| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Technical Developments |
1 From the Departments of Radiology (M.W.W., R.K.K., N.A.F., J.M.L, R.L.G.) and Medicine (A.P.V.), University of California, San Francisco, 505 Parnassus Avenue, Room M-361, San Francisco, CA 94143; and FeRx Incorporated, San Diego, Calif (J.K.). Received November 15, 2002; revision requested January 17, 2003; revision received April 9; accepted June 9. Address correspondence to M.W.W. (e-mail: mark.wilson@radiology.ucsf.edu).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMaterials and Methods ResultsDiscussionREFERENCES |
|---|
© RSNA, 2004
Index terms: Liver, interventional procedures, 761.1266 • Liver neoplasms, angiography, 761.1242 • Liver neoplasms, MR, 761.121411, 761.121412 • Liver neoplasms, therapy, 761.1266 • Magnetic resonance (MR), guidance
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMaterials and Methods ResultsDiscussionREFERENCES |
|---|
The ability to guide and monitor transcatheter intraarterial tumor therapies with intraprocedural MR imaging has yet to be accomplished. This is due to several factors, including the logistic difficulty of placing an arterial catheter with fluoroscopic guidance and then safely transporting the patient to a suitable MR imaging unit. An additional hurdle is that of selecting the appropriate therapeutic agent that can be delivered with a catheter and visualized with MR imaging. Hence, the purpose of this study was to evaluate our experience with a dual-modality interventional MR imaging/conventional angiography system for catheter-directed intraarterial delivery of a magnetic targeted chemotherapy agent in the treatment of inoperable HCC.
|
Materials and Methods |
|---|
|
TOPABSTRACTINTRODUCTIONMaterials and Methods ResultsDiscussionREFERENCES |
|---|
Patient 1, a 42-year-old woman who was seropositive for hepatitis B, had undergone left liver lobectomy for removal of a primary HCC 2 years before enrollment in this study. At routine follow-up 18 months later, she was found to have a recurrent 46.0-cm3 lesion in the right lobe of the liver. Patient 2 was a previously healthy 50-year-old man who was seronegative for hepatitis and had a newly diagnosed primary 365.0-cm3 HCC that spanned both hepatic lobes. Patient 3 was a 70-year-old man who was found to have several confluent HCC lesions with a combined size of 103.0 cm3 in the right lobe of the liver and a small lesion in the liver dome. He was deemed a poor candidate for transplantation because of the volume of the disease. Only the confluent lesions were treated. Patient 4 was a 71-year-old man with a history of chronic hepatitis B who was found to have a 57.0-cm3 HCC at computed tomography (CT) performed to evaluate the cause of mild transaminitis and mild thrombocytopenia. This patient refused to undergo surgical resection and chose transcatheter intraarterial therapy instead.
To induce conscious sedation, intravenous midazolam (Versed; Hoffmann-La Roche, Basel, Switzerland) and fentanyl (Sublimaze; Abbott Laboratories, Abbott Park, Ill) were administered to all patients. Midazolam was used to alleviate anxiety and the discomfort of remaining in the supine position for several hours during the procedure. Fentanyl-induced analgesia helped treat abdominal pain resulting from the MTC-DOX injection as well as lower back discomfort from remaining in a prolonged supine position.
Ferrous Carrier and Chemotherapy Agent
MTC-DOX is a suspension of 0.5–5.0 µm of magnetically active iron particles bound to activated carbon. The activated carbon allows subsequent binding of doxorubicin molecules to the iron particles through a reversible process of adsorption (11). A small 5-kG magnet is positioned at the skin surface adjacent to the tumor as MTC-DOX is injected intraarterially. This causes the agent to be drawn from the vascular space into the adjacent tumor tissues (Fig 1). Once outside the vasculature, doxorubicin dissociates from the particles, is absorbed by HCC cells, and results in a tumoricidal response (12). Ultimately, little or no microvascular embolization occurs. The feasibility of magnetic targeting and the safety of this agent have been documented in vivo (12).
|
|
|
|
Therapeutic Intervention
After baseline MR images had been obtained, the patients were transferred to the C-arm conventional angiography component of the dual-modality imaging system. Following sterile preparation of the right portion of the groin, cannulation of the right femoral artery was performed and a 5-F sheath (Cordis, Miami, Fla) was inserted. A Terumo guide wire (Terumo Medical, Somerset, NJ) and a 5-F cobra catheter (Cook, Bloomington, Ind) were advanced into the aorta and used to perform digital subtraction angiography (DSA) of the superior mesenteric, celiac, and common hepatic arteries. Delayed images were obtained to confirm patency of the portal vein in all patients.
After selective catheterization of the proper hepatic artery with the cobra catheter was performed, a coaxial microcatheter (Mass Transit; Cordis) was advanced selectively into the dominant arterial branch(es) supplying the tumor. The treatment of patient 3 required deviation from this protocol when angiography of the celiac artery revealed occlusion of the common hepatic artery. This required that we approach the proper hepatic artery via the superior mesenteric artery, the inferior pancreaticoduodenal artery, and the gastroduodenal artery arcade.
All patients received 60 mg of MTC-DOX suspension during the therapy session. For patients 1, 3, and 4, chemotherapy was divided into two separate injections, while patient 2 required three injections owing to the large volume of the tumor to be treated. MTC-DOX was premixed to a concentration of 0.7 mg/mL and was adminstered in 15-mg boluses at a rate of 2 mL/min.
Before the administration of MTC-DOX, a 5-kG external magnet (Magnet Sales, Culver City, Calif) was positioned over the patient’s abdomen at the approximate location of the tumor. Positioning of the external magnet was performed on the basis of results of preprocedural imaging (MR imaging and/or CT) and results of hepatic arterial angiography performed during the procedure to reveal tumor vascularity. Hepatic arterial angiography was performed in multiple projections to confirm which point on the body surface was closest to the tumor. This position was marked with a permanent marker, and the magnet was positioned on the patient’s skin with the mark at the center of the magnet. The magnet was left in place for 15 minutes after delivery of each dose. This "magnet retention phase" was designed to promote the transfer of the agent from the intravascular space into the adjacent tumor tissue, as was observed by Goodwin et al (12) in their toxicokinetic study. After each dose of MTC-DOX was given, continued patency of a hepatic artery branch supplying the tumor was ascertained by performing repeat DSA of the hepatic artery through the indwelling catheter. Hepatic angiography, MTC-DOX injection, magnet positioning, and postprocedural evaluation of vessel patency were performed by M.W.W., R.K.K., and N.A.F.
After the magnet retention phase was completed and with the hepatic artery catheter still in place, patients were transferred to the adjoining MR imaging unit, and repeat MR images were obtained. Each patient was subsequently transferred back to the C-arm angiography unit, where the hepatic artery catheters were repositioned to maximize the volume of affected tumor. Additional doses of MTC-DOX were administered after the hepatic catheters were repositioned. After the last dose of MTC-DOX, a final set of MR images was obtained. As a result, MR imaging was performed a total of three times in patients 1, 3, and 4 and four times in patient 2. The patients were followed up for survival and tumor response by M.W.W., A.P.V., and J.K.
Data Analysis
Areas of signal intensity loss on the single-shot gradient-echo MR images displayed the least amount of susceptibility artifact associated with iron deposition within the liver. Two radiologists (M.W.W. and N.A.F.) measured iron deposition and the cross-sectional areas of the liver and tumor on transverse single-shot spin-echo MR images at an Impax workstation (Agfa Healthcare, Mortsel, Belgium). The volumes of the liver, the tumor(s), and the signal intensity loss areas were calculated by multiplying the cross-sectional area of each section by the section thickness and adding the individual section volumes. Averages of all measurements were calculated.
The data were reported as the fraction of tumor and fraction of normal liver parenchyma affected by MTC-DOX, as described by the following equations:
|
|
|
Results |
|---|
|
TOPABSTRACTINTRODUCTIONMaterials and Methods ResultsDiscussionREFERENCES |
|---|
The extent of disease was evaluated in all patients on baseline MR images. Viable tumor tissue demonstrated intrinsically high signal intensity on T2-weighted MR images, as well as enhancement on the T1-weighted turbo spin-echo MR images obtained after administration of gadodiamide. Hepatic arterial angiography revealed hypervascular foci corresponding to the lesions detected at MR imaging. In all cases, a major feeding vessel or vessels could be successfully catheterized with the coaxial microcatheter. This allowed intratumoral injection of MTC-DOX in all patients.
Intraprocedural MR images obtained after each dose of the chemotherapeutic agent showed increasing areas of signal intensity loss owing to a magnetic susceptibility artifact caused by the iron. Nevertheless, during the procedure it was still possible to differentiate between regions of normal liver parenchyma with intermediate signal intensity, hyperintense unaffected tumor, and areas of signal intensity loss that demarcated regions of MTC-DOX deposition. In addition to having been deposited in the liver parenchyma, MTC-DOX was observed to have been deposited in the spleens of patients 1 and 3. All other imaged organs appeared to be free of the agent.
A representative case is that of patient 1, who had a 46.0-cm3 lesion in the right lobe of the liver (Fig 3, A). In this patient, the first postinjection MR images revealed incomplete coverage of the tumor with MTC-DOX (Fig 3, C). After the microcatheter was repositioned (Fig 3, B, D), coverage of the remainder of the lesion was successfully accomplished (Fig 3, E).
|
|
After the procedure was finished, the fractions of tumor and normal parenchyma volume affected by each MTC-DOX dose were determined. These data are summarized graphically in Figures 5 and 6. In patients 2–4, the fraction of tumor volume affected by MTC-DOX increased with each subsequent injection of the agent. The second MTC-DOX injection resulted in only a small increase in tumor coverage in patient 1. The final fraction of total tumor volume affected by MTC-DOX was 0.91 (41.8 of 46.0 cm3) in patient 1, 0.77 (281.3 of 365.0 cm3) in patient 2, 0.85 (87.4 of 103.0 cm3) in patient 3, and 0.64 (36.2 of 57.0 cm3) in patient 4. The fraction of normal liver parenchyma volume affected by the agent was 0.14 (171.0 of 1,234.0 cm3) in patient 1, 0.07 (150.4 of 2,148.0 cm3) in patient 2, 0.30 (270.0 of 916.0 cm3) in patient 3, and 0.25 (240.0 of 963.3 cm3) in patient 4.
|
|
Results of patient follow-up were as follows: Patient 1 had a substantial reduction in tumor size (from 5 to 3 cm in maximum diameter) and was still alive 17 months after treatment. Patient 2 was still alive and had a stable tumor size 12 months after treatment. Patient 2 also reported a substantial subjective reduction in tumor-associated abdominal pain during the follow-up period. Patient 3 was still alive and had a stable tumor size 9 months after treatment. Patient 4 was still alive 5 months after treatment, at which time he underwent radiofrequency ablation of the tumor.
|
Discussion |
|---|
|
TOPABSTRACTINTRODUCTIONMaterials and Methods ResultsDiscussionREFERENCES |
|---|
In the present study, tumors were characterized and treated by using both angiography and MR imaging during a single treatment session. Angiography helped characterize the major hepatic artery branches supplying the neoplasm. It was also used for catheter guidance, delivery of MTC-DOX, and documentation of target vessel patency after the injections were completed.
The MR imaging component complemented the angiographic component. It helped define tumor size and location before MTC-DOX administration. After MTC-DOX was delivered, MR imaging enabled immediate intraprocedural evaluation of the location of unaffected tumor, the size of the treated regions, and the extent of affected normal parenchyma. In addition to being deposited in the liver parenchyma, MTC-DOX was observed to have been deposited in the spleens of patients 1 and 3. All other imaged organs appeared to be free of the agent. In a single-dose toxicity study, Goodwin et al (12) reported evidence at postmortem examination of MTC-DOX deposition in areas of the spleen, lung parenchyma, gastric submucosa, pleura, and pericardium.
In the present study, regions of signal intensity loss on all MR images corresponding to MTC-DOX deposition within the tumors increased with each subsequent dose of MTC-DOX. The fraction of tumor volume covered by MTC-DOX at the end of the treatment session ranged from 0.64 to 0.91. Several factors may have diminished the affected tumor volume. Tumors were supplied by multiple hepatic artery branches. We relied on results of conventional angiography to locate the major arterial branches supplying the HCC lesions. Intraprocedural MR imaging revealed whether angiographically identified target branches were likely to be supplying the unaffected areas of the tumor. Only the largest vessels were targeted.
The position of the external magnet may also have affected tumor targeting. Although the location of target vessels varied, the position of the external magnet—overlying the center of the tumor—remained the same. This may have explained the reduced efficiency of lesion targeting in patient 2, who had a very large HCC volume. Altering the external magnet position relative to the particular target area may have increased overall lesion coverage. In patient 4, MTC-DOX covered only 0.64 of the tumor volume. A large necrotic area was evident in the center of the lesion on the initial MR images. At conventional angiography, this area was observed to be devoid of vascular supply. Lack of viable arterial supply to the lesion core is the most likely reason for low tumor coverage in patient 4.
Repositioning the catheter on the basis of intraprocedural MR imaging findings facilitated exclusion of normal liver tissue and improved targeting of the tumor. In patients 1 and 2, distribution of MTC-DOX to normal parenchyma was low. Larger areas of normal tissue were affected in patients 3 and 4; this probably resulted from less-optimal catheter placement or external magnet positioning. However, the intraprocedural MR imaging findings, at the very least, made us aware of this occurrence and provided reassurance that most of the agent was going where we intended it to go.
Drawbacks of the dual MR imaging/conventional angiography system in guiding HCC therapy with MTC-DOX include the long duration of the treatment sessions and the expense of obtaining multiple MR images. In our experience, the length of the treatment sessions was between 5 and 8 hours. During this period, the patient spent most of the time in the angiography suite, where diagnostic angiography, catheter manipulation, and MTC-DOX administration were performed. Radiation exposure lasted 60–90 minutes per treatment session, which was similar to the typical duration of radiation exposure at conventional chemoembolization. Patients spent a minority of the time in the MR imaging unit (approximately 30 minutes for each series of intraprocedural MR images). Owing to the presence of a shielded sliding door between the MR imaging and x-ray units, the MR imaging unit could be made available for other clinical examinations while angiography and MTC-DOX administration were performed.
Abdominal pain was reported by all patients during MTC-DOX injection but not after the procedure. Koda et al (10) reported that 17 of the first 22 patients to receive MTC-DOX experienced similar symptoms. To our knowledge, no report on MTC-DOX published to date provides an explanation for this phenomenon. Ischemia is unlikely to play a role, given the lack of evidence of embolization at postprocedure angiography and the absence of lasting abdominal discomfort.
Radiation dose is an important consideration in evaluating the utility of dual-modality systems. In our study, fluoroscopy and DSA were used intermittently for hepatic arterial angiography, external magnet positioning, catheter repositioning, and documentation of vessel patency after MTC-DOX infusion. Similarly, standard chemoembolization therapy involves hepatic arterial angiography, catheter repositioning, and documentation of the occlusion of supplying vessels. Although no external magnet is involved, fluoroscopy is used to monitor injection of the embolic agent in the latter procedure. Hence, although we did not actually quantify the radiation dose for this procedure, we believe it to be comparable to that of standard chemoembolization therapy.
MR imaging safety was addressed as well. A dedicated safety committee was formed at our institution to assess the safety of the procedures performed in the dual MR imaging/conventional angiography suite. The safety measures instituted by this committee were reviewed and endorsed by the institutional review board of our university. In the case of the portable magnet used for MTC-DOX targeting, it was determined that the sliding shielded doors dividing the MR imaging and x-ray units provided sufficient magnetic field shielding when they were sealed. In addition to providing radiofrequency shielding, the lead-lined doors acted as a physical barrier against an operator inadvertently bringing the portable magnet or metallic objects close to the MR imaging unit. To further increase safety, the switch initiating sliding door separation is kept under a locked panel, which is lifted only when the portable magnet and metallic objects are well outside the range of the magnetic field of the MR imaging unit.
It is important to note that a dual MR imaging/conventional angiography system is not essential for catheter-directed therapy. In fact, MTC-DOX infusion has been successfully performed in several patients without the benefit of such technology (10). Clinical trials to evaluate the efficacy of MTC-DOX are ongoing, and the findings of these trials are beyond the scope of this report.
We anticipate that the future role of MR imaging in guiding intraarterial tumor therapy will expand to include real-time MR imaging guidance of therapeutic agent administration. The feasibility of this process has already been demonstrated in vitro (15). If therapy with MTC-DOX and similar intraarterial therapies prove to be efficacious, use of an MR imaging/conventional angiography system may increase the accuracy of agent delivery and decrease the number of treatment sessions.
|
FOOTNOTES |
|---|
Author contributions: Guarantor of integrity of entire study, M.W.W; study concepts, M.W.W., R.K.K., A.P.V., J.M.L., J.K., R.L.G.; study design, all authors; literature research, M.W.W., N.A.F., A.P.V., J.K.; clinical studies, M.W.W., R.K.K., A.P.V., J.M.L., J.K., R.L.G.; experimental studies, M.W.W., R.K.K., N.A.F., A.P.V., J.K.; data acquisition, M.W.W., R.K.K., N.A.F., J.M.L., R.L.G.; data analysis/interpretation, all authors; statistical analysis, M.W.W., N.A.F.; manuscript preparation and editing, M.W.W., R.K.K., N.A.F.; manuscript revision/review, definition of intellectual content, and final version approval, all authors.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMaterials and Methods ResultsDiscussionREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  A. C. Larson, D. Wang, B. Atassi, K. T. Sato, R. K. Ryu, R. J. Lewandowski, A. A. Nemcek Jr, M. F. Mulcahy, L. M. Kulik, F. H. Miller, et al. Transcatheter Intraarterial Perfusion: MR Monitoring of Chemoembolization for Hepatocellular Carcinoma--Feasibility of Initial Clinical Translation Radiology, March 1, 2008; 246(3): 964 - 971. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. Wang, A. K. Bangash, T. K. Rhee, G. E. Woloschak, T. Paunesku, R. Salem, R. A. Omary, and A. C. Larson Liver Tumors: Monitoring Embolization in Rabbits with VX2 Tumors Transcatheter Intraarterial First-Pass Perfusion MR Imaging Radiology, October 1, 2007; 245(1): 130 - 139. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J.-P. Fortin-Ripoche, M. S. Martina, F. Gazeau, C. Menager, C. Wilhelm, J.-C. Bacri, S. Lesieur, and O. Clement Magnetic Targeting of Magnetoliposomes to Solid Tumors with MR Imaging Monitoring in Mice: Feasibility Radiology, May 1, 2006; 239(2): 415 - 424. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
= patient 1, 
= patient 2, 
= patient 3, x = patient 4.