Larissa V. Furtado, Jay L. Hess, and Bryan L. Betz
APPLICATIONS OF MOLECULAR DIAGNOSTICS IN ONCOLOGY
Molecular diagnostics is increasingly impacting a number of areas of cancer care delivery including diagnosis, prognosis, in predicting response to particular therapies, and in minimal residual disease monitoring. Each of these depends on detection or measurement of one or more disease-specific molecular biomarkers representing abnormalities in genetic or epigenetic pathways controlling cellular proliferation, differentiation, or cell death (Table 3.1). In addition, molecular diagnostics is beginning to play a role in predicting host metabolism of drugs—for example, in predicting fast versus slow thiopurine metabolizers using polymorphisms in the thiopurine methyltransferase (TPMT) allele and in use in dosing patients with thiopurine drugs.1 Molecular diagnostics has also had a major impact on assessing an engraftment after bone marrow transplantation and in tissue typing for bone marrow and solid organ transplantation.
The ideal cancer biomarker is only associated with the disease and not the normal state. The utility of the biomarker largely depends on what the clinical effect the biomarker predicts for, how large the effect is, and how strong the evidence is for the effect. For clinical application, biomarkers need a high level of analytic validity, clinical validity, and clinical utility. Analytic validity refers to the ability of the overall testing process to accurately detect and, in many cases, measure the biomarker. Clinical validity is the ability of a biomarker to predict a particular disease behavior or response to therapy. Clinical utility, arguably the most difficult to assess, addresses whether the information available from the biomarker is actually beneficial for patient care.
Biomarkers can take many forms including chromosomal translocations and other chromosomal rearrangements, gene amplification, copy number variation, point mutations, single nucleotide polymorphisms, changes in gene expression (including micro RNAs), and epigenetic alterations. Most biomarkers in widespread use represent either gain of function or loss of function alterations in key signaling pathways. Those that occur early and at a high frequency in tumors tend to be driver mutations, whose function is important for the cancer cell’s proliferation and/or survival. These are particularly useful as biomarkers because they often represent important therapeutic targets. However, cancer cells accumulate many genetic alterations, called passenger mutations, which tend to occur at a lower frequency overall and in a subset of a heterogeneous population of tumor cells that may contribute to the cancer phenotype but are not absolutely essential.2 Distinguishing passenger from driver mutations using various functional assays has become a major focus of translational research in cancer. The same biomarker may have utility in a variety of settings. For example, the detection of the BCR-ABL1 translocation, pathognomonic for chronic myelogenous leukemia (CML), is used for establishing the diagnosis, for the selection of therapy, and for monitoring for minimal residual disease during and after therapy.
Some of the most heavily used genetic biomarkers in cancer, particularly in hematologic malignancies, are chromosomal translocations. For certain diseases such as CML, detection of the BCR-ABL1translocation or in Burkitt lymphoma the immunoglobulin gene-MYC translocation is required, according to current World Health Organization (WHO) guidelines, to make the diagnosis. Identification of translocations is important in the diagnosis and subtyping of acute leukemias (e.g., detection of PML-RARA and variant translocations in acute promyelocytic leukemia) and is also extremely important for the diagnosis of sarcomas such as Ewing sarcoma. The discovery of chromosomal translocations, such as the TMPRSS-ETS in prostate cancer and ALK translocations in non–small-cell lung cancer, portends an importance of detecting translocations in solid tumors.3 Chromosomal translocations, especially for hematologic malignancies, have been traditionally detected by classical karyotyping. This approach has limitations; in particular, it requires viable, dividing cells, which are often not readily available from solid tumor biopsies. In addition, a significant proportion of chromosomal translocations are not detectable by conventional karyotyping. For example, 5% to 10% of CML cases lack detectable t(9;22) by G banding. Such “cryptic” translocations require other approaches for detection, which are to be discussed, including fluorescent in situ hybridization (FISH), polymerase chain reaction (PCR), as well as nucleic acid sequencing-based methods.
In certain settings, it can be helpful to detect if a population of cells is clonal. For example, in some lymphoid infiltrates, the cells are well differentiated and it can be difficult to determine whether these represent a reactive or neoplastic infiltrate. If dispersed, cells are available and these could be analyzed by flow cytometer to detect whether a monotypic population expressing either immunoglobulin kappa or lambda light chains is present. In theory, immunohistochemical staining (IHC) for immunoglobulin light chains could be used to assess clonality; however, in practice this is done with more sensitivity using RNA in situ hybridization for immunoglobulin kappa and lambda light chain transcripts. The most sensitive way to detect clonality in a B-cell population is to analyze the size of the break point cluster region that arises as a result of VDJ recombination by PCR. Reactive B cells will show a distribution in the size of the VDJ recombination for the IGH or IGK or IGL, whereas clonal cells will show a predominant band that represents the size of the VDJ region of the dominant clone. Similarly, sometimes it can be difficult to distinguish neoplastic from reactive T-cell infiltrates. Given the large number of T-cell antigen receptors, it is not as simple to detect clonality by IHC or flow cytometry in T-cell proliferations. One approach is to use aberrant loss of T-cell antigen expression to aid in the diagnosis of T-cell neoplasms. Another is to detect clonal rearrangement of the VDJ region of the T-cell receptor gamma (TCRγ) gene, which can be done by PCR on both fresh and formalin-fixed paraffin-embedded (FFPE) tissue.
Gene amplification is another important mechanism in cancer that has been found to have high utility in a subset of cancers. MYCN amplification occurs in approximately 40% of undifferentiated or poorly differentiated neuroblastoma subtypes,4,5 either appearing as double minute chromosomes or homogeneously staining regions. MYCN amplification is a very strong predictor of poor outcomes, particularly in patients with localized (stage 1 or stage 2) disease or in infants with stage 4S metastatic disease, where fewer than half of patients survive beyond 5 years.6
Use of other chromosome abnormalities has been largely limited to the diagnosis and prognostication of hematologic disorders. Roughly half of all myelodysplastic disorders show cytogenetically detectable chromosomal abnormalities, such as monosomy 5 or 7, partial chromosomal loss (5q-, 7q-), or complex chromosomal abnormalities. Certain abnormalities in isolation (e.g., 5q-) have a favorable prognosis, whereas many others (e.g., “complex” karyotypes with three or more abnormalities) carry a worse prognosis. Differences in ploidy have proven to be useful predictors in pediatric acute lymphocytic leukemia (ALL), with hyperdiploid cases (>50 chromosomes) showing a distinctly more favorable course compared with hypodiploid or near diploid cases.7 Overall, DNA ploidy can be assessed by flow cytometry. Specific chromosomal copy number alterations can be detected by conventional karyotyping, array hybridization methods, or FISH.
Copy number variation (CNV) represents the most common type of structural chromosomal alteration. Regions affected by CNVs range from approximately 1 kilobase to several megabases that are either amplified or deleted. It is estimated that about 0.4% of the genomes of healthy individuals differ in copy number.8 CNVs resulting in deletion of genes such as BRCA1, BRCA2, APC, mismatch repair genes, and TP53 have been implicated in a wide range of highly penetrant cancers.9,10 CNVs can be detected by a variety of means including FISH, comparative or array genomic hybridization, or virtual karyotypingusing single nucleotide polymorphism (SNP) arrays. Increasingly, CNV is detected using next-generation sequencing.
Large-scale sequencing of tumors has identified many mutations that are of potential prognostic and therapeutic significance. As will be discussed further, a wide range of strategies is available for the detection of point mutations (Fig. 3.1). It is important to recognize that many nucleotide variations occur at any given allele in populations. Formally, the term polymorphism is used to describe genetic differences present in ≥1% of the human population, whereas mutation describes less frequent differences. However, in practice, polymorphism is often used to describe a nonpathogenic genetic change, and mutation a deleterious change, regardless of their frequencies.
Mutations can be classified according to their effect in the structure of a gene. The most common of these disease-associated alterations are single nucleotide substitutions (point mutations); however, many deletions, insertions, gene rearrangements, gene amplification, and copy number variations have been identified that have clinical significance. Point mutations may affect promoters, splicing sites, or coding regions. Coding region mutations can be classified into three kinds, depending on the impact on the codon: missense mutation, a nucleotide change leads to the substitution of an amino acid to another; nonsense mutation, a nucleotide substitution causes premature termination of codons with protein truncation; and silent mutation, a nucleotide change does not change the coded amino acid.
Loss of function mutations, either through point mutations or deletions in tumor suppression genes such as APC and TP53, are the most common mutations in cancers. Tumor suppression genes require two-hit (biallelic) mutations that inactivate both copies of the gene in order to allow tumorigenesis to occur. The first hit is usually an inherited or somatic point mutation, and the second hit is assumed to be an acquired deletion mutation that deletes the second copy of the tumor suppression gene. Promoter methylation of tumor suppressor genes is an alternative route to tumorigenesis that, to date, has not been commonly employed for molecular diagnostics.
Oncogenes originate from the deregulation of genes that normally encode for proteins associated with cell growth, differentiation, apoptosis, and signal transduction (proto-oncogenes, [e.g., BRAF and KRAS]).Proto-oncogenes generally require only one gain of function or activating mutation to become oncogenic. Common mutation types that result in proto-oncogene activation include point mutations, gene amplifications, and chromosomal translocations. One example is mutations in the epidermal growth factor receptor (EGFR) that occur in lung cancer, which are almost exclusively seen in nonmucinous bronchoalveolar carcinomas. Somatic mutations of EGFR constitutively activate the receptor tyrosine kinase (TK). Importantly, responsiveness of tumors harboring these mutations to the inhibitor gefitinib is highly coordinated with a mutation of the EGFR TK domain.11,12
One of the challenges with using mutations as biomarkers is that there can be many nucleotide alterations that affect a given gene. For example, there are over 100 known different point mutations in EGFRreported in non–small-cell lung cancer. Many of these mutations occur at low frequency and have an unknown clinical significance.13,14 Another important concept is that the same driver oncogene may be mutated in a variety of different tumors. For example, lung cancers harbor a number of other different alterations that are common in other solid tumors, which generally occur at lower frequencies than EGFRmutations such as KRAS, BRAF, and HER2. Some lung cancers have translocations involving the ALK kinase gene. ALK, interestingly, is also activated by point mutations in a neuroblastoma as by translocation in anaplastic large cell lymphoma (Fig. 3.2). Hence, a therapy targeted to a genetic alteration in one cancer may demonstrate efficacy in other cancers.
The detection of mutations is also important in the evaluation of chemotherapy resistance. Roughly a third of CML patients are resistant to the frontline ABL1 kinase inhibitor imatinib, either at the time of initial treatment or, more commonly, secondarily. In cases of primary failure or secondary failure, over 100 different ABL1 mutations have been identified, including particularly common ones such as T315I and P loop mutations. While some mutations, such as Y253H, respond to second generation TK inhibitors (TKI), others, such as the T315I mutation, are noteworthy because they confer resistance not only to imatinib, but also to nilotinib and dasatinib.
Mutations are also used as important predictive biomarkers (Table 3.1). Two of the most notable examples are the use of the BRCA1 and BRCA2 mutation analysis for women with a strong family history of breast cancer. Over 200 mutations (loss of function point mutations, small deletions, or insertions) occur in BRCA genes, which are distributed across the genes necessitating full sequencing for their detection. The overall prevalence of these occur in about 0.1% of the general population.15,16 The lifetime risk of breast cancer for women carrying BRCA1 mutations is in the range of 47% to 66%, whereas for BRCA2mutations, it is in the range of 40% to 57%.17,18 In addition, the risk of other tumors including ovarian, fallopian, and pancreatic cancer is also increased. Detection of BRCA1 and BRCA2 mutations is, therefore, important for cancer prevention and risk reduction.
THE CLINICAL MOLECULAR DIAGNOSTICS LABORATORY: RULES AND REGULATIONS
Laboratories in the United States that perform molecular diagnostic testing are categorized as high-complexity laboratories under the Clinical Laboratory Improvement Amendments of 1988 (CLIA).19 The CLIA program sets the minimum administrative and technical standards that must be met in order to ensure quality laboratory testing. Most laboratories in the United States that perform clinical testing in humans are regulated under CLIA. CLIA-certified laboratories must be accredited by professional organizations such as the Joint Commission, the College of American Pathologists, or another agency officially approved by the Centers for Medicare & Medicaid Services (CMS), and must comply with CLIA standards and guidelines for quality assurance. Although the regulation of laboratory services is in the U.S. Food and Drug Administration’s (FDA) jurisdiction, the FDA has historically exercised enforcement discretion. Therefore, FDA approval is not currently required for clinical implementation of molecular tests as long as other regulations are met.20,21
SPECIMEN REQUIREMENTS FOR MOLECULAR DIAGNOSTICS
Samples typically received for molecular oncology testing include blood, bone marrow aspirates and biopsies, fluids, organ-specific fresh tissues in saline or tissue culture media such as Roswell Park Memorial Institute (RPMI), FFPE tissues, and cytology cell blocks. Molecular tests can be ordered electronically or through written requisition forms, but never through verbal requests only. All samples submitted for molecular testing need to be appropriately identified. Sample type, quantity, and specimen handling and transport requirements should conform to the laboratory’s stated requirements in order to ensure valid test results.
Blood and bone marrow samples should be drawn into anticoagulated tubes. The preferred anticoagulant for most molecular assays is ethylenediaminetetraacetic acid (EDTA; lavender). Other acceptable collection tubes include ACD (yellow) solutions A and B. Heparinized tubes are not preferred for most molecular tests because heparin inhibits the polymerase enzyme utilized in PCR, which may lead to assay failure. Blood and bone marrow samples can be transported at ambient temperature. Blood samples should never be frozen prior to separation of cellular elements because this causes hemolysis, which interferes with DNA amplification. Fluids should be transported on ice. Tissues should be frozen (preferred method) as soon as possible and sent on dry ice to minimize degradation. Fresh tissues in RPMI should be sent on ice or cold packs. Cells should be kept frozen and sent on dry ice; DNA samples can be sent at ambient temperature or on ice.
For FFPE tissue blocks, typical collection and handling procedures include cutting 4 to 6 microtome sections of 10-micron thickness each on uncoated slides, air-drying unstained sections at room temperature, and staining one of the slides with hematoxylin and eosin (H&E). A board-certified pathologist reviews the H&E slides to ensure the tissue block contains a sufficient quantity of neoplastic tumor cells, and circles an area on the H&E slide that will be used as a template to guide macrodissection or microdissection of the adjacent, unstained slides. The pathologist also provides an estimate of the percentage of neoplastic cells in the area that will be tested, which should exceed the established limit of detection (LOD) of the assay.
MOLECULAR DIAGNOSTICS TESTING PROCESS
The workflow of a molecular test begins with receipt and accessioning of the specimen in the clinical molecular diagnostics laboratory followed by extraction of the nucleic acid (DNA or RNA), test setup, detection of analyte (e.g., PCR products), data analysis, and result reporting to the patient medical record (Fig. 3.3).
An extraction of intact, moderately high-quality DNA is essential for molecular assays. For DNA extraction, the preferred age for blood, bone marrow, and fluid samples is less than 5 days; for frozen or fixed tissue, it is indefinite; and for fresh tissue, it is overnight. Although there is no age limit for the use of a fixed and embedded tissue specimen for analysis, older specimens may yield a lower quantity and quality of DNA. Because RNA is significantly more labile than DNA, the preferred age for blood and bone marrow is less than 48 hours (from time of collection). Tissue samples intended for an RNA analysis should be promptly processed in fresh state, snap frozen, or preserved with RNA stabilizing agents for transport.
Dedicated areas, equipment, and materials are designated for various stages of DNA and RNA extraction procedures. DNA and RNA isolation can be done by manual or automated methods. Currently, most clinical laboratories employ commercial protocols based on liquid- or solid-phase extractions. Nucleated cells are isolated from biological samples prior to nucleic acid extraction. White blood cells (WBC) can be isolated from blood and bone marrow samples by different methods. One method involves lysing the red blood cells with an ammonium chloride solution, which yields the total WBC population and other nucleated cells present. Another method involves a gradient preparation with a Ficoll solution, which yields the mononuclear cell population only. Sections of FFPE tissue blocks are prepared for DNA extraction by first removing the paraffin and disrupting the cell membranes with proteinase K digestion. Fresh and frozen tissues also undergo proteinase K digestion prior to nucleic acid extraction. DNA isolation protocols consist of several steps, including cell lysis, DNA purification by salting out the proteins and other debris (nonorganic method), or by solvent extractions of the proteins with phenol and chloroform solutions (organic method). The DNA is then precipitated out of the solution with isopropanol or ethanol. The pellet is washed with 70% to 80% ethanol and then solubilized in buffer, such as Tris-EDTA solution. Proteinase K can be added to assist in the disruption and to prevent nonspecific degradation of the DNA. RNase is sometimes added to eliminate contaminating RNA. The DNA yield is quantitated spectrophotometrically, and the DNA sample integrity is visually checked, if necessary, on an agarose gel followed by ethidium bromide staining. Intact DNA appears as a high–molecular-weight single band, whereas degraded DNA is identified as a smear of variably sized fragments. After extraction, the DNA is stored at 4°C prior to use in a PCR assay, and is then stored at –70°C after completion of the assay. Because the DNA extracted from formalin-fixed tissue is degraded to a variable extent, an analysis of the extraction product by gel electrophoresis is not informative. Yield and integrity of the extracted DNA is best assessed by an amplification control to ensure that the quality and quantity of input DNA is adequate to yield a valid result.
RNA isolation steps are similar to the ones described previously for DNA extraction. However, RNA is inherently less stable than DNA due to its single-strand conformation and susceptibility to degradation by RNase, which is ubiquitous in the environment. To ensure preservation of target RNA, special precautions are required, including the use of diethylpyrocarbonate (DEPC) water in all reagents used in RNA procedures, and special decontamination of work area and pipettes to prevent RNase contamination. The extracted RNA is usually degraded to a variable extent so that the analysis of the extraction product by gel electrophoresis is not informative. The quality of the RNA and its suitability for use in a reverse transcriptase polymerase chain reaction (RT-PCR)–based assay is assessed most appropriately by the demonstration of a positive result in an assay designed to detect the RNA transcripts for a “housekeeping gene,” such as ABL1 or GAPDH. Any RNA sample in which the 260/280-nm absorption ratio is below 1.9 or greater than 2.0 may contain contaminants and must be cleaned prior to analysis.
Following nucleic acid extraction, the assay is set up according to written procedures established during validation/verification of the assay by qualified laboratory staff. Dedicated areas, equipment, and materials are designated for various stages of the test (e.g., extraction, pre-PCR and post-PCR for amplification-based assays). For each molecular oncology test, appropriated positive and negative control specimens are included to each run as a matter of routine quality assessment. A no template (blank) control, containing the complete reaction mixture except for nucleic acids, is also included in amplification-based assays to evaluate for amplicon contamination in the assay reagents that may lead to inaccurate results. The controls are processed in the same manner as patient samples to ensure that established performance characteristics are being met for each step of the assay (extraction, amplification, and detection). All assay controls and overall performance of the run must be examined prior to interpretation of sample results. Following acceptance of the controls, results are electronically entered into reports. The final report is reviewed and signed by the laboratory director or a qualified designee who meets the same qualifications as the director, as defined by CLIA (see previous).
TECHNOLOGIES
Several traditional and emerging techniques are currently available for mutation detection in cancer (Table 3.2). In the era of personalized medicine, molecular oncology assays are rapidly moving from a mutational analysis of single genes toward a multigene panel analysis. As the number of “actionable” mutations such as ALK, EGFR, BRAF, and others increase, the use of next-generation sequencing platforms is expected to become much more widespread. Both traditional and emerging testing approaches have advantages and disadvantages that need to be balanced before a test platform is implemented into practice.
An important consideration when adding a new oncology test in the clinical laboratory menu is to define the intended use of the assay (e.g., diagnosis, prognosis, prediction of therapy response). The clinical utility of the assay, appropriate types of specimens, the spectrum of possible mutations that can be found in the genomic region of interest, and available methods for testing should also be determined. The laboratory director and ordering physicians should also discuss the estimated test volume, optimal reporting format, and required turnaround time for the proposed new test.21–23
Polymerase Chain Reaction
Polymerase chain reaction (PCR)24,25 is widely used in all molecular diagnostics laboratories for the rapid amplification of targeted DNA sequences. The reaction includes the specimen template DNA, forward and reverse primers (18 to 24 oligonucleotides long), Taq DNA polymerase, and each of the four nucleotides bases (dATP, dTTP, dCTP, dGTP). During PCR, selected genomic sequences undergo repetitive temperature cycling (sequential heat and cooling) that allows for denaturation of double-stranded DNA template, annealing of the primers to the targeted complementary sequences on the template, and extension of new strands of DNA by Taq polymerase from nucleotides, using the primers as the starting point. Each cycle doubles the copy number of PCR templates for the next round of polymerase activity, resulting in an exponential amplification of the selected target sequence. The PCR products (amplicons) are detected by electrophoresis or in real-time systems simultaneously to the amplification reaction (see real-time PCR, which follows).
PCR is specifically designed to work on DNA templates because the Taq polymerase does not recognize RNA as a starting material. Nonetheless, PCR can be adapted to RNA testing by including a reverse transcription step to convert a RNA sequence into its cognate cDNA sequence before the PCR reaction is performed (see reverse-transcription PCR, which follows). Multiplex PCR reactions can also be designed with multiple primers for simultaneous amplification of multiple genomic targets. PCR is a highly sensitive and specific technique that can be employed in different capacities for the detection of point mutations, small deletions, insertions and duplications, as well as gene rearrangements and clonality assessment. Limits of detection can reach 0.1% mutant allele or lower, which is important for the detection of somatic mutations in oncology because tumor specimens are usually composed of a mixture of tumor and normal cells. Reverse transcription PCR can also be used for the relative quantification of target RNA in minimal residual disease testing, such as BCR-ABL1 transcripts in CML. Another advantage of PCR is its ability to amplify small amounts of low quality FFPE-derived DNA. However, applications of PCR can be limited because it cannot amplify across large or highly repetitive genomic regions. Also, the PCR reaction can be inhibited by heparin or melanin if present in the extracted DNA, which may lead to assay failure. Finally, the risk of false positives due to specimen or amplicon contamination is an important issue when using PCR-based techniques; therefore, stringent laboratory procedures, as described previously, are used to minimize contamination. With the exception of hybridization assays, such as fluorescence in situ hybridization and genomic microarrays, PCR is the necessary initial step in all current molecular oncology assays.
Targeted Mutation Analysis Methods
Real-Time PCR (q-PCR)
In real-time PCR (q-PCR), the polymerase chain reaction is performed with a PCR reporter that is usually a fluorescent double-stranded DNA binding dye or a fluorescent reporter probe. The intensity of the fluorescence produced at each amplification cycle is monitored in real time, and both quantification and detection of targeted sequences is accomplished in the reaction tube as the PCR amplification proceeds.
The intensity of the fluorescent signal for a given DNA fragment (wild type or mutant) is correlated with its quantity, based on the PCR cycle in which the fluorescence rises above the background (crossing threshold [Ct] or crossing point [Cp]).26 The Ct value can be used for qualitative or quantitative analysis. Qualitative assays use the Ct as a cutoff for determining “presence” or “absence” of a given target in the reaction. A qualitative analysis by q-PCR is particularly useful for a targeted detection of point mutations that are located in mutational hotspots. Examples include the JAK2 V617F mutation, which is located within exon 14, and is found in several myeloproliferative neoplasms (polycythemia vera, essential thrombocythemia, and primary myelofibrosis),27 and the BRAF V600E,28 which is located within exon 15, and is found in various cancer types including melanomas and thyroid and lung cancers.
For a quantitative analysis, the Ct of standards with known template concentration is used to generate a standard curve to which Ct values of unknown samples are compared. The concentration of the unknown samples is then extrapolated from values from the standard curve. The quantity of amplicons produced in a PCR reaction is proportional to the prevalence of the targeted sequence; therefore, samples with a higher template concentration reaches the Ct at earlier PCR cycles than one with a low concentration of the amplified target. Quantitative q-PCR has high analytical sensitivity for the detection of low mutant allele burden. For that reason, this method has been widely utilized for monitoring minimal residual disease.
Allele-Specific PCR
Allele-specific PCR (AS-PCR) is a variant of conventional PCR. The method is based on the principle that Taq polymerase is incapable of catalyzing chain elongation in the presence of a mismatch between the 3′ end of the primer and the template DNA. Selective amplification by AS-PCR is achieved by designing a forward primer that matches the mutant sequence at the 3′ end primer. A second mismatch within the primer can be introduced at the adjacent -1 or -2 position to decrease the efficiency of mismatched amplification products. This will minimize the chance of amplifying and, therefore, detecting the wild-type target. AS-PCR is usually performed as two PCR reactions: one employing a forward primer specific for the mutant sequence, the other using a forward primer specific for the correspondent wild-type sequence. In this case, a common reverse primer is used for both reactions. Following amplification, the PCR products are detected by electrophoresis (capillary or agarose gel) or in q-PCR systems. The detection of adequate PCR product in the wild-type amplification reaction is important to control for adequate specimen quality and quantity, particularly when the specimen is negative in the mutation-specific PCR reaction.
AS-PCR is particularly useful for the detection of targeted point mutations. Multiplex AS-PCR reactions can be designed for the simultaneous detection of multiple mutations by including several mutation-specific primers. The method has high analytical sensitivity and specificity and can be easily deployed in most clinical laboratories. However, an important limitation is that this approach will not detect mutations other than those for which specific primers are designed. Therefore, it is utilized for highly recurrent mutations that occur at specific locations within genes, rather than for the detection of variable mutations that may occur throughout a gene.
Examples of AS-PCR applications in oncology include the detection of JAK2 V617F and MPL mutations in myeloproliferative neoplasms (primary myelofibrosis, essential thrombocythemia, and/or polycythemia vera),29 the BRAF V600E mutation,30 and KIT D816V mutations in cases of systemic mastocytosis and in acute myelogenous leukemia (AML).
Reverse Transcriptase PCR
RT-PCR is utilized for the detection and quantification of RNA transcripts. The first step for all amplification-based assays that use RNA as a starting material is reverse transcription of RNA into cDNA, because RNA is not a suitable substrate for Taq polymerase. In RT-PCR, RNA is isolated and reverse transcribed into cDNA by using a reverse transcriptase enzyme and one of the following: (1) random hexamer primers, which anneal randomly to RNA and reverse transcribe all RNA in the cell; (2) oligo dT primers, which anneal to the polyA tail of mRNA and reverse transcribe only mRNA; or (3) gene-specific primers that reverse transcribe only the target of interest. PCR is subsequently performed on the cDNA with forward and reverse primers specific to the gene(s) of interest. The RT-PCR products may then be analyzed by capillary electrophoresis or in real-time systems as in a standard PCR reaction.
RT-PCR is commonly used for detecting gene fusions during translocation analysis because breakpoints frequently occur within the intron of each partner gene and the precise intronic breakpoint locations may be variable. This variability complicates the design of primers used in DNA-based PCR assays. RT-PCR tests are advantageous because mature mRNA has intronic sequence spliced out, allowing for simplified primer design within the affected exon of each partner gene. In this setting, RT-PCR is useful in tests where both translocation partners are recurrent and only one or a few exons are involved in each partner gene. For instance, 95% of acute promyelocytic leukemia (APL) cases harbor the reciprocal t(15;17) chromosomal translocation and these breakpoints always occur within intron 2 of the RARA gene. By contrast, three distinct chromosome 15 breakpoints are involved, all occurring within the PMLgene: intron 6, exon 6, and intron 3. Because the breakpoints in the two genes are recurrent, most of the reported PML-RARA fusions can be detected by targeting these three transcript isoforms.
RT-PCR is the method of choice when high sensitivity is required to detect gene translocations. For example, PML-RARA transcript detection by RT-PCR can detect this fusion transcript down to 1 tumor cell in the background of 100,000 normal cells. Detecting low levels of fusion transcript can reveal relapse after consolidation and guide further treatment.31 RT-PCR can also be used to quantitate the amount of expression of a gene. One major application of RT-PCR in this setting includes quantitative detection of BCR-ABL1 fusion transcript for prognostication and minimal residual disease testing in CML (Fig. 3.4). In this setting, a three log decrease in BCR-ABL1 levels is associated with an improved outcome.32,33
Fragment Analysis
A fragment analysis is a PCR amplicon-sizing technique that is relevant for the detection of small- to medium-length–affecting mutations (deletions, insertions, and duplications). This is typically performed by capillary electrophoresis, which is capable of resolving length mutations from approximately 1 to 500 base pairs in size.
Fragment analysis represents a practical strategy because it enables comprehensive detection of a wide variety of possible length mutations and has high analytic sensitivity. Further, it can provide semiquantitative information regarding the relative amount of mutated alleles. Limitations of this approach include the inability to objectively quantitate mutant allele burdens, the inability to determine the exact change in nucleotide sequence, and the inability to detect non–length-affecting mutations such as substitution mutations.
Examples of fragment analysis applications in oncology include the detection of NPM1 insertion mutations (Fig. 3.5),34 EGFR exon 19 deletions, FLT3 internal tandem duplications, and JAK2 exon 12 mutations.35
High-Resolution Melting Curve Analysis
A high-resolution melting (HRM) curve analysis is a mutation screening method that allows for the detection of DNA sequence variations based on specific sequence-related melting profiles of PCR products.36 Because the melting property of DNA duplexes is dependent on the biophysical and chemical properties of the nucleotide sequences, mutant and wild-type DNA sequences can be differentiated from one another based on their melting characteristics.
An HRM analysis is preceded by a PCR. The reaction employs a pair of gene-specific forward and reverse primers, template DNA, and a reporter that can either be a double-stranded DNA binding dye or a fluorescent reporter probe. Following the last cycle of the PCR, the amplification products undergo a cooling step that generates homoduplexes (double-stranded molecules with perfect complementarity between alleles) and heteroduplexes (double-stranded molecules with sequence mismatch between alleles) followed by a heating step that denatures (i.e., melts) the double-stranded products. Heteroduplexes (mutant DNA) produce a melting profile different from that of wild-type samples (homoduplexes). In most cases, the reaction is performed in a q-PCR system that allows for an analysis of amplification and melting data in a close-tube format, thereby minimizing the risk of amplicon contamination.
An HRM analysis is useful for the qualitative detection of variable point mutations and small length-affecting mutations that occur within mutational hotspot regions. This method has high analytical sensitivity and can detect mutations even in a small fraction of alleles in a background of wild-type DNA. However, this assay does not characterize the specific sequence alteration in the mutant allele and may be challenging to interpret, especially for cases with mutation levels that approach the detection limit of the assay. Samples with a lower abundance of mutant alleles, and consequently a decreased fraction of heteroduplexes that produced fluorescence decay during the melting analysis, usually produce a melting curve that may not differ significantly from that of wild-type samples. Likewise, the detection of duplication mutations may be hampered by the similarity between the mutant and the duplicated wild-type genome sequences, which may produce only subtle differences in the melting behavior of the DNA duplexes, especially for samples with low mutant allele burden. Therefore, both the mutant sequence and the allelic burden play in the ability of an HRM analysis to detect mutations.37 Poor quality and impurity of genomic DNA may also lower the sensitivity of an HRM analysis.38 In instances of patients with a low mutant allelic burden, equivocal mutations identified by this approach may not be confirmable by an alternate method such as Sanger sequencing.
Examples of HRM applications in oncology include a mutational analysis of KRAS codons 12, 13, and 6139; a mutation screening of BRAF codon 60039; and the detection of JAK2 exon 12 mutations (Fig. 3.6).40
Sanger Sequencing
Mutations in single gene assays are commonly analyzed by targeted nucleic acid sequencing, most commonly by Sanger sequencing.41 This method, also known as dideoxy sequencing, is based on random incorporation of modified nucleotides (dideoxynucleotides [ddNTP]) into a DNA sequence during rounds of template extension that result in termination of the chain reaction at various fragment lengths. Because dideoxynucleotides lack a 3′ hydroxyl group on the DNA pentose ring, which is required for the addition of further nucleotides during extension of the new DNA strand, the chain reaction is terminated at different lengths with the random incorporation of ddNTPs to the sequence. In addition to the dideoxy modification, each ddNTP (ddATP, ddTTP, ddCTP, ddGTP) is labeled with fluorescent tags of different fluorescence wavelengths.
In this method, repetitive cycles of primer extension are performed using denatured PCR products (amplicons) as templates. Unlike PCR, in which both forward and reverse primers are added to the same reaction, in Sanger sequencing, the forward and reverse reactions are performed separately. Bidirectional sequencing is performed to ensure that the entire region of interest for each analysis is visualized adequately to produce unequivocal sequence readout. The sequencing products of increasing size are resolved by capillary electrophoresis, and the DNA sequence is determined by detection of the fluorescently labeled nucleotide sequences.
Sanger sequencing has the ability to detect a wide variety of nucleotide alterations in the DNA, including point mutations, deletions, insertions, and duplications. This technique is especially useful when mutations are scattered across the entire gene, when genes have not been sufficiently studied to determine mutational hot spots, or when it is relevant to determine the exact change in DNA sequence. Sanger sequencing can also provide semiquantitative information about mutation levels in a sample based on the evaluation of average peak drop values from forward and reverse mutant peaks on sequence chromatograms. Limitations of this approach include low throughput and limited diagnostic sensitivity. In general, heterozygous mutations at allelic levels lower than 20% may be difficult to detect by Sanger sequencing. This may be particularly problematic when testing for somatic mutations in oncogenes, such as JAK2 exon 12 in polycythemia vera, which may occur at low levels.35
Examples of Sanger sequencing applications in oncology include the detection of KIT mutations for gastrointestinal stromal tumors (GIST) and melanomas that arise from mucosal membranes and acral skin, EGFR mutations for non–small-cell lung cancers, and KRAS mutations for colorectal and lung carcinomas (Fig. 3.7).
Pyrosequencing
Pyrosequencing, also known as sequencing by synthesis, is based on the real-time detection of pyrophosphate release by nucleotide incorporation during DNA synthesis.42 In the pyrosequencing reaction, as nucleotides are added to the nucleic acid chain by polymerase, pyrophosphate molecules are released and subsequently converted to ATP by ATP sulfurylase. Light is produced by an ATP-driven luciferase reaction via oxidation of a luciferin molecule. The amount of light produced is proportional to the number of incorporated nucleotides in the sequence. When a nucleotide is not incorporated into the reaction, no pyrophosphate is released and the unused nucleotide is degraded by apyrase. Light is converted into peaks in a charge-coupled device (CCD) camera. Individual dNTP nucleotides are sequentially added to the reaction, and the sequence of nucleotides that produce chemiluminescent signals allow the template sequence to be determined. Mutations appear as new peaks in the pyrogram sequence or variations of the expected peak heights.43
Pyrosequencing is particularly useful for the detection of point mutations and insertion/deletion mutations that occur at short stretches in mutational hotspots. This method has higher analytical sensitivity than Sanger sequencing and can provide quantitative information about mutation levels in a sample. Pyrosequencing can also be used for the detection and quantification of gene-specific DNA methylation and gene copy number assessments. A microfluidic pyrosequencing platform is available for massive parallel sequencing. However, this method is not well suited for detecting mutations that are scattered across the entire gene because pyrosequencing read lengths are limited to ~100 to 250 base pairs.43
Examples of pyrosequencing applications in oncology include the mutational analysis of BRAF (codon 600),44,45 KRAS (codons 12, 13, 61),45 NRAS (codon 61),45 and the methylation analysis of MGMT in glioblastoma multiforme.46,47
Single Nucleotide Extension Assay (SNaPshot®)
The single nucleotide extension assay is a variant of dideoxy sequencing. This method consists of a single base extension of an unlabeled primer that anneals one base upstream to the relevant mutation with fluorophore-labeled dideoxynucleotides (ddNTP). Multiplexed reactions can be designed with multiple primers of differing lengths for simultaneous amplification of multiple genomic targets.48 Mutations are identified based on amplicon size and fluorophore color via capillary electrophoresis. When a mutation is present, an alternative dideoxynucleotide triphosphate is incorporated, resulting in a different colored peak with a different amplicon length than the expected wild-type one.
The single nucleotide extension assay is particularly useful for the simultaneous detection of recurrent point mutations. Clinically, it has been employed for analyses of mutational hotspots in multiple genes involved in melanomas, non–small-cell lung cancers, breast cancers, and metastatic colorectal cancers.49 The assay has higher analytical sensitivity than Sanger sequencing and can detect low-level mutations in FFPE-derived DNA, making it advantageous for biopsy specimens with limited tumor involvement. This assay, however, can only detect mutations that are immediately adjacent to the 3′ to the end of the primer.
Fluorescence In Situ Hybridization
FISH allows for the visualization of specific chromosome nucleic acid sequences within a cellular preparation. This method involves the annealing of a large single-stranded fluorophore-labeled oligonucleotide probe to complementary DNA target sequences within a tissue or cell preparation. The hybridization of the probe at the specific DNA region within a nucleus is visible by direct detection using fluorescence microscopy.
FISH can be used for the quantitative assessment of gene amplification or deletion and for the qualitative evaluation of gene rearrangements. Many oncologic FISH assays employ two probe types: locus specific probes, which are complementary to the gene of interest, and centromeric probes, which hybridize to the alpha-satellite regions near the centromere of a specific chromosome and help in the enumeration of the number of copies of that chromosome.
For the quantitative assessment of gene amplification, a locus-specific probe and a centromeric probe are labeled with two different fluorophores. The signals generated by each of these probes are counted and a ratio of the targeted gene to the chromosome copy number is calculated. The amount of signal produced by the locus-specific probe is proportional to the number of copies of the targeted gene in a cell. This type of gene amplification assay can be used for the detection of HER2 gene amplification as an adjunct to existing clinical and pathologic information as an aid in the assessment of stage II, node-positive breast cancer patients for whom Herceptin treatment is being considered. It can also be used for an assessment of MYCN amplification in neuroblastoma.
For the detection of deletion mutations, dual-probe hybridization is usually performed using locus-specific probes. For instance, for the detection of 1p/19q codeletion in oligodendrogliomas, locus-specific probe sets for 1p36 and 19q13, and 1q25 and 19p13 (control) are used. The frequencies of signal patterns for each of these loci are evaluated. A signal pattern with 1p and 19q signals that are less than control signals is consistent with deletion of these loci.
Gene rearrangements/chromosomal translocations in hematologic or solid malignancies can be tested using locus-specific dual-fusion or break-apart probes. Dual-color, dual-fusion translocation assays employ two probes that are located in two separate genes involved in a specific rearrangement. Each gene probe is labeled in a different color. This design detects translocations by the juxtaposition of both probe signals. Dual-color, dual-fusion translocation assays are very specific for detecting a selected translocation. But, it can only be used for detecting translocations that involve consistent partners, where both partners are known. Alternate translocations with different fusion partners are not detected by this approach. Examples of application of dual-fusion probes in oncology include for the detection of the IGH-BCL2 translocation that occurs in most follicular lymphomas and a subset of diffuse large B-cell lymphomas (Fig. 3.8) and for the detection of IGH-CCND1 rearrangements in mantle cell lymphomas.
In break-apart FISH assays, both dual-colored probes flank the breakpoint region in a single gene that represents the constant partner in the translocation. By this approach, rearranged alleles show two split signals, whereas normal alleles show fusion signals. This design is particularly useful for genes that fuse with multiple translocation partners (e.g., EWSR1 gene, which may undergo rearrangement with multiple partner genes, including FLI1, ERG, ETV1, FEV, and E1AF in Ewing sarcoma/primitive neuroectodermal tumor [PNET]; WT1 in desmoplastic small round cell tumors; CHN in extraskeletal myxoid chondrosarcoma; and ATF1 in clear cell sarcoma and angiomatoid fibrous histiocytoma).50 The disadvantage of this approach is that break-apart FISH does not allow for the identification of the “unknown” partner in the translocation.
FISH has the advantage of being applicable to a variety of specimen types, including FFPE tissue. Because probes are hybridized to tissue in situ, the tumor morphology is preserved, which allows for an interpretation of the assay even in the context of heterogeneous samples. However, FISH is a targeted approach that will only detect specific alterations. Because most probes are large (e.g., >100 kb), small deletions or insertions will not be detected. In addition, poor tissue fixation, fixation artifacts, nuclear truncation on tissue slides, and nuclear overlaps are potential pitfalls of this technique that may hamper interpretation. Some intrachromosomal rearrangements (e.g., RET-PTCand EML4-ALK) may be challenging to interpret by FISH due to subtle rearrangements of the probe signals on the same chromosome arm.
Methylation Analysis
Changes in the methylation status of cytosine in DNA regions enriched for the sequence CpG (also known as CpG islands) are early events in many cancers and permanent changes found in many tumors. The detection of aberrant methylation of cancer-related genes may aid in the diagnosis, prognosis, and/or determination of the metastatic potential of tumors.
The most common approaches for the detection of methylation are based on the conversion of unmethylated cytosine bases into uracil after sodium bisulfite treatment, which is then converted to thymidine during PCR. By this approach, bisulfite-treated methylated alleles have different DNA sequences as compared with their corresponding unmethylated alleles. The differences between methylated and unmethylated DNA sequences can be evaluated by several methods, including methylation-sensitive restriction enzyme analysis, methylation-specific PCR, semiquantitative q-PCR, Sanger sequencing, pyrosequencing, and next-generation sequencing.
The methylation status of oncogenic genes can also be assessed by methylation-sensitive multiplex ligation-dependent probe amplification (MS-MLPA) assay.51,52 MS-MLPA is a variant of multiplex PCR in which oligonucleotide probes hybridized to the targeted DNA samples are directly amplified using one pair of universal primers. This method is not based on bisulfite conversion of unmethylated cytosine bases into uracil. Instead, the target sequences detected by MS-MLPA probes contain a restriction site recognized by methylation-sensitive endonucleases. A probe amplification product will only be obtained if the CpG site is methylated because digested probes cannot be amplified during PCR. The level of methylation is determined by resolving PCR products by capillary electrophoresis and calculating the normalized ratio of each target probe peak area in both digested and undigested specimens. The ratio corresponds to the percentage of methylation present in the specimen.
Examples of applications of methylation analysis in oncology include an analysis of MLH1 promoter hypermethylation in microsatellite unstable sporadic colorectal carcinomas, an analysis of MGMTpromoter methylation status in glioblastoma multiforme patients treated with alkylating chemotherapy, and SEPT9 promoter methylation in DNA derived from blood plasma in colorectal cancer patients.53
Microsatellite Instability Analysis
Microsatellites are short, tandem-repeated DNA sequences with repeating units of one to six base pairs in length. Microsatellites are distributed throughout the human genome, and individual repeat loci often vary in length from one individual to another. Microsatellite instability (MSI) is the change in length of a microsatellite allele due to either insertion or deletion of repeating units and a failure of the DNA mismatch repair (MMR) system to repair these replication errors. This genomic instability arises in a variety of human neoplasms where tumor cells have a decreased ability to faithfully replicate DNA. MSI is particularly associated with colorectal cancer, where 15% to 20% of sporadic tumors show MSI, in contrast to the more common chromosomal instability (CIN) phenotype, with MSI status being an independent prognostic indicator. MSI analysis is also clinically useful in identifying patients at increased risk of hereditary nonpolyposis colorectal cancer (HNPCC)/Lynch syndrome, where a germline mutation of an MMR gene causes a familial predisposition to colorectal cancer. MSI analysis alone is not sufficient to make a diagnosis of a germline MMR mutation given the high rate of sporadic MSI-positive colorectal tumors, but a positive result is an indication for follow-up genetic testing and counseling.
In an MSI analysis, DNA is extracted from tumor tissue and the corresponding adjacent normal mucosa. The DNA is subjected to multiplex PCR using fluorescent-labeled primers for coamplification of five mononucleotide repeat markers for MSI determination and two pentanucleotide markers for confirming tumor/normal sample identity. The resulting PCR fragments are separated and detected using capillary electrophoresis. Allelic profiles of normal versus tumor tissues are compared, and MSI is scored as the presence of novel microsatellite lengths in tumor DNA compared to normal DNA. Instability in two or more out of five mononucleotide microsatellite markers in tumor DNA compared to normal DNA is defined as MSI-H (high). MSI-L (low) is defined as instability in one out of five mononucleotide markers in tumor DNA compared to normal DNA. Tumors with no instability (zero out of five altered mononucleotide markers) are defined as microsatellite stable (MSS).54,55
Loss of Heterozygosity Analysis
Loss of heterozygosity (LOH) is a common event in cancer that usually occurs due to deletion of a chromosome segment and results in a loss of one copy of an allele. LOH is a common occurrence in tumor suppressor genes and may contribute to tumorigenesis when the second allele is subsequently inactivated by a second “hit” due to mutation or deletion.
LOH studies are used to identify genomic imbalance in tumors, indicating possible sites of tumor suppressor gene (TSG) deletion. LOH studies can be done by multiplex PCR analysis of microsatellites (short tandem repeats [STRs]), FISH, and genomic microarrays). By PCR, microsatellites located in the vicinity of a tumor suppressor gene are used as surrogate markers for the presence of the gene of interest. DNA is extracted from tumor tissue and corresponding adjacent normal mucosa. The DNA is subjected to multiplex PCR using fluorescent-labeled STR primers. Peak height ratio of informative (nonhomozygous) alleles at each locus is calculated from both normal and tumor tissues. LOH is defined as the decrease in peak height of one of the two alleles, relative to the allele peak heights of the normal sample.
An example of applications of LOH studies in oncology include an analysis of 1p/19q loss in oligodendrogliomas, and an analysis of 1p loss in parathyroid carcinomas.
Whole Genome Analysis Methods
Next-Generation Sequencing
Next-generation sequencing (NGS), also known as massive parallel sequencing or deep sequencing, is an emerging technology that has revolutionized the speed, throughput, and cost of sequencing and has facilitated the discovery of clinically relevant genetic biomarkers for diagnosis, prognosis, and personalized therapeutics. By way of this technology, multiple genes or the entire exome or genome can be interrogated simultaneously in multiple parallel reactions instead of a single-gene basis as in Sanger sequencing or pyrosequencing. Currently, the most common NGS approach for cancer testing in the clinical setting employs targeted sequencing of specific genes and mutation hotspot regions. This targeted approach increases sensitivity for the detection of low-level mutations by increasing the depth of sequence coverage.
Presently, there are numerous NGS platforms that employ different sequencing technologies. A comprehensive review and comparison of NGS platforms is beyond the scope of this chapter and has been reviewed elsewhere.56,57 A generalized clinical workflow is shown (Fig. 3.9). Frequently, multiple DNA samples are individually barcoded and pooled together to leverage platform throughput. Pooled libraries are prepared and enriched, and single DNA molecules are arrayed in solid surfaces, glass slides, or beads and sequenced in situ using reversible DNA chain terminators or iterative cycles of oligonucleotide ligation. NGS signal outputs are based on luminescence, fluorescence, or changes in ion concentration. Robust bioinformatics pipelines are required for an alignment of reads to a reference genome sequence, variant calling, variant annotation, and to assist with result reporting.58
NGS can be used for the detection of single nucleotide variants, small insertions and deletions, translocations, inversions, alternative splicing, and copy number variations given sufficient depth of genomic DNA sequence (Fig. 3.10). Technical limitations of this technique include difficulty in sequencing guanine-cytosine (GC)–rich genomic regions, and erroneous sequencing of homologous DNA regions (e.g., pseudogenes) that may confound interpretation.
Examples of applications of NGS in oncology include small targeted panels (3 to 50 genes) for non–small-cell lung cancers, melanomas, colon cancers, and acute myeloid leukemias.57,59–62 Larger panels (50 to 500 genes) are increasingly being utilized, particularly in both clinical trials and research.
Massively parallel sequencing of RNA (RNA-Seq) can be used for determining sequence variants, alternative splicing, gene rearrangements, and allelic expression of mutant transcripts. To date, this technique has been used primarily for discovery rather than clinical applications, but it is likely to play an increasing role in clinical diagnostics as the technology improves. For transcriptome sequencing, the RNA must first be converted to cDNA, which is then fragmented and entered into library construction. After sequencing, reads are aligned to a reference genome, compared with known transcript sequences, or assembled de novo to construct a genome-scale transcription map. Expression levels are determined from the total number of sequence reads that map to the exons of a particular gene, normalized by the length of exons that can be uniquely mapped.56 Compared with genomic microarrays, RNA-Seq has a greater ability to distinguish RNA isoforms, determine allelic expression, and reveal sequence variants.
Chromatin immunoprecipitation with sequencing (ChIP-Seq) can be used to determine the genome-wide location of chromatin-binding transcription factors or specific epigenetic modifications of histones. This has proved to be a very powerful research tool, which to date has not been used for clinical diagnostics. Proteins in contact with genomic DNA are chemically cross-linked (usually with formaldehyde treatment) to their binding sites, the DNA is fragmented, and the proteins cross-linked with DNA are then immunoprecipitated with antibodies specific for the proteins (or specific epigenetic histone modification) of interest. The DNA harvested from the immunoprecipitate is converted into a library for NGS. The obtained reads are mapped to the reference genome of interest to generate a genome-wide protein binding map.63,64 ChIP-Seq is rapidly replacing chromatin immunoprecipitation and microarray hybridization (ChIP-on-chip) technology65 because of its higher sensitivity and resolution.66
Genomic Microarrays
High-density genomic microarrays are widely used for whole genome assessment of copy number changes, LOH, and genotyping. In array comparative genomic hybridization (aCGH), cloned genomic probes are arrayed onto glass slides and serves as targets for the competitive hybridization of normal and tumor DNA. In the aCGH reaction, tumor DNA and DNA from a normal control sample are labeled with different fluorophores. These samples are denatured and hybridized together to the arrayed single-strand probes. Digital imaging systems are used to quantify the relative fluorescence intensities of the labeled DNA probes that have hybridized to each target probe. The fluorescence ratio of the tumor and control hybridization signals is determined at different positions along the genome, which provides information on the relative copy number of sequences in the tumor genome as compared to the normal genome.67 This method is able to detect copy number variation, such as deletions, duplications, and gene amplification, but it cannot detect polymorphic allele changes.
An SNP array has the ability to detect LOH profiles in addition to high-resolution detection of copy number aberrations, such as amplifications and deletions. This method employs thousands of unique fluorescent-labeled nucleotide probe sequences arrayed on a chip to which a fragmented single-stranded specimen DNA binds to their complementary partners. Each SNP site is interrogated by complementary sets of probes containing perfect matches and mismatches to each SNP site. Each probe is associated with one of the two alleles of an SNP (also known as A and B). Relative fluorescence intensity depends on both the amount of target DNA in the sample, as well as the affinity between target and probe. An analysis of the raw fluorescence intensity is done by computational algorithms that convert the set of probe intensities into genotypes. Deleted genomic regions are identified as having an LOH associated with copy number reduction. A copy-neutral LOH is detected when SNPs expected to be heterozygous in the normal sample are detected as homozygous in the tumor sample without copy number variation. A copy neutral LOH may arise from somatic homologous recombination of a mutated tumor suppressor allele and its surrounding DNA that replaces the other allele (uniparental disomy [UPD]). SNP microarrays are the only genomic microarrays that are able to identify UPD. Array technologies cannot detect true balanced chromosome abnormalities and low-level mosaicism.
Examples of genomic microarrays applications in oncology include the detection of copy number variations and LOH in chronic lymphocytic leukemia68 and recurrent cytogenetic abnormalities in MDS (e.g., 5q-, -7 or 7q-, +8, 20q-).69
Expression Panels
Gene expression signatures of multiple cancer biomarkers are starting to be incorporated into clinical practice as an adjunct to clinical and pathologic information in diverse cancer management settings. An example of a multigene expression–based test in current use includes Oncotype DX, which is a quantitative RT-PCR–based assay that measures the expression of 21 genes in FFPE breast tumors. The test is designed to predict the potential benefit of chemotherapy and the likelihood of distant breast cancer recurrence in women with node negative or node positive, estrogen receptor (ER)-positive, and HER2-negative invasive breast cancer. This test has been in-corporated into current American Society of Clinical Oncology (ASCO) and National Comprehensive Cancer Network (NCCN) for breast cancer management.70 Prospective trials are in progress to evaluate other multigene tests for early stage breast cancer.
With the rapid advances in molecular diagnostic technologies, it is likely that many mutation- and expression-based panels analyzing hundreds if not thousands of genes, or even the complete genome or transcriptome, will enter widespread use. Some of the many challenges to address will be to provide evidence-based, actionable reports that guide the oncologist to more effective therapies, to learn from the results of such testing to improve the algorithms guiding therapy, to handle the incidental findings in such testing in an ethically responsible way, and ultimately, with the drugs available, to provide sufficient improvements in outcomes so that society will be willing to bear the costs.
REFERENCES
1. Lennard L, Cartwright CS, Wade R, et al. Thiopurine methyltransferase genotype-phenotype discordance and thiopurine active metabolite formation in childhood acute lymphoblastic leukaemia. Br J Clin Pharmacol 2013;76(1):125–136.
2. Haber DA, Settleman J. Cancer: drivers and passengers. Nature 2007;446(7132):145–146.
3. Hayashi T, Sudo J. Relieving effect of saline on cephaloridine nephrotoxicity in rats. Chem Pharm Bull (Tokyo) 1989;37(3):785–790.
4. Brodeur GM, Seeger RC, Schwab M, et al. Amplification of N-myc in untreated human neuroblastomas correlates with advanced disease stage. Science 1984;224(4653):1121–1124.
5. Seeger RC, Brodeur GM, Sather H, et al. Association of multiple copies of the N-myc oncogene with rapid progression of neuroblastomas. N Engl J Med 1985;313(18):1111–1116.
6. Weinstein JL, Katzenstein HM, Cohn SL. Advances in the diagnosis and treatment of neuroblastoma. Oncologist 2003;8(3):278–292.
7. Pui CH, Crist WM, Look AT. Biology and clinical significance of cytogenetic abnormalities in childhood acute lymphoblastic leukemia. Blood 1990;76(8):1449–1463.
8. Lee JA, Lupski JR. Genomic rearrangements and gene copy-number alterations as a cause of nervous system disorders. Neuron 2006;52(1):103–121.
9. Kuiper RP, Ligtenberg MJ, Hoogerbrugge N, et al. Germline copy number variation and cancer risk. Curr Opin Genet Dev 2010;20(3):282–289.
10. Shlien A, Malkin D. Copy number variations and cancer. Genome Med 2009;1(6):62.
11. Lynch TJ, Bell DW, Sordella R, et al. Activating mutations in the epidermal growth factor receptor underlying responsiveness of non–small-cell lung cancer to gefitinib. N Engl J Med 2004;350(21):2129–2139.
12. Paez JG, Janne PA, Lee JC, et al. EGFR mutations in lung cancer: correlation with clinical response to gefitinib therapy. Science 2004;304(5676):1497–1500.
13. Forbes SA, Bindal N, Bamford S, et al. COSMIC: mining complete cancer genomes in the Catalogue of Somatic Mutations in Cancer. Nucleic Acids Res 2011;39(Database issue):D945–950.
14. Van Allen EM, Wagle N, Levy MA. Clinical analysis and interpretation of cancer genome data. J Clin Oncol 2013;31(15):1825–1833.
15. Newman B, Mu H, Butler LM, et al. Frequency of breast cancer attributable to BRCA1 in a population-based series of American women. JAMA 1998;279(12):915–921.
16. Ford D, Easton DF, Stratton M, et al. Genetic heterogeneity and penetrance analysis of the BRCA1 and BRCA2 genes in breast cancer families. The Breast Cancer Linkage Consortium. Am J Hum Genet 1998;62(3):676–689.
17. Antoniou A, Pharoah PD, Narod S, et al. Average risks of breast and ovarian cancer associated with BRCA1 or BRCA2 mutations detected in case series unselected for family history: a combined analysis of 22 studies. Am J Hum Genet 2003;72(5):1117–1130.
18. Chen S, Iversen ES, Friebel T, et al. Characterization of BRCA1 and BRCA2 mutations in a large United States sample. J Clin Oncol 2006;24(6):863–871.
19. Bachner P, Hamlin W. Federal regulation of clinical laboratories and the Clinical Laboratory Improvement Amendments of 1988—Part II. Clin Lab Med 1993;13(4):987–994.
20. Halling KC, Schrijver I, Persons DL. Test verification and validation for molecular diagnostic assays. Arch Pathol Lab Med 2012;136(1):11–13.
21. Jennings L, Van Deerlin VM, Gulley ML, College of American Pathologists Molecular Pathology Resource C. Recommended principles and practices for validating clinical molecular pathology tests. Arch Pathol Lab Med 2009;133(5):743–755.
22. Jennings LJ, Smith FA, Halling KC, et al. Design and analytic validation of BCR-ABL1 quantitative reverse transcription polymerase chain reaction assay for monitoring minimal residual disease. Arch Pathol Lab Med 2012;136(1):33–40.
23. Pont-Kingdon G, Gedge F, Wooderchak-Donahue W, et al. Design and analytical validation of clinical DNA sequencing assays. Arch Pathol Lab Med 2012;136(1):41–46.
24. Saiki RK, Gelfand DH, Stoffel S, et al. Primer-directed enzymatic amplification of DNA with a thermostable DNA polymerase. Science 1988;239(4839):487–491.
25. Mullis KB. The unusual origin of the polymerase chain reaction. Sci Am 1990;262(4):56–61, 64–65.
26. Bernard PS, Wittwer CT. Real-time PCR technology for cancer diagnostics. Clin Chem 2002;48(8):1178–1185.
27. Bench AJ, Baxter EJ, Green AR. Methods for detecting mutations in the human JAK2 gene. Methods Mol Biol 2013;967:115–131.
28. Halait H, Demartin K, Shah S, et al. Analytical performance of a real-time PCR-based assay for V600 mutations in the BRAF gene, used as the companion diagnostic test for the novel BRAF inhibitor vemurafenib in metastatic melanoma. Diagn Mol Pathol 2012;21(1):1–8.
29. Furtado LV, Weigelin HC, Elenitoba-Johnson KS, et al. Detection of MPL mutations by a novel allele-specific PCR-based strategy. J Mol Diagn 2013;15(6):810–818.
30. Lang AH, Drexel H, Geller-Rhomberg S, et al. Optimized allele-specific real-time PCR assays for the detection of common mutations in KRAS and BRAF. J Mol Diagn 2011;13(1):23–28.
31. Wang ZY, Chen Z. Acute promyelocytic leukemia: from highly fatal to highly curable. Blood 2008;111(5):2505–2515.
32. O’Brien SG, Guilhot F, Larson RA, et al. Imatinib compared with interferon and low-dose cytarabine for newly diagnosed chronic-phase chronic myeloid leukemia. N Engl J Med 2003;348(11):994–1004.
33. Hughes TP, Kaeda J, Branford S, et al. Frequency of major molecular responses to imatinib or interferon alfa plus cytarabine in newly diagnosed chronic myeloid leukemia. N Engl J Med 2003;349(15):1423–1432.
34. Szankasi P, Jama M, Bahler DW. A new DNA-based test for detection of nucleophosmin exon 12 mutations by capillary electrophoresis. J Mol Diagn 2008;10(3):236–241.
35. Furtado LV, Weigelin HC, Elenitoba-Johnson KS, et al. A multiplexed fragment analysis-based assay for detection of JAK2 exon 12 mutations. J Mol Diagn 2013;15(5):592–599.
36. Reed GH, Kent JO, Wittwer CT. High-resolution DNA melting analysis for simple and efficient molecular diagnostics. Pharmacogenomics 2007;8(6):597–608.
37. Palais RA, Liew MA, Wittwer CT. Quantitative heteroduplex analysis for single nucleotide polymorphism genotyping. Anal Biochem 2005;346(1):167–175.
38. Carillo S, Henry L, Lippert E, et al. Nested high-resolution melting curve analysis a highly sensitive, reliable, and simple method for detection of JAK2 exon 12 mutations—clinical relevance in the monitoring of polycythemia. J Mol Diagn 2011;13(3):263–270.
39. Ney JT, Froehner S, Roesler A, et al. High-resolution melting analysis as a sensitive prescreening diagnostic tool to detect KRAS, BRAF, PIK3CA, and AKT1 mutations in formalin-fixed, paraffin-embedded tissues. Arch Pathol Lab Med 2012;136(9):983–992.
40. Jones AV, Cross NC, White HE, et al. Rapid identification of JAK2 exon 12 mutations using high resolution melting analysis. Haematologica Oct 2008;93(10):1560–1564.
41. Sanger F, Nicklen S, Coulson AR. DNA sequencing with chain-terminating inhibitors. Proc Natl Acad Sci U S A 1977;74(12):5463–5467.
42. Ronaghi M, Uhlén M, Nyrén P. A sequencing method based on real-time pyrophosphate. Science 1998;281(5375):363, 365.
43. Ronaghi M, Shokralla S, Gharizadeh B. Pyrosequencing for discovery and analysis of DNA sequence variations. Pharmacogenomics 2007;8(10):1437–1441.
44. Shigaki H, Baba Y, Watanabe M, et al. KRAS and BRAF mutations in 203 esophageal squamous cell carcinomas: pyrosequencing technology and literature review. Ann Surg Oncol 2013;20:485–491.
45. Vaughn CP, Zobell SD, Furtado LV, et al. Frequency of KRAS, BRAF, and NRAS mutations in colorectal cancer. Genes Chromosomes Cancer 2011;50(5):307–312.
46. Everhard S, Tost J, El Abdalaoui H, et al. Identification of regions correlating MGMT promoter methylation and gene expression in glioblastomas. Neuro Oncol 2009;11(4):348–356.
47. Mikeska T, Bock C, El-Maarri O, et al. Optimization of quantitative MGMT promoter methylation analysis using pyrosequencing and combined bisulfite restriction analysis. J Mol Diagn 2007;9(3):368–381.
48. Dias-Santagata D, Akhavanfard S, David SS, et al. Rapid targeted mutational analysis of human tumours: a clinical platform to guide personalized cancer medicine. EMBO Mol Med 2010;2(5):146–158.
49. Su Z, Dias-Santagata D, Duke M, et al. A platform for rapid detection of multiple oncogenic mutations with relevance to targeted therapy in non–small-cell lung cancer. J Mol Diagn 2011;13(1):74–84.
50. Lazar A, Abruzzo LV, Pollock RE, et al. Molecular diagnosis of sarcomas: chromosomal translocations in sarcomas. Arch Pathol Lab Med 2006;130(8):1199–1207.
51. Nygren AO, Ameziane N, Duarte HM, et al. Methylation-specific MLPA (MS-MLPA): simultaneous detection of CpG methylation and copy number changes of up to 40 sequences. Nucleic Acids Res 2005;33(14):e128.
52. Hömig-Hölzel C, Savola S. Multiplex ligation-dependent probe amplification (MLPA) in tumor diagnostics and prognostics. Diagn Mol Pathol 2012;21(4):189–206.
53. Warren JD, Xiong W, Bunker AM, et al. Septin 9 methylated DNA is a sensitive and specific blood test for colorectal cancer. BMC Med 2011;9:133.
54. Boland CR, Thibodeau SN, Hamilton SR, et al. A National Cancer Institute Workshop on Microsatellite Instability for cancer detection and familial predisposition: development of international criteria for the determination of microsatellite instability in colorectal cancer. Cancer Res 1998;58(22):5248–5257.
55. Umar A, Boland CR, Terdiman JP, et al. Revised Bethesda Guidelines for hereditary nonpolyposis colorectal cancer (Lynch syndrome) and microsatellite instability. J Natl Cancer Inst 2004;96(4):261–268.
56. Voelkerding KV, Dames SA, Durtschi JD. Next-generation sequencing: from basic research to diagnostics. Clin Chem 2009;55(4):641–658.
57. Cronin M, Ross JS. Comprehensive next-generation cancer genome sequencing in the era of targeted therapy and personalized oncology. Biomarker Med 2011;5(3):293–305.
58. Coonrod EM, Durtschi JD, Margraf RL, et al. Developing genome and exome sequencing for candidate gene identification in inherited disorders: an integrated technical and bioinformatics approach. Arch Pathol Lab Med 2013;137(3):415–433.
59. Grossmann V, Kohlmann A, Klein HU, et al. Targeted next-generation sequencing detects point mutations, insertions, deletions and balanced chromosomal rearrangements as well as identifies novel leukemia-specific fusion genes in a single procedure. Leukemia2011;25(4):671–680.
60. Marchetti A, Del Grammastro M, Filice G, et al. Complex mutations & subpopulations of deletions at exon 19 of EGFR in NSCLC revealed by next generation sequencing: potential clinical implications. PloS One 2012;7(7):e42164.
61. McCourt CM, McArt DG, Mills K, et al. Validation of next generation sequencing technologies in comparison to current diagnostic gold standards for BRAF, EGFR and KRAS mutational analysis. PloS One 2013;8(7):e69604.
62. Thol F, Kölking B, Damm F, et al. Next-generation sequencing for minimal residual disease monitoring in acute myeloid leukemia patients with FLT3-ITD or NPM1 mutations. Genes Chromosomes Cancer 2012;51(7):689–695.
63. Barski A, Cuddapah S, Cui K, et al. High-resolution profiling of histone methylations in the human genome. Cell 2007;129(4):823–837.
64. Schones DE, Zhao K. Genome-wide approaches to studying chromatin modifications. Nat Rev Genet 2008;9(3):179–191.
65. Ren B, Robert F, Wyrick JJ, et al. Genome-wide location and function of DNA binding proteins. Science 2000;290(5500):2306–2309.
66. Robertson G, Hirst M, Bainbridge M, et al. Genome-wide profiles of STAT1 DNA association using chromatin immunoprecipitation and massively parallel sequencing. Nat Methods 2007;4(8):651–657.
67. Shinawi M, Cheung SW. The array CGH and its clinical applications. Drug Discov Today 2008;13(17–18):760–770.
68. Iacobucci I, Lonetti A, Papayannidis C, Martinelli G. Use of single nucleotide polymorphism array technology to improve the identification of chromosomal lesions in leukemia. Curr Cancer Drug Targets 2013;13(7):791–810.
69. Ahmad A, Iqbal MA. Significance of genome-wide analysis of copy number alterations and UPD in myelodysplastic syndromes using combined CGH SNP arrays. Curr Med Chem 2012;19(22):3739–3747.
70. Goncalves R, Bose R. Using multigene tests to select treatment for early-stage breast cancer. J Natl Compr Canc Netw 2013;11(2):174–182.