Experts Disagree on Optimal Follow-up Time for Clinical Trial Participants
As cancers are detected at earlier stages, researchers need to plan to monitor study participants for much longer follow-up periods.
Increasingly, patients are surviving some cancers for a decade, or decades. Without prolonged follow-up, their late toxicities and experiences are lost to medical science.
Long-term follow-up of patients who participate in clinical cancer-prevention and cancer-treatment trials is an “important unmet need,” according to Jack Cuzick, PhD, FMedSci, FRCP(hon), head of the Centre for Cancer Prevention, director of the Wolfson Institute of Preventive Medicine, and John Snow Professor of Epidemiology at Queen Mary, University of London, in England.
“Provision for some sort of long-term follow-up needs to be made at the outset for trials, and this should be a requirement for their approval,” Dr. Cuzick wrote in an article published online in Annals of Oncology.1
“Ideally, all trial patients would be followed until death,” Dr. Cuzick said—though he is quick to acknowledge that follow-up durations will depend somewhat on the type of study in question.
The length of duration should depend on the trial's endpoints, argued medical oncologist Clifford Hudis, MD, chief of the Breast Medicine Service at Memorial Sloan Kettering Cancer Center in New York, NY.
“What endpoints we use to measure outcomes depends on the setting. In the adjuvant and curative settings, you may have to worry about long-term toxicities and secondary primary cancers that can be a consequence of curative therapy, or secondary cancers like leukemia, or other illnesses such as cardiac disease.”
The best duration for follw-up depends on “what questions you want to ask,” Dr. Hudis said.
For example, more than two-thirds of women who are diagnosed with breast cancer will survive long enough to die from another cause.1
As a consequence, the endpoints of disease-free survival and overall survival are no longer as critical as they once were.
“If the survival curves have separated sufficiently to generate a report at, say, year 6, it's very unlikely that this advantage will be undone at year 30,” Dr. Hudis cautioned. He called follow-up for follow-up's sake unjustifiable. “For toxicities, it's a different story,” Hudis said.
One example Dr. Cuzick mentioned was the increase in myocardial infarction and cardiac death rates after radiotherapy for breast cancer, which only emerged after 10 years of follow-up and was the major reason why the early radiotherapy trials did not show an impact on overall survival.
“Discovery of these late toxicities has led to a major change in radiotherapy planning and delivery, and the prospect from reduced cardiac toxicity and improved overall mortality using this modality is now high,” Cuzick said.
The rationale for long-term clinical trial follow-up is compelling, others have said.
“Long-term follow-up is absolutely essential, for a whole variety of reasons,” said Ian M. Thompson, Jr., MD, the Mays Family Foundation Distinguished University presidential chair, professor, and director of the Cancer Therapy and Research Center at the University of Texas Health Science Center at San Antonio, TX.
“Different things can happen over time. For example, an early signal suggesting benefit can wane over time and potential interventions can change that difference—especially with patients with long life expectancy.”
“The idea that we only care about a patient's health for 3 or 5 or 6 years, for the duration of a trial, that's very short-sighted,” Dr. Thompson added.
Long follow-up can be expensive and those costs are a hard sell, Dr. Thompson told Cancer Therapy Advisor. Clinical trials are very expensive, and prolonging the follow-up period can dramatically increase those costs.
“Industry doing a drug treatment trial simply cannot afford a study with a 20-year time horizon,” Dr. Thompson explained. “They're often not even interested in a follow-up beyond 3 or 5 years, because there's no return on investment for their drugs as the science rapidly changes.”
Long-term follow-up also involves logistical challenges. People move or die, and are lost to follow-up. Patient advocacy groups and social media strategies might help, Dr. Cuzick said—as can passive database follow-up, for many endpoints.
“A lot of cancer patients are in the Medicare system,” noted Dr. Thompson. “If they allow you to do so, then you can collect a huge amount of information passively at the only expense of paying Medicare for the data.”
Dr. Thompson pointed to a study led by Dawn Hershman, MD, MS, at the Columbia University Medical Center in New York, NY in which late effects associated with androgen deprivation therapy in prostate cancer were assessed by linking patient trial data with corresponding Medicare claims.2
“You could do that with private insurers, too,” Dr. Thompson said. “It's in their best interest, as well, as new therapies come along, to know the long-term impacts, because somebody's going to be paying for those.”
- Cuzick J. Statistical controversies in clinical research: long term follow up of clinical trials in cancer [published online ahead of print October 3, 2015]. Ann Oncol. doi: 10.1093/annonc/mdv392.
- Hershman DL, Unger JM, Wright JD, et al. Long-term consequences of intermittent and continuous androgen deprivation in older patients with metastatic prostate cancer. J Clin Oncol. 2015;33(suppl): Abstract 5008.