Josh Friedman , 2025-04-21 09:00:00
April 21, 2025
6 min read
Key takeaways:
- Common Sense Oncology has developed principles designed to improve randomized phase 3 trials and ensure meaningful results.
- Recommendations address use of certain endpoints, toxicity reporting and more.
A grassroots organization that formed with the goal of recalibrating cancer care around patient needs issued guidance intended to improve the design of randomized phase 3 trials.
The goal is to deliver results that are “meaningful,” not just statistically significant.
Common Sense Oncology — established in 2023 — is an international group of clinicians, patient advocates, researchers, editors and policymakers.
The group made nine recommendations intended to remove biases for randomized phase 3 clinical trials.
Suggestions include ensuring control treatments are always the standard of care, establishing OS or a validated surrogate as the primary endpoint, making health-related quality of life a significant endpoint, and reporting detailed censorship information.
“We have a number of large trials — often published in leading journals — that are subject to a lot of hidden bias, and that can lead to the approval of drugs that can have either no effect or a poor effect on patients,” Ian F. Tannock, MD, PhD, emeritus professor of medicine and medical biophysics at Princess Margaret Cancer Centre and University of Toronto, told Healio. “We want to criticize the treatments and the trials that have led to their approval so that we are not wasting a lot of resources on poor treatments and poor trials.
“We need trials to be better designed, better analyzed and better reported, and we are proposing checklists that we hope will achieve that,” he added. “We want to champion the trials and the treatments that really help patients.”
Common Sense Oncology
Common Sense Oncology formed on the foundation of three pillars: evidence generation, evidence interpretation and evidence communication, according to its website.
“Patients with cancer want to live longer lives and feel better,” organizing member Christopher Booth, MD, medical oncologist and director of the division of cancer care and epidemiology at Queen’s Cancer Research Institute, previously told Healio. “We think it’s important to advocate for treatments and systems that promote these priorities.”
Common Sense Oncology held its inaugural meeting in April 2023 and established four subgroups to improve care, one of which included accumulating data on cancer treatments, particularly randomized phase 3 trials.
“Everything is now run by the pharmaceutical industry — 90%-plus of phase 3 randomized, potentially practice-changing trials are run by pharma,” Tannock said. “The goal of pharma is to make a profit for their shareholders. Drug development has become a business, whereas when I was younger, these trials used to be done by academic groups.”
Healio previously reported that participation in industry-sponsored trials more than doubled over nearly 2 decades. In contrast, enrollment in federally sponsored trials has remained relatively stagnant.
“Along with that, there’s been a loosening of criteria, I think, for approving drugs,” Tannock said.
In response, Tannock and colleagues used their experience, the CONSORT checklist and published evidence to issue guidance they believe researchers should follow in the design and conduct of randomized phase 3 trials.
“We can’t just go on developing more and more drugs at these obscene prices that don’t do much,” Tannock said. “We can’t be spending such huge amounts on health care because if we do, there won’t be money left over to send your kids to school or mend the roads.”
Trial checklist
Common Sense Oncology’s issued the following nine recommendations:
- Control treatments should be the current standard of care;
- OS, or a validated surrogate, should be the primary endpoint;
- An “absolute” measure of benefit should be calculated, such as differences in OS;
- Health-related quality of life should at least be a secondary endpoint of all trials;
- Trials should include objective reporting of adverse events;
- Trials should be designed to determine “meaningful” differences in outcomes, not just statistically significant ones;
- Researchers should detail censoring data and conduct sensitivity analyses to determine if they impacted outcomes;
- Funding should be included so patients in control groups are offered experimental treatments if they are found to improve OS; and
- Trial reporting language should be easily understood.
Of those, three stood out to Tannock that “absolutely bias the results,” the first being censorship of dropouts.
Multiple trials report survival curves that note dropouts, but do not conduct sensitivity analyses regarding the effect they had on overall results.
“The reasons people drop out may be different in the experimental and control arms,” Tannock said. “You add a new drug in one arm of the trial; the new drug invariably adds toxicity. Patients who drop out on the experimental arm are often those with poor tolerance. They’re not doing well, and those patients are the ones who usually have poorer outcomes. You lose those patients, [and] you’re selecting out the better patients randomized to that arm. That makes the experimental arm look better.”
Conversely, the control arm may have patients drop out because they have better options.
“You still have patients who drop out because of toxicity, but it’s usually less,” Tannock said. “On the control arm, particularly in an open-label trial where people know what they’re getting, a number of patients go in because they want to get the experimental drug. Rightly or wrongly, they think it’s going to help them. If they don’t get it, they may say, ‘I don’t really want to continue here. Let me look for another treatment.’ The patients who drop out on the control arm are often those who are better informed and may have better outcomes. You’re losing the better prognosis patients in the control arm.”
“In some cases, it can completely negate what looks like a positive trial and make it a negative trial,” he added.
Tannock also emphasized too many trials do not make the control arm the standard of care for the trial.
“It’s occurred, for example, in a number of prostate cancer trials,” he said.
The third bias involves not funding crossover for individuals in control arms to take experimental therapies when shown to be beneficial.
“They recruit patients from many middle-income countries, such as Argentina, Russia, India, Brazil [and] China, and [some people] in those countries can’t afford to access that drug,” Tannock said. “You end up comparing the addition of the new drug earlier in the course of disease, without comparing it to the overall survival and quality of survival of patients who get what was the standard of therapy and the new drug later on.”
Additionally, Tannock highlighted the reliance on PFS and DFS as opposed to OS as a detriment — calling PFS as “a lousy endpoint” — and lack of or underreporting of quality-of-life measures.
“Most quality-of-life analyses have looked at the comparison between mean values of quality of life between patients,” he said. “I have no idea what mean quality of life means when you’ve got a group of patients, some of whom may be responding very well to treatment, others have progression of disease and everybody has a bit of toxicity. Mean quality of life is rather meaningless.
“We should make quality of life a responder criteria,” he added. “You pick an appropriate scale that has to be reported by the patients, and you ask the question: ‘What proportion of patients on each arm of the trial have a predefined improvement in a relevant patient-reported outcome?’ Trials have done that, but not very many. That’s so much easier to understand than this vague mean thing, which may be statistically relevant, but certainly isn’t clinically relevant.”
‘An uphill battle’
Members of Common Sense Oncology are in the process of submitting a paper that evaluated randomized cancer trials in 2023 to see how many met some — or all — of the checklist criteria.
They also are working on a paper evaluating quality of life in trials.
“We want to use those papers to try and encourage changes in the design, analysis and reporting of trials,” Tannock said. “The companies will only respond when they’re pushed to respond by journal editors and people in the registration agencies. That’s what we’ll be trying to do in the near future.”
Common Sense Oncology wants to improve standards for single-arm trials, as well.
The group wants clinicians to evaluate literature more thoroughly and is advocating for greater involvement of patients and patient advocates in the design of trials.
“It will be an uphill battle,” Tannock said. “Cancer drug development is a business. … I am optimistic that there will be improvements. I’m not naive enough to think that we’re going to solve all those problems that [the paper discusses], but I think we have to keep pushing.”
References:
For more information:
Ian F. Tannock, MD, PhD, can be reached at ian.tannock@uhn.ca.
Sources/Disclosures
Collapse
Disclosures:
Tannock reports no relevant financial disclosures. Please see the study for all other authors’ relevant financial disclosures.
