This news release summarizes a review of 65 randomized controlled trials and observational studies that included some 10,000 people with anal cancer. The analysis itself is important: it exposes the potential for misclassification of disease severity leading to over-treatment of many patients with radiation and chemotherapy, unnecessary side effects, and little or no improvement in outcomes for patients. But the release doesn’t include any of the weaknesses that enter in when results are based on data-mined meta-analysis of previously published studies.
The headline touting a “breakthrough in the “treatment” of anal cancer and the almost complete absence of data leaves readers with misleading or confusing information about problems in diagnosing and treating a rare but deadly and difficult disease.
Anal cancer, as noted above, is rare, about 8,200 new cases in the U.S. each year, compared to the 140,000 Americans diagnosed with colon cancer annually. But it can be lethal and treatments and their aftermath are often life-altering. News releases that claim that “specialists” have made a “breakthrough” in the treatment of the disease would be of high interest to those with the cancer or those at risk for it. When the claim is neither supported nor true, the public is ill-served. The study seems to imply that patients with lymph node positivity, or LNP, can be treated with less aggressive treatments, which is not shown in the data, but only implied in the results. Anal cancer that has spread to lymph nodes is associated with a lower chance of survival and is generally treated with radiotherapy.
There is no actual “treatment breakthrough” in the data that needs a price tag. The implication is that less treatment is acceptable, which should cost less. The release could have made this point but we won’t penalize it for not doing so.
The release states that the study “uncovered a phenomenon suggesting rates of lymph node spread are being overestimated, potentially leading to overtreatment of patients with chemo radiotherapy.” There are no data offered in support of that statement.
It also makes the claim that “These findings will help us to better understand how anal cancer patients should be treated, ultimately improving survival rates and quality of life.” Again, there are no numbers provided showing how survival rates and quality of life can be improved with a change in treatment.
The release should have attempted to put the main findings in meaningful context for lay readers.
The release makes refers to the harm of over treatment and side effects, but never quantifies them. A quote from a patient refers to “joint pain and continence issues” suffered by some patients but it’s unclear which treatments were responsible.
The release offers quotes praising “high profile results” that are likely to “play a vital part in improving patient care,” but offers no information about exactly what those results are, why they are so “vital” and what the research protocol involved. It quotes a patient (unclear if she was in the data set that was analyzed) about her treatment experience in very general terms.
Further, there are serious limitations in meta-analyses that are never mentioned or discussed. The meta-analyisis included both randomized controlled trials and observational studies, some of which did not report overall survival. This would limit the reliability of the evidence.
No mongering; anal cancer is a serous disorder.
The release notes funding is from the Bowel Disease Research Foundation (BDRF).
The only alternative the meta-analysis looked at was the difference between LNP and LNN (negative) anal cancer. It showed an association between LNP and lower survival.
The release states that the study is likely to have an impact on treatment plans of future patients, but we never learn what that impact might precisely be, and what next steps might be taken by the oncology treatment community to use the findings and correct misleading statistics on diagnosis and treatment.
The release is fairly clear that the meta-analysis has uncovered/confirmed that statistical misinformation can result from even the best intentioned efforts to improve diagnostic criteria and staging. It would have helped for the release to connect the dots and demonstrate what those data are and what their impact would have on various patients.
Some of the language, including the headline and quotes, is overblown and unattributed. There is no “breakthrough in treatment” inherent in the study or the findings. And calling a finding a “unique phenomenon” without telling readers what, exactly, is the evidence for the phenomenon is unfortunate.