Transhepatic arterial infusion chemotherapy using a combination of miriplatin and CDDP powder versus miriplatin alone in the treatment of hepatocellular carcinoma: a randomized controlled trial

This multicenter, open-label, randomized phase II trial was designed to compare the efficacy of miriplatin-TOCE monotherapy and miriplatin-TOCE/DDP-H-HAIC combination therapy and was conducted across 3 Japanese institutions. This study was approved by each Institutional Review Board and was registered on the University Hospital Medical Information Network Clinical Trials Registry ( http://​www.​umin.​ac.​jp/​ctr/​index.​htm, UMIN000004691). Written informed consent for publication of their clinical details and/or clinical images was obtained from the patient. A copy of the consent form is available for review by the Editor of this journal. The study protocol conformed to the ethical guidance of the 1975 Declaration of Helsinki.

Patients with HCC were considered eligible for the study if they fulfilled the following criteria: histologically and/or clinically diagnosed HCC: no indication or consent for surgical resection, radiofrequency ablation, radiation therapies, or liver transplantation according to the Japanese guidelines [ 12]; 20–80 years of age; at least one tumor larger than 10 mm that could be measured by CT and/or MRI as an enhanced nodule; an Eastern Cooperative Oncology performance status of 0–2; preserved organ function using indicators within the liver (Child–Pugh, score ≤ 8; total bilirubin, ≤3.0 mg/dl; albumin, ≥3.0 g/dl), blood (neutrophils, ≥1500/mm ³; platelets, ≥50,000/mm ³; hemoglobin, ≥8.0 g/dl), and kidneys (creatinine clearance, ≥50 ml/min adjusted for 1.73 m ² of body surface area by measuring creatinine concentrations in the blood and urine collected for a day); serum amylase, ≤324 IU/dl; and an interval of 4 weeks or longer after a previous treatment for HCC. Patients with the following features were considered ineligible: active infectious disease; massive pleural effusion and/or ascites refractory to treatments; active bleeding from the gastrointestinal tract; active concomitant cancer with invasive characteristics; severe mental disorder or hepatic encephalopathy; history of allergic reaction to iodine phase contrast and/or platinum agents; pregnant or lactating female; ongoing interferon therapy; difficulty with oral food intake; and other serious conditions judged to be inadequate by responsible physicians.

Under a local anesthesia, a 3 to 5-Fr hook or cobra-shaped catheter (GADELIUS Medical K.K., Japan; Terumo Corporation, Japan; Hanaco Medical Co., Ltd., Japan,) was introduced from the right femoral artery depending on arterial curvatures as anticipated by CT images and was anchored at the celiac trunk or common hepatic artery for the further insertion of 1.8/2.4-Fr (Carnelian Marvel, Tokai Medical Products, Japan) or 2.2/2.9-Fr (Carnelian ER, Tokai Medical Products, Japan) micro catheter, which was coaxially aligned with a 0.016-in. Radifocus™ wire (Terumo Corp., Tokyo, Japan). Patients were randomly assigned to receive either miriplatin monotherapy (TOCE) or miriplatin/DDP-H combination therapy (TOCE + HAIC). In a case of the TOCE group, a tip of the micro catheter was advanced as near as possible to the tumors, and 20 mg/ml of miriplatin dissolved in lipiodol was administered up to 120 mg/body, which was divided if the tumors were supplied via multiple vessels. For the TOCE + HAIC group, DDP-H was dissolved in saline at a concentration of 1.43 mg/ml and administered as HAIC at a rate of 3 mg/min through the hepatic arteries supplying the lobes of the liver containing tumors. If the tumors were located in the posterior, anterior, or both segments, a tip of the micro catheter was placed at the right hepatic artery, while the tip was placed at the left hepatic artery if the tumors were localized in the lateral, medial, or both segments. If the tumors were spread in both lobes, the tip of the catheter was positioned at the proper hepatic artery. After the administration of DDP-H, a successive TOCE was performed using miriplatin as described above. The dose of DDP-H was determined based on renal function. When creatinine clearance was 100 ml/min or less, DDP-H was infused up to the value of creatinine clearance [ 13]. For example, when creatinine clearance was 70 ml/min, 70 mg of DDP-H was infused in total. The highest dose of DDP-H was 100 mg/body. The corresponding treatment of each enrolled arm was repeated when the following CT/MRI study showed a stable disease/partial response or when hypervascular nodules appeared again in a case achieving complete response.

The anti-tumor response was evaluated by comparing CT/MRI images obtained before treatment and 3 months after treatment. The evaluation was performed in accordance with the modified Response Evaluation Criteria in Solid Tumors guideline [ 14]. The disease control rate (DCR) was defined as the percentage of treatment sessions in which anti-tumor effects were judged to be either a complete response, partial response or stable disease. The National Cancer Institute Common Terminology Criteria for Adverse Events (NCI-CTCAE) version 4.0 was used to assess toxicity. The corresponding treatment of the enrolled arm was repeated until disease progression and/or adverse events of grade ≥ 3 non-hematological or grade ≥ 4 hematological toxicities were observed. To evaluate the impact of each treatment on functional hepatic reserve, serum biomarkers were analyzed prior to each treatment session and 3 months after the final treatment. The values were analyzed after conversion to a percentage of the value prior to the initial treatment. We assessed three tumor markers, α-fetoprotein (AFP), des-γ-carboxy prothrombin (DCP), and the percentage of fucosylated AFP versus total AFP (L3%). AFP and DCP were converted first to a log scale value and then to a percentage value against the value prior to the initial treatment, while L3% was analyzed without any conversion. The primary endpoint was progression free survival (PFS), and DCR, overall survival (OS), the number of repetitive treatment sessions, the interval between the sessions, adverse events, and the dose of administered miriplatin were calculated at the time of the primary endpoint.

From 2010 to 2013, a total of 21 patients were screened for eligibility (Fig. 1). Based on a pretreatment evaluation (see Methods), 20 eligible patients were enrolled in this study and randomly assigned to one of the 2 study arms, either the TOCE monotherapy or TOCE + HAIC combination therapy. Among the 11 patients allocated to the TOCE + HAIC, one case was excluded from further follow-up because radio frequency ablation therapy was performed on the target before the first efficacy evaluation. The baseline characteristics of the patients showed no statistically significant differences between the 2 arms in terms of age, gender, performance status, etiology of background diseases, presence or absence of liver cirrhosis, Child-Pugh grade, tumor size, tumor number, TNM stage, stage of Barcelona-Clinic Liver Cancer staging classification (BCLC), or treatment history using DDP-H or TACE (Table 1).

The details of the treatments are summarized in Fig. 1 and Table 2. Three out of nine patients on TOCE (33%) and seven out of ten patients on TOCE + HAIC (70%) who received corresponding TOCE or TOCE + HAIC treatments repeatedly for fifteen or thirty-six sessions, respectively, approached a significant difference ( p = 0.057). The intervals between sessions, 94.5 and 90.5 days in TOCE and TOCE + HAIC, respectively, showed no significant difference ( p = 0.96). The dosage of miriplatin used in each session was not significantly different between the 2 arms ( p = 0.99).

The median periods of PFS were 91 days (range, 29–322 days) and 423 days (range, 34–1243 days) in the TOCE and TOCE + HAIC groups, respectively, with PFS being significantly longer in the combination therapy group as shown in Table 2 and Fig. 2a ( p = 0.025). DCRs based on assessments prior to each subsequent session were statistically higher for TOCE + HAIC (77.8%) than TOCE (40.0%) ( p = 0.0025). The representative images of the case treated with TOCE + HAIC that resulted in a complete response are shown in Fig. 3. Although no clear trend for AFP or L3% was observed during the course of either treatment, the increment ratio of DCP during the course of treatment was significantly larger in the monotherapy group, suggesting a better tumor control capability from the combination therapy ( p < 0.0001, Fig. 2b). However, the overall survival was not significantly different between the monotherapy and combination therapy groups, with median survival times of 706 days and 733 days, respectively ( p = 0.40, Table 2 and Fig. 2c).

The hematological and non-hematological toxicities were evaluated for each treatment session using NCI-CTCAE version 4.0 and are summarized in Table 3. No hematological or non-hematological adverse events of grade 2 or higher were observed. In the TOCE group, the white blood cell and neutrophil counts decreased in two cases (13%), elevated liver enzymes developed in three cases (20%), and nausea was reported in 2 cases (13%). All of these adverse events were categorized as grade 1. In the TOCE + HAIC group, a fever above 37 °C after 5 sessions was present in 2 cases (14%). One of these patients complained of anorexia after 3 sessions (8.3%) and showed hypoalbuminemia after 2 sessions (5.6%). These adverse events were all categorized as grade 1 and were resolved within 2 weeks without any specific treatments. There were no treatment-related deaths in either arm.

Although none of the serum concentrations of albumin, total bilirubin, creatinine, or prothrombin time-international normalized ratio (PT-INR) showed a significant trend during the treatment course (Fig. 4), a significant reduction in the serum cholinesterase concentration was observed in four out of ten cases receiving the combination therapy ( p < 0.0001, r = 0.86 Fig. 2d and Table 4). On the other hand, the level never dropped below 80% of the initial value in the other cases, including fifteen observation points in nine monotherapy cases, with the exception of one point at which the serum level of cholinesterase reached 68.9% 62 days after initial miriplatin administration. Notably, the two longest survivors, who lived more than three years after the initial combination treatments, never showed a reduction in cholinesterase below 80% of the pretreatment value throughout the entire course of treatment. The TNM and Child-Pugh stages did not significantly influence whether cholinesterase decreased below 80% ( p = 0.52, p = 1.00, respectively).