Occurrence of the same chromosome abnormalities in Ph+ and Ph- cells in chronic myeloid leukaemia. Evidence of a secondary origin of the Ph chromosome?

Imatinib mesylate (Gleevec; Novartis) has been successfully employed in the treatment of Philadelphia-positive chronic myeloid leukaemia (Ph+ CML) (Deininger et al, 2005). Although imatinib restores a polyclonal haemopoiesis in over 90% of patients, development of clonal chromosome abnormalities in Ph-negative cells (Ph) clonal evolution) occurs in about 15% of cases with a complete cytogenetic remission (CCR). This phenomenon has been very rarely observed in patients treated with interferon-a and/or cytotoxic agents. As the biological and clinical significance of these clones are still unclear, we investigated two cases of CML in chronic phase, not previously treated with genotoxic agents or drugs, in which imatinib therapy lead to the emergence of Ph) clones characterised by an abnormality initially disclosed in the Ph+ cells, a picture found only in two other cases in the literature (Gozzetti et al, 2003; Royer-Pokora et al, 2003). In patient 1, all metaphases at diagnosis carried both the Ph chromosome and deletion of the long arm of chromosome 5 [del(5q)] (Table I), which is associated either with myelodysplastic syndrome (MDS), characterised by a favourable prognosis, or with secondary acute leukaemia and an unfavourable prognosis. The apparently primary nature of the disorder and its benign clinical course, with complete haematological and cytogenetic responses, suggest that our patient suffered from the former type of disease; however, no dysplastic features were documented in any bone marrow cell line. During imatinib therapy, a progressive reduction of the Ph+ clone was achieved and emergence of 46,XY and 46,XY,del(5q) cells was observed. The patient is still alive and in CCR. In patient 2, four cytogenetic clones have been detected following the first cytogenetic investigation performed at diagnosis (Table I), with trisomy 8 occurring both in Ph+ and Ph) cells. After 5 months of treatment, the Ph), trisomy 8+ clone expanded and constituted the whole proliferating bone marrow cell population at the last cytogenetic control, carried out 2 months after therapy discontinuation when a resolution of leucopenia and thrombocytopenia was achieved. Because of an episode of heart failure and infective complications (otitis) imatinib was not re-started. A rapid worsening of haematological and general conditions was observed and the patient died 3 months later. Trisomy 8 is one of the most frequent chromosomal abnormalities in myeloid malignancies, occurring either as a sole, apparently primary, change or as a secondary alteration in about 10–20% of cytogeneticallyabnormal acute myeloid leukaemia, MDS and myeloproliferative disease. No features of dysplasia or acute leukaemia were seen in our patient, therefore we could not assign a clear significance to trisomy 8 in this case. Taken together, our findings indicate that the chromosomal aberrations revealed by imatinib, were markers of clones that existed prior to the Philadelphia chromosome, which actually implies a secondary origin of this genomic rearrangement. Furthermore, in patient 2, as the Ph was present alone or associated with trisomy 8 since the first cytogenetic control, it should have occurred in two different cells, with and without +8, suggesting a biclonal origin of the Ph chromosome and a possible derivation of all four bone marrow clones from leukaemia progenitor cells with normal karyotype. This model of leukaemic evolution agrees well with that reviewed by Feinberg et al (2006), according to which clonal chromosomal and/or molecular changes are secondary events arising from polyclonal epigenetically disrupted tumour-progenitor cells. Finally, as recent data in the literature prompt the importance of epigenetic phenomena in development, progression as well as therapy of CML (Issa et al, 2005; Roman-Gomez et al, 2005), we suggest that further investigation on the DNA methylation status of CML patients, especially in imatinibinduced CCRs, could provide more insights into the significance of Ph) abnormal clones and, in general, into the pathogenesis of CML