Thoughts on Optogenetics & Science Journalism, Part Two
As invariably seems to happen when I’m about to travel, I get involved in an online discussion (see: John Horgan, Scicurious, Paul Raeburn & my Opto thoughts Pt. 1) and then lose track. Belatedly, my follow-up thoughts:
In the disagreement between John Horgan and Dave Dobbs over the value of Helen Mayberg’s work identifying a discrete neurological target for deep-brain stimulation in people with severe depression, and which conceivably could be targeted for optogenetically-triggered therapies, I think Dobbs is right. There is some promise there: Mayberg’s results in humans are early & small-scale but intriguing, especially given the paucity of drug- or behavioral/cognition-based treatments for severe depression. The mouse models of severe depression in which optogenetic treatments are developed are, I would expect, far more useful than models of less-extreme (and more complicated) forms of depression (1). And I’m not overly concerned about Helen Mayberg’s conflicts of interest, namely her various neuromodulation patents. They’re very much worth keeping in mind, and going forward one would obviously want more rigorous results from non-interested parties, but they’re not disqualifying (2).
Anyways — reasonable people could certainly disagree about the degree of promise present in optogenetics for severe depression, but it seems to me at least somewhat promising. And this debate was a good reminder for me not to fall automatically back on, “We don’t understand the neurobiology of _____,” which I too often do. Until a few days ago, I wasn’t aware of Mayberg’s work or the advances in this research.
On Horgan’s larger sentiments, though, about optogenetics receiving more journalistic attention than it merits, and researcher hype being spread uncritically, I still strongly agree. As optogenetics defender Scicurious points out, and on which she’s in agreement with Horgan, optogenetics is a basic research technique (3). It’s low on the list of things to talk about if you want to talk about reducing the human toll of most major neurological illnesses — and I think it’s meaningful that most opto coverage comes from sci/tech people, not from health people.
Take something like the recent model-of-OCD work that sparked all this (4): If a research group had produced similar mouse results using cognitive therapy or a pill, I don’t think Scicurious or Ed Yong would have spent a minute on it (5). But … it’s optogenetics! Light + brain + genes = sexy!
Anyways — I’m rambling a bit here, and maybe turning prematurely into a curmudgeon. I guess I’ve got pretty burned out on med-tech-health the last few years, after experiences watching institutions approach health through a very determined, hard-tech-first lens that excludes or minimizes knowledge gained through other areas of research. And it bothers me because it skews the way people perceive health and research … and if this ultimately translates to funding priorities or health decisions, it can be counterproductive (6).
Look at how the Framingham Heart Study, which has been *fantastically fucking useful* and benefited millions of people, just had its budget cut by 40 percent — $4 million, an amount that seriously threatens its research capacity — by the National Heart, Lung & Blood Institute. The research group that produced one of the latest optogenetics sci-tech cycle stories is getting about $2 million from the NHLBI (7). Does covering the latter contribute to the former’s decline? Maybe that’s too simplistic, but in some way I feel like it does. And if my goal as a journalist is fundamentally to help people, then … I’m with Horgan. At least for now, I just can’t get very excited about optogenetics.
(1) Recommended reading: Daniel Engber’s superb “The Mouse Trap,” on the (usually) very limited and (often) overstated relevance of mouse models for human disease. Pretty much any responsible science or health journalist understands that treatment advances in mice and other animal models are, as a rule of thumb, relevant to researchers but not to the general public. To me, they’re worth covering only inasmuch as they bear on some larger question about a disease or condition and/or are intellectually compelling — “What a lovely mechanism!”
The various reasons for this, nicely summarized as “mice aren’t furry people,” include the temptation/necessity/inevitability of creating very simple models of very complex conditions. E.g., it’s easy to program a mouse to be grossly obese, but its metabolic processes won’t necessarily resemble those of a mouse that simply eats too much, much less a couch-potato human. Treatments that improve its health won’t affect a fat normal mouse the same way, much less a human couch potato.
(2) Financial conflicts of interests get all the attention, but I think reputational/attentional incentives (which can of course dovetail with financial) are just as bad. Once you’ve gone from giving your spiel to a few colleagues to a few million people, and you’ve become known as a salesperson for a particular idea, are you really going to question it rigorously? This holds for thinkers of all sorts, not just researchers. That said, I’m a little unnerved by the possibility that deep-brain stimulation for severe depression could segue into attempts at optogenetic “normal” depression therapies, as seems to be hinted at in Mayberg collaborator Andres Lozano’s loose talk (in a TED talk, natch) of severe depression as “sadness.”
(3) And as Dave Dobbs pointed out to me, a whole lot more elegant than the lesion studies used to test neuroanatomical function.
(4) The accompanying press release was titled, “Optogenetics is proving to be highly promising in the treatment of obsessive-compulsive disorders” — a patently ridiculous statement, and not the sort of thing that can be blamed on on the PR writer rather than the researchers themselves, who wrote in their Science paper’s abstract, “These findings raise promising potential for the design of targeted therapy for disorders involving excessive repetitive behavior.”
(5) A flip side to that argument is, “Well, maybe the optogenetics work deserved the coverage, just as drug- or cognitive intervention-based treatments would deserve coverage, too.” But the results have warning flags all over them: the genetic mechanism highlighted in mice, and ostensibly operative in humans, involves SAPAP3 — only attempts to find links between SAPAP3 and obsessive/compulsive behaviors in humans have turned up very limited correlations (see here, here, here.) It’s probably involved to some extent in some cases of OCD, but there’s a whole lot more going on, and any model in which SAPAP3 is central to OCD is probably not a good model for OCD.
(6) I suspect a lot of working researchers will agree with me: If you want to advance professionally, don’t concentrate on curing disease. Concentrate on developing a method that someone else might conceivably use to cure disease, and score the Nature and Science publications (and media coverage) that the NIH and research institutions value so much.
(7) Basic research is hugely important, but it’s also a bit of a sacred cow.