Express authorization of medicines, a double-edged sword

Cristina Roure

The approval of Aducanumab by accelerated route, by the Food and Drug Administration (FDA), for the treatment of Alzheimer's has been carried out against the unanimous opinion of its Advisory Committee on this matter, which has raised enormous dust. It is the first drug that attacks a supposed cause of the disease instead of its symptoms and has generated great expectations in a therapeutic area in which the gap is notable.

Aducanumab reduces amyloid-beta protein plaques in the brain, but the clinical trials have not shown any clinical benefit in patients. Reduction in amyloid burden has also not been shown to predictably correlate with cognitive improvement. The FDA's decision has led to the resignation of three of the 11 members of its Advisory Committee, the request to President Biden – made by a United States senator – to relieve the head of the FDA, and numerous criticisms in scientific journals and the media Communication. If you want to know more about this controversial topic, watch the 6-minute video produced by The Conversation.

Expedited or conditional authorization

The fast-track authorization of the FDA in the United States or the conditional authorization by the European Medicines Agency (EMA) in Europe are extraordinary mechanisms or "shortcuts" created in the 1990s so that useful drugs for serious diseases without treatment options, such as AIDS or cancer, will reach the market more quickly. The shortcut consists of the use of what are known as surrogate variables that do not properly measure the clinical benefit (live longer or better), but rather are approximations, biological markers that seek to indirectly predict clinical benefit more quickly and easily. Uncertainty about the true benefit to patients – and risk, let's not forget – is the price you pay for speed. Since clinical trials based on surrogate variables are generators of hypotheses and not evidence, it is expected that in the years following the conditional approval, this hypothesis will be confirmed or refuted by adequate confirmatory clinical trials, with real clinical variables of clinical benefit. and not subrogated. Its result should lead to regular approval if the hypothesis is confirmed, or to the withdrawal of the drug from the market, if not.

In 2002, the accelerated approval of Imatinib based on a 98% complete response (surrogate endpoint) in patients with chronic myeloid leukaemia made it possible that a treatment that subsequently demonstrated a dramatic increase in survival (true clinical benefit) to reach patients faster. By the regular way, it would have taken much longer to be approved. However, success stories like this are the exception rather than the norm. The results are often not as dramatic and confirmatory trials are often not carried out or are carried out too late, so that ineffective or even harmful drugs for patients remain on the market for years, often at exorbitant prices.

Surrogate variables of clinical benefit

Surrogate variables, as with all shortcuts, are irresistible. Being fast, convenient and cheap, they started being used in the early stages of the clinical research process (phase 2 trials) as rapid preliminary evidence of the presence/absence of clinical activity, thus saving time and money in the subsequent stages of research. But the presence of activity does not necessarily imply real clinical benefit. That is where “the trap” lies, since, in reality, they do not in themselves add any value to the patient. In the words of Adam Cifu, it is “something that patients did not know was important until their doctor told them about it”(1).

However, for someone desperate who suffers from a serious and incurable disease, the existence of some positive data, even if it is a mirage in analysis or an examination, is hope to cling to and therefore the surrogate variables are the perfect allies of commercial exaggeration and the generation of disproportionate expectations in patients. In approving Aducanumab, the FDA argued that the Alzheimer's Patients Association said it was willing to accept some uncertainty about clinical benefit in exchange for quick access to such a promising therapy. Is this argument enough for a regulatory agency that must be based on evidence to approve drugs without proven efficacy or safety? Do patients have a single voice? What about future patients? What about the conflicts of interest of the Association, which received considerable funding from the manufacturer of Aducanumab? As the authors of an excellent article that I highly recommend titled "FDA Approval and the Ethics of Desperation"(2) state, agencies need to listen to all stakeholders, but they should not dictate their decisions. Understandably, desperate patients and their families believe that "something is better than nothing", but agencies cannot fall into the ethic of despair and abdicate their mission to ensure that the safety and clinical efficacy of medicines are proven by evidence that meets one's minimum quality standards before being approved since this is its reason for existence.

Recently, regulatory agencies are abusing accelerated approval based on surrogate variables, especially in some pathologies such as cancer where 70-80% of approvals are based on progression-free survival or response rate and not on evidence of real clinical benefit(1,3). It is difficult to quickly demonstrate clinical benefit in a field such as oncology, but lowering the level of requirement so much leads to low-value health care that can directly or indirectly harm patients(4,5).

The risks of abuse of authorizations based on surrogate variables

• Uncertainty and low-quality evidence. A review of the risk of bias in the pivotal trials used in the approval of anticancer drugs by the EMA between 2014 and 2016 shows that 50% of the clinical trials had a risk of bias and that this was greater if the primary objective of the trial it used surrogate variables and not overall survival(6). Huseyin Naci, from the London School of Economics, describes it to us in less than 4 minutes in the following video:

• The probability that the evidence will be generated once the drug is available on the market, plummets because the incentives for research, which is expensive and lengthy, no longer exist. Why choose a “hard” clinical variable such as overall survival in confirmatory clinical trials if they will be longer and likely to have negative results? The industry claims that using “hard” clinical variables in trials is too expensive, but the question is: too expensive for whom? This supposed saving occurs at the cost of patients and society paying the price, bearing the economic cost and the health cost of keeping expensive and harmful drugs on the market for years.

• Risk of toxicity. By definition, trials for accelerated approval focus on efficacy as the primary objective, and, as these are trials of shorter duration, we must remember that they trade speed for uncertainty and that the risk of rare serious adverse events unnoticed is high. Rosiglitazone was approved in 2001 for the treatment of diabetes based exclusively on clinical trials with surrogate variables of glycaemia and glycosylated haemoglobin. Years later, in 2007, it was discovered that it produced, among other effects, an increased risk of myocardial infarction, but its final withdrawal did not occur until 2010, causing avoidable damage.

• Regulatory inertia and use in clinical practice despite the publication of negative results in confirmatory trials. Expedited authorization is rarely revoked, even if confirmatory trials do not demonstrate the expected clinical benefit. Of 18 indications with accelerated approval by the FDA and negative results in confirmatory trials since the year 2000, six currently remain unchanged in the drug's datasheet. The NCCN guidelines, followed by oncologists around the world, maintained the second recommendation (the second category of recommendation) in half of the indications despite the negative results of the confirmatory trial(5).

• Opportunity cost. Despite the uncertainty about their clinical benefit, fast-track approved drugs command high prices and, as this type of approval is more and more frequent, the economic impact is increasing. In the case of Aducanumab, the annual cost of treating a patient is 56,000 dollars and it is estimated that 6 million people would be suitable to receive it in the United States. This vast amount of resources could be used for more value-oriented interventions, such as providing support to the overburdened caregivers of these patients.

Demonstrating improvements in survival or quality of life is difficult and time-consuming. Accepting a certain degree of residual and transitory uncertainty in exchange for faster access to promising drugs seems reasonable in serious diseases that lack alternatives. However, when the level of demand is so low, the uncertainty about the clinical benefit is excessive and remains unresolved for years, it is worth asking if the regulatory agencies are fulfilling their mission of guaranteeing the safety and clinical efficacy of medicines to society and if we are not all contributing to health care of little value.

Bibliography

1. Vinayak K Prasad. Malignant: How Bad Policy and Bad Evidence Harm People with Cancer. Jo
2. Largent EA, Peterson A, Lynch HF. FDA Drug Approval and the Ethics of Desperation. JAMA Intern Med. Published online October 25, 2021. doi:10.1001/jamainternmed.2021.6045 
3. Schuster Bruce C, Brhlikova P, Heath J, McGettigan P. The use of validated and nonvalidated surrogate endpoints in two European Medicines Agency expedited approval pathways: A cross-sectional study of products authorised 2011-2018. PLOS Med. 2019 Sep 10;16(9):e1002873. doi: 10.1371/journal.pmed.1002873. PMID: 31504034; PMCID: PMC6736244. 
4. Yudkin JS, Lipska KJ, Montori VM. The idolatry of the surrogate. BMJ. 2011 Dec 28;343:d7995. doi: 10.1136/bmj.d7995. PMID: 22205706. 
5. Lenzer J, Brownlee S. Should regulatory authorities approve drugs based on surrogate endpoints? BMJ 2021; 374 :n2059 doi:10.1136/bmj.n2059
6. Bishal Gyawali Regulatory and clinical consequences of negative confirmatory trials of accelerated approval cancer drugs: retrospective obser-vational study BMJ 2021;374:n1959http://dx.doi.org/10.1136/bmj.n1959.