The Last Shoe Drops for Solanezumab

Just a tad too late for inclusion in the November issue of NeuroPerspective, the biomarker data for solanezumab were presented, at the CTAD meeting in Monte Carlo. As we had noted in a previous post (‘Chinese Data Torture’),  the data disclosure had been fragmented into three chunks, this was the last one.

What does the prodigal solanezumab biomarker data tell us? Of all the AD biomarkers utilized–beta amyloid in blood and plaque in brain, phosphorylated tau, brain volume changes–the only one that showed any difference at all was blood beta-amyloid, albeit not to the level of statistical significance. This is not a robust finding, given a program that enrolled over 2000 patients, and covered 18 months of treatment,

The most optimistic hope that one might take from this is that the biomarker might be related to the very modest slowing of cognitive decline observed in mild AD patients in this trial. Perhaps, though it’s difficult to get too enthused about a hypothesis that links a clinically insignificant  benefit to a statistically insignificant biomarker effect, particularly when all the other biomarkers come up empty.

Adding to the ambiguity is the fact that these biomarker results are close to the inverse of those reported for the bapineuzumab trials. There, the mAb reduced amyloid deposits in the brain and CSF phosphorylated tau, but apparently had no impact on beta-amyloid in the blood. And as has been previously discussed, bapi produced no changes in the rate of cognitive decline. Yet those presenting the bapi data state that the findings confirm ‘target engagement and an impact on neurodegeneration markers.’

There is a disconnect here. While these two mAbs differ in their binding-site and dosing regimen, the data around ‘target-engagement’ is internally contradictory. There is no consistency in the results for beta-amyloid plaque, soluble beta-amyloid in the blood, or tau in the CSF. The data apparently do agree that the use of a mAb has no impact on brain or hippocampal volume, in this, a disease of  neurodegeneration.  That is not entirely reassuring.

The Bottom Line is: A billion plus dollars later, the data for the use of monoclonal antibodies against beta-amyloid is a mess. Neither the target nor either of the payloads have been validated. Concluding that the barely discernible cognition effect in a subpopulation receiving solanezumab is due to the statistically insignificant lowering of blood AB is highly optimistic. It could be correct, but one barely has correlation here, let alone causality.Bapineuzumab is dead because of its safety profile, while solanezumab is only still breathing because there are some threads that raise the hope–and it is little more than that at present–that introducing it earlier in the course of the disease might reveal more impact. Since we do not yet know how to do so–our ability to reliably diagnose Alzheimer’s in mild-moderate stage is highly imperfect (15-20% of patients diagnosed with AD do not show AD signs post-mortem), let alone reliably diagnose prodromal AD–there is at present not even a clear road to assessing this.  One can choose to conduct the kind of long term prospective study in a vulnerable population like the Colombian trial of crenezumab, but that requires that we assume that such familial AD is a valid predictor of a drug’s behavior in sporadic AD. And there is no basis for that assumption.

One thing we do know: It would have been more efficient to present the solanezumab data in a single bolus. Titrating the dose did not improve its tolerability.

This entry was posted in Uncategorized and tagged , , , . Bookmark the permalink.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s