Aaron Gerds, MD
Deputy Director for Clinical Research, Cleveland Clinic Taussig Cancer Institute
Associate Professor, Cleveland Clinic Lerner College of Medicine of Case Western Reserve University
Here's an all-too-familiar occurrence at medical conferences: A researcher wraps up a presentation on an early-phase trial with, “in conclusion, kill-’em-all-amab demonstrated promising efficacy and was safe and well-tolerated.” While using the word “promising” to describe efficacy is akin to shooting the albatross, let’s sail that ship another day. Today, I’m more interested in the other part of the ubiquitous conclusion that claims a new therapy is “safe and tolerable.”
What is safe and tolerable? In truth, it is always a moving target and relies heavily on the individual patient and the situation. What is safe and tolerable for an adolescent or young adult may not be for someone in their eighth or ninth decade of life. Also, what is safe and tolerable differs depending on whether the therapy is used as a curative treatment or solely for palliation. The duration of therapy is another element of this equation. While nausea may be tolerable for a single week of chemotherapy administration, the prospect of taking a medication that causes nausea for years is unpalatable. Not only are these factors highly subjective, they are also intertwined. A young patient with acute myeloid leukemia will likely have a different view on a discrete course of curative, inpatient, induction chemotherapy than a geriatric patient who might opt for indefinite, palliative, single-agent
outpatient therapy.
The subjective nature of what is safe and tolerable for an individual patient calls for a conversation between that patient and his or her hematologist so they can review the anticipated side effects together. Of course, the way that side effects for a treatment are determined adds another element that may be colored by potential bias.
We all bring the baggage of bias to the interpretation of study results, and we all want the best possible outcomes. Companies sponsoring a study want to look good to investors and regulators. Medical writers want the sponsors who hired them to be happy with the abstract or manuscript. Investigators want the treatment that they tirelessly worked on to move forward in clinical development. Patients want the hope of the next big breakthrough. Sometimes these biases, unconscious or not, can drive the spin cycle of interpreting adverse events (AEs) and, in turn, influence what we think of as safe and tolerable.
I used to think that the long and distinguished list of potential side effects on a package insert was derived from a precise process conducted with the utmost of scientific rigor. When I began to work on clinical trials, however, I was surprised to find that the way AEs were recalled, recorded, and attributed to the study drug is a bit arbitrary. In fact, I can recall many instances in which the same AE for the same drug in the same patient were attributed differently by each investigator who saw that patient in clinic.
The next sequence of events is where things start to spin out of control and become even less clear. Although the collected AEs are put in a table for everyone to view, they may be sliced and diced in different ways. Tables that include all AEs observed during the study quickly get passed over in preference for tables of treatment-emergent AEs. These tables then get restricted to only grade 3 or 4 treatment-emergent AEs, then streamlined further to non-hematologic grade 3 or 4 treatment-emergent AEs, and then get reduced yet again to list non-gastrointestinal, non-hematologic grade 3 or 4 treatment-emergent AEs. Never mind that the grade 5 toxicities were never included in the table to begin with.
Pretty soon the only side effect being reported in the study results and registered by the medical conference attendees is the grade 2 nausea. Please don’t forget to take your pre-presentation antiemetic!
Several approaches can be taken to bring balance back to safety reporting. Simple patient-reported outcomes such as the Patient Global Impression of Change can help. The question “Are you better, worse, or the same?” can condense the overall value of a treatment into a single high-powered query. Likewise, the ratio of the number needed to treat to the number needed to harm (NNT/NNH) can give a sense of efficacy relative to toxicity. Patient-reported AEs (e.g., the Patient-Reported Outcomes version of the Common Terminology Criteria for Adverse Events, or PRO-CTCAE) can provide a patient-centered view of toxicity and overcome the superfluous steps between a patient experiencing an AE and it being recorded in a study’s electronic database. Moreover, electronic devices allow patients to log AEs in real time.
Given the subjective nature of what is considered “safe and tolerable,” how we present research results matters. Only you can stop the spin cycle. One strategy is to simply present the data – Just the facts, ma’am. Whether a new drug evaluated in a study demonstrates meaningful efficacy or safety is largely out of our control. What is in our control is how the results are presented. If a drug has excess toxicity in a study – or minimal efficacy, for that matter – it is important not to take it personally. You did not fail, the drug failed to meet expectations.
While there are many treatments with toxicity profiles that have stood the test of time and actually are safe and tolerable, there are countless up-and-coming therapies that tempt us to let the excitement of a new armament against disease overshadow the other edge of the sword: toxicity.
So, the next time you are at a medical meeting listening to presentations or walking though the poster hall and “safe and tolerable” emerges, take pause to make sure there isn’t also trouble.
Aaron Gerds, MD
Editor-in-Chief
The content of the Editor’s Corner is the opinion of the author and does not represent the official position of the American Society of Hematology unless so stated.
Have a comment about this editorial? Let us know what you think; we welcome your feedback. Email the editor your response, along with your full name and professional affiliation if you’d like us to consider publishing it, at [email protected].