- Email: [email protected]
151 Complications and patient satisfaction following breast implant reconstruction with and without radiotherapy
Proceedings
of the 41st Annual
ASTRO
Meeting
225
3. Tengs, T. et al., “Five-hundred life-saving interventions and their cost-effectiveness.” Risk Analysis, 15(3), 369.390, 1995. 4. Laupacis, A. et al., “How attractive does a new technology have to be to warrant adoption and utilization? Tentative guidelines for using clinical and economic evaluations.” CMAJ, 146(4), 473-481, 1992.
150
CAN COSTS BE MEASURED AND PREDICTED BY MODELLING CLINICAL TRIALS GROUP? ECONOMIC METHODOLOGICAL THERAPY ONCOLOGY GROUP (RTOG) STUDIES 90-03 AND
Owen
JB’,
Grigsby
PW’,
Caldwell
TM’,
Konski
A3, Johnson
D4, Demas
WITHIN A COOPERATIVE PILOT STUDIES OF RADIATION 91-04
W5, Movsas
B6, Jones CV’,
Wasserman
TZ
American College of Radiology, Philadelphia, PA, USA’; Washington University, St. Louis, MO, USA’; Toledo Radiological Associates, Toledo, OH, USA’; University of South Florida, Tampa, FL, USA4; Akron City Hospital, Akron, OH, USA5; Fox Chase Cancer Center, Philadelphia, PA, USA6; Radiological Associates of Sacramento, Sacramento, CA, USA’ Purpose/Objective: The aims are: 1) to measure radiation therapy treatment costs for patients treated in different arms of two randomized controlled clinical trials, 2) to compare the measured costs to those predicted by modeling the procedures expected based on the protocol, 3) to examine the distribution of costs among patients treated on the same arm for each arm of the two clinical trials, 4) to assess the feasibility of collecting economic data within randomized clinical trials conducted by a multi-center clinical trial cooperative group. Materials and Methods: RTOG 90-03 is a phase III study with four arms comparing radiation therapy treatment by standard fractionation, hyperhactionation, accelerated fractionation with a split, and accelerated fractionation with concomitant boost in patients with squamous cell carcinomas of the head and neck. RTOG 91-04 is a phase IlI study with two arms comparing accelerated hyperfractionation with standard fractionation in patients with brain metastases. The primary endpoints of both studies are survival, local control, and toxicity. The RTOG began its first economic pilot studies to measure resource costs as companion studies to 90-03 and 91-04. For each arm of each protocol expected quantities of Physicians’ Current Procedural Terminology (CPT) codes and their associated relative value units (RVU) were modeled, based on the treatment specified in the protocol. The institutions that had entered the largest number of patients on these studies were asked to provide data retrospectively on procedure codes charged to each case (including the CPT codes), quantities, and components. For each institution that submitted economic data, CPT codes and quantities for each patient were converted to the RVUs used for Medicare payments and these quantities were used to measure resource use. Total RVUs were calculated for each case. Cases were excluded if the radiation therapy quality control review judged them not to have been treated per protocol or with minor variation. Cases that were judged in economic data quality review to have incomplete economic data submitted were also excluded. Data analyses included the distribution of total RVUs per patient by study and by arm. Problems the institutions encountered in attempting to collect the requested data were recorded to assess the feasibility of conducting economic studies within a clinical bials cooperative group study. Results: Table 1 shows the relative value units per patient in each arm of the two studies. within the range predicted by the model for all arms of 90-03. The median and mean RVUs the predicted range. The distribution was narrow for both studies.
The median and mean RVUs were for both arms of 91-04 were above
Conclusion: The model was good at predicting resource use for patients who completed treatment per protocol, showing that modeling offers a reasonable approach to many components of economic studies. Actual economic data can be collected for critical cost items. Although some institutions experienced difficulty collecting retrospective data, this was largely due to the inability of many hospital computers to reconstruct analyzable data. It is feasible for many institutions to colleclt data this way; however, prospective collection of data is likely to allow a larger number of institutions groupwide to participate in future RTOG economic studies.
Study Arm Model Tl-^>:^c--
r‘-Lc”1cu”‘1
Median Mean Std. Deviation # Patients # Institutions
i 5 1
1 ( ,
90-03 1
1
205-269
1
I 1
2
3
322-395
I
208 217 63 20 4
COMPLICATIONS RECONSTRUCTION
Krueger EA’, Wilkins University of Michigan,
Total Relative Value Units per Patient 1 90-03 1 90-03 )
90-03
1
226-300
I
339 346 66 19 4
AND PATIENT WITH AND
EG’, Strawderman Ann Arbor, MI,
1
225-289
I
232 233 56 24 5
91-04
I
1
I
1 I
163-177
I
/
I
105 124 61 24 5
BREAST
91-04 2
54
236 245 49 25 5
SATISFACTION FOLLOWING WITHOUT RADIOTHERAPY
M i, Cederna USA’; William
1
4
1
188 203 75 28 5
IMPLANT
P’, Goldfarb S], Vicini FA’, Pierce LJ’ Beaumont Hospital, Royal Oak, MI, USA’
Purpose: To prospectively compare the rates of complications and patient satisfaction with mastectomy and a tissue expander/implant with and without radiotherapy.
among
breast
cancer
patients
treated
Materials and Methods: As part of the Michigan Breast Reconstruction Outcome Study (MBROS), breast cancer patients undergoing mastectomy with reconstruction were prospectively evaluated with respect to complications, general patient satisfaction with reconstruction, and aesthetic satisfaction. Included in this study were a cohort of women who chose reconstruction using an expander/implant. A subset of these patients received radiotherapy either before or after reconstmction. At one and two years post-operatively, a survey was administered which included seven items assessing both general
I. J. Radiation
226
Oncology
l
Biology
l
Physics
Volume
45, Number
3 Supplement
1999
satisfaction with their reconstruction and aesthetic satisfaction. Responses ranged from 1, indicating high satisfaction, to 5, reflecting low satisfaction. Only patients responding with a 1 or 2 for all of the items in the subscale were scored as “satisfied.” Complication data were also obtained at the same time points using hospital chart review. Any radiotherapy patients identified in the U of M Radiation Oncology data-base not included in the MBROS study were also included in the complication analysis. Results: Seventy-seven patients received an expander/implant reconstruction after mastectomy. Eighteen (23%) received radiation. For the radiotherapy patients, 50% received RT preceding the implant and 50% were irradiated following implant placement. The median dose delivered to the irradiated reconstructed breast, including boost, was 60 Gy (range 50.0-66.0 Gy) in 1.8 to 2.0 Gy fractions. With a median follow-up of 3 1.5 months from the date of surgery, the rates of complications were compared. Complications occurred in 72% (13/18) of the RT patients compared to 36% (21/59) in the no RT group (p=O.O06). The most common complications were infection and contracture, with infection occurring in 44% (8/18) of women with RT and 24% (14/59) without RT (p=O.13), and capsular contracture in 22% (4/18) and 10% (6/59), respectively, with and without RT (p=O.23). The rates of explantation varied significantly by group, with a 44% (S/18) explantation rate in the RT group versus 7% (4/59) in the no RT (p=O.O006). Sixty patients completed the satisfaction survey. For general satisfaction, 45% in the RT group were satisfied with their reconstruction compared to 58% in the no RT group, p=O.51. For aesthetics satisfaction, 36% of women in the RT group were pleased with their result compared to 24% without RT, p=O.46. When a multivariate logistic regression analysis was performed for the general satisfaction and aesthetics outcomes including both radiotherapy and complications, neither RT nor the rate of complications were found to significantly impact either endpoint. For general satisfaction, the odds ratio (OR) was 0.67 (CI 0.18-2.59) for RT/no RT versus 0.53 (CI 0.18-l 58) for complications/no complications; for aesthetics, the OR were 1.57 (CI 0.38-6.51) and 1.83 (CI 0.56-5.94): respectively. To offset potential bias for patients not completing the survey, we re-analyzed satisfaction data assuming “dissatisfaction” scores for surveys not completed. For general satisfaction. the OR was 0.57 (CI 0.17- 1.90) for RT/no RT and 0.41 (CI 0.15-l. 10) for complications/no complications. For aesthetics, the corresponding ratios were 1.0 (CI 0.26-3.86) and 1.3 (CI 0.42.4.29), respectively. Conclusion: Irradiated patients had a higher patients. Despite these differences, our pilot significantly different following radiotherapy in the present study and larger patient numbers in RT versus no RT patients in women who
152
RADIATION RESULTS
Rabinovitch University
EXPOSURE OF A NORTH
R, Carlsen of Coloi-ado
J, McCourt Health
TO WOMEN AMERICAN
RADlATION SURVEY
ONCOLOGISTS
DURING
PEGNANCY:
S: Pan Z
Sciences
Purpose: There is no long-term oncologists. This study evaluates
rate of expander/implant reconstruction failure and complications than non-irradiated data suggest that both general satisfaction and patient aesthetic satisfaction were not compared to patients who did not receive RT. Although statistical power was limited are needed to validate these results, this study suggests a comparable cosmetic outcome undergo successful implant reconstruction.
Center,
Denver,
CO, USA
data evaluating the effects of radiation exposure during the incidence, causes, and consequences of this exposure.
pregnancy
to female
radiation
Materials/Methods: 292 female ASTRO members in the United States and Canada were surveyed regarding individual pregnancy history: radiation exposure during pregnancy, and subsequent fetal outcome. Respondents were predominantly physicians (94%), and ranged in age from 31-76 years. Data from pregnancies carried only during residency training or postgraduate employment were evaluated, resulting in a total of 383 pregnancies carried by 182 women. Results: 21% of those surveyed reported radiation exposure during pregnancy corresponding to exposures in 25% of evaluated pregnancies. These incidents occurred between 1974-1998, with 43% of exposures occurring since l/90. Although women with a history of fetal exposure have worked in the radiation oncology field for a median of 13 years, 39% of exposures took place during residency or fellowship training. The overwhelming majority of incidents resulted from exposure to low dose-rate brachytherapy sources (86%). At the time of the exposure, 66% of women were aware of being pregnant. Factors described as contributing to activities resulting in fetal exposure included pressure to perform as a resident/fellow (38%) pregnancy unknown (27%) dose exposure of planned activity considered acceptable risk (9%), inadequate radiation safety education (6%), own carelessness (5%), lack of copractitioners to provide brachytherapy coverage (5%), and inadequate radiation safety supervision (4%). Outcomes of these 96 exposed pregnancies were 65 healthy newborns, 20 (218, 95% CI: 13.30%) spontaneous abortions, 7 (7.3%: 95% CI: 5.3.14.4%) congenital anomalies, 3 (3.1%, 95% CI: 0.65-8.90%) elective abortions, and 1 (l.O%, 95% CI: 0.03-5.70%) stillbirth; these results do not statistically vary from population norms. Children exposed in-utero currently range in age from 1 to 25 years, and none have been diagnosed with a childhood malignancy. Conclusions: Work-related radiation exposure to women radiation oncologists during pregnancy is not associated with a significantly increased risk of fetal demise, congenital birth defects or development of childhood malignancies. This conclusion, however, should be cautiously considered, since dose-monitoring data could not be confirmed by those surveyed. Residency training programs should assume greater responsibility for creating a work environment in which pregnant physicians in-training do not feel pressured to participate in activities which would result in fetal radiation exposure. Female radiation oncologists should be more vigilant in anticipating when they might be in the earliest weeks of pregnancy and avoid undesired radiation exposure risks. Information presented at meeting will be updated to include survey responses acquired after the date of abstract submission.
i 53
EFFECT REPAIR
Oldenburg Joint
NBE,
Center,for
OF RADIATION Calderwood Radiation
ON HUMAN
RADSO:
MECHANISMS
OF DOUBLE
STRANDED
DNA
SK Therapy,
Harvard
Medical
Double stranded (DS) DNA breaks result from direct breaks can lead to cell death or malignant transformation.
School,
Boston,
MA,
USA
and indirect effects of ionizing radiation. When left unrepaired, these Rad50 is a 153 kilodalton protein with ATP dependent DNA binding
of the 41st Annual
ASTRO
Meeting
225
3. Tengs, T. et al., “Five-hundred life-saving interventions and their cost-effectiveness.” Risk Analysis, 15(3), 369.390, 1995. 4. Laupacis, A. et al., “How attractive does a new technology have to be to warrant adoption and utilization? Tentative guidelines for using clinical and economic evaluations.” CMAJ, 146(4), 473-481, 1992.
150
CAN COSTS BE MEASURED AND PREDICTED BY MODELLING CLINICAL TRIALS GROUP? ECONOMIC METHODOLOGICAL THERAPY ONCOLOGY GROUP (RTOG) STUDIES 90-03 AND
Owen
JB’,
Grigsby
PW’,
Caldwell
TM’,
Konski
A3, Johnson
D4, Demas
WITHIN A COOPERATIVE PILOT STUDIES OF RADIATION 91-04
W5, Movsas
B6, Jones CV’,
Wasserman
TZ
American College of Radiology, Philadelphia, PA, USA’; Washington University, St. Louis, MO, USA’; Toledo Radiological Associates, Toledo, OH, USA’; University of South Florida, Tampa, FL, USA4; Akron City Hospital, Akron, OH, USA5; Fox Chase Cancer Center, Philadelphia, PA, USA6; Radiological Associates of Sacramento, Sacramento, CA, USA’ Purpose/Objective: The aims are: 1) to measure radiation therapy treatment costs for patients treated in different arms of two randomized controlled clinical trials, 2) to compare the measured costs to those predicted by modeling the procedures expected based on the protocol, 3) to examine the distribution of costs among patients treated on the same arm for each arm of the two clinical trials, 4) to assess the feasibility of collecting economic data within randomized clinical trials conducted by a multi-center clinical trial cooperative group. Materials and Methods: RTOG 90-03 is a phase III study with four arms comparing radiation therapy treatment by standard fractionation, hyperhactionation, accelerated fractionation with a split, and accelerated fractionation with concomitant boost in patients with squamous cell carcinomas of the head and neck. RTOG 91-04 is a phase IlI study with two arms comparing accelerated hyperfractionation with standard fractionation in patients with brain metastases. The primary endpoints of both studies are survival, local control, and toxicity. The RTOG began its first economic pilot studies to measure resource costs as companion studies to 90-03 and 91-04. For each arm of each protocol expected quantities of Physicians’ Current Procedural Terminology (CPT) codes and their associated relative value units (RVU) were modeled, based on the treatment specified in the protocol. The institutions that had entered the largest number of patients on these studies were asked to provide data retrospectively on procedure codes charged to each case (including the CPT codes), quantities, and components. For each institution that submitted economic data, CPT codes and quantities for each patient were converted to the RVUs used for Medicare payments and these quantities were used to measure resource use. Total RVUs were calculated for each case. Cases were excluded if the radiation therapy quality control review judged them not to have been treated per protocol or with minor variation. Cases that were judged in economic data quality review to have incomplete economic data submitted were also excluded. Data analyses included the distribution of total RVUs per patient by study and by arm. Problems the institutions encountered in attempting to collect the requested data were recorded to assess the feasibility of conducting economic studies within a clinical bials cooperative group study. Results: Table 1 shows the relative value units per patient in each arm of the two studies. within the range predicted by the model for all arms of 90-03. The median and mean RVUs the predicted range. The distribution was narrow for both studies.
The median and mean RVUs were for both arms of 91-04 were above
Conclusion: The model was good at predicting resource use for patients who completed treatment per protocol, showing that modeling offers a reasonable approach to many components of economic studies. Actual economic data can be collected for critical cost items. Although some institutions experienced difficulty collecting retrospective data, this was largely due to the inability of many hospital computers to reconstruct analyzable data. It is feasible for many institutions to colleclt data this way; however, prospective collection of data is likely to allow a larger number of institutions groupwide to participate in future RTOG economic studies.
Study Arm Model Tl-^>:^c--
r‘-Lc”1cu”‘1
Median Mean Std. Deviation # Patients # Institutions
i 5 1
1 ( ,
90-03 1
1
205-269
1
I 1
2
3
322-395
I
208 217 63 20 4
COMPLICATIONS RECONSTRUCTION
Krueger EA’, Wilkins University of Michigan,
Total Relative Value Units per Patient 1 90-03 1 90-03 )
90-03
1
226-300
I
339 346 66 19 4
AND PATIENT WITH AND
EG’, Strawderman Ann Arbor, MI,
1
225-289
I
232 233 56 24 5
91-04
I
1
I
1 I
163-177
I
/
I
105 124 61 24 5
BREAST
91-04 2
54
236 245 49 25 5
SATISFACTION FOLLOWING WITHOUT RADIOTHERAPY
M i, Cederna USA’; William
1
4
1
188 203 75 28 5
IMPLANT
P’, Goldfarb S], Vicini FA’, Pierce LJ’ Beaumont Hospital, Royal Oak, MI, USA’
Purpose: To prospectively compare the rates of complications and patient satisfaction with mastectomy and a tissue expander/implant with and without radiotherapy.
among
breast
cancer
patients
treated
Materials and Methods: As part of the Michigan Breast Reconstruction Outcome Study (MBROS), breast cancer patients undergoing mastectomy with reconstruction were prospectively evaluated with respect to complications, general patient satisfaction with reconstruction, and aesthetic satisfaction. Included in this study were a cohort of women who chose reconstruction using an expander/implant. A subset of these patients received radiotherapy either before or after reconstmction. At one and two years post-operatively, a survey was administered which included seven items assessing both general
I. J. Radiation
226
Oncology
l
Biology
l
Physics
Volume
45, Number
3 Supplement
1999
satisfaction with their reconstruction and aesthetic satisfaction. Responses ranged from 1, indicating high satisfaction, to 5, reflecting low satisfaction. Only patients responding with a 1 or 2 for all of the items in the subscale were scored as “satisfied.” Complication data were also obtained at the same time points using hospital chart review. Any radiotherapy patients identified in the U of M Radiation Oncology data-base not included in the MBROS study were also included in the complication analysis. Results: Seventy-seven patients received an expander/implant reconstruction after mastectomy. Eighteen (23%) received radiation. For the radiotherapy patients, 50% received RT preceding the implant and 50% were irradiated following implant placement. The median dose delivered to the irradiated reconstructed breast, including boost, was 60 Gy (range 50.0-66.0 Gy) in 1.8 to 2.0 Gy fractions. With a median follow-up of 3 1.5 months from the date of surgery, the rates of complications were compared. Complications occurred in 72% (13/18) of the RT patients compared to 36% (21/59) in the no RT group (p=O.O06). The most common complications were infection and contracture, with infection occurring in 44% (8/18) of women with RT and 24% (14/59) without RT (p=O.13), and capsular contracture in 22% (4/18) and 10% (6/59), respectively, with and without RT (p=O.23). The rates of explantation varied significantly by group, with a 44% (S/18) explantation rate in the RT group versus 7% (4/59) in the no RT (p=O.O006). Sixty patients completed the satisfaction survey. For general satisfaction, 45% in the RT group were satisfied with their reconstruction compared to 58% in the no RT group, p=O.51. For aesthetics satisfaction, 36% of women in the RT group were pleased with their result compared to 24% without RT, p=O.46. When a multivariate logistic regression analysis was performed for the general satisfaction and aesthetics outcomes including both radiotherapy and complications, neither RT nor the rate of complications were found to significantly impact either endpoint. For general satisfaction, the odds ratio (OR) was 0.67 (CI 0.18-2.59) for RT/no RT versus 0.53 (CI 0.18-l 58) for complications/no complications; for aesthetics, the OR were 1.57 (CI 0.38-6.51) and 1.83 (CI 0.56-5.94): respectively. To offset potential bias for patients not completing the survey, we re-analyzed satisfaction data assuming “dissatisfaction” scores for surveys not completed. For general satisfaction. the OR was 0.57 (CI 0.17- 1.90) for RT/no RT and 0.41 (CI 0.15-l. 10) for complications/no complications. For aesthetics, the corresponding ratios were 1.0 (CI 0.26-3.86) and 1.3 (CI 0.42.4.29), respectively. Conclusion: Irradiated patients had a higher patients. Despite these differences, our pilot significantly different following radiotherapy in the present study and larger patient numbers in RT versus no RT patients in women who
152
RADIATION RESULTS
Rabinovitch University
EXPOSURE OF A NORTH
R, Carlsen of Coloi-ado
J, McCourt Health
TO WOMEN AMERICAN
RADlATION SURVEY
ONCOLOGISTS
DURING
PEGNANCY:
S: Pan Z
Sciences
Purpose: There is no long-term oncologists. This study evaluates
rate of expander/implant reconstruction failure and complications than non-irradiated data suggest that both general satisfaction and patient aesthetic satisfaction were not compared to patients who did not receive RT. Although statistical power was limited are needed to validate these results, this study suggests a comparable cosmetic outcome undergo successful implant reconstruction.
Center,
Denver,
CO, USA
data evaluating the effects of radiation exposure during the incidence, causes, and consequences of this exposure.
pregnancy
to female
radiation
Materials/Methods: 292 female ASTRO members in the United States and Canada were surveyed regarding individual pregnancy history: radiation exposure during pregnancy, and subsequent fetal outcome. Respondents were predominantly physicians (94%), and ranged in age from 31-76 years. Data from pregnancies carried only during residency training or postgraduate employment were evaluated, resulting in a total of 383 pregnancies carried by 182 women. Results: 21% of those surveyed reported radiation exposure during pregnancy corresponding to exposures in 25% of evaluated pregnancies. These incidents occurred between 1974-1998, with 43% of exposures occurring since l/90. Although women with a history of fetal exposure have worked in the radiation oncology field for a median of 13 years, 39% of exposures took place during residency or fellowship training. The overwhelming majority of incidents resulted from exposure to low dose-rate brachytherapy sources (86%). At the time of the exposure, 66% of women were aware of being pregnant. Factors described as contributing to activities resulting in fetal exposure included pressure to perform as a resident/fellow (38%) pregnancy unknown (27%) dose exposure of planned activity considered acceptable risk (9%), inadequate radiation safety education (6%), own carelessness (5%), lack of copractitioners to provide brachytherapy coverage (5%), and inadequate radiation safety supervision (4%). Outcomes of these 96 exposed pregnancies were 65 healthy newborns, 20 (218, 95% CI: 13.30%) spontaneous abortions, 7 (7.3%: 95% CI: 5.3.14.4%) congenital anomalies, 3 (3.1%, 95% CI: 0.65-8.90%) elective abortions, and 1 (l.O%, 95% CI: 0.03-5.70%) stillbirth; these results do not statistically vary from population norms. Children exposed in-utero currently range in age from 1 to 25 years, and none have been diagnosed with a childhood malignancy. Conclusions: Work-related radiation exposure to women radiation oncologists during pregnancy is not associated with a significantly increased risk of fetal demise, congenital birth defects or development of childhood malignancies. This conclusion, however, should be cautiously considered, since dose-monitoring data could not be confirmed by those surveyed. Residency training programs should assume greater responsibility for creating a work environment in which pregnant physicians in-training do not feel pressured to participate in activities which would result in fetal radiation exposure. Female radiation oncologists should be more vigilant in anticipating when they might be in the earliest weeks of pregnancy and avoid undesired radiation exposure risks. Information presented at meeting will be updated to include survey responses acquired after the date of abstract submission.
i 53
EFFECT REPAIR
Oldenburg Joint
NBE,
Center,for
OF RADIATION Calderwood Radiation
ON HUMAN
RADSO:
MECHANISMS
OF DOUBLE
STRANDED
DNA
SK Therapy,
Harvard
Medical
Double stranded (DS) DNA breaks result from direct breaks can lead to cell death or malignant transformation.
School,
Boston,
MA,
USA
and indirect effects of ionizing radiation. When left unrepaired, these Rad50 is a 153 kilodalton protein with ATP dependent DNA binding