There seem to be a range of responses to this question so far, but it is now quite clear that NGS will increasingly make inroads into clinical practice, especially because of continually decreasing cost, and the ongoing improvements in knowledge/databases related to cancer genes and related pathways, driver mutations and the software/algorithms to reliably identify them, and actionable targets. One simply has to look at the incredible story of Dr Lukas Wartman here at Washington University. He is a leukemia oncologist diagnosed a few years ago with that very disease. Rapid sequencing found a problem with FLT3 and Sutent, a FLT3 inhibitor, put him into remission. His story was widely publicized, e.g. in the NYT ( http://www.nytimes.com/2012/07/08/health/in-gene-sequencing-treatment-for-leukemia-glimpses-of-the-future.html?pagewanted=all&_r=0 ) As time goes on, you will see more stories like this.

NGS has not become a part of routine oncology practice so far primarily because the equipment is not freely available at a majority of cancer centers at the present.  But it seems to be a promising tool that might change the ways we shall be treating cancer patients in the years to come. Next decade should see NGS to become an important part in prevention and management of several cancers. 

Although NGS is a big hit in recent Oncology studies, it is still not considered as a routine clinical practice due to the high price (US$ ~2000) for just sequencing not including sequencing analysis/interpretation. Therefore, for detecting SNPs and specific gene mutations there are  more economic alternatives other than using NGS. Besides, identifying the mutated genotypes is the first step for diagnosis and whether the mutated genes have corresponding actionable treatments is another issue. Hopefully when the NGS price is getting lower, more people are willing to choose NGS as routine diagnosis.

Thanks all of you for your comments. I was opened this question because NGS is available for my team at a reasonable cost. However the information obtained by NGS is so huge that pragmatically speaking it will save much time and efforts to hear some suggestions from teams which already used NGS in clinical setting.

Before you spend so much money on sequencing, you might just use a microscope and look at the chromosomes in a typical carcinoma cell. The genomes of carcinoma cells are highly unstable, at the chromosome and gene level. The current cancer sequencing mania will lead us nowhere...because it delivers no targets for therapy. You cannot fight chaos by searching for targets, which are constantly changing.

Dear Dr. Stindl,

Unfortunately we cannot see the mutation of interest through microscope. There is enough high-grade evidence that sequencing is indicated for mutations in some genes as BRCA-1&2, p53, mismatch-repair group and others both in patients with various types of cancer and their relatives. The information obtained is proven to be beneficial for both treatment planning and preventive measures. My question is related to NGS use beyond widely accepted indications in regard to offer more personalized treatment/surveillance options. We just see the door of Precision Medicine now, it will take time to open it.

short answer no.

longer answer: it will depend on cost effectiveness in my health system.

if it is not cost effective it will not be provided through the national system and use will be small. at present given the exhorbitant drug cost and the significant test cost I am not hopeful.  eg Canadians have decided they will not provide testing and therapy for alk mutated lung cancer for that reason [and this isn't even ngs];I am sure our numbers will be like theirs.  

Dear Dr Julianov,

"Precision Medicine" only works, if the targets are not constantly changing. Unfortunately this is the case in carcinoma. Nearly every carcinoma cell within a tumor harbors different mutations. This approach is hopeless ;-)

I am currently implementing targeted panel sequencing in a diagnostic setting. Targeted sequencing of current and emerging actionable targets is cost effective and with Roche's EZ capture system the library can be adapted very easily with minimal time involved in validation. We are not the only one to do this: MSKCC, MDA, UHN Toronto just to name a few. The EORTC is currently investigating how to convince health care agencies in Europe to reimburse panel sequencing. Similar activities are starting in Canada. For now you need some cash in your pocket to do this. Targeted seq is not a hopeless attempt to identify current drug able targets and those that will select patients for trials. I don't think that WES will be standard of care in a molecular path centre, but may be helpful in case of lymphomas.

I highlighted the relevant sentences in the attached Nature article. Believe me, sequencing biopsy samples of carcinomas is hopeless...

If we ignore the fact that this paper does not answer the original question, and instead ask is there any data suggesting mutation analysis can be useful a quick medline search will show many papers where treatment is better if a driver mutation is identified and targeted in lung cancer, melanoma et cetera. As one example see M Kris presentation at ASCO 2013 J Clin Oncol 31, 2013 (suppl; abstr 8019).

Is the issue of dramatic genetic heterogeneity and ongoing mutation a problem: absolutely! Are there substantial issues with development of resistance because of this: totally! Is cost effectiveness a major issue: yes!

But is sequencing biopsy samples "hopeless": demonstrably not.

I attached a figure of another recent Nature article. The problem is that depending on the location of the biopsy, you will find completely different "driver mutations".

The value of targeted genetic profiling of diagnostic, prognostic and predictive biomarkers  in a clinical setting will not only be addressed in the Winther trial by the WIN consortium, but also by the molecular profiling efforts of the EORTC (the SPECTA fora). Winther-like protocols for sequencing-guided combination therapies are being developed.  

The canadian NCIC CTG group and the EORTC in Europe are currently working on guidelines for minimal requirements of the sample, reporting and QC/QA in a clinical setting. One thing is clear: You need the appropiate infrastructure (including automation with liquid handlers) to run these type of assays and will most likely be limited to CLIA or equivalently accredited molecular diagnostics aboratories.

As I wrote before, negotiations with health care systems are ungoing for reimbursement of panel sequencing because this strategy is already cost effective compared to consequtive single gene testing when there is a need to analyze more than 2 genes.

WES is a totally different question and don't see this finding its way in routine diagnostics.

@Rohan Chaubal

If you could just look for mutations in a known "cancer gene", it would be a reasonable approach. However, a sequencing study, lead by Bert Vogelstein, found that 1718 genes, representing 9.4% of all known coding genes in the human genome, had at least one non-silent mutation in either a breast or colorectal cancer (out of 11 breast and 11 colorectal tumor samples). On average, the number of non-silent gene mutations per tumor was shown to be between 77 and 101 [1]. Accordingly, the authors concluded that the genomic landscapes of (breast and colorectal) cancers are composed of a handful of commonly mutated gene ‘‘mountains’’ and a much larger number of gene ‘‘hills’’ that are mutated at low frequency.

1. Wood LD, Parsons DW, Jones S, Lin J, Sjoblom T, Leary RJ, et al. The genomic landscapes of human breast and colorectal cancers. Science 2007;318:1108–13.

This is old knowledge, but things got even worse. Now the "driver mutations" are also found in non-coding DNA regions. See attached file...

I have use it, of course, not as for daily practice, not for all patients. I have used sometimes in patients that have failed to the approved therapies according to histopathological findings without positive results, in those cases I re-biopsy (of course we all know about tumoral heterogeneity) and I used drugs according to the affected gens not to pathological report. I have some successful results with my patients. One big issue is the high cost of the test.

here, many doctors use it , but only for research

http://www.triconference.com/mmtc_content.aspx?id=109881&libID=109838

I guess it depends on what you are sequencing. The technology is getting cheaper every day and soon more labs going to use it. Routine is is different unless NGS is used for Targeted sequencing. But exome sequencing yield lots of information that needs additional efforts and longer time that is not valuable for routine work.

FoundationOne has been at the forefront of applying NGS for clinical and research purposes (http://foundationone.com/). The company Foundation Medicine advertises to physicians and healthcare providers, and there are some case studies listed in the Publication section on the website. The sequencing is targeted, and some discussion on the advantages of this approach is available at http://www.genomeweb.com/clinical-genomics/cleveland-clinic-using-foundation-medicine-test-study-impact-sequencing-cancer-c. 

There seem to be a range of responses to this question so far, but it is now quite clear that NGS will increasingly make inroads into clinical practice, especially because of continually decreasing cost, and the ongoing improvements in knowledge/databases related to cancer genes and related pathways, driver mutations and the software/algorithms to reliably identify them, and actionable targets. One simply has to look at the incredible story of Dr Lukas Wartman here at Washington University. He is a leukemia oncologist diagnosed a few years ago with that very disease. Rapid sequencing found a problem with FLT3 and Sutent, a FLT3 inhibitor, put him into remission. His story was widely publicized, e.g. in the NYT ( http://www.nytimes.com/2012/07/08/health/in-gene-sequencing-treatment-for-leukemia-glimpses-of-the-future.html?pagewanted=all&_r=0 ) As time goes on, you will see more stories like this.

Аmazing story! Thanks for sharing Michael.

I think the original question was clear on was whether the NGS techniques are routinely used in the clinical setting. The answer is even clearer: it is not in widespread use. Now I think it is inevitable to happen at the end. It is clear that the NSG technical impact or will impact on the diagnosis, prognosis and possible treatment of many diseases difficult clinical management. One issue that I think has not been discussed, the NGS techniques will also improve the genetic counseling of these patients. The clinical future is of them also

I agree that NGS will eventually impact genetic counseling too, but that will come somewhat later after NGS starts to be used more widely as a routine diagnostic tool. Before that can happen, the cost has to be reduced probably at least one more order of magnitude yet. (That sounds like an enormous hurdle, but it must be considered in light of the history of sequencing technology in which the cost has already plummeted by a remarkable 5 orders of magnitude: about $1B for the first human genome in 2001 to about $10K today). Application of NGS for cancer will certainly precede its use for counseling (indeed, it has started already) because (1) there are, in relative terms, a much smaller number of people and (2) those people already show an urgent need to have their genomes sequenced. Stay tuned for what is in store in this exciting field.

I agree with the statement by Dr. Michael Wendl and bring something more I recommend the article "Whole-genome secuencing in health care" (Carla G van El et al, 2013) which includes the recommendations of the European Society of Human Genetics on this very exciting.

http://www.nature.com/ejhg/journal/v21/n6/full/ejhg201346a.html

CTCA uses genomic screening and testing for use to better match targeted therapies with the patients cancer and increased therapy effectiveness.

We recently started using targeted panels and exome sequencing in the clinical practice mostly for targeted therapies in lung and colon cancer as well as melanoma. It is a rapidly changing field that is destined to grow especially as price continue to drop.  The real benefit will be as we are able to gather more exome or whole genome data on cancer genomes.  We will be better able to determine tumour heterogeneity and evaluate reasons for treatment failure.  Eventually this will be a front end analysis that will be more important than histopathology and radiology.

I want to share this: I had a patient with pancreatic cancer stage IV (adenocarcinoma) since 2 years ago, I have been treating her as guidelines says. Now she is under progression. She has ECOG 0 (without any symptom/she does not look like a patient). I ask for a new biopsia, I order NGS, the results were: mutation on JAK/STAT and KRAS. Now I am going to give her an oral JAK inhibitors (ruxolitinib). I hope she reach remission.