A narrative review of cancer molecular diagnostics: past, present, and future : Journal of Bio-X Research

Secondary Logo

Journal Logo

Review Articles

A narrative review of cancer molecular diagnostics: past, present, and future

Yao, Jinjuana,*; Zhai, Qihui (Jim)b

Author Information
Journal of Bio-XResearch 5(4):p 145-150, December 2022. | DOI: 10.1097/JBR.0000000000000136
  • Open

Abstract

Introduction

Since the first human genome project was completed in 2003,[1] cancer medicine entered the Genomic Era. Molecular diagnostics was incorporated in modern pathology and cancer work-up to identify genetic alterations associated with diagnosis, prognosis, and treatment. Genomic medicine, including targeted therapy and immunotherapies, is becoming important components of cancer treatment. Here, we review the past, present, and possible future of molecular diagnostics, aiming to provide a comprehensive summarization of molecular diagnostic technologies, testing platforms, and applications in cancer management.

Retrieval strategy

Literature review was electronically performed using PubMed database. English language and full-text articles published between 2003 and 2021 were included in this non-systematic review. The authors searched the PubMed database to identify relevant publications. The literature search strategy was conducted as follows (1) cancer, (2) molecular diagnostics, (3) technologies, and (4) future development. The authors screened the reference list of included studies to identify other potentially useful studies. First, the authors screened the titles and abstracts, and then the full texts for keywords, such as “cancer molecular diagnostic,” and “molecular technologies” to find those that were potentially suitable. The data extraction process focused on the information about relevant literature.

The evolution of molecular diagnostic methodologies

Molecular diagnostics is a collection of techniques that can be used in the context of various diseases to detect genetic alterations, to assist in diagnosis, classification, and progression prediction, and to monitor treatment response.[2]

Molecular diagnostics techniques first began to be developed in research laboratories in the middle of last century. Recombinant DNA and cDNA cloning were the technologies used to analyze gene sequences in the early phase of molecular diagnosis.[3] Sanger sequencing was created in 1977 and became the gold standard for gene sequencing in clinical laboratories for several decades. Subsequently, polymerase chain reaction (PCR) was invented, and many associated methods were also developed. The first next-generation sequencing (NGS) instrument was developed in 2000, while the launch of the Illumina HiSeq and ThermoFisher Ion Torrent sequencers in 2010 fully opened the door to large-scale DNA sequencing. Subsequently, innovation focused on long-read sequencing, also known as third-generation sequencing. The Nanopore sequencer is a representative third-generation sequencer that may reshape the landscape of nucleic acid sequencing and molecular diagnostics in the near future.[4]

Current molecular diagnostic methodologies

Sanger sequencing

Sanger sequencing is also called chain termination sequencing. It was the most popular sequencing method before NGS became dominant and is still considered the gold standard. Frederick Sanger won the 1980 Nobel Prize in Chemistry for developing this approach to nucleic acid sequencing.[5] The sequencing reaction includes primers, DNA polymerase, deoxynucleic acid, and dideoxynucleic acid. In the process of sequence extension, dideoxynucleic acid is randomly integrated, resulting in termination of sequence extension. The collection of fragments of various lengths are then subjected to capillary electrophoresis, and the bases are read based on the dideoxynucleic acid fluorescence signals.[6]

Sanger sequencing can rapidly detect single-gene or single-locus mutations, such as EGFR and KRAS mutations in lung cancer, KIT and PDGFRA mutations in gastrointestinal stromal tumor. However, it is time-consuming and low-throughput. In addition, the sensitivity is not optimal; when applied to cancer diagnosis, at least 50% of tumor content are required to be distinguished from the background and achieve a sensitivity of 25%.

Using locked nucleic acid probes can block sequencing of wild-type fragments, thereby greatly increasing the limit of detection to 1% or even lower.[7]

Polymerase chain reaction

PCR was invented in 1983 by Dr Mullis, the winner of the 1993 Nobel Prize in Chemistry.[8] The reaction includes a double-stranded DNA template, primers, nucleotides, and DNA polymerase. After denaturation of the double-stranded template, the primers bind and extend from the 5′ to the 3′ end. The newly generated copies of the DNA are used as templates for further replication so that the original template is amplified exponentially in a chain reaction. The amplified product can be visualized directly by gel electrophoresis; subjected to fragment analysis and restriction fragment length polymorphism (RFLP) analyses based on the size of the PCR fragments[9]; or subjected to melting curve analysis based on its dissociation properties.[10] The PCR products can also be sequenced by Sanger sequencing, pyrosequencing, single-base extension, or NGS. Most of these methods incorporate fluorescent tags into the PCR product that are detected by optical systems to identify the genomic changes.

Fragment analysis and RFLP are variations on the same technique. Both involve PCR amplification of a template with fluorescently labeled primers followed by capillary electrophoresis to sort PCR products by fragment size (length). Fragment analysis allows for the rapid detection of small and medium-sized insertions and deletions (50 bases to hundreds of bases long), some of which could be challenging for other technologies to detect or could be easily missed by massive parallel sequencing. RFLP is a simple, rapid, and cost-effective way to detect the presence of single-nucleotide variants or methylation at a given site using sequence-specific restriction enzymes that cut PCR fragments based on the presence of specific palindromic sequences.

The examples of fragment analysis and RFLP in cancer molecular diagnostics include the clonality studies of lymphomas, the detection of NPM1 and FLT3 mutations in acute myeloid leukemia.

Quantitative PCR

Quantitative PCR evolved from PCR. In comparison with PCR, which detects the end-product of amplification, quantitative PCR analyzes the number of DNA copies in the exponential phase of the PCR reaction, which is a more accurate reflection of the starting amount of template.[11] Quantitative reverse-transcription PCR (qRT-PCR) is a combination of reverse transcription (RT) of RNA to complementary DNA (cDNA) and quantitative detection that is frequently used in gene expression analysis.[12] These methods use probes with fluorescent dyes that are measured by an optical system; then, standard curves or comparative thresholds are used to calculate the number of copies in the starting material. qRT-PCR is widely used for minimal disease detection in chronic myeloid leukemia to monitor the BCR::ABL1 fusion, in acute myeloid leukemia for NPM1 mutation, RUNX1::RUNX1T1 and PML::RARA fusions.

Digital PCR

Digital PCR (dPCR) is a relatively new technique that is used to directly quantify the amplified nucleic acid. In contrast to traditional PCR, which amplifies the entire sample in a single reaction, dPCR amplification is partitioned such that thousands and millions of partitions are generated from a single sample.[13] The competitive inhibition that is normally a factor in a standard PCR reaction is decreased, and the sensitivity of detection is improved. dPCR is widely used in absolute quantification, detection of copy number variation, gene expression analysis, and mutation identification. Because of its high sensitivity, it is also used to analyze liquid biopsy specimens.

Next-generation sequencing

NGS was developed more than a decade ago with the aim of increasing detection capacity and capturing all genetic alterations at the same time. In recent years, it has come to be widely applied in clinical settings, using many different sequencing technologies or platforms.[14]

The most used platforms are the Illumina and Ion Torrent Semiconductor sequencers.[15] These two sequencers have different chemistries. Illumina detects fluorescence signals, while Ion Torrent detects current change. In Illumina sequencing, the templates are copied on a flow cell using four differently colored fluorescently tagged deoxyribonucleotide triphosphates (dNTPs). During one round of reaction, only the base complementary to the template is incorporated into the sequencing primer or growing chain. The florescent base is excited by a laser, and its unique emission spectrum is captured by the built-in camera. The sequence of the template is determined based on the readout of the signals that occur at the same position in sequential pictures.

In contrast, the Ion Torrent platform is not an optical system, but rather uses current as a signal. The DNA library fragments are clonally amplified on the surface of a bead. The sequence of the fragment in each bead is read in a semiconductor chip with micromachined wells that have an ion sensitive layer and an ion sensor to detect the hydrogen ions that are released during the incorporation of the deoxyribonucleotide triphosphate into the template DNA.

In clinical laboratories, enrichment methods are used to select regions of interest for targeted sequencing. Two methods of enrichment that are currently used are hybrid capture and amplicon capture. These technologies can enhance the assay sensitivity, lower cost, shorter turnaround time, and better support for therapeutic decision-making and patient management, in comparison with single gene assays.[16]

Targeted NGS assays are becoming an important integral part of cancer driver mutation detection, as well as the assessment of microsatellite instability and tumor mutation burden.

RNA sequencing

RNA sequencing (RNA-seq) is a technique that detects gene fusions and analyzes gene expression using NGS.[17] Gene fusions, especially those involving the tyrosine kinase domain of growth factors, are known driver mutations for many cancer types. These mutations are diagnostic, targetable, or both.[18,19] However, clinical use of targeted DNA-seq is limited in that this technique cannot detect all structural variants, because these events commonly involve introns, which are too long to tile, contain unmappable repetitive elements, or have genomic breakpoints in alternative introns that are not covered by the panel design. RNA-seq can capture the junction of exons from the two fusion partner genes and offers a direct approach to detecting fusions. In addition, fusion genes may have higher expression at the RNA level, which increases the sensitivity of RNA-seq.[19] Targeted RNA-seq typically involving using sequence-specific primers to known partner genes and universal primers for unknown genes to further increase the detection sensitivity.[19]

Liquid biopsy/cell-free DNA assay

Liquid biopsy, also known as a cell-free DNA (cfDNA) assay, detects circulating tumor DNA (ctDNA) in the blood. In recent years, this technique has been developed for use in cancer monitoring, evaluation of drug response, diagnosis, and even early detection. Liquid biopsy provides a comprehensive analysis of genomic alterations in both the primary tumor and distant metastases.[20] It is much less invasive than tissue biopsy, and samples can be collected multiple times through the disease process, providing a dynamic picture of evolutions in genetic cancer alterations. Depending on the design, liquid biopsy can be a single-gene assay or can involve a small or large panel of genes to detect single-nucleotide variations, small insertions and deletions, structural variants, and microsatellite instability. In cancer patients, the proportion of ctDNA to total cfDNA varies considerably and depends on tumor type, stage, and size. Some tumor types are more prone to shed than others. Usually, the later the stage and the larger the size of the tumor, the higher the proportion of ctDNA in the circulation is. Besides blood, cfDNA can also be obtained from cerebrospinal fluid, urine, and other body fluids.

The molecular diagnostic technologies described above are summarized in Table 1.

Table 1 - A comparison of common molecular techniques
Molecular techniques Variant types
SNVs Small indels CNV SVs Sensitivity (%)
Sanger sequencing 25
fragment analysis and RFLP ± 5
Allele-specific PCR 1–5
qPCR ± ± <1
qRT-PCR 0.001
dPCR ± <1
NGS-Amplicon capture ± 5–10
NGS-Hybridization capture 2–5
NGS-liquid biopsy ± <1
NGS-RNA sequencing ± 5
±=means may or may not be able to detect those variant types; or with limited ability of detection, CNV=copy number variation, NGS=next-generation sequencing, RFLP=restriction fragment length polymorphism, SNV=single-nucleotide variant, SV=structural variant.

Future molecular diagnostic technologies

Nanopore sequencing

Nanopore sequencing technology was invented in 2014. Currently, it is used in scientific research. Unlike other sequencing technologies that were already available in the clinical laboratories, Nanopore sequencing does not require PCR amplification and can generate long reads (10–100kb), reduce cost and amplification errors, and improve de novo assembly and mapping quality. Nanopore sequencing devices are portable, fast, and affordable.[21] In the future, clinical use of this technique is expected to yield breakthroughs in germline mutation detection, virology, and gene fusion testing. In addition, because these devices are portable, it is expected that this technique will also be used in the field at point of care. The disadvantages of Nanopore are its relatively low accuracy and precision, both of which need to be improved before clinical use.[22]

Single-cell sequencing

Single-cell sequencing technology is also increasing in popularity. Single-cell DNA sequencing detects DNA sequences and genomic alterations at the level of individual cells, captures information regarding spatial and temporal heterogeneity within a given tumor, and provides information related to tumor evolution, relapse, and metastasis.[23] Single-cell sequencing of RNA or epigenetic modifications further elucidates phenotypic changes by providing information about gene and protein expression.[24] Genomic, transcriptomic, and epigenetic information obtained at the level of the individual cell is becoming the basis for targeted molecular approaches in cancer therapy.

The disadvantage of single-cell sequencing is the requirement for fresh or frozen tumor tissue to isolate individual tumor cells. Its application to hematological diseases, especially myeloid malignancies, is more extensive through the use of flow cytometry cell sorting and selection.[25,26]

Clustered regularly interspaced short palindromic repeat technology

Clustered regularly interspaced short palindromic repeat (CRISPR) gene editing technology is an emerging tool that is used for cancer mutation detection. Genetic alterations at the DNA and RNA levels are detected by using CAS12 and CAS13 in combination with the detector and Sherlock technologies.[27–29] CRISPR technology can be used to shear wild-type alleles to enrich the mutant allele, which, when followed by amplification and NGS, selectively improves the detection sensitivity for low-frequency mutations.

CRISPR will be a valuable addition to clinical molecular diagnostics because of its clean reads, stability, portability, and low cost.

Machine learning and artificial intelligence

Machine learning (ML) and artificial intelligence (AI) approaches have been developed to infer tumor origin based on large-panel sequencing data, including hotspot mutations, insertions and deletions, focal or genome-wide copy number alterations, structural variants, mutational signatures, and clinical parameters. Computational biologists from Memorial Sloan Kettering (MSK) Cancer Center designed and trained an algorithmic classifier using comprehensive genomic profiling data from 7791 tumors representing 22 cancer types, generated during clinical prospective tumor sequencing, using the MSK-IMPACT (integrated mutation profiling of actionable cancer targets) targeted panel. They reported that the correct tumor type was predicted in 73.8% of the total training set (5748/7791 patients) and in 74.1% of an independent cohort (8623/11,644 patients). The accurate prediction rate in plasma cell-free DNA reached 75.0%. These findings indicate that AI technology can further enhance the utility of molecular testing by providing additional tumor origin information, especially in those tumors that have been classified as cancer of unknown primary based on histologic and immunophenotypic features.[30,31]

The evolution of molecular testing principles

Molecular testing principles for solid tumors have evolved with the discovery of cancer biomarkers and the development of corresponding treatments.

In 1998, trastuzumab was approved to treat HER2-positive metastatic breast cancer,[32] which was the first targeted therapy approved by the Food and Drug Administration (FDA). In 2004, treatment of lung adenocarcinomas with activating mutations in EGFR with gefitinib and erlotinib was reported.[33–35] Since then, more tumor-specific genetic alterations and associated drugs have been identified and approved, including the KRAS resistance mutation and cetuximab in colorectal cancer[36] and the BRAF V600E mutation and vemurafenib in melanoma.[37] In addition, more biomarkers in the same tumor types continue to be identified, such as mutations in ALK, ROS1, KRAS, BRAF, RET, and many more in lung adenocarcinoma.[38]

Molecular diagnostics began as single-gene, single-platform testing >20 years ago.[39] However, with the expansion of cancer biomarkers and targeted therapies, traditional low-throughput testing approaches cannot capture all relevant biomarkers within the time frame needed for clinical management and from limited tissue samples. Molecular diagnostics laboratories have therefore had to develop more comprehensive platforms to capture all clinically indicated alterations. Foundation One and MSK-IMPACT, representative examples of large, targeted NGS panels, were launched in 2011 and 2014, respectively. These two panels apply the pan-cancer strategy, testing all solid tumors for a large panel of cancer genes, regardless of tumor type and whether the tumor has known biomarkers. This type of strategy has identified many patients with different cancer types who could potentially benefit from basket clinical trials. Since 2017, the FDA has approved three pan-cancer biomarkers and associated drugs based on basket clinical trials, including high microsatellite instability and pembrolizumab,[40]NTRK fusion and larotrectinib,[41,42] high tumor mutation burden and pembrolizumab (Fig. 1).[43,44] In addition, more pan-cancer biomarker candidates are emerging biomarkers, including BRCA1/BRCA2, PD-L1, RET, FGFR, and so on.[45]

F1
Figure 1.:
FDA CDx approval and targeted treatment. ABL1=ABL Proto-Oncogene 1, BCR=B-cell receptor, CDK=cyclin dependent kinase, CDx=companion diagnostics, CMS=Centers for Medicare & Medicaid Services, EFGR=epidermal growth factor receptor, FDA=US Food and Drug Administration; FGFR: fibroblast growth factor receptor, KRAS=Kirsten rat sarcoma virus, MET=mesenchymal epithelial transition factor receptor, MSI=microsatellite instability, MSK-IMPACT=Memorial Sloan Kettering-Integrated Mutation Profiling of Actionable Cancer Targets, NGS=next-generation sequencing, NTRK=neurotrophic tyrosine receptor kinase, NYS=New York State, RET=rearranged during transfection.

Currently, molecular laboratories need to build up a portfolio of testing platforms to meet clinical requirements of being both rapid and comprehensive, which is not currently the case for either single-gene or cancer panel testing. Therefore, an algorithm is occasionally needed for testing stratification, especially in tumor types with many targets; for example, in lung adenocarcinoma-single gene testing is used to identify EGFR, ALK, or KRAS alterations, followed by NGS testing for the detection of mutations, copy number and structural variants in other driver genes, as well as the assessment of microsatellite instability tumor mutation burden to complete the biomarker identification.

The role of pathologists in the evolution of molecular diagnostics

Molecular genetic pathology was created as a joint subspecialty of the American Board of Medical Genetics and Genomics together with the American Board of Pathology in 1999 to provide high-quality training for physicians in the rapidly expanding field of molecular diagnosis.[46]

As an integral part of cancer diagnosis and treatment, pathology plays a critical role in patient care. Before the genomic era, surgery, radiation, and chemotherapy were the three main treatment options pursued based on histology and TNM staging. The incorporation of molecular diagnostic technologies and molecular pathology into the field of pathology has brought cancer patients the hope of a cure by providing essential information regarding genomic alterations to guide targeted and immune therapies.

In addition to providing up-to-date training for molecular pathologists, the molecular genetic pathology subspecialty of pathology has been continuously evolving to incorporate new tests and new platforms that are both rapid and comprehensive to cover all the indicated biomarkers in all tumor types and help guide clinical decision-making and selection of targeted therapies or immunotherapies (Fig. 2). Based on results from comprehensive genomic profiling, cancer patients can be grouped based on their genomic alterations, rather than their tumor types or tissue origins. These patients will be managed with targeted therapy, immunotherapy, or chemoradiation.

F2
Figure 2.:
Comprehensive genomic profiling. MSI=microsatellite instability, TMB=tumor mutation burden.

In addition, clinical laboratories are required to meet local and/or federal regulations.[47] For any assay to be applied clinically, validation of its accuracy, precision, sensitivity (lower limit of detection), and specificity are critical. To be approved as a companion diagnostic (CDx) by the FDA, the clinical validity and utility also need to be verified. Currently, the majority of clinical molecular assays are used as laboratory-developed tests, and only a small portion have received clearance or approval from FDA to serve as a CDx. Selected FDA-cleared or -approved large-panel NGS platforms are summarized in Table 2.

Table 2 - FDA-approved large panel NGS testing platforms (selected list)
Platform Genes assessed FDA approval Year Mutations Somatic/germline Sample type
MSK- IMPACT (Memorial Sloan Kettering) 505 Authorization 2017 SNVs, Indels and MSI Somatic FFPE
FoundationOne CDX (Foundation Medicine) 324 Clearance 2017 SNVs, Indels, CNAs, gene fusions, MSI, and TMB Somatic FFPE
Omics Core
(NantHealth, Inc.)
468
Whole exome
Authorization 2019 SNVs, Indels, select CNAs and gene fusions, MSI and TMB (whole exome) Somatic FFPE
PGDx elio™ (Personal Genome Diagnostics) 505 Authorization 2019 SNVs, Indels, CNAs, gene fusions, MSI and TMB Somatic FFPE
Guardant360® CDx (Guardant Health, Inc.) 74 Clearance 2020 SNVs (74), CNAs (18), fusion (6) Somatic Plasma
FoundationOne® Liquid CDx (Foundation Medicine) 311 Clearance 2020 SNVs and Indels (311), CNV (3), fusions (4) Somatic Plasma
Helix The Exome+
Assay
Whole exome Authorization 2021 SNVs and small indels Germline Saliva
±=means may or may not be able to detect those variant types; or with limited ability of detection, CNV=copy number variation, FFPE=formalin-fixed, paraffin-embedded, MSI=microsatellite instability, NGS=next-generation sequencing, RFLP=restriction fragment length polymorphism, SNV=single-nucleotide variant, SV=structural variant, TMB=tumor mutation burden.

Limitations

The limitations of this review include: possible incomplete retrieval of all relevant publications and bias of NGS-based technologies due to the authors experience and expertise.

The future of molecular diagnostics

In the past few years, the field of molecular diagnostics has undergone rapid, substantial growth, and it will continue to grow in the future. Accurate, sensitive, and rapid detection will help facilitate initial diagnosis and disease monitoring. The further development of molecular diagnostics in the field of early detection will hopefully shape the future of cancer management.

With increasing knowledge and standardization of regulations, molecular diagnostics will continue to offer additional advantages in clinical practice and patient management.

Acknowledgments

None.

Author contributions

Both authors participated in literature search, data analysis, manuscript writing and review, and approved the final version of the manuscript.

Financial support

None.

Conflicts of interest

JY reviews and reports the testing results of MSK-IMPACT, which is a FDA-approved class II in vitro diagnostic medical device. QZ has no relevant financial interest in the products or companies described in this article.

References

[1]. Collins FS, Morgan M, Patrinos A. The Human Genome Project: lessons from large-scale biology. Science 2003;300:286–290.
[2]. Netto GJ, Saad RD, Dysert PA 2nd. Diagnostic molecular pathology: current techniques and clinical applications, part I. Proc (Bayl Univ Med Cent) 2003;16:379–383.
[3]. Patrinos GP, Danielson PB, Ansorge WJ. Chapter 1—molecular diagnostics: past, present, and future. In: Patrinos GP, ed. Molecular Diagnostics (Third Edition). Academic Press2017:1–11.
[4]. Heather JM, Chain B. The sequence of sequencers: the history of sequencing DNA. Genomics 2016;107:1–8.
[5]. Shampo MA, Kyle RA. Walter Gilbert—1980 nobel prize for chemistry. Mayo Clin Proc 2003;78:588.
[6]. Sanger F, Nicklen S, Coulson AR. DNA sequencing with chain-terminating inhibitors. Proc Natl Acad Sci USA 1977;74:5463–5467.
[7]. Ishige T, Itoga S, Matsushita K, et al. Locked nucleic acid probe enhances Sanger sequencing sensitivity and improves diagnostic accuracy of high-resolution melting-based KRAS mutational analysis. Clin Chim Acta 2016;457:75–80.
[8]. Mullis KB, Faloona FA. Specific synthesis of DNA in vitro via a polymerase-catalyzed chain reaction. Methods Enzymol 1987;155:335–350.
[9]. Jarcho J. Restriction fragment length polymorphism analysis. Curr Protoc Hum Genet 2001;Chapter 2:Unit 2.7.
[10]. Pryor RJ, Wittwer CT. Real-time polymerase chain reaction and melting curve analysis. Methods Mol Biol 2006;336:19–32.
[11]. Hirakawa Y, Medh RD, Metzenberg S. Quantitative polymerase chain reaction analysis by deconvolution of internal standard. BMC Mol Biol 2010;11:30.
[12]. Gadkar V, Filion M. New developments in quantitative real-time polymerase chain reaction technology. Curr Issues Mol Biol 2014;16:1–6.
[13]. Taylor SC, Laperriere G, Germain H. Droplet Digital PCR versus qPCR for gene expression analysis with low abundant targets: from variable nonsense to publication quality data. Sci Rep 2017;7:2409.
[14]. Slatko BE, Gardner AF, Ausubel FM. Overview of next-generation sequencing technologies. Curr Protoc Mol Biol 2018;122:e59.
[15]. McCombie WR, McPherson JD, Mardis ER. Next-generation sequencing technologies. Cold Spring Harb Perspect Med 2019;9:a036798.
[16]. Mamanova L, Coffey AJ, Scott CE, et al. Target-enrichment strategies for next-generation sequencing. Nat Methods 2010;7:111–118.
[17]. Heyer EE, Deveson IW, Wooi D, et al. Diagnosis of fusion genes using targeted RNA sequencing. Nat Commun 2019;10:1388.
[18]. Stransky N, Cerami E, Schalm S, et al. The landscape of kinase fusions in cancer. Nat Commun 2014;5:4846.
[19]. Benayed R, Offin M, Mullaney K, et al. High yield of rna sequencing for targetable kinase fusions in lung adenocarcinomas with no mitogenic driver alteration detected by DNA sequencing and low tumor mutation burden. Clin Cancer Res 2019;25:4712–4722.
[20]. Stewart CM, Kothari PD, Mouliere F, et al. The value of cell-free DNA for molecular pathology. J Pathol 2018;244:616–627.
[21]. Branton D, Deamer DW, Marziali A, et al. The potential and challenges of nanopore sequencing. Nat Biotechnol 2008;26:1146–1153.
[22]. Bowden R, Davies RW, Heger A, et al. Sequencing of human genomes with nanopore technology. Nat Commun 2019;10:1869.
[23]. Baslan T, Hicks J. Unravelling biology and shifting paradigms in cancer with single-cell sequencing. Nat Rev Cancer 2017;17:557–569.
[24]. Clark SJ, Lee HJ, Smallwood SA, et al. Single-cell epigenomics: powerful new methods for understanding gene regulation and cell identity. Genome Biol 2016;17:72.
[25]. Brierley CK, Mead AJ. Single-cell sequencing in hematology. Curr Opin Oncol 2020;32:139–145.
[26]. Paguirigan AL, Smith J, Meshinchi S, et al. Single-cell genotyping demonstrates complex clonal diversity in acute myeloid leukemia. Sci Transl Med 2015;7:281re–281r2.
[27]. Mintz RL, Gao MA, Lo K, et al. CRISPR technology for breast cancer: diagnostics, modeling, and therapy. Adv Biosyst 2018;2:1800132.
[28]. Yan F, Wang W, Zhang J. CRISPR-Cas12 and Cas13: the lesser known siblings of CRISPR-Cas9. Cell Biol Toxicol 2019;35:489–492.
[29]. Shmakov SA, Sitnik V, Makarova KS, et al. The CRISPR spacer space is dominated by sequences from species-specific mobilomes. mBio 2017;8 (5):e01397–e01317.
[30]. Kleppe A, Skrede OJ, De Raedt S, et al. Designing deep learning studies in cancer diagnostics. Nat Rev Cancer 2021;21:199–211.
[31]. Penson A, Camacho N, Zheng Y, et al. Development of genome-derived tumor type prediction to inform clinical cancer care. JAMA Oncol 2020;6:84–91.
[32]. Thor A. HER2--a discussion of testing approaches in the USA. Ann Oncol 2001;12(Suppl 1):S101–S107.
[33]. Pao W, Miller V, Zakowski M, et al. EGF receptor gene mutations are common in lung cancers from “never smokers” and are associated with sensitivity of tumors to gefitinib and erlotinib. Proc Natl Acad Sci USA 2004;101:13306–13311.
[34]. Paez JG, Jänne PA, Lee JC, et al. EGFR mutations in lung cancer: correlation with clinical response to gefitinib therapy. Science 2004;304:1497–1500.
[35]. 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;350:2129–2139.
[36]. Lièvre A, Bachet JB, Le Corre D, et al. KRAS mutation status is predictive of response to cetuximab therapy in colorectal cancer. Cancer Res 2006;66:3992–3995.
[37]. Chapman PB, Hauschild A, Robert C, et al.; BRIM-3 Study Group. Improved survival with vemurafenib in melanoma with BRAF V600E mutation. N Engl J Med 2011;364:2507–2516.
[38]. Lindeman NI, Cagle PT, Aisner DL, et al. Updated molecular testing guideline for the selection of lung cancer patients for treatment with targeted tyrosine kinase inhibitors: guideline from the College of American Pathologists, the International Association for the Study of Lung Cancer, and the Association for Molecular Pathology. J Thorac Oncol 2018;13:323–358.
[39]. Sobel ME, Bagg A, Caliendo AM, et al. The evolution of molecular genetic pathology: advancing 20th-century diagnostic methods into potent tools for the new millennium. J Mol Diagn 2008;10:480–483.
[40]. Marcus L, Lemery SJ, Keegan P, et al. FDA approval summary: pembrolizumab for the treatment of microsatellite instability-high solid tumors. Clin Cancer Res 2019;25:3753–3758.
[41]. Scott LJ. Larotrectinib: first global approval. Drugs 2019;79:201–206.
[42]. U.S. Food and Drug Administration. FDA approves larotrectinib for solid tumors with NTRK gene fusions. https://www.fda.gov/drugs/fda-approves-larotrectinib-solid-tumors-ntrk-gene-fusions. Accessed November 26, 2018.
[43]. Marcus L, Fashoyin-Aje LA, Donoghue M, et al. FDA approval summary: pembrolizumab for the treatment of tumor mutational burden-high solid tumors. Clin Cancer Res 2021;27:4685–4689.
[44]. U.S. Food and Drug Administration. FDA approves pembrolizumab for adults and children with TMB-H solid tumors. https://bit.ly/30QEt40. Accessed June 16, 2020.
[45]. Yao J, Arcila ME, Ladanyi M, et al. Pan-cancer biomarkers: changing the landscape of molecular testing. Arch Pathol Lab Med 2021;145:692–698.
[46]. Byers PH. Molecular genetic pathology: coming of age in the molecular world. J Mol Diagn 1999;1:3–4.
[47]. Code of Federal Regulation: Laboratory requirements 2022. https://www.ecfr.gov/current/title-42/chapter-IV/subchapter-G/part-493. Accessed November 18, 2022.
Keywords:

cancer genomics; molecular diagnostics; past; present; future

Copyright © 2022 The Chinese Medical Association, Published by Wolters Kluwer Health, Inc.