By Ramy Sedhom
As a first year oncology fellow, I am eager to help patients suffering from cancer. And although some cancer drugs offer large, definite benefit, many drugs improve outcomes only marginally. The medical community and laypersons recognize the need to develop therapeutics that is clinically meaningful. The American Society of Clinical Oncology (ASCO) defines clinically meaningful by the benchmark of improvements in overall and progression-free survival. What has been most surprising thus far is how small some of these relative improvements can be. The ethics and trajectory of our current standard of practice concerns me.
A major area of concern is the practice of prescribing chemotherapeutic agents on a standard that veers from the definition established by our medical societies. Dreicer and colleagues evaluated approved drugs from randomized trials against the thresholds defined by oncologic societies, both in America and in Europe, and sought whether authors explicitly endorsed or stated a meaningful benefit for a particular cancer indication. Their results should raise concerns for anyone who takes seriously the premise of informed decision-making and patient communication.
End points of the clinical trials included in their study included changes in overall survival (OS), progression free survival (PFS), and other clinical endpoints between intervention and control arms that were deemed having meaningful benefit. They compared the author’s usage against the oncologic society guidelines.
After sorting through 559 articles in 2014 and 2015, they identified 53 claims of meaningful benefit in randomized studies. One was in the adjuvant setting for non-small cell lung cancer, three were neoadjuvant in either rectal or urothelial cancer, and 49 were in the advanced or metastatic setting. Of the claims in these populations, 25 described median PFS improvement, 14 described median OS improvement, and 10 used another measure of benefit.
Interestingly enough, the median improvement in OS thought to constitute a meaningful benefit was 2.2 months (range 0.33–5.7 months). The median improvement in PFS thought to be meaningful was 4.0 months (range 0.2–14.7 months). Among the 14 claims of meaningful benefit based on median OS, 6 (43% 95% CI 18–71%) met ASCO guidelines. Among 25 claims of meaningful benefit based on PFS, 17 (68% 95% CI 47–85%) met ASCO guidelines.
Not emphasized enough in our training is the premise that trials should be designed in a way that includes the minimum clinically important difference, taking into account the tolerability and toxicity of new treatments. The opinions of oncologists as to what constitutes a minimal important difference is likely to vary, which makes adoption of new therapeutics in standard practice difficult. For example, an increase in median survival by less than 1 month for patients with advanced-stage cancer may not be regarded by many as clinically important, unless the new agent is less toxic than standard treatment or is an added therapeutic when limited options exist.
The results described above should warrant caution for all treating and future oncologists, given that the phrase “meaningful benefit” may not always meet consensus guidelines. This has major implications for our patients, who are often in a very vulnerable time in their lives. It raises concerns about the integrity of shared decision making. Given that the ASCO thresholds are modest, we may be setting the bar too low for our patients. These “meaningful benefits” come with significant costs in quality of life and financial toxicity. They also distract from much needed conversations about death and dying. We have likely under-explored what patients themselves consider meaningfully beneficial. I wonder how often these should serve as externally valid metrics for professional societies.
The opinion of oncologists on what constitutes a meaningful benefit should be placed in the context of the opinion of patients and families about what is meaningful for them, and arguably, in the context of what can be reasonably afforded. In the first few months of fellowship, I am often left wondering, if patients new the marginal benefit of what they are offered, would they forego the toxicities of therapy to spend more time with their families? From the societal perspective, all new-targeted drugs are very expensive. It raises the question of whether the current health care economy can continue to be reasonably supported by Western economies. Only time will tell.
Author(s): Ramy Sedhom
Affiliations: Sidney Kimmel Comprehensive Cancer Center, The Johns Hopkins School of Medicine, Baltimore, Maryland, USA.
Competing interests: None.