Secondary leukaemias are common, accounting for more than 40% of all patien
ts with acute myeloid leukaemia (AML) or myelodysplastic syndrome (MDS). A
clinical history of exposure to haematotoxins or radiation is helpful; howe
ver, many older patients are diagnosed with leukaemia with no antecedent hi
story of exposure. These patients' disease show a remarkably similar phenot
ype to classic therapy-related leukaemia. The specific cytogenetic abnormal
ities common to MDS, alkylating-agent-related AML and poor-prognosis AML (3
q-, -5, 5q-, -7, 7q-, +8, +9, 11q-, 12p-, -18, -19,20q-, +21, t(1;7), t(2;1
1)), probably reflect a common pathogenesis distinct from that of other de
novo AMLs, although the pathogenetic pathway has yet to be elucidated. Poss
ibly, tumour suppressor genes are implicated and genomic instability may be
a cause of multiple unbalanced chromosomal translocations or deletions. Ty
pically, these patients are either elderly or have a history of exposure to
alkylating agents or environmental exposure 5-7 years prior to diagnosis.
Another distinct entity affects the mixed lineage leukaemia (MLL) gene loca
ted on 11q23. These account for about 3% of patients with therapy-related l
eukaemia and have a short latency period from exposure, usually to an inhib
itor of topoisomerase II. Other therapy-related patients with t(8:21), invl
6 or t( 15;17) translocations should be treated as any other de novo AML wi
th similar cytogenetics. In summary, the major prognostic factor is related
to the pathogenetic mechanisms of the leukaemia. Cytogenetics and molecula
r features are a better predictor of outcome than patient history. Patients
should receive standard induction therapy. However, the long-term outcome
is relatively poor; the best results being obtained among patients undergoi
ng allogeneic transplantation.