Times Have Changed, But SOPHiA Hasn’t (NASDAQ:SOPH)

Investment Thesis

When Swiss entrepreneur Dr. Jurgi Camblong established SOPHiA GENETICS (NASDAQ:SOPH) in the early years of the 2010s, his goal was to make the world’s disparate genomic databases more accessible to healthcare providers, drug developers, and academic institutions. Many of these databases are housed in public libraries with varying and often incompatible structures.

At the time, the gene diagnostic testing industry wasn’t very advanced — it was largely limited to research and widely absent from standard-of-practice clinical guidelines. This shouldn’t come as a surprise given that, at the time, sequencing a genome costs nearly $30,000, limiting insurance coverage, and distorting the economic valuation of these tests, a critical aspect in adopting new healthcare technologies.

The problem for SOPH is times have changed, and it seems that the gene testing industry has developed in a way that makes the company’s mission less viable than it once was. The asset and value proposal of Sophia DDM, manifested in the genetic biobank data platform, is not as valuable as it once was; every gene diagnostic company now has a biobank, given the explosion in gene testing volumes, as lower costs created more access. These companies are also offering updated, well-organized, clinically-validated biobanks developed in CLIA-certified labs, directly competing with SOPH. Even biopharma companies who, unlike gene testing businesses, don’t have built-in synergies to build a genome biobank are venturing into this field. For example, Natera (NTRA), Invitae (NVTA), and Myriad (MYGN) collect and store genomic data of their customers to build their bioinformatics business; Eli Lilly (LLY) found a way to build its own database by paying for comprehensive genomic profiling of all cancer patients (Thyroid, Lung and other indications).

When SOPH was founded, the correlations between gene makeup and cancer risk were patentable, and running big-data analysis algorithms on genomic databanks could yield substantial financial rewards. Uncovering the impact of a point or structural mutation on health was also patentable. Today, neither of these is, at least in the US, the largest gene testing market, after the US Supreme Court rulings in Mayo Medical Laboratories v. Prometheus Laboratories (2012) and Association for Molecular Pathology v. Myriad Genetics (2013), respectively.

Finally, the decentralized approach to genomic data analysis is inconsistent with current trends, manifested in increasing regulatory oversight over Laboratory Developed Tests “LDT.” For example, obtaining a CLIA-waived authorization from the CMS (the de facto marketing authorization for the industry) requires a CLIA certificate, granting access for lab inspections and demonstrating test reproducibility given a standardized and streamlined manufacturing process. All these requirements pose practical limits on using the Sophia DDM platform for drug development and clinical use, as discussed in more detail below.

Our hold rating mirrors SOPH’s obscure market position and an unclear path to profitability, weighed against its large addressable market (if the business model were realized) and relatively low valuation.

Market Position and Business Model

When the Coronavirus hit the world in March 2020, it became a priority for governments around the world to prevent the spread of the virus by tracing and isolating those infected. Behind the scenes, a global community of scientists came together to 1) devise a testing solution for diagnosis and 2) develop a vaccine for prevention, utilizing genetic sequencing as a critical component of their research. These scientists rushed to sequence as many coronavirus genomes as possible before performing a genetic alignment against known virus samples (some from previous pandemics) in order to identify the similarities and differences between COVID-19 and other viruses of similar family/phylogeny. These samples are stored in genetic biobanks containing thousands of known virus genetic sequences, structures, and functional impact. One way to access these biobanks, record and analyze virus genomic data is through the Sophia DDM platform, which is what the University of Sassari did through its partnership with SOPH, giving us a practical example of what the company does and how it monetizes its platform.

This article doesn’t dismiss the value proposition of SOPH but instead questions its commercial prospects. As mentioned above, biobanks and bioinformatics are critical components of any genetics-based testing and drug development endeavor, which explains why SOPH has amassed a small but impressive list of customers in that space in recent years.

However, SOPH’s target market behavior and trends suggest the two are deviating in different directions. For example, gene diagnostic testing companies already have their own genomic biobanks and are, in fact, driving the commercialization of bioinformatics through advances in AI and machine learning techniques, directly competing with SOPH. Second, it is clear that these once-potential clients are advancing their products without the need for the Sophia DDM database, which despite being on the market since 2011, occupies a very small market share.

Beyond gene testing companies, SOPH’s business model also suffers from other drawbacks impacting its clinical clients (healthcare providers). First, its solution is for Research Only Use and Not for Clinical use. Now, I’m not naive to the fact that many providers use ROU devices in patient care. During SOPH’s Investor Day, a hospital representative in Brazil talked about how it utilizes Sophia DDM software in patient care.

Nonetheless, I also believe that any use of ROU devices is also limited to a confirmatory tool or assistant procedure rather than as a primary diagnostic machine.

Americans are a very litigious society. Patients are quick to drag healthcare providers to court if they have even the slightest suspicion that something was done incorrectly, motivated by the money that a lawsuit can generate, as Dr. Leiyu Shi, associate professor at John Hopkins, explains in his book, “Delivering Health Care in America.” This is a significant risk for using ROU devices in the treatment course.

This brings us back to the obscurity of SOPH’s business model. The company, which defines itself as a software company, is now venturing to get a test approved for the market. It would be interesting to see if this works out. Sample preparation is the first and one of the most critical steps of genetic testing. Given that Sophia DDM is a data analysis tool, it would be interesting to see how the FDA or the CMS would classify a testing device whose only purpose is to aid the analysis process. To me, this is like building a database and reporting system to sell on the front end without making any effort to develop the backend analytic capability that would enable the system to generate useful results.

Summary

SOPH needs to communicate better with stakeholders about their core business and the core product they have developed so far. Management’s enthusiasm over Sophia DDM conceals serious questions about its market position and commercialization plan, at least in the magnitude pitched to investors.

Management states that its primary target market is US healthcare providers but failed to reconcile this proposition with the fact that its product is an ROU. Its pitch to gene testing companies is also flawed, given of the dozen gene testing companies that I follow, each one of them has a biobank of data collected from their patients, and many are considering leveraging this data into a new bioinformatics segment for biopharma companies.