Clinicians can use simple multiples of median overall survival time in clinical trials to estimate survival percentiles for metastatic colorectal cancer, researchers reported online in Annals of Oncology.
"Previous work has suggested that clinicians often overestimate prognosis, and we hope that this study and others will provide more objective estimates on which to base discussions with patients and their families," said Dr. Matt Williams of Imperial College Healthcare NHS Trust, London, and his associates.
Prior studies of incurable breast and lung cancer have shown that multiples of median overall survival (OS) accurately reflected time points when 90%, 75%, 25%, and 10% of patients were alive, the researchers said. To assess this approach for estimated survival in metastatic colorectal cancer, they analyzed 46 MEDLINE-indexed, randomized, phase III clinical trials of chemotherapeutic agents published between 2000 and 2011 (Ann. Onc. 2014 [doi:10.1093/annonc/mdu372]).
The median OS for the 29,011 patients in the trials was 16.5 months, the researchers said. The investigators then estimated 90%, 75%, 25% and 10% survival times by multiplying the median OS by 0.25, 0.5, 2, and 3, they explained.
The estimated survival time fell within 75%-133% of actual survival time 88% of the time, the researchers reported, adding that the lowest concordance was for the earliest or 90% survival level. The low rate of agreement at the 90% survival time point may be due in part to the narrow time window at this survival point, the investigators speculated. "Our definition of agreement allows for approximately 5 weeks’ difference between the actual and predicted values, which is a narrow time window and thus more susceptible to random variation," they said.
"Taken with previous studies, these findings are likely to be valid across a large range of patients," Dr. Williams and his associates commented. "We would encourage clinicians to think of prognosis as a trajectory and to consider quoting survival ranges instead of point estimates, in discussions with patients."
The findings mirror prior work in breast and lung cancer, even though these cancers have very different survival times, the investigators noted. For this reason, the correlations are likely to hold true even if median OS for metastatic colorectal cancer increases in future trials, they said.
The authors said they received no funding for the work and had no conflicts of interest.