Developed in Japan in the 1980s, TACE became the most frequent treatment for unresectable hepatocellular carcinoma (HCC) in patients with preserved hepatic function after 2002, when two radiochemotherapies (RCTs) showed survival benefits for HCC patients who underwent conventional Lipiodol-based TACE (cTACE). Nowadays, nearly half of HCC patients undergo this procedure during their clinical history. In the last ten years, cTACE has been challenged by an alternative procedure, drug-eluting beads-TACE (DEB-TACE), after the introduction of calibrated embolizing microspheres loaded with a chemotherapeutic agent. DEB-TACE is considered to be less toxic and better standardized than cTACE, with reported no differences in patient survival. Since 2006, DEB-TACE has become the standard in many centers worldwide. Though, the need of adding doxorubicin to small beads embolization alone (TAE) remains unsettled. Though cTACE/DEB-TACE and TAE have been compared in several RCTs, no study demonstrated a clear survival benefit associated with the former.

Our study aims to compare first-line DEB-TACE and TAE on a random sample of HCC with the hypothesis that the addition of drug to embolization with small size beads is not associated with a survival benefit when compared to embolization alone performed with tiny calibrated microspheres. HCC is considered a chemo-resistant tumor and to date there is no clear evidence of benefits in associating anticancer agents to TAE. On the other hand, the optimal size of embolic agents has still to be defined. A comparative evaluation of TACE and TAE is essential for two additional reasons: a) it is still unclear whether side effects following embolization procedures are related to the embolization itself, to drug addition or both; b) DEB-TACE procedure is more expensive than TAE and, given the current attention on cancer-related health care cost control, identification of opportunities for cost savings in HCC treatments of an increasingly common cancer would be valuable.

Specific Aims

Specific aim 1 Compare time to progression (TTP) treated with TAE and DEB-TACE in a homogeneous hepatocellular carcinoma (HCC) patient population

Specific aim 2 Compare

• safety - severe adverse events (SAE)
• radiologic tumor response (mRECIST)
• overall survival (OS)
• cost-effectiveness after the entire follow-up

Study plan

Experimental design This is an interventional with drug, multicenter, prospective, randomized open label study.

Experimental design Aim 1 The study was designed, in relation to the primary endpoint, as an equivalence trial between two intra-arterial HCC treatments, i.e. DEB-TACE and TAE. The TTP considered as the reference value for the DEB-TACE arm is 9 months, on the basis of the results of our previous multicentric experience. The study is designed as an equivalence trial on the primary endpoint. The standard deviation of TTP is expected not to exceed 6 months, through accurate selection criteria of included patients. The equivalence limit is set to no more than 5 months between the two arms. Thus, using appropriate formulae each arm will be formed by 69 patients (alpha:0.05; beta:0.80). Taking into account a 10% of drop-out, the final sample size per each arm will be of 77 patients (154 total).

Experimental design Aim 2

• Safety. Adverse events (AEs) and severe adverse events (SAEs) will be monitored and recorded. AE will be assessed during and after each treatment and at all follow-up visits and graded according to the NCI Common Terminology Criteria for AE (CTCAE) version 3.0. The AEs occurring within 4 weeks after TACE will be considered as treatment related. SAE incidence in both arms will be ascertained assuming a reduction of SAE from 25% after TACE to 19% after TAE (25% of reduction). Such an assumption would require 607 patients per arm for being confirmed. To accomplish this task, the O'Brien-Fleming stopping boundaries will be applied considering an analysis when 69 patients for each arm will be obtained and follow-up ended. Applying stopping boundaries to this sample size, a nominal p-value <0.001 (z-score: 3.15) is needed to confirm the assumption. If the p-value will be above this threshold (even if <0.05) the null hypothesis will be not refused.
• Efficacy. The response will be evaluated by CT or MRI as local (per-lesion) response and overall (per-patient) response, according to mRECIST, at 1 month after each TACE and, thereafter, every 3 months, for at least 2 years.
• Survival. Through the selection previously described, we expected a standard deviation of the mean survival of no more than 10 months. To design the present equivalence trial we also expected a difference in the mean survival not greater than 5 months (equivalence limit) between the two groups. Using formulas proposed by Julious et al., each arm will be formed by 69 patients (alpha: 0.05; beta: 0.80). To obtain more robust estimate and accounting for possible dropout from the study a 10% of patients will be added to the initial sample size, resulting in 77 patients for each arm.
• Cost-effectiveness will be assessed from a third payer perspective, thus, including only direct costs of the procedures and related costs (hospitalization, imaging, etc.). The effectiveness will be assessed by measuring the life-expectancy. Incremental cost-effectiveness ratio (ICER) will be used to evaluate the cost-effectiveness of one treatment over the counterpart.

Study population The study involves the enrollment of a total of 154 patients, 77 per randomization arm, with a HCC diagnosis, according to the guidelines of the American Association for the Study of the Liver Disease (AASLD), as in clinical practice. The enrolled patients will have to meet the inclusion criteria and sign the informed consent for the participation in this study. At the time of enrollment, demographic, clinical and radiological data will be collected, as in standard clinical practice.

Treatment

Treatment study In the present study, only patients in the DEB-TACE arm (Arm A) will receive intra-arterial hepatic chemotherapy (Doxorubicin). Both treatment arms, on the other hand, will receive arterial embolization of the branches that vascularize the tumor lesions.

Arm A: DEB-TACE or chemoembolization with microspheres. The chemotherapy used in this arm is the Doxorubicin that will be carried into the tumor by Embozene TANDEM® (Boston Scientific) microspheres. TANDEM® embozene microspheres are made of non-resorbable, biocompatible, hydrogel microspheres, subjected to precision calibration and coated with an inorganic perfluorate polymer (Polyzene®-F). Thanks to their design, the microspheres can be loaded with drugs, such as doxorubicin, in order to administer a local, controlled and constant dose of the drug to the tumor sites affected after embolization.

TANDEM® embozene microspheres are available in three different sizes, in 2 ml and 3 ml volumes of product and are supplied in preloaded vials and syringes.The maximum loadable amount on the hydrospheres 50 mg of Doxorubicina for ml. The maximum injectable dose of Doxorubicin for each treatment is 150 mg and therefore the maximum amount of microspheres that can be used for each treatment is 3 ml. The size of microspheres that will be used in this study is up to 100 ± 25 μm.

Arm B: TAE or embolization with microspheres. The TAE will be performed with Embozene microspheres (Boston Scientific). Embozene microspheres are spherical particles of hydrogel, precisely calibrated, biocompatible, non-absorbable and coated with a perfluorinated inorganic polymer (Polyzene-F). This medical device is available in various sizes; in this study, to avoid that the particle size exceeds the one used for the treatment of arm A, it will be possible to use microspheres with dimensions up to 100 μm (range 75 ± 125 μm).

Embozene microspheres are offered in vials containing 1 ml of suspended product in physiological saline solution for apyrogenic sterile transport. The total volume of the Embozene microspheres, including the transport solution, is about 7 ml. To this product it is necessary to add an appropriate amount of non-ionic contrast medium in order to obtain a homogeneous suspension and with good visibility during the injection under fluoroscopy.

In enrolled patients, regardless of the treatment arm, the possible presence of extra-hepatic shunts, such as gastric and pulmonary arteries arising from the hepatic branches, should be appropriately assessed. As in standard radiological practice in case of intra-arterial hepatic treatments, if these shunts are macroscopically evident, the preventive closure of the same vessels will be evaluated, using the devices that will be more suitable on a case-by-case basis. If these shunts are not treatable by occlusive devices, the enrolled patient will not be treated in the clinical study and the patient will be considered as drop-out. Both treatments will be repeated "on demand" after demonstration to the imaging of the presence of vital tumor ie absence of complete response (complete response, CR) or in case of intrahepatic distal recurrence at follow-up.

Safety assessment The evaluation of the safety of the experimental treatment will consist in the monitoring and recording of AEs and SAEs.

Inclusion Criteria:

- First level eligibility criteria include

1. HCC diagnosis according to the AASLD criteria [26];
2. patients ≥18 years;
3. HCC unsuitable for curative treatment or had failed/recurred after resection/ablation diagnosed according to the AASLD criteria;
4. no previous treatment in target lesions (prior treatments including resection on non-target lesions will be accepted);
5. Child-Pugh class A or B (max score 7);
6. ECOG Performance Status (PS) <2;
7. target lesion measurable according to mRECIST.

Second level eligibility criteria include:

1. The modified hepatoma arterial-embolization prognostic score (m-HAP-II), based on bilirubin, albumin, serum alpha fetoprotein, tumor number and tumor size, divides patients in 4 classes (A, B, C, D) with different survivals and is useful for prognosis stratification. The first selection criteria will be m-HAP-II classes B or C fulfilment.

2. The UNOS/TNM stage: only patients with T1, T2, T3 and T4 tumors will be included.

   These two main criteria will be used for stratification of patients prior to randomization in order to obtain identical prevalence of m-HAP-II classes B/C and of UNOS/TNM stages from T1 to T4a

3. obtaining of the informed consent.

Exclusion Criteria:

1. infiltrative HCC;
2. neoplastic branch or main portal vein invasion;
3. equivocal hepatic lesion;
4. advanced liver disease (bilirubin levels >2.5 mg dl-1, albumin <30 g l-1, platelets <50 x 109/L, INR >1.5);
5. ascites and/or F3 oesophageal varices;
6. other tumors in the previous 5 years;
7. technical contraindications to arteriography or TACE.