A 75yr man with a two year history of diabetes mellitus (NIDDM) was referred by the cardiologists for urgent haematological investigation of a mild, normochromic anaemia (Hb 100g/L, MCV 95fl, WBC 4.9x109/L, platelets 276x109/L) associated with a high serum ferritin (1419μg/L). Vitamin B12 and serum folate levels were normal. He had developed symptoms of tiredeness, breathlessness and fatigue and was said to be taking oral iron supplements. A bone marrow biopsy report described the erythropoiesis as "micronormoblastic with dyserythropoiesis" and 40% of the nucleated cells were erythroid. The trephine showed no abnormal infiltrate and all three lineages were represented. A bone marrow karyotype analysis reported the loss of the Y chromosome in 65% metaphases as the sole abnormality. The patient was lost to follow up by the haematologists until he presented again five years later with symptoms of anaemia. At this time, he was found to have a severe microcytic, hypochromic anaemia (Hb 57g/L, MCV 67.6fl, MCH 18.3pg), a broad red cell size distribution (RDW 29.1%), normal WBC 4.7x109/L and platelets 306x109/L, and increased serum ferritin (1223μg/L). His biochemistry profile was essentially unchanged. Physical examination revealed no lymph nodes and no abdominal organomegaly. His red blood cells were markedly dimorphic with the occasional siderocyte containing multiple Perls’ positive granules. On this occasion his bone marrow karyotype showed 45X-Y in all metaphases without any additional abnormalities and Perls’ stain of the marrow smear confirmed the presence of storage iron and ring sideroblasts. He was started on folic acid (5mgtd) and pyridoxine (100mgbd) and a blood sample was referred for sequence analysis of the erythroid-specific ALA Synthase gene (ALAS2) in buffy coat genomic DNA. All ALAS2 exons, intron-exon boundaries, 900bp promoter and at least 100bp 3′flanking sequence of genomic DNA were PCR amplified and sequenced. This showed, as the only variation, an apparent heterozygosity for the 561 CGC→CAC variation predicting Arg170His described previously by other investigators in two female heterozygotes with sideroblastic anaemia that was unresponsive to pyridoxine. PCR amplification of ALAS2 exon5 from a repeat peripheral blood DNA sample taken four months later confirmed the presence of the two sequences, with a small increase in the proportion of the variant allele of uncertain significance. Peripheral blood reticulocyte RNA was prepared and RTPCR using primers specific for ALAS2mRNA showed expression only of the variant Arg170His allele. In order to test if this was an acquired variation PCR amplification of ALAS2 exon 5 in DNA extracted from two buccal smears yielded a product containing only the usual sequence. The patient experienced no benefit and demonstrated no haematological response to pyridoxine. Furthermore, although he remains on pyridoxine and folic acid, he is now receiving regular blood transfusions. To our knowledge this is the first molecular demonstration of acquired sideroblastic anaemia caused by a somatic mutation in the erythroid-specific ALA Synthase gene. In addition, the severity and refractory nature of this patient’s acquired anaemia confirms the markedly deleterious nature of this mutation.