Of 48 cancer drugs approved between 2009-2013, 57% of uses showed no benefits and some benefits were ‘clinically meaningless’, says BMJ study
Most cancer drugs that have recently arrived on the market have come with little evidence that they boost the survival or wellbeing of patients, research reveals.
Forty-eight cancer drugs were approved by the European Medicines Agency between 2009 and 2013 for use as treatments in 68 different situations.
But the study, which looked at the clinical trials associated with the drugs, reveals that at the time the therapies became available there was no conclusive evidence that they improved survival in almost two-thirds of the situations for which they were approved.
In only 10% of the uses did the drugs improve quality of life. Overall 57% of uses showed no benefits for either survival or quality of life.
Cancer treatment: sorting the good news from the hype Read more
The team then looked to see whether the picture improved over time.
Huseyin Naci, assistant professor of health policy at the London School of Economics, and a co-author of the study, published in the British Medical Journal, said: “We wanted to see once [the drugs] were already on the market did they actually generate some evidence to show that they improved or extended life?”
The team found that after a follow-up period of between three to eight years, 49% of approved uses were linked to no clear sign of improvement in survival or quality of life. Where survival benefits were shown, the team said these were clinically meaningless in almost half of the cases.
“What we find very surprising is that not very many studies are looking at overall survival or quality of life as their [primary] objective,” said Naci. He said that instead most of the studies examined indirect measures, such as x-rays or laboratory tests that were assumed to offer clues as to a drug’s survival benefits.
The immune system: the great new hope of cancer treatment Read more
He added: “Unfortunately the expectation is that once the drugs are on the market then companies will be investing in [longer term] trials to then demonstrate overall survival benefits. But unfortunately these trials are not necessarily taken up and conducted.”
Naci said the findings did not mean patients should worry. “I think it is very important that no one is alarmed,” he said.
Carl Heneghan, professor of evidence-based medicine at University of Oxford, described the lack of drug improvement with regard to survival as disappointing, and called for a more rigorous approach to evaluating cancer drugs. “It is hard to understand why half the drugs were approved in the first place if they provide no clinically meaningful benefit,” he said.
But Winette van der Graaf, professor of personalised oncology at the Institute of Cancer Research, said that making decisions based on smaller studies looking at benefits other than overall survival, were important in making sure new treatments were swiftly made available to patients.
“In my area of research [on] rare cancers the level of evidence called for here is very hard to obtain, meaning that these patients would find it extremely difficult to gain access to new treatments,” she said, adding that large trials looking directly at survival could be expensive and lengthy.
“Ideally, studies should try also measuring early markers of treatment failure, so that health authorities can make well-balanced decisions.”
Emma Greenwood, Cancer Research UK’s director of policy, warned that the study did not necessarily reflect the situation in the UK where Nice (the National Institute for Health and Care Excellence) played an important role in deciding which drugs were available to patients.
“The study does highlight the importance of using real-world evidence from patients, on top of data from clinical trials, to build our understanding of how drugs work in a real-life setting. We’re already starting to see this happen through the cancer drugs fund in England, where patients can access promising new drugs while more data is collected on their effectiveness.”
The_Peyote_Coyote on October 5th, 2017 at 12:44 UTC »
It's funny, I read this with a neuro background and thought "almost one in two new cancer drugs work?! That's amazing!"
bobbi21 on October 5th, 2017 at 12:24 UTC »
Medical oncologist here.
Just reading the abstract and the article, it should be noted that this is measuring overall survival. This should be the most important marker but it's sometimes hard to get that early on for drugs, especially if people live for a long time with cancer. Prostate cancer for example, people routinely live 5-10 years with it. So if you want approval of a new drug and all you look at is overall survival, you're going to have to wait 10 years to finish the study and get approvals before the drug is available, even if in 100% of patients you see 0 sign of cancer anywhere in the body after a month of treatment.
Now there are many situations where improvements on imaging don't correlate with overall survival, which is why overall survival is still the gold standard, but having a bit of wiggle room in the definition for drugs that show tremendous imaging benefits is fair since you have a pretty high likelihood of having survival benefits in that case (although not 100% of course).
Having 57% of drugs being in that category does seem a bit high to me but we really have to look at these individually.
Also the median overall survival benefit does have to be taken with a grain of salt as well. Occasionally you see minimal benefit in the median survival but when you look at like the 5 year survival you see huge benefits. We get those situations when you have a drug that only works in a % of patients. You get like 10% of patients living 5 years longer on the drug but 90% of patients show no benefit at all, so the median (patient who had a response just in the middle) shows basically no benefit since they're part of that 90% that shows no benefit. Drug is still incredibly useful for those 10% so of course it should be approved (with attempted to figure out which 10% respond of course if possible) but when you do studies like this, it makes the numbers look pretty bad.
mrjobby on October 5th, 2017 at 11:35 UTC »
Haem/onc pharmacist here, based in UK, work for the NHS. Firstly - articles like this are welcomed, a huge amount of research and money is being poured into new anti-cancer therapies, and it is certainly an exciting (and challenging) time to work within cancer services.
That said, the benefits often quoted by pharma when launching a new drug often do not translate into the clinical setting, as stated by this article. New biological therapies have a market in the billions, and there are growing sentiments to get tough with how much benefit is gained from some newer agents. 'Real world outcomes' or 'big data' are now being gathered and analysed to try to curb the expenditure on cancer medicines and ensure best value for money in the health service.
I should point out that 'approved by the European Medicines Agency' is part of the licensing and commercial launch of new drugs - individual nations and cancer networks will evaluate the evidence individually before advocating the use of a new drug in whatever setting (NICE/SMC in the UK. Not all drugs approved by the EMA will be used by the NHS.