Critics Suggest FDA Approving Aduhelm Will Erode the “Public Trust”: What About Patients’ Trust?

June 24, 2021By James E. Valentine

For the last 13 years, this blogger has been at the center of what has now been dubbed “patient-focused drug development.” For 6 years, I served as a patient liaison within FDA in what was then called the Office of Special Health Issues.  This was the office (originally the AIDS Relations Staff) that was established in the 1980’s in response to the HIV/AIDS activism movement, the first function within the Agency with a mission to engage with patient stakeholders.  Over the years, this function expanded to cancer (renamed the Office of AIDS and Special Health Issues) and, ultimately, all serious and life-threatening diseases (dropping the AIDS nomenclature).  As part of this lean and mean team, I worked to incorporate patient viewpoints into study design, results interpretation, and, ultimately, benefit-risk decision-making.  In 2012 we at the Agency, under Dr. Janet Woodcock’s leadership, identified the need for more systematic patient engagement — a way to improve representativeness and be more proactive rather than reactive to interacting with patient communities.  So, we crafted the Patient-Focused Drug Development, or PFDD, a commitment that was agreed to in PDUFA V.  This launched us into a new era where we acknowledged that patients are experts in their lived experience and have important preferences to share. To ignore this wealth of expertise would be a huge missed opportunity to inform drug development and review.  Even since leaving FDA, as the PFDD initiative transitioned to externally-led meetings, my HP&M colleagues and I have helped nearly three dozen patient communities plan externally-led PFDD meetings, and that is only half of the work we do to help facilitate what the 21st Century Cures Act codified as “patient experience data.”

I have been boots on the ground in watching the culture shift both within FDA and in industry, as the value of the patient voice has been recognized as legitimate.  I have witnessed it inform whether clinical trial results are clinically meaningful (or not) as well as the selection and even development of primary endpoints.  Patient stakeholders have become partners in a decision-making process that serves as the gatekeeper to drugs – including both their benefits and their risks – but also will determine whether the status quo will remain (i.e., the morbidity, oftentimes irreversible and progressive, and mortality associated with their condition).  As someone who studied in the fields of public health, medical sociology, and biomedical ethics, I was impressed with how quickly the pharmaceutical and biotechnology arena caught up with what took many decades to occur in the context of the physician-patient relationship.  For too long, patients’ perspectives and own preferences were not considered in the provision of care by physicians at the patient level.  The practice of medicine was viewed as paternalistic.  Now there is a whole body of research that supports patients as partners in their own care, and the practice of medicine has largely corrected course.

This is what strikes me as so contradictory about the various critics who are suggesting that FDA violated the public trust when approving Aduhelm (aducanumab-avwa) for Alzheimer’s disease earlier this month (see our previous coverage of the approval here). In the last week, we have seen FDA advisory committee members resign in protest (see coverage here) and consumer advocates call for resignations of FDA officials (see coverage here). These self-proclaimed consumer and public health advocates, all who are physicians, feel they are in a better position to inform and/or judge FDA’s decision than the patients who will ultimately be the ones affected by this regulatory decision.  It harkens to the paternalism in medicine of the past.

As the ultimate stakeholder when it comes to drug approval decisions, shouldn’t we instead consider whether this builds or erodes patients’ trust in FDA?

Patient Input in the Approval of Aduhelm: Setting the Context for Accelerated Approval

While the FDA’s summary basis of approval, including supporting memos from both the Director of the Office of New Drugs (OND) and the Director of the Center for Drug Evaluation and Research (CDER), were just released (as I was writing this very post), FDA previously made clear statements about the role of patient stakeholders. In her closing statement during a press briefing on the date of the approval, CDER Director Dr. Patrizia Cavazzoni stated:

I wanted to thank the patients living with Alzheimer’s, their caregivers and health care professionals who have reached out to us over the past several months regarding the Aduhelm data. While our decisions are based on science and the data in the application, we have listened to and carefully considered the needs of the people who are living with this devastating, debilitating and incurable disease. The data supports patients and caregivers having the choice to use this drug.

What Dr. Cavazzoni lays out is the traditional way in which patients have informed drug benefit-risk decision-making: by setting the “clinical context” of the disease.  This clinical context encompasses two major considerations: 1) an analysis of the disease condition, including the severity of the condition, and 2) the degree of unmet medical need. This frames the assessment of benefits and risks, in a way, calibrating the application of the regulatory framework to the specific condition at hand.  For example, we know that cancer patients have a high tolerance for toxicity when offered a chance at improved survival, a risk tolerance that would not be shared by individuals suffering from the common cold.  We have also learned from patients with progressive conditions, like Pompe disease, that slowing progression of their otherwise universally permanently progressive condition would be of tremendous value to them – they do not need to see reduced symptom burden or improved functioning for a treatment to be meaningful.

However, the scope of patient input goes beyond setting the clinical context.  When a disease is serious or life-threatening and it has an unmet medical need, because there are serious risks associated with not approving a drug (i.e., serious morbidity, oftentimes irreversible and progressive, and risk of mortality), FDA must also balance the risks of making a type 2 error, or false negative conclusion from the data in hand.  Patients also have a role to play in setting the context for how much risk of a type 2 error they are willing to tolerate.  Or, said differently, patients can express their tolerance for less certainty of a treatment benefit.

Because no data set is completely reliable in establishing a drug’s benefit, and because no drug is universally effective for all patients, there is inherent uncertainty in every drug approval.  The question is: how much uncertainty is acceptable?  The Federal Food, Drug, and Cosmetic Act states that the evidence needs to be that “which it could fairly and reasonably be concluded…that the drug will have the effect…” (Section 505(d)).  Patients with different conditions, or even at different stages/severity of the same condition, may have different tolerance for less certainty (similar to the differences in risk tolerance discussed above).  This is why it is important for relevant patient populations to provide input as to what degree of certainty is needed for there to be a fair and reasonable conclusion of drug benefit for their condition.

One regulatory pathway that embodies this concept of “less certainty” is Accelerated Approval. These approvals, like with Aduhelm, may be granted when a benefit is demonstrated on a surrogate endpoint that is merely “reasonably likely…to predict clinical benefit” (i.e., it is not established using a validated surrogate endpoint that would support traditional approval, e.g., FEV1 in asthma or serum LDL in hypercholesterolemia).  Instead, the ultimate clinical benefit must be confirmed in a post-approval study.  If Accelerated Approval works to help expedite drugs to the market as it’s intended, then we should not expect all drugs approved on these surrogates to have clinical benefit confirmed.  It is in acknowledgement of this inherent uncertainty that FDA has provisions for expedited removal of these drugs if benefit is not confirmed post-approval.

We see that FDA considered Alzheimer’s patients’ input on tolerance for less certainty of treatment benefit for Aduhelm when Dr. Peter Stein, OND Director, shared during the press briefing:

We heard very clearly from patients that they are willing to accept some uncertainty to have access to a drug that could provide meaningful benefit in preventing the progression of this disease which as we all know can have very devastating consequences.

It is clear that this patient input helped inform the Agency’s conclusion that it was appropriate to consider approval based on an unvalidated surrogate (as long as it could conclude the surrogate was “reasonably likely to predict” clinical benefit), even without the ultimate showing of clinical benefit at the time of approval.  Patients expressed a willingness to take on the risks of a drug, even if benefits were less certain, because this uncertain chance of benefit and the associated risks of the drug was worth it to them compared to the alternative: waiting several years for another study, during which time there will be irreversible, progressive neurodegeneration.  Losing one’s identity, or possibly one’s life.  This input was further acknowledged by Dr. Stein in the concluding paragraph of his Concurrence Memorandum (at 9-10):

…approval using this pathway requires a tradeoff: patients and physicians who choose to use the drug must be willing to accept some residual uncertainty regarding clinical benefit – and therefore be willing to take a drug that may ultimately prove ineffective – along with the risks of the drug, in order to gain earlier access to a potentially valuable treatment. FDA staff has heard from patients that many would consider this a reasonable tradeoff and would opt to accept the uncertainty and risks for a drug that is likely, but not confirmed, to provide clinical benefit, potentially slowing the progression of their disease. Based upon this, I agree with the Office of Neuroscience that the benefit-risk balance for aducanumab is positive…

Closing Thoughts: Building Patients’ Trust & Digging into the Aduhelm Summary Basis of Approval

These statements by FDA officials surrounding the approval of Aduhelm, articulating how patient input set the context for their approval decision, should build trust with all patients.  This transparency surrounding one approval decision exemplifies that patient input matters.  This trust, with patients, is the trust that FDA should be most concerned about.

While we made some initial reflections of the merits of the Aduhelm approval in our June 8th blog post, we have not yet fully reviewed the SBA that was just released.  However, in the coming days we hope to do so and may provide our thoughts in subsequent posts.  In our work across many sponsors with many therapeutic modalities for many indications, we see how the many FDA review divisions apply the Accelerated Approval “reasonably likely to predict” standard required of a surrogate endpoint and, even more so, the application of the “substantial evidence of effectiveness” standard required of all new drug approvals, accelerated or traditional.  It is through this lens that we hope to evaluate the Aduhelm approval, to consider whether it “erodes” FDA’s standards.  However, as we collectively review these FDA review documents, we should all consider the patient input that FDA has shared with us and have that serve as our context for review as well.

Further Resources on Patient Input

For those of you who may now be intrigued about the evolution of patient input in drug development and regulatory decision-making, I would point you to some of our FDA Law Blog posts from over the years on this important topic: