How do we know whether a drug works or not?

It seems to be a simple question. But in Parkinson's not so simple to answer based on how clinical trial results are interpreted and publicized.

The Clinical Trials Lessons

Let's look at two examples - GLP-1 agonists and anti-alpha-synuclein antibodies. To do so let's remember that the primary means of assessing treatments in Parkinson's is via the UPDRS (Unified Parkinson's Disease Rating Scale). The current version has 4 sections with Parts II and III most often used as outcome measures in clinical trials. Part II is a series of questions answered by a patient to reflect how they feel or function. Part III is a series of questions answered by a doctor to reflect how they believe the patient is feeling or functioning. And each looks at the disease from a wide range of perspectives. The score for each question are added to get a score for each Part, with the change over time reflecting benefit or worsening.

I'm not going to discuss the validity or advantages/disadvantages of the UPDRS. But it is important to understand that the FDA is more interested in how a patient describes the effects of a therapy than they are about how a doctor interprets the effects of a therapy. The patients' experiences are of paramount importance to regulators.

Trials of GLP-1 agonists and anti-alpha-synuclein antibodies have something in common. In phase 2 trials, physicians assessed the effects on patients as showing improvement on the UPDRS Part III, while at the same time, the patients reported no change in how they felt or functioned on the UPDRS Part II. Despite failing to improve how patients experienced the effects of these drugs, both advanced into Phase 3 (or at least another Phase 2 study). At a certain point, shouldn't we consider whether we are pursuing the wrong therapies or looking at the disease the wrong way?

When developing drugs it is critically important to make decisions based on data and on what the FDA will require, rather than what we want or hope to see. I don't understand the decision-making processes that lead to really, really smart people deciding to continue on these paths. I do recognize that being critical of these folks may not be in my best interest, as they may be advisors to the investors I am targeting! But we need to follow the path the data illuminates.

The FDA's Point of View

The FDA is clear: they will consider a drug to be beneficial when it favorably affects how patients "feel, function or survive." Note that a physician assessment is not part of the FDA's requirements as summarized in its three-part definition.

While the UPDRS is an accepted way to assess therapies, due to its long track-record and careful study, it does not directly assess how patients "feel, function or survive," but rather represents how patients feel about their function.

In meetings with the FDA, I proposed to assess RB-190's impact using direct measures of function. They seemed open to considering such measures to prove efficacy. Of course, we need to conduct the study and complete a rigorous analysis of the data. And then we'd need to replicate the findings in a Phase 3 clinical trial. At that point, with meaningful improvements in motor function (the effect I expect to see in the short term) measured by how a patient feels or functions, the drug should be approvable by the FDA, because that would be a major part of the "proof" that the drug works for people with Parkinson's.

Later, following initial FDA approval, and then with a deep-pocketed partner or large investor, I would conduct a larger, long-term trial to demonstrate that dopamine reduction with RB-190 halts or reverses disease progression. Such data would also show that the drug works.

The key is that we are starting the clinical trial program with a rigorous plan to meet FDA's requirements and FDA's preferences for how to show RB-190 works - first on motor function and then for reversing disease progression - and doing so in a manner that demonstrates meaningful benefit for people with Parkinson's.