Mutations of PTEN and the PI3K-AKT pathway in T-ALL. (A) Sequencing of PTEN in 44 of the primary samples shown in Figure 1 identified nonsynonymous sequence alterations in 12 of these samples, all of which were predicted to disrupt the PTEN protein within an 18-amino acid region of the C2 domain. Note that the specific mutations in cases 14 and 27 were impossible to determine because of the presence of 2 simultaneous frameshift sequences. (B) Targeted sequencing of PIK3R1, PIK3CA, and AKT1-3 exons known to be mutated in human cancer identified nonsynonymous sequence alterations in PTEN and the PI3K-AKT pathway in 47.7% of primary T-ALL cases. Lesions within the PTEN-PI3K-AKT pathway were mutually exclusive. Abnormalities in the NF1 and RAS genes were also identified but were not solely associated with PTEN-PI3K-AKT pathway abnormalities. *Novel in-frame insertion/deletions. (C-D) Kaplan-Meier event-free survival curves for the 44 cases analyzed by CGH and sequencing demonstrate that, overall, genetic alterations of the PTEN-PI3K-AKT pathway did not predict event-free survival, whereas deletions of PTEN were significantly associated with early treatment failure.

PTEN deletions in T-ALL. (A) Array CGH was performed with genomic DNA from diagnostic specimens collected from 47 children with T-ALL. The data are shown as a dChip plot of CGH segmented log₂ copy number ratios at the PTEN locus. The red box denotes the location of the PTEN coding sequence. White arrowheads point to cases with segmented log₂ copy number ratios of less than −0.5 involving the PTEN coding sequence. Two samples on which CGH was unsuccessful (T-ALL 36 and 37) were excluded from analysis. (B-D) Raw CGH data from representative patient samples. Red lines represent the segmented log₂ copy number ratio shown in panel A. (E-G) FISH analysis of representative cases confirmed the deletions identified by CGH. Orange, PTEN probe; green, centromere 10 probe. Images were obtained with an Axio Imager A1 fluorescence microscope (Carl Zeiss) using a 100× Alpha Apochromatic Plan oil-immersion objective (Carl Zeiss), a JAI CV-M4+CL progressive scan camera (JAI Inc), and Genus acquisition software version 3.92 build 7 (Genetix USA). Note that the Genus cytogenetic image acquisition software applies an automated “thresholding” algorithm that sets a signal intensity threshold below which any signal is considered background and thus excluded from the final composite image. During image acquisition, all images generated by the software were compared to the view from the microscope to confirm that they were fully representative. (B,E) Homozygous deletions had log₂ ratios of −1.26 (case 44) and −4.11 (case 45). Cells were available for FISH on case 44 and clearly showed homozygous loss of PTEN. (C,F) The CGH detection of a heterozygous deletion in case 34 (log₂ ratio, −0.54) correlated with the detection of PTEN deletion by FISH on 1 allele in 21% of the cells examined. (D,G) Case 21 retained both PTEN alleles intact by FISH and CGH.