Volume 90 Number 1S Supplement 2014 DIR (p < 0.005). Similarly, the value between reference CT and CT of ICBT was significantly ameliorated from 0.60 to 0.87 (p < 0.005). Simple rectal D2cc ranged from 70.5Gy to 88.5, and DIR-based rectal D2cc from 65.5Gy to 87.4Gy. Ten of 14 patients (71%) had higher simple rectal D2cc than DIR-based rectal D2cc, and the mean dose using simple rectal D2cc was 1.3 Gy higher than that using DIR-based rectal D2cc. Conclusions: DIR method may allow us to track more accurate total dose from multiple treatment planning CT of radical radiotherapy for uterine cervical cancer. The clinical impact of difference between simple rectal D2cc and DIR-based rectal D2cc warrants further investigation. Scientific Abstract 2681; Table Procedures of DIR method 1. Making treatment plans 2. Replacement of CT values using in-house software to improve CT images noise 3. DIR of CT images in Velocity AI 4. DIR of dose distribution converted to EQD2 in Velocity AI 5. Integration of dose distributions of each EBRT and ICBT in Velocity AI
Author Disclosure: K. Hayashi: None. F. Isohashi: None. Y. Akino: None. N. Wakai: None. A. Koba: None. H. Yamaguchi: None. S. Baek: None. H. Ookubo: None. K. Tamari: None. O. Suzuki: None. Y. Seo: None. Y. Otani: None. I. Sumida: None. Y. Yoshioka: None. K. Ogawa: None.
2682 Early Outcomes and Dose-Volume Parameters for CT-Based Treatment Planning in Brachytherapy for Cervical Cancer With Severe Vaginal Invasion or Vaginal Cancer K. Murofushi,1 N. Kitamura,1 Y. Yagi,1 T. Kozuka,1 N. Takeshima,1 H. Sakurai,2 and M. Oguchi1; 1The Cancer Institute Hospital of Japanese Foundation for Cancer Research, Tokyo, Japan, 2Tsukuba University Hospital, Ibaraki, Japan Purpose/Objective(s): In majority of Japanese institutions, point A dose with whole pelvis irradiation (WP) plus HDR brachytherapy (BT) for cervical cancer (CC) is about 60 Gyab10. Intracavitary brachytherapy (ICT) is difficult to irradiate the sufficinent dose for the tumor with >5 mm vaginal invasion sparing organs at risk (OAR). If thickness is >5mm, interstitial brachytherapy (IBT) is recommended. We reported that early outcomes and DVH parameters for CC with severe vaginal invasion and vaginal cancer (VC) treated with CT based treatment planning in ICT or IBT. Materials/Methods: We analyzed the patients (pts) with CC involving lower half of vagina and >5mm vaginal invasion or VC recognized >5mm thickness. Twenty two of 232 pts with CC or VC treated between July 2010 and June 2013 in our institution were eligible. WP was performed to 30.6-40 Gy, center shield was 5.4-19.8 Gy. The point A dose of 20-24 Gy/ 4 fr was irradiated by HDR-ICT using tadem-cylinder. The dose of 24-30 Gy/ 4-5 fr for 85% BDIS was irradiated by HDR-IBT (42 Gy/7 fr for HDR-IBT alone). Before BT, vaginal invasion was measured by MRI. ICT was selected if vaginal invasion became 5 mm. In the case of ICT, the dose of target and OAR were calculated on planning CT at first ICT. The dose to bladder and rectum was analyzed by DVH parameters (D2cc). Total doses, including WP and the total fractioned BT, were biologically normalized to conventional 2-Gy fractions (a/b 10 Gy for target, 3 Gy for OAR). Results: Median age was 60 years old (34-78). 14 pts were treated with CC (T3a/T3b/T4 5/8/1), 8 with VC (T1/T2 2/6). Median vaginal tumor thickness was 20 mm (10-40). Five pts were treated with RT alone, 17 with CCRT. Ten pts were treated with ICT, 12 with IBT. Two were treated with IBT alone. Median treatment time was 50 days (3-64). Median follow up period was 25.7 months (8.2-43.2). Complete response was obtained 22 pts (100%). Four pts experienced recurrence (local/distant/both 1/2/1). Two pts treated with IBT plus EBRT experienced local recurrence. Two-year OS, DFS, pelvic recurrence free survival and distant metastasis free survival rates were 92.9%, 85.9%, 95.5% and 85.9%, respectively. Late rectal
Poster Viewing Abstracts S481 complications were recorded with 4 pts (Gr2/Gr3 3/1). The mean HRCTV D90 was 66.3 Gyab10 (58.6-76.5), the volume of the prescribed dose was 45.0 cc (31.0-143.6). The mean D2cc for the bladder and rectum were 68.2 Gy ab3 (40.8-86.1) and 61.7 Gy (42.4-82.0). The case with 82.0 Gyab3 (D2cc for the rectum) developed Gr3 late rectal complication. Conclusions: Excellent OS, local control and low toxicity rates were obtained, if the ICT or IBT was selected for each patient appropriately according to their vaginal tumor thickness measured by MRI before BT (threshold: 5 mm) and the dose of >60 Gyab10 was administered for HRCTV D90 according to the Japanese treatment protocol. Author Disclosure: K. Murofushi: None. N. Kitamura: None. Y. Yagi: None. T. Kozuka: None. N. Takeshima: None. H. Sakurai: None. M. Oguchi: None.
2683 Introducing PET/CT in Cervical Cancer Staging Procedures Leads to Stage Migration and Selection Bias H. Hansen,1 A. Loft,2 A. Berthelsen,1,2 I. Christensen,3 C. Høgdall,4 and S. Engelholm1; 1Rigshospitalet, Department of Oncology, Copenhagen, Denmark, 2Rigshospitalet, Department of Clinical Physiology, Nuclear Medicine & PET, Copenhagen, Denmark, 3 Rigshospitalet, The Finsen Laboratory, Copenhagen, Denmark, 4 Rigshospitalet, Department of Gynecology, Copenhagen, Denmark Purpose/Objective(s): In cancer of the uterine cervix, lymph node metastases are associated with a poor prognosis. Even so, the International Federation of Gynecology and Obstetrics (FIGO) does not take into account diagnostic results by methods such as PET/CT, as these are not readily available everywhere. Undetected lymph node metastases can lead to radiotherapy-plans that will not optimally treat the patient. Thus, any difference in the underlying prevalence of false negative scans between CT and PET/CT may be reflected in treatment outcomes. We report on treatment outcomes for the group of cervical cancer patient with no lymph node metastases, before and after the introduction of PET/CT at our department. Materials/Methods: A single institution retrospective analysis of 301 patients with a histopathological diagnose of cervical cancer receiving curative intended chemoradiotherapy. Treatment was given according to department protocol for patients without lymph node metastases based on either pre-treatment high quality CT, or -PET/CT. Specialists in radiology and nuclear medicine reviewed scans blinded. Patients were stratified into two subgroups: PET/CT and Non-PET/CT. Patient characteristics (table 1) and treatment outcomes were acquired from our intra department patient database containing information on treatments since January 2000. Scientific Abstract 2683; Table Patient characteristics (Squamous Cell Carcinoma [SCC], AdenoCarcinoma [AC], Undifferentiated [UD]) Imaging Method No. of Patients Median Age (mean) Range Stage 1B 2A 2B 3A 3B 4A 4B Histopathology SCC AC ASC UD Rare Unknown Primary Tumor Size, mm Range Performance Status (RTOG) Range Concomitant Chemo
Non-PET/CT
PET/CT
Total
P values
105 196 301 56.0 (27-86) 50.4 (23-88) 52.3 (23-88) .003 .08 14 (13%) 4 (4%) 38 (36%) 3 (3%) 44 (42%) 1 (1%) 1 (1%)
25 (13%) 15 (8%) 92 (47%) 2 (1%) 53 (27%) 7 (4%) 2 (1%)
39 (13%) 19 (6%) 130 (43%) 5 (2%) 97 (32%) 8 (3%) 3 (1%)
83 (79%) 8 (8%) 4 (4%) 4 (4%) 3 (3%) 3 (3%) 50 (10-100)
150 (77%) 24 (12%) 5 (3%) 6 (3%) 11 (6%) 50 (15-150)
.13 233 (77%) 32 (11%) 9 (3%) 10 (3%) 14 (5%) 3 (1%) 50 (10-150) .31
0 (0-4)
0 (0-2)
0 (0-4)
.56
78 (74%)
176 (90%)
254 (84%)
<.001
S482
International Journal of Radiation Oncology Biology Physics
Results: A significant difference of 19% (C.I. 16%-28%), 17% (12%24%), and 10% (6%-18%) for 5-year overall-, disease free-, and disease specific survival respectively was observed between the two patient groups. This remained significant in both univariate- and multivariate analyses of overall survival, including age, FIGO stage, performance status, BMI, and histopathology; hazard ratio 0.61 (0.42-0.90), p Z 0.012. Conclusions: Although not allowed to influence the FIGO stage, the use of PET/CT in pre-radiotherapy diagnostics proved to be a significant covariate that can lead to unreported stage migration. It needs to be taken into account when designing and reporting on clinical trials in order to avoid selection bias. Author Disclosure: H. Hansen: None. A. Loft: None. A. Berthelsen: None. I. Christensen: None. C. Høgdall: None. S. Engelholm: None.
determine if normal tissue motion in the time interval between imaging and treatment impacts the dose to critical organs. Materials/Methods: Fourteen patients enrolled from December 2012 to August 2013 on a prospective study evaluating image-based brachytherapy for cervical cancer with available 3D image reconstruction in all cardinal planes were included in this analysis. Each patient underwent a non-contrast pelvic CT and MRI following placement of the applicator and prior to initiation of pulsed-dose rate brachytherapy. The average time elapsed between CT and MRI was 87 minutes (range 43151 minutes). The external contours of the normal organs at risk (OARs), including bladder, rectum, and sigmoid, were delineated on both image sets in Oncentra brachytherapy planning system. The minimum dose delivered to the most irradiated 2 cm3 (D2cc) and 1 cm3 (D1cc) for all OARs was recorded on each image set. The D2cc volume for each OAR was then converted to a 3D structure. The hottest point within the D2cc volume was plotted on a DICOM coordinate system, with the origin of the coordinate system designated as the intersection between the tandem and the top of the ovoids or cylinder. The distance formula was used to calculate each point’s change in position between image sets. Results: Nine patients were treated with a tandem and ovoid applicator and five patients were treated with a tandem and cylinders. The average difference in D2cc bladder between the two scans was 2.4 Gy (range 0.1-6 Gy); the difference in D1cc bladder was 2.8 Gy (range 0.09-8.0 Gy). The average displacement of the hottest point within the bladder D2cc volume was 8.3 mm (range 1.2-20.0 mm). For the rectum and sigmoid, the average difference in D2cc dose was 1.8 Gy (range 0.4-7.4 Gy) and 1.0 Gy (range 0.5-3.5 Gy), while the average difference in D1cc dose was 2.1 Gy (range 0.3-8.5 Gy) and 1.1 Gy (0.07-4.2 Gy), respectively. The average displacement was 11.7 mm (range 2.9-56.4 mm) for the rectum and 11.2 mm (range 2.3-28.1 mm) for the sigmoid. For the sigmoid and bladder, there was no evidence that imaging modality impacted the OAR contour, since higher doses were equally likely to be observed on the CT or MRI image set. However, the rectal D2cc and D1cc were significantly higher on CT images as compared to MRI, suggesting that the use of different imaging modalities at each time point may limit accurate assessment of the effects of organ motion on rectal dose in this study. Conclusions: Internal organ motion impacts the dose and volume of critical structures treated during brachytherapy. Efforts to limit organ motion and reduce the time between imaging and treatment will improve accurate assessment of normal tissues doses. Author Disclosure: C.W. Swanick: None. K.O. Castle: None. M.G. Cunningham: None. S. Vedam: None. G.M. Rauch: None. A. Jhingran: None. P.J. Eifel: None. A.H. Klopp: None.
2684 Contribution of Imaging in Diagnosis of Local Recurrence After Radiation Therapy for Cervical Cancer A. Kubo, H. Ikushima, T. Kawanaka, S. Furutani, and H. Masafumi; University of Tokushima, Tokushima, Japan Purpose/Objective(s): To assess diagnostic ability and typical finding of the magnetic resonance imaging and 18F-fluorodeoxyglucose positron emission tomography for local recurrence after radiation therapy of cervical cancer. Materials/Methods: Diagnosis imaging and pathological findings of 120 patients who received radiation therapy for uterine cervical cancer were retrospectively evaluated. Twenty four patients were in stage Ib, 42 in stage II, 45 in stage III, and 9 in stage IV, respectively. Weekly CDDP was administered concurrently for 93 patients. All patients underwent MRI before and 3 months after radiation therapy, and every 6 months thereafter. FDG PET/CT was routinely performed before radiation therapy, and 3 months after radiation therapy, and when necessary thereafter. Median follow-up period was 22 (range: 1 - 58) months. Results: Twenty-three of 120 patients were diagnosed as local recurrence by imaging using MRI and FDG PET/CT, and 18 of 23 patients were finally diagnosed as local recurrence pathologically. The median time of local recurrence after completion of radiation therapy was 4 (range: 3 - 11) months. The sensitivity and specificity of MRI for local recurrence was 86% and 60%, respectively, and two patients showed false positive. Typical findings of local recurrence on MRI were re-appearance of high signal intensity on T2 weighted images at former tumor site, mass lesion with peripheral enhancement on dynamic contrast enhanced images, and high intensity lesion on diffusion weighted images. One of two false positive findings on MRI was a persistent high signal intensity on T2 weighted images and another is an abnormal early enhancement on dynamic contrast enhanced images. Necrosis at the central tumor site and neovascularization at peripheral tumor site is typical pathological findings of local recurrence of cervical cancer, and it was shown as ring-like enhancement of the tumor on dynamic contrast enhanced images of MRI. Some patients showed false negative on 18FDG-PET/CT, and its sensitivity was 73%. However, no false positive was found on 18FDG-PET/CT, and specificity was 100%. Conclusions: MRI has high sensitivity in diagnosis for local recurrence after radiation therapy for cervical cancer, and 18FDG-PET/CT contribute to exempt false positive findings on MRI. Author Disclosure: A. Kubo: None. H. Ikushima: None. T. Kawanaka: None. S. Furutani: None. H. Masafumi: None.
2685 Impact of Normal Tissue Motion on Dose to Critical Structures During Intracavitary Brachytherapy C.W. Swanick, K.O. Castle, M.G. Cunningham, S. Vedam, G.M. Rauch, A. Jhingran, P.J. Eifel, and A.H. Klopp; M.D. Anderson Cancer Center, Houston, TX Purpose/Objective(s): Accurate assessment of normal tissue doses is critical in 3D image-based brachytherapy. The purpose of this study was to
2686 Simultaneous Integrated Boost to Pelvic and Para-Aortic Nodes From Cervical Cancer Improves the Dosimetric Therapeutic Ratio J.M. Boyle,1 J.A. Dorth,2 O.I. Craciunescu,3 K. Light,4 J.R. Roper,5 and J.P. Chino6; 1Duke University, Durham, NC, 2University Hospitals Seidman Cancer Center, Cleveland, OH, 3Duke University Medical Center, Durham, NC, 4Duke University, Durham, NC, 5Emory University, Atlanta, GA, 6Duke University, Durham, NC Purpose/Objective(s): Dose escalation to pelvic and paraaortic nodes is a challenge for radiation treatment planning due to the proximity of critical normal tissues. A simultaneous integrated boost (SIB) has several potential advantages: (1) reduction in overall treatment time limiting tumor repopulation, (2) higher dose per fraction resulting in a higher biologically equivalent dose (BED) and (3) increased conformity resulting in lower dose to normal tissues. The purpose of this dosimetric study was to compare SIB with a sequential boost (SB). Materials/Methods: Ten women with node positive cervical cancer were planned with both SIB and SB IMRT. Two target volumes were used: an elective volume (PTV45) to be treated in 25 fractions of 1.8 Gy and a boost to involved nodes (PTV55). In the SB plans this was