Page 145 - 2014 Printable Abstract Book
P. 145
presentation, treatment, and locoregional recurrence (LRR) patterns in breast cancer patients diagnosed
below age 40 and in those above age 75. Results: Breast cancer patients diagnosed below the age of 40
were more likely to present with a symptomatic breast mass (88% vs. 37%, p<.001) and have Stage II or III
disease (79% vs. 35%, p < 0.001) compared to women diagnosed above age 75. Younger subjects were
more likely to receive chemotherapy (p<.001) and mastectomy treatment (p<.001) than older subjects.
Women less than 40 years of age had significantly higher rates of LRR (20% vs. 7% in older patients, p=.004)
and distant metastasis (18% vs. 5% in older patients, p.003). Among those who recurred locoregionally,
younger patients developed more regional nodal recurrences than older women (44% vs. 29%), while
older patients tended to develop more in-breast and chest wall recurrences (86% vs. 64%) although
neither of these differences reached statistical significance. On multivariable analysis, patient age less
than 40 was the only significant predictor of LRR (p=.011) in a model which included chemotherapy
treatment, surgery type (mastectomy vs. lumpectomy), mammography, tumor stage, grade and receptor
status. Conclusion: Age less than 40 was the most significant predictor of LRR in our patient population.
These patients are at high risk of LRR and DM despite more aggressive treatment including chemotherapy
and mastectomy. Future studies aimed at enhancing local therapies in this patient population are
warranted.
(PS2-16) Heart disease mortality among the Atomic-bomb survivors - the Life Span Study, 1950-2008.
2
1;2
2
2
2
2
Ikuno Takahashi ; Yukiko Shimizu ; Eric J. Grant ; Ritsu Sakata ; Atsuko Sadakane ; and Kotaro Ozasa
1
Department of Clinical Studies, Radiation Effects Research Foundation, Hiroshima, Japan and department
2
of Epidemiology, Radiation Effects Research Foundation, Hiroshima, Japan
[Purpose] Results from the Life Span Study (LSS) at the Radiation Effects Research Foundation
have indicated that total body irradiation could be related with a linear dose response for deaths from
broad-based heart disease categories, but dose responses for subtype-specific heart disease deaths have
not yet been determined. The broad category of heart disease consists of different diseases with different
etiologies. The International Classification of Diseases (ICD) codes have been revised a number of times
since the initiation of the LSS; dose response analyses for subtype-specific heart disease mortality may be
affected by these changes and therefore calendar-specific analyses are required.
[Method] During the observation period of 1950-2008, atomic-bomb radiation dose-response
associations for heart disease subtypes were examined, including: ischemic heart disease (IHD),
hypertensive organ damage (HOD), valvular heart disease (VHD), and heart failure (HF), among the 86,610
LSS cohort members with DS02 individual dose estimation. Subtype-specific risks of heart diseases were
analyzed for 1950-68, 1969-80, 1981-1994, and 1995-2008, corresponding to the ICD transitions. The
outcome is mortality from heart diseases designated as the underlying cause of death in death certificates.
Poisson regression model was used to calculate excess relative risk (ERR).
[Results] Significant dose associations with HOD, VHD and HF deaths were observed in 1950-2008; no
significant evidence for departures from linearity were shown. Regarding IHD, no models were statistically
significant, but a quadratic model fit the data nominally better. For period-specific analyses, the significant
ERRs of HOD and rheumatic VHD were limited to the earliest period (1950-1968). In the latest period
(1995-2008), when death diagnoses may be most reliable due to medical development and improvement
of coding rule, elevating ERRs were suggested for non-rheumatic VHD (ERR per Gy=0.75 (95% CI; 0.02,
1.92), p=0.04). [Conclusion] Based on detailed analyses of LSS subjects, elevating risks for non-rheumatic
VHD were observed in the most recent period since 1995.
143 | P a g e
below age 40 and in those above age 75. Results: Breast cancer patients diagnosed below the age of 40
were more likely to present with a symptomatic breast mass (88% vs. 37%, p<.001) and have Stage II or III
disease (79% vs. 35%, p < 0.001) compared to women diagnosed above age 75. Younger subjects were
more likely to receive chemotherapy (p<.001) and mastectomy treatment (p<.001) than older subjects.
Women less than 40 years of age had significantly higher rates of LRR (20% vs. 7% in older patients, p=.004)
and distant metastasis (18% vs. 5% in older patients, p.003). Among those who recurred locoregionally,
younger patients developed more regional nodal recurrences than older women (44% vs. 29%), while
older patients tended to develop more in-breast and chest wall recurrences (86% vs. 64%) although
neither of these differences reached statistical significance. On multivariable analysis, patient age less
than 40 was the only significant predictor of LRR (p=.011) in a model which included chemotherapy
treatment, surgery type (mastectomy vs. lumpectomy), mammography, tumor stage, grade and receptor
status. Conclusion: Age less than 40 was the most significant predictor of LRR in our patient population.
These patients are at high risk of LRR and DM despite more aggressive treatment including chemotherapy
and mastectomy. Future studies aimed at enhancing local therapies in this patient population are
warranted.
(PS2-16) Heart disease mortality among the Atomic-bomb survivors - the Life Span Study, 1950-2008.
2
1;2
2
2
2
2
Ikuno Takahashi ; Yukiko Shimizu ; Eric J. Grant ; Ritsu Sakata ; Atsuko Sadakane ; and Kotaro Ozasa
1
Department of Clinical Studies, Radiation Effects Research Foundation, Hiroshima, Japan and department
2
of Epidemiology, Radiation Effects Research Foundation, Hiroshima, Japan
[Purpose] Results from the Life Span Study (LSS) at the Radiation Effects Research Foundation
have indicated that total body irradiation could be related with a linear dose response for deaths from
broad-based heart disease categories, but dose responses for subtype-specific heart disease deaths have
not yet been determined. The broad category of heart disease consists of different diseases with different
etiologies. The International Classification of Diseases (ICD) codes have been revised a number of times
since the initiation of the LSS; dose response analyses for subtype-specific heart disease mortality may be
affected by these changes and therefore calendar-specific analyses are required.
[Method] During the observation period of 1950-2008, atomic-bomb radiation dose-response
associations for heart disease subtypes were examined, including: ischemic heart disease (IHD),
hypertensive organ damage (HOD), valvular heart disease (VHD), and heart failure (HF), among the 86,610
LSS cohort members with DS02 individual dose estimation. Subtype-specific risks of heart diseases were
analyzed for 1950-68, 1969-80, 1981-1994, and 1995-2008, corresponding to the ICD transitions. The
outcome is mortality from heart diseases designated as the underlying cause of death in death certificates.
Poisson regression model was used to calculate excess relative risk (ERR).
[Results] Significant dose associations with HOD, VHD and HF deaths were observed in 1950-2008; no
significant evidence for departures from linearity were shown. Regarding IHD, no models were statistically
significant, but a quadratic model fit the data nominally better. For period-specific analyses, the significant
ERRs of HOD and rheumatic VHD were limited to the earliest period (1950-1968). In the latest period
(1995-2008), when death diagnoses may be most reliable due to medical development and improvement
of coding rule, elevating ERRs were suggested for non-rheumatic VHD (ERR per Gy=0.75 (95% CI; 0.02,
1.92), p=0.04). [Conclusion] Based on detailed analyses of LSS subjects, elevating risks for non-rheumatic
VHD were observed in the most recent period since 1995.
143 | P a g e
