On Sunday night, the New England Patriots topped the Seattle Seahawks in the most-watched (and most tweeted) television event in American history: Super Bowl XLIX. History tells us that biotech stocks are going higher this year as a result.
We looked at the Nasdaq Biotechnology Index’s (NBI) performance going all the way back to 1994, which marked the first full year since the index’s creation. In the resulting 20-year dataset, 80% of the time in which the AFC champions won the Super Bowl, the NBI ended the year higher than where it began in January.
In half of the AFC’s winning years, biotech stocks returned double-digit gains over the course of the year.
We can tell that in the years in which the AFC champion won and the total game score (scores combined) was greater than 50 points, the mean 1-year NBI return was a whopping 45%. The total score this past weekend: 52. Clearly, biotech investors should be celebrating the Patriots’ big win.
If you landed on this article searching for a combination of “how to make money” and “NFL Super Bowl” keep reading: we’re about to BLOW YOUR MIND.
Here it is: the correlation between which NFL conference wins the Super Bowl any given year and the likelihood that biotech stocks rally is about as strong as the correlation between the weather outside and how loudly you howl when you stub your toe in a doorway. That is to say – there is no correlation, whatsoever.
Likewise, small biotech companies often try to correlate a product that Executives’ livelihood [often] depends on to any piece of data that will support ongoing employment and [often] a handsome bonus. In biotech we call this “data mining” or “data dredging” – looking for a positive trend in the results of a clinical trial, often failed studies.
Data mining, post-hoc analysis, subgroup stratification and the sort are all red flags for biotech investors.
At PropThink, our first advice to new and aspiring biotech investors is to be skeptical. Be deeply skeptical. There’s not a management team out there that’s not bullish on their own company, and they’re not going to be the ones to debunk the bull thesis or walk you through why you shouldn’t own their stock.
To illustrate this point, here’s an old but quintessential example of data mining at its finest:
Apricus Biosciences (APRI), formerly known as NexMed, developed NM100060, a topical treatment of onychomycosis, for much of the last decade, to treat nail fungal infection. NexMed signed commercial rights to the drug over to the pharma giant Novartis (NVS) in 2005 for a payment of $4 million upfront and $47 million in contingent milestone payments. At the time, it wasn’t a bad deal for the company.
Then Novartis completed two double-blind and placebo-controlled phase 3 clinical trials of NM100060 in 2008. The studies failed to meet their intended endpoint, and a year later a third study being run in Europe flopped as well. Novartis handed the NM100060 rights back to Apricus in the summer of 2009, indicating that it would not pursue approval in the U.S. or Europe.
Two years later in June of 2011, Apricus announced a “reanalysis” of the failed phase 3 trials, “[…] a combined post-hoc analysis of two randomized, double-blind, vehicle controlled, multicenter, parallel group Phase III studies to assess the efficacy, safety and tolerability of MycoVa [NM100060 had been renamed by this point], demonstrated statistically significant results in mycological cure, resulting in the eradication of nail fungus in favor of active treatment in patients who did not present with comorbid tinea pedis (athlete’s foot), as these patients are considered at higher risk of reinfection.”
That certainly sounds scientific and rigorous. What the company is saying, however, is that MyCova works better in patients with onchomycosis who did not also have athlete’s foot. The original MycoVa studies failed, so Apricus dug up a small subset of patients in which the drug did appear to work. Investors will never know how many other subcategories Apricus cast aside in its mining expedition to find a positive signal. They may as well have said that the drug was likely to work because a NFL player was named MVP that year. Get it? The subgroup benefit was [likely] completely coincidental.
Apricus’ rehashed data (we call it a post-hoc analysis) is not an adequate approach to proving that MycoVa works in onchomycosis without athlete’s foot. Regardless, Apricus promised to meet with regulators and seek an FDA approval.
Fast forward three years to the present day and MycoVa has essentially been buried. The FDA suggested that Apricus run more studies, and the company never moved forward with a regulatory filing in Europe. Though the company talked up a shot at Canadian approval for almost a year, nothing ever panned out. Apricus effectively “deep sixed” MyCova. Meanwhile, here’s the correlation between that and the Company’s share price (yes, over the long-run fundamentals do matter).
Another pro tip: be skeptical of companies that say they’re doing it all, or that they have a cure-all in one drug. At the time of the MycoVa “reanalysis”, Apricus billed itself as a company focused on “Sexual Dysfunction, Oncology, Dermatology, Autoimmune, Pain, Anti-Infectives, Diabetes and Cosmeceuticals among others.” Ebola anyone?
Today Apricus is “advancing innovative medicines in urology and rheumatology.” Apricus’ stock has lost more than 95% of its value (see chart above) since going public almost two decades ago, and racked up an accumulated loss of $268 million since inception.
For a more recent example of egregious data mining, turn to Celsion Corp (CLSN), which is currently running a 550-patient phase 3 trial based on subgroup from a previous failed phase 3 trial.
Celsion’s lead product candidate, ThermoDox, is a heat-activated formulation of the approved oncology drug doxorubicin. Healthcare providers use radiofrequency ablation (RFA) to “activate” the Thermodox product in a specific part of the body. In the HEAT study, completed in early 2013, Thermodox failed to improve Progression Free Survival (PFS) in patients with hepatocellular carcinoma.
Months later, Celsion made the case that a subgroup of patients in the HEAT study who had received more than 45 minutes of radiofrequency ablation actually lived longer and had tumor shrinkage when compared to patients in the control arm of the study.
Celsion subsequently designed another study, called OPTIMA, with this “optimal” process: at least 45 minutes of RFA. The study will enroll 550 patients with HCC, just like the HEAT study, and began enrolling patients in the third quarter of last year.
We won’t know for another couple of years how OPTIMA plays out, but we’re deeply skeptical of this data dredging tactic.
For the NFL-loving, number-crunching investor now armed with this little nugget of information, we’re offering a special discount on our Premium service. Save $100 when you use promo code NFLMyBiotech and learn how we navigated healthcare stocks to return 54% last year, one trade a time. We’ll provide you the information you need to know when to buy, sell, and how to limit risk in your portfolio – and most importantly, how to make money investing in some of the best technologies.
Heck, if you’re not approaching biotechnology stocks with a skeptic’s eye, you may as well plan your investments based on the outcome of the Super Bowl. After all, the data show that it works 80% of the time.
