- Email: [email protected]
No evidence of increased radiation-induced late effects in breast cancer patients carrying ATM gene mutations
S204
I. J. Radiation Oncology
● Biology ● Physics
Volume 57, Number 2, Supplement, 2003
Conclusions: To our knowledge, this is the first rapid assay of intrinsic radiosensitivity confirmed prospectively. CD8 T-lymphocyte apoptosis alone can predict significantly the differences in radiation-induced late toxicity between individuals, and could be used as a rapid screen for genetically hypersensitive patients. Patients in future dose escalation studies could be stratified using the apoptosis assay.
132
No Evidence of Increased Radiation-Induced Late Effects in Breast Cancer Patients Carrying ATM Gene Mutations
M. Bremer,1 T. Doerk,2 R. Bendix-Waltes,1 C. Sohn,2 J.H. Karstens1 1 Radiation Oncology, Medical School Hannover, Hannover, Germany, 2Obstetrics and Gynecology, Medical School Hannover, Hannover, Germany Purpose/Objective: The possible impact of ATM heterozygosity for clinical radiosensitivity in breast cancer patients remains a matter of debate while clinical data are scarce. A report on increased late subcutaneous toxicity after radiation therapy (RT) in breast cancer patients carrying ATM gene mutations has raised concerns about RT as part of the management in these patients (Iannuzzi et al. IJROBP 2002;52:606 – 613). Materials/Methods: Between September 1995 and December 2002 genomic DNA samples were collected from 1100 breast cancer patients receiving adjuvant RT at our department. Using mutation-specific assays, we screened for frequent ATM gene mutations in this hospital-based cohort. A retrospective analysis of acute and late radiation-related toxicity for skin and subcutaneous normal tissue was performed in patients identified as AT heterozygotes applying common toxicity criteria (CTC) and LENT/SOMA scoring criteria, respectively. Results: Eleven patients were identified to be heterozygous for a pathogenic ATM gene mutation (splicing mutation 1066-6T⬎G: 8, frameshift mutation 3801delG: 2, missense mutation S2592C: 1). The median follow-up after the end of adjuvant RT was 5 years (range: 1.8 –7.2). Two patients presented with bilateral disease. Ten patients had received at least one course of RT: 9 patients (10 breasts) following breast conservative surgery and one patient following mastectomy. One patient failed distantly and was subsequently irradiated at four different sites for bone and brain metastases. There was no evidence of increased clinical radiosensitivity with no grade 3– 4 acute or late skin or subcutaneous reactions being observed. Local relapse occurred in one single patient who had declined adjuvant RT following breast conservative therapy. Conclusions: Heterozygosity for a pathogenic ATM mutation did not predict for abnormal acute or late radiation related toxicity in our breast cancer patient cohort. Our results do not provide evidence for a relative contraindication to adjuvant RT in AT heterozygotes. Due to their increased cellular radiosensitivity these patients may differentially benefit from RT and qualify for dose and volume reduction trials.
133
Radiosensitivity Can Be Predicted by DNA-End Binding Complex Analysis
C.W. Stevens,1 S.M. Ismail,2 S. Prithivirajsingh,2 T. Buchholz,1 M. Story2 1 Radiation Oncology, U.T. M. D. Anderson Cancer Center, Houston, TX, 2Experimental Radiation Oncology, U.T. M. D. Anderson Cancer Center, Houston, TX Purpose/Objective: Predicting the radiosensitivity of normal and tumor cells has great potential importance for radiation oncology. The surviving fraction after 2Gy (SF2) has been shown in many instances to predict normal tissue toxicity and tumor control probability. SF2 measurements, however, are time consuming, expensive, and very lab-dependent. We hypothesized that an analysis of proteins that bind to DNA double-strand breaks could predict radiosensitivity, and would be more practical than SF2 determination. Materials/Methods: DNA end binding complexes (DNA-EBCs) were identified by mixing nuclear protein extracts with a radiolabeled oligonucleotide, in the presence of a vast excess of supercoiled DNA (without ends, to remove non-end-specific binding proteins like histones), followed by electrophoresis and autoradiography. DNA-EBCs from 21 primary fibroblast and 15 tumor cell lines with a variety of radiosensitivities (SF2 range 0.01– 0.40 and 0.30 – 0.80 respectively) were then correlated with SF2. To identify protein components of the DNA-EBCs, antibodies to DNA repair proteins were added to the DNA-EBC reaction. Super-shifting of DNA-EBCs demonstrated the presence of the specific protein. The nuclear protein levels of each band A component were determined in triplicate by western analysis. Results: Ten DNA-EBCs were identified. DNA-EBC analysis identified a rapidly migrating ATM-containing band (called: band-A) whose density correlated with SF2. The density of other bands, or total DNA-EBC binding, correlated more poorly with SF2 (r2⬍0.45). There were no differences in DNA-EBC migration pattern noted. Band-A density correlated well with SF2 (0.02 ⬍ SF2 ⬍ 0.41) in 21 primary fibroblast lines (r2⫽0.77). The DNA-EBC pattern of peripheral blood lymphocytes was identical to that of fibroblasts, and the band-A density was identical in matched pairs of lymphocytes/fibroblasts from an individual. Band-A density also correlated with SF2 (0.35 ⬍ SF2 ⬍ 0.80) in 15 human tumor cell lines (r2⫽0.91). When normal and tumor cell results are combined (Fig.), there is an excellent correlation between band-A density and Sf2 (r2⫽0.85). Mixing studies demonstrated that band-A density predicted SF2 of tumor cells even when contaminated with normal cells, or (in the case of xenografts) rodent cells. DNA-EBC analysis also predicted radiosensitization by two unrelated radiosensitizers: the COX-2 inhibitor SC-236, and the histone deacetylase inhibitor sodium butyrate. Supershift analysis revealed that band-A contains the following proteins: ATM, Ku70, DNA Ligase III, Rpa32, Rpa14, DNA ligase IV, XRCC4, WRN, BLM, RAD51 and p53. In a subset of primary cells that were relatively radiosensitive because of BRCA1 mutation, there was no correlation between the levels of any of these proteins and band-A density or SF2. Conclusions: Band-A density predicts SF2 of primary and tumor cells, and can predict radiosensitizing effects of biologic agents. The band-A density of lymphocytes is identical to that of fibroblasts from the same individual, demonstrating that band-A analysis of lymphocytes would be a good predictor of normal tissue SF2. DNA-EBC analysis may be a clinically useful test to predict toxicity and tumor control in individual patients.
I. J. Radiation Oncology
● Biology ● Physics
Volume 57, Number 2, Supplement, 2003
Conclusions: To our knowledge, this is the first rapid assay of intrinsic radiosensitivity confirmed prospectively. CD8 T-lymphocyte apoptosis alone can predict significantly the differences in radiation-induced late toxicity between individuals, and could be used as a rapid screen for genetically hypersensitive patients. Patients in future dose escalation studies could be stratified using the apoptosis assay.
132
No Evidence of Increased Radiation-Induced Late Effects in Breast Cancer Patients Carrying ATM Gene Mutations
M. Bremer,1 T. Doerk,2 R. Bendix-Waltes,1 C. Sohn,2 J.H. Karstens1 1 Radiation Oncology, Medical School Hannover, Hannover, Germany, 2Obstetrics and Gynecology, Medical School Hannover, Hannover, Germany Purpose/Objective: The possible impact of ATM heterozygosity for clinical radiosensitivity in breast cancer patients remains a matter of debate while clinical data are scarce. A report on increased late subcutaneous toxicity after radiation therapy (RT) in breast cancer patients carrying ATM gene mutations has raised concerns about RT as part of the management in these patients (Iannuzzi et al. IJROBP 2002;52:606 – 613). Materials/Methods: Between September 1995 and December 2002 genomic DNA samples were collected from 1100 breast cancer patients receiving adjuvant RT at our department. Using mutation-specific assays, we screened for frequent ATM gene mutations in this hospital-based cohort. A retrospective analysis of acute and late radiation-related toxicity for skin and subcutaneous normal tissue was performed in patients identified as AT heterozygotes applying common toxicity criteria (CTC) and LENT/SOMA scoring criteria, respectively. Results: Eleven patients were identified to be heterozygous for a pathogenic ATM gene mutation (splicing mutation 1066-6T⬎G: 8, frameshift mutation 3801delG: 2, missense mutation S2592C: 1). The median follow-up after the end of adjuvant RT was 5 years (range: 1.8 –7.2). Two patients presented with bilateral disease. Ten patients had received at least one course of RT: 9 patients (10 breasts) following breast conservative surgery and one patient following mastectomy. One patient failed distantly and was subsequently irradiated at four different sites for bone and brain metastases. There was no evidence of increased clinical radiosensitivity with no grade 3– 4 acute or late skin or subcutaneous reactions being observed. Local relapse occurred in one single patient who had declined adjuvant RT following breast conservative therapy. Conclusions: Heterozygosity for a pathogenic ATM mutation did not predict for abnormal acute or late radiation related toxicity in our breast cancer patient cohort. Our results do not provide evidence for a relative contraindication to adjuvant RT in AT heterozygotes. Due to their increased cellular radiosensitivity these patients may differentially benefit from RT and qualify for dose and volume reduction trials.
133
Radiosensitivity Can Be Predicted by DNA-End Binding Complex Analysis
C.W. Stevens,1 S.M. Ismail,2 S. Prithivirajsingh,2 T. Buchholz,1 M. Story2 1 Radiation Oncology, U.T. M. D. Anderson Cancer Center, Houston, TX, 2Experimental Radiation Oncology, U.T. M. D. Anderson Cancer Center, Houston, TX Purpose/Objective: Predicting the radiosensitivity of normal and tumor cells has great potential importance for radiation oncology. The surviving fraction after 2Gy (SF2) has been shown in many instances to predict normal tissue toxicity and tumor control probability. SF2 measurements, however, are time consuming, expensive, and very lab-dependent. We hypothesized that an analysis of proteins that bind to DNA double-strand breaks could predict radiosensitivity, and would be more practical than SF2 determination. Materials/Methods: DNA end binding complexes (DNA-EBCs) were identified by mixing nuclear protein extracts with a radiolabeled oligonucleotide, in the presence of a vast excess of supercoiled DNA (without ends, to remove non-end-specific binding proteins like histones), followed by electrophoresis and autoradiography. DNA-EBCs from 21 primary fibroblast and 15 tumor cell lines with a variety of radiosensitivities (SF2 range 0.01– 0.40 and 0.30 – 0.80 respectively) were then correlated with SF2. To identify protein components of the DNA-EBCs, antibodies to DNA repair proteins were added to the DNA-EBC reaction. Super-shifting of DNA-EBCs demonstrated the presence of the specific protein. The nuclear protein levels of each band A component were determined in triplicate by western analysis. Results: Ten DNA-EBCs were identified. DNA-EBC analysis identified a rapidly migrating ATM-containing band (called: band-A) whose density correlated with SF2. The density of other bands, or total DNA-EBC binding, correlated more poorly with SF2 (r2⬍0.45). There were no differences in DNA-EBC migration pattern noted. Band-A density correlated well with SF2 (0.02 ⬍ SF2 ⬍ 0.41) in 21 primary fibroblast lines (r2⫽0.77). The DNA-EBC pattern of peripheral blood lymphocytes was identical to that of fibroblasts, and the band-A density was identical in matched pairs of lymphocytes/fibroblasts from an individual. Band-A density also correlated with SF2 (0.35 ⬍ SF2 ⬍ 0.80) in 15 human tumor cell lines (r2⫽0.91). When normal and tumor cell results are combined (Fig.), there is an excellent correlation between band-A density and Sf2 (r2⫽0.85). Mixing studies demonstrated that band-A density predicted SF2 of tumor cells even when contaminated with normal cells, or (in the case of xenografts) rodent cells. DNA-EBC analysis also predicted radiosensitization by two unrelated radiosensitizers: the COX-2 inhibitor SC-236, and the histone deacetylase inhibitor sodium butyrate. Supershift analysis revealed that band-A contains the following proteins: ATM, Ku70, DNA Ligase III, Rpa32, Rpa14, DNA ligase IV, XRCC4, WRN, BLM, RAD51 and p53. In a subset of primary cells that were relatively radiosensitive because of BRCA1 mutation, there was no correlation between the levels of any of these proteins and band-A density or SF2. Conclusions: Band-A density predicts SF2 of primary and tumor cells, and can predict radiosensitizing effects of biologic agents. The band-A density of lymphocytes is identical to that of fibroblasts from the same individual, demonstrating that band-A analysis of lymphocytes would be a good predictor of normal tissue SF2. DNA-EBC analysis may be a clinically useful test to predict toxicity and tumor control in individual patients.