Opko Health (NYSE:OPK) had a big selloff Friday. The selloff and a negative article here at Seeking Alpha made me reexamine my investment reasoning.
The article was by Richard Pearson and his facts seem to be accurate, but his conclusions do not.
Pearson wrote that insider selling, weak buying from the CEO, and new stock issued from pipeline purchases and liquidity financing made a good case to short the stock. After review, these reasons don't make a good short case after all. But more troubling was the allegation that the near term pipeline was deficient, including the prostate cancer screening test. This is what I want to shed some light on below, because, barring another negative article, it seems like the next short term catalyst for the stock.
Quick review of Pearson's case:
Insider Selling addressed well and in a timely fashion by the company (company press release); one sale doesn't make a trend.
CEO Weak Stock Buying: This is too subjective to be part of a short case - is he buying symbolically? Methodically? It seems true that he was buying less, recently, but maybe he put his money into the convertible bond instead of stock - I would. Now it appears he was a buyer on Friday.
Big Overhang: This is a positive! I like that they went out and used the stock as currency to buy pipeline. That's what I want them to do. Well done, OPK! This stock is like a venture capital fund, and management knows more about what looks promising than I do, so I want them to run my VC bio-tech money. 50mm out of 300mm shares is a good deal. It's like they bought a 5 year call for 17% of their market cap. The potential sales figures sound crazy, but we'll see, do an expected value analysis and discount the result, it will certainly be more than the $340mm purchase price.
Prostate Cancer Test: This is what I want to write about here, and is the main point of my article. If they don't have any hope for the prostate cancer test, then I really don't want to own the stock here. Richard Pearson's article made it seem this way, but let's look a little closer.
Pearson cited a Barron's article from September of last year as his source for making the claim that the prostate cancer test wasn't working "as planned". Let's take a look at that article, and take what it says at face value. The article is by Bill Alpert and was written before the two recent acquisitions, and I'm not going to get into a tangent about valuations, just one thought: if you are going to cite old valuation claims, you need to keep them in the context of being calculated with old information.
Bill Alpert's article has a tone that sounds neutral to me (certainly not scathing), he calls OPK a "VC Portfolio", and he cites a May 2010 Journal of Clinical Oncology editorial claiming OPK's prostate cancer test missed "14% of high grade cancers, and was not likely to change prostate cancer mortality". This sounds very negative, doesn't it? Even worse, it prefaces this with a claim that current tests get higher accuracy. This would be very troubling indeed, if you didn't know anything about bio statistics and about prostate cancer in general. But let's simply believe the claims as stated.
Let's do a quick refresher on Sensitivity & Specificity, and Prostate Cancer:
Sensitivity = (number of True Positives)/(number of people tested with the disease). It is the percent of the diseased people that you correctly identify. If the prostate cancer test missed 14%, then the sensitivity = 86%.
Specificity = (number of True Negatives)/(number of people tested without the disease). This is the number that tells us how many false positives we have via the complement. False positives are a big deal in the prostate cancer business because biopsies, the followup step, can be harmful.
Prostate Cancer: Biopsies do a lot of damage because the prostate is so small, and it is hard to perform one well. There is a debate about which does more harm (people dying from undetected prostate cancer, or damage from bad biopsies on false positive patients); it is not clear that any screening decreases mortality. Screening seems to be here to stay in the US, because it is a standard quality measure and because prostate cancer is a big business for urologists. Not all prostate cancers are alike; there is no one size fits all.
With that in mind, lets get the big picture right and review those seemingly negative statements about OPK's prostate cancer test.
The big picture is that if a company can create a test that is more accurate it will be a success. More accurate means increasing sensitivity and specificity (increasing the % of cancers detected and decreasing false positives). Mortality is irrelevant.
The Barron's article gives the OPK sensitivity at 86% for high grade prostate cancer. At a PSA threshold of 4.0, the standard PSA test gives a sensitivity close to 40%. Which is more accurate? OPK's is.
SWOG, a cancer organization, has a nice chart on PSA Test Sensitivity available on its website from a Journal of American Medical Association 2005 issue. Link to SWOG chart on PSA sensitivity. It shows the sensitivity and specificity for PSA levels of 1.1, 2.6, and 4.1. The higher the PSA, the more likely it is that prostate cancer is there too. 4.0 is a traditional threshold for determining if a biopsy is required (to find out for sure). Only at the 1.1 level is there a higher sensitivity than OPK's at 93, but the specificity is an unacceptable 39%. This means that if this were the standard, as the articles seem to mislead us, 61% of people without the disease would be subjected to a biopsy. Why stop there? Why not biopsy all males every year? Obviously this line of thinking is absurd. There has to be a balance of detection threshold and the number of false positives, and that is why the 1.1 threshold is not used, and that is why what the cited Journal editorial said was misleading.
There is no mention of false positives in the Barron's article from last fall. If the rate of false positives in OPKO's new test is higher than the current standard, then it won't be adopted. But I have to believe that if it was, they would have mentioned it. The only information in the article was about sensitivity.
The kicker for OPK's intended prostate cancer test is that the urologist supposedly gets to bill for the test instead of sending it out to a lab.
Conclusion: OPK is like a venture capital fund that invests in biotechs, and should be valued as such. It's all about probabilities of future sales, and the next catalyst is the prostate cancer test. I don't know if the prostate cancer test will be successful, but based on the information portrayed as bad, it looks good. This is a great example of how information can be accurate, but misleading.