GigaAssay technology spun out of the University of Nevada Las Vegas connects mutations to health
Over the course of the past three months with school-aged children underfoot, there have been many opportunities to talk about what life was like when I was growing up and how that contrasts with life today. Yet the questions that kids have remind me that these were the same questions I had while growing up – what the future would look like in terms of what I would be doing many years hence, who I could interact with, wondering what life would be like. Of course those musings would lead to other terrifying thoughts, such as what kind of work and career I would have, who I would marry, and what my kids would be like. These thoughts could be too much to bear for a 10 year-old, or a 16 year-old for that matter.
While lying on the grass looking up at the clouds with the boys a few weeks ago, we spent several minutes looking at clouds and picking out shapes. It reminded me of those long summer days growing up in Santa Monica, with little to do over the summer but ride my bike to different places and keep myself out of trouble, and sometimes just lying on the grass looking at the clouds and wondering. Kids today still wonder – what kind of work will I do, where will I live, what will life be like and so many other unknowns.
Careers are funny things
Everyone makes things up as we go along, we think we have a master plan yet so many circumstances dictate otherwise, and what may seem like a good idea at one moment in life many years later we reconsider. That idea years ago still is a good idea but in hindsight well there are a lot of other things that could have been taken more seriously as a risk or as a problem.
We make choices based upon our best understanding, and our best thinking through the possibilities both good and not-so-good. And of course change is not easy.
I announce today that I took a new role as Chief Commercial Officer of Heligenics Inc., a startup company based in Las Vegas Nevada. (I joke about it as someone going gambling in Las Vegas.) And I have a few thoughts about startups and Heligenics to share with you.
A few thoughts about startups
There are three key criteria to evaluate when looking at any new role and company, regardless of size. Yet these three criteria become especially important when in the early months and years of a company’s commercial effort.
The three criteria are the people, the technology, and the problem in the market that that technology aims to solve.
The people are first and foremost. When you think about it, everything we do and everything we have is from and through people. We apply for a job, that is a person who you are applying to (the hiring manager). We buy groceries, that is many people who have provided those vegetables (from growing them to transporting them to storing them and then to put them in the grocery aisle at the market). And the interconnectedness of people has been especially highlighted during this pandemic we are still working our way through.
While looking at website traffic (and Google Analytics has a neat alluvial plot illustrating visually how visitors access your website and what pages they view), over the years it is clear to me how popular the ‘Management Team’ link is. It is one of the first things you want to know about a company, next to what they offer is who is behind this new company you never heard of before.
The people are most important – they will set the culture, the norm, the standards and habits and practices that define a workplace. The people are most important – they will accelerate technology development, they will prioritize market- and customer-responsiveness, they will set the standards of quality of the final product. The people are most important – if their technology isn’t good enough they will make it good enough, if an exogenous event occurs like a worldwide pandemic they will do what needs to be done, if they have to pivot their business plan for any one of a myriad of reasons, they will. The people are most important – everything the company is or will become is because of people.
Heligenics currently has only a handful of employees, and is looking for a good Ph.D.-level molecular biologist by the way. You can check out the JD here. (Oops, that’s company-speak for ‘Job Description’, for what that’s worth.)
Finding product and market fit
The technology Heligenics has developed is called the GigaAssay, and looks at the effect of single point mutations on the function of a given protein or the transcriptional activity of a promoter in a human cell-line context. By mutagenizing every base in the region of interest (called saturating mutagenesis), a cell-based assay is read out by flow cytometry and the relative activity to wild-type protein activity (or transcriptional expression activity) can be quantitatively measured.
We are taking what is normally a laborious, mutation-by-mutation approach and scaling it up by several orders of magnitude. It can take months to analyze a set of a few dozens of mutations in a particular gene of interest; we can analyze 100’s of thousands of mutations in that timeframe.
Any assay that has a fluorescent reporter can take advantage of this technology; it can be for transcription (expression level), cell division, cell death, most deleterious amino acid substitution, most conserved substitution and many others.
Three main application areas for this technology are for drug discovery, for failed drug rescue, and for clinical cancer mutation interpretation.
A big problem to solve in drug discovery
A well-known problem in drug discovery is that the majority of medications developed are only effective for a fraction of the population, due to genetic variation between individuals. A often-used example is with the common blood-thinner called Warfarin, where genetic tests have been developed to look at particular genotypes in the cytochrome P450 gene complex that metabolizes compounds in the liver. Variation in different regions of the complex (with names like CYP2C9 and VKORC1) will determine what dosage to use, which in the case of Warfarin a mis-dosage can affect the health of the individual.
Yet thanks to the inexpensive cost of Warfarin as a drug, and the economics of genetic testing, a test for the appropriate Warfarin dosage is not commonly performed. Rather, physicians will simply dose on the basis of empirical feedback from the patient. In other words, basic trial-and-error.
Natural variation is likely one of the causes of vastly different responses to SARS-CoV-2 infection, however a large international GWAS (genome-wide association study) has yet to conclusively identify variants in the population of affected individuals they have data from that determine COVID-19 disease severity. Work is ongoing with a project called The COVID-19 Host Genetics Initiative, a worthwhile effort; another effort to keep track of is what 23andMe is doing with their dataset, recently reporting susceptibility by ABO bloodtype, where Type O blood are 9 to 18% less susceptible to SARS-CoV-2 infection.
These results give important clues about the mechanism of how individuals vary in terms of susceptibility to SARS-CoV-2, which could lead to new therapeutics or protective (prophylactic) agents.
With a given drug and drug target, that drug target will have SNPs (single nucleotide polymorphisms) both in its promoter region (affecting that drug target’s protein expression) as well as in its coding region (affecting that drug target’s native activity). Is there a way to know what variants will have a particular effect on the drug target’s activity? Before Heligenics, it would take months and years to analyze a few dozen mutations in the promoter and/or coding regions.
Now a drug company can measure directly the effect of all potential variants across the entire promoter and coding regions for their protein target of interest.
The following figure was reproduced from the Heligenics website’s drug development product page , where the level of protein activity is measured relative to wild-type, of literally thousands of different mutated proteins. The Gene Mutation / function Library data (GML for short) for a gene is illustrated below.
What can a drug developer do with this information? One would be to turn their ‘one-size-fits-all’ compound in development into a personalized one, based upon the genotype of the individual who is to receive it. Thus in a clinical trial where the safety profile or efficacy of the drug would be adversely affected by a few ‘bad’ genotypes (that is, mutations those individuals carry), the clinical trial could be designed to specifically exclude individuals with specific genetic variants. By excluding these individuals with the mutations that cause the drug to be unsafe or cause bad side effects, as well as mutations that cause the drug to be ineffective, means the chances for a successful clinical trial have measurably improved.
The current percentage of new drug approvals having a companion diagnostic component is currently 45%, and with overall drug approval rates hovering in the 15% range (from Phase I through Phase III approval).
Another problem to solve with failed clinical trials
With an overall ~85% failure rate and each drug candidate representing years and many millions of dollars of development effort, it is estimated there is over a trillion dollars of failed drug candidates in storage among pharmaceutical and biopharmaceutical companies. Given the variation between individuals resulting in poor side-effects (affecting what is called that drug’s safety profile), or ineffectiveness, could these compounds be brought back with a genetic companion diagnostic?
By looking at the function of all potential mutations in the target protein in the context of drug activity, Heligenics technology could bring these failed candidates into a redesigned clinical trial with a simple companion diagnostic, looking for germline SNP variants to exclude from the trial. By turning a ‘one-size-fits-all’ therapeutic agent into a targeted, genetically-driven, precision medicine, the immense R&D sunk costs could be recovered.
A third problem area in cancer diagnostics
In cancer diagnostics companion diagnostic approval is a clear trend, with the majority of oncology drug approvals having companion diagnostics with them, including the first ‘tissue agnostic’ therapies approved. That is, only the molecular change of a NTRK fusion gene regardless of tissue type will qualify for treatment with larotrectinib (VITRAKVI).
And these are part of large gene panels assayed via NGS, and as it is sequencing there are a number of variants that appear that have unknown meaning. Called a VUS (a Variant of Unknown Significance), only a limited number of mutations are approved as a companion diagnostic for the indication. The others are simply unknown, and progress has been slow to reduce the number of VUS results returned by test providers.
For example the first therapy targeting the oncogene PIK3CA for breast cancer (specifically hormone receptor positive, HER2-negative and PIK3CA mutated cancers) was approved about a year ago called alpelisib (PIQRAY), developed by Novartis. The QIAGEN companion diagnostic, approved at the same time, is a real-time PCR assay that tests for 11 mutations in the PIK3CA gene; however this only represents 40% of the observed mutations in breast cancer in PIK3CA.
Looking it up in Jackson Laboratory’s Clinical Knowledge Base, there are 277 entries of mutations in the PIK3CA gene, of which a full 154 (55%) have an unknown impact.
While there are many common mutations with known function, do you want to be that individual patient with a rare PIK3CA mutation in metastatic breast cancer without clear guidance to the treating oncologist on the meaning of that particular mutation?
A favor to ask
If you’ve read this far, I first of all want to thank you for the support over the years!
If you are or know a pharmaceutical development scientist or executive who may find Heligenics interesting, could you make an introduction me via this post or directly via email? My email address is dyuzuki (@) heligenics (dot) com.