PB1914 PURGING WITH CHLORAMBUCIL TO PREVENT INFUSION‐RELATED REACTIONS BEFORE OBINUTUZUMAB ADMINISTRATION: A MONOCENTRIC PILOT EXPERIENCE.

Background: The efficacy and safety of Obinutuzumab (G), in combination with chlorambucil (Clb‐G) in patients with treatment‐naïve chronic lymphocytic leukemia (CLL) and comorbidities was demonstrated in the CLL11 study by improved progression‐free survival and overall survival compared with Clb plus rituximab (R) or Clb alone, leading to the FDA approval for its use in combination with Clb in patients with previously untreated CLL in 2013. Nevertheless, the Clb dose used in CLL11 was lower than the other European experiences of its use in associations with Rituximab, and, moreover, has emerged that the infusion of G was burdened by a high incidence of infusion‐related reactions (IRRs). IRRs are among the worst complications of the frontline G‐Clb treatment to deal with in urgency, complicating collectively up to 87% of the infusions (67% grade 1‐2, 20% grade 3‐4); reducing the disease burden before the administration of G could be a solution to prevent the occurrence of these adverse events (AE). Aims: Given the previous experience of our group, we used, in the same setting of patients as the CLL11, an increased dose of Clb in association with G, looking at safety and response. Methods: We retrospectively analysed a pilot cohort of ten untreated CLL patients (median age 73 years) with comorbidities, treated with G‐Clb regimen. Chlorambucil 1 mg/Kg every 28 days was administered at a standard daily dose of 10 mg for 8 cycles, with 2 cycles of purging before the start of the G administration. Obinutuzumab was infused intravenously from the third cycle onwards according to the manufacturer. An independent‐samples t‐test was conducted to compare complete blood counts (CBCs) before and after the first 2 cycles of Clb. Results: At 2 months after the end of the therapy (median observation time of 10 months) the overall response rate (ORR) was 80%, with the partial responses (PR) accounting for the 60% and the complete responses (CR) for the remaining 20%. In the safety analysis, grade 3 or 4 treatment‐emergent AE occurred in 50% of patients; most common were neutropenia (30%), thrombocytopenia (20%), infections (20%) and anaemia (10%). Three patients experienced grade 1‐2 IRR (30%), but no grade 3‐4 IRRs or clinical TLS have been detected in our cohort. One patient was hospitalized for AE (spondylodiscitis). When comparing CBCs before and after the first 2 cycles of Clb, we observed a statistically significant reduction of peripheral white blood cells and lymphocytes from a mean of 58900/mmc to 20270/mmc and 50310/mmc to 15710/mmc respectively (p = 0.034), while no significant differences were detected in haemoglobin levels, neutrophil or platelets counts. Two patients had to permanently cease treatment for AEs (pulmonary toxicity and spondylodiscitis). Two patients needed a dose reduction of Clb, and 1 both G and Clb. Median dose of Clb was 520 mg for each patient (210‐640 mg) with median Clb of 78 mg for each cycle. Summary/Conclusion: Our real‐life pilot analysis confirmed that the use of a Clb purging regimen is a valid option to minimize the IRR, since no grade 3‐4 were observed. Haematological/extrahaematological toxicity and response rates were similar to already published experiences. These results are encouraging, but further studies with wider cohorts are needed to confirm them.