Editor’s note on April 17, 2006: We would like to commend Nature for its news report (fee req.) of April 13, 2006, about the claims of the Beijing neurosurgeon Huang Hongyun, which has confirmed the serious doubts raised by our correspondent in this January 2005 article.
Anesthetize the rat. Lay it belly down. Shave a patch along its spine and cut to the bone. Do a laminectomy, that is, take the bone off a short length of the back of the spine, exposing the spinal cord. Suspend a 10-gram rod above the spinal cord, at a height of 12.5 millimeters, or 25, or 50 millimeters. Let it drop.
The result will be a bruise, or more technically, a contusion, of the rat’s spinal cord. The bruise interrupts nerve transmission, paralyzing some muscles and blocking sensation. The location and severity of the damage will depend on the site of the blow and the height of the drop – and the consequent behavioral changes are reproducible. The procedure was developed in the early 1990s in the laboratory of Wise Young, a neurologist then working at New York University and now at Rutgers. He wanted to create a model for spinal-cord injury, in order to test and evaluate proposed treatments to repair the damage and restore some degree of function. Not long before, three scientists at Ohio State University had devised a rating scale for precise scoring of loss of function in spinal-cord injury. Young adapted the scale to his rat model, based on how well or poorly an injured creature could walk. In 1995, he showed that the behavioral rating varies in direct proportion with tissue damage at the injury site. In a recent conversation, he said, “This was the first behavioral outcome measure that correlated with morphological damage in the spinal cord.” Although no one measure is universally accepted in spinal-cord-injury work, Young said, “This comes close.”
The spinal cord is remarkably well protected, by bone and by its tough outer layer, the dura. In humans, only about 10 percent of spinal-cord injuries, caused by mishaps like a bullet through the spine, interrupt the cord completely. Ninety percent are contusions. Nerves in the adult central nervous system, including the spinal cord, do not spontaneously regenerate. Some nerves in the peripheral system, however, can – importantly, in the presence of Schwann cells, a type of cell that provides an environment favorable to new growth of nerve axons. Many attempts have been made to transplant such cells into damaged spinal cords, to promote regeneration, but they have all failed.
Enter olfactory ensheathing glial cells – bearing the hope of a way to fix, or at least to ameliorate, spinal-cord injuries. In 1984, Ron Doucette, at the University of Saskatchewan, described a new kind of cell, which he had found in the olfactory nerve and the olfactory bulb. The olfactory nerve is the only central-nervous-system nerve that continually regenerates throughout adult life. It is made up of neurons that arise in the mucous tissue of the nose and run the short distance to the olfactory bulb, one of the most primitive parts of the brain.
We sniff substances all the time that are toxic to these neurons, which die and must be replaced. New ones are constantly being generated. They send axons up the olfactory nerve to establish fresh connections to the bulb. Doucette’s new-found cell produces a particular protein that marks it as a glial cell – a class of support cells, which include Schwann cells, that surround neurons. The surface of Doucette’s cell carries what are called cell-adhesion molecules, which attract growing axons. In the years after his discovery, Doucette isolated these cells and learned to grow them in tissue culture. He found that they wrap around axons and promote their growth: hence the name, olfactory ensheathing glial cells. In 1990, Doucette proposed that they are the principal reason the olfactory nerve can regenerate. Then and today, he has been pursuing how exactly Schwann cells and ensheathing cells do what they do.
The exciting question was whether the glial cells might encourage regrowth of spinal-cord neurons. Several scientists jumped on it, conspicuously Almudena Ramón-Cueto of the Universidad Autónoma de Madrid in Spain and Geoffrey Raisman at the National Institute of Medical Research in London .
Ramón-Cueto first tried cutting the peripheral nerves of rats at the point, called the spinal root, where they connect with the spinal cord. Such injuries are crippling. Normally the nerves will not grow back into the spinal cord. She then transplanted some of the creatures’ own olfactory ensheathing glial cells into the region of the root, and in 1994, she claimed that this allowed the nerves to regenerate their connections. She then went to work with Mary Bunge of the Miami Project to Cure Paralysis, which is at the University of Miami. Bunge’s main approach has been to graft Schwann cells into rats’ spinal cords, bridging spinal lesions, and then to try various measures, including drugs in different combinations, to get them to grow. In 1998, she and Ramón-Cueto injected adult-rat olfactory ensheathing glial cells into the areas at each end of the Schwann bridges. They reported that six weeks after the combined graftings, spinal-cord axons were growing through the Schwann cell bridges and beyond – and that the ensheathing cells had migrated, accompanying growing axons through and alongside the Schwann bridges.
Raisman, meanwhile, was also experimenting with olfactory ensheathing glial cells. In 1985, he had suggested that these cells had special properties that enabled them to repair central-nervous-system neurons. Now, in a clever experiment, he used a thin electrode to burn through rat spinal cords on one side only, at a point that left the creatures able to use only one forepaw. Before the operation, he had trained the rats to reach through a hole for pellets of food with their forepaws, using one or the other with equal facility; afterwards, they were unable to reach with the affected limb but could use the other normally. He then transplanted into the spinal lesions a mixture of cell types, including olfactory ensheathing glial cells. In 1997, Raisman and colleagues reported in Science that as early as ten days after the transplants, spinal-cord axons sprouted and grew across the lesions. Two to three months after the transplants, of a group of seven rats, four were able to use either forepaw as adeptly as normal rats. Dissection showed that these four had regrown spinal-cord axons across the lesions.
In 2000, after returning to Spain, Ramón-Cueto published a paper in the journal Neuron asserting that when she transected the spinal cords of rats and injected olfactory ensheathing glial cells into the lesions, many of the rats recovered some locomotor function. The degree of regeneration and recovery was slight, and some raised questions about exactly how she did the tests. Yet the paper had impact.
The pressure is now intense to get to clinical trials. The United States alone has on the order of 200,000 patients with spinal-cord injuries. (Their plight was dramatized by Christopher Reeve, the quadriplegic Superman and spinal-cord campaigner, who died on October 10, 2004.) Raisman is pushing toward trials, as is Ramón-Cueto. In June 2003, Raisman told the BBC, “My guess is we are probably two to three years away. It could be less.” A group in Brisbane, Australia, led by Alan Mackay-Sim, has duplicated the rat experiments with ensheathing cells and is at the stage of exploratory clinical trials; Carlos Lima, from the Egaz Moniz Hospital in Lisbon, has treated a small number of patients. Yet extreme caution is obviously necessary: the procedure raises great scientific, medical, regulatory, and ethical problems. In a recent telephone conversation, Doucette emphasized repeatedly that the basic physiology is still not understood. “Just putting the cells in and saying, ‘Oh, great, we’ve got some functional recovery,’ and then moving on to the next step, to me isn’t satisfactory. I want to know how it happened. Why. And how you can control it,” he said. He went on: “My view is that I think we’re probably five, ten years away. In terms of being at a stage where I’m confident we know enough about what’s going on.”
Enter Dr. Huang Hongyun.
In 1999, a Chinese neurosurgeon named Huang Hongyun arrived at New York University School of Medicine from Beijing, wanting to work with Wise Young and learn about spinal-cord injury. Young had moved to Rutgers, so Huang followed him there. “He wanted to know what to do,” Young said. “Studies recently published had claimed that olfactory ensheathing glial cells transplanted into spinal cords would regenerate rats and improve locomotor recovery. I was skeptical about some of the results. They were mostly based on, I thought, fairly questionable behavioral outcome measures. So I suggested to him, Why don’t we do it in our spinal-cord injury model?” Huang worked with Young for several years, then moved back to Beijing, becoming chair of neurosurgery at Chaoyang Hospital.
Almost at once, Huang began operating on human patients with injured spinal cords. In March 2003, he and colleagues submitted a four-page paper to the Chinese Medical Journal, which published it in October of the same year.
The journal is something of a historical oddity. It comes out monthly, about 100 pages an issue, entirely in English except for contributors’ names. It was founded in 1887 by missionaries who wanted to bring Western medical methods and standards to China and needed an English-language publication that would present the best of modern Chinese medical research and clinical practice. In the first half of the 20th century, it was well respected; after the Communist takeover of mainland China, it declined badly. Only in the last five years or so has the journal begun to regain quality and the respect of non-Chinese scientists. But scientists do not consider the journal to be peer reviewed – at least, not to Western standards. Submitted manuscripts may be looked over by various senior medical-faculty members, but if anything, this is a liability, for a uniquely Chinese reason: Confucian tradition still inculcates profound respect for elders. To turn down a paper submitted by a senior person would be an act of disrespect.
Huang’s paper reported results of surgery on 171 patients, 139 male and 32 female, ranging in age from 2 to 64 years, with the average age just under 35. All had suffered extensive paralysis and loss of sensation. The time since injury was at least six months and as much as 18 years. All had received previous therapy of one sort or another, for example, administration of nerve-growth factors and surgery, if that had been necessary to relieve pressure on the spinal cord. A requirement was that magnetic-resonance imaging showed no gap in the spinal cord and no compression.
The surgical procedure*, which the paper described in detail, is essentially to perform a laminectomy at the site of the damage, open the dura, and inject ensheathing cells. These Huang derived from olfactory bulbs. Although the paper does not mention this, in later discussions Huang has said that the cells come from fetuses aborted in the fourth month of pregnancy. (But they are not stem cells, as has sometimes been reported.) He grows them for two weeks in cell culture, as he learned to do in Young’s lab. He then injects 50 microliters of a cell suspension, approximately half a million cells, into the spinal cord, next to the ends of the lesion.
Before the operation, patients were assessed for degree of paralysis and for sensitivity to light touch and to pinpricks, following an international standard. They were reassessed between two and eight weeks later. The paper claimed that patients made significant if relatively slight improvement in these measures. However, the data are scanty and impossible to evaluate reliably. The subjects are grouped by age but not differentiated further, not even, say, into male and female. The paper describes no individual cases. It offers no before and after scores, just degrees of improvement, and these as averages within each age group. It says nothing about possible deleterious effects, not even that there were none. It reports no long-term outcomes.
Sound and Fury
The report drew immediate and intense attention. Discussion groups sprang up on the Internet; within weeks, thousands of patients from the United States and elsewhere had got in touch with Huang. First to report the story in print was Jerome Groopman, at the New Yorker, in a profile of Christopher Reeve published on November 10, 2003; he described a range of animal experiments that Reeve was following, including Young’s and especially Ramón-Cueto’s, and gave five paragraphs to the promise of Huang’s work and some of its problems.
In February 2004, in Vancouver, British Columbia, a consortium called the International Campaign for Cures of Spinal Cord Injury Paralysis held a two-day international workshop on clinical trials. Several speakers presented preliminary results of treatments involving drugs. Three spoke of clinical trials involving olfactory ensheathing glial cells, surgically implanted. Mackay-Sim, from Brisbane, described an initial human trial testing the safety of his procedure. He used ensheathing cells from each patient’s own mucosa, purified and grown for six weeks in culture, then injected at 40 small sites in and around the patient’s spinal lesions. Four patients got transplants; four got placebos. His assessments before and afterwards were elaborate and blind, the best in the business so far. Results were not yet in. Lima, from Lisbon, reported that he had treated seven patients by taking portions of their own olfactory mucosae, containing many sorts of cells, and transplanting these directly into spinal-cord lesions. Improvements were minimal, and one patient got worse. Lima used no placebos, and assessments were not blinded.
Huang reported his work – announcing that he had now given fetal-olfactory-ensheathing-cell transplants to more than 300 patients, including a number of Americans and other Westerners. Some patients, Huang said, showed improvements two or three days after the operation, although all experimental evidence said that nerves could not regrow that fast. He had tried no placebos; his assessments were unblinded and were thought rudimentary. He reported no adverse consequences, although with so many cases that was implausible. Follow-up was minimal and never conducted more than a few months after the procedure. The ethical risks were obvious and considerable.
James Guest and Eva Widerstrom-Noga, both physicians working with the Miami Project to Cure Paralysis, attended the Vancouver meeting. They came home with grave reservations; nonetheless, Bunge and her colleagues decided they needed to know more. They invited Huang to come to Miami.
Media attention built. On April 13, the Detroit Free Press ran a story about the Rehabilitation Institute of Michigan, located on the campus of the Detroit Medical Center. The previous fall, the institute had announced it would screen patients for possible operations in China or Portugal. After that, two patients had gone abroad, Robert Smith to Beijing and Erica Nader to Lisbon. While in the United States, Huang had visited the Rehabilitation Institute. Now with a waiting list approaching a hundred, the institute said that in August it would open an outpatient center where applicants would be evaluated and patients returning from China or Portugal would be monitored. The institute was already following up with Smith and Nader, and the newspaper’s account of their progress, though cushioned with language like “steady progress” and “long road to recovery,” was glowing.
That same day, public broadcasting stations aired an hour-long program called “Miracle Cell,” part of the starry-eyed series Innovation. Though it didn’t mention Huang, the program presented Lima’s work in Lisbon, enthusiastically overstating the progress his patients had made, and gave Raisman in London a platform from which to announce his plans for clinical trials. “Miracle Cell” repeatedly confused fetal olfactory ensheathing glial cells with stem cells.
Huang lectured at the Miami Project on May 5, 2004. Guest arranged to visit him for 10 days in July, accompanied by Tie Qian, a physician specializing in physical medicine and rehabilitation with the Miami Veterans Affairs Medical Center who is Chinese and speaks the language.
The second week in June, Tim Johnson, a reporter for the Knight Ridder News Service, filed an article from Beijing about Huang, his hospital, and his claims. It was picked up by a number of papers in the chain, including the Lexington, KY, Herald-Leader and the Miami Herald. On July 30, the Scientist, a weekly magazine of science news and features, carried an article about Huang. The Asian edition of Time ran a similar story from Beijing in its August 16 issue.
On August 27, the Chicago Tribune ran an article by Michael Lev that began, “A Chinese neurosurgeon has been besieged by desperate Americans willing to pay $25,000 for an implant of cells from aborted fetuses, a controversial and scientifically unproven procedure.” The piece was more thorough than most in voicing the uncertainties and reservations about Huang’s claims. Yet febrile publicity and desperate hope were by that time driving the public response. In Lev’s article Huang claimed that he had performed 450 transplants, while the waiting list for his procedure had grown to more than a thousand, including a hundred Americans.
A House Call
At half past eight on the morning of Friday, September 10, 2004, a meeting began at the laboratories of Massachusetts General Hospital. Huang was to speak. The meeting was limited to physicians and scientists. The chief organizer was Robert H. Brown, a professor of neurology at Harvard Medical School and director of the Day Laboratory of Neuromuscular Research at Mass. General. I had spoken with him by telephone early in the week; he told me he was skeptical.
Huang is of medium height, with a receding chin, and seemed somewhat diffident. His English is limited and strongly accented. He was there, it turned out, not to present his work on spinal-cord injury but to discuss another project that, he said, he had begun 18 months earlier. The title of his talk was “Olfactory Ensheathing Cell Transplantation for Amyotrophic Lateral Sclerosis.” ALS is the devastating nerve disorder better known as Lou Gehrig’s disease. (The accounts in the Scientist, Time Asia, and the Chicago Tribune had mentioned Huang’s turn to ALS.) Huang offered some minimal PowerPoint slides. His summary claim, at beginning and end: “OEC transplantation is safe, feasible, and rapidly improves partial function. Results are observable in two or three days, and improvement continues for two to three months. The mechanism is unclear.” However, his data were shockingly thin – indeed, insultingly so, I came to think. He finished up with half a dozen brief, blurry before-and-after videos of six of what he said had been a set of eight ALS patients, newly able to walk, or to stand, or to sit up, or to move the tongue enough to talk, if indistinctly. Each was followed by charts depicting nerve function before and after the transplant surgery.
His audience treated him with caution and courtesy, while its skepticism and impatience steadily increased. Much of the simplest factual information – pre-data, one might call it – was missing. Halfway through the question period, I asked several questions. When did his work with ALS patients begin? January 2003, he answered. But the videos carried dates, and these were as recent as mid-August 2004, just three weeks earlier. How many patients had he treated? He gave no clear answer; after follow-up questions from others, the likely number seemed to be 10 or 11 – until he said there had been “about 40.” Did they all get fetal cells? No answer.
As the questioning went on, problems with Huang’s methodology seemed to emerge, chiefly the lack of rigorous pre- and postoperative evaluation of patients’ functioning, the lack of controls, and, above all, the total absence of follow-up beyond a few months.
On his home ground, Huang is more assured, smoother. Indeed, with Chinese visitors and with patients, he evinces a certain quiet charisma. Chaoyang Hospital, Beijing, is part of a set of gray, grimy stone buildings around a gated courtyard, with no clear indication of which is its main entrance. Huang’s office is on the hospital’s top floor, but we met on the second, in a serviceable workroom with a central set of tables and, around the walls, shelves haphazardly filled with equipment and supplies. The room sits at the head of a dim corridor along which open, on either side, wards with six beds each, some empty, some occupied by patients, though not all are spinal-cord cases. The patients are surrounded by members of their families – as is customary in China, where much of patients’ care falls to relatives.
Huang and I discussed his procedures in detail. Some who had heard him in the United States wondered whether the cells he implanted were a raw mixture or purified. “We get the olfactory bulb out,” Huang said. “Of course, mixture. Then we culture them and purify them.” The dose for a spinal-cord patient is one million cells, “90 percent OEG cells.” Had he published anything about safety? He ducked the question at first, then said that the cells caused “no long-term fever.” He elaborated: “No problems with the cells; maybe we have complications of the surgery – infection of the area, leakage of the cerebrospinal fluid. The general complications of other surgery.”
How much did patients gain? Again, he ducked. Before and after the procedure, he said, patients were evaluated by three doctors, according to standard protocols, for movement, for control of the anal sphincter, and for sensitivity to touch and pinprick. Did any patients have adverse reactions? “Ah, a very complicated question.” But then, “In actual score, no patient got worse.” But the degree of improvement? These patients are in bad shape, he said. “Any improvement is a bonus.” Any complete cures? “I don’t think it is possible to cure this disease.” Even when progress is minimal and gradual, Huang said, it is valuable. “Complete chronic injury, no chance to get 100 percent.”
Critics in the United States have suggested that any patient with spinal-cord injury or, for that matter, ALS who comes to a medical center for some major procedure will probably get a variety of other treatments at the time, and this by itself might provoke temporary improvement. Did patients at Chaoyang Hospital get other treatment as well – such as physical therapy or other rehabilitative help? “No,” Huang said. Physical therapy is not routine in China. “They go home.”
What about follow-up? “They start to improve in two or three days. Then we follow them in two to four weeks. Then another follow-up three to six months.” But what about the longer term? Again, critics have held that patients ought to be tracked for at least two years. Huang hesitated. Then, “Chinese patients very poor. They go home.” He said he could not get in touch with them again.
Had Huang tried to publish other papers, and in peer-reviewed Western journals? Several, he said, but so far no response. He was collaborating on a paper with Guest and Qian from the Miami Project. During their visit to China, Huang said, “They evaluated one patient” before and after surgery. “Totally paralyzed. After surgery, can do this, can do this” – he was making small finger and hand motions. How quick was the recovery? “Second day after, Dr. Guest and Dr. Qian saw some difference.” What could possibly be the mechanism for change that soon? “In front of all eyes, we saw some change, even though they know we couldn’t explain it.” In mid-October, Guest sent the completed case report to Huang, but a month later Huang still had not found time to look at it.
Huang told me that hospital policy prohibited my watching the surgery. Guest and Qian during their visit examined 12 spinal-injury patients. They formally assessed six of them before and after and indeed observed four operations. They acknowledge that some of the patients demonstrated a degree of modest improvement in motor and sensory function – and that the improvement occurred surprisingly soon. However, two patients showed “wound breakdown,” one of them suffering “a reduction in leg function.” A third patient came down with meningitis. “The Chinese clinicians did not record these complications in the medical record,” asserts an unpublished report by the Miami Project. Although Guest and Qian did watch surgery and observe patients, they were not allowed into the laboratory where the cells for transplantation were prepared and had no way to know the content of the putative human fetal olfactory-bulb cultures – not even whether the material transplanted actually contained ensheathing cells. Guest adds, “We did see one set of cultures that showed robust cell growth and morphology that could be ensheathing glia. They were very healthy cultures. We viewed them in Dr. Huang’s clinical office.” The chief problem they saw, however, was the lack of long-term follow-up, including full records of any adverse effects.
To me, the most disturbing sign was Huang’s evasiveness. He pleaded repeatedly that patients needed to be treated: “These are suffering, dying people. I am a surgeon. The first thing is to save lives and alleviate suffering.” Though this sentiment may be genuine in Huang’s case, such evasions are a classic mark of the charlatan. Alternatively, he asserted that important types of controls (for example, surgery that mimicked the operation but injected not cells but salt water) would be dangerous and unethical. He insisted repeatedly that the procedure is safe.
Huang’s methodology is a moving target: from work with spinal-cord lesions to ALS, from injections into the spinal cord to injections into brain tissue. Critics have demanded that the procedure involve a fixed quantity of injected cells, one or a few standard points of injection, and significant blinded controls, and that evaluation follow a standardized protocol, including, for example, rigorous pre- and postoperative physiological tests that measure such properties as breathing, muscle tone, and strength.
Despite the defects in Huang’s work, no definitive judgment is yet possible. Wise Young is a cautious advocate. He notes, “There are really no randomized clinical trials for any of the current neurosurgical procedures.” Regarding Huang’s work, “The big debate right now is, What is the level of evidence that’s necessary and sufficient to take something to clinical trial?” Meanwhile, though, when dealing with spinal-cord patients and their families, “My official recommendation is that they should wait. Many of them ignore me; they go on ahead to do it anyway.” Mary Bunge and her colleagues at the Miami Project find Huang’s claims frustrating. “Presently, Dr. Huang’s project by research standards in the United States is not a clinical trial but is a clinical treatment series. The treatment series does not meet the design standards for a clinical trial that would allow for definitive results to be obtained.” Yet they call for “independent and impartial assessment of the risks and benefits of this cell therapy.” Meanwhile, though, “Miami Project faculty do not endorse this procedure and at this time would not advise individuals to undergo this surgical transplantation strategy. While some people with SCI will view these current experimental procedures abroad as their only hope, by participating they may be putting themselves in harm’s way.”
The science of Dr. Huang Hongyun raises to our awareness this deep tension over standards of evidence and the ethics of clinical practice.
I saw Huang the afternoon of October 20, 2004. A correspondent from the Mobile, AL, Register, Karen Tolkkinen, was also in Beijing, Huang said; he was to treat several Americans with ALS that week, and one was from Alabama. That evening, he operated on Ronnie Abdinoor, a 47-year-old from New Hampshire. On October 29, Tolkkinen reported in the Register that Abdinoor had died.
The big new idea for making self-driving cars that can go anywhere
The mainstream approach to driverless cars is slow and difficult. These startups think going all-in on AI will get there faster.
Inside Charm Industrial’s big bet on corn stalks for carbon removal
The startup used plant matter and bio-oil to sequester thousands of tons of carbon. The question now is how reliable, scalable, and economical this approach will prove.
The dark secret behind those cute AI-generated animal images
Google Brain has revealed its own image-making AI, called Imagen. But don't expect to see anything that isn't wholesome.
The hype around DeepMind’s new AI model misses what’s actually cool about it
Some worry that the chatter about these tools is doing the whole field a disservice.
Get the latest updates from
MIT Technology Review
Discover special offers, top stories, upcoming events, and more.