Ataxia telangiectasia (A-T) is a rare autosomal recessive disorder caused by biallelic mutations in the Ataxia Telangiectasia-mutated gene. A-T shows a complex phenotype ranging from early-onset progressive neurodegeneration to immunodeficiencies, high incidence of infections, and tumors. Unfortunately, no therapy is up to now available for treating this condition. Recently, the short term treatment of ataxia-telangiectasia patients with glucocorticoids was shown to improve their neurological symptoms and possibly reverse cerebellar atrophy. Thus, corticosteroids represent an attractive approach for the treatment of this neurodegenerative disease. However, the molecular mechanism involved in glucocorticoid action in A-T is yet unknown. The aim of our work is to construct cDNA libraries containing those genes which are transactivated by the glucocorticoid analogue, dexamethasone, in A-T human cells. For this purpose, suppression subtractive hybridization has been performed on ATM-null lymphoblastoid cell transcriptome extracted following drug administration. Annotation of whole genes contained in the libraries has been obtained by coupling subtractive hybridization with microarray analysis. Positive transcripts have been validated by quantitative PCR. Through in silico analyses, identified genes have been classified on the basis of the pathway in which they are involved, being able to address signaling required for dexamethasone action. Most of the induced transcripts are involved in metabolic processes and regulation of cellular processes. Our results can help to unravel the mechanism of glucocorticoid action in the reversion of A-T phenotype. Moreover, the induction of a specific region of the ATM transcript has been identified as putative biomarker predictive of dexamethasone efficacy on ataxic patients.