It was back in June of 2016 when Forbes had revised its valuation of Theranos from $4.5 billion to $0 following a series of investigations and allegations that the company’s blood tests were inaccurate. And yet, earlier this month Theranos made headlines again when The The Wall Street Journal published more information on violations of policies and procedures, which it topped off with commentary that the company is now “on life support“. Specifically, The Wall Street Journal had revealed that Theranos employees improperly operated blood testing machines and that the company did not ensure that all patients who may have received potentially inaccurate blood test results were notified. Although Theranos had shifted its focus and is now developing the miniLab– a tabletop laboratory which it will sell to health care providers– the original vision of running hundreds of tests using just a finger prick remains appealing. The question then becomes ‘what are the technology gaps that prevented Theranos’s original promise from becoming a reality?’ To better understand this gap we outline the state of innovation of blood diagnostics today:
Theranos’s claim was that it could perform 30 tests with one drop of blood, so how large of a volume did Theranos consider a “drop” — their Nanotainer was shown with about 300 µl in it normally. Many finger-sticks take blood volumes on the order of 30 µl while sufficient blood to form a drop is an order of magnitude higher, around 300 µl, which matches the company’s claim. However, different test methods will require different minimum blood volumes, with the minimum volumes around 20 µl, and the company’s first tests were lower volume tests for blood type, iron, cholesterol, and herpes (their only approved test). Without multiplexing their tests, Theranos could only get half of their original claim of 30 tests. This is where the company’s story falls apart when it compares to other innovation currently occurring in blood diagnostics. This innovation is in both sample collection in addition to detection, where many molecular diagnostic tests require larger volumes of blood (200 µl minimum) for a sufficient amount of target material to be in the sample. Other small volume samples monitoring for glucose or vitamins can be done with lower volumes (about 20 µl).
For blood collection, Tasso (client registration required), for example, is developing a painless blood collection system that utilizes microfluidics to obtain blood samples. The company’s offering competes with large incumbents such as BD, Medtronic, Greiner, Sekisui, and Terumo, in addition to other startup players such as Seventh Sense Biosystems and Velano Vascular. All of these companies produce consumables for blood collection; however, these commercially available offerings do little to help the pre-analytical steps apart from the addition of some additives. If future collection devices facilitate pre-analytical processes without significant complexity or costs, this will spur further innovation in point of care tests.
On the diagnostics side, there is a general demand for innovating sensor systems that can detect more analytes; with higher precision and accuracy, faster, and at lower concentration levels. While many large players including Abbott Laboratories, BD, bioMérieux, Danaher, Qiagen, Roche, and Siemens are active in this space, smaller players like Curetis (client registration required) have also developed platforms. All of these devices will need to either use a cytometer and/or multiplex their assays since even detecting a simple disease requires testing for multiple genotypes. Also, these tests will require higher volumes of blood, which further calls into question Theranos’s initial claims.
Even comparing Theranos to other noninvasive disease management or diagnostics developers that analyze other bodily fluids, like tears, sweat, or saliva, the company claims did not match the current innovation. It’s important to realize that at this time, the accuracy and reliability of these approaches are often still limited; one commonly cited challenge is that analyte concentration in bodily fluids is much lower than in blood. In addition, there’s often a non-trivial time lag issue, so measurements tend to be inconsistent. For optical analyses, motion artifacts and low signal to noise ratio can often be challenging.
Readers should learn from the example of Theranos and be wary of any company’s claims that seem at odds with the current trends within an industry, especially if the company does not provide the data to back up these claims. While Theranos is a glaring example, there are other companies whose claims may not stand up to a thorough vetting.
By: Noa Ghersin and Thomas Dawidczyk