- Email: [email protected]
The effect of high dose rate brachytherapy dwell sequence on cell survival
Proceedings of the 43rd Annual ASTRO Meeting
the accelerated and the conventional fractionation schemes. A good qualitative agreement with the laboratory findings has been observed in all cases. Consequently, the hypothesis that “advanced in silico models can substantially enhance the radiotherapy optimization process” is further strengthened. Currently, our group is developing extensions of the presented algorithms so that efficient descriptions of the corresponding in vivo cases are achieved. Selected Model Assumptions: 1. A cytokinetic model, according to which a tumor cell when cycling passes through the phases G1, S, G2, and Mitosis, is considered. After mitosis is completed, each one of the resultant cells re-enters G1 if oxygen and nutrient supply in its current position is adequate. Otherwise, it enters the resting G0 phase. It can stay there for a limited time TG0 if oxygen and nutrient supply are inadequate. Subsequently, it enters the necrotic phase leading to cell death unless the local environment becomes adequate before the expiration of TG0. In the latter case, the cell re-enters G1. In addition to the previously described pathway, there is always a probability that the cell dies due to ageing, damage and apoptosis. 2. If the minimum distance between a proliferating (cycling) tumor cell and the nutrient medium becomes greater than three cell layers, the tumor cell enters the G0 phase and subsequently undergoes necrosis and lysis. 3. Cells in any cell cycle phase are more radiosensitive than hypoxic cells residing in G0. Cells in the S phase are more radioresistant than cells in any other cycle phase G1, G2 and M). Three different sets of values for the and parameters of the LQ model are assumed: one set for the proliferating cell cycle phases except the S phase, a second one for the S phase and a third one for the resting G0 phase. 4. The response of each cell to irradiation is described by the linear-quadratic model. 5. A three-dimensional mesh quantizing the volume occupied by the cell culture is used. Each geometrical cell of the mesh can be occupied by a single tumor cell, nutrient medium or products of cell death. 6. Tumor expansion is computationally achieved by shifting a cell chain from the newly occupied mesh cell towards the external environment of the tumor. The opposite rule has been introduced for tumor shrinkage. 7. Time is quantized and measured in h. The Case of SCLC In Vitro: The simulation model has been tested by considering the case of SCLC spheroids. Both the accelerated (2x2 Gy/d x 5d/wk, TD⫽60 Gy) and the conventional (1x2 Gy/d, 5 d/wk, TD⫽60 Gy) fractionation schemes have been simulated. A good qualitative agreement with laboratory experience has been observed. Quantitative improvements of the model are currently under investigation
2032
Biodosimetric Assessment of Micronuclei in Lymphocytes of Patients with Pelvic Radiotherapy
1
T.K. Lee , K.F. O’Brien2, R.R. Allison1, J.L. Naves1, A.L. Wiley3, U.L. Karlsson1 1 Radiation Oncology, Brody Sch. Med. at ECU, Greenville, NC, 2Biostatistics, ECU School of Allied Health, Greenville, NC, 3Radiation Oncology, The Watson Clinic, Lakeland, FL To further verify the applicability of the micronuclei (MN) assay for clinical biodosimetry, we measured the MN yield in cytokinesis-blocked (CB) peripheral blood lymphocytes (PBL) of 12 patients with primary cancers of either prostate (n⫽9) or testis (n⫽3). These patients had no previous chemotherapy or radiotherapy (xRT). They were treated with standardized schemes of fractionated pelvic xRT. Before xRT, and at 1-3 random time-point during the course of xRT, blood samples were collected from each patient for the following purposes: (1) to assess the impact of pelvic xRT on the MN yield; and (2) to verify the relationship between the MN yield and the estimated equivalent (EQ) total-body absorbed dose. The number of xRT fractions,cumulative tumor doses, and EQ total-body absorbed doses of these patients represented in a wide range. We found that (1) before xRT, the mean(SD) yield of MN baseline level was 13.6 (6.9) per 1000 binucleated PBL. However, during the course of xRT (3-37 fractions), it increased to 84.3 (59.3). The ratio of relative increment of MN yield ranged from 0.9-17.0 folds above that of the respective baseline levels in each patient; and (2) MN yield (Y) increased with the estimated EQ dose (D) as Y⫽7.8⫹13.6D (r⫽0.8, p⫽0.0001). The distributions of MN yield were overdispersed in 11 patients, relative to dispersion expected under a Poisson model. Our findings indicate that the MN yield in CB PBL of cancer patients offers a reliable biodosimeter for EQ total-body absorbed radiation dose estimation.
2033
The Effect of High Dose Rate Brachytherapy Dwell Sequence on Cell Survival
M.R. Arnfield, P.S. Lin, M.A. Manning, D.W. Arthur, B.D. Kavanagh, R.D. Zwicker, D.A. Buck, R.K. Schmidt-Ullrich Department of Radiation Oncology, Medical College of Virginia Hospitals, Richmond, VA Purpose: During a high dose rate (HDR) brachytherapy treatment, as the source steps through different dwell positions the dose rate at any fixed point within the implant varies, since the distance between the point and the source continually changes. The instantaneous dose rate may vary by a factor of 100 or more, in a complex dwell position sequence. Two different points which receive the same total dose may have received that dose with a very different sequence of dose rates. Any effects due to the complex changes in dose rate, including the sequence of dose delivery, are ignored. We investigated the effects of dwell sequence in which dose is delivered at two different dose rates, representative of dose rates which occur during an HDR treatment. Materials and Methods: V-79 Chinese hamster cells were cultured and suspended in a 1.0 cm3 microfuge tube to serve as the target. An acrylic phantom device was fabricated with two fixed source dwell positions near and far from the target, to provide high (⬎12 Gy/hr) and intermediate (2-12 Gy/hr) dose rates. The radiation doses were delivered using an afterloading HDR Ir-192 device (Nucletron) while cells were maintained at 37o in an immersion bath. The clonogenicity of V-79 cells was measured following exposures to irradiation sequences consisting of either a high dose rate component followed by an intermediate dose rate component (H-I arm), or the reverse (I-H arm). In either arm the total dose and the dose ratio were the same, only the order in which the high or intermediate dose rate components of the dose were delivered was changed. Results: Results from one representative experiment are shown in Table 1. When the intermediate dose rate component was
241
242
I. J. Radiation Oncology
● Biology ● Physics
Volume 51, Number 3, Supplement 1, 2001
given before the high dose rate component, there was increased survival. All data pairs from three experiments showed greater survival for the I-H arm than the H-I arm by amounts ranging from 4% to 24%. Simple linear quadratic models such as the Lea-Catchside model, which is invariant to time reversal of irradiation sequence, do not predict these results. Conclusion: These results suggest that targets receiving the same total dose of radiation during an HDR implant may not experience the same biological effect. This may be related to induced radioresistance and/or sub-lethal damage repair. Potential strategies to compensate for this effect include randomization of dwell position sequences. Cumulative wtd. avg. dose (Gy)
HDR wtd. avg. dose (Gy)
IDR wtd. Avg. dose (Gy)
SurvivalHI
SurvivalIH
p-value
1.7 3.4 5.1 6.8 8.5
0.72 1.45 2.18 2.90 3.62
0.97 1.94 2.91 3.88 4.85
0.694⫾0.067 0.648⫾0.061 0.473⫾0.055 0.408⫾0.044 0.251⫾0.028
0.747⫾0.062 0.678⫾0.108 0.505⫾0.074 0.446⫾0.046 0.289⫾0.052
0.03 0.55 0.16 0.10 0.05
2034
A Prospective Comparison of Breast Pain in Patients Participating in a Randomized Trial of Breast Conserving Surgery and Tamoxifen with or without Radiation Therapy
G. Rayan1, K.A. Vallis1,2, L.A. Dawson3, A. Bezjak1,2, A.W. Fyles1,2, A. Lau1, Q. Yi1 1 Radiation Oncology, Princess Margaret Hospital, Toronto, ON, Canada, 2University of Toronto, Toronto, ON, Canada, 3 University of Michigan Medical Center, Ann Arbor, MI Purpose: To determine the incidence and severity of breast pain in women treated with breast conserving surgery (BCS) and tamoxifen (TAM) with and without radiation therapy (RT) and to determine whether breast pain affects quality of life (QoL). Materials and Methods: A randomized clinical trial (RCT) of BCS ⫹ TAM with or without breast RT was carried out at the Princess Margaret Hospital between 1993 and 1999. The RCT was designed to evaluate the need for RT in addition to TAM in women ⱖ 50 yr treated with BCS (with negative excision margins) for T1-2, node negative breast cancer. A companion study to assess breast pain was carried out during the last 2 years of accrual to the RCT. Participants in the breast pain study completed questionnaires within one week of entry into the RCT (baseline) and at 3, 6 and 12 months. The questionnaires used were the short-form McGill pain questionnaire (SF-MPQ), the European Organisation for Research and Treatment of Cancer (EORTC) quality of life questionnaire (QLQ-C30) and the EORTC breast cancer module questionnaire (QLQ-BR23). Medical records were reviewed to obtain demographic, pathologic and surgical data. The SF-MPQ pain score was generated by summarizing intensity values for all descriptors and the EORTC QLQ-C30 and QLQ-BR23 were scored according to the EORTC scoring manual. Analysis was performed according to an equivalency design. Means and their difference between the two treatment groups were calculated for the SF-MPQ pain score and for each functional and symptom scale of QLQ-C30 and QLQ-BR23. Acute radiation toxicity was scored according to the RTOG/ EORTC system. Results: Eighty-six patients participated in the breast pain study; 41 received RT ⫹ TAM and 45 received TAM alone. The median age was 70 yr (range, 51-80 yr). There was no statistically significant difference between the groups with respect to T-stage, histologic type, number of surgical procedures or complications. Baseline pain and QoL scores were similar for the two groups. At 3 months, patients receiving RT experienced more breast pain compared to those receiving TAM alone but this did not reach statistical significance. At 3 months the mean SF-MPQ pain scores for the RT ⫹ TAM and TAM groups were 2.39 and 1.83 respectively, with a difference of 0.56 (95% CI, -0.95-2.05) (p⫽0.47). At 12 months, pain scores were lower in both groups with a difference of 0.20 (95% CI, 0.86-1.27) (p⫽0.71). There was no statistically significant difference between the treatment groups with respect to mean scores for the functional or symptom scales of the QLQ-C30 or QLQ-BR23 questionnaires at baseline, 3, 6 or 12 months. The global health status / QoL scale was very similar for the two groups at each time point. Acute radiation toxicity scores did not correlate with breast pain or QoL scores at 12 months. Conclusion: These results suggest that breast RT does not significantly contribute to breast pain or adversely impact QoL up to 12 months following treatment.
2035
Investigating HER2/neu Status as a Predictor of Outcome in Stage I/II Breast Cancer Following Breast Conservation Therapy
J.L. Rembert1, E.E. Harris1, D.J. Schultz2, M. Feldman3, L.J. Solin1 1 Radiation Oncology, University of Pennsylvania, Philadelphia, PA, 2Mathematics, Millersville University, Millersville, PA, 3 Pathology, University of Pennsylvania, Philadelphia, PA Purpose: HER2/neu status has been a useful factor in selecting systemic breast cancer therapies. However, its role relative to radiation therapy is unknown. The purpose of our study was to evaluate HER2/neu status as a predictive factor for outcome following breast conservation surgery with definitive radiation therapy in early stage breast cancer. Materials and Methods: A total of 353 women with a diagnosis of stage I/II primary unilateral breast carcinoma were treated with definitive radiotherapy after breast conservation surgery between 1990-1995 at the Hospital of the University of Pennsylvania. Of these, 150 patients (42%) had a known HER2/neu status, and in the majority of the remaining patients, their tissue was not tested for HER2/neu expression. Of the patients who were tested, 43 (29%) were positive. All positive tumors were re-stained using the recently FDA approved HercepTestTM (Dako Corp, Carpinteria, CA). Median follow-up was 4.4 years(range: ⬍1-10.8 years). Results: HER2/neu status did not correlate with pathologic tumor stage or nodal status, with T2 tumors found in 21% of HER2 positive patients and 18% of HER2/neu negative patients (p⫽0.25). HER2/neu was overexpressed in 21% of node positive patients and 20% of node negative patients (p⫽0.86). HER2/neu status did not correlate with histologic subtype (p⫽0.49),
the accelerated and the conventional fractionation schemes. A good qualitative agreement with the laboratory findings has been observed in all cases. Consequently, the hypothesis that “advanced in silico models can substantially enhance the radiotherapy optimization process” is further strengthened. Currently, our group is developing extensions of the presented algorithms so that efficient descriptions of the corresponding in vivo cases are achieved. Selected Model Assumptions: 1. A cytokinetic model, according to which a tumor cell when cycling passes through the phases G1, S, G2, and Mitosis, is considered. After mitosis is completed, each one of the resultant cells re-enters G1 if oxygen and nutrient supply in its current position is adequate. Otherwise, it enters the resting G0 phase. It can stay there for a limited time TG0 if oxygen and nutrient supply are inadequate. Subsequently, it enters the necrotic phase leading to cell death unless the local environment becomes adequate before the expiration of TG0. In the latter case, the cell re-enters G1. In addition to the previously described pathway, there is always a probability that the cell dies due to ageing, damage and apoptosis. 2. If the minimum distance between a proliferating (cycling) tumor cell and the nutrient medium becomes greater than three cell layers, the tumor cell enters the G0 phase and subsequently undergoes necrosis and lysis. 3. Cells in any cell cycle phase are more radiosensitive than hypoxic cells residing in G0. Cells in the S phase are more radioresistant than cells in any other cycle phase G1, G2 and M). Three different sets of values for the and parameters of the LQ model are assumed: one set for the proliferating cell cycle phases except the S phase, a second one for the S phase and a third one for the resting G0 phase. 4. The response of each cell to irradiation is described by the linear-quadratic model. 5. A three-dimensional mesh quantizing the volume occupied by the cell culture is used. Each geometrical cell of the mesh can be occupied by a single tumor cell, nutrient medium or products of cell death. 6. Tumor expansion is computationally achieved by shifting a cell chain from the newly occupied mesh cell towards the external environment of the tumor. The opposite rule has been introduced for tumor shrinkage. 7. Time is quantized and measured in h. The Case of SCLC In Vitro: The simulation model has been tested by considering the case of SCLC spheroids. Both the accelerated (2x2 Gy/d x 5d/wk, TD⫽60 Gy) and the conventional (1x2 Gy/d, 5 d/wk, TD⫽60 Gy) fractionation schemes have been simulated. A good qualitative agreement with laboratory experience has been observed. Quantitative improvements of the model are currently under investigation
2032
Biodosimetric Assessment of Micronuclei in Lymphocytes of Patients with Pelvic Radiotherapy
1
T.K. Lee , K.F. O’Brien2, R.R. Allison1, J.L. Naves1, A.L. Wiley3, U.L. Karlsson1 1 Radiation Oncology, Brody Sch. Med. at ECU, Greenville, NC, 2Biostatistics, ECU School of Allied Health, Greenville, NC, 3Radiation Oncology, The Watson Clinic, Lakeland, FL To further verify the applicability of the micronuclei (MN) assay for clinical biodosimetry, we measured the MN yield in cytokinesis-blocked (CB) peripheral blood lymphocytes (PBL) of 12 patients with primary cancers of either prostate (n⫽9) or testis (n⫽3). These patients had no previous chemotherapy or radiotherapy (xRT). They were treated with standardized schemes of fractionated pelvic xRT. Before xRT, and at 1-3 random time-point during the course of xRT, blood samples were collected from each patient for the following purposes: (1) to assess the impact of pelvic xRT on the MN yield; and (2) to verify the relationship between the MN yield and the estimated equivalent (EQ) total-body absorbed dose. The number of xRT fractions,cumulative tumor doses, and EQ total-body absorbed doses of these patients represented in a wide range. We found that (1) before xRT, the mean(SD) yield of MN baseline level was 13.6 (6.9) per 1000 binucleated PBL. However, during the course of xRT (3-37 fractions), it increased to 84.3 (59.3). The ratio of relative increment of MN yield ranged from 0.9-17.0 folds above that of the respective baseline levels in each patient; and (2) MN yield (Y) increased with the estimated EQ dose (D) as Y⫽7.8⫹13.6D (r⫽0.8, p⫽0.0001). The distributions of MN yield were overdispersed in 11 patients, relative to dispersion expected under a Poisson model. Our findings indicate that the MN yield in CB PBL of cancer patients offers a reliable biodosimeter for EQ total-body absorbed radiation dose estimation.
2033
The Effect of High Dose Rate Brachytherapy Dwell Sequence on Cell Survival
M.R. Arnfield, P.S. Lin, M.A. Manning, D.W. Arthur, B.D. Kavanagh, R.D. Zwicker, D.A. Buck, R.K. Schmidt-Ullrich Department of Radiation Oncology, Medical College of Virginia Hospitals, Richmond, VA Purpose: During a high dose rate (HDR) brachytherapy treatment, as the source steps through different dwell positions the dose rate at any fixed point within the implant varies, since the distance between the point and the source continually changes. The instantaneous dose rate may vary by a factor of 100 or more, in a complex dwell position sequence. Two different points which receive the same total dose may have received that dose with a very different sequence of dose rates. Any effects due to the complex changes in dose rate, including the sequence of dose delivery, are ignored. We investigated the effects of dwell sequence in which dose is delivered at two different dose rates, representative of dose rates which occur during an HDR treatment. Materials and Methods: V-79 Chinese hamster cells were cultured and suspended in a 1.0 cm3 microfuge tube to serve as the target. An acrylic phantom device was fabricated with two fixed source dwell positions near and far from the target, to provide high (⬎12 Gy/hr) and intermediate (2-12 Gy/hr) dose rates. The radiation doses were delivered using an afterloading HDR Ir-192 device (Nucletron) while cells were maintained at 37o in an immersion bath. The clonogenicity of V-79 cells was measured following exposures to irradiation sequences consisting of either a high dose rate component followed by an intermediate dose rate component (H-I arm), or the reverse (I-H arm). In either arm the total dose and the dose ratio were the same, only the order in which the high or intermediate dose rate components of the dose were delivered was changed. Results: Results from one representative experiment are shown in Table 1. When the intermediate dose rate component was
241
242
I. J. Radiation Oncology
● Biology ● Physics
Volume 51, Number 3, Supplement 1, 2001
given before the high dose rate component, there was increased survival. All data pairs from three experiments showed greater survival for the I-H arm than the H-I arm by amounts ranging from 4% to 24%. Simple linear quadratic models such as the Lea-Catchside model, which is invariant to time reversal of irradiation sequence, do not predict these results. Conclusion: These results suggest that targets receiving the same total dose of radiation during an HDR implant may not experience the same biological effect. This may be related to induced radioresistance and/or sub-lethal damage repair. Potential strategies to compensate for this effect include randomization of dwell position sequences. Cumulative wtd. avg. dose (Gy)
HDR wtd. avg. dose (Gy)
IDR wtd. Avg. dose (Gy)
SurvivalHI
SurvivalIH
p-value
1.7 3.4 5.1 6.8 8.5
0.72 1.45 2.18 2.90 3.62
0.97 1.94 2.91 3.88 4.85
0.694⫾0.067 0.648⫾0.061 0.473⫾0.055 0.408⫾0.044 0.251⫾0.028
0.747⫾0.062 0.678⫾0.108 0.505⫾0.074 0.446⫾0.046 0.289⫾0.052
0.03 0.55 0.16 0.10 0.05
2034
A Prospective Comparison of Breast Pain in Patients Participating in a Randomized Trial of Breast Conserving Surgery and Tamoxifen with or without Radiation Therapy
G. Rayan1, K.A. Vallis1,2, L.A. Dawson3, A. Bezjak1,2, A.W. Fyles1,2, A. Lau1, Q. Yi1 1 Radiation Oncology, Princess Margaret Hospital, Toronto, ON, Canada, 2University of Toronto, Toronto, ON, Canada, 3 University of Michigan Medical Center, Ann Arbor, MI Purpose: To determine the incidence and severity of breast pain in women treated with breast conserving surgery (BCS) and tamoxifen (TAM) with and without radiation therapy (RT) and to determine whether breast pain affects quality of life (QoL). Materials and Methods: A randomized clinical trial (RCT) of BCS ⫹ TAM with or without breast RT was carried out at the Princess Margaret Hospital between 1993 and 1999. The RCT was designed to evaluate the need for RT in addition to TAM in women ⱖ 50 yr treated with BCS (with negative excision margins) for T1-2, node negative breast cancer. A companion study to assess breast pain was carried out during the last 2 years of accrual to the RCT. Participants in the breast pain study completed questionnaires within one week of entry into the RCT (baseline) and at 3, 6 and 12 months. The questionnaires used were the short-form McGill pain questionnaire (SF-MPQ), the European Organisation for Research and Treatment of Cancer (EORTC) quality of life questionnaire (QLQ-C30) and the EORTC breast cancer module questionnaire (QLQ-BR23). Medical records were reviewed to obtain demographic, pathologic and surgical data. The SF-MPQ pain score was generated by summarizing intensity values for all descriptors and the EORTC QLQ-C30 and QLQ-BR23 were scored according to the EORTC scoring manual. Analysis was performed according to an equivalency design. Means and their difference between the two treatment groups were calculated for the SF-MPQ pain score and for each functional and symptom scale of QLQ-C30 and QLQ-BR23. Acute radiation toxicity was scored according to the RTOG/ EORTC system. Results: Eighty-six patients participated in the breast pain study; 41 received RT ⫹ TAM and 45 received TAM alone. The median age was 70 yr (range, 51-80 yr). There was no statistically significant difference between the groups with respect to T-stage, histologic type, number of surgical procedures or complications. Baseline pain and QoL scores were similar for the two groups. At 3 months, patients receiving RT experienced more breast pain compared to those receiving TAM alone but this did not reach statistical significance. At 3 months the mean SF-MPQ pain scores for the RT ⫹ TAM and TAM groups were 2.39 and 1.83 respectively, with a difference of 0.56 (95% CI, -0.95-2.05) (p⫽0.47). At 12 months, pain scores were lower in both groups with a difference of 0.20 (95% CI, 0.86-1.27) (p⫽0.71). There was no statistically significant difference between the treatment groups with respect to mean scores for the functional or symptom scales of the QLQ-C30 or QLQ-BR23 questionnaires at baseline, 3, 6 or 12 months. The global health status / QoL scale was very similar for the two groups at each time point. Acute radiation toxicity scores did not correlate with breast pain or QoL scores at 12 months. Conclusion: These results suggest that breast RT does not significantly contribute to breast pain or adversely impact QoL up to 12 months following treatment.
2035
Investigating HER2/neu Status as a Predictor of Outcome in Stage I/II Breast Cancer Following Breast Conservation Therapy
J.L. Rembert1, E.E. Harris1, D.J. Schultz2, M. Feldman3, L.J. Solin1 1 Radiation Oncology, University of Pennsylvania, Philadelphia, PA, 2Mathematics, Millersville University, Millersville, PA, 3 Pathology, University of Pennsylvania, Philadelphia, PA Purpose: HER2/neu status has been a useful factor in selecting systemic breast cancer therapies. However, its role relative to radiation therapy is unknown. The purpose of our study was to evaluate HER2/neu status as a predictive factor for outcome following breast conservation surgery with definitive radiation therapy in early stage breast cancer. Materials and Methods: A total of 353 women with a diagnosis of stage I/II primary unilateral breast carcinoma were treated with definitive radiotherapy after breast conservation surgery between 1990-1995 at the Hospital of the University of Pennsylvania. Of these, 150 patients (42%) had a known HER2/neu status, and in the majority of the remaining patients, their tissue was not tested for HER2/neu expression. Of the patients who were tested, 43 (29%) were positive. All positive tumors were re-stained using the recently FDA approved HercepTestTM (Dako Corp, Carpinteria, CA). Median follow-up was 4.4 years(range: ⬍1-10.8 years). Results: HER2/neu status did not correlate with pathologic tumor stage or nodal status, with T2 tumors found in 21% of HER2 positive patients and 18% of HER2/neu negative patients (p⫽0.25). HER2/neu was overexpressed in 21% of node positive patients and 20% of node negative patients (p⫽0.86). HER2/neu status did not correlate with histologic subtype (p⫽0.49),