Advances in Anatomic Pathology:
Implementing Genomic Medicine in Pathology
Williams, Eli S. PhD; Hegde, Madhuri PhD, FACMG
Department of Human Genetics, Emory University School of Medicine, Atlanta, GA
The authors have no funding or conflicts of interest to disclose.
Reprints: Madhuri Hegde, PhD, FACMG, Emory Genetics Laboratory, Department of Human Genetics, Emory University School of Medicine, Atlanta, GA 30322 (e-mail: firstname.lastname@example.org).
The finished sequence of the Human Genome Project, published 50 years after Watson and Crick’s seminal paper on the structure of DNA, pushed human genetics into the public eye and ushered in the genomic era. A significant, if overlooked, aspect of the race to complete the genome was the technology that propelled scientists to the finish line. DNA sequencing technologies have become more standardized, automated, and capable of higher throughput. This technology has continued to grow at an astounding rate in the decade since the Human Genome Project was completed. Today, massively parallel sequencing, or next-generation sequencing (NGS), allows the detection of genetic variants across the entire genome. This ability has led to the identification of new causes of disease and is changing the way we categorize, treat, and manage disease. NGS approaches such as whole-exome sequencing and whole-genome sequencing are rapidly becoming an affordable genetic testing strategy for the clinical laboratory. One test can now provide vast amounts of health information pertaining not only to the disease of interest, but information that may also predict adult-onset disease, reveal carrier status for a rare disease and predict drug responsiveness. The issue of what to do with these incidental findings, along with questions pertaining to NGS testing strategies, data interpretation and storage, and applying genetic testing results into patient care, remains without a clear answer. This review will explore these issues and others relevant to the implementation of NGS in the clinical laboratory.
Genetics is a growing part of public healthcare awareness, thanks in large part to the mainstream attention by the Human Genome Project and the movement toward personalized medicine. Critical to personalized medicine is the ability to survey an individual’s genome for genetic risk factors and integrate this knowledge with other health information to generate a unified plan of medical care and lifestyle choices. While still in the early stages, personalized genomics is rapidly becoming achievable through the implementation of next-generation sequencing (NGS) technologies. Also known as massively parallel sequencing, NGS dramatically increases the scope of genetic tests from a single gene or small gene panel to the entire genome. Already, NGS has revolutionized genetic research by allowing a wider net to be cast, leading to the identification of novel genes for clinically diagnosed diseases such as MLL2 for Kabuki syndrome1 or by providing a complete genetic characterization of complex genetic diseases such as autism.2–4
NGS is beginning to move into the clinical realm as well. At the time of writing, 8 clinical laboratories in the United States are offering whole-exome sequencing (WES) in the setting of rare disease. WES—the sequencing of all coding exons in the human genome (roughly 1% of the entire genome)—offers a potential end to the “diagnostic odyssey” that some individuals with rare genetic diseases or atypical presentations face. Moreover, WES and whole-genome sequencing (WGS) have been shown to be fast enough and cheap enough to designate cancer patients to appropriate clinical trials,5 and a growing number of cancer centers are incorporating WGS into the testing algorithm. The use of NGS on cell-free fetal DNA in maternal plasma has allowed noninvasive prenatal testing for specific chromosome aneuploidies.6–10 Infectious disease diagnosis is also rapidly changing in light of the ability to sequence an entire organism, thereby detecting subtle genetic changes that may direct treatment for specific diseases.11,12 NGS testing approaches are able to provide answers that were previously unattainable due to the scope of the question.
It is important that the groundwork for NGS be properly laid to ensure that the potential of this technology is realized. Numerous issues must be addressed before NGS can be fully implemented into the clinical laboratory. The cost per base of sequencing is currently plummeting as a result of NGS, and the $1000 genome will soon be a reality. However, the upfront investment is risky, particularly for smaller laboratories, as NGS sequencers are expensive, rapidly evolving instruments. Some clinical laboratories are choosing to outsource NGS testing to CLIA-certified genetic facilities, which have undergone significant modifications to establish NGS technology. The testing strategy must also be considered. Is it necessary to perform WES or is a targeted gene panel sufficient? NGS produces massive amounts of data, with 1 individual’s annotated genome comprising terabytes of information.13 These data require the same quality assurance and careful analysis as single-gene tests, a daunting task with current tools and databases.
The larger question of how NGS will fit into the current testing algorithm must also be addressed. NGS will not be a standalone test, but rather a part of larger integrated test model for a variety of diseases. Cancer diagnosis will incorporate clinical and histopathologic information with NGS results to gain a complete picture of the malignancy providing superior diagnostic accuracy and therapeutic decision making. Rare disease diagnosis will require the integration of comprehensive clinical information and relevant testing results (eg, biochemical testing) to facilitate the accurate interpretation of NGS results. Only by integrating genetics into the current clinical testing algorithm will we be able to move toward true personalized medicine.
NEXT-GENERATION SEQUENCING IN RARE DISEASE
The list of genes associated with human disease has grown steadily since the completion of the Human Genome Project, and will continue to grow with increased implementation of NGS. The most recent release of Human Gene Mutation Database (HGMD) catalogues mutations in over 5000 genes related with human disease.14 Clinical genetic testing laboratories are expanding the number of single-gene tests offered to keep abreast of the field, and Sanger sequencing remains the primary sequencing technology present in clinical laboratories today. Also, Sanger sequencing serves as the standard method by which NGS data are compared and validated. Sanger sequencing will remain important, even as NGS moves into the clinical setting. Clear phenotypic indication of a particular disease, for example, rhizomelia and macrocephaly in a newborn, warrants single-gene testing, in this case FGFR3 gene sequencing, with mutations present in 99% of individuals with achondroplasia. The single-gene testing approach eliminates the problems of incidental findings and eases interpretation, but may lead patients on the “diagnostic odyssey” if the phenotype is ambiguous.
Genetic contributions for complex, multifactorial disease such as autism are becoming better understood, in large part due to WES of cohorts of parent-child trios.2–4 Diseases like cardiomyopathy (CM) are also conducive to NGS approaches because of difficulty in precisely phenotyping CM due to the spectrum of disease and the large number of genes involved, over 50 to date. Single-gene testing for autism or CM would be expensive, time-consuming, and ultimately may not find the underlying genetic cause for disease. These complex genetic diseases necessitate testing a large number of genes simultaneously in a cost-effective manner. NGS approaches are well suited for the diagnosis of these diseases and others, but several issues must be addressed before it can become commonplace in the clinical laboratory.
Lynch syndrome testing is a model for the incorporation of genetic and pathologic testing modalities. Lynch syndrome is caused by mutations in DNA mismatch repair genes and individuals with Lynch syndrome are predisposed to malignancies including colorectal cancer (CRC), endometrial cancer, and other associated cancers. Unlike most hereditary cancers, where a clinical suspicion leads to genetic testing of a causative gene, Lynch syndrome poses unique problems due to the genetic heterogeneity and mutation spectrum of the disease. Mutations in MLH1, MSH2, MSH6, PMS2, and EPCAM (which results in MSH2 silencing) have been described in Lynch syndrome; with mutations in MLH1 and MSH2 accounting for approximately 90% of cases.15–20 Moreover, current genetic testing methods fail to detect deep intronic mutations or some promoter mutations. Overcoming these obstacles is possible by the incorporation of screening for tumor microsatellite instability, which arises as a result of mutations in DNA mismatch repair gene mutation,21–23 and immunohistochemical analysis of the protein products of the genes. Data from microsatellite instability and immunohistochemical testing used in conjunction with molecular sequencing provide a comprehensive approach to Lynch syndrome diagnosis.
Whole-exome Sequencing Versus Targeted Next-generation Sequencing Panels
A decision must be made on what NGS testing strategy will be used. Two options have emerged in clinical laboratories as the primary testing modalities for NGS: whole-exome or targeted gene panels (Table 1). A WES approach offers the benefit of a larger spectrum of genes being tested by a single test, in practice however, complete coverage of all coding exons in the genome is currently unattainable. Recent comparative studies between WES platforms have shown that 10% to 20% of targeted bases will not get the 20× read-depth required for clinical confidence and interpretation, with cancer genomes likely requiring much higher read depth in the region of 1000×.5,24 Narrowing the scope to clinically relevant genes does little to improve the coverage. Ninety percent of the 5231 unique genes listed in OMIM, HGMD, GWAS, and CGP are targeted by WES, with approximately 85% of these targeted genes receiving adequate coverage.25 The technical issues for these lapses in coverage are 2-fold—first there is incomplete coverage of the target, that is, the exome, with many genes lacking coverage of at least 1 exon (M.H., unpublished data, 2012), second is the unsequenceable regions of the genome. Although the target enrichment strategies will continue to improve through the release of new platforms, the unsequenceable regions of the genome will remain challenging. A variety of causes contribute to sequencing problems including repetitive sequences and the ability of a particular region of the genome to form secondary structures due to high G-C or A-T content.26,27 Currently, the cost and time investment is simply too high to fill-in underperforming regions of the exome, and this limitation must be appreciated.
In contrast to WES stand the NGS targeted panels offering comprehensive sequencing of all genes included in the panel with the regions refractory to NGS, proactively analyzed by Sanger sequencing. Clinical laboratories are using NGS to offer phenotype-driven targeted panels composed of clinically relevant genes for diseases such as autism, CM, or cancer. Through validation of this subset of genes, problematic exons can be Sanger sequenced to efficiently offer large “clinically complete” gene panels. A limitation to targeted panels is the rigidity of the test and the biased approach of testing only a selected number of genes. As the field is rapidly evolving, new genes may be associated with a clinical phenotype requiring redesign and revalidation of the panel. Also, while the clinical laboratory is not focused on identifying new disease-related genes, the number of clinical genetic tests, including clinical exomes, offers a huge data source for identifying these genetic associations, as has been shown for microarrays and copy number variation.28,29 Clear advantages exist to a targeted NGS panel in addition to the complete clinical coverage of all genes on the panel. Incidental findings, an ethical challenge for WES approaches30 will be dramatically reduced or completely eliminated and the number of variants of unknown clinical significance (VOUS) will decrease. Together these 2 aspects will significantly reduce the time needed to interpret and report test result, thereby reducing cost and increasing turn-around time.
NEXT-GENERATION SEQUENCING IN CANCER TESTING
The identification of the Philadelphia chromosome in patients with chronic myeloid leukemia was the first time a specific genetic (chromosome) abnormality was linked to a specific neoplastic disease.31 Cytogenetic techniques remain important in the diagnosis and management of malignancy, particularly hematological malignancies.32 Greater than 75% of patients with hematological malignancies show an acquired structural or numerical change using conventional cytogenetic techniques including chromosome analysis and fluorescence in situ hybridization.33 There are limitations to cytogenetic approaches, particularly the resolution of the test, which is unable to detect base substitutions and indels, and unable to precisely identify the mutation responsible for malignant transformation. Advances in the detection of copy number variation and the implementation of SNP arrays into clinical practice has improved the tools available to a pathologist. SNP arrays provide the ability to detect copy-neutral loss of heterozygosity (LOH), which is seen in many cancers as a result of ongoing genomic instability. These regions of LOH can constitute the second-hit in a tumor suppressor gene.34 However, SNP arrays are still unable to detect gene fusions, and the utility of these diagnostic tools may be limited to certain cancer types such as chronic lymphocytic leukemia and myelodysplastic syndrome, both of which are characterized by chromosome aneuploidy and reported to arise with LOH.35–37
Detecting Mutations: Drivers Versus Passengers
The number of somatic mutations in a cancer genome is typically in the order of thousands, with the majority of adult cancers in the 1000 to 10,000 range.38,39 Of these mutations, a small number will be driver mutations, those changes that confer a growth advantage to the cell through the alteration of a fundamental cellular process.40 The remainders are so-called passenger mutations, which are not believed to contribute to tumorigenesis, and have accumulated over the course of a lifetime of exposure to mutagens (eg, UV light) or during the deregulated mitoses of malignant transformation. The ability to distinguish the driver mutations from the passenger mutations will determine the clinical utility of NGS in cancer.
The number of genes that contain driver mutations stands at over 400,41 most of which are dominantly acting (ie, oncogenes). Undoubtedly many more remain to be discovered, as the search for cancer genes began in the cellular processes most associated with cancer. Recent systematic approaches including high-resolution array comparative genomic hybridization and sequencing have identified new cancer genes, many of them from pathways known be associated with malignancy, such as the RAS/MAPK pathway seen altered among most cancer types and the JNK signaling pathway in HR+ breast cancer.38,42,43 The protein products of these mutations often serve as targets for anticancer drugs, and the mutations themselves can be predictive of therapy outcome in some malignancies.
KRAS mutations at key sites, most commonly in codons 12 and 13, are seen in approximately 30% to 40% of CRC and result in the constitutive activation of KRAS-associated signaling.44,45 The presence of mutant KRAS in CRC correlates with a poor prognosis.46,47 Randomized clinical trials have demonstrated EGFR-targeted therapeutics ineffective in patients with KRAS mutations45,48 making KRAS a useful biomarker in therapy prediction. BRAF mutations in melanomas are another well-characterized example of biomarker for prognosis and treatment response. BRAF mutations are seen in approximately half of melanoma cases49 with 1 mutation, V600E (substitution of the codon for valine at position 600 with glutamic acid), accounting for 90% of these cases. Recent phase 3 randomized clinical trials have shown increased rates of overall and progression-free survival in previously untreated V600E mutation-positive melanoma using the BRAF kinase inhibitor vemurafenib compared to standard of care.50
Genetic characterization of those tumors that fail to respond to BRAF inhibitors has yielded more insight into the molecular mechanisms of tumors and identified mutations responsible for the resistance. Recent phase 1 and phase 2 trials combining BRAF inhibition with MEK inhibition has shown promise in addressing this problem.51 However, the molecular markers of one tissue type do not easily translate to other tissues. The V600E mutation is common in multiple cancer types, such as CRC where it constitutes approximately 10% of cases.52 V600E-positive CRCs have limited response to vemurafenib suggesting the contribution of other factors in determining treatment response and prognosis such as microenvironment, additional acquired mutations, and even the patient’s genetic background.53
NGS approaches in cancer will allow the capture of all the genetic changes in a particular malignancy, allowing for the identification of known driver mutations, as well as prognostic and therapeutic biomarkers. In addition, NGS detects tumor heterogeneity and constitutional nucleotide changes that may affect disease course and not tested for through standard testing algorithms. Putting all the pieces back together, combining traditional pathologic and cytogenetic approaches with NGS allows for a more robust picture of 1 disease in 1 individual, which is personalized medicine.
Data Processing and Variant Interpretation
The massive amounts of data produced by NGS is perhaps the most daunting challenge facing clinical laboratories as protocols for data filtering, interpretation, reporting, and storage must be established. After the targeted regions are sequenced, the reads must be mapped back to a reference sequence in a computationally intensive process, which generates a list of sites where the sample sequence does not match the reference or control sequence. These sites of nucleotide changes must then be filtered to generate a manageable list of nucleotide changes for the genetic professional to analyze and make decisions as to clinically relevant changes. Although there are several commercially available data filtering software packages, many laboratories choose to develop their own bioinformatics pipeline in-house. Population-specific genetic variation is not yet well catalogued and the frequency of rare deleterious mutations in healthy populations is not well understood or defined. The present databases, such as dbSNP, HGMD, 1000 genomes, Exome Variant Server, and others can guide the interpretation of variants, but have become contaminated with true mutations (in the case of dbSNP) and benign polymorphisms (in the case of HGMD). In the case of cancer testing, the cancer sequence will be compared back with normal sequence from the same individual to facilitate the filtering of unique polymorphisms in that individual but separating the driver from the passenger mutation is not easy because of heterogenous nature of the tumors and the their ability to mutate and become resistant to therapy. These factors and others make the interpretation of nucleotide changes intensive and often inconclusive.
The increased number of genes sequenced in a single-test necessarily results in an increase in the number of VOUS, particularly if genes unrelated to the phenotype or poorly characterized are interrogated. VOUS provide unique challenges to the laboratorian and clinician as they must be reported, but by definition their impact is currently unknown. A VOUS may very well be related to the phenotype or, in the case of cancer, with the tumor type, but there is insufficient evidence at the time of report to say that the change is definitively linked to disease. In time, the significance of the VOUS may become clear, and laboratories must be diligent in reclassifying VOUS as this information becomes available. MacArthur et al54 recently estimated approximately 100 genuine loss of function mutations per healthy individuals, with about 20 genes completely inactivated. The apparent redundancy in the human genome highlights the challenges in discerning true causative mutations from ultra-rare polymorphisms that have no clinical impact.
In addition to the rare deleterious variants discussed above, WES will detect clinically relevant mutations unrelated to the disease tested for, that is, secondary or incidental findings, as well as identify carrier status, risk alleles, and pharmacogenetic markers. Guidelines for reporting such variants are still evolving, and a significant investment in counseling time will be needed to properly inform the individual about the implications of these data not just on them, but on their family as well.
Although there are limitations, the advantages to WES and WGS are clear, and there have been many published successes in a variety of human disease categories. Congenital disorders of glycosylation (CDGs) are multisystem disorders with onset in infancy, which claims the life of 20% of patients within the first 5 years of life. Clinical diagnosis of CDGs is challenging due to the wide spectrum of symptoms at presentation and the absence of reliable biomarkers for many forms of disease. Further challenging diagnosis is the involvement of a significant percentage of the genome, estimated at 1% to 2%, in glycosylation. WES and WGS strategies are well suited to address these types of genetic heterogenous diseases. In a recent study by Jones et al,55 targeted NGS analysis of 25 CDG genes failed to uncover a causative mutation in child with a biochemical analysis suggestive of CDG. WES revealed 2 mutations in the DDOST gene, which was not previously associated with CDGs. Sanger sequencing of the variants and functional analysis confirmed these nucleotide changes as the causative mutations in this individual.
Critical to the success of NGS testing is the ability to interpret the numerous variants found through this powerful testing approach. Integrating information from all clinical areas will be essential in providing a complete picture of the disease. The increase in the number of nucleotide changes will put a burden on the pathologist to interpret these novel changes, many of which will not have been previously reported. Without clinical context or phenotype information, this task is not possible. Historically, molecular geneticists want to know they are in the appropriate gene when interpreting a variant. At times even for single-gene testing this is a problem even though the ordering physician has made a clinical decision based on the phenotype. Expanding the number of genes adds further complexity to data interpretation. Variants may be found in multiple clinically relevant genes that must be categorized.
The importance of phenotype data to this process cannot be overstated in inherited and somatic diseases. The pathologist will play a critical role in this process as they have the expertise to select the relevant cells for genomic analysis for a particular disease and aid in interpretation of the sequence changes (Fig. 1).
Clinical Utility of Next-generation Sequencing in Oncology
Pilot studies investigating the feasibility of NGS approaches are already underway. Roychowdhury et al5 integrated data obtained from WGS and WES sequencing of tumor, WES of normal tissue from the same individual, and RNA-sequencing (RNA-seq). RNA-seq evaluates the mRNA expression in the tumor, providing corroborative information on gene expression, point mutations, and gene fusion events. Identification of these gene fusion events, whether by detecting the abnormal transcript through RNA-seq or by detecting the fusion gene through WGS, contributes substantially to the understanding of the disease. WGS was successfully used to identify a cryptic fusion oncogene in a patient with acute promyelocytic leukemia with no X-RARA fusion identified by traditional methods.56 In this case, WGS was able to detect PML-RARA fusion created by an insertion event, thereby altering medical care of the patient. WGS is able to deliver clinically actionable results in an appropriate time frame of 4 to 7 weeks.5,56 The implementation of WGS in the clinical setting highlighted several anticipated challenges, the same challenges that have been described here and elsewhere: identifying patients who would benefit from such an approach, the handling of incidental findings (counseling issues), and the interpretation of results.
It is important to realize that there are several key differences in the NGS workflow for rare disease testing compared to cancer testing (Fig. 2). Germline studies require the patient sequence to be mapped and compared with a reference sequence, most typically the NCBI or Genome Reference Consortium. Comparison to this reference genome allows for the identification of single nucleotide changes, insertions, and deletions in the patient genome that then may be classified as benign, pathogenic, or of unknown clinical significance. However, this reference sequence contains gaps and sequencing errors, which can lead to improper identification of a nucleotide change. Data quality may also suffer as a result of the removal of a number of accurate, high-quality reads that may not align properly with a flawed reference sequence. Somatic testing relies on the use of normal tissue from the same individual to serve as the reference. This approach leads to a clear advantage over the germline approach in the interpretation of variants, as variants seen in both normal and malignant tissue can be quickly excluded from consideration. Cancer testing will also require NGS not once, but multiple times over the course of disease to monitor disease progression and/or treatment response.
NGS is revolutionizing clinical genomic testing for cancer, rare disease, prenatal, and infectious diseases. Pathologists are well suited to usher this technology into the clinical laboratory given their diagnostic expertise and familiarity with molecular testing, including genomic analysis. However, NGS testing presents a new set of challenges for the clinical laboratory and its implementation will require significant upfront investment, both in time and personnel commitments and financial commitments. Critical decisions will need to be made concerning the testing strategy, WES or targeted panels, and these decisions will need to be based on an individual’s laboratory test menu and sample type. Significant computational resources will be needed to process the volumes of data generated by NGS tests. Consistent variant classification will depend on the accessibility to well-annotated, up-to-date clinical databases. The cost of these resources is decreasing making NGS more feasible for many pathology laboratories.
As we move into the next era of clinical genomic testing, training of future pathologists must be geared to include a comprehensive understanding of genomics in medicine. The incorporation of NGS technology will only increase as prices decrease and resources improve. This increase will necessarily result in a better understanding of the human genome, leading to improved diagnosis of rare disease, more effective targeted therapies, and a greater appreciation of one’s genetic background to the progression and management of disease. Together with other genetic professionals, pathologists are poised to provide effective therapeutic options for patients based not only on genomics, but on the unique clinical picture of each individual.
1. Ng SB, Bigham AW, Buckingham KJ, et al .Exome sequencing identifies MLL2 mutations as a cause of Kabuki syndrome.Nat Genet. 2010; 42:790–793.
2. Neale BM, Kou Y, Liu L, et al .Patterns and rates of exonic de novo mutations in autism spectrum disorders.Nature. 2012; 485:242–245.
3. O’Roak BJ, Vives L, Girirajan S, et al .Sporadic autism exomes reveal a highly interconnected protein network of de novo mutations.Nature. 2012; 485:246–250.
4. Sanders SJ, Murtha MT, Gupta AR, et al .De novo mutations revealed by whole-exome sequencing are strongly associated with autism.Nature. 2012; 485:237–241.
5. Roychowdhury S, Iyer MK, Robinson DR, et al .Personalized oncology through integrative high-throughput sequencing: a pilot study.Sci Transl Med. 2011; 3:111ra121
6. Chiu RW, Chan KC, Gao Y, et al .Noninvasive prenatal diagnosis of fetal chromosomal aneuploidy by massively parallel genomic sequencing of DNA in maternal plasma.Proc Natl Acad Sci USA. 2008; 105:20458–20463.
7. Fan HC, Blumenfeld YJ, Chitkara U, et al .Noninvasive diagnosis of fetal aneuploidy by shotgun sequencing DNA from maternal blood.Proc Natl Acad Sci USA. 2008; 105:16266–16271.
8. Palomaki GE, Deciu C, Kloza EM, et al .DNA sequencing of maternal plasma reliably identifies trisomy 18 and trisomy 13 as well as Down syndrome: an international collaborative study.Genet Med. 2012; 14:296–305.
9. Palomaki GE, Kloza EM, Lambert-Messerlian GM, et al .DNA sequencing of maternal plasma to detect Down syndrome: an international clinical validation study.Genet Med. 2011; 13:913–920.
10. Sparks AB, Struble CA, Wang ET, et al .Noninvasive prenatal detection and selective analysis of cell-free DNA obtained from maternal blood: evaluation for trisomy 21 and trisomy 18.Am J Obstet Gynecol. 2012; 206:319 e1–319 e9.
11. Highlander SK, Hulten KG, Qin X, et al .Subtle genetic changes enhance virulence of methicillin resistant and sensitive Staphylococcus aureus. BMC Microbiol. 2007; 7:99
12. Holt KE, Parkhill J, Mazzoni CJ, et al .High-throughput sequencing provides insights into genome variation and evolution in Salmonella typhi. Nat Genet. 2008; 40:987–993.
13. Richter BG, Sexton DP .Managing and analyzing next-generation sequence data.PLoS Comput Biol. 2009; 5:e1000369
14. Stenson PD, Mort M, Ball EV, et al .The Human Gene Mutation Database: 2008 update.Genome Med. 2009; 1:13
15. Peltomaki P .Role of DNA mismatch repair defects in the pathogenesis of human cancer.J Clin Oncol. 2003; 21:1174–1179.
16. Miyaki M, Konishi M, Tanaka K, et al .Germline mutation of MSH6 as the cause of hereditary nonpolyposis colorectal cancer.Nat Genet. 1997; 17:271–272.
17. Berends MJ, Wu Y, Sijmons RH, et al .Molecular and clinical characteristics of MSH6 variants: an analysis of 25 index carriers of a germline variant.Am J Hum Genet. 2002; 70:26–37.
18. Goel A, Nguyen TP, Leung HC, et al .De novo constitutional MLH1 epimutations confer early-onset colorectal cancer in two new sporadic Lynch syndrome cases, with derivation of the epimutation on the paternal allele in one.Int J Cancer. 2011; 128:869–878.
19. Senter L, Clendenning M, Sotamaa K, et al .The clinical phenotype of Lynch syndrome due to germ-line PMS2 mutations.Gastroenterology. 2008; 135:419–428.
20. Kuiper RP, Vissers LE, Venkatachalam R, et al .Recurrence and variability of germline EPCAM deletions in Lynch syndrome.Hum Mutat. 2011; 32:407–414.
21. Thibodeau SN, Bren G, Schaid Microsatellite D .instability in cancer of the proximal colon.Science. 1993; 260:816–819.
22. Aaltonen LA, Peltomaki P, Leach FS, et al .Clues to the pathogenesis of familial colorectal cancer.Science. 1993; 260:812–816.
23. Ionov Y, Peinado MA, Malkhosyan S, et al .Ubiquitous somatic mutations in simple repeated sequences reveal a new mechanism for colonic carcinogenesis.Nature. 1993; 363:558–561.
24. Clark MJ, Chen R, Lam HY, et al .Performance comparison of exome DNA sequencing technologies.Nat Biotechnol. 2011; 29:908–914.
25. Asan, Xu Y, Jiang H, et al .Comprehensive comparison of three commercial human whole-exome capture platforms.Genome Biol. 2011; 12:R95
26. Razin SV, Ioudinkova ES, Trifonov EN, et al .Non-clonability correlates with genomic instability: a case study of a unique DNA region.J Mol Biol. 2001; 307:481–486.
27. Sipos B, Massingham T, Stutz AM, et al .An improved protocol for sequencing of repetitive genomic regions and structural variations using mutagenesis and next generation sequencing.PLoS One. 2012; 7:e43359
28. Kaminsky EB, Kaul V, Paschall J, et al .An evidence-based approach to establish the functional and clinical significance of copy number variants in intellectual and developmental disabilities.Genet Med. 2011; 13:777–784.
29. Cooper GM, Coe BP, Girirajan S, et al .A copy number variation morbidity map of developmental delay.Nat Genet. 2011; 43:838–846.
30. Kaye J, Boddington P, de Vries J, et al .Ethical implications of the use of whole genome methods in medical research.Eur J Hum Genet. 2010; 18:398–403.
31. Nowell PC, Hungerford DA .Chromosome studies on normal and leukemic human leukocytes.J Natl Cancer Inst. 1960; 25:85–109.
32. Dewald GW .Cytogenetic and FISH studies in myelodysplasia, acute myeloid leukemia, chronic lymphocytic leukemia and lymphoma.Int J Hematol. 2002; 76:suppl 2 65–74.
33. Kolialexi A, Tsangaris GT, Kitsiou S, et al .Impact of cytogenetic and molecular cytogenetic studies on hematologic malignancies.Anticancer Res. 2005; 25:2979–2983.
34. Foulkes WD .Inherited susceptibility to common cancers.N Engl J Med. 2008; 359:2143–2153.
35. Dohner H, Stilgenbauer S, Benner A, et al .Genomic aberrations and survival in chronic lymphocytic leukemia.N Engl J Med. 2000; 343:1910–1916.
36. Maciejewski JP, Mufti GJ .Whole genome scanning as a cytogenetic tool in hematologic malignancies.Blood. 2008; 112:965–974.
37. Edelmann J, Holzmann K, Miller F, et al .High-resolution genomic profiling of chronic lymphocytic leukemia reveals new recurrent genomic alterations.Blood. 2012; 120:4783–4794.
38. Greenman C, Stephens P, Smith R, et al .Patterns of somatic mutation in human cancer genomes.Nature. 2007; 446:153–158.
39. Wood LD, Parsons DW, Jones S, et al .The genomic landscapes of human breast and colorectal cancers.Science. 2007; 318:1108–1113.
40. Hanahan D, Weinberg RA .The hallmarks of cancer.Cell. 2000; 100:57–70.
41. Futreal PA, Coin L, Marshall M, et al .A census of human cancer genes.Nat Rev Cancer. 2004; 4:177–183.
42. Pleasance ED, Cheetham RK, Stephens PJ, et al .A comprehensive catalogue of somatic mutations from a human cancer genome.Nature. 2010; 463:191–196.
43. Kan Z, Jaiswal BS, Stinson J, et al .Diverse somatic mutation patterns and pathway alterations in human cancers.Nature. 2010; 466:869–873.
44. Samowitz WS, Curtin K, Schaffer D, et al .Relationship of Ki-ras mutations in colon cancers to tumor location, stage, and survival: a population-based study.Cancer Epidemiol Biomarkers Prev. 2000; 9:1193–1197.
45. Amado RG, Wolf M, Peeters M, et al .Wild-type KRAS is required for panitumumab efficacy in patients with metastatic colorectal cancer.J Clin Oncol. 2008; 26:1626–1634.
46. Andreyev HJ, Norman AR, Cunningham D, et al .Kirsten ras mutations in patients with colorectal cancer: the “RASCAL II” study.Br J Cancer. 2001; 85:692–696.
47. Esteller M, Gonzalez S, Risques RA, et al .K-ras and p16 aberrations confer poor prognosis in human colorectal cancer.J Clin Oncol. 2001; 19:299–304.
48. Van Cutsem E, Kohne CH, Hitre E, et al .Cetuximab and chemotherapy as initial treatment for metastatic colorectal cancer.N Engl J Med. 2009; 360:1408–1417.
49. Safaee Ardekani G, Jafarnejad SM, Tan L, et al .The prognostic value of BRAF mutation in colorectal cancer and melanoma: a systematic review and meta-analysis.PLoS One. 2012; 7:e47054
50. Chapman PB, Hauschild A, Robert C, et al .Improved survival with vemurafenib in melanoma with BRAF V600E mutation.N Engl J Med. 2011; 364:2507–2516.
51. Flaherty KT, Infante JR, Daud A, et al .Combined BRAF and MEK inhibition in melanoma with BRAF V600 mutations.N Engl J Med. 2012; 367:1694–1703.
52. English DR, Young JP, Simpson JA, et al .Ethnicity and risk for colorectal cancers showing somatic BRAF V600E mutation or CpG island methylator phenotype.Cancer Epidemiol Biomarkers Prev. 2008; 17:1774–1780.
53. Prahallad A, Sun C, Huang S, et al .Unresponsiveness of colon cancer to BRAF(V600E) inhibition through feedback activation of EGFR.Nature. 2012; 483:100–103.
54. MacArthur DG, Balasubramanian S, Frankish A, et al .A systematic survey of loss-of-function variants in human protein-coding genes.Science. 2012; 335:823–828.
55. Jones MA, Ng BG, Bhide S, et al .DDOST mutations identified by whole-exome sequencing are implicated in congenital disorders of glycosylation.Am J Hum Genet. 2012; 90:363–368.
56. Welch JS, Westervelt P, Ding L, et al .Use of whole-genome sequencing to diagnose a cryptic fusion oncogene.JAMA. 2011; 305:1577–1584.
next-generation sequencing; genetic testing; cancer; rare disease; clinical testing; personalized genomics; pathology
Copyright © 2013 by Lippincott Williams & Wilkins
Highlight selected keywords in the article text.