At this year’s International Alzheimer’s Conference, the results of the Phase 3 clinical drug trial for LMTM, a drug targeting the build up of Tau protein, were announced. These results have been eagerly awaited, as LMTM is the first drug targeting Tau protein to reach advanced clinical testing. Alzheimer’s is characterized by abnormal build up of two proteins, Abeta and tau. Until now, pharmaceutical companies have focused, unsuccessfully, on reducing Abeta. In light of these repeated failures, scientists have begun to focus on reducing Tau protein.

The Trial Design. There were 891 people in this trial who were divided into three groups. One group received a high dose of the drug, one received a lower dose of the drug, and the control group received a placebo. The trial lasted 15 months and the participants were tested for cognitive and functional status. They also underwent scans to evaluate brain atrophy.

Results. The primary results of the trial indicated that the drug failed to slow either cognitive or functional decline in people with mild to moderate Alzheimer’s. The disease progression curves among all groups were essentially identical.  However, the pharmaceutical company separately evaluated the outcomes among the 15% of participants who were not already on a medication to moderate Alzheimer symptoms. Among this group, the pharmaceutical company announced strikingly positive results: up to an 80% reduction in progression.

Objections of the Larger Research Community. Researchers and clinicians not involved in the trial immediately took exception to this secondary “finding”, unequivocally stating that the suggestion of benefit for those not already on medication was without merit, and making these points:

• Participants not on medication to moderate symptoms might either be naturally slow progressors, or lack medical care. Either reason would make these participants outliers in the study.

• The secondary statistical analysis did not compare the results of the un-medicated participants who received LMTM to the results of the un-medicated participants who had not received the drug, but to all those who received the placebo, a deeply flawed statistical design.

• The trial participants with the APOE4 gene, which increases the likelihood of developing Alzheimer’s, were underrepresented as a proxy for the population as a whole.

• None of the trial participants were screened for the presence of tau protein.

• The results of the trial have not yet been published in a peer-reviewed journal.

Larger implications.

Until the basic science underlying the disease process is better understood, we are likely to have drug trials based on flawed assumptions and inadequate clinical trial designs. Those whose trials fail are under pressure to recover some of the costs associated with the trial, and one avenue is through a massaging of a subset of the data to suggest efficacy in a smaller target group. Without the vigilance and courage of the scientific community, research time and money can be diverted by those seeking to recoup investments through analyzes that are not supported by rigorous statistical analysis.