Researchers sceptical about Alzheimer’s drug that ‘slows disease by 80 per cent’

A new drug could slow the progress of Alzheimer’s disease by as much as 80 per cent, a study has suggested.

However, other researchers have expressed scepticism.

The drug, called LMTX, is the first to target ‘tau tangles’ — clumps of proteins that accumulate and spread in the brains of people with Alzheimer’s. Most previous drugs have targeted beta-amyloid, a different protein whose accumulation is also a marker of the disease.

During trials on 891 patients taking LMTX, participants were given memory tests. An 85 per cent slower decline was observed in patients just taking LMTX over 15 months than those taking a placebo. Brain scans also showed they had lost 38 per cent less brain matter.

However, only 135 people in the trial were exclusively taking LMTX. The drug was not effective in patients taking other dementia medication as well.

Professor Claude Wischik, the drug’s inventor, told the Daily Mail: ‘This is the first glimmer of what a disease-modifying drug looks like in practice. We have really remarkable results. Nobody has come to this point before. It is not just a big clinical effect, it is a hard-edged effect on the rate of brain shrinkage.’

But David Knopman, a professor of neurology at the Mayo Clinic, told Forbes that ‘what ultimately counts is the primary analysis’. Under the terms of the primary analysis the study failed. Only when excluding those patients who took LMTX along with other Alzheimer’s drugs could the trial be deemed a success.

Instant analysis
It is easy to be sceptical about media stories of ‘cures’ for serious diseases, as we seem to be constantly bombarded with them. This story shows considerable promise, but there are important drawbacks.

One is the sample size. Many of the results were excluded because the drug did not work when administered with other dementia drugs. This reduces the number of relevant study subjects from 891 to 135, which diminishes the study’s reliability. From a positive standpoint, we have useful extra information about the drug — that it is ineffective in the context of these other drugs. But it doesn’t tell us why it is ineffective, and therefore what else could potentially impact its efficacy.

This doesn’t negate the importance of the findings of the trial, but it does mean that further research is needed before this drug is proclaimed as a miracle cure.
MB
Research score: 3/5


  • westlondoner11

    This article has been amended after Ben Goldacre and others made objections to the original version, but I still have grave reservations about the way you have framed this story.

    For instance, there is no indication anywhere in this article of the overall outcome of the RCT (randomised controlled trial). The RCT was a failure: the researchers found that the drug had no overall effect in the sample population – it was no better than placebo in preventing mental decline in Alzheimer’s patients. American news coverage reflects that fact: the New York Times, which accurately reported the story, gave it the headline ‘Alzheimer’s Drug LMTX Falters in Final Stage of Trials’.

    Here are a couple of quotes from the NYT story, which does an admirable job of presenting the facts:

    ‘“I have to say that the results that we saw here were, to me, more disappointing than not,” Dr. David Knopman, a neurologist at the Mayo Clinic, said in moderating a news conference at the meeting in Toronto.

    Dr. Rachelle Doody, director of the Alzheimer’s disease and Memory Disorders Center at Baylor College of Medicine, agreed. “To present it to the public now as a promising approach seems unjustified,” she said.’

    Your ‘Instant Analysis’ hints at the manifold problems of this study in the sentence ‘Many of the results were excluded because the drug did not work when administered with other dementia drugs.’ But you can’t just exclude results when it suits you; the article fails to point out that a subset analysis of this kind is a hallmark of bad research. RCTs should define their primary outcomes at the beginning and then report them at the end. Redefining the terms of the study in mid-stream is a way of salvaging a positive outcome from a negative result.

    You also omit another very important caveat. In a trial of this kind there is of course a control group who did not receive the new drug. The 15% of patients who were found to receive some benefit from the drug were those who were taking no other anti-dementia medication. It would be natural – and normal practice – to compare these patients only with those in the control group who were receiving no medication at all. Instead of doing this, the researchers chose to compare them with the WHOLE control group, including those who were taking other anti-dementia medication. So the comparison between subset and control was unusual and ought to have been questioned.

    Pharmaceutical companies naturally try to put the best spin on a failed clinical trial. The first sentence of this article does little more than regurgitate their press release, when it should be reporting what was a failure. Alzheimer’s is a uniquely horrible disease which causes sufferers and their families immense distress. Presenting a failed trial as some sort of success, however minor, gives false hope where it may not be warranted. I’m disappointed that the Spectator did not look at the detail of this study and report it appropriately.