Impact of Socioeconomic Status & Race/Ethnicity On Survival in Diffuse Large B-Cell Lymphoma (DLBCL): A Population-Based Study.

Despite advances in treatment and a well-characterized prognostic index, significant heterogeneity remains in DLBCL survival. Preliminary data suggest a potential survival disparity based on race/ethnicity or socioeconomic status (SES). To evaluate the impact of these and other variables on survival we performed an analysis in the ethnically diverse population-based California Cancer Registry (CCR). We utilized Neighborhood SES, an index of 7 census measures of education, income, occupation & cost of living, based on the residential census-block group at diagnosis. Each census-block group comprises ∼1500 residents. Neighborhood SES has been shown to be significantly associated with survival after Follicular Lymphoma (JCO 27:3044, 2009).

All pts with DLBCL (ICD-O-3 codes 9680 & 9684) diagnosed from Jan 1988 to Dec 2007 and reported to CCR were included in the analysis, including n=16,892 diagnosed from 1988-2000, and n=11,916 from 2001-2007 (total study pop'n =28,808). HIV/AIDS pts were excluded, as were n=63 with Mediastinal LBCL & n=10 with primary effusion lymphoma. The mean age was 63 yrs, and the cohort was 53% male. Between time periods, there was a relative increase in Hispanic pts [15.4% (1988-2000) to 20.8% (2001-2007), p<0.001], and a 4% increase in advanced stage from 42% (1988-2000) to 46% (2001-2007) (p<0.001). Neighborhood SES was stratified into quintiles from lowest (SES-1) to highest (SES-5), the pt distribution was: SES-1, 14%; SES-2, 18%; SES-3, 21%; SES-4, 23%; and SES-5, 24%.

To evaluate the impact of prognostic factors (particularly diagnosis period, SES, and race/ethnicity) on overall survival (OS) & disease-specific survival (DSS) we used Cox proportional hazards regression to calculate hazard ratios (HR) for death with 95% CI's. Multivariate regression models included variables significant at p<0.15 in univariate models or with a priori hypotheses for inclusion. Results are presented by stage at diagnosis [Localized/Regional (LocReg) vs. Advanced (ADV)].

There was a significant improvement in OS in patients diagnosed after 2001 for both LocReg (HR 0.87, 95%CI 0.82-0.91, p<0.001) and ADV stage (HR 0.69, 95%CI 0.66-0.72, p<0.001), which correlates with the introduction of rituximab into therapy for DLBCL. As expected, age >60 years was associated with a significantly worse OS for LocReg (HR 3.06, 95%CI 2.90-3.24) and ADV stage (2.02, 95%CI 1.93-2.12). Females also had significantly better OS compared with males (Loc-Reg - HR 0.90, 95%CI 0.86-0.94; ADV - HR 0.89, 95%CI 0.85-0.93). There was no significant impact of race/ethnicity on survival with the exception of non-Hispanic Asian/Pacific Islanders (NH A/PI) with ADV stage, for whom OS was significantly inferior compared with whites (HR 1.18, 95%CI 1.09-1.27, p<0.001).

Compared with the highest quintile (SES-5), there was a significant effect of lower neighborhood SES on OS and DSS (see Table).

There has been a significant improvement in survival after DLBCL since 2001, but patients in the lowest SES-1 quintile have a 34% higher risk of death from any cause and 20% higher risk for death from lymphoma than those in the highest SES-5. In this model, race/ethnicity did not have a significant impact on survival with the exception of NH A/PI with ADV stage. Studies to understand and address these socioeconomic disparities are urgently required in order to extend the improvements in DLBCL survival more effectively.

Foran:Genentech: Honoraria, Research Funding.