Significant Survival Disparity Between Chronic Lymphocytic Leukemia (CLL) and Stage IV Small Lymphocytic Lymphoma (SLL) Patients: Analysis Using the Surveillance, Epidemiology and End Results (SEER) Database

Introduction: CLL and SLL are identical entities based on the World Health Organization classification of tumors due to similar tumor immunophenotype. Clinically, CLL is considered the leukemic phase of or stage IV SLL. However, population-based data comparing the natural history and outcome of these 2 diseases are lacking.

Methods: We reviewed the entire SEER database of patients with CLL (n=30,961) and stage IV SLL (n=5,151) diagnosed from 1985–2005 and compared their demographics, incidence, and survival. Age-adjusted incidence rate calculations were expressed per 100,000 persons in 5 year increments and obtained from SEER 9 database. Specific treatment data outside of surgery and radiation are not collected by SEER, so we did not include analysis according to treatment received.

Results: CLL had a slightly higher proportion of males than SLL (59.6 vs 58.2%, P =.05). The mean age at diagnosis was also higher in CLL (69.9 vs 67.6 years; P <.001). The age at diagnosis did not change over time for both diseases. Race distribution was similar in both diseases with predominance of whites (89.7%). Blacks were generally diagnosed at a younger age whether CLL (67.3 vs 70.0 yrs; P <.001) or SLL (64.3 vs 67.9 yrs; P <.001). The overall incidence rates were stable over time for CLL and SLL at 5.4 and 0.9, respectively. However, there was a decline in the incidence of CLL among blacks (4.7 in 1986–1990 vs 3.7 in 2001–2005, P <.001). The median overall survival (OS) for CLL and SLL were 77 and 63 months, respectively (P <.001). The median OS improved over time for SLL (P <.001) but not for CLL (P =.36). The incremental improvements in median OS between the 1985–1989 and 1996–2000 eras were 4 and 30 months for CLL and SLL, respectively. Multivariate analysis using separate datasets for CLL and SLL showed that increasing age at diagnosis (60–69 years, HR: 1.6, 1.7; 70–79 years, HR: 2.7, 2.8; ≥80 years, HR: 6.0, 5.1), male sex (HR: 1.3, 1.3), and black race (HR: 1.7, 1.5) were associated with worse OS (all P ≤.001). Better OS was associated with later diagnosis years (1993–1996, HR: 0.91, 0.76; 1997–2000, HR: 0.93, 0.66; 2001–2005, HR: 0.88, 0.55; all P ≤.01). Multivariate analysis of combined CLL and SLL datasets additionally showed that SLL histology was associated with a worse OS (HR: 1.3; P <.001). Similar results were obtained when we performed analyses using disease-specific survival as endpoint.

Conclusions: Contrary to current assumption, our population-based study suggests that patients with CLL have a significantly better outcome compared to stage IV SLL. This survival disparity is not due to lead time bias and is narrowing over time. Future studies should investigate whether this is due to different treatment patterns or biology.