Positioning Errors Assessed With Kilovoltage Cone-Beam CT For Online Image-Guided Prostate Radiotherapy And Its Impact On Dose Escalation

ObjectiveTo assess errors and margins associated with kVCBCT online-guided intermediate to high risk prostate cancer patients' setup and to evaluate dosimetric impact for prostate cancer as a function of the reduction of planning target volume(PTV) margin.MethodsPartⅠ:Between August 2008 and February 2009,stage T2-3N0M0 prostate cancer patients undergoing intensity modulated radiotherapy and kVCBCT at least two times weekly were enrolled in this study.After positioning the patients using the skin marks, a CBCT scan was acquired and the setup error was determined by fusing the CBCT with the pretreatment simulation CT using an auto grey-scale registration software. The patient's setup was then corrected by moving the table accordingly if necessary (≥2mm).A second CBCT scan was acquired immediately after the correction to evaluate the residual error.A PTV was derived to account for the measured residual error.The differences were statistically tested with paired student t-test.All tests were two-tailed and p value of less than 0.05 was considered significant.PartⅡ:Inverse intensity modulated treatment plans for six intermediate to high risk prostate cancer patients were evaluated using normal PTV margin(10mm around the seminal vesicles and the prostate except 5mm towards the rectum) and reduced PTV margin(data derived from PartⅠ) with dose prescriptions of 75.6Gy,79.2Gy and 82.8Gy.After the whole treatment,dose-volume histograms(DVH) were produced and tumor control probability(TCP) and normal tissue complication probabilities (NTCPs) were calculated.For all plans,the minimum dose,the mean dose and the dose coverage of PTV were quantified.In addition,the dose to the normal organs such as rectum,bladder and bilateral femoral head were compared using some dosimetric endpoints.We found the optimal dose escalation schemes by comparing these radiobiological metrics.ResultsPartⅠ:Of the 10 enrolled patients,7 were included in the analysis.From 198 kVCBCT images,the random and systematic positioning errors and calculated PTV margins without correction in mm were:a) Lateral 3.1,2.1,9.3 b) Longitudinal 1.5, 1.8,5.1 c) Vertical 4.2,3.7,13.0.In total,treatment repositioning was required in every direction on 92.8%of the fractions imaged.Initial setup error and residual error were both largest in anterior-posterior(AP) compared to left-right(LR) and superior-inferior(SI).After correction,the range of residual error is -4～3mm.The random and systematic positioning errors and calculated PTV margin without correction in mm were:a) Lateral 1.1,0.9,3.5 b) Longitudinal 0.7,1.1,2.5 c) Vertical 1.1,1.3,3.7.Both of≥3mm or 5mm displacement and 3D vector is reduced after the use of kVCBCT guided radiotherapy.PartⅡ:In this study,at doses of 75.6Gy and 79.2Gy,reducing the margin from 10/5mm to 4/3mm resulted in a higher minimum dose of PTV1;although at a dose of 82.8Gy it didn't show the significance(p=0.079),there was still the same trend.No improvement in dose coverage of PTV2 was seen.At each prescription dose with a reduced margin,the dosimetric endpoints of rectum such as D_mean,V50 and V70 were improved and NTCPrw was decreased by approximately 8%(75.6Gy) and 19% (79.2Gy and 82.8Gy),respectively.As to bladder,the dosimetric endpoints of two patients were under the dose constraint after margin reduction,while exceeding dose constraint with normal margin.D_max of bilateral femoral head was significant lower at each prescription dose with a tighter margin(p＜0.05).In general,the improvement of TR_mod was significant;however,there seemed to be large interpatient variability in the changes of TR_mod when using a tighter PTV margin.ConclusionPartⅠ:In the definitive radiotherapy of patients with intermediate and high risk prostate cancer,the initial setup error and residual error is largest in AP compared to LR and SI.Positioning error can be reduced significantly with kVCBCT guided radiotherapy.On the basis of the residual error measurements,the margin required after online CBCT correction for the patients enrolled in the study would be appoximatively 4～3mm.PartⅡ:In the definitive radiotherapy of patients with intermediate and high risk prostate cancer,PTV margin reduction in this study is effective in protection of critical normal organ such as rectum,bladder and femoral heads,while showing limited influence in dose coverage and TCP of PTV.The benefit of tighter PTV margin size can be significant for some specific patients,supporting the use of patient-specific margin and planning.The improvement of TR_mod cannot be achieved by dose escalation without reduced PTV margin.In general,the optimal TR_mod is achieved at the dose of 79.2Gy with 4/3mm PTV margin in this study.