907    

Background:  The introduction of HAART has dramatically reduced the incidence of primary brain lymphoma and other AIDS-associated cancers. Survival of most HIV-infected persons has also been dramatically prolonged. For certain opportunistic infections or cancers, however, no survival benefits seem attributable to HAART. In this study, we evaluated the impact of HAART on survival of patients with AIDS-associated primary brain lymphoma.     

Methods:  Between January 1992 and June 2004, 118 HIV-infected patients with primary brain lymphoma were consecutively admitted at 11 HIV referral clinic centers throughout Italy and were included in this study. By univariate analysis, the Kaplan-Meier method was used to estimate survival probability, while differences among strata were statistically tested by means of the log-rank test. Hazard ratios (HR) and their 95% confidence intervals (CI) were used to compute crude and adjusted risks of death, according to Cox’s model.     

Results:  The estimated median survival time since primary brain lymphoma diagnosis did not significantly decreased between pre- and post-HAART period:  62 days (95% CI 39 to 91) in 1992–1996 and 57 days (95% CI 24 to 74) in 2000–2004 (p = 0.63). There was no difference in survival between primary brain lymphoma patients never treated with HAART (median 62 days) and patients treated with HAART before the diagnosis of primary brain lymphoma (median 60 days). Conversely, the survival time was significantly longer in patients who started HAART after primary brain lymphoma diagnosis (median 90 days) (p = 0.01). After adjustment for HAART, patients with histologically confirmed primary brain lymphoma showed an higher median survival time than those with clinical or instrumental primary brain lymphoma diagnosis (p < 0.001). By multivariate analysis, starting HAART after primary brain lymphoma diagnosis in naïve patients (HR 0.41; 95% CI 0.25 to 0.67) and treatment of primary brain lymphoma with radiotherapy or chemotherapy (HR 0.37; 95% CI 0.25 to 0.56) were associated with a statistically significant reduced risk of death.

Conclusions:  In contrast to the protective effect of HAART on primary brain lymphoma incidence, our data suggest that survival of this cancer remained poor even in the late HAART era and ART offered a survival advantage only to a restricted group of PBL patients, i.e., to naïve patients exposed to HAART after primary brain lymphoma diagnosis. Timely integration of HAART with conventional chemotherapy or radiotherapy represents the current optimal therapeutic approach for patients with AIDS-associated primary brain lymphoma, even though alternative therapeutic strategies are warranted.

Keywords: primary brain lymphoma; HAART; survival