“Amateurs focus on identifying their weakness and improving them. Professionals focus on their strengths and on finding people who are strong where they are weak.” Shane Parrish, Farnham Street Blog
I am happy to announce today that I have joined Pillar Biosciences in Natick MA. (For those not familiar with Boston geography, Natick is about 30 miles due west of the Boston Logan Airport off of I-90, and is adjacent to Framingham.) I will be serving in a global market development role, which naturally in a small organization will mean several different functions.
Functions of a Market Development Manager
One main function is business development, which is a key role for an emerging business. This is a role that requires sales skills but at the executive level; a VP of R&D may find value in a company’s assay technology, and want to incorporate it into their future product development pipeline. Then legal counsel on both sides needs to navigate confidentiality agreements with input from business development, R&D people will get involved with doing a pilot study or providing other kinds of data. Lastly licensing or other commercial terms need negotiation and agreement between all parties.
Another core function will be pre-launch marketing. This means honing the value proposition and positioning by working with our collaborators, customers and our internal people to drive key messaging, and get it right. For those familiar with the “4 P’s of Marketing” know what I’m talking about here, with the key 5th P being positioning.
A third core function will be content marketing. Once the primary and secondary waterfall messaging is established, the messaging needs to go out, and the web is a primary opportunity to get the message out through social media channels, whether a corporate blog, LinkedIn, and Twitter. In addition I’ll try a few new things, such as making short online technical presentations, try my hand at ‘explainer video’ animations, and help setup a digital marketing infrastructure. (For those unfamiliar with digital marketing, think of useful emails you receive from vendors that are not just ads but actually something you can use in your work.)
In case you were wondering, I will continue to post here at yuzuki.org on other topics unrelated to Pillar’s focus on cancer clinical genomic test development.
The Founders of Pillar Biosciences
Pillar Biosciences was founded in 2014 by two scientists involved in the development of the IQuum Laboratory-In-A-Tube (or LIAT). The Marlboro MA-based startup company’s goal was to produce a molecular diagnostic device that could detect nucleic acid at the point of care (POC). The IQuum company was able to get FDA approval for an influenza test in 2011 and was acquired by the Roche Corporation in 2014 for $275M plus $175M in milestones.
In 2015 at the American Association for Clinical Chemistry (AACC) held in Philadelphia, Roche launched their cobas LIAT system. A small device (“smaller than a shoebox”), a sample goes in, reagents are sequentially added, a real-time PCR amplification is performed by the same instrument, and a quantitation of the analyte obtained. More information about the cobas LIAT is here.
After IQuum, the co-founders of Pillar Biosciences Dr. Gang Song (LinkedIn profile here) and Dr. Zhaohui Wang (LinkedIn Profile here) joined Genzyme Genetics in 2008 in MA, and continued at Laboratory Corporation of America after they acquired Genzyme Genetics in 2010.
Pillar Biosciences’ Goal
Pillar Biosciences’ goal is to make the overall clinical genomics workflow in the molecular pathology laboratory easily implemented. The Pillar Bioscience website reads: “Our Mission: Enable Affordable Precision Medicine”.
There are several obstacles in the current market. One main one is financial: enrichment assays (i.e. targeted sequencing library preparation) are expensive. If you speak with a few molecular pathology laboratories you quickly discover the financial pressures they are under, with scant insurance reimbursement for these tests as clinical utility for these personalized medicine tests remain to be convincingly demonstrated.
Another obstacle involves laboratory personnel: a high amount of expertise and training is required to run these assays. Medical technologists, who are highly trained professionals, may simply be unfamiliar with the complexities of a genomic assay such as a Illumina TruSight Tumor 15 assay running on a MiSeq, compared to an automated clinical chemistry analyzer.
A last obstacle is also related to laboratory personnel: a high amount of expertise and training is required to analyze and report on the data generated. The existing bioinformatics market is highly fragmented, and the clinical genomics laboratory wants something simple, easily implemented, without the cadre of highly trained specialists one might find at a large academic cancer center for example.
Pillar Biosciences Technology
Pillar Biosciences has developed a multiplexed, single-tube overlapping PCR amplification technology. This method is called SLIMamp and stands for Stem-Loop Inhibition Mediated amplification.
As an aside, for those that may not be familiar with PCR you need a single opposing primer pair to amplify a single target region. In a multiplex reaction, if you want to enrich overlapping target regions you need to separate them into different physical vials, because with a total of four primer pairs in a single region you can come up with a total of four potential amplicons.
Two of the amplicons are ones you want, and two are the ones you do not want. The figure below shows ones you want in green, the ones you don’t want in blue. Due to amplification preference and efficiency, the smallest unwanted product can easily overpower and outcompete for limited amplification reaction resources.
By using a novel stem-loop complementary between the two unmatched primers for the smallest PCR amplicon, the Pillar Biosciences technology enables overlapping multiplex PCR amplification. For further details, the method was recently published here in PLOS One. From what I’ve been told (by one ‘outsider’ who has studied these methods for years) many methods require specialized nucleotides or other special modifications which will also suppress the undesired PCR product. Why have they not been adopted in the targeted sequencing PCR enrichment market? Frankly because they are expensive to produce.
Another key feature of the patented Pillar PCR technology is the lack of a need for a ligation step. This is important for two reasons: the first is that ligases are notoriously inefficient. For readers here who are not laboratory people, talk to anyone who has done molecular biology and they’ll have at least one nightmare ligase story to share. I have several, but will spare readers here who have stories of their own.
With a relatively inefficient ligase as part of the workflow, you will need more starting material to begin with, and you will need time for the reaction to take place. You may say the starting material is PCR amplified product to begin with, but if the efficiency is say 70% there are 30% of the unique starting molecules that never get into your library for sequencing that represents missing data. I have been told that the efficiency number is much lower, on the order of 50%.
According to a joint poster at last year’s Association for Molecular Pathology conference in Charlotte VA, Pillar collaborated with Dartmouth-Hitchcock Cancer Center’s Dr. Greg Tsongalis and Dr. Francine de Abreu and presented a poster called “Evaluation of the Pillar NGS SLIMamp Lung and Colon Hot Spots Panel”. If you’d like to take a look at the abstract, here is a PDF.
It was a side-by-side comparison between Thermo Fisher Scientific’s Ion Torrent Ampliseq Cancer Hotspot Panel v2 and a Pillar Bioscience 20-gene panel. The total elapsed time comparison (from purified DNA to ready-to-sequence library) was 17 hours for Ampliseq versus 7 hours for Pillar.
The process is faster, and a simpler workflow. A side-benefit of a simpler workflow is then the ability to automate that much easier. Given the fragmented state of liquid handling automation, easy-to-automate is important to achieve the stated goal of getting NGS technology adopted quickly in the clinical lab for cancer patient care.
Pillar Biosciences Accuracy
With a BRCA panel covering 17.7 kb of both genes, their PLOS One paper shows >99% mapping to hg19, >99% mapping to the targeted exons, and >98% within 0.2x of the mean coverage from 34 samples. There were no false positives and no false negatives reported. The supplemental data online has several spreadsheets to dig into if you want the particulars.
In the 2016 AMP poster, Dartmouth analyzed 15 FFPE samples across the two platforms, and the AmpliSeq Cancer Hotspot v2 had 3 variant calls it missed of 30 variants (two single base-pair deletions and a single point mutation) but Pillar detected all 30. In addition, Dartmouth compared 50ng to 5ng FFPE input material using the same samples on the Pillar assay, and determined 100% concordance between the input amounts.
I will be very busy in the coming weeks getting up to speed as Pillar Biosciences is getting ready for a formal product launch. However, currently enrichment kits for the Illumina platform (MiSeq etc) are available on a trial basis. If you are a clinical genomics lab running Illumina libraries send me a note (my email is yuzukid at pillar-biosciences.com). If you are an assay developer looking for a flexible and efficient technology feel free to reach out to me. And watch this space for updates!