The more people know about the scientific uncertainties around a recently approved cancer drug, the less interest they may have in trying it, a new study suggested.

The study, including 6000 US adults, found that learning about important evidence gaps for a new cancer drug — unclear survival benefits or long-term risks, for instance — dimmed participants’ interest and trust in the treatment.

The research looked only at hypotheticals, asking participants to imagine they were a patient with advanced cancer deciding whether to try a newly approved drug. But experts said the findings, published in JAMA Internal Medicine, suggest that frank discussions about treatment uncertainties are likely to be meaningful to patients.

That will be increasingly relevant as more cancer therapies come to the market through accelerated approvals, according to lead study author Avi Cherla, PhD, research fellow at Harvard Medical School in Boston.

That FDA pathway allows drugs to be approved on the basis of surrogate endpoints rather than evidence of survival benefits (though postapproval confirmatory trials are required).

While it’s important to expedite new treatments to patients with serious diseases, Cherla said, that can also create a “tension” where therapies enter practice with key questions still unanswered.

Meanwhile, that reality may be obscured: In a prior study, Cherla and his colleagues found that fewer than half of important clinical trial uncertainties for new cancer drugs were included in the prescribing information for clinicians. That prompted the new work, Cherla said.

The study included 6272 US adults who were not required to have any personal experience with cancer, though nearly two thirds did. In most cases, they had an immediate family member or close friend who’d been diagnosed with cancer; less than 10% had ever been diagnosed themselves.

First, Cherla’s team came up with plain-language descriptions of the main scientific uncertainties around a hypothetical drug called Zenova — modeled after adagrasib (Krazati), which won accelerated approval in 2022 for certain patients with advanced non-small cell lung cancer (NSCLC).

Some of those uncertainties were that Zenova was approved on the basis of a single-arm trial, meaning it wasn’t shown to be better than any other treatment; the trial was short, so the long-term benefits and risks are unknown; and the trial showed only that the drug shrank patients’ tumors, not that it extended their lives or helped them feel better.

Study participants first read a vignette about a 38-year-old woman named Alex who was being treated for NSCLC. The scenario explained that less than 10% of people in her situation live for more than 5 years. Alex’s doctor asked whether she was interested in trying Zenova, which was covered by her insurance.

Initially, participants were asked to imagine they were Alex and rate how likely they would be to take the drug. Then they were split into two groups: In study 1, participants were randomly assigned to learn about one Zenova uncertainty; in study 2, they were randomly assigned to learn about one, two, three, or four uncertainties, or to a control group presented with none of that information.

In study 1, people were less likely to say yes to Zenova after hearing about one uncertainty. Across the group, willingness declined by a mean of 10 percentage points, from 67% to 57%.

In study 2, hypothetical decisions to take Zenova decreased with each additional uncertainty presented, reaching a 19-percentage-point reduction with four uncertainties compared with the control group. Participants’ ratings of the drug’s perceived effectiveness also dropped after they learned of multiple uncertainties — as did their trust in the information and in the FDA itself.

The findings highlight challenges that clinicians and patients face in making informed decisions amid the wave of new cancer therapy approvals, said Paul Han, MD, MPH, professor of bioethics and decision sciences at Geisinger College of Health Sciences in Scranton, Pennsylvania. He previously worked as a senior scientist at the National Cancer Institute and wasn’t involved in the new study.

No one, however, is suggesting that important information be withheld from patients. Han, as well as an editorial accompanying the study, expressed concern that some clinicians may be reluctant to discuss uncertainty around treatments — when, in fact, it’s “essential to disclose.”

In addition, people who actually have cancer may respond differently to such ambiguities than people in this study did, according to Han.

“In my experience taking care of patients, and also in the research that I’ve done in the past, people with cancer look at things very differently,” he said. “And there’s evidence even that they respond to uncertainty about a drug differently.”

In particular, Han said, people who have advanced cancer and have run out of treatment options sometimes “like” uncertainty because questions around the effects of a treatment can provide a “source of hope.”

According to Han, the medical field needs to find ways to help patients manage uncertainty — so that they’re not “simply scared off from making a choice that might be in their interest.” One potential approach, he noted, is normalization, which involves assuring patients that scientific uncertainty is not an anomaly, but an expected part of the research process.

“If this [study] is good evidence that being more extensive and precise about disclosing uncertainty is going to have a dissuasive effect on patients, then the reaction is not to talk less about uncertainty,” Han said. “It’s to come up with ways to help patients manage that uncertainty.”

The study was funded by Arnold Ventures and the London School of Economics and Political Science Phelan US Centre. Cherla and Han had no relevant financial disclosures.