PS1270 HIGH INCIDENCE OF CYTOMEGALOVIRUS (CMV) REACTIVATION IN HEMATOLOGICAL PATIENTS TREATED WITH BENDAMUSTINE BASED CHEMOTERAPY REGIMENS

Background: Bendamustine (BENDA) alone or in combination with other drugs is being increasingly used in the treatment of chronic lymphocytic leukaemia (CLL), non‐Hodgkin lymphomas (NHL), and multiple myeloma (MM) due to its safety and efficacy. Variable infection rates both in BENDA monotherapy and in BENDA‐containing regimens have been documented. Limited data are available on cytomegalovirus (CMV) reactivations and infections in patients with hematologic malignancies undergoing BENDA therapy. Aims: We retrospectively analysed for CMV reactivation 120 consecutive patients who underwent therapy with BENDA alone (6%), combined with rituximab (81.5%) and combined with rituximab and dexamethasone. Our cohort included 95 B‐NHL (49 indolent e 46 aggressive B‐LNH), 5 T‐NHL, 18 CLL and 2 HL with a median age of 72 years (range 47–88) and a female to male ratio of 0.67 (female/male: 48/72). BENDA was used as first line therapy in 94 patients (78%) and as a salvage therapy in 26 patients (22%). Methods: CMV monitoring was performed on plasma samples by a real‐time TaqMan CMV‐DNA polymerase chain reaction (PCR) assay, according to the manufacturers’ recommendations (Roche). Quantitative CMV DNA was detected every 3 weeks before starting the next BENDA cycle, while CMV DNA monitoring was performed once a week in cases of CMV reactivation defined by plasma CMV DNA levels higher than 137 IU/ml. Results: CMV reactivation was detected in 29 patients (24%) among the 120 patients analysed for CMV‐DNA, with a median CMV peak viral load of 1835 UI/ml. Reactivation occurred after 55 days (range: 11–313 days) since the first BENDA administration in 27 B‐LNH (10 aggressive and 17 indolent B‐LNH), 1 T‐LNH and 1 CLL. Reactivation rates were different in each BENDA protocols: 1/7 (14.3%) in patients treated with BENDA, 21/98 (28.6%) in patients treated with R‐BENDA and 7/15 (46.7%) in patients treated with RD‐BENDA. All patients showed a deep lymphocytopenia at the time of CMV reactivation with a median of 345 lymphocytes/mm 3 . We furthermore analyzed potential risk factors for CMV reactivation during a BENDA containing treatment such as: age of the patients, diagnosis, type of BENDA regimen used, CMV‐specific IgG levels, nadir of total lymphocytes and total IgG during treatment, T CD4+ lymphocytes levels. The nadir of total lymphocytes and total IgG resulted significantly associated to CMV reactivation in a multivariate analysis (P‐value = 0.0001 and P‐value = 0.0008, respectively). Summary/Conclusion: Using these variables, we built a risk score for CMV reactivation which can help the physician to assess the risk of CMV reactivation in each patient and eventually start CMV‐DNA monitoring, prophylaxis or preemptive therapy. This single‐center study provides further evidence that BENDA regimens for lymphoproliferative diseases are associated with a high CMV reactivation rate and therefore a strict surveillance for CMV is required during their use.