https://orcid.org/0000-0002-6262-1123
TO THE EDITOR:
In a recent issue of Blood Advances, Venditti et al1 report on recommendations generated by the European LeukemiaNet (ELN) on fitness assessment in adults with acute myeloid leukemia (AML). A steering committee conducted a literature review to generate the 21 statements then voted upon by the entire panel. This panel, comprised of physicians with expertise in AML management, geriatricians, and patient advocates, utilized a 2-round Delphi poll approach and a Likert scale for level of agreement. Consensus required ≥70% agreement (ratings of 4 or 5) or disagreement (ratings of 1 or 2) with an eventual 19 of 21 statements reaching consensus. These statements provide important recommendations on approaches to define fitness assessment, timing of such evaluations, and the influence of fitness on treatment decision-making.
Considering initial trials for older adults (age ≥60 years) in the 1980s led to debate about the utility of antileukemic therapy, if any, in these patients, the recommendations by Venditti et al highlight just how far our field has come on the topic of age.2 Specifically, statement 6, “Age, as a single parameter, should not be considered for the definition of fitness,” was met with 92% agreement. Furthermore, the panel recommends a comprehensive approach, at the very least, assessing functional capacity, comorbidities, and performance (statement 1, 96% level of agreement). Specifically, the panel recommends use of comprehensive geriatric assessment (CGA) before initiating therapy to guide decision-making and identify vulnerabilities that could be intervened upon. These recommendations align with prior expert guidance from the International Society of Geriatric Oncology and with the guideline recommendations for conducting CGA prior to systemic therapy for older adults with any cancer type.3,4
The authors, however, go further to provide a recommendation regarding timing of therapy initiation, emphasizing the opportunity to conduct both assessments of disease biology and CGA. Ninety-six percent of the panel agreed that time from diagnosis to treatment of up to 3 weeks does not seem to affect short- and long-term outcomes in most patients (statement 14). This recommendation provides further reassurance that there is not only time to conduct a CGA but time to use the information to engage in aggressive supportive care in an effort to optimize fitness. Finally, the ELN panel endorsed the importance of measuring fitness across the entirety of a patient’s treatment course, so-called “dynamic fitness assessment,” both to inform modulation of treatment intensity (statement 16, 100% agreement) and candidacy of subsequent lines of therapy, including allogeneic stem cell transplant (statement 4, 81% agreement). Dynamic fitness assessment is also highly relevant to patients as many older adults consider functional capacity after the initiation of therapy an important determining factor in making their treatment decision.5
However, the present report stops short in answering some of the practical questions facing oncologists on a day-to-day basis. Most importantly, what combination of scores on CGA measures should be used to personalize therapeutic decision-making in the current evolving treatment landscape, and how can this be implemented in a clinical setting both at diagnosis and longitudinally? There are available data to guide oncologists, for example, functional deficits and cognition may indicate vulnerability in the context of hypomethylating agent therapy,6 whereas cognitive impairment and objectively measured physical function are associated with poorer outcomes in the setting of intensive therapy.7,8 However, the authors correctly state that there is currently no standardized, validated fitness criteria used to guide treatment decision-making. Why does this gap in knowledge exist? To date, the inclusion of a CGA or any fitness assessment tool beyond performance status has not been systematically incorporated in sentinel therapeutic trials in AML. It is intuitive to oncologists, and the panel agrees (statement 2, 92% agreement) that characterization of fitness must be made in the context of specific treatments. Thus, given the landscape of AML treatment continues to rapidly evolve, validation of fitness measures can only be accomplished by integrating a standardized assessment tool such as a CGA into clinical trials in real-time. A retrospective approach will always leave us wondering whether our patients are “fit” for a new therapeutic option in clinical practice.
Broad incorporation of a CGA into therapeutic trials would set the stage for multiple additional future directions drawn from the recommendations by Venditti et al. For example, once relevant CGA measures are identified from prospective observational data, a randomized study of fitness-guided treatment allocation vs standard of care would identify the predictive value of such an approach. A recent phase 2, single-center trial provides compelling early evidence for this concept. Utilizing a CGA combined with disease biology to allocate an intensive or less intensive regimen, in comparison to a historical control, Bhatt et al9 demonstrated lower 90-day mortality rates (21.9% vs 40%; P < .05) among 73 older adults with AML. Results of this study need to be confirmed in a randomized trial but it highlights the opportunity to integrate fitness into the paradigm of precision medicine.9 Studies testing CGA-guided fitness optimization strategies have also shown promising results in patients with hematologic malignancies undergoing allogeneic stem cell transplant and chimeric antigen receptor T-cell therapy and should be evaluated in the frontline AML setting to enhance or maintain fitness during treatment.10,11 Finally, despite a 100% panel agreement in the importance of dynamic fitness assessment, only 3 studies have reported on the topic.12-14 Future studies assessing the utility of conducting CGA sequentially during treatment can enhance understanding of treatment tolerability and quality of life, while also providing targets for intervention to optimize survivorship and eligibility for subsequent therapies. Finally, no agreement was reached (statement 21, level of agreement 65%) regarding the early incorporation of palliative-care specialists into the care of patients with AML. The expertise of palliative-care specialists is a major untapped resource for symptom management and fitness optimization for patients with AML worth further investigation.
The present ELN recommendations join a recent call for validated fitness measures at diagnosis and longitudinally in patients with AML.15 A precision medicine approach to understand AML disease biology has become standard-of-care leading to multiple drug approvals and improvement in patient outcomes. Now is the time to apply the same level of attention, both in funding and scientific rigor, to testing and implementing proactive fitness assessment strategies to improve outcomes for older adults with AML.
Contribution: S.J.Y. conceptualized and wrote the manuscript; and H.D.K. conceptualized and edited the manuscript.
Conflict-of-interest disclosure: The authors declare no competing financial interests.
Correspondence: Samuel J. Yates, Section of Hematology/Oncology, Department of Medicine, University of Chicago, 5841 S Maryland Ave, MC7082, Chicago, IL 60637; email: samuel.yates@uchospitals.edu.
