SHOW ME THE DATA!…

Posted on Wednesday 10 October 2012


Shares of Eli Lilly surged today after investigators spelled out the encouraging solanezumab data they plucked out of a secondary analysis of a group of patients with a mild form of the memory-wasting disease. As Lilly conceded earlier, both of its late-stage studies for solanezumab failed to produce a statistically significant response for mild-to-moderate patients on two key endpoints. Simply put, the study failed. But after a secondary analysis of the first study showed that there was a 42% reduction in the rate of cognitive decline among a subpopulation of patients in the solanezumab arm with only a mild form of the disease, investigators decided to hunt for confirmation of that endpoint in a second Phase III. They didn’t find it, seeing the numbers fall short of statistical significance after switching from one measure [ADAS- Cog11] to another [ADAS-Cog14]. But by "pooling" the data they came up with a 34% reduction in cognitive decline in that particular group. None of the data indicated a significant reduction in the rate of functional decline.

The independent analyses of the EXPEDITION study data was conducted by the Alzheimer’s Disease Cooperative Study, an academic research consortium and presented at the annual meeting of the American Neurological Association. Solanezumab is designed to reduce levels of beta amyloid, a toxic protein that accumulates in the brains of Alzheimer’s patients. Eli Lilly execs pushed hard to assert that the data represented an encouraging advance in the field, which has been blighted by a string of high-profile failures. But in the drug development world, data derived from secondary analysis is considered a notoriously unreliable gauge of efficacy. Lilly’s second study produced evidence of only a weak response among mild patients, conflicting with the response seen in the first study. In the world of clinical drug development, Lilly has only demonstrated that it detected a weak signal in a particular subpopulation.

Lilly’s investigators, as well as patient groups desperate for some sign of progress, highlighted the positive. "Alzheimer’s disease research has been extremely challenging," said Rachelle Doody, a professor at Baylor. "The data results from the solanezumab Phase III trials were encouraging to the ADCS team. These results represent an important step for the medical, academic, and scientific communities in understanding brain amyloid as a target of AD therapies." Lilly execs agreed, "Our next step is to talk to the regulators to see what this means, although we think this is a step forward in getting to a treatment for Alzheimer’s," David Ricks, president of Lilly Bio-Medicines, told The Wall Street Journal. That sentiment helped drive up Lilly shares 5% this afternoon as investors bet on the first encouraging sign of success in years of expensive studies.

The $60 billion question here is whether Eli Lilly and its science team has the kind of data indicating that they are on a solid path to an approval. The likely answer is they don’t and will have to gamble on at least one more big study to put this theory to the test again. Another potentially complicating fact: Investigators found that patients in the drug arm experienced a significantly higher risk of angina. "This isn’t the home run that is going to get you a medication by the end of the year, but it is very encouraging because with subsequent studies this could be a viable therapeutic option," Maria Carrillo, the VP of medical and scientific relations at the Alzheimer’s Association, tells Bloomberg. "We need more options for our patients and families. We are very encouraged by what this could mean."

Here we go again, "Lilly’s investigators, as well as patient groups desperate for some sign of progress, highlighted the positive" [drug company enthusiasm + high patient group expectations = danger, danger]. And this story is more noisy on the Financial web-sites than the medical. It’s easy to be even more skeptical based on "But by ‘pooling’ the data they came up with a 34% reduction in cognitive decline in that particular group. None of the data indicated a significant reduction in the rate of functional decline" [drug company enthusiasm + high patient group expectations + secondary analysis of pooled data = danger, danger, danger]. But the hingepoint for me is that there’s no data from either study on clinicaltrials.gov. This is going to be a fine test case for SHOW ME THE DATA! It’s telling that "The $60 billion question here is whether Eli Lilly and its science team has the kind of data indicating that they are on a solid path to an approval" rather than "Does it work in Alzheimer’s Disease?" But I guess that’s expected.
Experimental Lilly drug may help slow mental decline in people with mild Alzheimer’s disease
Washington Post
[AP]
October 8, 2012

… Solanezumab aims to bind to and help clear the sticky deposits that clog patients’ brains. The two studies each had about 1,000 patients, about two-thirds with mild disease and one-third with moderately severe Alzheimer’s, in 16 countries. Their average age was 75.

The main measures were two tests — one reflecting language, memory and thinking and the other, ability to perform daily activities such as eating and grooming. The combined results on the mild disease patients showed a nearly 2-point difference in the roughly 90-point score on thinking abilities. Previous studies suggest that a change of 3 to 4 on the score is needed to show a clinical benefit, like an improvement in how well patients can take care of themselves.

“It’s a small difference,” Dr. Rachelle Doody of Baylor College of Medicine said of the drug’s effect. She heads a nationwide research network funded by the National Institute on Aging that did an independent analysis of Lilly’s results on the studies and presented them Monday at the conference. Still, “you slow the decline” with the drug, she said. Independent experts cautioned that the improvement was small, and needs to be verified in another study.

SHOW ME THE DATA!
  1.  
    October 10, 2012 | 9:42 PM
     

    Solanezumab? Satan’s playmate? Who names these drugs?

  2.  
    Stan
    October 10, 2012 | 11:47 PM
     

    The propaganda data illusion & empty promise game never ends…The Pharmaceutical Corporate Money Train is leaving the station once again…toot toot….

  3.  
    October 11, 2012 | 12:03 AM
     

    Wiley,

    I wondered about the name too. I found this, for whatever it means [WHO INN]. I guess solanezumab is easier to say than:

    immunoglobuline G1-kappa, anti-[Homo sapiens amyloïde-bêta (Abeta) peptide monomère soluble], anticorps monoclonal humanisé; chaîne lourde gamma1 [VH humanisé (Homo sapiens IGHV3-23*04(87.60%) -(IGHD)-IGHJ4*01) [8.8.5] (1-112) -Homo sapiens IGHG1*01, CH3 K130>del (113-441)], (215-219′)-disulfure avec la chaîne légère kappa (1’-219’) [V-KAPPA humanisé (Homo sapiens IGKV2-30*01 (90.00%) -IGKJ1*01) [11.3.9] (1′-112′) -Homo sapiens IGKC*01 (113′-219′)]; dimère (221-221″:224-224″)-bisdisulfure neuroprotecteur

    It’s a synthetic monoclonal antibody against beta-amyloid – the “goo” thought to be the problem in Alzheimer’s…

  4.  
    annonymous
    October 11, 2012 | 1:34 AM
     

    http://www.nytimes.com/2012/10/11/business/glaxo-opens-door-to-data-on-its-research.html

    Not enough, but at least it indicates this issue may be gaining traction.

    Still, the silence from the professional societies is deafening.

  5.  
    Bernard Carroll
    October 12, 2012 | 8:20 AM
     

    Dr. Nardo put his finger on the central issue with this observation: It’s telling that “The $60 billion question here is whether Eli Lilly and its science team has the kind of data indicating that they are on a solid path to an approval” rather than “Does it work in Alzheimer’s Disease?” The central issue is the perverse incentive of billions in profit from a drug that doesn’t do very much.

    An FDA approval is the lowest bar possible, especially when nominal statistical significance is valued over good clinimetric evidence. That is what I have called the Cheshire cat effect – the only evidence that the drug was there is a disembodied p value, grinning in statistical space, with no connection to clinical reality (see PubMed ID 15169734).

Sorry, the comment form is closed at this time.