Patient-reported outcomes for the myelodysplastic syndromes: a new MDS-specific measure of quality of life

To the editor:

In their recommendations for diagnosis and treatment of the myelodysplastic syndromes (MDS),¹  the European Leukemia Net point out the crucial role of quality of life (QoL) assessment, as affected patients are most often elderly and have a largely incurable disease. We agree that rigorous assessment of QoL is highly valuable²  and that measuring patient-reported outcomes is key to managing this disease.³  Preference for involvement in treatment decisions by MDS patients can also be associated with their QoL, and it is therefore crucial to rely on solid measurements⁴ ; however, as pointed out by the authors, so far, most QoL analyses for MDS have used generic or cancer-specific instruments.⁵  A disease-specific measure would allow for more sensitive assessment of the impact of MDS interventions.^6,7 

With the input of MDS patients, caregivers, providers, and QoL experts, we have created the Quality of Life in Myelodysplasia Scale version 1 (QUALMS-1),⁸  a new MDS-specific QoL measure developed in the era of disease-modifying agents. Individual and combined focus groups were conducted with members of the MDS community associated with the Dana-Farber Cancer Institute to identify MDS-relevant QoL domains and associated question topics. A draft 38-item scale was then constructed, using an algorithm that weighted the overall score by incorporating a greater number of questions from more highly ranked domains (Table 1).

Not unexpectedly, fatigue was ranked as the most important domain. None of the 12 domains was ranked significantly differently by patients/caregivers vs providers. The 2 groups ranked only 5 of 60 question topics differently (“too tired for routine tasks”: providers higher, P = .05; “limited support beyond the family”: providers higher; P = .02”; “organizing life around transfusion/doctor appointments”: providers higher, P = .03; “bruising”: patients/caregivers higher, P = .05; “anger over diagnosis”: providers higher, P = .03). The high level of agreement in the rankings of domains and question topics between MDS patients/caregivers and providers suggests that the QoL of MDS patients is compromised, and in a reliably measureable fashion.

We next piloted the QUALMS-1 with a new cohort of MDS patients at the Dana-Farber Cancer Institute (n = 20) including behavioral coding followed by a structured cognitive interview using Jobe’s framework.⁹  We interviewed patients individually and reassessed after every 5, refining the QUALMS-1 after each group. By patient 17, saturation was achieved: no questions were found to be confusing, upsetting, irrelevant, or intrusive. The mean time to complete the instrument was a minimally burdensome 7.5 minutes. We are currently field testing the QUALMS-1 in a larger, international cohort of patients (n = 180) to assess its validity, reliability, and responsiveness.

We hope that, once validated, the QUALMS-1 will impact the lives of MDS patients by facilitating the approval of disease-altering therapies, especially given the increasing focus by the FDA on patient-reported outcomes.¹⁰  It will also permit physicians to more specifically advise patients about the risks and benefits of new therapies in the context of their current QoL. Finally, the implementation of the QUALMS-1 will provide more precise measurement of QoL in these patients to detect actual differences between treatments being tested, if these truly exist.