Carbapenem-resistant A. baumannii are prevalent in low- and middle-income countries such as Egypt, but little is known about the molecular epidemiology and mechanisms of resistance in these settings. Here we characterise carbapenem-resistant A. baumannii from Alexandria, Egypt, and place it in a regional context. 54 carbapenem-resistant isolates from Alexandria Main University Hospital, Egypt, collected between 2010 and 2015 were genome sequenced using Illumina technology. Genomes were de novo assembled and annotated. Genomes for 36 isolates from the Middle East region were downloaded from GenBank. Core gene compliment was determined using Roary, and analyses of recombination were performed in Gubbins. MLST sequence type and antibiotic resistance genes were identified. The majority of Egyptian isolates belonged to one of 3 major clades, corresponding to Pasteur MLST clonal complex (CCPAS) 1, CCPAS2 and sequence type (STPAS) 158. Strains belonging to STPAS158 have been reported almost exclusively from North Africa, the Middle East and Pakistan, and may represent a region-specific lineage. All isolates carried an oxa23 gene, six carried blaNDM-1, and one carried blaNDM-2. The oxa23 gene was located on a variety of different mobile elements, with Tn2006 predominant in CCPAS2 strains, and Tn2008 predominant in other lineages. Of particular concern, in 8 of the 11 CCPAS1 strains, the carbapenemase gene was located in a temperate bacteriophage phiOXA, previously identified only once before in a CCPAS1 clone from the US military. The carbapenem-resistant A. baumannii population in Alexandria Main University hospital is very diverse, and indicates an endemic circulating population, including a region-specific lineage. The major mechanism for oxa23 dissemination in CCPAS1 isolates appears to be a bacteriophage, presenting new concerns about the ability of these carbapenemases to spread throughout the bacterial population.Data SummaryThe whole genome shotgun sequences of the isolates from this study have been deposited at DDBJ/ENA/GenBank under the BioProject accession number PRJNA659545. The individual genome accession numbers for each isolate are as follows: A1a, JACSUC000000000; A2, JACSUB000000000; A4, JACSVQ000000000; A5, JACSUA000000000; A6, JACSTZ000000000; A7-T, JACSVP000000000; A8-T, JACSVO000000000; A8a, JACSTY000000000; A9, JACSTX000000000; A10, JACSTW000000000; A10a, JACSTV000000000; A11a, JACSTU000000000; A13a, JACSTT000000000; A14a, JACSTS000000000; A15, JACSTR000000000; A16, JACSTQ000000000; A18, JACSTP000000000; A21, JACSVN000000000; A22, JACSTO000000000; A27, JACSTN000000000; A30, JACSTM000000000; A31, JACSTL000000000; A34, JACSTK000000000; A35, JACSTJ000000000; A36, JACSTI000000000; A39, JACSTH000000000; A40, JACSTG000000000; A41, JACSTF000000000; A42, JACSTE000000000; A43, JACSTD000000000; A44, JACSTC000000000; A45, JACSTB000000000; A46, JACSTA000000000; A64, JACSSZ000000000; A68, JACSSY000000000; A69, JACSSX000000000; A70, JACSSW000000000; A71, JACSVM000000000; A72, JACSSV000000000; A73, JACSSU000000000; A74, JACSST000000000; A75, JACSSS000000000; A78, JACSSR000000000; A82, JACSSQ000000000; A83, JACSVL000000000; A84, JACSSP000000000; A85, JACSSO000000000; A86, JACSVK000000000; A87, JACSSN000000000; A88, JACSSM000000000; A89, JACSSL000000000; A92, JACSSK000000000; A5910, JACSSJ000000000; A6135, JACSVJ000000000.