Loss of the Y chromosome (-Y) in bone marrow cells frequently occurs in elderly patients. It is thus unclear whether loss of the Y chromosome, when found in Philadelphiapositive cells, should be overlooked, or whether it represents a clonal evolution (a criterion of cytogenetic acceleration of the disease) or an additional cytogenetic abnormality (ACA), considered of little importance in patients treated with imatinib mesylate (IM). We have collected diagnostic characteristics, clinical, cytogenetic and molecular evolution data from 29 patients diagnosed with CML from 1991 to April 2007 in 6 participating centres, who presented a loss of the Y chromosome at diagnosis (n=20) or during evolution (n=9) and who were treated with imatinib in first (n=19) or second (n=10) line. They were compared to patients with a Philadelphia chromosome as sole anomaly, diagnosed at similar dates in the same centres, thus benefitting from the same management. Age and Sokal score at diagnosis were similar. In both -Y and control groups, two thirds of patients had received prior treatment (mostly interferon, hydroxyurea and aracytine). Median follow-ups were respectively 54 months (range: 7–146) and 58 months (range: 13–149), censored if therapy was changed for a second generation tyrosine-kinase inhibitor or bone marrow transplantation. The time to achieve complete cytogenetic remission (CCR) (22 months vs 7, p=0.02) and major molecular response (MMR) (not reached vs 30 months, p=0.0015) were significantly higher in -Y patients. Acceleration (8/29 vs 1/29, p=0.003), acutisation (4/29 vs 1/29, p=0.1) and death (3/29 vs 1/29) were more frequent in the -Y group, but owing to the rare occurrence of these events, statistical significance was clear for acceleration only. Interestingly, prognosis (time to CCR, MMR and occurrence of acceleration) was worse if loss of the Y chromosome was present in a subclone (suggesting a clonal evolution) rather than all mitoses (ACA) and also when it was observed at diagnosis rather than during the evolution of the disease. Consequently, survival was significantly reduced (55 months vs not reached, p=0.009) in those patients for whom -Y was present in a subclone at diagnosis. In conclusion, loss of the Y chromosome in Philadelphia-positive cells is an important finding since it affects the prognosis of CML patients, especially, but not exclusively, when present at diagnosis in a subclone. In this case, loss of the Y chromosome even predicts a decreased survival after imatinib treatment.

Disclosures: No relevant conflicts of interest to declare.

Author notes

Corresponding author