EVLA was defined as ultrasound-documented absence of flow in the treated vein. Treatment failure was defined as ultrasonographic evidence of flow at any point in the treated vein segment at one week or beyond the treatment date. Two- tailed t-test was performed for statistical analysis and the null hypothesis was rejected at p<0.05. RESULTS: 86 treated veins were followed with ultrasound over I - 14 months (mean 7.4 months). During this time, 65/ 86 (76%) of treated veins remained closed. In this group of successfully treated veins, average energy delivered was 59.7 joules/cm (range 22 - 131 joules/cm). The average energy delivered to the 20/86 (24%) veins in the failure group was 46 joules/cm (range 23 - 69 joules/cm). This difference in delivered energy was statistically significant (p<.02). No treatment failures were identified in patients who received 70 joules/cm or more. CONCLUSION: EVLA is an effective method of saphenous vein ablation. Higher delivered energy is associated with successful EVLA, particularly when >70 joules /cm are delivered. Scientific Session 37 Oncology Monday, March 29, 2004 4:30 PM - 6:00 PM Moderator(s): Thierry de Baere, MD Catherine M. Tuite, MD 4:30PM AbstractNo_199 MR-Guided Percutaneous Cryoablation of Renal Tumors: Results of Midterm Follow Up_ T. Mogami; Radiology, Jikei University, Kashiwa Hospital, Kashiwa, Chiba, Japan-J. Harada-M. Dohi-K. Fukuda-K. Kishimoto PURPOSE: Cryoablation provides a method of focal destruction of targeted tissue while preserving most of the surrounding tissue. MRI has high sensitivity to detect ice ball as signal loss area. The purpose of our study is to evaluate the midterm results of percutaneous cryoablation for renal tumors under horizontal open MRl guidance. MATERIALS AND METHODS: Thirteen patients with renal tumor underwent percutaneous cryoablation with local anesthesia using a horizontal open MRl system. The size of tumor was radiographically documented as 4 cm or less in diameter. One to four cryoprobes with 2 or 3 mm in diameter were advanced into the tumor under MR fluoroscopy. During cryoablation, growth of ice ball was monitored by MR images. Two freeze-thaw cycles were performed for cryoablation. Follow up dynamic CT and laboratory examinations were done for evaluation of cryoablation. RESULTS: All patients tolerated the procedure well. The maximum tumor size ranged from 2.0 cm to 4.0 cm (mean 2.7 cm). Follow-up CT images were available for all 13 patients. The follow-up period ranged from 17 months to 31 months (mean 24.0 months). MR images demonstrate the ice ball as sharply marginated signal loss area that expand and engulfed the tumor with clear contrast between the ice ball and surrounding normal tissue. Complete tumor necrosis was confirmed in 10 of 13 patients as non-enhanced area on follow up dynamic CT. Three patients showed residual or recurrent tumor on follow-up CT. One patient underwent additional cryoablation, and complete tumor necrosis was obtained. Other 2 patients underwent partial nephrectomy. Two complications occurred, one perirenal hematoma and one pleural effusion. Both complications resolved spontaneously without surgical or other interventions. On follow up CT images, cryoablated area and tumor almost disappeared in 3. CONCLUSION: MR-guided percutaneous renal cryoablation demonstrated its feasibility and minimal morbidity. MR- guided cryoablation can be used as a safe procedure for nephron-sparing surgery. 4:41 PM Abstract No_ 200 Hepatocellular Carcinoma Treated with Radiofrequency Ablation: An Early Evaluation with Magnetic Resonance Imaging_ A.A. Khankan, Osaka Un iversity Graduate School of Medicine, Sui/a, Osaka, Japan-T. Murakami-M. Ma/sushita-r. Aoki-H. Onishi-K. Osuga, ef al. PURPOSE: Previous animal studies on the effect of radiofrequency ablation (RFA) in liver usi ng MRl revealed two concentric zones on T I-weighted image (T I WI). The central heterogenous high intensity zone corresponded to the coagulative necrosis, while the peripheral low intensity zone corresponded to hemorrhage and edema. The purpose of this study is to determine the usefulness of the two zones pattern of MRl in early evaluation of efficacy of RF ablation for hepatocellular carcinoma (HCC). MATERIALS AND METHODS: Twenty-three patients with 28 HCC nodules (9-26 mm, mean 18 mm in diameter) treated with percutaneous RFA under US guidance underwent MRl with T I WI (SPGR, TRITE 70-200/2.3) one day and dynamic CT one week after the treatment, respectively. For both imaging modalities, appearance of the treated area (two zones on T I WI and unenhancement on dynamic CT) and ablation efficacy (complete, nearly complete, and incomplete) were assessed each time by correlating MR images with dynamic CT images. Follow up with MRI and dynamic CT was performed 6 months after the treatment, and local recurrence was evaluated. RESULTS: Complete (n=18) and nearly complete (n=9) ablated nodules corresponded with tumor location within the central zone and within both central and peripheral zones, respectively, on one day Tl WI. Those ablated nodules located within the nonenhanced areas on one week dynamic CT after treatment. Treated areas shown on T I WI were correlated well with the findings of dynamic CT, anq included the previously identified HCC nodules. Conversely, the incomplete ablated nodule (n= I) located out of the zones on Tl WI and showed enhancement on dynamic CT. Follow up studies at 6 month revealed persistent complete necrosis in all complete ablated nodules without recurrence at ·the site of ablation. Three of nine nearly complete ablated nodules had developed local recurrence shown on both MRI and dynamic CT. CONCLUSION: Despite the small number of patients, it is suggested that one day Tl WI is useful for early assessment of the efficacy of RFA for HCC according to the zone pattern. 4:52PM Abstract No_ 201 Successful Use of Arteriography to Determine Duration of Neoadjuvant Therapy and PI-edict Necrosis in Osteosarcoma_ B.A. Jamroz, Presbyterian/ St. Luke s Medical Center, Denver, CO, USA -J. W Cullen -SL Stevens -W Madsen-J. Hinshaw-R.M. Wilkins, et al. PURPOSE: To investigate our ability to use serial arteriography to assess tumor response, predict necrosis, and individualize the duration of an intra-arterial (IA) neoadjuvant S213
Transcript
EVLA was defined as ultrasound-documented absence of flow in the treated vein. Treatment failure was defined as ultrasonographic evidence of flow at any point in the treated vein segment at one week or beyond the treatment date. Twotailed t-test was performed for statistical analysis and the null hypothesis was rejected at p<0.05 .
RESULTS: 86 treated veins were followed with ultrasound over I - 14 months (mean 7.4 months). During this time, 65/ 86 (76%) of treated veins remained closed. In this group of successfully treated veins, average energy delivered was 59.7 joules/cm (range 22 - 131 joules/cm). The average energy delivered to the 20/86 (24%) veins in the failure group was 46 joules/cm (range 23 - 69 joules/cm). This difference in delivered energy was statistically significant (p<.02). No treatment failures were identified in patients who received 70 joules/cm or more.
CONCLUSION: EVLA is an effective method of saphenous vein ablation. Higher delivered energy is associated with successful EVLA, particularly when >70 joules/cm are delivered.
Scientific Session 37 Oncology
Monday, March 29, 2004 4:30 PM - 6:00 PM Moderator(s): Thierry de Baere, MD
Catherine M. Tuite, MD
4:30PM AbstractNo_199
MR-Guided Percutaneous Cryoablation of Renal Tumors: Results of Midterm Follow Up_ T. Mogami; Radiology, Jikei University, Kashiwa Hospital, Kashiwa, Chiba, Japan-J. Harada-M. Dohi-K. Fukuda-K. Kishimoto
PURPOSE: Cryoablation provides a method of focal destruction of targeted tissue while preserving most of the surrounding tissue. MRI has high sensitivity to detect ice ball as signal loss area. The purpose of our study is to evaluate the midterm results of per cutaneous cryoablation for renal tumors under horizontal open MRl guidance.
MATERIALS AND METHODS: Thirteen patients with renal tumor underwent percutaneous cryoablation with local anesthesia using a horizontal open MRl system. The size of tumor was radiographically documented as 4 cm or less in diameter. One to four cryoprobes with 2 or 3 mm in diameter were advanced into the tumor under MR fluoroscopy. During cryoablation, growth of ice ball was monitored by MR images. Two freeze-thaw cycles were performed for cryoablation. Follow up dynamic CT and laboratory examinations were done for evaluation of cryoablation.
RESULTS: All patients tolerated the procedure well. The maximum tumor size ranged from 2.0 cm to 4.0 cm (mean 2.7 cm). Follow-up CT images were available for all 13 patients. The follow-up period ranged from 17 months to 31 months (mean 24.0 months). MR images demonstrate the ice ball as sharply marginated signal loss area that expand and engulfed the tumor with clear contrast between the ice ball and surrounding normal tissue. Complete tumor necrosis was confirmed in 10 of 13 patients as non-enhanced area on follow up dynamic CT. Three patients showed residual or recurrent tumor on follow-up CT. One patient underwent additional cryoablation, and complete tumor necrosis was obtained. Other 2 patients underwent partial nephrectomy. Two complications occurred, one perirenal hematoma and one pleural effusion.
Both complications resolved spontaneously without surgical or other interventions. On follow up CT images, cryoablated area and tumor almost disappeared in 3.
CONCLUSION: MR-guided percutaneous renal cryoablation demonstrated its feasibility and minimal morbidity. MRguided cryoablation can be used as a safe procedure for nephron-sparing surgery.
4:41 PM Abstract No_ 200
Hepatocellular Carcinoma Treated with Radiofrequency Ablation: An Early Evaluation with Magnetic Resonance Imaging_ A.A. Khankan, Osaka Un iversity Graduate School of Medicine, Sui/a, Osaka, Japan-T. Murakami-M. Ma/sushita-r. Aoki-H. Onishi-K. Osuga, e f al.
PURPOSE: Previous animal studies on the effect of radiofrequency ablation (RFA) in liver using MRl revealed two concentric zones on T I-weighted image (T I WI). The central heterogenous high intensity zone corresponded to the coagulative necrosis, while the peripheral low intensity zone corresponded to hemorrhage and edema. The purpose of this study is to determine the usefulness of the two zones pattern of MRl in early evaluation of efficacy of RF ablation for hepatocellular carcinoma (HCC).
MATERIALS AND METHODS: Twenty-three patients with 28 HCC nodules (9-26 mm, mean 18 mm in diameter) treated with percutaneous RFA under US guidance underwent MRl with T I WI (SPGR, TRITE 70-200/2.3) one day and dynamic CT one week after the treatment, respectively. For both imaging modalities, appearance of the treated area (two zones on T I WI and unenhancement on dynamic CT) and ablation efficacy (complete, nearly complete, and incomplete) were assessed each time by correlating MR images with dynamic CT images. Follow up with MRI and dynamic CT was performed 6 months after the treatment, and local recurrence was evaluated.
RESULTS: Complete (n=18) and nearly complete (n=9) ablated nodules corresponded with tumor location within the central zone and within both central and peripheral zones, respectively, on one day Tl WI. Those ablated nodules located within the nonenhanced areas on one week dynamic CT after treatment. Treated areas shown on T I WI were correlated well with the findings of dynamic CT, anq included the previously identified HCC nodules. Conversely, the incomplete ablated nodule (n= I) located out of the zones on Tl WI and showed enhancement on dynamic CT. Follow up studies at 6 month revealed persistent complete necrosis in all complete ablated nodules without recurrence at · the site of ablation. Three of nine nearly complete ablated nodules had developed local recurrence shown on both MRI and dynamic CT.
CONCLUSION: Despite the small number of patients, it is suggested that one day Tl WI is useful for early assessment of the efficacy of RFA for HCC according to the zone pattern.
4:52PM Abstract No_ 201
Successful Use of Arteriography to Determine Duration of Neoadjuvant Therapy and PI-edict Necrosis in Osteosarcoma_ B.A. Jamroz, Presbyterian/St. Luke s Medical Center, Denver, CO, USA -J. W Cullen -SL Stevens -W Madsen-J. Hinshaw-R.M. Wilkins, et al.
PURPOSE: To investigate our ability to use serial arteriography to assess tumor response, predict necrosis, and individualize the duration of an intra-arterial (IA) neoadjuvant
S213
chemotherapy protocol in patients with biopsy-proven high grade osteosarcoma and malignant fibrohistiocytoma of bone.
MATERIALS AND METHODS: Between July 1987 and March 2003, 108 patients completed a protocol using neoadjuvant IV doxorubicinlIA cisplatin. Patients were eligible regardless of age, stage or site, but normal renal and cardiac function were necessary. A minimum of3 IA cycles followed by definitive surgery were required for inclusion in the final analysis. [A dose/duration was increased for tumors> I Ocm. Initial arteriograms were evaluated for neovascularity, and % decrease was determined on subsequent arteriograms. An interdisciplinary team reached consensus on the final arteriographic result prior to definitive surgery. Cycles continued until I of 3 criteria were met: I) ~90% decreased neovascularity, 2) plateau of effect, or 3) no response.
RESULTS: We performed 404 IA procedures : 42 patients underwent 3 IA cycles, 52 had 4 and 14 required 5. There was a 2.5% complication rate (soft tissue necrosis-7, hematoma-2, arterial spasm-I). Ninety-three patients (86%) reached >90% decreased neovascularity. A good histologic response (tumor necrosis ~90%) was achieved in 81 .5%. The accuracy of serial arteriograms in predicting necrosis was 90%. Sensitivity and specificity were 96.6% and 60%, respectively. Univariate analysis was performed using the following variables: age at diagnosis, tumor site, size, stage, histology, initial neovascularity, and number of IA cycles. The only significant variable was the # of 1A cycles. When 5 were given, prediction of necrosis was less sensitive (p=0.03) and less accurate (p=0.02).
CONCL USjON: Serial arteriography was highly sensitive and accurately predictive of % tumor necrosis. This individually modified, dose-intensified neoadjuvant protocol yielded an excellent histologic response rate and was associated with minimal complications. Future endeavors should involve a multi-institutional study using this unique approach.
5:03 PM Abstract No. 202
Liver Malignancies: CT-Guided Brachytherapy in Patients Unfavorable for Radiofrequency or Laser Ablation. J. Ricke, Charile Virchow-Klinikum, Berlin, Germany'P Wust·G. Wieners'M Pech·R. Felix
PURPOSE: To evaluate CT-guided bracbytherapy in patients with very large liver malignancies or in liver tumors located adjacent to the liver hilum.
MATERIALS AND METHODS: We treated 20 patients (19 liver metastases and I cholangiocellular carcinoma) with CT guided high-dose-rate (HDR) brachytherapy us ing a I 92Iridium source. All patients demonstrated no functional liver degradation prior to irradiation. We stratified patients with either liver tumors > 5 cm (group A, n= II) or liver tumors < 5 cm adjacent to the liver hilum (group B, n=9). Dose planning for brachytherapy was performed using threedimensional CT data acquired after percutaneous applicator positioning. MRi follow up was performed 6 weeks and every 3 months post intervention.
RESULTS: The mean tumor diameter was 7.7 cm (5.5 -10.8 cm) in group A, 3.6 cm (2.2 -4.9 cm) in group B. On average, a minimal dose of 17 Gy in the target volume was applied (12 - 25 Gy). Severe side effects were recorded in 2 patients (10%). One patient demonstrated obstructive jaundice due to tumor edema after irradiation of a metastasis adjacent to the bile duct bifurcation. One patient developed intraabdominal hemorrhage which was treated by a single blood transfusion.
S214 We commonly encountered moderate increases of liver
enzymes and bilirubin without clinical relevance. The median follow up was 9 months. In group A (tumors> 5 cm), primary local tumor control after 6 and 12 months was 75 and 40%. Three of 4 local recurrences were successfully re-irradiated leading to a primary assisted local control of 91 % after 12 months. In group B, the local control rate after 12 months was still 100%.
CONCLUSION: CT guided brachytherapy based on individual : dose plans using 3D CT data sets is effective in liver : malignancies measuring up to 10 cm. Adjacent location ofthe ! liver hilum is no limiting factor.
5:14PM Abstract No. 203
CT Radiofrequency Ablation for Un resectable NSCLC: Short and Long Term Results at One Year. A.B. Wallace, David Geffen School of Medicine at UCLA, Los Angeles, CA, USA -R.D. Suh -J. G. Goldin
PURPOSE: To evaluate both short and long term results at one year ofCT-guided percutaneous radiofrequency ablation (RFA) in the treatment ofunresectable primary non-small cell lung carcinoma (NSCLC).
!viA TERIALS AND METHODS: From October 1, 2000, sixteen patients (9 males and 7 females; average age 64.2 years) with 19 biopsy-proven, unresectable (comorbid cardiopulmonary disease , refractory to conventional chemotherapy and/or radiotherapy, and local recurrence) non- ' small cell lung carcinoma were treated with RFA. Pre-RFA lung function ranged from 15%-98% FEY 1. Stage of disease at RFA were as follows: IA (4/16), IB (4/16), 2B (1/16), 3B/ 4 (1 / 16), and 6/16 were treated following failure of conventional treatment. Three patients received neoadjuvant chemotherapy or chemotherapy prior to RF A, and six patients received adjuvant or salvage chemotherapy following RFA.i Each patient was followed up for one year with serial CT scans at 1, 4, 6 and 12 months. Histopathology cell types included adenocarcinoma (15/19), bronchioalveolar (2/19), poorly differentiated (1119) and NSCLC with neuroendocrine features (1119).
RESULTS: Initial technical success was achieved in all sites (19119) in all patients. Short term complications included pneumothorax (11119 of which only 4/11 required catheter placement), symptomatic pleural effusion (1 / 19), prolonged pain (2/16 patients), dyspnea (2/16 patients), and significant contrast reaction (1116 patients at 3 month followup CT scan). Long term outcomes revealed progressive disease (5/16 patients) and death (3/16) at 2, 7 and 8 months post-RFA, none of which directly related to the procedure. Sites of progressive disease included 2 local and 4 metastatic sites.
CONCLUSION: Although technically feasible and safe, RFA as a sole treatment modality for cure for NSCLC is questionable. However, further clinical trials, in which patient selection can be optimized and the role of adjuvant therapies with RFA investigated, need to be performed.
5:25PM Abstract No. 204
Histopathologic Outcome of Neoadjuvant Image-Guided Therapy of Hepatocellular Carcinoma. H. Marin, University of Pennsylvania, Philadelphia, PA, USA-M.e. SoulenoT W Clark-e.M. Tuite-E.E. Furth·K. Olthoff
PURPOSE: The only curative treatment for hepatocellular carcinoma (HCC) is orthotopic liver transplantation (OLT). Acceptable indications for OLT are either a single mass <Scm
or three tumors < 3cm. Arresting tumor growth in patients waiting for OLT or downstaging to make patients candidates for OLT can be attained by neoadjuvant methods like chemoembolization (CE), radioembolization (RE), chemical (CA) or radio frequency ablation (RFA). We evaluated the histopathologic response in explant specimens to image-guided therapy of HCC.
MATERIALS AND METHODS: 23 patients with 35 HCC nodules underwent neoadjuvant image-guided therapy 88± I 0 I days (range 1-392) prior to transplant, including CE (7 nodules in 4 patients), RE (8 nodules in 4 patients), RFA (9 nodules in 7 patients), CA (3 nodules in 2 patients) and combined chemoembolization and ablation (8 nodules in 6 patients). 151 23 patients not transplanted within 30 days had interval gadolinium-enhanced MRI at I month and every 3 months untilOLT.
RESULTS: Residual viable tumor was seen on MRI in 8 out of 15 patients with post-treatment images (14 of 21 treated nodules). 5 new HCC nodules appeared on MRI in 4 patients. Mean time from latest MR1 to OLT was 35 days. At explant, pathology revealed 15 additional radiologically-occult HCC nodules in 8 patients, for a total of 55 nodules. Among the 35 tumors treated with neo-adjuvant therapy, in 26 (74%) viable tumor was identified at microscopy. Viability of the treated tumor was noted in 3/4 patients with CE, 4/4 with RE, 6/7 with RFA, 2/2 with CA and 5/6 with combined treatment. In addition, remote lesions were identified in the explants in 5/9 patients treated with CA or RFA, in 2/8 patients with regional therapy (RE or CE) and in ]/6 patient with combined treatment. Among the patients with completely necrotic tumor on MRI, I had viable satellite nodules and 2 had viable remote lesions.
CONCLUSION: Viable tumor is identified histologically in 74% of treated HCC nodules and somewhere in the explanted liver in 19/23 (83%) of patients treated with image-guided therapies, despite apparently complete radiological response in 47% (7115 patients) on MRI. MRJ overestimates the complete response rate of HCC to image-guided therapy.
5:36PM Abstract No. 205
PET in Patients Selection and Follow-Up for RFA of Liver Metastases: Work in Progress. F Orsi, European Institute o/Oncology, Milan, Italy·L. Mon/ardini ·D. Paolo ·G. Bonomo·L. Travaini·S. Rizzo, e/ af.
PURPOSE: The aim of this study is to assess the role of PET in patient's selection and follow-up for locoregional treatment such as RFA. PET and CT are performed before and after liver RFA in patients with hepatic metastases.
MATERIALS AND METHODS: Sixteen patients with liver metastases (6 M, 10F; age: 49-79; average: 64) have been enrolled for liver RFA in a multi-modal therapeutic strategy (systemic and local). All patients were selected clinically (by a multi-disciplinary team) and instrumentally (by CT and PET). Twenty treatments were performed in local anaesthesia. Instrumental follow-up was then obtained by CT (I and 3 months after treatment) and by PET (at least 2 months after liver ablation). CT and PET were then compared in order to obtain data about local and systemic pathological ftndings.
RESULTS: PET and CT showed the same results in fifteen treatments: both negative in nine and both positive for local recurrence and/or persistent disease in six. PET was positive in spite of CT negative after five treatments: PET and CT showed discordant findings in 25% of cases. Moreover in
seven patients PET showed extra-hepatic systemic detection, modifying following treatment planning.
CONCLUSION: PET detected local recurrence and/or residual disease after hepatic RFA earlier than CT in 25% of cases. The association of findings from both PET and CT seems to be more effective in local and systemic treatment management than CT alone in patient with liver metastases. However we need further evaluation, more experience and longer followup to assess the real clinical benefit from combined CT and PET follow-up after RFA.
Percu taneous Intervention · for Non-Ma tur ing Arteriovenous (AV) Fistulae. B. PeynirciQg/u, Indiafla University, Sr..·hoo{ of Medicine, Indianapolis, IN, USA ·C. Mc.Lerman ·B. Hankins'M Kohli·M.S . .fohn on·J. Namys!ow kj, et at.
PURPOSE: To determ ine tbe ,Success of angioplasty (PTA) and collateral embolization for non-maturing or Ihalftmctioning fistulae. . .
MATERlALSAND METHODS: 196 PT and 8 embolizations were perfonned in 102 patients from Augu I 1995 to MlIrcq 2002. Six prisoners were excluded due to fRB rules. Techniea·1 ~uccess was defined as residual stenosi < 30% wjth al~ improved thrill. Clinical success was defined as the ability to dialyze Ulrough the fi lula. The mean number ofintervenlions; 3, .6, and 12 month prima.ry and econdary patencie ; and the mean life of the fistulae were calcu lated.
RESULTS.·There were 2± I int rvelltions per patient achieving a mean primary patency of 12± 13 months and a fis tula l ife of 16±14 months from the time offir t intervention. Technical success was achieved in 179 of 196 PTA (9 1 %). Of 102 patients, 72 had interventions tbat re ulted in the u se ofthei ~ fistula (70.6%). Primary and secondary p<!ttmcics are listed i the tabfe. the complication rate was 4.4%. ComplicatiOLls :ve~e 2 puncture site hematomas, 2 dissections of which on wa occlusive, 2 ruptures of which one was stented, 1 prolonged bleeding and 2 mild to moderate vasospasms. [n 30 patie nts who were referred fo r non-ma.l.uring fi stu lae, intervention restllted in fistula use in 14 (47%). 1 maintail1cd fistula use until transplant and 3 are currently using the fi tula (at 22, 24, & 25 months). Mean primary patency was 7±7 l11011thS wi th m ean tistu la life of9±8 monlhs. 5 of 8 patients who had collaterals embolized were able to dialyze wi th lhe fistu la more Ihan once. Mean primary patency was 5±4 months with mean fistula life of 1101: 13.5 monlhs.
CONCLUSION: Angiop lasty and embolization of collatcrals can make malfunctioning fistulae u efu l and can prolong.the useful li fe of native fistulae for more th.an I year 50% of the time. SllccessfuJ intervention in non-maturing fistulae carr [esult in a useable fistula but long-term patency is less than for fistulae that have matured. .
