Next Generation Sequencing In Clinical Molecular Diagnosis of Cancer: A Journey towards Personalized Diagnosis and Therapeutics
Suraksha Genomics (R&D Division of Suraksha Diagnostic), Salt Lake, Kolkata, West Bengal, India.
- *Corresponding Author:
- Tridip Chatterjee
Suraksha Genomics (R & D Division of Suraksha Diagnostic)
DD-18/1. Sector 1, Salt Lake, Kolkata-700064
West Bengal, India.
Tel: +9198313252580; 91-033-6619-1000
E-mail: [email protected]; [email protected]
Accepted date October 17, 2016
Most cancers are genetically complex and involve a series of patho-genomic DNA mutations or genetic hits. The order of development of these genetic hits is extremely diverse and this diversity in order of development of genomic mutations makes the diagnosis of most types of cancers very much challenging. So, each cancer needs to be treated as a single novel case. So, a new era has arrived when personalized diagnosis of cancer can be offered with development of massively parallel sequencing technology, popularly known as Next- Generation Sequencing. With this technology, it is quite possible to look into the substantial part of the genome of a cancer cell within a very short period of time and at an affordable cost.
Next Generation sequencing represents an effective way for capturing a substantially large amount of genetic information about a cancer. Most Next-Generation Sequencing technologies involve sequencing by synthesis, while other use sequencing by ligation. Each DNA fragment to be sequenced is actually bound to an array. DNA polymerase then adds labeled nucleotides sequentially to it. Then, a high-resolution camera captures the signal from each nucleotide, as it got integrated. The camera also takes notes on the spatial coordinates and time. Then the sequence at each spot can be inferred by a computer program/ software to generate a contiguous DNA sequence, which is termed as read. Multiple Next- Generation Sequencing technology uses different ways to capture the signal and make a read. Illumina GA/Miseq, HiSeq uses fluorescent chemistry, where as Life Technologies Ion PGM, Ion Proton uses semiconductor based chemistry to capture the signal. In nutshell, Next-Generation Sequencing has revolutionized the sequencing and cost has brought down to quite an affordable level, so that it can be implemented in clinical settings for regular diagnostic procedures in molecular pathology lab.
Next generation sequencing, Molecular diagnosis, Cancer, Personalized diagnosis.
The journey of a cell towards carcinogenesis is very unique and hence very interesting process to observe. Only for few cancer types like chronic myelogenous leukemia (CML) only single patho-genomic DNA mutation (often termed as genetic hit), is sufficient to take to the cell towards carcinogenesis. So, molecular detection of these types of malignancies is somehow easier and straight forward.
But, most of the cancers are involved with a series of patho-genomic DNA mutations or many genetic hits. For these types of cancer, which follows the multi-hit model, molecular detection becomes very difficult and challenging, because there is no specific order of development of the patho-genomic DNA mutations or genetic hits . No one can predict which mutation will develop first, which second and which will develop at last, as the order of developing the patho-genomic mutations is highly diverse and differs from one another. This huge diversity in order of development of genomic mutations makes the diagnosis of most types of cancers very much challenging . Most cancers are genetically complex and it is better to define it by the activation of specific signaling pathways, rather than relying upon a defined set of mutations . So, each cancer needs to be treated as a single novel case. Any generalized strategy of diagnosis or prognosis will not be able to give exact information of the path of the journey of a cell towards carcinogenesis.
Projects for looking at various cancer genomes are inspired by the success of Human Genome Project and increased affordability and reliability of different sequencing technologies and platforms. Affordable and reliable sequencing techniques have integrated the genome science with clinical diagnostic practice. So, we have moved into a new era of personalized diagnosis of cancers. Now-adays it is quite possible to look into the substantial part of the genome of a cancer cell within a very short period of time, by the development of massively parallel sequencing technology, popularly known as Next-Generation Sequencing (NGS).
Next Generation Sequencing Technologies and Platforms
Frederick Sanger develop the di-deoxy or chain-termination sequencing (Sanger sequencing) in 1977. Since then, the sequencing technology has been continuously upgrading in very rapid space. Now, sequencing has arrived in the era of massively parallel sequencing technology, popularly known as 'Next Generation Sequencing'. All NGS technologies are basically based on the model of massively parallel, high-throughput sequencing. It was first successfully developed by the 454 sequencing platform (acquired by Roche) in 2005. Presently, the high throughput NGS platforms are represented by companies like Roche, Illumina, Life Technologies, etc. All of them have their own amplification technology for sequencing, detection technology and limitation of maximum output (Table 1). From the Table 1, it is quite evident that, the throughput of Roche 454 GS Junior (i.e., ~400 Mb/run) and 454 GS FLX (i.e., ~700 Mb/run) are substantially lower than that of the other platforms (i.e., >100 Gb). Though different NGS platforms follow different sequencing chemistries and throughputs, but all of them offer a sequencing cost at USD (0.07-10)/Mb sharply contrast to USD 2400/Mb by conventional Sanger sequencing .
The availability of compact low/medium throughput sequencers in the past few years, exemplified by the 454 GS Junior, MiSeq and Ion PGM launched by Roche, Illumina and Life Technologies (presently taken over by Thermo Fisher Scientific Corporation) respectively. Along with the high throughput NGS platforms, these bench-top sequencers, as mentioned above, offer a lower data throughput (from tens of MB to several GB) at low costs, which are well suited for clinical applications. More importantly, in terms of the sequencing turn-around time. For the some versions of Ion PGM versions, raw sequencing data are available within 3 hours, while high throughput NGS platforms usually take more than a week to complete a sequencing run. There is an increased amount of data evaluating the clinical applications of these bench-top sequencers for detecting mutations in human cancer genes with promising results [5-7] (Table 1).
Next Generation Sequencing in Cancer Diagnosis
As cancer is a genetic disease driven by heritable or somatic mutations, new DNA sequencing technologies will have significant impact on the detection, management and treatment of the disease. Next Generation sequencing is continuously enabling worldwide collaborative efforts, such as the International Genome Consortium [ICGC]  and The Cancer Genome Atlas [TCGA] project  to catalogue the genomic landscape of thousands of cancer genomes across many disease types. Several other studies contributing to the consortia have been published [10-12]. These discoveries from individual studies will ultimately lead to better understanding of pathogenesis of the disease, leading to a new era of molecular pathology and personalized medicine . It is quite practical to imagine that every patient will sequence both their constitutional and cancer genomes to compare the changes in order to monitor disease progression, which will finally enable the clinician to perform an accurate molecular sub-typing of the disease and use molecularly guided therapies for accurate treatment.
Today the medical diagnosis has arrived to an era, when samples are no longer need to be handled differently from standard diagnostic specimens and recent advances have even enabled complex genomic data to be derived from a patient’s peripheral blood. The concept of precision medicine goes hand in hand with an accurate understanding of the cancer genome as determined by NGS. Many molecular pathology laboratories are now thinking of bringing the sequencing platforms, for making the transition to NGS from conventional approach.
|Company||Platform||Sequencing||Amplification||Run time||Read length||Max. Output|
|Roche||454 GS Junior||SBS Pyro||emPCR||10 h||400 bp (SE, PE)||35 Mb|
|454 GS FLX+||SBS Pyro||emPCR||10?23 h||700 bp (SE, PE)||700 Mb|
|Illumina||MiSeq||SBS RDT||Bridge PCR||4?27 h||36?151 bp (SE, PE)||>1 Gb|
|GAIIx||SBS RDT||Bridge PCR||2?14 days||36?151 bp (SE, PE)||=95 Gb|
|HiSeq 1000||SBS RDT||Bridge PCR||1.5?8.5 days||36?101 bp (SE, PE)||=150 Gb|
|HiScanSQ||SBS RDT||Bridge PCR||1.5?8.5 days||100 bp (SE, PE)||=150 Gb|
|HiSeq 2000||SBS RDT||Bridge PCR||2.5?11 days||36?101 bp (SE, PE)||=300 Gb|
|Life Technologies/Thermo Fisher Scientific Corporation||Ion Torrent||SBS H+||emPCR||2 h+||316+318 chip
>100 bp (SE)
|5500||SBL||emPCR||2?7 days||35?75 bp (SE, PE)||77 Gb|
|5500XL (4hp)||SBL||emPCR||2?7 days||35?75 bp (SE, PE)||155 Gb|
Note: PE: Paired-End Read; SE: Single-End Read; bp: base pairs; Mb: Megabases; Gb: Gigabases; SBL: Sequencing-By-Ligation; em PCR: emulsion PCR; Pyro: Pyro-sequencing; SBS: Sequencing By Synthesis; RDT: Reverse Dye Terminator Chemistry; H+: Hydrogen Ion Detection Chemistry.
Table 1. Different NGS platform and sequencing technologies.
Clinical Relevance of NGS Data in Cancer
Once the set of alterations are indentified within the tumor of any patient, many cases will yield only a small set of clinically relevant events. Along with that, a long list of sequencing variants will also appear which will have uncertain significance. Here come the roles of bioinformaticians, who can develop algorithms for interpretation of NGS data that can automate the clinical relevance of the alterations, which will enable more rapid clinical interpretation of genomic sequencing data of cancers. If such approaches get matured, the laboratory will be able to come up with better tumor-specific markers and susceptibility sequences from NGS data, enabling probabilistic approaches for ranking genomic alterations related to different tumors by clinical relevance.
|Database||Host Institute||Based on||Reference|
|Personalized Cancer Therapy (PCT)||MD Anderson||Gene|||
|My Cancer Genome||Vanderbilt||Disease|||
|DGIdb||Washington University||Drug/gene interaction|||
Table 2.Databases for interpretation of somatic mutations in cancer
|Drug||Cancer Type||DNA mutation|
|Cetuximab||Colon cancer||KRAS codon 12, 13|
|Vemurafenib, Dabrafenib||Melanoma||BRAF V600E|
|Imatinib, Nilotinib, Dasatinib, Bosutinib||Chronic myelogenous leukemia||BCR-ABL1 fusion|
|Ponatinib||Chronic myelogenous leukemia||BCR-ABL1 fusion|
|T315I resistance mutation|
|Erlotinib, Afatinib||Lung adenocarcinoma||EGFR|
|Exon 19 deletions|
|Olaparib||Ovarian cancer||BRCA1 and BRCA2 mutations|
|Crizotinib||Lung cancer||ALK gene fusions|
Table 3.FDA-approved drugs for Cancers correlated with DNA mutation.
Furthermore, there are many databases that can be evaluated and accessed to test the clinical significance of mutations. The first level approach of analysis can be the variation, observed in any clinical tumor sample has been seen reported before in published papers. Understandably, the driver mutations are likely to recur in multiple tumor types and patients. The common databases are listed in Table 2.
Among the databases listed in Table 2 Catalog of Somatic Mutations in Man (COSMIC) [14,15] and TCGA (available for data exploration at multiple sites) [16,17] are more common. Information about cancer therapies and prognostic information about development of cancers can be found at a number of databases. MD Anderson’s Cancer Institute host a database for Personalized Cancer Therapy [18,19]. Broad Institute’s develop their own database TARGET [15,16]. Vanderbilt holds database, named ‘My Cancer Genome’ [20,21]. Other databases include TARGET [22,23], IntOGen [24,25], DGIdb [26,27] and CIViC [28,29]. Each database has specialties in their organizations of clinically relevant research derived information links to relevant primary literatures. In nutshell, if someone is having a sequence data (by NGS), derived from some tumor cells of patients, he is not in a clueless position today. There are enough literature support and databases packed with clinically relevant information, which can help to annotate any sequence data derived from cancer genome. There can be novel variations, but that will also add up in the future literature and update is ever continuing process.
Finally, NGS technologies, that require both germ-line and somatic testing (e.g. whole-genome and whole-exome sequencing), the American College of Medical Genetics has released the guidelines after clearly mentioning which variants should be reported to patients always, regardless of whether these are relevant to the presenting illness  or not. This detailed reporting of list of genomic variations has developed the need of through genetic knowledge of the consultant oncologists, as finally they need to interpret the diagnosis report. Since most of reported genes are also involved non-cancer-related syndromes, there is an increasing demand for oncologists, who are prepared to receive the results that turn up unexpected inherited genetic issues .
There are different ways that NGS can help the clinicians for better combating with cancers. It starts from diagnosis to choice of proper therapeutics to each specific patient. Tumor subtypes, which were defined by morphologic criteria only just a few years ago, they are now defined by genetic mutations, either exclusively or inclusively. For example, a study by Honeyman et al. fibrolamellar hepatocellular carcinoma patients showed recurrent inframe fusion between DNAJB1 and PRKACA .
Next, NGS can also help in choosing the appropriate ‘targeted therapy’ for any particular patient or type of cancer, as an increasing number of therapies are available today. So, choice of therapy should be based on data available for therapeutic outcome correlated with DNA level genomic mutations, based on DNA sequencing results, as listed in Table 3. Patients, who do not get the facility for proper detection of mutations, can never be given targeted drug. For them, clinicians have to choose drugs based on assumption, which may benefit the patient, but can actually be also harmed by inappropriately targeted therapies . So, proper diagnosis plays a key role in choice of specific targeted therapies (Table 3).
NGS data can also benefit, when a patient stops responding to a specific targeted therapy, after developing known resistance mutations. In some cases instances, the resistant mutations may be limited to one or a few loci. For example, resistance to EGFR targeted therapies in cancer quite frequently involves a single point mutation and can actually be overcome by merely switching to another different agent . However, glioblastoma can become resistant to EGFR targeted therapies via a more complicated epigenetic regulation .
If a patient has failed to respond in conventional therapy, NGS can be immensely helpful to identify and enroll them into the appropriate clinical trial. The US National Cancer Institute has recognized the potential of the NGS adequately followed by targeted therapeutic approach by setting up Molecular Analysis for Therapy Choice (MATCH) Program . In this program, biopsies from tumors from 3000 patients will undergo NGS to identify individuals whose tumors have genetic abnormalities that may respond to some selected targeted drugs. Then, as many as 1000 patients will be assigned to one of the phase II trials. The patients will be assigned based on the genetic abnormality or mutations that are thought to be driving their cancer, not on their cancer types . In nutshell, NGS allows the clinicians to see more complete overview of tumor dynamics and decide accurate personalized therapeutics with lesser side effects, as compare to conventional therapy.
- Walker SP, Palma-Dias R, Wood EM, Shekleton P, Giles ML (2013) Cytomegalovirus in pregnancy: To screen or not to screen. BMC Pregnancy Childbirth 13: 96.
- Coll O, Benoist G, Ville Y, Weisman LE, Botet F, et al. (2009) Guidelines on CMV congenital infection. J Perinat Med 37: 433-445.
- Yinon Y, Farine D, Yudin MH (2010) Screening, diagnosis, and management of cytomegalovirus infection in pregnancy. ObstetGynecolSurv 65: 736-743.
- Seror J, Bordes P, Luton D (2013) Routine screening for CMV during pregnancy: Practices assessment in Ile-de-France. GynecolObstetFertil 41: 578-582.
- Gyselaers W, Jonckheer P, Ahmadzai N, Ansari MT, Carville S, et al. (2015) What are the recommended clinical assessment and screening tests during pregnancy? Appendix. Good Clinical Practice (GCP) Brussels: Belgian Health Care Knowledge Centre (KCE). KCE Reports p: 248.
- Hughes R, Aitken E, Anderson J, Barry C, Benton M, et al. (2016) Antenatal care for uncomplicated pregnancies. NICE guidelines, clinical guidelines CG62.
- Akkerman D, Cleland L, Croft G, Eskuchen K, Heim C, et al. (2012) Routine prenatal care. Bloomington (MN): Institute for Clinical Systems Improvement (ICSI) p: 115.
- Yinon Y, Farine D, Yudin MH, Gagnon R, Hudon L, et al. (2010) Cytomegalovirus infection in pregnancy. Journal of Obstetrics and Gynaecology Canada 32: 348-354.
- Palasanthiran P, Starr M, Jones C, Giles M (2014) Management of perinatal infections. Australasian Society for Infectious Diseases-Endorsed by RANZCOG 5-10.
- Carrera JM, Mallafré J, Serra B (2007) Recommendations and guidelines for perinatal medicine. World association of perinatal medicine - Matres Mundi International pp: 120-133.
- Pfeifer S, Goldberg J, McClure DR, Lobo R, Thomas M, et al. (2013) Recommendations for gamete and embryo donation: A committee opinion. The Practice Committee of the American Society for Reproductive Medicine and the Practice Committee of the Society for Assisted Reproductive Technology 99: 47-621.
- Adler SP (2011) Screening for cytomegalovirus during pregnancy. Infect Dis ObstetGynecol,pp: 1-9.
- Soper DE (2013) Congenital cytomegalovirus infection: An obstetrician's point of view. Clin Infect Dis 57: 171-173.