Cancer trials aimed at 'surrogate' targets miss bigger mark — study
Find a random person on the street and ask them what a cancer drug should do. What are the odds they don’t say “it should help patients live healthier and longer”?
A confluence of forces have pushed clinical trials away from that seemingly central question, with developers and patient groups betting on the promise that aiming at more subtle measures can help bring needed therapies to market faster. But a new study from the British Medical Journal suggests the conventional wisdom may be right and that trend, among others, have led cancer drugs to hit the market based off studies that have a high risk of bias.
Researchers led by London School of Economics Professor Huseyin Naci tracked cancer drug approvals by the European Medical Association over three years from 2014 to 2016. Putting aside 13 trials not based on a randomized control model — which have their own set of issues but can be useful for rare diseases — and two trials without published data, they evaluated 39 trials that formed the basis of drug approvals in that period according to a standard criteria of bias and found: 19 (49%) were at high risk of bias and the ones that didn’t look at improving survival were more likely to be at risk. Only 10 trials primarily looked at improving survival.
Among them, 20% of trials that looked at overall survival had a high risk of bias, compared to 55% of the rest.
Other factors besides endpoint were at play in studies at risk for bias, including missing data and red flags in how data were measured, such as the trial lacking a blinded assessment.
But in their discussion of the results, the authors focused in part on the paucity of trials that looked head-on at whether a drug will help patients survive, warning that trials that look at other “surrogate” endpoints speed up development but might mislead patients and bring drugs that don’t work. Evaluating the study in a BMJ opinion column, Barbara Mintzes and Agnes Vitry took similar aim.
“Uncertainty and exaggeration of the evidence that supports approval of cancer drugs causes direct harm if patients risk severe or fatal adverse effects without likely benefit,” they wrote. “Or forgo more effective and safer treatments.”
The push toward these “surrogate” measures has come from both developers, eager to get their treatments to market, and patient groups hoping to secure access to new remedies as fast as possible. Targeting overall survival on average takes about an extra year, time many cancer patients don’t have.
A federal law, the 21st Century Cures Act, that went into effect in 2017 codified that push. Yet data on trials that avoid the central question of elongating and improving quality of life have been mixed.
A 2018 JAMA Internal Medicine review of 52 articles covering 38 trials on a total of 14,000 patients with 12 different cancers from 2000 through 2016 found no significant association between progression-free survival and health-related quality of life.
The study also comes as one of many critically evaluating existing scientific literature in the wake of the replication crisis that broke out most publicly in psychology and has rippled across the sciences. In 2012 Amgen’s head of research, Glenn Begley, published a paper in Nature revealing a 10-year company effort to reproduce 53 “landmark” studies in oncology. They got positive results in 6.