Page 398 - 2014 Printable Abstract Book
P. 398
The large-scale sequencing of de novo AML has fundamentally shaped our understanding of its
clonal evolution (TCGA, NEJM, 2013). In healthy individuals, hematopoietic stem cells (HSCs) accumulate
mutations with age resulting in a genetically heterogeneous HSC population. By age 50, each individual
HSC has on average 5 unique coding somatic mutations (Welch et al, Cell, 2012). De novo AML evolves
through the acquisition of select “driver” mutations followed by clonal expansion. Cytotoxic therapy can
alter the evolution of these leukemias (Ding et al. Nature, 2012). Therapy-related AML/MDS (t-AML/MDS)
is a well described complication of chemotherapy and/or radiotherapy. Compared to de novo AML, t-AML
is more likely to be associated with high-risk chromosomal abnormalities and worse clinical outcomes.
How exposure to cytotoxic therapy affects the evolution of t-AML/t-MDS remains unknown. These
disorders have a distinct mutational spectrum compared to their de novo counterparts. In particular, they
are enriched in somatic TP53 mutations. We sequenced the genome of 22 cases of t-AML. Surprisingly,
they had a similar number of somatic single nucleotide variants and a similar percentage of
chemotherapy-associated transversions as de novo AML. This suggests that cytotoxic therapy is not
directly inducing somatic TP53 mutations through genome-wide DNA damage.
Instead, we considered the hypothesis that TP53 somatic mutations may be age-related and occur prior
to cytotoxic therapy. Hematopoietic cells, having acquired these mutations, would be resistant to
chemotherapy and preferentially expand after treatment. Using a newly developed high sensitivity
sequencing approach, we discovered functional TP53 somatic mutations in small populations of
peripheral blood leukocytes in individuals never previously exposed to cytotoxic therapy. In 3 patients
with t-AML/t-MDS, we identified the exact TP53 mutation clonal at diagnosis in prior banked specimens
obtained 3-6 years earlier, in one case prior to any cytotoxic therapy exposure. In mouse models, Tp53+/-
HSCs outcompete their wild-type counterparts under the selective pressure of radiation (Bondar et al. Cell
Stem Cell, 2010) or certain chemotherapies. The early acquisition of TP53 somatic mutations may explain
many of the characteristic features of t-AML.
(SNE03) Genes and Radiation: Does genetic predisposition matter in radiation-induced breast cancer?
2
3
1
2
Jonine Bernstein ; Dan Stram ; Duncan Thomas ; and Roy Shore, Memorial Sloan Kettering Cancer Center,
1
New York, NY ; department of Preventive Medicine, Keck School of Medicine, University of Southern
3
2
California, Los Angeles, CA ; and Radiation Effects Research Foundation, New York, NY
Broadening genome-wide association studies (GWAS) to encompass alleles that are involved in
gene-environment interactions is important for the understanding of the full range of genetic influence
on cancer risk and for adding to our basic knowledge of cancer etiology. In radiation epidemiology the
discovery of alleles that modify individual genetic sensitivity to the carcinogenic effects of radiation
exposure would have potential consequences for radiation protection policy as well as for clinical practice.
The Women’s Environment Cancer and Radiation Epidemiology (WECARE) Study includes a two stage
GWAS in which women with second primary breast cancers in the contralateral breast (CBC cases) are
compared to matched controls with unilateral breast cancer only (UBC controls), for the purpose of finding
genetic variants that either increase susceptibility to CBC on their own or interact with radiation exposure
(estimated from treatment records for the first cancer) to increase CBC risk. At present the WECARE Study
has completed both stages of a two-stage genotyping design: Stage 1 is a GWAS study, in which
approximately 1 million SNPs were genotyped using the Illumina OMNI platform. After removing
genotyping failures, a total of 641 CBCcases and 1247 matched UBC controls comprised Stage 1. Stage 2
consists of 816 women with CBC and 816 UBC controls. These women were genotyped for SNPs that had
clonal evolution (TCGA, NEJM, 2013). In healthy individuals, hematopoietic stem cells (HSCs) accumulate
mutations with age resulting in a genetically heterogeneous HSC population. By age 50, each individual
HSC has on average 5 unique coding somatic mutations (Welch et al, Cell, 2012). De novo AML evolves
through the acquisition of select “driver” mutations followed by clonal expansion. Cytotoxic therapy can
alter the evolution of these leukemias (Ding et al. Nature, 2012). Therapy-related AML/MDS (t-AML/MDS)
is a well described complication of chemotherapy and/or radiotherapy. Compared to de novo AML, t-AML
is more likely to be associated with high-risk chromosomal abnormalities and worse clinical outcomes.
How exposure to cytotoxic therapy affects the evolution of t-AML/t-MDS remains unknown. These
disorders have a distinct mutational spectrum compared to their de novo counterparts. In particular, they
are enriched in somatic TP53 mutations. We sequenced the genome of 22 cases of t-AML. Surprisingly,
they had a similar number of somatic single nucleotide variants and a similar percentage of
chemotherapy-associated transversions as de novo AML. This suggests that cytotoxic therapy is not
directly inducing somatic TP53 mutations through genome-wide DNA damage.
Instead, we considered the hypothesis that TP53 somatic mutations may be age-related and occur prior
to cytotoxic therapy. Hematopoietic cells, having acquired these mutations, would be resistant to
chemotherapy and preferentially expand after treatment. Using a newly developed high sensitivity
sequencing approach, we discovered functional TP53 somatic mutations in small populations of
peripheral blood leukocytes in individuals never previously exposed to cytotoxic therapy. In 3 patients
with t-AML/t-MDS, we identified the exact TP53 mutation clonal at diagnosis in prior banked specimens
obtained 3-6 years earlier, in one case prior to any cytotoxic therapy exposure. In mouse models, Tp53+/-
HSCs outcompete their wild-type counterparts under the selective pressure of radiation (Bondar et al. Cell
Stem Cell, 2010) or certain chemotherapies. The early acquisition of TP53 somatic mutations may explain
many of the characteristic features of t-AML.
(SNE03) Genes and Radiation: Does genetic predisposition matter in radiation-induced breast cancer?
2
3
1
2
Jonine Bernstein ; Dan Stram ; Duncan Thomas ; and Roy Shore, Memorial Sloan Kettering Cancer Center,
1
New York, NY ; department of Preventive Medicine, Keck School of Medicine, University of Southern
3
2
California, Los Angeles, CA ; and Radiation Effects Research Foundation, New York, NY
Broadening genome-wide association studies (GWAS) to encompass alleles that are involved in
gene-environment interactions is important for the understanding of the full range of genetic influence
on cancer risk and for adding to our basic knowledge of cancer etiology. In radiation epidemiology the
discovery of alleles that modify individual genetic sensitivity to the carcinogenic effects of radiation
exposure would have potential consequences for radiation protection policy as well as for clinical practice.
The Women’s Environment Cancer and Radiation Epidemiology (WECARE) Study includes a two stage
GWAS in which women with second primary breast cancers in the contralateral breast (CBC cases) are
compared to matched controls with unilateral breast cancer only (UBC controls), for the purpose of finding
genetic variants that either increase susceptibility to CBC on their own or interact with radiation exposure
(estimated from treatment records for the first cancer) to increase CBC risk. At present the WECARE Study
has completed both stages of a two-stage genotyping design: Stage 1 is a GWAS study, in which
approximately 1 million SNPs were genotyped using the Illumina OMNI platform. After removing
genotyping failures, a total of 641 CBCcases and 1247 matched UBC controls comprised Stage 1. Stage 2
consists of 816 women with CBC and 816 UBC controls. These women were genotyped for SNPs that had
