The inherited bone marrow failure syndromes (IBMFS) are a heterogeneous group of rare genetic disorders with distinctive phenotypic and laboratory abnormalities. Patients with IBMFS, including Fanconi Anemia (FA), Diamond-Blackfan Anemia (DBA), Shwachman-Diamond Syndrome (SDS), and Dyskeratosis Congenita (DC), have an increased risk of developing aplastic anemia (AA), myelodysplastic syndrome (MDS), and acute myeloid leukemia (AML). To determine the incidence, types, and possible clinical significance of abnormal bone marrow karyotypes among patients with IBMFS, we are conducting prospective, serial, routine and molecular cytogenetic analyses of marrow. We hypothesize that abnormal marrow karyotypes, without other evidence of MDS (significant cytopenias or morphologic dyspoiesis), may not predict an adverse outcome. Analyses have included centrally-reviewed marrow morphology, G-banded karyotype analysis, fluorescence in situ hybridization (FISH), and comparative genomic hybridization (CGH). Patients have been followed for up to eight years. Clonal chromosome abnormalities have been detected in six of 15 (40 percent) patients with FA, none of 17 (0 percent) patients with DBA, two of 4 (50 percent) patients with SDS, and two of 11 (18 percent) patients with DC. G-banding is the best method for detecting abnormal clones, and FISH provides additional information; CGH is the least sensitive method. Approximately half of the clonal abnormalities documented among our patients with IBMFS are different from those commonly found in patients with sporadic MDS or AML. Abnormal clones have been found in IBMFS patients who are clinically stable and do not have morphologic MDS. Furthermore, some of these clones have waxed and waned over time. These data suggest a more conservative approach to therapy may be indicated in IBMFS patients who are known to be sensitive to cytotoxic chemotherapy and radiotherapy. To determine the prognostic and biological significance of abnormal bone marrow karyotypes in IBMFS patients, large collaborative databases including morphologic and clinical data are needed.