- Email: [email protected]
The pancreas cancer genome project
Abstracts
37
S E X - R E L A T E D D I F F E R E N C E S A N D L A C K O F i(12p) F O U N D IN C Y T O G E N E T I C C H A R A C T E R I Z A T I O N O F 82 P E D I A T R I C G E R M C E L L TUMORS. KJ Bassev L9, HJ Lawce 1,9, DC Arthur 2,9, DK Kalousek 3,9, M Krailo 9, 17,Giller4,9 P Rogers 3.9, C Stolar 5,9, H lssacs 6,9, S Heifetz 7,9, R Womer 8,9, and RE Magenis 1,9. lOregon Health Sciences University (OHSU), 2University of Minnesota, 3University of British Columbia, 4University of Iowa, 5Babies Hospital, New York City, 6Children's Hospital of Los Angeles, 7Indiana University, aChildren's Hospital of Philadelphia, 9Children's Cancer Group. Germ cell tumors (GCT) are rare tumors in children, accounting for less than 3 percent of all childhood cancers. Because of the rarity of this pediatric tumor type, few cytogenetic studies have been done. We report here on a series of 91 tumors submitted for cytogenetic analysis as part of two clinical trials conducted by the Children's Cancer Group (CCG). Sixty-three of the 91 tumors were processed at OHSU; the remainder were studied at other participating CCG institutions. Constitutional studies for some of the patients were done at institutional laboratories. Tumors were mechanically and enzymatically dissociated and placed into culture, as well as harvested directly if possible. Only those specimens with a culture duration of 10 days or less were used in the analysis of the data. Five female tumors and four male tumors were deemed ineligible because of extended culture time, a subsequent diagnosis inconsistent with germ cell tumor, or the age of the patient. A total of 82 tumors (30 male and 52 female) were included in the analysis. The ages of the patients ranged from newborn to 18 years. The patient population included 5 patients with confirmed constitutional abnormalities. These were three 47,XXY, one 46,X,inv(Y)(pl 1.2ql 1.2), one 47,XXX, and one 47,XY,+21. Both sexes have tumors that are characterized by predominantly normal or simple karyotypes. 44 tumors had a karyotype that was either normal (46,XX, 46,XY) or had the same abnormality as the constitutional karyotype. 13 of these were saccroccygeal tumors from children less than 1 year old. 5 teratomas were (23,X)x2. The most common abnormalities seen in the tumors varied depending on the sex of the patient. Male tumors most frequently showed +21,+7,-11, and lp. Only gain of 21 was seen in simple kuryotypos. Three such karyotypes were from tumors with an XXY complement, which was confirmed to be constitutional in two cases. All of the tumors from males with constitutional abnormalities were extragonadal, ain of i(12p) was seen primarily in post-pubertal male tumors (only one exception was noted). The most common abnormalities in female tumors were + 1q,+2,+3,+8,+14, and -4. All but the -4 were present in both simple and complex karyotypes; loss of 4 was only found in complex karyotypes suggesting a potential role in progression. Gain of 14 was most common in simple karyotypes, gains of 2 and 3 were present in roughly equal frequencies between simple and complex karyotypos, and gains of lq and 8 were more often associated with complex karyotypes. Breakpoint analysis demonstrated clustering around the 12 centromere due to the i(12p) in post-pubertal male tumors. Clustering was observed at lp36, 1p32, Iq11-23, 8(124, 15pll, and 1 8 p l l . l - p l l . 2 . These results suggest that i(12p) is not associated with pre-pubertal male GCT and that sexrelated differences exist for both tumor-related abnormalities and the incidence of constitutional abnormalities.
38
THE PANCREAS CANCER GENOME PROJECT. SE Kern. S Schutte, M Goggins, E Rozenbhim, W Hilgers, CJ Yeo, RH Hruban. The Johns Hopkins University School of Medicine, Baltimore. Descriptive science can often be the more efficient approach to discovery. Thus, we aim to describe the altered genomes found in pancreatic cancer. Results of molecular genetic studies indicate a rich collage of specific mutations and a promising hunting grounds for further gene mutations. The molecular genetic alterations of pancmalic carcinoma include somatic K-ras mutations in 90 percent, p l 6 inactivation in 85 percent, p53 mutations in 75 percent, and DPC4 inactivation in 50 percent. Germline BRCA2 mutations are reasonably common, being found in 7 percent of tumors which are clinically sporadic. Gcrnflthe p l 6 mutations are rare and represent an overlap syndrome with familial melanoma. Recurrent chromosomal structural alterations are difficult to diseem, and the occasional reports of gene amplification have not been reproduced to date. At least 64 percent of pancreatic cancers are known to carry a homozygous deletion at a tumor-suppressor gene, and at least 19 percent have two or more such deletions. Homozygous deletions have been cenu'al to the discovery of several tamor-supprassor genes. Now, at least two methods are available for the efficient discovery of novel homozygous deletions. 1) The representational difference analysis (RDA) allows the direct cloning of homozygous deletions in an unbiased manner. We reported the map of a homozygoos deletion in a pancreatic cancer, cloned by RDA and located within the region linked to BRCA2. Because the small size of the homozygous deletion allows the focussed construction of a detailed map, the random markers of the map contained the first published sequences of the BRCA2 gene, which was subsequently defined by others. 2) Genome scanning with spaced markers can be used to explore a chromosomal man suspected of an undiscovered tumor-supprassor gene. We localized a hotspot of homozygous deletion of 18q in 30 percent of pancreatic cancers. Only one gene encompassed all deletions, and the DPC4 gene was found to have intragenic mutations in another 20 percent of these cancers. DPC4 is a homolog of the Drosophila Mad gene, involved in TGF-b supeffamily signal tmnsduction. An alleltuype of pancreatic cancer showed that 40 percent of the chromosomal arms had suffered loss of heterozygosity, and that more than a dozen arms had losses in over 40 percent of tumors. Some additional high-prevalence mutational targets might therefore remain. Realistically, however, one might suspect that most tumor-suppressor genes involve small subsets of each tumor type, and that a better understanding will await a picture of cancer that emphasizes pathways rather than individual gene members.
15 5
39 Patrick Brown's abstract w a s not available at the time of printing. 40
D E T E C T I O N O F H E T E R O Z Y G O U S M U T A T I O N S IN B R C A I U S I N G H I G H D E N S I T Y O L I G O N U C L E O T I D E A R R A Y S AND T W O - C O L O R F L U O R E S C E N C E ANALYSIS JG Hacia I , LC Brody 1, MS Cbee 2, SA Fodor 2, and FS Collins 1 1National Center for Human Genome Research, National Institutes of Health, Bethesda, MD; 2Affymetrix, Inc., Santa Clara, CA. The ability to scan a large gene rapidly and accurately for all possible heterozygous mutations in large numbers of patient samples will be critical for the future of medicine. High density arrays consisting of over 96,600 oligonucleotides 20-nt in length were designed to screen for a wide range of beterozygous mutations in the 3.45-kb exon 11 of the BRCAI gene. Reference and test samples were cohybridized to these arrays and differences in hybridization patterns quantitated by two-color analysis. Fourteen of fifteen patient samples with known exon 11 mutations were accurately diagnosed, and no false positive mutations were identified in 20 control samples. Eight single nucleotide polymorphisms were also readily detected. DNA chip-based assays may provide a valuable new technology for highthroughput cost-efficient detection of genetic alterations.
37
S E X - R E L A T E D D I F F E R E N C E S A N D L A C K O F i(12p) F O U N D IN C Y T O G E N E T I C C H A R A C T E R I Z A T I O N O F 82 P E D I A T R I C G E R M C E L L TUMORS. KJ Bassev L9, HJ Lawce 1,9, DC Arthur 2,9, DK Kalousek 3,9, M Krailo 9, 17,Giller4,9 P Rogers 3.9, C Stolar 5,9, H lssacs 6,9, S Heifetz 7,9, R Womer 8,9, and RE Magenis 1,9. lOregon Health Sciences University (OHSU), 2University of Minnesota, 3University of British Columbia, 4University of Iowa, 5Babies Hospital, New York City, 6Children's Hospital of Los Angeles, 7Indiana University, aChildren's Hospital of Philadelphia, 9Children's Cancer Group. Germ cell tumors (GCT) are rare tumors in children, accounting for less than 3 percent of all childhood cancers. Because of the rarity of this pediatric tumor type, few cytogenetic studies have been done. We report here on a series of 91 tumors submitted for cytogenetic analysis as part of two clinical trials conducted by the Children's Cancer Group (CCG). Sixty-three of the 91 tumors were processed at OHSU; the remainder were studied at other participating CCG institutions. Constitutional studies for some of the patients were done at institutional laboratories. Tumors were mechanically and enzymatically dissociated and placed into culture, as well as harvested directly if possible. Only those specimens with a culture duration of 10 days or less were used in the analysis of the data. Five female tumors and four male tumors were deemed ineligible because of extended culture time, a subsequent diagnosis inconsistent with germ cell tumor, or the age of the patient. A total of 82 tumors (30 male and 52 female) were included in the analysis. The ages of the patients ranged from newborn to 18 years. The patient population included 5 patients with confirmed constitutional abnormalities. These were three 47,XXY, one 46,X,inv(Y)(pl 1.2ql 1.2), one 47,XXX, and one 47,XY,+21. Both sexes have tumors that are characterized by predominantly normal or simple karyotypes. 44 tumors had a karyotype that was either normal (46,XX, 46,XY) or had the same abnormality as the constitutional karyotype. 13 of these were saccroccygeal tumors from children less than 1 year old. 5 teratomas were (23,X)x2. The most common abnormalities seen in the tumors varied depending on the sex of the patient. Male tumors most frequently showed +21,+7,-11, and lp. Only gain of 21 was seen in simple kuryotypos. Three such karyotypes were from tumors with an XXY complement, which was confirmed to be constitutional in two cases. All of the tumors from males with constitutional abnormalities were extragonadal, ain of i(12p) was seen primarily in post-pubertal male tumors (only one exception was noted). The most common abnormalities in female tumors were + 1q,+2,+3,+8,+14, and -4. All but the -4 were present in both simple and complex karyotypes; loss of 4 was only found in complex karyotypes suggesting a potential role in progression. Gain of 14 was most common in simple karyotypes, gains of 2 and 3 were present in roughly equal frequencies between simple and complex karyotypos, and gains of lq and 8 were more often associated with complex karyotypes. Breakpoint analysis demonstrated clustering around the 12 centromere due to the i(12p) in post-pubertal male tumors. Clustering was observed at lp36, 1p32, Iq11-23, 8(124, 15pll, and 1 8 p l l . l - p l l . 2 . These results suggest that i(12p) is not associated with pre-pubertal male GCT and that sexrelated differences exist for both tumor-related abnormalities and the incidence of constitutional abnormalities.
38
THE PANCREAS CANCER GENOME PROJECT. SE Kern. S Schutte, M Goggins, E Rozenbhim, W Hilgers, CJ Yeo, RH Hruban. The Johns Hopkins University School of Medicine, Baltimore. Descriptive science can often be the more efficient approach to discovery. Thus, we aim to describe the altered genomes found in pancreatic cancer. Results of molecular genetic studies indicate a rich collage of specific mutations and a promising hunting grounds for further gene mutations. The molecular genetic alterations of pancmalic carcinoma include somatic K-ras mutations in 90 percent, p l 6 inactivation in 85 percent, p53 mutations in 75 percent, and DPC4 inactivation in 50 percent. Germline BRCA2 mutations are reasonably common, being found in 7 percent of tumors which are clinically sporadic. Gcrnflthe p l 6 mutations are rare and represent an overlap syndrome with familial melanoma. Recurrent chromosomal structural alterations are difficult to diseem, and the occasional reports of gene amplification have not been reproduced to date. At least 64 percent of pancreatic cancers are known to carry a homozygous deletion at a tumor-suppressor gene, and at least 19 percent have two or more such deletions. Homozygous deletions have been cenu'al to the discovery of several tamor-supprassor genes. Now, at least two methods are available for the efficient discovery of novel homozygous deletions. 1) The representational difference analysis (RDA) allows the direct cloning of homozygous deletions in an unbiased manner. We reported the map of a homozygoos deletion in a pancreatic cancer, cloned by RDA and located within the region linked to BRCA2. Because the small size of the homozygous deletion allows the focussed construction of a detailed map, the random markers of the map contained the first published sequences of the BRCA2 gene, which was subsequently defined by others. 2) Genome scanning with spaced markers can be used to explore a chromosomal man suspected of an undiscovered tumor-supprassor gene. We localized a hotspot of homozygous deletion of 18q in 30 percent of pancreatic cancers. Only one gene encompassed all deletions, and the DPC4 gene was found to have intragenic mutations in another 20 percent of these cancers. DPC4 is a homolog of the Drosophila Mad gene, involved in TGF-b supeffamily signal tmnsduction. An alleltuype of pancreatic cancer showed that 40 percent of the chromosomal arms had suffered loss of heterozygosity, and that more than a dozen arms had losses in over 40 percent of tumors. Some additional high-prevalence mutational targets might therefore remain. Realistically, however, one might suspect that most tumor-suppressor genes involve small subsets of each tumor type, and that a better understanding will await a picture of cancer that emphasizes pathways rather than individual gene members.
15 5
39 Patrick Brown's abstract w a s not available at the time of printing. 40
D E T E C T I O N O F H E T E R O Z Y G O U S M U T A T I O N S IN B R C A I U S I N G H I G H D E N S I T Y O L I G O N U C L E O T I D E A R R A Y S AND T W O - C O L O R F L U O R E S C E N C E ANALYSIS JG Hacia I , LC Brody 1, MS Cbee 2, SA Fodor 2, and FS Collins 1 1National Center for Human Genome Research, National Institutes of Health, Bethesda, MD; 2Affymetrix, Inc., Santa Clara, CA. The ability to scan a large gene rapidly and accurately for all possible heterozygous mutations in large numbers of patient samples will be critical for the future of medicine. High density arrays consisting of over 96,600 oligonucleotides 20-nt in length were designed to screen for a wide range of beterozygous mutations in the 3.45-kb exon 11 of the BRCAI gene. Reference and test samples were cohybridized to these arrays and differences in hybridization patterns quantitated by two-color analysis. Fourteen of fifteen patient samples with known exon 11 mutations were accurately diagnosed, and no false positive mutations were identified in 20 control samples. Eight single nucleotide polymorphisms were also readily detected. DNA chip-based assays may provide a valuable new technology for highthroughput cost-efficient detection of genetic alterations.