In the aftermath of the Theranos downfall, dozens of pundits have performed virtual autopsies of where CEO Elizabeth Holmes went wrong, generally highlighting a lack of full disclosure about the technology its labs were using. In the beginning of July, Interleukin Genetics announced that it was laying off more than half of its staff amid a decision to halt sales of its genetic test for periodontal disease risk. Just a few weeks later, Courtagen Life Sciences announced it would no longer offer its next-generation sequencing (NGS)-based tests for neurological disorders due to difficulties with reimbursement and the regulatory environment.
These three companies and others like them share a common thread: an (apparent) failure to plan appropriately for their IVDs from the beginning.
When Investment Rhymes with Canada
Canada has a proud history of achievement in the areas of science and technology, and the field of biomanufacturing and life sciences is no exception.
Developing and commercializing an in vitro diagnostic (IVD) is a complex enterprise involving numerous tasks accomplished by a variety of specialized and knowledgeable individuals. Generally, the process can be broken down into product concept development and validation, diagnostic development/engineering, regulatory requirements, market access planning, and commercialization planning and operations. Each of these overarching processes is similarly subdivided into tasks and deliverables. Importantly, while these are often thought of as discrete tasks, in reality they are significantly intertwined and overlap in time (or should be).
I think that the failures of Theranos, Interleukin Genetics, and Courtagen Life Sciences highlight mistakes that are made commonly by IVD companies and can serve as examples for companies focused on remaining relevant in the extremely competitive IVD industry.
Theranos
Elizabeth Holmes founded Theranos in 2003. Without an inside source, dissecting exactly what went awry for Theranos relies on the limited information released by the company at various conferences, recently unsealed court documents, and the analysis of others, including reporter John Carreyrou from the Wall Street Journal, who has published a series of exposés about the company. According to company filings, Theranos had secured nearly $7 million by the end of 2004 and $400 million just a decade later with partnerships inked with the chains Safeway and Walgreens, the Cleveland Clinic, and Pennsylvania-based insurer Capital BlueCross. With a valuation estimated at $9 billion, for a time, Theranos seemed on track to revolutionize the lab industry, a storied outlook that came crashing down, beginning in 2015 with a Journal of the American Medical Association viewpoint by Stanford’s John Ioannidis. Several articles have reported the secrecy surrounding Theranos’ technology, unreliable lab results, use of other manufacturers’ equipment for patient tests, a lack of peer-reviewed published data, and eventual involvement of the FDA.
One thing appears certain after the fallout: Theranos was unable to bring to fruition a device that could accurately measure the hundreds (or even dozens) of analytes it claimed in a fingertip’s worth of blood.
Several articles have reported the secrecy surrounding Theranos’ technology, unreliable lab results, use of other manufacturers’ equipment for patient tests, a lack of peer-reviewed published data, and eventual involvement of the FDA. One thing appears certain after the fallout: Theranos was unable to bring to fruition a device that could accurately measure the hundreds (or even dozens) of analytes it claimed in a fingertip’s worth of blood.
Using a development framework can ensure that a product doesn’t reach launch before it’s been thoroughly validated and verified to meet user needs and any regulatory requirements. From an outsider’s perspective, this seems to be at the heart of Theranos’ problems. Two basic engineering models are typically used for device development: the waterfall and concurrent models. Both have their benefits and drawbacks, but the concurrent model works well with good manufacturing practices and design controls. Communication between the various development teams means that problems can be addressed early. Technical experts are involved from the beginning and can work with designers to ensure the final product is both technically possible and matches the design concept, minimizing potential product redesigns and their associated costs.
Adding the stage-gate method to the development framework further ensures that a device won’t reach final production stages or launch before being thoroughly validated. Specific deliverables must be completed prior to moving on to the next stage in the development process. This, in particular, may have been able to prevent some of the challenges Theranos faced, by ensuring that not only was the technology feasible, but that prototypes were rigorously assessed and any functional irregularities fixed prior to final manufacturing and public release.
Interleukin Genetics
In a 2009 MIT Technology Review article, Interleukin Genetics was highlighted as proof that genetic testing was becoming mainstream. The company, which went public in 2003, sold its tests for heart disease and periodontal disease risks, vitamin metabolism, and weight management directly to consumers and through direct-sales company Amway as part of a 2009 deal. In 2015, Interleukin landed a partnership with New York University School of Medicine to develop a test to predict patients at risk for progression of osteoarthritis.
However, the company faced a number of challenges. Daniel Macarthur, a well-known geneticist currently at the Broad Institute of Harvard and MIT offered up a stinging rebuke of the company’s offerings. A 2015 FDA letter about its tests and subsequent meetings with the Agency resulted in the company transitioning its tests away from a direct-to-consumer model to one where a physician or dentist was required to order a test.
Despite its relationship with Amway and long-standing direct-to-consumer sales model and partnerships with academic medical centers, the company struggled to find profitability and failed to tap into the large genetic testing market. At the beginning of July, a confluence of events, including a looming debt payment, led to the company’s decision to halt performing certain genetic tests and lay off the majority of its small staff as it pursued “strategic alternatives”. Three weeks later, the company announced it would liquidate assets and seek interested buyers in its intellectual property and CLIA-certified laboratory.
With a significant percentage of the population affected by periodontal disease, heart disease, and obesity, Interleukin would seem poised to reap the benefits of a market increasingly interested in genetic testing. Despite this potential, the company appears to have been unable to articulate to the public the value in their genetic tests, a problem compounded by negative publicity from doctors and scientists in the field. Furthermore, an educational brief published by the American Dental Association states “At this time, neither genotyping nor microbial testing are recommended as a routine dental procedure to identify the presence, absence, or severity of the disease [periodontal disease]. Clinical measurements … remain the single best method for assessing disease”. For Interleukin Genetics, this position by a professional society and the requirement that the test be ordered through a physician or dentist, made increased sales unlikely in the foreseeable future.
One of the most important components of IVD development and commercialization occurs before product design is underway: identifying the unmet need. This involves not only pinpointing what the current problem is, but talking with healthcare providers to learn where the problems are in their clinical workflow and whether they see the need for your product. More extensive preliminary research could have clarified the ADA’s stance on using genetics for risk prediction and suggested additional research studies that would be needed for such tests to be adopted into clinical practice.
Courtagen Life Sciences
Courtagen Life Sciences launched in 2010 after acquiring the assets of Decision Biomarkers, eventually renaming as Courtagen. The privately-owned company provided genetic tests for neurological disorders including autism, intellectual disability, and epilepsy. Courtagen acquired Medicinal Genomics in 2012, a company that investigates the clinical use of cannabis and cannabis genomics. In early 2017, Courtagen announced the expansion of its test portfolio, adding 17 new tests for neurological and related conditions. However, several months later in late July, the company announced that it was halting its diagnostic testing business and would take over the Medicinal Genomics Corporation name. In its press release, Courtagen cited the regulatory and reimbursement landscapes for genetic testing which “make the goal of achieving long-term profitability in the diagnostic space extremely difficult”.
Courtagen correctly calls out some of the challenges facing diagnostics companies, particularly those which offer genetic and genomic tests. Payers are notoriously unlikely to reimburse for these tests and broadly label them ‘experimental’ or ‘unproven.’ If patients are routinely denied coverage for these tests, providers will be hesitant to prescribe them, except in specific situations, such as for oncology, where the clinical utility of the test is well known. This sets up a negative feedback cycle and as Courtagen noted, it can be difficult for a company to find financial success. But these challenges are not insurmountable and with careful planning during device development, can be often overcome.
Thorough understanding of the different regulatory pathways, from the costs to the length of time from submission to a decision by the FDA and the benefits and drawbacks to developing a product as an LDT, are essential and should be considered from the beginning of the design phase. Similarly, knowing how to position a device and what studies are needed to obtain a positive coverage determination from payers and a fair reimbursement level commensurate with the value the device yields should be among the top issues addressed during development. Although the regulatory landscape is evolving, even devices manufactured as LDTs may be impacted by a future tiered regulatory path based on risk.
Realistic expectations are essential when forecasting both the adoption of the device and the length of time needed for payers to reimburse for the test to ensure adequate working capital. Genomic Health’s Oncotype DX is a case in point. The company was founded in 2000 and a proof of concept was completed to analyze gene expression profiles from FFPE tumor tissue in 2001. Genomic Health focused the indication for its first diagnostic test to breast cancer, presenting two studies at scientific conferences in 2003. The following year (2004), the lab received CLIA certification. But commercial breakthrough for Oncotype DX followed a longer timeline. Payers granted positive coverage decisions beginning in 2005, with public payer coverage in 2006 by a local Medicare contractor. Notably, the company published numerous clinical studies of Oncotype DX which supported the eventual recommendations for the test by the National Comprehensive Cancer Network and the American Society of Clinical Oncology. Published clinical studies like those performed by Genomic Health are essential for any IVD company looking to obtain private and public payer coverage.
Without an inside view of Theranos, Interleukin Genetics, and Courtagen Life Sciences, analyzing what each could have done differently to prevent their eventual demise is left largely to conjecture based on previously published reports. In each case, critical steps in the overall IVD development process appear to have been overlooked or not fully fleshed out. Strict adherence to a manufacturing framework can prevent a device from launching before it’s validated, while research performed during concept design can verify the IVD will fit into a clinical workflow and address an unmet need. All companies should understand how their product would be impacted by any changes in the regulatory environment, regardless of the pathway chosen for development and be prepared for a lengthy delay between product launch, uptake by users, and payer coverage decisions. Though success in this competitive industry is never assured, steps can be taken during IVD design and development to stack the odds in your favor.
Photo: MilosJokic, Getty Images
Harry is the author of two related books: Commercializing Novel IVD’s; A Comprehensive Manual for Success and MoneyBall Medicine: Thriving in the New Data-Driven Healthcare Market.
