This autosomal recessive disorder, with a birth incidence of about 1 in 300,000, is characterized by the development of cerebellar ataxia in the first decade, along with choreoathetosis, dysarthria, and abnormalities of ocular movements. Mental retardation is not usually a feature. The neurological features are progressive, leading to confinement to a wheelchair in the second decade of life. Oculocutaneous telangiectasia develops in childhood (often after ataxia is apparent) and then spreads to involve other exposed cutaneous areas. Vitiligo, café-au-lait spots, and macular hyperpigmentation may occur. The development of acanthosis nigricans is associated with the development of neoplasia. An immune deficiency occurs, with disordered B cell and T helper cell function, thymic hypoplasia, reduced levels of IgA (70 %) and IgE (80 %), and reduced T cells. Frequent bacterial (pulmonary or sinus) infections occur secondary to immunodeficiency. The serum alpha-fetoprotein is consistently elevated. There is an inconsistently increased incidence of spontaneous structural chromosomal aberrations (30–50-fold) (chromatid gaps, breaks and interchanges, and telomere fusions) in cultured white cells and fibroblasts, and this is markedly increased by exposure to X- radiation and radiomimetic agents. Peripheral blood lymphocytes may show abnormal clones of cells with a stable cytogenetic rearrangement, usually involving chromosome 14, particularly involving the T cell receptor genes on 14q11, 7q14, and 7q35. Clones of cells with these translocations may develop into T cell promyelocytic leukemia. Other translocations involve the immunoglobulin genes in B lymphocytes. In vivo sensitivity to X-rays is also observed. Recombination is increased by a factor of 30–200 (Tomanin et al. 1989; Peterson et al. 1992; Viniou et al. 2001; Sun et al. 2002).
There is a 30–40 % risk of malignancy developing, most frequently Hodgkin and non-Hodgkin lymphoma (60 %) and lymphoblastic T cell leukemia (27 %) (Johnson 1989). The lymphoreticular neoplasms develop before the age of 16 years, and epithelial carcinomas (including medulloblastomas; gastric, basal cell, hepatocellular, parotid, laryngeal, skin, and breast carcinomas; uterine leiomyomas; and ovarian dysgerminoma) may develop in patients surviving longer (Spector et al. 1982). Severe reactions have been described to standard doses of radiotherapy in this condition.
The ATM gene was cloned in 1995 (Savitsky et al. 1995). Its product, ATM, has a central role in orchestrating the response to double-strand breaks. It is a serine/threonine kinase that mediates checkpoint regulation and homologous repair by phosphorylating a number of proteins; without it, cells display aberrant cell-cycle progression and increased chromosomal breakage, especially when exposed to ionizing radiation. McConville et al. (1996) and others have reported families with a milder clinical and cellular phenotype due to certain ataxia-telangiectasia mutations (Sariozzi et al. 2002; Chessa 2003) causing reduced ATM function. The ATM carrier frequency is estimated to be about 1 %, and early studies showed that these heterozygotes have an increased risk of cancer, particularly breast cancer (Morrell et al. 1990). This was supported by studies conducted after the gene was identified (Athma et al. 1996; Thomson et al. 2005). These epidemiological studies were confirmed directly when Renwick et al. identified 12 deleterious mutations in 443 affected cases compared to 2 out of 521 controls resulting in an estimated relative risk of 2.37 (Renwick et al. 2006). One limitation of the study however was that the significance of 35 of 37 missense mutations identified could not be determined. To try to solve this problem, Tavtigian et al. used an in silico approach to test the hypothesis that missense mutations in evolutionary conserved residues would be more common among breast cancer patients than in controls. Suggestive of a dominant-negative model, they found that missense mutations ranked highly deleterious were associated with an even greater risk of breast cancer than truncation or splice site mutations, demonstrating that not all variants have equal risk predisposition (Tavtigian et al. 2009; Thorensten et al. 2003), but this remains controversial.