AIDS-Related Vs. Non-AIDS-Related Diffuse Large B-Cell Lymphoma. Analysis of the County Hospital AIDS Malignancy Project (CHAMP)

Introduction: Stroger Hospital of Cook County (CCH) and the Ruth M. Rothstein CORE Center (CC) are the largest health providers for HIV+ patients (pts) in Chicago and among the largest in the United States. Together, CCH and CC treat over 5,500 HIV+ individuals per year and 60 newly diagnosed HIV-associated cancers yearly. In addition, CCH is the largest safety net hospital in Illinois; in the calendar year 2010, it had over 32,000 outpatient hematology/oncology clinic visits. The County Hospital AIDS malignancy project (CHAMP) is a prospective database of all HIV-associated hematological malignancies from 1995 to present. Here we analyzed the demographic, histological, germinal center status, outcome data, and HIV characteristics of pts with AIDS-related diffuse large B-cell lymphoma (AR-DLBCL). As a control, we compared these data to DLBCL pts not infected with HIV, intra-institutionally instead of national historical controls to better understand the AR-DLBCL population of the inner-city, a mostly minority, and one of the largest growing HIV populations. 

Methods: We screened the CCH cancer registry from 2001 to 2014, and 240 non-HIV pts were identified to have DLBCL. We searched the CHAMP database for cases of DLBCL from the same period. From 196 lymphomas, 51 were identified. Only pts who had a confirmed pathology report as DLBCL were included in this study. We subsequently identified the HIV characteristics, overall survival (OS), and patient demographics for all pts. In addition to OS data in both cohorts, we compared outcomes based on chemotherapy, stage, germinal center (defined by the HANS algorithm), and IPI status.

Statistics: Non-parametric Fisher's exact test was used to examine the difference in proportion of pts in both the HIV and non-HIV arm. Survival data were analyzed using Kaplan-Meier analysis and Cox Proportional Hazards model.

Results: The M:F ratio was 7.5:1 in AR-DLBCL and 1.9:1 in the non-HIV cohort (P<0.001). The median CD4+ count for the HIV population was 107 cells/mm³. Patients with germinal center DLBCL had a median CD4+ count of 196 cells/mm³ vs. non-germinal of 87 cells/mm³. The racial composition of the 2 cohorts showed a significantly higher percentage of African-Americans (AA) in the HIV cohort, 51% vs. 35% in the non-HIV, (p<0.04) and no difference in Caucasian or Hispanic populations. The AR-DLBCL population was younger. Only 4% of the HIV pts were 60 years and older compared to 32% in the non-HIV cohort. The HIV positive pts tended to present with more advanced disease. Seventy two % vs. 53% in the non-HIV cohort presented with stage III/IV disease (p <0.01). Only 18% vs. 38% in the non-HIV cohort presented with stage I/II disease (p <0.006). Of pts tested, there were no differences in GC (68 (n=15) vs. 66% (n=77) or NGC (32 (n=7) vs. 34% (n=39)) between HIV and non-HIV cohorts. The OS of the AR-DLBCL at 5 years was 54% vs. 77% in the non-HIV cohort (p<0.003). However, since 2009, pts with AR-DLBCL have been treated with daEPOCHR (Sparano et al, Blood 2010) at CCH, and since then no difference in OS compared to the non-HIV cohort has been identified. At 5 year follow up, the OS of AR-DLBCL was 81% treated with daEPOCHR (n=11) vs. 77% (n=108) in the non-HIV population (p 0.081). Comparing survival between each cohort by stage there was a trend to poorer outcomes in the AR-DLBCL for stage III/IV disease, 58% vs. 75% 5-year survival (p 0.064). Similarly, there was little difference in comparing each cohort by IPI score due to the small number of pts in each IPI category in the HIV cohort.

Conclusions: In assessing the cases of AR-DLBCL, it appears to be a disease of AA (51% vs. 35%) men (88% vs. 53%) compared to the non-HIV population at the same institution. But when compared to the general population of Chicago, where AA constitute only 33% of the population, the disparity is more pronounced. Patients are also younger, 4% vs. 32% over 60 years of age. In addition, the AR-DLBCL pts present with more advanced disease (72% vs. 53%). As a whole, the AR-DLBCL cohort had a poorer OS survival than the non-HIV population, at 5 years, 54% vs. 77%. Importantly, pts with AR-DLBCL had the same OS as the non-HIV cohort, 81% vs. 77%, with a 5 year follow up when treated with daEPOCHR despite HIV pts presenting with a higher stage. Thus, education and screening in the AA community is needed, as this is an AA male disease in the inner city. It is also clear that treatment with daEPOCHR should be the therapy of choice when treating AR-DLBCL.