TL;DR
This study demonstrates that socio-demographic factors, such as education level and rural residency, significantly impact treatment response and outcomes in patients with chronic myeloid leukemia (CML) treated with tyrosine kinase inhibitors (TKIs). The researchers developed novel predictive models by combining these socio-demographic factors with established clinical risk scores, offering a more comprehensive approach to patient stratification and personalized treatment strategies.
What Problem Does This Paper Address?
Chronic myeloid leukemia (CML) in its chronic phase (CML-CP) is a manageable condition with tyrosine kinase inhibitors (TKIs). However, a subset of patients experiences treatment failure, disease progression, or suboptimal responses. While clinical risk stratification scores like Sokal, Hasford, EUTOS, and ELTS exist to predict outcomes, they primarily focus on clinical and biological markers. Emerging evidence suggests that socio-demographic factors also play a crucial role, but their combined impact with clinical factors in predicting TKI treatment response and long-term outcomes in CML-CP patients hasn't been fully integrated into predictive models. This paper aims to bridge this gap by exploring the influence of socio-demographic variables alongside clinical factors and developing a more holistic predictive framework.
Methods
The researchers conducted a retrospective analysis of 1414 adult CML-CP patients diagnosed and treated with first-line TKIs (imatinib, nilotinib, or dasatinib) between January 2006 and December 2020. They collected comprehensive socio-demographic data (age, sex, education level, marital status, and household registration) and clinical information (including ELTS score, white blood cell count, hemoglobin levels, and comorbidities).
To identify independent predictors of treatment response and outcomes, they employed Cox regression models. Treatment responses were defined by cytogenetic and molecular milestones (CCyR, MMR, MR⁴, MR⁴.⁵), while outcomes included failure-free survival (FFS), progression-free survival (PFS), overall survival (OS), and CML-related survival (CML-OS).
Crucially, the study then developed novel predictive models by assigning scores based on the identified independent socio-demographic and clinical predictors. These models were used to stratify patients into different risk subgroups, allowing for an assessment of their predictive power.
Datasets
The study utilized a dataset of 1414 newly diagnosed adult CML-CP patients from Peking University People's Hospital. This dataset included detailed socio-demographic and clinical information collected retrospectively from January 2006 to December 2020. No new datasets were introduced.
Key Insights and Ideas
- Socio-demographics Matter: The study strongly emphasizes that factors beyond clinical markers significantly influence CML treatment outcomes. Lower educational attainment and rural residency were consistently associated with poorer treatment responses and survival.
- Integrated Prediction is Superior: The core innovation is the development of predictive models that combine socio-demographic and clinical variables. This integrated approach proved more effective in stratifying patients and predicting their responses and outcomes compared to relying on clinical factors alone.
- Specific Factors Identified: Key socio-demographic predictors for poorer outcomes included lower education levels, male sex, rural household registration, being single/divorced/widowed, and the presence of comorbidities. Clinical factors like high initial WBC count and low hemoglobin levels also remained significant predictors.
- Personalized Risk Stratification: The newly developed models allow for a more nuanced risk stratification, creating distinct subgroups with statistically significant differences in treatment responses and survival rates. This can aid clinicians in tailoring treatment intensity and follow-up strategies.
Key Results
- A total of 1414 CML-CP patients were analyzed.
- Multivariate analysis revealed that lower educational level and intermediate/high ELTS risk were significantly associated with lower rates of complete cytogenetic response (CCyR), major molecular response (MMR), and deeper molecular responses (MR⁴, MR⁴.⁵), as well as inferior FFS, PFS, OS, and CML-OS.
- Additional significant predictors for poorer responses included male sex, rural residency, high initial WBC count (≥120×10⁹/L), low hemoglobin (<115 g/L), and first-line imatinib treatment.
- For survival outcomes, being single, divorced, or widowed, rural residency, high initial WBC, low hemoglobin, and comorbidities were associated with worse FFS, PFS, OS, and/or CML-OS.
- The newly developed combined models successfully stratified patients into risk subgroups with significant differences in treatment responses (CCyR, MMR, MR⁴, MR⁴.⁵) and survival (FFS, PFS, OS, CML-OS) (all P<0.001).
Limitations and Future Work
The authors acknowledge several limitations:
- Retrospective Design: The study's retrospective nature means it cannot establish definitive causality.
- Heterogeneous Treatment: The inclusion of patients treated with both imatinib and second-generation TKIs as first-line therapy introduces variability.
- Single-Center Bias: Data from a single institution might limit the generalizability of findings, particularly regarding the "household registration" variable.
- Dynamic Factors: Socio-demographic information like education and marital status can change over time.
- Adherence and Monitoring: The study could not definitively confirm patient adherence to medication or regularity of monitoring.
Future work could involve prospective studies to validate these findings and further refine the predictive models. Investigating the mechanisms by which socio-demographic factors influence treatment adherence and biological responses would also be valuable. The authors suggest that clinicians should pay closer attention to patients' socio-demographic characteristics to identify at-risk individuals and provide targeted education and management to improve overall outcomes.
Summary updated