At a Glance
Lung cancer is a major health problem in the United States and the world. According to the American Cancer Society’s annual statistical report, lung cancer is the most prevalent cancer in American men and women, with an estimated incidence of 220,000 new diagnoses in 2011. Furthermore, the case mortality rate of lung cancer is second only to pancreatic cancer, and approximately 85% of patients with lung cancer die of the disease. This combination of high frequency and high lethality makes lung cancer the leading cause of cancer-related deaths in both men and women, responsible for more deaths than breast, prostate, colon, and pancreatic cancers combined. Lung cancer is similarly a leading cause of cancer morbidity and mortality around the world.
Diagnosis is suspected by clinical findings, which can be somewhat nonspecific initially: cough, dyspnea, and pleuritic pain. X-rays show an area of consolidation with or without discrete mass lesion or regions of “ground glass” opacification. Higher-resolution imaging studies, such as CT or MRI, may show discrete mass lesions or enlargement of lymph nodes.
In some instances, lung tumors secrete active peptide hormones or trigger formation of antibodies that cross-react with normal tissues and present as autoimmune phenomena. Some patients present initially with signs and symptoms of one of these paraneoplastic syndromes, rather than those of primary lung disease.
Clinical laboratory testing has a very limited role in the diagnosis and management of lung cancer. Tumors that present with paraneoplastic syndromes may be suspected with the demonstration of increased concentrations of peptide hormones secreted by the tumor or by the demonstration of antibodies associated with paraneoplastic syndromes, such as the Eaton-Lambert Myasthenic syndrome. However, this is a minority of lung cancers.
Where laboratory diagnostics have made tremendous recent strides in lung cancer is in the use of molecular diagnostic tests to guide therapy selection. This began with the 2004 discovery of activating mutations in the epidermal growth factor receptor (EGFR) oncogene in a subset of lung adenocarcinomas, with further observation that tumors with these mutations are uniquely sensitive to pharmacologic blockage of the EGFR tyrosine kinase binding pocket. Patients with these mutations show dramatic clinical response to treatment with two ATP-mimetic tyrosine kinase inhibitors (TKIs), gefitinib, and erlotinib.
In advanced stage, median survival extends from approximately 6 months to approximately 2 years if patients with EGFR mutations are treated with these TKIs. Moreover, the TKIs are better tolerated, with fewer side effects, than conventional platinum-based chemotherapy. However, randomized controlled trials published in 2009 and 2010 showed that patients without EGFR mutations have poorer outcomes if they are treated with TKIs, rather than with platinum-based chemotherapy, establishing that TKIs should not be administered empirically to all patients and that EGFR mutation testing is essential for proper management.
These mutations are preferentially found in females, Asians, and patients who have never smoked tobacco products, but these clinical factors are insufficiently sensitive for use as screening criteria, and patients of any gender, ethnicity, or smoking history should be tested for EGFR mutations. Histologic selection is limited to exclusion of pure squamous cell carcinomas and pure small cell carcinomas, which have only shown EGFR mutations in rare case reports. Tumors with any adenocarcinoma component, whether pure or mixed, should be tested.
Two mutations account for about 90% of cases: a somewhat variable in-frame deletion in exon 19 that typically involves most or all of a DNA segment at codons 746-750 and a single point mutation that changes a leucine to arginine at codon 858 (i.e., L858R mutation) in codon 21. Other mutations accounting for the remaining 10% of mutant tumors include variable mutations at codon 719 in exon 18 (G719C, G719S, G719A) and a series of uncommon mutations in exon 20 that include variable in-frame insertion/duplications around codon 770 and two point mutations, S768I and T790M. Importantly, these mutations in exon 20, although oncogenic, confer resistance to the TKIs, rather than sensitivity. The T790M, in particular, is commonly acquired in patients after successful response to therapy and is the most common cause of eventual relapse.
EGFR mutations are detected in about 20% of lung adenocarcinomas (about 10% of all lung cancers), and earlier studies have shown that KRAS mutations (point mutations in codons 12, 13, 61) occur in 30% of lung adenocarcinomas. These mutations are, essentially, mutually exclusive so that approximately half of lung adenocarcinomas are driven by one of these two oncogenes. Unfortunately, no effective treatment option exists for KRAS mutant lung cancers, and KRAS testing is of limited use in lung cancer, except as a possible “rule out” for EGFR testing in a multi-test algorithm. Because KRAS is technically easier to test than EGFR, some laboratories may screen EGFR test requests with a quick KRAS test first.
Another subtype of lung cancer that has a successful targeted therapy contains a chromosomal abnormality, inversion of part of chromosome 2 that results in dysregulation of the ALK (anaplastic lymphoma kinase) gene, typically by fusion with EML4. Early trials have shown about 60% of patients with ALK rearrangements, as detected by fluorescence in situ hybridization (FISH), respond to a targeted inhibitor of the ALK kinase, crizotinib. Although the frequency of ALK rearrangement in lung adenocarcinoma is only about 5% and phase III trials have yet been completed, the early trials were sufficiently promising that the U.S. Food and Drug Administration (FDA) approved crizotinib for lung adenocarcinomas with evidence of ALK rearrangement by FISH in August, 2011.
Patients with ALK rearrangements are, like patients with EGFR mutations, typically nonsmokers. However, gender and ethnicity seem less important as predictive factors for ALK rearrangements, whereas younger age is a more important risk factor. However, as with EGFR, the only risk factor of sufficient sensitivity to determine which patients to test is histology. Like EGFR mutations, ALK rearrangements are not seen in squamous or small cell carcinomas of the lung.(Table 1)
|Mutation||Mutation Type||Mutation Expression|
|EGFR||Wild type||Resistant to EGFR TKI|
|EGFR exon 18||G719 mutation||Sensitive to EGFR TKI|
|EGFR exon 19||Deletion involving ELREA (codons 746-75)||Sensitive to EGFR TKI|
|EGFR exon 20||T790M or insertion||Resistant to EGFR TKI|
|EGFR exon 21||L858R mutaion||Sensitive to EGFR TKI|
|KRAS||G12 or G13 mutation||Resistant to EGFR TKI|
|ALK FISH||ALK rearrangement in <15% of cells||Resistant to ALK TKI|
|ALK FISH||ALK rearrangement in >15% of cells||Sensitive to ALK TKI|
Are There Any Factors That Might Affect the Lab Results? In particular, does your patient take any medications – OTC drugs or Herbals – that might affect the lab results?
Numerous techniques can be employed to detect EGFR mutations, all beginning with the polymerase chain reaction (PCR). The PCR products can be analyzed by Sanger dideoxyterminator sequencing, Pyrosequencing-by-synthesis, probe hybridization, single nucleotide extension with fluorescent or mass spectrometric detection, real-time PCR, melt curve analysis, restriction digestion, denaturing HPLC, and other techniques. A comprehensive review of these is beyond the scope of this chapter; however, a few points are worth mentioning.
The gold-standard technique is Sanger sequencing. This was the predicate method used in the studies that demonstrated the clinical value of the diagnostic assay. This method is versatile and able to detect mutations in all four of the critical exons (18-21). However, analytical sensitivity is poor, and samples with admixed benign and reactive elements, such as inflammatory cells, fibroblasts, and adjacent normal lung tissue, can give false-negative results.
In general, Sanger sequencing performs well when the malignant cells outnumber these benign elements, but, when the content of cancer cells falls below 50% of total cellularity, this method can begin to fail. This often necessitates manual dissection of samples by a pathologist or a specially trained technologist prior to analysis, which extends the turnaround time and is still often insufficient, particularly for small biopsies or cytology samples that are the mainstay of diagnostic samples obtained in advanced stage lung cancer.
More sensitive methods are available, such as allele-specific PCR or sequencing after PCR with suppression of wild-type amplification (e.g., with peptide nucleic acid (PMA) or locked nucleic acid (LNA) clamps) or laser capture micro-dissection of individual cancer cells. These methods can detect mutations in samples with cancer cell contents of 10% or lower, sometimes less than 1%. However, although these methods enable testing of many of the small samples that would be insufficient for unmodified Sanger sequencing, they also confer a risk of false-positive results.
Novel mutations of uncertain significance have been reported with these methods, often unable to be confirmed on repeat analysis. Artifactual mutations are a risk as well. Furthermore, these methods can detect minor subpopulations with a mutation in the absence of overall mutant background, and the clinical significance of this is unclear. This is particularly confusing if an untreated patient is discovered to have a sensitizing EGFR mutation alongside a T790M resistance mutation.
ALK testing is typically performed by FISH, and a commercial “split apart” assay (Vysis) initially developed for anaplastic lymphoma (the disease in which ALK abnormalities were first discovered) was used as the predicate device in the early clinical trials showing the value of this marker and has been approved by the FDA as a companion diagnostic to select patients for crizotinib therapy. The rearrangement is a bit subtle, and an experienced interpreter is recommended for the differentiation between a true split (positive) signal due to chromosomal rearrangement and narrow split signals seen in the absence of a true rearrangement.
Standard interpretive criteria are that a true positive signal contains a split that measures at least two probe diameters of separation. Immunohistochemistry has shown some promise as a surrogate marker for ALK rearrangement, but antibodies are not commercially available as of the summer of 2011. RT-PCR shows some promise as well, although the diversity of rearrangement breakpoints and occasional variant rearrangements makes this technically challenging and somewhat less practical.
Both tests are typically performed on formalin-fixed paraffin-embedded tissue samples. Alternative fixatives may inhibit the tests, particularly those containing heavy metals (e.g., Zenker’s solution). DNA is damaged by tissue treatments with strong acids, such as Bouin’s fixative, or decalcifying solutions used to soften bone samples, and these treatments interfere with these tests. Although blood testing is uncommon in this setting, heparin is a potent inhibitor of PCR and interferes with EGFR testing. However, the tissue concentrations achieved by patients receiving heparin are insufficient to cause a problem, and administration of heparin to a patient is not a contraindication to testing.
What Lab Results Are Absolutely Confirmatory?
Definitive diagnosis of lung cancer requires microscopic demonstration of cancer cells in a fluid or tissue sample. Unlike many other cancers, lung cancer commonly presents in advanced stage, and the diagnostic sample is often small: bronchoscopic or transthoracic needle biopsy, pleural fluid cytology, and biopsy of metastatic sites, including lymph nodes, predominates. Advanced stage disease is often surgically incurable, so large tumor resections are somewhat uncommon, relative to other common cancers, such as colon, prostate, and breast.
Histologically, the overwhelming majority of lung cancers are carcinomas, classically divided into two broad categories: small cell carcinoma and nonsmall cell carcinoma (NSCLC). Small cell carcinomas account for about 15% of lung carcinomas. They are typically centrally located and show evidence of neuroendocrine differentiation, with secretory granules demonstrated by electron microscopy (unnecessary and essentially obsolete in a modern pathology lab), immunoreactivity for neuropeptides, such as chromogranin or synaptophysin, and somewhat uniform rounded nuclei with sparse cytoplasm, finely stippled “salt-and-pepper” chromatin, and molding of nuclei at boundaries between adjacent cells.
These tumors are resistant to chemotherapy and are typically treated with radiation therapy. They are aggressive and have poor survival. Small cell carcinomas can be hormonally active and are associated with several paraneoplastic syndromes.
Approximately 85% of lung carcinomas are of the NSCLC type, roughly evenly divided between adenocarcinomas and squamous cell carcinomas, with additional rare subtypes (e.g., large cell carcinoma, salivary duct-like carcinoma) accounting for a small remaining fraction. Not uncommonly, NSCLC’s contain mixed histology, with patterns of both adenocarcinoma and squamous carcinoma; sometimes regions of small cell histology are also present.
Pure squamous cell carcinomas consist of polygonal cells with distinct cytoplasmic borders and rudimentary intercellular desmosomes. Extracellular keratin may be present, often in whorls. Tumor cells infiltrate adjacent tissue in jagged nests of solid cancer. Immunohistochemistry demonstrates high molecular weight keratins and p63, without expression of TTF1.
By contrast, pure adenocarcinoma consists of cuboidal-columnar cells that may be arranged in acinar or tubular patterns, although solid growth can also occur. Intra- and extracellular mucin may be present. Immunohistochemistry demonstrates low molecular weight keratins and nuclear expression of TTF-1, without expression of p63.
A distinct subtype of adenocarcinoma is the so-called bronchioloalveolar carcinoma (BAC) of the lung in which cancer cells spread throughout the lining of alveolar spaces (“lepidic spread”) without infiltrating into the parenchyma. As such, it is considered an adenocarcinoma in-situ and has a favorable prognosis when compared to invasive adenocarcinomas. Special stains for elastin or basement membrane collagens may be needed to convincingly demonstrate that a BAC does not contain regions of microscopic invasion of parenchyma.
With regard specifically to EGFR-mutant and ALK-rearranged lung adenocarcinomas, the definitive tests are Sanger sequencing and FISH, respectively, as mentioned previously.
Mortality in lung cancer is very closely associated with disease stage. Early stage disease (confined to lung or with spread to local lymph nodes) is treated surgically, with 5-year survival rates in the 30-50% range. Advanced stage lung cancer (spread to contralateral or extrathoracic nodes or to distant sites) has a much grimmer prognosis, with 5-year survival rates less than 10%. Accordingly, early diagnosis should be accompanied by imaging studies to assess the possibility of metastatic disease.
Beyond EGFR, KRAS, and ALK, other oncogene mutations have been reported in lung adenocarcinomas, although each are relatively uncommon, accounting for 1-5% of the total, and none have established therapeutic implications. Mutually exclusive activating mutations have been found in BRAF, ERBB2, AKT1, NRAS, and MEK, in total accounting for 10-15% of lung adenocarcinomas. MET amplification is found in approximately 20% of EGFR mutant cancers that relapse on therapy. Targeted therapies designed to inhibit these mutant proteins are in development or clinical trials, but are neither standard of care as of the summer of 2011 nor testing for these mutations.
Far less is known about targeted therapies for squamous cell carcinomas and small cell carcinomas, which are still managed by cytotoxic chemotherapy or radiation therapy, respectively. FGFR1 amplification is a common finding in squamous cell carcinoma, as are PIK3CA mutation and mutation in FGFR2. However, these have yet to garner application as either diagnostic or therapeutic targets outside the research lab.
Copyright © 2017, 2013 Decision Support in Medicine, LLC. All rights reserved.
No sponsor or advertiser has participated in, approved or paid for the content provided by Decision Support in Medicine LLC. The Licensed Content is the property of and copyrighted by DSM.
- At a Glance
- Are There Any Factors That Might Affect the Lab Results? In particular, does your patient take any medications - OTC drugs or Herbals - that might affect the lab results?
- What Lab Results Are Absolutely Confirmatory?