The safety and efficacy of transarterial chemoembolization combined with sorafenib and sorafenib mono-therapy in patients with BCLC stage B/C hepatocellular carcinoma

This study was approved by the Institutional Review Board of Guangxi Medical University and was conducted in accordance with the Declaration of Helsinki and internationally accepted ethical guidelines. During their admission for surgery, the patients enrolled in this study provided written informed consent for their information to be stored in hospital databases and used for research. During data collection, patient records were anonymized. Patient admission and consent procedures have been described previously [21].

This retrospective study included 104 patients with HCC between August 2004 and November 2014. Patients treated with TACE and sorafenib were included in the TACE + sorafenib group (n = 56); patients who were treated only with sorafenib were included in the sorafenib group (n = 48). All patients were diagnosed with HCC based on the criteria of the European Association for the Study of the Liver [22].

The inclusion criteria were: (a) 18–75 years old; (b) HCC classified as either unresectable BCLC stage B or BCLC stage C [23]; and(c) liver function classified as Child-Pugh class A or B.

Patients were excluded from the study if they had any of the following: (a) malignant tumors of other organ systems; (b) HCC of Child-Pugh class C; or (c) any contraindication for therapy with TACE (e.g., complete obstruction of the portal vein) or sorafenib (e.g., allergy to sorafenib).

The following information was obtained for all patients included in the analysis: disease history; physical examination findings; results of serum laboratory tests (total bilirubin, TBil; albumin, ALB; alanine aminotransferase, ALT; platelet count, PLT; prothrombin time, PT; α-fetoprotein level, AFP; and hepatitis B virus surface antigen, HBsAg); and results of radiologic investigations (computed tomography, CT; magnetic resonance imaging, MRI; and/or Doppler ultrasound).

PVTT was confirmed by radiologic investigations (a filling defect sign in CT or MRI images; or ultrasonographic features of a mass in the portal vein). PVTT type was defined according to a previous study [24] as follows: type I, tumor thrombus (TT) involving segmental branches of the portal vein or above; type II, TT involving the right/left portal vein; type III, TT involving the main portal vein trunk; or type IV, TT involving the superior mesenteric vein or inferior vena cava.

Portal vein hypertension (PVH) was defined as the presence of esophageal varices and/or a platelet count <100,000 /μL in association with splenomegaly.

We used the Seldinger technique [25] and introduced a 4.1-French RC1 catheter into the tumor feeding artery. Afterwards, we carefully identified the number, location, size and branches of the tumor. A mixture of 10–20 mL iodized oil, gelfoam particles with 30–50 mg doxorubicin and 50–100 mg cisplatinum were injected into the arterial branches. The number of TACE cycles administered ranged from 1 to 6, with TACE repeated at 1-month intervals, depending on the patients’ liver function and tumor shrinkage.

Sorafenib was administered orally from the beginning of the treatment period (i.e. treatment was initiated before TACE was performed in those receiving combination therapy) at a dosage of 400 mg twice daily (Bayer HealthCare AG, 200 mg/pill). The sorafenib dose was adjusted if adverse drug events (ADEs) developed. If grade I or 2 ADEs (National Cancer Institute Common Terminology Criteria for Adverse Events version 3.0; [26]) occurred, we adopted a wait-and-see policy. Usually these ADEs disappeared spontaneously, but if they persisted the drug was either reduced in dosage or discontinued. When grade 3 or 4 ADEs occurred, the oral dose was reduced to 200 mg per day. If the ADEs had not disappeared or decreased in severity 1 week after dose adjustment, it was recommended that the patient stop receiving sorafenib until the symptoms had alleviated or disappeared. In patients receiving combination therapy, treatment with sorafenib was continued during and after the performance of TACE.

Patients were asked to return to the hospital for follow-up every 1–2 months after discharge. During each follow-up, blood tests and radiologic investigations were performed as at baseline. Tumor response was recorded during every follow-up and classified (based on the best response) after 6 months, according to the Modified Response Evaluation Criteria in Solid Tumors for HCC (mRECIST) [27, 28], as either complete response (CR), partial response (PR), stable disease (SD) or progressive disease (PD). Patients lost to follow-up were excluded from the final analysis.

The primary outcome measures in our study were OS and PFS.PFS was defined as the duration from patient discharge to disease progression (according to mRECIST guideline). Secondary outcome measures were tumor response and the occurrence of ADEs.

SPSS 18.0 (IBM, Chicago, USA) was used for statistical analysis. A P value <0.05 was defined as the threshold of statistical significance. Normally distributed data are expressed as the mean ± standard deviation (SD), non-normally distributed data are expressed as median (range), and enumeration data are expressed as n (%). Differences in outcomes between the two therapy groups were assessed for significance using independent-samples t-tests or χ² tests. The Kaplan–Meier method was used to evaluate the effects of patient characteristics on OS and PFS. Factors significantly associated with OS or PFS were identified by multivariate analysis using a stepwise Cox model, with calculation of hazard ratios (HRs) and 95% confidence intervals (CIs). In addition to the type of therapy used (TACE + sorafenib versus sorafenib), the other factors entered into the multivariate analysis were patient age, patient gender (male versus female), tumor number, tumor diameter, vascular invasion (present versus absent), metastasis (present versus absent), Child-Pugh stage (A versus B), and AFP level (< 400 ng/mL versus ≥400 ng/mL). These other parameters were chosen so as to be representative of factors known to be associated with HCC progression or patient survival. Additional variables related to these factors were not included in the multivariate analysis (for example, other parameters related to liver function were excluded as they are related to Child-Pugh stage). A subgroup analysis based on PVTT status was conducted to try and identify whether a subset of patients might benefit more from combination therapy with TACE and sorafenib. An additional analysis was also performed to determine whether tumor size influenced OS and PFS.

From August 2004 to November 2014, a total of 104 patients with HCC (mean age, 49.02 ± 12.29 years) were included in this retrospective study, including 94 males and 10 females. All patients’ data are attached in the Additional file 1 (organized file). Forty-eight patients received sorafenib mono-therapy while 56 patients received sorafenib plus TACE therapy. The baseline demographic and clinical characteristics were similar between the two treatment groups, except that patients in the TACE + sorafenib group had a significantly higher level of ALB (P = 0.028) and proportionally more patients with Child-Pugh class A disease (P = 0.017). There were no therapy-related deaths, and in-hospital mortality was zero (Table 1).

Median OS was 22.0 months (95% CI: 14.1–29.9 months) in the TACE + sorafenib group and 18.0 months (95% CI: 11.8–24.2 months) in the sorafenib group, with no significant difference between groups (P = 0.223; Fig. 1 and Table 2). However, median PFS was significantly longer in the TACE + sorafenib group (8.0 months; 95% CI: 3.4–12.6) than in the sorafenib group (6.0 months; 95% CI: 3.3–8.7 months; P = 0.004; Fig. 1 and Table 2), indicating that combination therapy was more effective than sorafenib mono-therapy at limiting disease progression.

Data for tumor response at 6 months were available for 40 patients in the sorafenib group and 48 patients in the TACE + sorafenib group (Table 3). There were no significant differences between treatment groups in the CR rate (P = 1.000), PR rate (P = 0.502), SD rate (P = 0.574), PD rate (P = 0.906) and OR rate (P = 0.425). Furthermore, subgroup analysis on the basis of the presence (i.e., types I, II, III or IV) or absence of PVTT also showed no statistical differences between the sorafenib and TACE + sorafenib groups in the tumor response 6 months after treatments (all P > 0.05; Table 4). This suggests that the two treatment regimens were similar with regard to reducing tumor size.

There were no significant differences between the sorafenib and TACE + sorafenib groups in the incidences of grade I, II, III and IV ADEs (all P > 0.05), and all ADEs were tolerable. Grade III ADEs occurred in 14 patients in the sorafenib group and 13 patients in the TACE + sorafenib group, while no Grade IV ADEs were observed (Table 5). Symptoms in patients with grade III ADEs disappeared or were alleviated following adjustment of the sorafenib dose or administration of symptomatic supportive treatments. These findings indicate that the addition of TACE to sorafenib therapy does not result in a notable increase in the incidence or severity of ADEs.

Multivariate Cox regression analysis identified use of TACE + sorafenib combination therapy (HR = 0.498, 95% CI = 0.278–0.892, P = 0.019), no vascular invasion (HR = 0.354, 95% CI = 0.183–0.685, P = 0.002) and Child-Pugh class A (HR = 0.308, 95% CI = 0.141–0.674, P = 0.003) as independent factors predicting better OS, while tumor number (HR = 1.286, 95% CI = 1.031–1.604, P = 0.026) was an independent factor predicting poorer OS (Table 6). Similarly, use of TACE + sorafenib combination therapy (HR = 0.461, 95% CI = 0.273–0.780, P = 0.004) and no vascular invasion (HR = 0.557, 95% CI = 0.314–0.988, P = 0.045) were independent factors predicting a better PFS (Table 7).

The observation that tumor diameter was not an independent predictor of OS and PFS in the multivariate analysis was perhaps unexpected. One possibility we considered was that OS and PFS might only be influenced by tumor size once the tumor exceeded a certain diameter. To explore this possibility, OS and PFS were further analyzed based on different tumor diameters (Table 8 and Fig. 2); the cutoff value of5 cm was based on that used in the TNM classification, while the additional higher cutoff value of 7 cm was arbitrarily chosen. Median OS was 44.0 months (95% CI: 21.624–66.376) in patients with a tumor diameter < 5 cm and 17.0 months (95% CI: 11.806–22.194) in patients with a tumor diameter ≥ 5 cm (P = 0.004; Fig. 2a); in contrast, PFS did not differ between the two groups (8.0 months versus 7.0 months, respectively, P = 0.268; Fig. 2b). Patients with a tumor diameter < 7 cm had a median OS of 38.0 months (95% CI: 20.228–55.772) and a median PFS of 9.0 months (95% CI: 6.003–11.997), while patients with a tumor diameter ≥ 7 cm had a median OS of 14 months (95% CI: 10.409–17.591) and a median PFS of 5.0 months (95% CI: 3.007–6.993); both OS and PFS differed significantly between the two groups (P < 0.05; Fig. 2c and d).