U.S. Expert Consensus on Defining Intolerance to Tyrosine Kinase Inhibitor Treatment in Chronic Phase Chronic Myeloid Leukemia (CML)

Introduction: The incidence of CML in the United States (U.S) is 1-2 cases per 100,000 persons per year and the prevalence has risen, in part due to the effectiveness of BCR-ABL1 tyrosine kinase inhibitor (TKI) therapy at improving survival of patients with chronic phase (CP) CML. The goals of CML therapy have recently expanded to treatment-free remission (TFR), improving quality of life (QoL), and reducing long-term toxicities. There is a lack of a clear definition of treatment intolerance, making it difficult for clinicians to know when to recommend modifications to TKI therapy. We developed an expert consensus-based definition of treatment intolerance in CP-CML to help inform these decisions.

Methods: We used a double-blinded RAND/UCLA modified Delphi panel method, which involves 2 rounds of expert ratings with a live discussion in between, to define intolerance to CML treatment. The panel included 13 U.S. panelists (8 hematologist/oncologists, 1 hematologist, 1 pharmacist, 1 advanced practice provider, 1 PhD CML researcher, and 1 CML patient) who were first provided a literature review on current treatment of CML and strategies for managing TKI tolerability. Each panelist then independently rated 480 patient scenarios using 1-9 scales regarding a) the patient’s intolerance to therapy (1=definitely tolerant to 9=definitely intolerant) and b) the appropriateness of three different TKI management strategies (no change, hold or dose reduce, switch or discontinue) (1= highly inappropriate to 9= highly appropriate). Scenarios were created by considering factors including line and generation of therapy, intended treatment duration (i.e., TFR or chronic therapy), length of time on TKI, and interference of adverse events (AEs) on daily activities. In developing the ratings, panelists chose to evaluate AEs in aggregate, so no singular AE was included. Recurrence of AEs was conceptually discussed as frequency with which AEs in general interfered with daily activities within the past week (defined as rarely, sometimes, and often). An example scenario includes a patient taking 1^st line, 2^nd generation TKI therapy, whose goal is TFR, and who develops an AE during months 3-6 after initiating TKI therapy which has, in the past week, often interfered with daily activities. Severity of laboratory abnormalities were also considered. Areas of agreement and disagreement (≥2 panelist ratings of 1 to 3 and ≥2 panelist ratings of 7 to 9 within the same scenario) were discussed at a virtual meeting after which ratings were repeated. Second-round ratings formed the basis of expert consensus.

Results: Panelists agreed on majority of ratings. They disagreed on 1% of scenarios defining intolerance in CML and TKI management strategies in second round ratings, compared to 17% of scenarios in first round ratings. Across scenarios, the more frequently AEs interfered with QoL, the more likely panelists were to consider patients intolerant to treatment and TKI modifications to be appropriate. For example, in a patient taking 1^st line, 2^nd generation TKI therapy, whose goal is TFR, and who develops an AE during months 3-6 after initiating TKI therapy which has, in the past week, often interfered with daily activities, panelists agreed that following supportive measures, it is appropriate to consider this patient intolerant to TKI therapy and to hold or dose reduce TKI therapy. They also agreed that these measures may also be appropriate in the same patients whose AE sometimes interfered with daily activities. In contrast, in the same patients whose AE rarely interfered with daily activities, panelists agreed these patients can be considered tolerant of TKI therapy and that it is inappropriate to hold or dose reduce the TKI. Detailed ratings of intolerance and management recommendations will be presented.

Conclusions: A diverse panel of experts used the modified Delphi panel to define intolerance to TKI therapy. Expert agreement on both the definition of intolerance and appropriate responses was high and increased after discussion. Experts agreed that the extent of AE interference with daily activities is a key parameter in defining the degree of treatment intolerance and that there is a direct relationship between the extent of this interference and the appropriateness of holding, dose reducing, and switching or discontinuing TKI therapy. We hope these findings can help provide clarity and support clinician decision-making.