- Email: [email protected]
Predictors for chronic urinary toxicity following treatment of prostate cancer with 3-d conformal radiotherapy: Dose-volume analysis of a phase II dose escalation study
Proceedings of the 46th Annual ASTRO Meeting
2171
Position of the Prostate Relative to Pelvic Bony Anatomy Based on Intraprostatic Gold Fiducial Markers and Electronic Portal Imaging
J. M. Schallenkamp,1 M.G. Herman,1 J. J. Kruse,1 T. M. Pisansky1 1
Division of Radiation Oncology, Mayo Clinic College of Medicine, Rochester, MN
Purpose/Objective: To describe the relative positions and motions of the prostate, pelvic bony anatomy, and intraprostatic gold fiducial markers based on daily on-line localization of the prostate during external beam radiotherapy (EBRT). Materials/Methods: Twenty patients with clinically localized prostate adenocarcinoma were treated with definitive EBRT to the prostate alone, or prostate and seminal vesicles, according to a daily target localization protocol using 3 or 4 intraprostatic gold fiducial markers. Daily pre-treatment electronic portal images (EPIs) were acquired for localization and through-treatment EPIs were obtained for each treatment field. The patients’ pelvic bony anatomy and each intraprostatic gold marker were identified on simulation digitally reconstructed radiographs (DRRs), and then used for matching to pre-treatment and through-treatment EPIs. The markers as a group represent the position of the prostate while the distance between them was measured to monitor migration. These data provide the basis for a quantitative inter-fractional analysis of the motion of the prostate, its position relative to the bony anatomy, and the stability of individual intraprostatic fiducial markers. Intra-fractional motion for both bony anatomy and prostate was determined by comparing results from opposing fields acquired during each fraction. Treatment planning margins were calculated in three-dimensions (3D) and along each body axis with and without using daily localization. Results: A total of 22,266 data points were obtained from daily pre-treatment and through-treatment EPIs. The pre-correction 3D average displacement for the prostate of 5.6 mm improved to 2.8 mm after using the localization protocol, which provided for beam re-alignment only if the pre-treatment 3D displacement was ⱖ5 mm. The bony anatomy 3D average displacement was 4.4 mm both before and after localization to the prostate (p ⫽ 0.46). Along the superior-inferior (SI), anterior-posterior (AP), and right-left (RL) axes, the average prostate displacement improved from 2.5, 3.7, and 1.9 mm before localization to 1.4, 1.6, and 1.1 mm after (p ⬍ 0.001 [all]). The pre- to post-correction position of the bony anatomic landmarks worsened from 1.7 to 2.5 mm (p ⬍ 0.001) in the SI axis, remained statistically unchanged at 2.8 mm (p ⫽ 0.39) in the AP axis, and improved from 2.0 to 1.2 mm in the RL axis (p ⬍ 0.001). There was no significant intrafractional displacement of the prostate position or the bony anatomic landmarks. Ninety-six inter-marker distances were followed daily. Seventy-nine percent had a standard deviation of less than 1 mm and 96% were less than 1.5 mm. The 3D uniform margin required for 99% coverage of the clinical target volume (CTV) was 4.6 mm before localization and 2.3 mm after. Individual axis margins were 3.3, 2.9, and 3.5 mm in the SI, AP, and RL axes before localization and 1.5, 1.9, 1.8 mm after localization. Conclusions: Significant interfractional motion exists for patients’ prostate and pelvic bony anatomy during EBRT. However, these structures move independently and, therefore, the pelvic bony anatomy should not be used as a surrogate for prostate motion. Fiducial markers are stable within the prostate and can allow significant margin reduction when used for on-line EPI-based localization of the prostate gland.
2172
Predictors for Chronic Urinary Toxicity Following Treatment of Prostate Cancer with 3-D Conformal Radiotherapy: Dose-Volume Analysis of a Phase II Dose Escalation Study
A. R. Harsolia,1 C. E. Vargas,1 L. L. Kestin,1 D. Yan,1 D. S. Brabbins,1 D. M. Lockman,1 J. Liang,1 G. S. Gustafson,1 P. Y. Chen,1 F. A. Vicini,1 J. W. Wong,1 A. A. Martinez1 1
Radiation Oncology, William Beaumont Hospital, Royal Oak, MI
Purpose/Objective: The relationship between chronic urinary toxicity and the dose to the volume of irradiated bladder and prostate remains unclear. We analyzed our experience treating localized prostate cancer with three-dimensional (3-D) conformal adaptive radiotherapy (RT) in our phase II dose escalation trial to identify factors predictive for chronic urinary toxicity. Materials/Methods: From 1999 –2002, 331 patients with clinical stage T1-T3 (NO, MO) prostate cancer were prospectively treated with 3-D conformal RT to a median dose of 79.7 Gy to the isocenter and 75.6 Gy to the confidence limited planning target volume (cl-PTV) in 1.8 Gy fractions. A patient specific cl-PTV was constructed based on multiple CT scans. Urethrograms were done to identify the urethral beak and bladder neck. For each case, the bladder (bladder solid) was contoured (median volume: 145 cc). The bladder wall was defined as the bladder solid with a 3-mm wall thickness (median volume: 28cc). Toxicity was quantified using the NCI Common Toxicity Criteria 2.0. Multiple clinical and dose-volume endpoints were evaluated. Median follow-up was 1.6 years. Results: The 3 year rates of grade ⱖ2 and grade ⱖ3 chronic urinary toxicity were 15% and 3.6% respectively. Thirty patients (10%) experienced grade ⱖ2 chronic urinary toxicity (8.2% urinary frequency, 3.1% urinary retention, 1.2% hemorrhagic cystitis, 0.8% urethral stricture, and 0.4% urinary incontinence) at a median interval of 1.2 years. Eight patients (2.3%) experienced grade ⱖ3 chronic urinary toxicity at a median interval of 1.1 years. One patient had grade 4 urinary toxicity. Bladder solid volume, prostate volume, cl-PTV, and the bladder wall volume were all significantly associated with urinary toxicity. The bladder wall receiving ⱖ30 Gy (V30) and ⱖ82 Gy (V82) were significant predictors for grade ⱖ3 chronic urinary toxicity and grade’3 chronic urinary retention (see Table). Grade ⱖ2 (p ⫽ 0.001) and grade ⱖ3 (p ⫽ 0.03) acute urinary toxicity were predictive of grade ⱖ2 chronic urinary toxicity. The same applied for acute grade ⱖ2 (p ⫽ 0.05) and ⱖ3 (p ⬍ 0.001) urinary toxicity being predictive for chronic grade ⱖ3 urinary toxicity. Conclusions: Bladder wall dose-volume endpoints are predictive for the development of urinary toxicity. Acute urinary toxicity is predictive for chronic urinary toxicity. Bladder wall V30 and V82 can be used to predict the risk of chronic urinary toxicity and can be used as an aid in determining optimal prostate RT dose. Our recommendations based on this data set is to attempt to limit the bladder wall V30 to ⬍30cc and the V82 to ⬍7cc.
S437
S438
I. J. Radiation Oncology
2173
● Biology ● Physics
Volume 60, Number 1, Supplement, 2004
Standard (STD) or Hyperfractionated (HFX) Conformal Radiation Therapy (CRT) in Prostate Cancer: A Prospective Trial Evaluating Toxicities and Early Biochemical Control
R. Valdagni,1,2 C. Italia,2 P. Montanaro,2 A. Lanceni,2 P. Lattuada,2 T. Magnani,3 C. Fiorino,4 A. Nahum5 1
Direzione Scientifica, Istituto Nazionale Tumori, Milan, Italy, 2Radiation Oncology, Casa di Cura San Pio X, Milan, Italy, Associazione Ricerca Oncologica Multidisciplinare, Milan, Italy, 4Medical Physics, Istituto Scientifico San Raffaele, Milan, Italy, 5Radiation Physics, University Hospital, Copenhagen, Denmark
3
Purpose/Objective: In 1993 we activated a prospective, non randomized trial in prostate cancer using CRT and 2 different fractionation regimens. The end points were to compare genito-urinary (GU) and gastro-intestinal (GI) toxicities (tox) as well as biochemical control (bRFS), utilizing a STD (2.0 Gy daily) or a HFX schedule. HFX (1.2 Gy BID) was chosen as a radiobiological method to try to reduce late tox without compromising local control (1). Materials/Methods: 370 consecutive patients (pts) entered the study in the period January 1993-January 2003. 209 were treated with STD and 161 with HFX CRT. 179 pts (87%) in the STD and 151 (94%) in the HFX were evaluable for late tox and bRFS analysis, while all 370 pts were considered for acute tox evaluation. Pre-treatment clinical variables (STD vs. HFX: median age 71 vs. 69; T1-2 64.8% vs. 60.4%; median iPSA 11.2 vs. 12 ng/ml; GPS ⱖ 7 39% vs. 40%) and treatment characteristics (pelvis irradiation 43% vs. 44%; androgen deprivation 79% vs. 71%) were not statistically different in the 2 groups. CRT consisted of a 4-field technique for prostate and/or pelvic nodes and a 5-field boost with rectal shielding. Median doses were 74 Gy and 79.2 Gy (EQD2:73.9 Gy, isoeffective for tumor control assuming ␣/ ⫽ 10) for STD and HFX patients, respectively. Median follow-up was 29.4 months (STD: 25.2 mos; HFX: 37.7 mos; p ⬍ 0.01 t-test). GU/GI acute and late tox (RTOG-EORTC scoring) and 5-year bRFS were compared in the 2 fractionation regimens using univariate (Chi-square, log-rank test) and multivariate analyses (Cox regression hazard model). Results: Acute ⱖ grade 2 GU tox was higher in the STD group (48.6% vs. 37.3% in the HFX group, p ⫽ 0.03) while no significant difference was found for acute GI tox. Late ⱖ grade 2 GU and GI tox were lower in the HFX group (5-year actuarial rate: GU: 10.1% vs. 20.3%, p ⫽ 0.05; GI: 6.0% vs. 10.6%, p ⫽ 0.18). 5-year bRFS were 70% (⫾13.8%, 95% CI) and 82.6 % (⫾7.2%) for STD and HFX, respectively (p ⫽ 0.44); a trend favouring HFX was found in the subgroup of pts who did not receive hormonal therapy (5-year bRFS: 85.9% ⫾ 12.4% vs. 63.9% ⫾ 23.8%, p ⫽ 0.15). Multivariate analysis revealed that only Pisansky risk groups and age were statistically related to bRFS but not fractionation regimens. Using the Nahum-Chapman TLCP model and prostate parameter set, which includes hypoxia, the TLCPs are approximately equal for the 2 regimens, whereas assuming ␣/ ⫽ 1.5 and no hypoxia we obtain 73% for the STD group but only 36% for the HFX group. Conclusions: As expected from radiobiology, HFX reduces GI and GU late tox. Concerning early bRFS, our clinical findings suggest that HFX is, at least, not detrimental with respect to standard fractionation when delivering an isoeffective total dose equivalent to 2 Gy/fr (considering ␣/ ⫽ 10). Whilst accepting that caution is appropriate due to the relatively short follow-up, this result seems to be inconsistent with a low ␣/ ratio for prostate cancer. The trend of a non-detrimental, possibly improved, clinical outcome with HFX is more consistent with the idea that ␣/ is relatively high and that some prostate tumours are hypoxic (2– 4), with hyperfractionation probably reducing this hypoxic proportion through increased reoxygenation. If the bRFS with HFX is genuinely higher than with STD, then a plausible explanation is that hyperfractionation reduces this hypoxic proportion through increased reoxygenation. 1. Forman JD et al. Radiother. Oncol. 1993;27:203–208 2. Nahum AE et al. Int. J. Radiat. Oncol. Biol. Phys. 2003;57:391– 491 3. Movsas B et al. Urology 2002;60:634 – 639 4. Parker C et al. Int. J. Radiat. Oncol. Biol. Phys. 58:750 –757, 2004 This work was kindly supported by Fondazione I. Monzino P01 grant.
2171
Position of the Prostate Relative to Pelvic Bony Anatomy Based on Intraprostatic Gold Fiducial Markers and Electronic Portal Imaging
J. M. Schallenkamp,1 M.G. Herman,1 J. J. Kruse,1 T. M. Pisansky1 1
Division of Radiation Oncology, Mayo Clinic College of Medicine, Rochester, MN
Purpose/Objective: To describe the relative positions and motions of the prostate, pelvic bony anatomy, and intraprostatic gold fiducial markers based on daily on-line localization of the prostate during external beam radiotherapy (EBRT). Materials/Methods: Twenty patients with clinically localized prostate adenocarcinoma were treated with definitive EBRT to the prostate alone, or prostate and seminal vesicles, according to a daily target localization protocol using 3 or 4 intraprostatic gold fiducial markers. Daily pre-treatment electronic portal images (EPIs) were acquired for localization and through-treatment EPIs were obtained for each treatment field. The patients’ pelvic bony anatomy and each intraprostatic gold marker were identified on simulation digitally reconstructed radiographs (DRRs), and then used for matching to pre-treatment and through-treatment EPIs. The markers as a group represent the position of the prostate while the distance between them was measured to monitor migration. These data provide the basis for a quantitative inter-fractional analysis of the motion of the prostate, its position relative to the bony anatomy, and the stability of individual intraprostatic fiducial markers. Intra-fractional motion for both bony anatomy and prostate was determined by comparing results from opposing fields acquired during each fraction. Treatment planning margins were calculated in three-dimensions (3D) and along each body axis with and without using daily localization. Results: A total of 22,266 data points were obtained from daily pre-treatment and through-treatment EPIs. The pre-correction 3D average displacement for the prostate of 5.6 mm improved to 2.8 mm after using the localization protocol, which provided for beam re-alignment only if the pre-treatment 3D displacement was ⱖ5 mm. The bony anatomy 3D average displacement was 4.4 mm both before and after localization to the prostate (p ⫽ 0.46). Along the superior-inferior (SI), anterior-posterior (AP), and right-left (RL) axes, the average prostate displacement improved from 2.5, 3.7, and 1.9 mm before localization to 1.4, 1.6, and 1.1 mm after (p ⬍ 0.001 [all]). The pre- to post-correction position of the bony anatomic landmarks worsened from 1.7 to 2.5 mm (p ⬍ 0.001) in the SI axis, remained statistically unchanged at 2.8 mm (p ⫽ 0.39) in the AP axis, and improved from 2.0 to 1.2 mm in the RL axis (p ⬍ 0.001). There was no significant intrafractional displacement of the prostate position or the bony anatomic landmarks. Ninety-six inter-marker distances were followed daily. Seventy-nine percent had a standard deviation of less than 1 mm and 96% were less than 1.5 mm. The 3D uniform margin required for 99% coverage of the clinical target volume (CTV) was 4.6 mm before localization and 2.3 mm after. Individual axis margins were 3.3, 2.9, and 3.5 mm in the SI, AP, and RL axes before localization and 1.5, 1.9, 1.8 mm after localization. Conclusions: Significant interfractional motion exists for patients’ prostate and pelvic bony anatomy during EBRT. However, these structures move independently and, therefore, the pelvic bony anatomy should not be used as a surrogate for prostate motion. Fiducial markers are stable within the prostate and can allow significant margin reduction when used for on-line EPI-based localization of the prostate gland.
2172
Predictors for Chronic Urinary Toxicity Following Treatment of Prostate Cancer with 3-D Conformal Radiotherapy: Dose-Volume Analysis of a Phase II Dose Escalation Study
A. R. Harsolia,1 C. E. Vargas,1 L. L. Kestin,1 D. Yan,1 D. S. Brabbins,1 D. M. Lockman,1 J. Liang,1 G. S. Gustafson,1 P. Y. Chen,1 F. A. Vicini,1 J. W. Wong,1 A. A. Martinez1 1
Radiation Oncology, William Beaumont Hospital, Royal Oak, MI
Purpose/Objective: The relationship between chronic urinary toxicity and the dose to the volume of irradiated bladder and prostate remains unclear. We analyzed our experience treating localized prostate cancer with three-dimensional (3-D) conformal adaptive radiotherapy (RT) in our phase II dose escalation trial to identify factors predictive for chronic urinary toxicity. Materials/Methods: From 1999 –2002, 331 patients with clinical stage T1-T3 (NO, MO) prostate cancer were prospectively treated with 3-D conformal RT to a median dose of 79.7 Gy to the isocenter and 75.6 Gy to the confidence limited planning target volume (cl-PTV) in 1.8 Gy fractions. A patient specific cl-PTV was constructed based on multiple CT scans. Urethrograms were done to identify the urethral beak and bladder neck. For each case, the bladder (bladder solid) was contoured (median volume: 145 cc). The bladder wall was defined as the bladder solid with a 3-mm wall thickness (median volume: 28cc). Toxicity was quantified using the NCI Common Toxicity Criteria 2.0. Multiple clinical and dose-volume endpoints were evaluated. Median follow-up was 1.6 years. Results: The 3 year rates of grade ⱖ2 and grade ⱖ3 chronic urinary toxicity were 15% and 3.6% respectively. Thirty patients (10%) experienced grade ⱖ2 chronic urinary toxicity (8.2% urinary frequency, 3.1% urinary retention, 1.2% hemorrhagic cystitis, 0.8% urethral stricture, and 0.4% urinary incontinence) at a median interval of 1.2 years. Eight patients (2.3%) experienced grade ⱖ3 chronic urinary toxicity at a median interval of 1.1 years. One patient had grade 4 urinary toxicity. Bladder solid volume, prostate volume, cl-PTV, and the bladder wall volume were all significantly associated with urinary toxicity. The bladder wall receiving ⱖ30 Gy (V30) and ⱖ82 Gy (V82) were significant predictors for grade ⱖ3 chronic urinary toxicity and grade’3 chronic urinary retention (see Table). Grade ⱖ2 (p ⫽ 0.001) and grade ⱖ3 (p ⫽ 0.03) acute urinary toxicity were predictive of grade ⱖ2 chronic urinary toxicity. The same applied for acute grade ⱖ2 (p ⫽ 0.05) and ⱖ3 (p ⬍ 0.001) urinary toxicity being predictive for chronic grade ⱖ3 urinary toxicity. Conclusions: Bladder wall dose-volume endpoints are predictive for the development of urinary toxicity. Acute urinary toxicity is predictive for chronic urinary toxicity. Bladder wall V30 and V82 can be used to predict the risk of chronic urinary toxicity and can be used as an aid in determining optimal prostate RT dose. Our recommendations based on this data set is to attempt to limit the bladder wall V30 to ⬍30cc and the V82 to ⬍7cc.
S437
S438
I. J. Radiation Oncology
2173
● Biology ● Physics
Volume 60, Number 1, Supplement, 2004
Standard (STD) or Hyperfractionated (HFX) Conformal Radiation Therapy (CRT) in Prostate Cancer: A Prospective Trial Evaluating Toxicities and Early Biochemical Control
R. Valdagni,1,2 C. Italia,2 P. Montanaro,2 A. Lanceni,2 P. Lattuada,2 T. Magnani,3 C. Fiorino,4 A. Nahum5 1
Direzione Scientifica, Istituto Nazionale Tumori, Milan, Italy, 2Radiation Oncology, Casa di Cura San Pio X, Milan, Italy, Associazione Ricerca Oncologica Multidisciplinare, Milan, Italy, 4Medical Physics, Istituto Scientifico San Raffaele, Milan, Italy, 5Radiation Physics, University Hospital, Copenhagen, Denmark
3
Purpose/Objective: In 1993 we activated a prospective, non randomized trial in prostate cancer using CRT and 2 different fractionation regimens. The end points were to compare genito-urinary (GU) and gastro-intestinal (GI) toxicities (tox) as well as biochemical control (bRFS), utilizing a STD (2.0 Gy daily) or a HFX schedule. HFX (1.2 Gy BID) was chosen as a radiobiological method to try to reduce late tox without compromising local control (1). Materials/Methods: 370 consecutive patients (pts) entered the study in the period January 1993-January 2003. 209 were treated with STD and 161 with HFX CRT. 179 pts (87%) in the STD and 151 (94%) in the HFX were evaluable for late tox and bRFS analysis, while all 370 pts were considered for acute tox evaluation. Pre-treatment clinical variables (STD vs. HFX: median age 71 vs. 69; T1-2 64.8% vs. 60.4%; median iPSA 11.2 vs. 12 ng/ml; GPS ⱖ 7 39% vs. 40%) and treatment characteristics (pelvis irradiation 43% vs. 44%; androgen deprivation 79% vs. 71%) were not statistically different in the 2 groups. CRT consisted of a 4-field technique for prostate and/or pelvic nodes and a 5-field boost with rectal shielding. Median doses were 74 Gy and 79.2 Gy (EQD2:73.9 Gy, isoeffective for tumor control assuming ␣/ ⫽ 10) for STD and HFX patients, respectively. Median follow-up was 29.4 months (STD: 25.2 mos; HFX: 37.7 mos; p ⬍ 0.01 t-test). GU/GI acute and late tox (RTOG-EORTC scoring) and 5-year bRFS were compared in the 2 fractionation regimens using univariate (Chi-square, log-rank test) and multivariate analyses (Cox regression hazard model). Results: Acute ⱖ grade 2 GU tox was higher in the STD group (48.6% vs. 37.3% in the HFX group, p ⫽ 0.03) while no significant difference was found for acute GI tox. Late ⱖ grade 2 GU and GI tox were lower in the HFX group (5-year actuarial rate: GU: 10.1% vs. 20.3%, p ⫽ 0.05; GI: 6.0% vs. 10.6%, p ⫽ 0.18). 5-year bRFS were 70% (⫾13.8%, 95% CI) and 82.6 % (⫾7.2%) for STD and HFX, respectively (p ⫽ 0.44); a trend favouring HFX was found in the subgroup of pts who did not receive hormonal therapy (5-year bRFS: 85.9% ⫾ 12.4% vs. 63.9% ⫾ 23.8%, p ⫽ 0.15). Multivariate analysis revealed that only Pisansky risk groups and age were statistically related to bRFS but not fractionation regimens. Using the Nahum-Chapman TLCP model and prostate parameter set, which includes hypoxia, the TLCPs are approximately equal for the 2 regimens, whereas assuming ␣/ ⫽ 1.5 and no hypoxia we obtain 73% for the STD group but only 36% for the HFX group. Conclusions: As expected from radiobiology, HFX reduces GI and GU late tox. Concerning early bRFS, our clinical findings suggest that HFX is, at least, not detrimental with respect to standard fractionation when delivering an isoeffective total dose equivalent to 2 Gy/fr (considering ␣/ ⫽ 10). Whilst accepting that caution is appropriate due to the relatively short follow-up, this result seems to be inconsistent with a low ␣/ ratio for prostate cancer. The trend of a non-detrimental, possibly improved, clinical outcome with HFX is more consistent with the idea that ␣/ is relatively high and that some prostate tumours are hypoxic (2– 4), with hyperfractionation probably reducing this hypoxic proportion through increased reoxygenation. If the bRFS with HFX is genuinely higher than with STD, then a plausible explanation is that hyperfractionation reduces this hypoxic proportion through increased reoxygenation. 1. Forman JD et al. Radiother. Oncol. 1993;27:203–208 2. Nahum AE et al. Int. J. Radiat. Oncol. Biol. Phys. 2003;57:391– 491 3. Movsas B et al. Urology 2002;60:634 – 639 4. Parker C et al. Int. J. Radiat. Oncol. Biol. Phys. 58:750 –757, 2004 This work was kindly supported by Fondazione I. Monzino P01 grant.