system
for lung cancer, and its goal is to classify patients into groups based on the extent
of disease. This system relies heavily on the pathologic evaluation of the primary
tumor (T), regional nodes (N) and distant metastases (M). Lymph nodes in the mediastinum
are the most common sites of metastases. Recent advances in molecular diagnostics
has great potential to increase the accuracy of detection of both primary tumors and
metastatic lymph nodes by enabling detection of very small numbers of cancer cells
(”micrometastases”) using minimally invasive sampling techniques such as image-guided
fine needle aspiration (FNA).
Mediastinal lymph nodes evaluation is a critical component of staging NSCLC patients.
Lymph nodes larger than 1 cm in diameter by CT are presumed to contain metastatic
disease. However, when compared with surgical pathology, CT misses mediastinal lymph
node metastases (false negative) in approximately 13 % of patients and incorrectly
suggests it (false positive) in approximately 50 % [3]. Because of this limitation, mediastinoscopy has, for years, been accepted as the
”gold standard” for mediastinal staging. Mediastinoscopy requires general anesthesia
and specialized expertise to perform safely, and therefore is not universally employed.
Recently, positron emission tomography (PET) has emerged as a non-invasive method
for evaluating cancer stage. Recent studies suggest that the sensitivity and specificity
of PET may be in excess of 80 - 90 % [4]
[5]. Because PET relies on the accumulation of a radioactive tracer in proportion to
cellular metabolic activity, its ability to detect microscopic tumor deposits and
distinguish between inflammation and cancer is limited. Moreover, PET and CT scan
cannot make a tissue diagnosis.
The presence or absence of mediastinal lymph node metastases has been classically
dependent on histological analysis alone. However, survival statistics indicate clearly
that reliance on histology is inadequate. Following presumably curative surgical resection,
the five-year survival rate for patients with pathologic stage I disease (no histologic
evidence of lymph node metastases) is only 62 %. For patients with metastatic disease
identified in hilar lymph nodes but not mediastinal lymph nodes (stage II), the five-year
survival rate falls to only 42 % [6]. These figures clearly suggest that histological evaluation of mediastinal lymph
nodes may miss metastatic disease in a large proportion of patients with NSCLC. Studies
have shown that serial sectioning and immunohistochemical staining increase the sensitivity
of detection of metastatic disease, and that the presence of metastatic disease detected
in this fashion is associated with worse survival [7]
[8]
[9]. Although serial sectioning provides the ability to detect clinically significant
metastatic disease, it is extremely time-consuming and expensive and is thus is not
practical performed on routine basis.
Standard therapies for patients with NSCLC include surgery, chemotherapy and radiation
therapy, depending on the stage of disease. Surgery is considered most appropriate
for patients in whom disease is confined to the lung and hilar lymph nodes (stages
I and II). For patients with metastatic disease to mediastinal lymph nodes (stage
III), the benefit of surgery as primary therapy is questionable and combined chemo-radiotherapy
may be most appropriate [10]. Better preoperative staging would reduce the rate of unnecessary surgical exploration
and morbidity.
Role of endoscopic ultrasound
Role of endoscopic ultrasound
Endoscopic ultrasound (EUS) is a minimally invasive method capable of sampling mediastinal
lymph nodes by fine needle aspiration and can be performed under conscious sedation
on an outpatient basis. It has been shown that EUS-FNA is a safe and accurate method
of staging lymph nodes in the mediastinum of NSCLC patients [3]
[11]
[12].
Experience at multiple institutions has shown that EUS-FNA with cytology improves
the accuracy of detection of mediastinal metastases compared to CT and PET [3]
[11]
[13]
[15]
[16]
[17]. In addition, EUS-FNA is more cost effective than other surgical methods of lymph
node sampling such as mediastinoscopy [14].
Tissue samples obtained by EUS-FNA are typically analyzed by standard cytopathology
techniques (thin prep, Papanicolau staining), but can also be analyzed for molecular
markers of cancer cells that may not be detectable by cytological methods.
Molecular studies of lymph nodes
Molecular studies of lymph nodes
Molecular characterization of lymph node tissue is a rapidly expanding field. Standard
cytological or histological methods usually fail to detect early ”micrometastases”
in lymph nodes. Many of these tumor cells can be detected using molecular analysis
for lung epithelial cell specific proteins or mRNA in lymph nodes. Detection of these
”micrometastases” using molecular techniques such as immunohistochemistry or reverse
transcriptase-polymerase chain reaction (rtPCR) has been shown to improve the detection
of metastatic epithelial cells in the lymph nodes and bone marrow of patients with
lung carcinoma [18]
[19]
[20]. Using this method, several markers have been identified such as telomerase [21] and KS Œ, which can accurately identify signatures of metastatic disease in cytopathology
negative lymph nodes [22]. These markers will be discussed in more details in this context. These studies
suggest that up to 50 % of patients with histologically normal lymph nodes have micrometastases
detected by molecular techniques. The presence of micrometastases has also been shown
to have a significant impact on long-term survival even among patients with pathologically
”normal” lymph nodes.
Gene marker expression and RT-PCR
Gene marker expression and RT-PCR
Altered gene expression associated with malignant transformation provides an opportunity
to identify the presence of malignant cells by detecting mRNA transcripts that would
otherwise not be present in lymph nodes. For example, cytokeratin (CK) genes are normally
expressed in epithelial but not lymphoid cells, so the presence of cytokeratin mRNA
in a lymph node suggests the presence of metastatic cells of epithelial origin. RT-PCR
is highly sensitive technique that allows for detection of rare gene transcripts in
tissue samples, and has been reported to be capable of detecting one cancer cell per
106 normal cells [23]
[24]. Furthermore, metastatic spread produces ectopic expression of tissue-specific genes.
Clear advantages of RT-PCR include minimal tissue requirements, sensitivity of detection,
and potential cost efficiency [18]
[25]
[26]
[27]
[28]
[29]. Studies with other cancers have shown that RT-PCR is capable of highly sensitive
detection of metastases. A good example of this is the application of this technology
to improve staging for patients with breast cancer [26]. Similar work with lung cancer has been limited. RT-PCR has been used to detect
muc1 transcripts (a mucopolysaccharide gene associated with respiratory epithelium) in
histologically negative mediastinal lymph nodes from patients with resected NSCLC
[18]. More recently, RT-PCR for p53 and K-ras was shown to improve detection of occult
lymph node metastasis compared to immunohistochemistry for cytokeratins [27].
Although yet to be definitely proven for lung cancer, there is considerable evidence
to suggest that molecular staging is clinically relevant. An obvious example is hematologic
malignancies, where molecular staging has had considerable success. This has been
used to monitor residual disease in the peripheral blood and thus guiding the need
for further therapy [28]
[29]
[30]
[31]. With melanoma, recent studies suggest that RT-PCR is more sensitive than immunohistochemistry
for detection of metastatic melanoma in sentinel lymph nodes [32].
Specific genes that are highly expressed in NSCLC have been identified recently. Muc1 was the first marker used for the detection of metastatic NSCLC, but there have been
concerns about the specificity of its detection by RT-PCR. With respect to muc1, we found significant overlap in mRNA expression between normal and malignant lymph
nodes, suggesting that muc1 lacks specificity in NSCLC. In contrast, the specificity of lunx appeared very high. This result is consistent with Iwao and colleagues [33], who were the first to show that lunx was highly expressed in NSCLC tumors but not in normal lymph nodes. These findings
provide evidence that lunx expression might serve as a useful molecular marker for detection of NSCLC lymph
node metastases.
Another molecular marker with very high specificity for NSCLC is KS1/4, a gene that encodes a glycoprotein expressed on epithelial cells [34]. The protein is recognized by the monoclonal antibody Ber-EP4. Immunohistochemical
staining with Ber-EP4 has shown that KS1/4 expression is specific for epithelial cells and is present on epithelial cancers
(carcinomas) [35]. Interestingly, antibodies against Ber-EP4 have shown promise in clinical trials
for colorectal cancer. In a study of 189 patients with resected stage III colorectal
cancer, treatment with edrecolomab (an antibody to Ber-EP4) resulted in a 32 % increase
in overall survival compared with no treatment (P < 0.01), and decreased the tumor
recurrence rate by 23 % (P < 0.04) [36]. These results raise the possibility that edrecolomab therapy may have activity
in NSCLC patients.
Telomerase expression
Telomerase expression
Telomerase is an enzyme expressed in the rapidly dividing cancerous cells of the lung.
Activation of the telomerase gene has been analyzed using a variety of techniques.
Most commonly, the activity of the telomerase is measured by using telomerase harvested
from the cell extract to add telomeric repeats to the end of a primer followed by
amplification of the product using the polymerase chain reaction (TRAP). However,
this technique requires approximately 103 cells [37]. Because of the small size of the samples, and potentially the few malignant cells
in samples obtained by EUS, human telomerase catalytic subunit gene (hTERT) and RT-PCR
have been used to detect molecular evidence of malignant lymph nodes in lung cancer
[21]. hTERT encodes a protein that is the rate-limiting determinant of the enzymatic
activity of human telomerase and expression appears to develop early in the course
of tumorigenesis [38]. In addition, expression of this gene has been shown to correlate with telomerase
activity [39]
[40], metastatic disease [41], and shorter survival in non-small cell lung cancer [40]
[42]
[43].
Evidence of telomerase over-expression was noted in both pathologically positive and
negative mediastinal lymph nodes from patients with NSCLC after EUS-FNA, and the absence
of telomerase expression in all normal control lymph nodes. Almost one third of pathologically
negative lymph nodes over-express telomerase. The clinical relevance of these findings
remains to be determined by survival studies, but the pattern of expression suggests
that analysis of hTERT expression may prove useful in detecting micrometastatic disease.
It does not appear that rtPCR ”positivity” in some histologically negative lymph nodes
is predictive of other histologically positive lymph nodes at the time of complete
surgical lymphadenectomy.
Telomerase represents a unique target for molecular staging in lung cancer. Not only
is telomerase expressed in the majority of lung cancers, but also by all histologic
types of lung cancer, and expression in non-small cell lung cancer has been shown
to be independently predictive of a worse prognosis [44]. However, recent studies in animal models of lung inflammation have put into question
the specificity of telomerase as marker of malignancy. Injury to rat lungs caused
by both silica and bleomycin have been shown to be associated with the development
of increased telomerase activity in lung fibroblasts [45]
[46], raising the question as to whether telomerase might be increased in lungs inflamed
by other agents such as cigarette smoke. Yet in two studies of patients with known
non-small cell lung cancer analysis of tumor normal pairs showed no evidence of telomerase
gene expression in any of the paired normal specimens (total of 218 samples for both
studies) [44]
[47]. Evaluation of lymph nodes for the presence of metastatic disease using telomerase
remains controversial. While high levels of telomerase activity in hyperplastic lymph
nodes and tonsils have been reported [48], others have found a significant difference in telomerase activity between benign
and malignant lymph nodes with high levels of activity in malignant and only low levels
of activity in benign nodes [49].
These data suggest that detection of hTERT gene expression in lymph nodes may identify
micrometastatic disease when normal cytologic and histologic techniques fail. Further
studies are needed to determine the clinical significance of expression of hTERT in
pathologically negative lymph nodes on survival, and whether chemo-radiotherapy is
beneficial to patients with rtPCR evidence but not histological evidence of mediastinal
lymph node metastases (rtPCR stage III).
EGFR gene: A new target
EGFR gene: A new target
The epidermal growth factor receptor family of genes (EGFR) encodes widely expressed transmembrane molecules that have been implicated in the
development and progression of cancer. The EGFR gene is frequently expressed in solid tumors, and this may correlate with a poor
clinical outcome in certain cancers [50].
Non-small-cell lung cancer frequently expresses EGFR [51]
[52]
[53], and for this reason, it is of considerable interest for clinical trials of inhibitors
of the tyrosine kinase of EGFR [54]. The kinase inhibitors erlotinib (Tarceva, OSI Pharmaceuticals, Melville, NY, USA)
and gefitinib (Iressa, AstraZeneca, Wilmington, DE, USA) have been studied most extensively
in clinical trials [54]
[55]
[56]. Somatic mutations in the region of EGFR that encodes the tyrosine kinase domain
of the receptor (exons 18 through 21) have been identified in lung cancer [57]
[58], and many studies suggest that they can be used to predict responsiveness to gefitinib
and erlotinib [57]
[58]
[59]
[60]
[61]
[62]
[63]. Risk factors for such mutations include female gender, Asians, and patients who
have never smoked.
A recent study evaluated the effect of EGFR expression in NSCLC on responsiveness
to erlotinib and survival [64]. Patients with EGFR mutation were found to be more responsive to this agent, though
it did not demonstrate a significant survival benefit compared to the placebo group.
Further studies are underway to confirm this effect on survival. It remains to be
determined if mutational analysis is necessary to identify patients in whom treatment
with EGFR inhibitors is appropriate.
Conclusion
Conclusion
There in increasing evidence that molecular diagnostic methods allow detection of
tumor specific genes in lung cancer specimens. The most promising methods include
real time reverse transcriptase PCR in conjunction with minimally invasive fine needle
aspiration samples of normal lymph nodes in lung cancer, patients. Promising markers
for detection of micrometastases to lymph nodes include KS Πand telomerase. Current
research is evaluating the effect of EGFR mutations on response to treatment and survival.
Several important issues remain unanswered though. Since overexpression alone is not
a definitive evidence of micrometastatic disease, the presence of these markers needs
to be correlated directly with clinical outcome such as disease recurrence and survival.
However, the most promising aspect of this technology is the broad applicability to
other tumors. Combining the ability to easily access tissue with EUS-FNA and modern
genomic and proteomic methods offers a rich area for further research and clinical
applications.
