Addition of Lomustine in the Treatment of Elderly Patients with Acute Myeloblastic Leukemia Improves Survival with Acceptably Increased Infectious Toxicity. a Filo Study

The LAMSA2007 trial of the French collaborative FILO group (Clinicaltrial.gov ID, NCT00590837) demonstrated the improvement of treatment results brought by the addition of lomustine for elderly (60 years old or more) fit patients with acute myeloblastic leukemia. Of 448 patients included, 222 received lomustine. Lomustine (200 mg/m² D1) was added to a standard induction (8mg/m² D1-5 idarubicin, 100 mg/m² D1-7 cytarabine) then to a consolidation schedule (8mg/m² D1-3 idarubicin, 50 mg/m² x2 D1-5 cytarabine subcut. with or without lomustine 80 mg D1) and 6 mini-reinductions (idarubicin 8 mg/m² D1 IV, cytarabine 50 mg/m²/12h D1-D5 with or without lomustine 40 mg D1), followed by a 6-months oral maintenance with 6-mercaptopurine and methotrexate.

The risk of adding lomustine was to increase toxicities, mainly because of a prolonged period of neutropenia and therefore an increased risk of infectious deaths. Indeed, neutropenia lasted longer for patients receiving lomustine at all stages of therapy, from two more days during induction up to five more days during first reinduction. Surrogate markers of infection showed a higher risk for patients receiving lomustine, mainly during the first two courses of chemotherapy (induction and consolidation). During the induction phase (N=424), 358 patients presented with febrile events (50% in each group of treatment) and 265 infections were documented (135 in the lomustine group). Lower infection rates occurred during consolidation: out of 424 patients, 127 presented at least 1 fever episode over 38.5°C (79 in the lomustine group and 48 in patients without lomustine) and 103 were microbiologically documented (67 in patients receiving lomustine and 36 in patients without). The median number of days with at least two intravenous antibiotics was significantly higher for lomustine-receiving patients (median 19 days [range 0-98] versus 17 days [range 0-46] p=0.005 at induction; 3 days [range 0-49] versus 0 days [range 0-25] p<0.0001 at consolidation). The median duration of fever over 38.5°C was also significantly longer during consolidation (1 day [range 0-16] versus 0 [range 0-8], p = 0.0002) but not during induction (3 days in both groups, [range 0-44] for patients receiving lomustine and [0-24] for patients without lomustine). There was no statistical difference in the median days of treatment with anti-fungal drugs (14 days [range 0-98] vs. 13 days [range 0-56]) neither at induction nor during consolidation (0 days in both groups, ranges 0-50 and 0-65).

These infectious episodes were however related neither to severity nor mortality: 16 patients (9 receiving lomustine and 7 without) died because of infections during the induction phase and 8 patients deceased from infections during the consolidation period (5 in the lomustine group). Regarding severity, 25 patients suffered from a septic shock (15 in the lomustine group), but only 9 patients died from that complication (6 in the lomustine group). During the consolidation phase, 17 patients presented a septic shock (11 while on lomustine, but only 3 deceased, and 6 without lomustine, and only one was fatal).

As for microbiological documentation, the most frequent pathogens (Aspergillus, Candida, Staphylococcus, Enterobacteries, Pseudomonas) had a similar incidence in both arms. Interestingly, only Staphyloccus (25 vs 15 at induction and 27 vs 17 at consolidation) and Streptococcus (15 vs 6 at induction and 7 vs none at consolidation) seemed to be more frequent in the lomustine arm, yet without statistical significance. Infections from Enterobacter were surprisingly more frequent in the control arm at induction (36 vs 45, NS) yet more frequent in the lomustine arm at consolidation (31 vs 16, p=0.02).

During the 6 mini-reinductions, less infections occurred (18 during R1, 7 in R2, 12 in R3, 5 in R4, 6 in R5 and 8 in R6), with no treatment-related death.

In this trial, lomustine significantly improved the rate of complete remission after 1 cycle (75% vs 85% p= 0.01) as well as event-free survival (47% [39 - 54] vs 29% [22 - 37; p= 0.02]). We therefore keep recommending its association with standard chemotherapy in AML elderly patients, yet propose to abandon the consolidation phase because of increased toxicity.