Scientific Session 19 Oncology

RESULTS: The catheter tip was usually placed at the SVC/ RAjunction (n=63), with the remainder in the proximal RA (n=8) or proximal subclavian v (n=I). Paresthesia ofthe forearm required port removal in 2 children . Delayed complications included site infection (n=2), catheter infection (n=3), cellulitis or necrosis (n=2), thrombosis (n=I), and phlebitis (n=I). Thirty ports were removed electively after 60-1 '6 20 portdays (mean= 321 d) and 11 non-electively after 15- 150 portdays (mean=60 d). Thirty-two ports remain functioning after 15-1440 port-days (mean=450 d). CONCLUSION: Peripheral tunneled venous pon insenion is . safe and effective in children, and is a cost effective alternative when done with conscious sedation in Interventional Radiology. 12:52 PM

Abstract No. 95

Ultrasound Guided Percutaneous Liver Biopsies in Infants. J.A. Schwartz, The Hospital/or Sick Children. Toronto, ON, Canada·J. Amaral'P Chait·B. Connolly-M. Temple'P John, eta!' PURPOSE: This study reviews our experience in ultrasound guided percutaneous liver biopsies (PLB) in chi.ldren under I year of age in a tertiary pediatric center. MATERIALS AND METHODS: Retrospective analysis of 45 biopsies performed in 41 infants from 0611999 to 06/2003 was performed and relevant data collected including demographics, needle type and size, number of samples, pathology results and complications. RESULTS: The group studied included 26 males and 15 females with a median age of 83 days (29 to 348 days) and a mean weight of 4.3kg (2-8 .3kg). 29 procedures (64%) were done under GA. An 18G end cut needle was used in 35 (78%) cases. Coaxial technique was utilized only in 2 cases and one tract embolization was performed . The average number of passes was 1.8 (1-4) obtaining an average of 2.2 cores of tissue (1-4). All the biopsies provided sufficient tissue for pathology analysis. Two patients had PLB repeated due to inconclusive pathology reports, even though good samples were provided in the first biopsy. One transplant patient had 3 PLB to exclude rejection. One major bleed needing surgical intervention (2.2%) and 2 minor complications (4.4%) happened. No deaths occurred related to the procedure. CONCLUSION: PLB in infants is a good and efficient diagnostic tool but should be pelformed with caution and by experienced physicians. 1:03PM

Abstract No. 96

Pediatric Patient Doses in Interventional Neuroradiology. N. Swoboda, MacMaster University, Hamilton, ON,

Canada·B. Connolly·E. Charkot·J. Smith ·D. Armstrong PURPOSE: Neuroangiography is a major component of neurointerventional techniques (diagnostic & therapeutic). Procedures may be long and complex. Although radiation doses are important (especially in children), they receive little attention / emphasis. Our aim was to evaluate radiation doses during neuroangiography procedures using a PEMNET online dosimeter, which measures entrance dose. MATERIALS AND METHODS: Retrospective analysis of recorded doses in 6 I children on a biplane neuroangiography system. Doses were categorized as from fluoroscopy (FL) or digital subtraction angiography (DSA); frontal (PA) or lateral (Lat) planes;.diagnostic (DxP) or therapeutic procedures (TxP).

RESULTS: 49 procedures were diagnostic and 12 were therapeutic. The mean fluoroscopy time was 12.7 minutes PA and 7.4 Lat. The median number of DSA frames/patient was 266 PA and 194 Lat. For DxPs the median dose/pt was 215mGy PA and 109 mGy Lat [max 1182mGy and 347mGy respectively]. For TxPs the median dose/pt was 398mGy PA and 187mGy Lat [max 1267mGy and 756mGy respectively]. DSA accounted for 78% of patients' dose. Procedural doses do not exceed threshold doses for deterministic effects (2Gy), although returning patients would easily exceed these limits. Computed estimates of dose equivalents for brain, lens and thyroid were calculated using the MonteCarlo method. CONCLUSION: Awareness of cumulative doses is important during neuroangiography. These numbers establish a baseline upon which improvements to reduce dose can be sought and evaluated.

Scientific Session 19 Oncology Saturday, March 27, 2004 12:30 PM - 2:00 PM Moderator(s): Chieh-Min Fan, MD Jeer Sandlnt, MD 12:30 PM

Abstract No. 97

FEATURED ABSTRACT

Commentator: Damian Dupuy, MD Lung Radiofrequency Abl a ti on with and withou t Bronchial Occlusion: Experimental Study in Porcine Lungs. F Oshima, Mie University School of Medicine, Edobashi T.~u, Mie, Japan"K. Yamakadc)'A. Nakatsuka'lvL Akeboshi'M. Makita·K. Takeda, et al. PURPOSE: Blood flow is an efficient mechanism for heat loss by convection when radiofrequency (RF) ablation is performed in the solid organs such as liver and kidney. In addition to perfusion, there may be the influence·ofventilation on thermal lesion volume when RF ablation is perfolmed in ~he lung. Balloon occlusion of the main bronchus blocks ventilation and lowers· perfusion in the lung. This study was undertaken to compare the thermal lesion volumes the volume ofthermallesions when bronchial occlusion is and is not used. MATERIALS AND METHODS: RF ablation was performed in six pigs. A straight 17G internally cooled·tip electrode with a 2cm exposed tip was inserted into the center of the inferior lobe oflung under the biplane ,f luoroscopic guidance. In each animal, RF ablation was performed for 12 minutes with balloon occlusion ofthe main bronchus in one lung and without balloon occ1usionin the contralateral lung. The tissue temperature around the electrode tip was measured immediately after RF application , The volume ofthe thermal lesion was compared by histo logy in both groups of lungs ablated with and without bronchial occlusion. RESULTS: The ablated tissue temperature became higher i~ the lungs with bronchial occlusion than those without bronchiall occlusion (51 ±7°C vs. 44± 2°C, p<0 ,06), Thermal lesion volume was significantly larger in the lungs with bronchial occlusio n than th os.e withoLl.t bronchial occlusion (6,535±1 , I 14mm l vs. 3,368±676 mm3, p
Sl77

12:45 PM

Abstract No. 98

Prospective Evaluation of Radiofrequency Ablation of Lung Tumors: Work in Progress. T. de Baere, Institut Gustave Roussy, Villejuij, Val de marne, France ·J Palussiere ·A. Auperin' V. Kuoch

PURPOSE: To define feasibility, safety, local efficacy and describe CT image finding ofRF ablation oflung tumors. MATERIALS AND METHODS: During 18 years, 58 patients aged 27 to 81 with lung tumors meeting inclusion criteria (non surgical tumors less than 4 cm, 5 or less in number, MEV superior to I IIsec, no other tumor location) were treated under CT guidance with radiofrequency using 140 LeVeen needles and RF 3000 generator with algorithm dedicated to the lung. Follow-up included respiratory test at 30 days, CT 24 to 48 hours after RF, and every two months during I year. RESULTS: 9 primaries, and III metastases required a mean of 1.33 RF deliveries. Pneumothorax occurred in 50% of patient, 29% required no treatment, 13% were managed by aspiration, and only 8% required chest tube placement. I and 32 days after treatment, ground glass opacities and pleural effu s ion were present in 77% and 64% respectively. Hemoptysis occurred in 7% of case and could last as long as 3 weeks. Pneumopathy requiring antibiotics was present in 9% . MEV was not significantly modified after treatment. Mean tumor size increased from 17mm before radiofrequency, to 32rnrn immediately after, then 39mm early after RF, and decreased to 25, 23 and 20mm at respectively 2, 4 and 12 months ..6'of the 110 treated tumors had recurred after a mean follow-up of 7 months. CONCLUSION: RF ablation of lung tumors is feasible and safe. Assessment oflocal efficacy needs further investigation. 12:56 PM

Abstract No. 99

Pathological Assessment of Radiofrequency Ablation of Pulmonary Metastases in Humans - Preliminary Experience. J.D. laskolka, Mount Sinai Hospital, Toronto, ON, Canada·MR. Asch·E.B. Hayeems'M Tsao·T.K. Waddell·JR. Kachura, et al.

PURPOSE: To evaluate the efficacy of radiofrequency ablation (RFA) of pulmonary metastases in humans using pathological assessment and an ablate and delayed resection protocol. MATERIALS AND METHODS: Three female patients scheduled to undergo surgical resection of multiple pulmonary metastases were prospectively enrolled in the study which bad approval from the local institutional review board. Each patient underwent CT-guided percutaneous RFA of a single lesion, under general anesthesia, 2 to 4 weeks prior to surgical resection. Primary tumors included colorectai carcinoma (2) and soft-tissue sarcoma (I). Lesion sizes ranged from 1.2 2.5 cm (mean 1.9 cm). All procedures were performed on an outpatient basis with an expandable multitined electrode (Boston Scientific, Natick, MA) using the manufacturers suggested algorithm. The ablated lesion was analyzed histopathologically using routine hematoxylin and eosin and immunohistochemistry for the proliferative marker MIB- l to assess for local tumor necrosis.

S178

RESULTS: The RFA procedure was technically successful, with rise in impedance being achieved in all cases. There were no symptomatic procedure-related complications. Histopathological analysis demonstrated that each treated metastasis was entirely encompassed by the tbermal lesion.

Normal lung parenchyma and neoplastic cells at the periphery of the tumor showed complete coagulative necrosis. However, 50-75% of central neoplastic cells in all 3 patients retained normal staining characteristics at the time of review.

CONCLUSION: Percutaneous RFA of pulmonary metastases was technically successful and safe in our patients. However histopathologic analysis of the treated lesions suggested incomplete tumor necrosis which is discordant with tbe animal literature and otber preliminary human studies. This may be due to nonuniform distribution of heat within tbe tumor or our method of histopathological analysis. Differentiation between live and necrotic cells may be difficult without special stains and may explain the apparent viability oftreated tumors in our series. Further evaluation of methods of pathologic assessment for tumor viability or reevaluation of current heating algorithms during RFA is recommended. 1:07PM

Abstract No. 100

Magnetic Targeted Delivery of Mitomycin C Administered by Pulmonary Artery Infusion to Swine Lung. MJ Sands, Cleveland Clinic Foundation, Cleveland, OH, USA- Y. Dong ·C. Peterson-G. Tapolsky- Y. Li·T. Leakakos

PURPOSE: To evaluate the toxicity of mitomycin adsorbed to Magnetic Targeted Carriers (MTC-MMC) and administered by pulmonary artery infusion to swine lung under magnetic guidance as a potentiallocoregional treament of non-small-cell lung cancer. MATERIALS AND METHODS: Safety of locoregional administration of MTC-MMC delivered by the pulmonary artery route was evaluated in a 28 day toxicity study conducted in 18 swine. MMC was adsorbed to the activated carbon component ofMTC microparticles containing metallic iron that provided magnetic susceptibility. Pulmonary arteries were catheterized under fluoroscopic guidance for delivery of escalating doses ofMTC-MMC or MTC and MMC alone to a specified lung area. An external permanent rare earth magnet was used during MTC-MMC administration to retain the test material in the targeted area. The animals were evaluated over 4-weeks following treatment for changes in clinical signs, body weight, and clinical pathology. Pharmacokinetics, gross necropsies and histopathological examinations were included. RESULTS: There was no mortality or clinical signs. Changes in clinical chemistry, hematology and coagulation parameters were generally minor and transient. MMC showed linear pharmacokinetics and was rapidly distributed in plasma. Systemic levels were reduced in MTC-MMC arms . Treatment-related effects seen upon histopathological examination were limited to targeted lung where MTC particles were observed in alveolar capillaries. One animal from the high dose group of MTC-MMC had multiple pulmonary granulomas in the targeted lung witb MTC particles in their centers. MTCs were not observed in locations other than the targeted lung lobes. Specifically, no MTC particles were noted within the reticulo-endothelial system. CONCLUSION: Pulmonary artery delivery of MTC-MMC resulted in site-specific localization of MTC particles in the targeted lung lobe. Biologically significant gross and microscopic lesions were limited to the targeted areas oflung in a single animal administered 80 mg MTC- (4mg) MMC. MTC particles were confined to the targeted region of tbe lung 28 days after dosingindicating retention ofMTCs in that area. Safety and targeting efficiency support further development.

1:18 PM

Abstract No. 101

Radiofrequency Ablation of Un resectable Metastatic Lung Thmors from Colorectal Cancer: Initial Experience. M Akeboshi, Mie University School of Medicine. Tsu, Mie, Japan-K. Yamakado ·H. Takaki ·A. Nakalsuka·K. Matsumura·K. Takeda PURPOSE: There have been no useful therapeutic options other than surgery for the treatment of lung metastases from colorectal cancer. This study was prospectively undertaken to determine whether lung rad iofrequency (RF) ablation can be a useful therapeutic option to treat unresectable metastatic lung neoplasms from colorectal cancer. A1A TERJALS AND METHODS: 14 consecutive patients with 28 metastatic lung tumors from colorectal cancer underwent RF ablation. Twnor sizes ranged from 0. 8cm to S.Ocm with the mean size of2.2±I.Ocm. RF ablation was percutaneously done under the CT-t1uoroscopic guidance. Initial therapeutic response was evaluated by FDG-PET and contrast-enhanced CT studies. Disappearance of radioactivity in the tumor on PET images and no tumor enhancement on CT images were considered to be complete tumor necrosis.

RESULTS: RF ablation was technically successful in all lesions. Complete tumor necrosis was achieved in 19 out of 28 tumors (68%) after initial RF treatment. There was a significant difference in the rate of complete tumor necrosi s between tumors 2.Scm or less in a diameter and those larger than 2.Scm (100% vs. 43 %, p
1:29PM

Abstract No. 102

Percutaneous Hepatic Perfusion for Primary and Secondary Liver Thmors- Interventional Radiology Initial Experience. Z. Neeman, National Institutes ofHealth, Bethesda, MD, USA-B.J. Wood·A . Kam·R. Chang·J. Ping Pang·R.H. Alexander, et al. P URPOSE: The prevalence, morbidity, and lack of effective systemic treatments for liver tumors (primary and secondary) have spawned efforts to establi sh more organ-specific techniques. Since the majority of dose-limiting toxicities for systemic chemotherapies are not hepatic, organ-specific techniques that would prevent systemic exposure could potentially allow for higher dosing regimens for neoplasms of the liver. MATERJALS AND METHODS: The procedure is performed under general anesthesia and takes about 3 hours. During the patient's fust session, the gastroduodenal artery is documented and embolized using coils, to prevent reflux ofthe high dose melphalan and injury of small bowel. The common! proper hepatic artery is catheterized, and melphalan (I.S mg/kg) is infused during 30 minutes. Hepatic venous bemofiltration is achieved by placement of a double-balloon, IVC filter catheter system (Delcath system, Delcath Inc. Stamford, CT), isolating and diverting the hepatic venous outflow through an external activated charcoal filter(Delcath systems, Inc.) and back into the systemic circulation via the right internal jugular vein for another 30 minutes.

RESULTS: A total of 62 treatments were performed on 2S patients . Only several minor complications related to heparinization (hemorrhage into metastasis, venous access site hematomas) and high dose melphalan (low white count and low platelets) are described . CONCLUSION: This report seeks to address technical issues related to interventional radiology and radiology practice, encountered using PHP with melphalan in phase I trials for patients with primary and secondary liver tumors. Delivery of melphalan is possible via this percutaneous system, is relatively safe with minor complications related to highdose heparin, and with limited manageable systemic toxicity. From the interventional radiology technique stand-point it has proven to be a safe procedure, supporting continuing further clinical studies using this percutaneous approach.

Scientific Session 20 Transplantation Saturday, March 27, 2004 12:30 PM - 2:00 PM Moderator{s}: Matthew A. Mauro, MD Jonathan Susman, MD

12:30PM

Abstract No. 103

FEATl'IU:D AB.I.;TRA(, I'

I

Commentator: Janelle D. Durham \11>, !\IBA

r ercutaneous Intraportal Islet Cell Transplantation for rrype I Diabetes.

1M. Jung, Universiry ofPennsylvania, Philadelphia, PA, USA ·Me. Soulen·.1.F Markmann ·D.B. Freiman'R.D. ~hlan f..y-Goldberg· .0. Trerolola, et al. PURPOSE: To report o ur experience withpercutaneou intraportal islet cell transplantation. lATERIALS AND METHODS: I I patients un derwent 20 islet cell transplantations between September, 2001 aud July, 2003. Inclusion criteria were type J diabetes for 5 years, rnultipl e hospitalizations for severe hypoglycemia, and 1;Jugetectable C-peplide level. The portal vein wa catheterized 'underllitra ound ancVor t1uoroscopic guidance. 1 let cells were slowly infused through a SF catheter whil e monitoring portal pressure. The tract was embolizcd wi th Gelfoam. Patients w ho rem ained in ulin dependent unde rwent add itional transplant procedure . RESULTS: Ten of 1 J patients completed the treatment protocol; one had to stop immune suppres ion because of an unrelated infection . 7 patients underw ent two transplant procedure, wi th a, median interval of 2 months between proced ure (rang 2 -13); one pat ient underwe nt a th ird procedure. Portal pressures were 10 ± 3.S mmHg before iri fu ion and 12 ± 3.0 mmHg after (p=.OOI, paired l-te t).The only com pl icat ion was a perihepatic hematoma requiring transfusion. Itrasou nd guidance was un uccessflli during three repeat transplant procedures due to increased hepatic echogenicity obscuring tbe port,al vein. Seven patients were imaged with MRI after transplantation; 3/7 had a uniq ue pattern of periportal hepati c steatosis, all with functioning grafts. The remaining 4/7 patients. who did not show hepatic ~tcatosis . bad non-functional grafts at the time of the MRI. N b ine patients ach!eved independence from exogenous iJl ulih. f these, 3 remalO mdcpendent at 2, 7, and' I I month ; 2 of th e e 3 had previou Iy attai ned ins ul in independence for periods of8 and 9 month . The remain ing 6 out 0[ 9 patients ~ere independent of ins ul in for a medi an of 6 mOllths (range 1-13 months). The patient who did nOI achie e insulin independence had his requirement reduced by 66%.

S179