Relapse of HHV8-positive multicentric Castleman disease following rituximab-based therapy in HIV-positive patients.

Successful treatment of HIV-associated multicentric Castleman disease (HIV+MCD) with rituximab-based approaches has dramatically improved survival and reduced the risk of human herpesvirus 8 (HHV8)-associated lymphoma. Longer term outcomes including relapse rates have not been described and are important to establish the potential role of maintenance therapy. A prospective cohort of 84 patients with biopsy-proven HIV+MCD were treated with risk-stratified rituximab-based therapy. Four patients (5%) died of refractory HIV+MCD and 80 achieved clinical remission. The median follow-up for the 80 patients was 6.9 years and their 5-year overall survival was 92% (95% confidence interval [CI], 85 to 99). Eighteen have relapsed (all histologically confirmed), including 5 with concomitant HHV8-associated lymphoma and MCD at relapse. The 5-year relapse-free survival is 82% (95% CI, 72 to 92). No clinical or laboratory findings that were present at MCD diagnosis predicted subsequent relapse, and the median time to first relapse was 30 months (maximum, 10 years). There were no significant differences in clinicopathological features at initial diagnosis and at relapse. All patients were successfully retreated at relapse with rituximab-based therapy. Only 1 patient died of relapsed MCD (at fifth relapse 9.4 years after initial diagnosis). Despite the use of rituximab, the risk of developing HHV8-associated lymphoma was significantly elevated in this cohort, with an incidence of 11.4/1000 person-years. The relatively low relapse rate and high salvage rates at relapse reduce the potential benefit of maintenance therapy; this should only be advocated in the context of a clinical trial.

No evidence for a polyomavirus association or aetiology in AIDS-associated nonsmall cell lung cancer.

AIDS-associated lung cancer has an increasing incidence, unaccounted for by smoking, and occurs consistently at a younger age than matched controls. We investigated whether known and new cancer-associated polyomaviruses, including the newly identified Merkel cell virus, may play a role in its etiopathogenesis. Although viruses target conserved pathways in cellular evolution, we are unable to suggest that the viruses studied here induce novel mechanisms of oncogenic dysregulation in AIDS-associated lung cancer.