The world of science and technology journalism sometimes seem to operate under a rule of: "Hype first, and think later."
The MIT Technology Review magazine recently had an article that was a bad example of reporting about a risky new treatment. The article had the title "Brain-cell transplants are the newest experimental epilepsy treatment," and sounds as if it was written by someone trying to raise the stock price of a company called Neurona Therapeutics. We hear about some new technique tried on only five people, in which an attempt is made to treat epilepsy by inserting brain cells into one of the most sensitive parts of the brain.
The approach mentioned is one that takes the exact opposite path that physicians have been using for decades in treating epilepsy. Epilepsy is a mysterious brain disease that causes seizures, which are like electrical storms in the brain. The exact cause of epilepsy is unknown. Nowadays most epilepsy is successfully treated by prescribing medications. A minority of patients have what is called drug-resistant epilepsy. The standard treatment for very bad drug-resistant epilepsy is surgical removal of part of the brain.
In the most severe cases of epilepsy, doctors can perform a hemispherectomy that may involve removal of half of the brain. But most of the times a much less drastic operation is performed. Since most epilepsy is what is called temporal lobe epilepsy (in which seizures come from the part of the brain called the temporal lobe), drug-resistant epilepsy can usually be treated by an operation called a temporal lobectomy. Such an operation involves removing much of one of the temporal lobes of the brain. Several thousands of such operations occur every year in the US. Most of the the time the operation works very well, without serious side effects.
Since doctors have a good surgical treatment for drug-resistant epilepsy, why would the MIT Technology Review be promoting some scantily-tested experimental treatment taking a "Johnny-come-lately" approach that is the exact opposite of the standard medical technique, a treatment involving adding brain cells rather than removing them? The author article tells us this about mainstream surgery for epilepsy:
"While this kind of surgery can stop seizures permanently, it comes with the risk of 'major cognitive consequences.' People can lose memories, or even their vision."
The article gives us no references or quotes backing up this statement, which seems to be very misleading in regard to the claim about losing memories. Removal of one of the brain's temporal lobes to treat epilepsy was done thousands of times a year in the decades before the year 2003, and also in the past two decades. Reports of severe memory problems following removal of the temporal lobe are actually rare. A 2003 scientific paper tells us how rare how such cases are:
"Davies and Weeks (1993) did report one case of postoperative amnesia in a series of 58 cases of unilateral temporal lobectomy, whereas Walczak et al. (1990) found one case of marked deterioration in memory from a preoperative normal state in their series of 100 patients who underwent such surgery. Rausch and Langfitt (1992) estimated that, on the basis of their series, 'the prevalence of patients at risk for postoperative amnesia who otherwise met criteria for surgery fell between one and four out of 218’ (p. 508), and Jones‐Gotman et al. (1993) noted that 'the base‐rate of post‐resection amnesia, were all patients operated on without prior screening with the amobarbital procedure, may be less than 1%, (p. 447).' "
The authors of the 2003 paper state, "We were able to locate nine definite cases of amnesia following unilateral temporal lobe surgery in the English‐language literature." This is not very worrying, given that (1) many thousands of operations of this type were done before the year 2003; (2) doctors or scientists often loosely use the term "amnesia" for any of a large variety of memory performance problems, and typically use such a term for cases that are something other than a severe loss of learned or episodic memories; (3) there are many reasons why a person may have memory problems, and a few people having memory problem after a particular operation does not show the operation caused such a memory problem.
In fact, in Table 3 of the paper we are given the details of the impairments of these nine cases of claimed amnesia; and none of them sound like a case of loss of knowledge or episodic memories (with the possible exception of case 5, which is not well-described). We hear about what seem like rather minor memory performance shortfalls. A 2009 study carefully testing "before and after" memory results for 82 patients who had surgery for temporal lobe epilepsy tells us this: "The main finding of this study is that, at variance from the picture emerging from short-term follow-up studies, longer-term memory outcome after TLE [temporal lobe epilepsy] surgery seems to be good, as after 2 years memory performance was equal to or better than baseline [before surgery] in most patients." A review of 911 surgical operations for epilepsy (looking for bad effects) mentions no case of amnesia or memory loss or memory deterioration.
There is no robust evidence that surgical treatment for temporal lobe epilepsy causes loss of memories. Trying to justify some radical surgical treatment that is the opposite of the approach doctors have long taken for treating epilepsy (an experimental technique involving adding brain cells rather than removing them), our MIT article seems to have missed the mark. The facts about how taking out large chunks of the brain are even more shocking than those discussed above. It seems that contrary to the dogma that your brain stores your memories, you can remove not just the temporal lobe, but even an entire half of the brain without causing people to lose their memories. See my post here for a justification of that claim.
We get some hype in the MIT article, not justified by any published research results of the study mentioned. We read this:
"The treatment, developed by Neurona Therapeutics, is shaping up as a breakthrough for stem-cell technology. That's the idea of using embryonic human cells, or cells converted to an embryonic-like state, to manufacture young, healthy tissue."
Ironically, to the left of this claim we have to a link to an article entitled, "After 25 years of hype, embryonic stem cells are still waiting for their moment." The MIT article is part of that 25 years of hype. The only evidence it gives is to cite a group of only five people who have had these Neurona Therapeutics cell implants. We hear that for four of these people their seizures have decreased through some brain cell implants that did something involving a chemical called GABA.
But these are self-reports, which are often unreliable. And reporting the frequency of seizures is particularly unreliable, given that when a seizure is finished, a person may be left with no memory of it occurring. (Most seizures are much less severe than the type of dramatic seizures that used to be called "grand mal" seizures but are now called tonic-clonic seizures.) The self-reports are also anecdotal word-of-mouth stuff, rather than published research results.
One of the quotes in the MIT article is a misleading slur about people with epilepsy. The article states, "People with epilepsy have a hard time remembering things, but some of the volunteers can now recall an entire series of pictures." That leaves us with the impression that people with epilepsy have very bad memory, and cannot even recall a series of pictures. That is misleading prose that is defaming 1% of the population.
In scientific studies (often using too-small samples for reliable results), people with epilepsy tend to perform almost as well on memory tests as people without epilepsy. Often the epileptic patients chosen for such studies are those with particularly severe epilepsy, which may create a misimpression about the general population of epileptics. There is reason to suspect that a study randomly selecting 1000 patients prescribed epilepsy medication (and using IQ-matched controls) would show little difference in memory scores. One scientific paper says, "There are, however, reports that suggest that memory may not be affected in PWE [people with epilepsy] when compared with controls if the PWE and controls are matched with respect to intelligence quotients." One of the studies on this topic that used a larger sample size was one doing cognitive tests on 176 children with epilepsy, comparing with 113 control subjects; it was entitled "Educational Underachievement in Children With Epilepsy: A Model to Predict the Effects of Epilepsy on Educational Achievement." The study reported, "The children were assessed with a test battery consisting of tests for educational achievement, cognitive tests and tests for reaction time, and tests for memory and intelligence....For memory, none of the analyses showed statistically significant effects."
Can we imagine what an uproar would be created if some article tried to claim that people of some racial minority have trouble remembering things? But it seems that without any uproar a mainstream article can do defamatory stereotyping of 1% of the population by saying "people with epilepsy have a hard time remembering things." The vast majority of people with epilepsy have good memories and function well in school, work and society.
The link here takes you an FDA page describing the clinical trial being done. We are told the study will be completed in May of 2026. We are told subjects will "take medicines to partially suppress their immune system (aimed to prevent the body from rejecting the cells) for 1 year," which sounds like a significant risk. We are told no results have been posted for this clinical trial. So why is the MIT Technology Review hyping this study, which has not even published any results?
The MIT Technology Review story tells us something that makes no sense under "brains make minds" claims. It tells us that a few people were given "inhibitory interneurons" and that "the job of these neurons is to quell brain activity." But we are also told (without any specifics) that for the handful given such cells there have been "improvements in cognitive tests." If your brain makes your mind, why would quelling brain activity result in improvements in cognitive tests?
A story last year in the MIT Technology Review describes 25 years of hype about stem cells, and tells us, "Yet today, more than two decades later, there are no treatments on the market based on these cells. Not one." It seems we've been grossly misinformed for decades about stem cells. And quite a few other scientific topics. By now an FDA page entitled "Approved Cellular and Gene Therapy Products" lists a small number of treatments, some of which are supposedly stem cell treatments; but they are all cancer treatments or blood treatments.
I can imagine a conversation that might occur between a hypothetical scientist Smith and a hypothetical scientist Jones working at some company that is testing stem cell implants:
Smith: Don't think of an embryo as a "developing human being" but merely as a bank of cells we can rob.
Jones: So what's the plan?
Smith: We will take cells from an embryo, grow them in our lab, and then inject them into a sensitive part of the body, to see what happens.
Jones: But that will require some very sick human "guinea pigs" willing to roll the dice.
Smith: They can always be found.
Jones: But you know, this fooling around with stem cells has never worked very well. They've been playing with stem cells for 25 years, and there is still almost nothing in the way of treatments. What if our sizable investments in this company fail to pay off?
Smith: Don't worry. To keep stock prices high, we can always feed the right type of gossip to certain "pushover" journalists eager for a "science gloriously marches onward" story.
Jones: But you can play that game only so long. One day you have to publish results.
Smith: Don't worry. Large stock investments can be conveniently sold BEFORE the first results are published, when the stock price is still high.
