What happened in 2016?
Oxford Nanopore’s PromethION. Illumina’s panel with Amgen. Thermo Fisher’s panel with Norvartis/Pfizer. IBM Watson replacing doctors. Genapsys’ fundraising. Qiagen’s GeneReader win against Illumina. Biocartis’ taking Europe by storm with Idylla…it was a busy year!
I wrote a post back in January listing the trends and my predictions of things that are happening or going to happen in next generation sequencing in 2016 (and 1 that isn’t). So I think it’s about time I reviewed which of these came true!
Skip to the review:
- There will be a larger emphasis on clinical interpretation of sequencing data
- There will be a continuing trend for simple, affordable and smaller ‘plug and play’ systems (POC)
- Multigene panels will be used clinically
- There will be an emphasis on liquid biopsies and cell free DNA
- There will be further definition of clinical standards and guidelines
- There will be a change in the fundamental chemistry of sequencing
- Whole genome sequencing will be more popular
- There will be a decrease in the price of sequencing a genome to….$50??
There will be a larger emphasis on clinical interpretation of sequencing data
Okay, this was an easy one to predict you say? Sure I probably agree with you, however, I don’t think anyone could have predicted where 2016 went with the idea of ‘big data’ with emphasis on IBM Watson… Doctors in particular have this big data challenge and this is where IBM Watson Health and its AI technology comes in. It is humanly impossible for an oncologist to know all that information.
Big Data – The Big Business of Storing Human Genomes
I interviewed Brandon Kruse in 2016, a serial entrepreneur in the IT industry and also the founder of a company (agtc.io) that stores sequencing data in the life science community on exactly this topic. Brandon spoke about the challenges and solutions to large quantities of genomic data. You can read my interview with him here.
This is about providing that insight to the oncologist for a shared decision-making process to help make sure the best treatment options are considered.
There will be a continuing trend for simple, affordable and smaller ‘plug and play’ systems (POC)
Wow oh wow. Did this trend continue and go on even further! I take my hat off to what Nanopore and Biocartis have achieved this year. I was fortunate enough to get a demonstration of the MinION and Idylla (respective platforms of the manufacturers) at AMP this year and whilst don’t share a single molecule of the same chemistry (I’m sorry for this cheesey joke…click here to tweet if you get it!) they are remarkably similar in their end desires of bringing ‘plug and play’ to the masses. Oxford Nanopore albeit in a super-affordable and flexible way to their loyal fans whilst Biocartis instead focusing on the clinical applications and the art of simplicity.
— OxfordNanoporeEvents (@NanoporeConf) 2 December 2016
What would ONT want regulatory approval for you ask? You don’t need much of an imagination to see these being placed in a GP’s surgery at the point of care… Combine this with the little rumour that the MinION can reliable detect ctDNA in blood down to 5%…watch this space!
Last, but not least forgetting the PromethION and the RNA-Seq capabilities of the MinION: without the need for a cDNA library preparation step this may be the best thing since sliced bread… although, with only ~14 platforms of the PromethION currently installed in 2016, it’s still a far cry from the ~8,000 Illumina platforms and their dominance.
Biocartis Idylla which is perhaps the best plug and play RT-PCR to have come to the market in 2015. Although labs should be cautious in adopting these ‘black box’ platforms for several reasons, furthermore having an open-source flexibility coupled with a simple and efficient user interface can help improve NGS data analysis. Also keep in mind that it takes only 4 to 5 years for a sequencer’s specifications to become obsolete.
Did I mention the MiniSeq was available too?
Multigene panels will be used clinically
As pharmaceutical companies shift toward developing new targeted therapies as opposed to ‘one-size-fits-all’ drugs, there is a clear need to more efficiently stratify cancer patients and tailor their treatments to enable better health outcomes.
The two big pieces of news to hit in 2016 were from the ‘big 2’ of Illumina and Thermo Fisher:
Illumina and Amgen
Thermo Fisher and Norvartis/Pfizer
Thermo Fisher and its Oncomine submission to the FDA with Norvartis and Pfizer – November 2016
A trend which will undoubtedly continue in 2017…
There will be an emphasis on liquid biopsies and cell free DNA
A HUGE trend in 2016, not just for liquid biopsies but also for non-invasive prenatal testing (NIPT). What we currently know about therapies is that, whilst they may work well at the beginning, progression is universal and happens in up to two-thirds of the EGFR-mutant patients that will eventually have a second site mutation EGFR T790M or EGFR C797S. In patients who are on a first-generation TKI who experience clinical progression, it’s worthwhile doing a plasma test specifically to identify the T790M mutation.
Where cfDNA is heading – my interview with Lao Saal, the new CEO of SAGA Diagnostics
I previously interviewed Lao Saal in a blog post I wrote, who’s now coincidently the new CEO of SAGA Diagnostics AB. You can read it here
The utility and versatility of cfDNA assays for disease monitoring, earlier detection than solid tumor biopsies and prenatal applications (eg. Trisomy and aneuploidy) means that this will continue to grow strongly in 2017. Astrazeneca is also very keen on labs getting cfDNA tests set up for both EGFR T790M and EGFR C797S: http://www.nature.com/nature/journal/v534/n7605/abs/nature17960.html
What’s next? How about cell free RNA….??
There will be further definition of clinical standards and guidelines
The reliability of laboratory developed tests (LDTs), which have been largely unregulated, is inconsistent and could put patients at risk. Last year, his office released a report listing 20 LDTs that had produced clinically erroneous and even invalid results.
Definition: Laboratory Developed Tests (LDTs) – Whilst most common laboratory tests are commercial tests, manufactured and marketed to multiple labs, some tests are developed, evaluated, and validated within one particular laboratory.
Attending the Association of Molecular Pathology (AMP) 2016 meeting this year, there was a lot of talk around new standards and guidelines, especially now with our new president…
Regardless of the outcome, there is sure to be more regulations, guidelines and standards in the future to improve the industry.
There will be a change in the fundamental chemistry of sequencing
Okay, I called semi-conductor sequencing and maybe was wrong. On paper at least it has the best specifications: low cost platforms, high reliability (platform due to not needing an optical sensor) high throughput and the backing of Thermo giant…but it just hasn’t happened. Why? One word: homopolymer. Or perhaps better put another way: aaaaaaaaaaaaaaaaaaa(h).
Illumina sequencing technology, sequencing by synthesis (SBS), is the most successful and widely adopted next-generation sequencing (NGS) technology worldwide currently. However, there has recently been a dramatic rise in the use of other technologies such as: nanopores, electronics, microfluidics, real-time technology, and others.
QIAGEN Releases GeneReader for Clinical Sequencing in Cancer
Although not a ‘fundamental’ change in chemistry, or any chemistry whatsoever in fact! The simple application and simple message from Qiagen will appeal to many small clinical labs.
It’s biggest selling point? “A lab does not need a bioinformatician to process the results.” – Wow.
November 9, 2015 | QIAGEN’s GeneReader DNA sequencing system was finally unveiled last week in Austin, Tex., at the annual meeting of the Association for Molecular Pathology. The company had first planned to launch the GeneReader in 2014, but ran into delays during early access testing.
Very simple and very effective. Time will tell.
What will the chemistry change to? Well, semi-conductor sequencing or SMRT may be a good bet. If this new technology can deliver a robust point-of-care or field deployable sequencer then there will be a huge wealth of benefits, including: faster sequencing, portability, simplicity and a decreased cost. Semi-conductors will also be more robust outside of a traditional sequencing lab as the optics are gone. Conversely, Single Molecule Real-Time (SMRT) sequencing from PacBio and Oxford Nanopore Technologies appears to be the biggest trend in 2016, certainly garnering a lot of support on twitter…
An example of how ONT is taking social media by storm:
— Paul Hoskisson (@PaulHoskisson) 7 December 2016
The clinical use is yet to be elucidated and Illumina will undoubtedly be thinking about what their response to this technology…
Whole genome sequencing will be more popular
A comparison of gene-panel, exome and whole genome sequencing
Some people love WGS, some people hate it…and it’s caused a bit of a stir on social media.
— BarbaraJennings (@GeneticsMBBS) January 23, 2016
You can read my previous post on it here.
Main conclusion: The popularity of WGS is relative to its cost. Leading me nicely onto my next prediction…
There will be a decrease in the price of sequencing a genome to….$50??
Okay, this one may have been VERY aspirational. We’re not close, at all.
Genapsys has previously been pegged to be the company to achieve this with their GENIUS (Gene Electronic Nano-Integrated Ultra-Sensitive) however nothing had yet to come out of this start up. To note, they have recently (15th november), received $37M Series B Financing….so we shall see.
Ever since 2014 when the HiSeq X Ten was released, there appear to be somewhat of a stumbling block (perhaps the laws of physics…) to go below this $1,000 mark.
Flatley’s breakdown of projected HiSeq X Ten sequencing costs included the cost of reagents needed to run the machine ($797 per genome), the depreciated cost of the machine itself ($137 per genome), and the costs of paying technicians to run the machines and of preparing samples for sequencing ($55‒65 per genome). But it left out the overhead costs that academic centers must pay, such as the costs of electricity needed to run the machines.
-TruSeq Nano DNA or TruSeq DNA PCR-Free library preparation
-Includes DNA and library QC
-150bp paired-end read length
≥100 Gb raw data
≥75% bases above Q30
-30x coverage for human DNA (as defined by Illumina)
-Delivery of FASTQ data via online portal DNAnexus with clinical-grade security and privacy
Maybe we need a chemistry revolution to break this $1k barrier?