It's worth noting that (many) blood tests don't need FDA approval. This makes Theranos' approach unusual: they are getting FDA approval, but not publishing the (easier and cheaper to do) tests.

I don't know enough to judge this one in any sensible way. I thought this Forbes article[1] was good, and the author said:

How much is real and how much is hype? I’m not going to provide a final answer here. I still have no idea how Theranos’ technology works. But I’m more confident than I was before after spending a day talking to the company’s partners.

Here are a few reasons for that confidence: The Food and Drug Administration has just posted a detailed explanation of its decision to clear Theranos’ one approved test, and, to my eyes, it does give some validation of their technology

[1] http://www.forbes.com/sites/matthewherper/2015/07/15/giving-...

According to this, they are being forced to seek FDA approval as well as go under review from other regulatory bodies:

http://www.wsj.com/articles/hot-startup-theranos-dials-back-...

I say forced because you made it sound like they had gone out of their way to seek approval when actually they were not.

That's not quite true. They don't need expensive clinical trials like a drug developer would, but there is a regulatory process for getting FDA approval for diagnostic tests that lets companies self-regulate, as long as an FDA approved "lab manager" signs off on the internal validation process run by the company. They would only have to fall back to regular approval processes if there is no "golden standard" diagnostic test which to compare their test with.

I'm a bit dubious of using their partners as a source for the technologies effectiveness though. From the insurer's point of view they are getting extremely cheap blood tests from a company using the same equipment and methods as any other lab. Thats a great thing for the customer and the insurer but why would either care whether shifting to the proprietary technology gets approved?

That test is for a herpes simplex virus? That is positive/negative result, rather than a quantitative assay.

I'm not entirely convinced that what Theranos is promising is even that great.

A couple weeks ago I got a blood test. I walked into the test center, which was staffed by a single person. I waited in line (there was one person in front of me), and told the person my name. She walked me over to a chair, drew enough blood for a whole bunch of tests, and that was it. I was in and out in 12 minutes (I timed it), and it was almost completely painless.

This wasn't Theranos. It was "Lab Corporation", an utterly non-unicornish testing center, and the local branch was staffed by a competent phlebotomist. It turns out that, once you stick a tiny IV needle in, you can draw a few vials of blood very quickly and painlessly.

In contrast, I once had a test at Theranos. The tech poked my finger with the magic lance (which stung more than the IV needle), put the little magic capillary collection widget over the drop of blood on my finger, and squeezed REALLY FUCKING HARD, repeatedly. My finger tingled weirdly for a few hours, because the REALLY FUCKING HARD squeeze hurt and probably broke a few capillaries. And the process took considerably longer than the blood draw at Lab Corporation. See, a little needle with a vacutainer at the other end draws quite a few mL of blood very easily, but a little finger prick struggles for more than a drop or two.

(Next time you give blood and get the finger-prick hematocrit test first, look at the tiny amount of blood that they get. Then imagine the rather large multiple of that amount that you'd need to fill the little Theranos collection widget. Ouch!)

I suppose if I were afraid of needles, then maybe the squeeze-the-finger-really-hard approach would have been an acceptable idea.

From experience, a blood sample which has been milked out of a finger is quite different to a venous blood sample.

(I am an anaestheologist and often perform or ask nurses to perform post-op point of care pinprick blood counts to look for anaemia in patients who have had big surgeries. We do 2 readings to check for error, and I often find that samples which have been squeezed out have a haemoglobin value 10-20% different to those which are drawn from a vein due to expression of tissue fluid.)

It seems that this isn't an issue for Theranos's FDA-approved HSV-1 assay. See Table 16 and Figure 1: http://www.accessdata.fda.gov/cdrh_docs/reviews/K143236.pdf

However, the skill level of whoever performed the fingerpricks for that data may be significantly higher than the average technician at Theranos's wellness centers. And of course we have no idea about the hundreds of other tests. There seems to be some other correlation data on Theranos's site (https://www.theranos.com/our-lab - scroll down to 'Representative Clinical Correlations'), but there's no captions or text to provide context.

Theranos definitely has an uphill battle against the ubiquity and cost efficiency of offsite testing labs but their technology is significantly different because it is an onsite test with a tiny amount of blood.

The infrastructure to support the kind of experience you're used to at Lab Corporation or Quest Diagnostics does not exist in much of the world. With the ubiquity of electricity and wireless broadband, a charity like Doctors Without Borders or Red Cross can drop a Theranos test device in a developing country and only have to worry about delivery of consumables to the location instead of shipping samples out, which requires a reliable postal infrastructure instead of a once a month delivery by airplane, boat, train, truck, or rickshaw for that matter.

I'm excited by Theranos' technology because the FDA has strict limits on how much blood you can draw from patients during a clinical trial. As our therapeutics get more and more complicated, we'll need a lot more data but the FDA will not raise this limit much, if at all, in the future. This is a known issue that you have to design around when running clinical trials, especially if you're developing something like a cancer cocktail where you might be looking for half a dozen metabolites and intermediary molecules in extremely low concentrations that each need a vial of blood. This isn't Theranos' target market (intermediaries/metabolites can be much harder to design tests for) but their tech's ability to test onsite with a tiny amount of blood might allow them to move into this market.

You're not thinking hard enough. Imagine you could do blood tests weekly from home cheaply? It could change all of medicine.

Uh, 25x revenue? Are you conflating money raised with revenue? Because Theranos has not made anywhere close to 400m over the entire life of their company...

I apologize, I didn't look up any estimates/figures for Theranos and I think I mistook the funding number for revenue. I also haven't gone to Walgreens and taken a theranos test, although I do suspect that if Theranos is actually at Walgreens their ratio of valuation to revenue does not end with a divide by zero error.

Regardless, there are so many biotech IPOs based on literally zero revenue that this isn't even worth mentioning. Off the top of my head, these are the public biotech companies that I know of that went public with no revenue to speak of: Genentech (yes, THAT Genentech, biggest acquisition in histoty at over $50 billion by Roche), Celgene, and in the last year alone, Axovant Sciences, CytomX, Strongbridge, MyoCardia, RegenX, and Global Blood Therapeutics (I'm almost certain that "in the last year" in actually "in the last few months")

Fantastic comment. Totally agree. Biotech, pharma, you're a zero revenue company until that massively expensive FDA approval comes through. Investors are aware of this, and it is an all or nothing play.