A 67-year-old man was admitted to hospital for investigation of recent weakness, dyspnoea and vomiting. A blood count showed: haemoglobin: 65 g/l, white cell count: 24AE1 · 10/l (41% blasts and 20% granulocytes), platelet count: 42 · 10/l. Coagulation tests were consistent with a moderate activation of coagulation (prothrombin time: 16 s, detectable d-dimers), without overt disseminated intravascular coagulation. The bone marrow aspirate was hypercellular with rare megakaryocytes. The major population consisted of blast cells (71%) 20–25 lm in diameter, with a large, regular nucleus displaying one or several prominent nucleoli. The cytoplasm was hypergranular, occasionally showing multiple Auer rods. Maturing granulocytic cells (14% of nucleated cells) were dysplastic: hypogranular cytoplasm, pseudo-Pelger-Huet anomaly. Interestingly, Auer rods were also seen in maturing neutrophils, some of which displayed Auer rod faggots/ bundles. Immunophenotyping was unremarkable (myeloperoxidase, CD13,CD33, CD34 and no lineage infidelity). Since the presence of Auer rod bundles, more generally found in blasts than in maturing cells at diagnosis, is highly suggestive of a diagnosis of acute promyelocytic leukaemia (APL), this diagnosis was initially considered probable despite the atypical morphology of the blasts. Indeed, the morphology of leukaemic blasts may be unusual and dysgranulopoiesis has been reported in APL with variant RARA rearrangements [t(11;17) or t(5;17)]. Cytogenetic analyses however did not confirm APL, but revealed a complex karyotype in which a chromosome 11 was split into three fragments: 46, XY,der(6) t(6;11)(p21;q13),der(8)t(8;6)(q24;p21),der(9)del(9)(p12)del(9) (q12),der(11)t(6;11)(p21;q13)t(11;17)(p13;p11),der(17)t(11;17) (p13;p11), + 1 5 dmin. The absence of RARA involvement was further demonstrated by fluorescence in situ hybridization and reverse transcription polymerase chain reaction. FISH with a MYC-IGH probe however, revealed an impressive amplification of MYC carried by double minute chromosomes (red spots, lower right image). Before the karyotype was available, treatment as for APL was started with cytarabine, an anthracycline, and all-transretinoic acid (ATRA). When a diagnosis of APL was ruled out by cytogenetic analysis, ATRA (which had not induced any sign of blast differentiation) was discontinued. After the induction-treatment-related aplasia, peripheral blood blasts reappeared and rapidly increased and the patient died within a few days. Blast cells with bundles of Auer rods are a common observation in APL. In this disease, ‘faggots’ can also be seen in mature cells after ATRA-induced differentiation. In nonpromyelocytic leukaemias, faggot cells are rarely observed, often in association with MYC amplification. However our observation underlines the lack of specificity of Auer rod bundles for APL and the interest of detecting MYC amplification in this context.