Hepatic Malignancies: Rationale for Local and Regional Therapies

Thursday, March 31, 2005 Clinical Oncology: What the IR Needs to Know (SY) Coordinators: Thierry de Baere, MD, Damian Dupuy, MD, Catherine M. Tuite, MD, Michael C. Soulen, MD

Objectives: Upon completion of this symposium, the attendee should be able to: 1. Describe the biology, epidemiology, and natural histOly of common solid tumors.

2. Work up, diagnose and stage patients with cancers who may be candidates for image-guided therapy.

1:40 p.m. The Original Image-Guided Therapy: The Role of Radiation in Tumors Amenable to IGT Thomas DiPetril!o, MD Rhode Island Hospital ProVidence, Rl 2:00 p.m. BREAK

Who To Treat, When and How: Integrating IGT into the Global Care of Cancer Patients Moderator: Damian E. DUpuy, MD

3. Integrate image-guided therapy (IUD with chemotherapy, radiation, and surgery for specific solid cancers.

2:15 p.m.

4. Clinically evaluate and care for patients with cancer and cancer treatment-related toxicities.

Hepatic Metastases: Medical Oncology Paolo Hoff, MD 2:30 p.m.

Principles Of Oncology: What You Need to Know for the Initial Consultation And Why It Matters Moderator: Catherine M. TUite, MD 12:00 p.m. General Assessment of the Cancer Patient Catherine M. Tuite, MD Hospital of the University of Pennsylvania Philadelphia, PA 12:20 p.m. Assessment of the HCC Patient Riad Salem, MD, MBA Northwestern Memorial Hospital Chicago,IL 12:40 p.m. Chemotherapy: What We Use, What They Use, and Why You Should Care Paolo Hoff, MD 1:00 p.m. When and How to Image for Tumor Response: cr/MRI After IGT David Lu, MD Dumont UCLA Liver Cancer Centre CA See Limanond P, Zimmerman P, Raman SS, Kadell BM, Lu DS. Interpretation of CT and MRI after radiofrequency ablation of hepatic malignancies. AJR 2003; 181:1635-1640 1:20 p.m. When and How to Image for Tumor Response: PET After IGT Homer Macapinlac, MD

P4

Hepatic Malignancies: Rationale for Local and Regional Therapies Thierry de Baere, MD Institut Gustave Roussy Villejuif, France Due to the relative inefficacy of general treatment of liver tumors, there is a large place for so called "local" and "regional" therapies in this field. It is very difficult to make a clear cut difference between what is called a "local" treatment and a "regional" treatment and both words are used in the literature without clear significance. Local most often means targeting the tumor, while regional means targeting the organ or the region of the disease. The goal of all these treatments is to target the tumor as accurately and selectively as possible. Because we are not able to be so selective with the tumor, we enlarged treatment to healthy parenchyma around it. If we thought about ablative therapies, we took safety margins, and somewhat transformed a local to loco-regional treatment. If we thought about chemoembolization, we tried to be as selective as possible to go from a regional treatment to the liver into a local treatment by targeting the lobe, the segment or even the subsegment bearing the tumor according to our technical possibilities, and probably providing a loco-regional treatment as well. Consequently, we will use the term loco-regional in the syllabus both for ablative and intra-arterial techniques even if their rational is different. Numerous and various loco-regional treatments have been used for many years in cancer management including: radiation therapy, brachytherapy, regional chemotherapy delivery (intra-arterial, intra-peritoneal, ...), and obviously surgery. These treatments can provide dramatic results and cure the patient in some occasions when the disease is limited. Indeed, surgery, a locoregional treatment, provides the best hope for cure and

disease free survival in large series of patients bearing liver tumor. However, only a small subset of patients are amenable to surgery. Among patients bearing primary or metastatic liver tumors, no more than 20% will undergo a surgical resection. Furthermore, 70% of patients who underwent surgery will develop new liver tumors during the first 3 years of follow-up. These recurrences will be most often distant from the site of resection, but local recurrences at the site of resection are also reported. One drawback of surgery is that this technique is most often a one shot technique due to higher rate of complications, and higher technical difficulties for subsequent hepatectomies, even if iterative hepatectomies have been reported feasible and effective. Consequently, there is a place for a local treatment that can be easily repeated in a disease that will recure in more than half of patients. Nonresectable liver tumors are often targeted with various local treatments either because the disease is located only in the liver, or the disease is predominant to the liver and life expectancy is relied to liver tumor control. In metastatic disease, the liver is often a stepwise pattern of metastatic progression, and local treatment will often be associated with systemic treatment to prevent the spread of disease to other organs. For example, lungs have been reported as a common location of occurrence of new tumors or progression during intraarterial hepatic chemotherapy performed for metastases from colon cancer even when local efficacy have been proven. Livers can be targeted through a vascular approach, injecting the active product through feeding vessels or with a direct approach to the tumor by delivering the trearment inside of the parenchyma. Vascular Approach Intra-arterial injection takes advantage of the fact that the liver vascularization is 30% arterial and 70% portal, while tumors growing in the liver are fed nearly exclusively by the arterial blood. Thus a drug injected in the hepatic artery will reach the rumor, and occlusion of arterial branches will cause severe ischemia to the tumors with minor damages to the healthy liver, due to the residual blood supply to the healthy liver through the portal vein. However, there are major differences between various tumors. For example, HCC and metastases from neuroendocrine tumors (NET) share a very important arterial feeding while metastases from colorectal cancer demonstrate a poor arterial supply, even if this arterial supply is still predominant over the portal supply. Thus, HCC and NET metastases will be easier to target with embolic material than colorectal cancer metastases. At last, there is an exception to arterial supply to liver tumors at the very early phase of the metastatic process, when seeding cells reach the liver through the portal circulation and are fed by portal blood. The chemotherapeutic drugs that will be injected intravascular to the liver are the same ones that are available for systemic use. Indeed, no drug today is

specially designed for intra-arterial use. Various carriers will be used trying to improve their pharmacokinetic. Carriers will also be used in order to deliver radiation therapy as selectively as possible to the liver. Hepatic arterial chemotherapy, transarterial chemoembolization, hepatic injection of loaded emboIs (drug and radiation), inb'a-portal drug delivelY, and isolated liver perfusion are the main techniques using a vascular approach to the liver. Intra-arterial Hepatic Chemotherapy (IAHC) IAHC has the main advantage of increasing drug concentrations in tumor deposits, thus resulting in a significant increase in response rates because many tumors display a steep dose-response curve. The advantage for such an intra-arterial route is proportional to first pass extraction of the drug by the liver and inversely proportional to body clearance of the drug. Consequently, the choice of the drug is of utmost importance. FUDR has been extensively used for IAHC because it is extracted by the liver at more than 95% during the first pass, with an increase of exposure of the liver 100 to 300 times higher than with a systemic perfusion. When compared to IV perfusion the estimated increase in liver exposure by IAHC are about 20 fold for THP adryamicine, 5 to 10 fold for 5FU, 4 to 7-fold for cisplatin, 6 to 8 fold for mitomycine, 4 fold for oxaliplatinum, and only 2 fold for doxorubicin. All clinical trials using 5 FU or FUDR have demonstrated a better response rate for IAHC than for IV treatments. However, only two trials have demonstrated a moderate benefit in survival OJ. Intra-arterial was more or less abandoned at the time IV irinotecan and oxaliplatinum proved to give equivalent response rate to intra-arterial 5-FU. Recently a French multicentric trial has used these new drugs intrarterially. IAHC was performed using 100 mg/m2 of oxaliplatinum repeated every second week with overall response rate of 64% (95%C1: 440/0-81%) in heavily pretreated patients. In this study, a very promising 45% response rate has been observed in patients sustaining a progression following intravenous treatment with the same drug [2]. Today percutaneous placed ports prOVide an equivalent patency to surgical ones with respective patency of 6.5 ::':: 4.2 and 4.3 ::':: 3.4 courses of chemotherapy in the only comparative study published [3], The largest series of percutaneous catheter implantations reported a patency rate of 91%, 81%, and 58% at respectively 6 months, 1 year and 2 years, allOWing 3 to 102 courses of chemotherapy (mean = 35) (40). Consequently IAHC can be safely and effiCiently delivered without surgical implantation. In our practice there are two main populations who benefit from IAHC through percutaneous implanted ports. Patients who do not respond to one or two lines of systemic chemotherapy, and we have underlined that the chances of responding to a second line after failure of a FolFox or a FolFiri is below 7%. The second population is made of patients who are borderline for resection and in whom a maximum efficacy is wanted to

P5

make them respectable. IAHC can also be thought as an adjuvant therapy to local ablation as it will be discussed later on this syllanus. Trans-arterial Chemo-embolization (TACE) In fact, there are many various techniques under the word TACE. The most common procedure is the intraarterial injection of chemotherapy emulsified with Lipiodol® followed by injection of embolic material. The rational for using the emulsion of an oily contrast medium and aqueous chemotherapy instead of the pure drug is due to the fact that oil drops injected in the arterial flow have a propensity to go through the largest arteries without entering the small ones due to tensioactif surface forces as demonstrated by in vivo microscopy studies [4]. Because hypervascular tumors have larger vessels, Iipiodol will go preferentially to the tumors. This selectivity to the tumor is measured in an experimental model to be between 4 and 10 fold higher than for nonnalliver, with the best results obtained for the water in oil emulsions [5]. Another advantage of water in oil emulsions over oil in water emulsion is that in the internal phase of the emulsion, the drug is contained inside of the oil drops, and will then reach tumor vessels where it will be released [4]. In such a setting, ratio of drug concentration in the tumor to the healthy liver can be as high as 100 [6, 7J. Furthermore, Lipiodol is responsible for temporary embolization of the arteries and of the portal system [8], in that it reaches through the peribilairy anastomosis [9], thus increasing dwell time between drug and tumor. This oily portal embolization can damage healthy liver parenchyma when volumes higher than 20 ml are injected [10], At last Lipiodol can go through the endothelium of tumor vessels, and is found in the interstitial space as well as in tumor cells [11]. Embolization after chemo-lipiodol increase the efficacy of treatment by providing additional ischemia to the highly hypervascularized tumor usually targeted with this treatment. Such embolizations have been reported to induce failure of transmembrane pumps, thus increasing drug retention inside of cells [12]. This combination of chemolipiodol + embolization provides longer survival than embolization alone in hepatocellular carcinoma patients (3).

Loaded Embols Recently two main types of loaded embols have been proposed: glass spheres loaded with Ytrium-199 and gelatin spheres loaded with doxorubicin. The rationale of these loaded embols is to concentrate a therapeutic stimuli (radiation therapy or toxic drug) in the liver by trapping the embols in the arterial circulation. Ischemia produced by embolization is supposed to increase drug toxicity or the effect of radiation therapies. Very little is known about the ideal size and volume of the embol needed to achieve the desired level of ischemia to the tumor, and to provide the best selectivity possible to the tumor when compared to healthy liver. The pharmaco-

P6

kinetic advantage to the tumor using gelatin spheres loaded with doxorubicin was demonstrated in an experimental model of a VX2 tumor where a peak concentration of 400 nMoles of doxorubicin per gram of tissue was measured at Day 4, while no doxorubicin was measured in the tissue after intra-arterial injection of doxorubicin alone (J Geschwind, unpublished data). Glass spheres loaded with Yttrium-90 have been used since the late 1980's. Yttrium-90 is a pure beta emitter that allows an average range of radiation in tissue of 2.5 mm with a maximum of 1 em. This allows us to release the patient immediately after treatment. The target dose of irradiation is 100-150 Gy, because therapeutic efficacy has been demonstrated superior for patients receiving more than 100 Gy [14]. Assessment of the arterial shunt to the lung is crucial because lungs cannot tolerate more than 30 Gy, and lung toxicity was a cause of death. At last, 13IIodine-labelled lipiodol has been injected intra-arterially for nonresectable HCC with portal vein thrombosis trying to take advantage of the selectivity of Lipiodol for the tumor in order to provide preferential radiation therapy. Increase of survival was very mild [15] and toxicity was very high in these fragile patients [16], On the other hand, adjuvant intra-arterial 13IIodinelabelled lipiodol after complete resection of HCC has been demonstrated to lower recurrences from 59% to 28% at 34 months in a randomized trial [17], and this was a trend confirmed in a second trial (18). Intra-portal Chemotherapy Intra-portal chemotherapy has been experimented with in order to avoid seeding and implantation of tumor cells through the ponal branches. It has been used as a postoperative treatment, after surgical resection of primary colon cancer, because it has been established that during surgical resection, the number of tumor cells circulating increases, and theoretically increases the risk of seeding to the liver. Postoperative portal vein infusion via a catheter implanted by the surgeon in the portal vein system increased the 5-year survival in a small series, but this benefit was not confirmed in two large randomized trials. Direct Approach Rational for direct approach of liver tumors is to deliver a therapeutic effect or product that cannot be sustained by systemic therapy and even by intra-arterial delively, to the tumor. Today most of the treatments are physical destruction, and radiofrequency ranks first among them because it has been demonstrated to be more efficient than chemical treatment either in metastases or hepatocellular carcinomas. In addition, radiofrequency provides a more significantly tumor free survival than alcohol injection [19]. Physical destruction is superior because the area of distribution of the therapeutic effect is more precise than with chemical treatment. Nevertheless tllese physical treatments remain limited in size because diffusion of the therapeutic effects far from the

applicator is difficult to obtain. Consequently, size of the target tumor is a main limitation to success of these physical destructions. Overlapping ablations are difficult to precisely deliver to cover a complete volume, and are time consuming, and did not proof efficiency equal to single application. Trying to overcome these difficulties, some researchers have tried to obtain a synergy between a physical destruction and another treatment which will favorize tissue destruction where the physical treatment provides only sub lethal effect, namely in a rim at the periphery of the destruction zone. It has been demonstrated experimentally and in clinical practice that IV chemotherapy combined to RF can increase the volume of ablation [20). The rational of such a direct approach is very close to surgical resection, but instead of removing the tumor, it destroys it and leaves it in place. Thus imaging follow-up is of major importance and no imaging technique today has demonstrated accuracy in predictive efficacy early after treatment. Consequently, follow-up is performed during several months before treatment success can be confirmed. The advantages of ablation compared to surgery are a lower morbidity and mortality. Main disadvantages of these ablation techniques are to be very dependant on imaging guidance, size, and location of the tumor. The ideal surgery or the ideal ablation must be able to remove or to destroy all targeted tumors. Both surgery and ablation sometimes failed, and we recently compared evaluation of radiofrequency, wedge resection and anatomical resection in our surgical experience of radiofrequency. The liver metastases were treated by anatomic resection (40 patients, 213 LMs) when large, by wedge resection (64 patients, 99 LMs) when peripheral and small, and by RFA (88 patients, 227 LMs) when central and small. The median follow-up was 27.6 months (range, 15-74 months), and a total of 539 LMs were treated (median of 5 per patient). The local recurrence rates were 5.7% for the 227 RFAs, 7.1% for the 99 wedge resections, and 12.5% for the 40 anatomic resections. These very promising results of radiofrequency have to be considered taking into account that the patients in this series had major tumor burden, anatomical resection was difficult, and wedge resection tried to save as much liver parenchyma as possible. Because local treatment most often targets tumor in patients with advanced disease, namely metastatic patients, adjuvant chemotherapy to fight against nondetectable disease must be considered. The advantage of adjuvant chemotherapy has been demonstrated very clearly in a randomized trial for intra-arterial adjuvant chemotherapy [21) after complete resection of colon cancer metastases, but seems real also for IV chemotherapy. Advantage of systemic chemotherapy is to fight against liver disease undepicted at the time of local treatment, as well as extra-hepatic disease which is often the cause of progression when intra-arterial chemotherapy alone is used [22).

Conclusion Locoregional treatments often provide very good local or regional tumor control. They carry the best hopes for cure. Combination with systemic adjuvant treatment should be considered to increase local efficacy and to lower distant relapse. Today they are used in patients with highly advanced cancer, and must target in the future an earlier stage of disease which is currently treated with surgery. Trials are needed for validation but randomized trials will be difficult to build.

References 1. Meta-Analysis Group in Cancer. Reappraisal of he-

patic arterial infusion in the treatment of nonresectable liver metastases from colorectal cancer. ] Natl Cancer Inst 1996; 252-258 2. Boige V, Lacombe S, de Baere T. Hepatic arterial

infusion of oxaliplatin combined with 5FU and folinic acid in non resectable liver metastasis of colorectal cancer: a promising option for failures to systemic chemotherapy. ]CO 2003; 22 Proceeding of ASCO 2003: 291 3. Aldrighetti L, Arru M, Angeli E, et al. Percutaneous vs. surgical placement of hepatic altery indwelling catheters for regional chemotherapy. Hepatogastroenterology 2002; 49: 513-517 4. de Baere T, Dufaux], Roche A, et al. Circulatory alterations induced by intra-arterial injection of iodized oil and emulsions of iodized oil and Doxorubicin: Experimental study. Radiology 1995; 194: 16S170 5. de Baere T, Zhang X, Aubert B, et al. Quantification of tumor uptake of iodized oils and emulsions of iodized oils: Experimental study. Radiology 1996; 731-735 6. Konno T. Targeting chemotherapy for hepatoma: arterial administration of anticancer drugs dissolved in Lipiodol. Eur.]. Cancer 1992; 28: 403-409 7. Egawa H, Maki A, Mori K, et al. Effects of intraarterial chemotherapy with a new lipophilic anticancer agent, estradiol-chlorambucil (KM2210), dissolved in Iipiodol on experimental liver tumor in rats.] Surg Oncol 1990; 44: 109-14 8. de Baere T, Denys A, Briquet R, Chevalier P, Laurent A, Roche A. Modification of arterial and portal hemodynamic after injection of iodized oil in the hepatic artery: experimental study.] Vasc Interv Radiol 1998; 9: 305-310 9. Kan Z, Ivancev K, Lunderquist A. Peribiliary plexaImportant pathways for shunting of iodized oil and silicon rubber solution from the hepatic artery to the portal vein. An experimental study in rat. Invest Radiol 1994; 29: 671-676 10. Chung], Park], Han], et al. Hepatic tumors: Predis-

P7

posing factors for complications of transcatheter oily chemoembolization. Radiology 1996; 198: 33-40 11. Bhattacharya S, Dhillon AP, Winslet MC, et al. Human liver cancer cells and endothelial cells incorporate iodised oil. Br] Cancer 1996; 73: 877-81 12. Kmskal ]B, Hlatky L, Hahnfeldt P, Teramoto K, Stokes KR, Clouse ME. In vivo and in vitro analysis of the effectiveness of doxombicin combined with temporary arterial occlusion in liver tumors. ] Vasc Interv Radiol 1993; 4: 741-7 13. Llovet ]M, Real MI, Montana X, et al. Arterial embolisation or chemoembolisation versus symptomatic treatment in patients with unresectable hepatocellular carcinoma: a randomised controlled trial. Lancet 2002; 359: 1734-1739 14. Salem R, Lewandowski R, Roberts C, et al. Use of Yttrium-90 glass microspheres (TheraSphere) for the treatment of unresectable hepatocellular carcinoma in patients with portal vein thrombosis.] Vasc Interv Radiol 2004; 15: 335-345 15. Raoul]L, Guyader D, Bretagne]F, et al. Randomized controlled trial for hepatocellular carcinoma with portal vein thrombosis: intra-arterial iodine-131-iodized oil versus medical support. ] Nucl Med 1994; 35: 1782-1787 16. de Baere T, Taourel P, Tubiana ], et al. Hepatic intra-arterial 131-I-iodized oil for treatment of hepatocellular carcinoma with portal thrombosis. Radiology 1999; 212: 665-668 17. Lau WY, Leung TW, Ho SK, et al. Adjuvant intraarterial iodine-131-1abelled lipiodol for resectable hepatocellular carcinoma: a prospective randomised trial. Lancet 1999; 353: 797-801 18. Boucher E, Corbinais S, Rolland Y, et al. Adjuvant intra-arterial injection of iodine-131-labeled lipiodol after resection of hepatocellular carcinoma. Hepatology 2003; 38: 1237-1241 19. Lencioni RA, Allgaier HP, Cioni D, et al. Small hepatocellular carcinoma in cirrhosis: randomized comparison of radio-frequency thermal ablation versus percutaneous ethanol injection. Radiology 2003; 228: 235-240. 20. Goldberg SN, Kamel IR, Kruskal ]B, et al. Radiofrequency ablation of hepatic tumors: increased tumor destruction with adjuvant liposomal doxorubicin therapy. A]R Am] Roentgenol2002; 179: 93--101. 21. Kemeny N, Huang Y, Cohen Al\1, et al. Hepatic arterial infusion of chemotherapy after resection of hepatic metastases from colorectal cancer. N Engl ] Med 1999; 341: 2039-2048 22. Fallik D, Ychou M, ]acob], et al. Hepatic arterial infusion using pirarubicin combined with systemic chemotherapy: a phase II study in patients with

P8

nonresectable liver metastases from colorectal cancer. Ann Oncol 2003; 14: 856-63

2:45 p.m. Renal: Urologic Therapies TED 3:00 p.m.

Renal: Ablation and Embolization Timothy W.I. Clark, MD Hospital of the University Of Pennsylvania Philadelphia, PA

3:15 p.m. Lung: Medical Oncology Neil E. Ready, MD Lung cancer is a common and virulent disease with an estimated 164,000 new cases and 156,900 deaths annually in the US (1). The magnitude of the burden of lung cancer is exemplified by the fact that the number of deaths annually from lung cancer approximates the total number of deaths from the second through fifth leading causes of cancer mortality combined (colorectal, breast, prostate and pancreatic cancers). Non-small cell carcinomas (adenocarcinoma, squamous cell and large cell carcinomas) account for 80-85% of lung cancers and have been traditionally considered collectively for purposes of treatment. As our experience with molecular-targeted therapy grows, future therapy may be directed at specific molecular abnormalities associated with the different subtypes of non-small cell lung carcinoma (NSCLC). This review will cover the medical and multi-modality treatment of NSCLC. Stage I cancers have no lymph node involvement and stage II cancers have involvement of intrapulmonary or ipsilateral hilar lymph nodes. Complete surgical resection with a minimum of a lobectomy is the treatment of choice for patients with stage I and II NSCLC. Conformal high dose radiation, wedge resection and image guided ablation are some of the modalities being studied for the treatment of early stage NSCLC that can not be treated with standard surgery due to severe comorbid medical disease. Five-year survival is apprOXimately 60-70% for stage I and 40-50% for stage II cancers following complete surgical resection. Many patients will relapse following surgery. The results from recently reported large randomized trials have shown that adjuvant chemotherapy improves disease free and overall survival. IALT was a large randomized, multinational trial for stage IB, II and IlIA NSCLC in which 1800 patients were randomized to observation or several cycles of cisplatin based chemotherapy (2). IAiT showed a statistically significant 5% improvement in disease free survival and 4% improvement in over all survival. The NCI Canada trial BRIO evaluated adjuvant cisplatin and vinorelbine combination chemotherapy versus observation in stage IB and II NSCLC (3). The arm that received adjuvant treatment had a 15% improvement in overall survival due to a de-