This post is about Seracare’s new direction for Precision Medicine (in particular Precision Oncology), and the launch of a new product the Seraseq Solid Tumor Mutation Mix-I (AF20). But I’ll start first with a story from my days at Illumina.
It was ten years ago this week that I last launched a product. It was the Illumina Infinium Human-1 Genotyping BeadChip, and it took months of hard work, many core-team meetings, plenty of long days and many hallway discussions about the finest details of the product. Long days and many meetings go together: how can you get the things you commit to in a meeting ever done if you go to many meetings?
And in case anyone is wondering, yes that is the job of a Product Manager – you get to work hard and work closely with a team from across different areas of the company (research, manufacturing, quality, support, and engineering) and you get to experience many ups-and-downs, and learn things about yourself and others through this experience that you carry with you. These may be hard days, but they are good ones.
The product was a game-changer, in that it gave Affymetrix quite a competitor from that first 108K genotype product (one sample per BeadChip, scanned on the humble BeadStation). Alas the technical details become fuzzy (was that a 3 micron bead?) and it was such an interesting time in my career going from Product Management into field sales shortly after, selling the new product I just helped develop. There was real satisfaction in seeing customers at the NIH excited about ramping up for their first Genome Wide Association Studies (GWAS). Checking the GWAS Catalog, as of February 20, 2015 there were no less than 2,111 publications and 15,396 associated SNPs. Remarkable.
Throwback Thursday: A classic from the 2005 launch of Illumina Infinium Arrays! Happy 10th anniversary, Infinium. pic.twitter.com/ehmcf6sDqo
— Illumina (@illumina) July 2, 2015
And in my new role at SeraCare, a new product launches today, to help enable Precision Medicine in the field of oncology. In particular, the new product assists laboratories in their own NGS-based laboratory-developed tests (LDTs) with reference materials. It was only weeks of hard work, not that many meetings (the development had largely been done by the time I came on-board), and a few hallway discussions. And from what I can gather, SeraCare has been somewhat accustomed to selling directly as an OEM supplier to major diagnostic manufacturers for molecular diagnostics of infectious diseases such as HIV and HCV reference materials. Going ‘direct to the customer’ wasn’t something that they are geared toward, nor doing much in terms of digital marketing. New management and direction has changed that, and I’m now a part of this new Precision Medicine group.
With any new product area there are a host of new complications, a new competitive landscape, new people to work with and new unknowns, but the product has been finished and the appropriate decisions made. (In my humble opinion, the right ones too, but time will certainly tell.) Some of the hardest decisions a marketing professional has to make is what to call the product, and what to set the price at. Next on that list is producing the appropriate ‘sales collateral’, which in marketing parlance mean a mix of product PDFs (sometimes called ‘sales sheets’), Powerpoint presentations (one for internal training of the commercial folks, another for prospective customers, and a third for the general company), and outbound messaging (including a press release, webpage copy, email messaging, trade show messaging, social media in there too). On top of this is setting up training and process for technical support.
And so the Seraseq™ Solid Tumor Mutation Mix-I (AF20) launches. It’s a mixture of 26 synthetic targets of ‘clinically actionable’ mutations frequently found in solid tumors, compatible with both the Ion AmpliSeq™ Cancer Hotspot Panel v2 and the Illumina TruSeq™ Cancer Panel (and also compatible with other enrichment platforms, such as Agilent’s SureSelect™ or NimbleGen’s EZCap™). At an allele frequency of 20%, the 26 target sequences are quantitated with digital PCR, and you can easily track run-to-run process variation as well as look at different types of mutation variation detection (i.e. indels or SNVs as part of homopolymer stretches). A unique 6-base insert called an “Internal Quality Marker” within 25 bases of the variant provides an easy way to view the synthetic target reads compared to the wild-type GM24385 background, and may also provide a unique way to look at run contamination.
By the way, this was developed in conjunction with the US National Cancer Institute, which whom SeraCare entered into a Cooperative Research And Development Agreement (aka “CRADA”). You can read more about that here.
Currently existing clinical pathology laboratories that have implemented their own hotspot panels will create their own blends of cancer cell line DNA, and use those for process reference materials. However there is time and effort involved with batch-to-batch variation, in addition to the concerns about stability over time. For the Seraseq Solid Tumor Mutation Mix, it is stable for 2 years at -20C, and stable up to 10 freeze-thaw cycles.
Variation from run-to-run at a set allele-frequency was data I personally had not seen before. (And this is someone who sold the Illumina Genome Analyzer before there was a ‘GAII’ or ‘GAIIx’.) The same assay, the same instrument, the same sample, but different results over time, this is one target (BRAF) of the 26 targets in the Mutation Mix over 28 different runs.
In case you aren’t familiar with Levey-Jennings plots, they look at the mean (blue) with an upper and lower levels at typically 2 standard deviations (SD) over time. And as you can readily see from the chart, with the same sample at a measured allele frequency of 24%, it varies from 21% to 28%, about a 4% swing either way. And if the assay changes (different primers, different assay, different instrument) that same BRAF (COSMIC ID 476) certainly may have an average allele frequency (AF) of 18% instead of 24%, and vary from 14% to 22%. All from the exact same input target sample, due to many subtle differences in the massively parallel sequencing process.
One concern with purified genomic DNA with synthetic targets mixed into it arises: is it a true process control? In the context of a clinical laboratory, shouldn’t the sample mimic FFPE (Formalin Fixed, Paraffin Embedded) tissue as closely as possible?
This is a fair objection, one to which I can say we’re actively working on. But that is for Major Launch Number Two, and we’re only on Major Launch Number One for now. Certainly lots of discussion (and meetings!) around what the market (and customers) really want, and value.
Lastly a word about SeraCare Life Sciences: they recognize the future of precision medicine will need high-quality source of reference materials for routine NGS-based applications in diagnostics that are already changing the course of healthcare. We are in a major transition in taking all the findings from large oncology genomics studies (such as the excellent work of The Cancer Genome Atlas and the International Cancer Genome Consortium) and applying them, in addition to several large clinical trials underway (such as the recently-announced NCI-MATCH and ASCO’s TAPUR). This work is ongoing, and a readily visible outcome of the application of personalized medicine. And the need for precision oncology reference materials will only grow.
For those who have read thus far, visit the new website! And if you are doing work in clinical oncology, consider giving it a try! Stay tuned here as well; there are lots of things going on in the world of genomics I’d like to write about still.