SEATTLEA response to highly active antiretroviral therapy (HAART)
independently predicts the outcome of AIDS-related lymphoma, underscoring the
importance of giving HIV-positive patients this therapy whenever possible,
Christian Hoffmann, MD, of the University of Kiel, Germany, said at the 9th
Conference on Retroviruses and Opportunistic Infections (abstract 619W).
Previous studies have yielded conflicting findings about the
effect of HAART on the survival of patients with AIDS-related lymphoma, Dr.
Hoffman said. Recently, he and his colleagues found that HAART dramatically
improves the survival of patients with AIDS and primary central nervous system
lymphoma (AIDS 15:2119-2127, 2001). Nonetheless, few studies have looked at
associations between the survival of HIV-infected patients with lymphoma and
the actual receipt and effectiveness of both HAART and curative lymphoma
"Due to the poor prognosis of AIDS patients with lymphoma
before HAART, many patients and/or physicians possibly might have shown a
therapeutic nihilism even after 1996 when HAART became available. Therefore, in
studying these patients, one should consider whether they really received HAART
and chemotherapy or not, and, if yes, if the HAART and chemotherapy were
successful or not," he said.
Dr. Hoffmann and his colleagues analyzed clinical and
laboratory predictors of survival in a cohort of 221 patients diagnosed with
AIDS-related lymphoma between 1990 and 2001 and treated with chemotherapy or
radiation therapy. In 13% of patients, the lymphoma was Hodgkin’s disease. At
the time of diagnosis of lymphoma, the patients’ median CD4 count was
140/µL, and AIDS-defining events had occurred in 33%.
During follow-up, 32% of 104 evaluable patients received and
had a response to HAART (an increase in the CD4 cell count of more than 100/µL
and/or at least one viral load of less than 500 copies/mL within 2 years of
diagnosis of AIDS-related lymphoma), and 59% of patients in the entire cohort
who received polychemotherapy had a complete remission (an absence of
measurable disease for at least 4 weeks on CT scans).
In multivariate analysis, the risk of mortality was reduced
significantly in patients with a response to HAART (relative hazard, 0.30) and
in patients with a complete remission (vs partial remission or progression) of
their lymphoma (relative hazard, 0.26). The risk of mortality was increased in
patients with a prior AIDS-defining event (relative hazard, 1.93) and in those
with extranodal (vs nodal) manifestations of lymphoma (relative hazard, 2.92).
In contrast, factors historically associated with poor survival
of AIDS-related lymphoma (low CD4 cell count, elevated lactate dehydrogenase
level, advanced stage, B symptoms, and others) were not significantly
associated with mortality in multivariate analysis.