I don't think they purport to serve as or appoint anyone to serve as judge and jury for "sufficient" improvement. The concern is the approval of drugs which provide no improvement.
I don't think they purport to serve as or appoint anyone to serve as judge and jury for "sufficient" improvement. The concern is the approval of drugs which provide no improvement.
But one study found that 72 new anticancer drugs approved by the FDA between 2004 and 2014 prolonged survival for an average of 2.1 months. A 2017 report concluded that “most cancer drug approvals have not been shown to, or do not, improve clinically relevant end points,” including survival and quality of life. The authors worried that “the FDA may be approving many costly, toxic drugs that do not improve overall survival.”
Therefore, the implication is that drugs which prolonged survival by an average of 2.1 months (which was the FDA's track record for approvals between '04 and '14) are not sufficiently "clinically relevant" to justify approval (presumably the 2017 report quoted from includes these drugs).
Therefore the question remains: if small but incremental improvements (av 2.1m) aren't sufficient to warrant approval, what level of improvement would be sufficient? And who are the right people to draw this rather important line in the sand?
We need to talk about death and we need to talk about what people want when they die. A few people do want to struggle and gain every extra day they can no matter how painful that's going to be for them, but that's not what most people want.
In contrast, if you've got (say) three months to live, the prospect of an extra few months with a reasonable quality of life, if probably something many people would take.
I can't speak about FDA, but I can speak about NICE and the Cancer Drugs Fund.
NICE was set up to look at the cost effectiveness of medications and treatments. They were refusing to recommend new cancer drugs. This was politically tricky, so the Cancer Drugs Fund was set up to provide novel meds.
We spent over £1bn on it, and most of the drugs provided no meaningful benefit and caused harm. https://www.bmj.com/content/357/bmj.j2097
You'd need to explore the data behind the individual approvals to comment on statistical validity. But without going that deep, you can be assured that the FDA is generally pretty conservative, and employs very robust statistical methodology.
We don't know how many patients received the 72 drugs approved over a ten year period but we do know the potential customer base for the drugs is massive. Millions of people could have taken the drugs, you could easily get a statistically significant result.
It seems that maybe you are considering a smaller variance with all the drugs tightly clustered around the 2.1 month average where the central question becomes - what's the cutoff and who decides?
As to who decides where the line should be drawn in the US it is the patient and in the UK it is the government (NHS). Ideally the patient should give informed consent for a treatment regime, that means realistic assessments of likely outcomes sans hype.
I was able to access the paper in question, and of the 71 trials included in their analysis, two trials reported a negative overall survival (OS) outcome (-0.3m and -0.6m) and one trial reported exactly 0m overall survival gain. There were also 23 trials (I think - quick eyeballing) where the overall survival gain wasn't available or reported. Everything else had a positive overall survival benefit reported.
In most cases (2/3 cases with a negative or null survival benefit reported, and a majority of the cases where no survival benefit was reported) there was a positive improvement reported in progression-free survival (PFS).
So with the exception of a single case (Premetrexed, approved in 2008 for 1L NSCLC with 0m PFS benefit and -0.3 OS benefit - maybe it was better tolerated than the comparator??) there's seems to be a positive benefit to be found somewhere, small as it may have been in some cases.
And yes, it's essentially a data-set with a lot of clustering of OS benefit around a low single-digit number of months, with a few higher outliers.
EDIT: FWIW, it would also be very uncommon (as borne out by the dataset) for FDA to approve a drug which didn't offer a clear statistically-significant improvement in (usually) OS or (more rarely) PFS.
No, this is wrong. UK patients have the same access to private healthcare as Americans.
https://www.telegraph.co.uk/news/2017/04/03/terminally-boy-d...
The NHS decides what's necessary care and what's not, then they're supported by the courts. The illusion of "same access as the US" was shattered by Charlie Gard when reality collided with our ideals.
In the US: The doctors come up with a plan. The parents disagree. It is the parents who need to go to the courts to get a ruling.
Charlie Gard's case has absolutely nothing to do with cost. In the UK children are humans and humans have rights, and those rights are protected by the courts. Parents do not get to own children as possessions. The courts will focus on the child's best interests, because of the Paramountcy Principle which flows from Article 3(3) of the EU Treaty and Article 24 of the Charter of Fundamental Rights of the European Union, which flows from the UNCRC article 3(1). Charlie Gard's case is about someone who cannot make a choice for themself, and the courts focussing on that person's best interests, not on the interests of that person's relatives.
Charlie Gard had RRM2B MDS. This has fewer than 50 cases worldwide. There is no cure. There is no treatment. The discussion is whether to allow him to die with pain relief, or allow medical experimentation that has no hope of success but which will cause pain. Unsurprisingly the courts decided it was in his best interests to be allowed to die with pain relief.
http://blogs.bmj.com/medical-ethics/2017/07/07/never-let-an-...
You may wish to read the judgments, because almost everything written by the US press was wrong.
https://www.judiciary.uk/judgments/great-ormond-street-hospi...
https://www.judiciary.uk/judgments/great-ormond-street-hospi...