# # # #
Today’s post is a work of fictional creative writing, because obviously no self-respecting, by-the-books, “public safety comes first” health care regulator would knowingly chose to put themselves into the ridiculous position herein described.
Call it a cautionary tale if you like.
Any aspects of this tale that sound remotely familiar to real world events are “reasonably likely” to be purely accidental and the work of the author’s warped ‘post-truth’ imagination. In reality, we all know that no health care regulator would be this foolish.
# # # #
A Crazy Mess Carol
Stave one – The ghost of Crazy Mess past
A biotech company named Biowin had been developing a novel therapy for medical use called “Arukiddingmab”. After successful Phase 1 testing (the drug was well tolerated and exhibited some interesting preliminary data), the company approached the health regulator (the Government funded entity whose sole task is to ensure the public safety regarding medical therapies) and asked if they could skip the prudent Phase 2 “learn & confirm” stage of clinical testing, and jump directly into the critical test of efficacy Phase 3 step of clinical evaluation – to determine if their Arukiddingmab treatment was effective. The company had limited clinical data, but the move would help bring the treatment to patients quicker and of course save the company significant resources and time.
The regulator agreed to this idea, promoting it as a potential example of their “access accelerator” initiative. Biowin pressed ahead with their Phase 3 clinical programme, initiating 2 large trials involving several thousand people. As the trials rolled on, however, it became apparent to the independent data monitoring committee that the treatment was unlikely to demonstrate any effect. The initial data suggested that Arukiddingmab was having no impact compared to a placebo treatment. The independent data monitoring committee recommended stopping the trials, which Biowin reluctantly agreed to do. They made a press statement and the patient community expressed their disappointment.
BUT, as the trials wound down, extra data was collected, and during the post-study analysis of that data it became apparent to the Biowin researchers that the treatment had had a small effect in one of the two studies. Excited by this result, the company put out a press release – and the patient community expressed their excitement. Biowin also went to the regulator, asking if their drug could be approved based on this single result.
The regulator looked at the data. They noted that the positive result only occurred in one of the two studies. They delayed making a decision by asking a panel of experts for their opinions on the situation. The panel debated the data, but unanimously agreed (10 out of 10 votes) that the drug should not be approved based on the available information. The regulator’s own biostatistical review of the data even found significant holes in the data.
At the same time, the patient community got very vocal and lobbied the regulator to approve Arukiddingmab, to make it available for use.
Stave two – The ghost of Crazy Mess present
Despite the recommendation of the expert scientific panel and seemingly ignoring the conflicting result from the second Phase 3 trial, the regulator has decided to approve Arukiddingmab (based on ‘surrogate endpoints’ rather than any actual clinical benefit) – stating that it is “reasonably likely that the effect of the drug will result in clinical benefits”. This statement has left the research committee utterly befuddled – “‘Reasonably likely’ is not how science works, especially when patient safety is in question” suggests one scientist. And the disconnect between surrogate endpoints and any clinical benefit looks like a lowering of the bar.
To confuse the situation further, the regulator has also decided to broaden the category of patients who can be prescribed the treatment, beyond the specifically defined subtype that Arukiddingmab was tested on in the two Phase 3 clinical trials.
Arukiddingmab was only tested in a specific subtype of patient in both of the Phase 3 trials, and the positive result was observed in only a subset of those particular patients in only one of the studies. Scientists have suggested that if the treatment is useful in a certain group of patients, that needs to be established before any approval. They suggest more data is required to know exactly which patients might benefit – and they worry that approval for a broader population could put patient safety at risk.
The solution, the regulator counters, is for the approval to be dependent on post-regulatory approval “Phase 4” data collection, that will determine if Arukiddingmab is actually having any effect. Questioning this unorthodox approach, however, many researchers express concern because Phase 4 data is tricky to collect as the studies are typically not as well controlled as more rigid Phase 2 or 3 clinical trials. In addition, given the broader range of patients being administered the drug, commentators suggests that it sounds a lot more like a rather desperate fishing trip as opposed to a strictly scientific investigation confirming a result found in a subset of patients. The regulator, however, is pressing on by giving Biowin 10 years to present data demonstrating that Arukindingmab is actually working. After lifting their jaws off the floor, observers can be heard breathlessly repeating the words “10 years?!?!?”.
In the wake of the approval and subsequent announcements, numerous members of the independent scientific committee are resigning their positions in disgust, stating that none of this is appropriate, scientific, or the actions of a responsible health regulator.
Meanwhile, as the dust settled on the celebration party, Biowin has announced that they will be making Arukiddingmab available to patients at a cost of $50,000 per year. Due to a lack of available data regarding long-term outcomes (greater than 12 months), patients may need to take the treatment for the rest of their lives. To lessen the financial strain of these costs, Biowin is making deals with financial support companies, giving less well off patient families the opportunity to borrow the funds required to cover the cost of treatment.
Patient support groups are naturally excited by the idea of a new treatment being available (even if the evidence that it works is thin at best), but they are already expressing some outrage at the excessive cost of treatment (independent cost assessment bodies have suggested a more appropriate cost would be just $5,000 per year).
Many doctors suggest that they won’t be prescribing Arukiddingmab to their patients due to the lack of supporting data. Their patients are scrambling to find practitioners who are prepared to prescribe it. And many industry observers worry that health insurance company might refuse to fund Arukiddingmab, which could put pressure on state health care providers.
Stave three – The ghost of Crazy Mess yet to come
Immediately there will be pressure on other regulatory bodies around the world to approve Arukiddingmab, based on… (the unwise decision of another regulator?). Regional patient groups will lobby their respective regulators in order to have accelerated access to the new treatment.
Six months after the regulators decision, the results of the two Phase 3 trials will be published in a high impact journal (behind a paywall, in a language inaccessible to patients). The research community will dig deep into the results and ask lots of questions – both to the investigators and the regulator.
Within 12 months of the approval, clinical trials around the world will report significant numbers of participants dropping out so that they can start to take Arukiddingmab. Several trials will be halted completely, while others will be slow in getting started, struggle to maintain momentum, and ultimately run out of funding. The pace of research will be reduced as the patient community jumps on Arukiddingmab in the sheer hope that it might actually have a small effect on their disease.
Other biotech companies will come forward to the regulator with Arukiddingmab-like treatments. Some of them will be asking to be given “access accelerator” status (because that is what Biowin got), while others will be asking to skip Phase 2 (because that is what Biowin got). Several biotech firms who previously had Arukindingmab-like treatments that failed in clinical trials will conduct post-hoc analysis on their old data and remarkably find small positive effects that they will take to the regulator as justification for approval (because that is what Biowin got). They will argue that post-regulatory approval “Phase 4” data can be collected (across vast subtypes of patients) to help determine if their drug actually works (because that is what… oh, you get the idea).
During this same period of time, the patient groups that tirelessly lobbied the regulator for access to Arukiddingmab, will start lobbying the regulator again due to the excessive cost of the treatment. Families will find that their insurance companies refuse to cover the treatment under their health care plans, and they will be forced to take on significant financial strain to cover the cost of a treatment in the hope that it might only have a small impact of the progression of the condition, if they are in a yet-to-be-determined subtype of the disease.
Initially, Biowin shareholders will toast bumper profits as their new ‘blockbuster’ brings in considerable revenue in the first few years, but the lowered bar for regulatory approval will quickly lead to additional competition from other Arukiddingmab-like treatments that could also possibly work, and this will lower subsequent Biowin profits. The company’s lawyers will be preparing the groundwork for potential litigation if the drug is ultimately found to have no effect (or worse). But the company’s executives will rest assured as the decision on situation won’t matter for another 5-10 years of bonuses (and arguments could be made for extending that particular time frame).
Stave four – How do you remove an egg out of an omelette?
The alarmist tale of Arukiddingmab portrays a massive gamble by a health regulator to appease the needs of the masses. It involves reckless and (more likely) potentially dangerous decisions – beyond just the safety of the patients. And it raises lots of topics for further discussion/debate:
- The desperation of the patient community should help to fuel research, but not determine how we interpret it.
- Compassionate access is a good thing, as long as it is done correctly and based on solid data.
- A lack of alignment between clinical outcomes and biological effect cannot become the norm.
- Marginal positive data is just as likely to be the anomaly as negative data.
- Post hoc analysis is only a hypothesis-generating exercise, that should lead to further investigation (not wishful regulatory approvals).
The story also provides a case study from which other more prudent and far-sighted health regulators can learn from. Actions have consequence – intended and unintended. Awareness and appreciation of this singular fact is important.
What solutions could there be for the regulator in this story? They could:
- Admit an error and scale back the approved indication – instead of broadening the population of people able to be administered the treatment, firstly prove that Arukiddingmab actually works in the sub type where the initial data suggested an effect. The biotech firm can then test other cohorts on their own time.
- In future, establish efficacy before approval – otherwise you are no better than the snake oil merchants. Do not shift risk onto the patient and healthcare system.
- Never use wishful words like “reasonably likely” – in science we don’t move from fact to fact, but from observation to observation. Admittedly there is a degree of grey involved. When it comes to public safety, however, the facts should be as clear as possible, the parameters well established, and the situation should be de-risked. Being the gate keeper is not just a title.
Enough of this. Creative writing is boring – time to get back to reality.
Needless to say, endgame scenarios of disease modification for neurodegenerative conditions keeps me awake at night (Click here for a previous SoPD post on this topic).
Back to the research in the next post.
All of the material on this website is licensed under a
Creative Commons Attribution 4.0 International License
You can do whatever you like with it!
The banner for today’s post was sourced from glasgowartclub.