Changing patterns and comparative outcomes of “watch-and-wait” versus immediate therapy in CLL: A propensity-weighted SEER analysis, 2000–2020

Abstract

Abstract Background: Chronic lymphocytic leukemia (CLL) exhibits a broad clinical spectrum, ranging from indolent lymphocytosis to aggressive disease requiring prompt intervention. Trials from the 1990s demonstrated that early chemotherapy in asymptomatic patients conferred no survival benefit and increased toxicity. Although targeted agents have since transformed CLL care, the real-world adoption and long-term survival impact of a “watch-and-wait” strategy remain poorly defined. We evaluated national trends in initial management, identified demographic and socioeconomic predictors of observation, and assessed its association with overall survival. Methods: We conducted a retrospective cohort study using SEER 18 Registries (2000–2020), identifying adults (age ≥20) with CLL via ICD-O-3 codes. Cases with missing or nonnumeric survival times were excluded. Patients were classified as receiving “watch-and-wait” if both chemotherapy and radiation flags were absent; all others were categorized as immediate treatment. Annual proportions of patients managed with observation were calculated, and segmented regression identified temporal inflection points. To reduce confounding by indication, we estimated propensity scores using sociodemographic factors and year of diagnosis, then applied stabilized inverse-probability weights truncated at the 1st and 99th percentiles. Weighted Cox proportional hazards models with robust variance estimation were used to estimate hazard ratios (HRs) for death. Covariate balance was assessed using standardized mean differences. Sensitivity analyses included CLL-specific mortality modeling. Results: Among 96,836 patients, 83,702 (86.5%) were initially observed and 12,907 (13.5%) were treated within 12 months. Observation rates increased from 79.1% in 2000 to 87.3% in 2020. Segmented regression identified a significant inflection in early 2002 (estimate: 2002.24; SE 0.98), with adoption slopes of +1.11 percentage points/year during 2000–2002 (p=0.02) and +0.38 thereafter (p <0.01). Racial disparities narrowed: in 2000, observation rates ranged from 67.2% in non-Hispanic Black to 80.3% among non-Hispanic White patients; by 2020, all groups exceeded 87%. After weighting, covariates were well-balanced (standardized mean differences <0.10). Watch-and-wait was associated with a lower hazard of death compared to immediate treatment (HR 0.66; 95% CI 0.64–0.68; p <0.001). No significant interaction by treatment era (pre- vs. post-2014) was observed (p= 0.71). In time-varying models, the survival advantage increased over time (HR 0.69 at 1 month; HR 0.64 at 240 months). Sensitivity analyses yielded consistent results. Conclusions: In this nationally representative cohort, watch-and-wait has become the predominant initial strategy for CLL and is associated with improved survival. Demographic disparities in its application have narrowed over time. These findings reinforce the safety and efficacy of active surveillance in asymptomatic patients, even amid evolving therapies. Our results support current guideline recommendations and highlight the value of real-world data in shaping equitable, high-value care in CLL.