John L. Berk, MD: There are new therapies that are coming along. Before those new therapies come along, is there any other unmet need that you can see, if we have these agents that are quite powerful? And will it continue to improve? What do you need, diagnostically, that’s going to help you determine when therapies absolutely should begin?

P. James B. Dyck, MD: I think this is where clinical judgment is still very important. We don’t have a nice biomarker to say that you now, clinically, are developing amyloid polyneuropathy or amyloid cardiomyopathy; or it may be better in cardiomyopathy than polyneuropathy. So, symptomatic onset and the battery of tests and a neurologic exam. The neurologic exam is still very important in these patients.

John L. Berk, MD: So you’re advocating, at the moment, that there be end-organ changes before we actually have the capacity to treat?

P. James B. Dyck, MD: Yes. And again, for diflunisal, maybe not. But certainly for the silencing drugs, at $450,000 a year, we can’t be putting patients who haven’t demonstrated they have a neuropathy yet on these drugs.
John L. Berk, MD: Michael, what’s the biomarker we’re missing?

Michael J. Polydefkis, MD: Well, they’re approved for hATTR [hereditary amyloid transthyretin amyloidosis] neuropathy, so you’re a little bit restricted to what the label says. But if it were up to me, I would be treating people very early on. That’s what I would want as a patient, but the costs and the other factors complicate issues.

John L. Berk, MD: So for all 3 of you, the question then is: if you have a capacity to detect when the variant protein begins to misfold, would that tool be sufficient to guide you in instituting therapeutics? Akshay, what do you think about that?

Akshay S. Desai, MD, MPH: Well, I think that might mark the beginning of the vulnerability to development of disease. But I think as has been mentioned, one of the challenges is really not understanding exactly the nature of the pathophysiologic progression here. I think we have some broad models for what happens, but whether that step itself, and if we could key in on it as the signal to initiate, of the timing of therapeutic initiation, I think is still a hard question to answer. I can step back a little bit and look to my colleagues here because I think for us, really the trigger to initiate therapy would be the onset of heart failure symptoms. It would be hard to argue to initiate therapy well in advance of that, given the data we have.

But I think the challenge for us is that we often see these patients and don’t recognize that they’re there. As I’ve learned today, the index of suspicion needs to be pretty high. There are some auxiliary clues in the clinical history and in the family history that we should try and integrate, and maybe those are our current biomarkers—symptom onset and a detailed context analysis.

John L. Berk, MD: I think one of the things we have learned from these clinical trials is that the people who are earlier in the spectrum of disease seem to benefit most from the interventions. Having a biomarker that would permit us to intervene early, I think, is going to be critical, not only for cardiac disease where there’s a huge burden of amyloid, but also in the neuropathy, where there’s less evidence histologically of amyloid burden and maybe a little more toxicity from the oligomers and the aggregates that circulate.

It’s been a real pleasure talking with you, Michael, Akshay, and Jim. You represent a huge amount of amyloid innovation, and I look forward to doing this again sometime. Thanks a lot.