“The plural of anecdote is not evidence.”
It is an adage that is quoted often in medical and drug development circles as a stalwart against crackpots, snake-oil salesmen and patients desperate for a Hail Mary miracle when all of the standard options have failed to alleviate their suffering. It has all the hallmarks of the parental comeback: “And if your friend Billy decided to jump off a bridge, you’d jump off, too?”
As a recidivist biochemist and long-time science and medicine writer, I spend an inordinate amount of time repeating this adage to friends who post the craziest things on Facebook or Twitter. Rarely, on deeper scrutiny, do these medical miracles show signs of scientific thinking or investigative rigor. More often than not, the evidence of success is a series of undocumented testimonials about how much weight someone lost or the renewed joy of playing piano or traveling with a loved one. (Not unlike most advertising by pharmaceutical companies.)
By the same token, on rare occasions, the lack of a clinical trial does not necessarily mean a lack of clinical relevance.
This morning, I watched a news item about a young girl who suffered an extreme form of epilepsy, a form that sent her into seizures every few minutes. From birth, this poor girl was crushed by hundreds of seizures every day, and none of the typical medications given for epilepsy or pain seemed to work for very long, if at all. Her mother was desperate.
For reasons not clearly explained in the news item, the mother decided to try her daughter on oils extracted from medicinal marijuana (infused into coconut oil). It worked. The daughter’s seizures ceased within hours and have continued to be controlled for two years and running.
Is this evidence of a new treatment for her condition? No. It is an anecdote of one patient’s experience.
But for at least this one patient, it is a clinically relevant anecdote.
Back in the 1920s, a compound used in the vulcanization of rubber was also found to induce an intense sensitivity to alcohol consumption in factory workers; an instant hangover. Eventually, the compound was identified as disulfiram, and by the 1960s, it was marketed as an aversion therapy to treat alcoholism.
But the anecdotal echoes didn’t stop there, according to Stephen Marcus, CEO of Cantex Pharmaceuticals, with whom I spoke at AACR 2016 in New Orleans.
“There were scattered reports about anticancer activity over the years,” Marcus recounts, “but about 10 or 15 years ago, the University of Utah discovered that if you combine disulfiram with heavy metals, it chelates the heavy metals and that the chelate has very potent anticancer properties.”
Subsequent studies showed that when combined with copper, disulfiram has very potent activity in glioblastoma (GBM). Having purchased the IP from the University of Utah, Cantex is now working with Washington University in St. Louis and has initiated trials in recurrent GBM.
“The wonderful thing about this is that it will not be a $100,000 drug,” he enthuses. “We’re repurposing an old drug, and it will be more affordable than most things.”
Marcus is the first to admit, however, that despite having some ideas of what the drug does in the cell (e.g., proteasome inhibition and induction of reactive oxygen radicals), they do not have a clear mechanism of action, which he says puts it on par with most chemotherapeutics.
“For many of the chemotherapy drugs, the mechanism of action is unclear,” he explains. “Sometimes people will cite a mechanism of action, but then others will say then why would it work in Cancer A but not Cancer B.”
“So there’s a lot that’s unknown; it’s just empirical,” he continues. “I think empiricism is going to play a big role in future cancer drug development; finding something that works or somebody who has an unusual way of looking at something.”
In some ways, it seems, it is less about making lemonade out of lemons so much as making lemonade out of quartz crystals. Just because we don’t understand why something works in an unexpected application, we shouldn’t automatically dismiss any suggestions that it does.
Ironically, it is who we are, as much of what we know as the modern pharmaceutical industry was built on such serendipitous observations and anecdotal folklore.
So, while it remains true that the plural of anecdotes is not evidence, it can be a good starting point. And because, in so many ways, the outcome is more important than the process in healthcare, we have a duty to at least considering that these stories might be simply the first superficial signs of a mountain of undiscovered evidence.
Randall C Willis can be reached by email at firstname.lastname@example.org