To the Editor:

We read with interest the article by Adams et al.1 summarizing microdeletions and microduplication syndromes with cancer predisposition detected by array-based comparative genomic hybridization. We would like to add an additional, potentially recognizable microdeletion to this list, 2p16.2 p21.

This deletion was first described by Sanders et al.2 in a child with developmental delay and mild facial dysmorphism including microcephaly and hypotelorism. Retrospectively, these facial features are possibly due to deletion of SIX3 (2p21) mutations which are responsible for holoprosencephaly 2, which has an extremely variable phenotype even within the same family.3 Patients with loss of SIX3 may have a severe central nervous system defect that precludes survival, while others may survive into adult life and be at risk for the Lynch syndrome, due to deletion of one or two mismatch repair genes, MSH2 and MSH6. For example, Lucci-Cordisco et al.4 described a 37-year-old woman with a history of developmental delay and mild dysmorphic features who developed cancer. A routine peripheral blood karyotype was normal. However, the colon tumor exhibited microsatellite instability, and more detailed study revealed microdeletion 2p16.21 including the MSH2 gene.

We recently evaluated a 21-year-old man with developmental delay, mental retardation, short stature, microcephaly, and hypotelorism. Peripheral blood chromosome analysis performed in infancy as well as a head computed tomography scan were reportedly normal. Because of concern about heritability evoked by his two normal sibs, an array-based comparative genomic hybridization was performed (CMDX, Irvine, CA), which showed deletion of 2p16.3p21, including all three relevant genes, SIX3, MSH2, and MSH6. Both parents were normal.

Most patients with the Lynch syndrome do not develop colon cancer or any of the other tumors within the Lynch syndrome spectrum until the third decade. Our patient had a history of unexplained bouts of diarrhea. A colonoscopy was performed, fortunately revealing only an inflammatory pseudopolyp. We provided a screening protocol for the Lynch syndrome to the parents and referring physician.