Stent placement combined with intraluminal radiofrequency ablation and hepatic arterial infusion chemotherapy for advanced biliary tract cancers with biliary obstruction: a multicentre, retrospective, controlled study

Biliary tract cancers (BTCs) are classified into intrahepatic cholangiocarcinoma, perihilar cholangiocarcinoma (pCCA), distal cholangiocarcinoma, gall bladder cancer, and carcinoma of the ampulla according to the anatomical structure [1, 2]. The incidence and mortality of BTCs have progressively increased worldwide in recent decades [3, 4]. For early-stage disease, liver transplantation and radical resection seem to be potential curative options [5]. Unfortunately, approximately 60–70% of patients are diagnosed in advanced stages, and > 50% of advanced BTCs (Ad-BTCs) are complicated by biliary obstruction (BO) [6], which can rapidly result in liver failure and cachexia with an extremely poor prognosis [7]. The median survival of patients with Ad-BTCs and BO is 4.8 months, with a 5-year survival rate of 10% [8]. Metal stent placement is considered the most effective palliative treatment to improve the quality of life and survival [9]. However, 1-year recurrence rates for BO reach 68%, with the main cause for recurrence being local tumour progression [10, 11]. Therefore, tumour growth control is critical for maintaining stent patency and prolonging survival in cases of Ad-BTCs with BO.

Radiofrequency ablation (RFA) is widely used to treat solid tumours, particularly hepatocellular carcinoma [12]. Intraluminal RFA (intra-RFA), based on the same principle of energy generation, has been used to manage malignant BO (MBO) in recent years [13]. The high thermal energy generated by radiofrequency electrodes can effectively hamper intraluminal tumour growth [14]. It was reported that intra-RFA could prolong stent patency and survival in patients with BTCs and BO, although its efficacy needs improvement [15]. In a meta-analysis, the pooled weighted mean difference in stent patency was only 1.7 months [16]. Moreover, intra-RFA cannot reach residual tumours and lesions located outside the biliary tract, which may quickly progress and trigger stent re-stenosis.

Hepatic arterial infusion chemotherapy (HAIC) can effectively increase the local concentration of anti-tumour drugs and decrease the systemic distribution of agents [17]. Compared with systemic chemotherapy, HAIC can significantly reduce adverse events (AEs) and achieve more potent local anti-tumour effects [18]. Patients with advanced hepatocellular carcinoma can achieve high local tumour control rates and yield long-term benefits from HAIC [19, 20], and retrospective studies suggest that HAIC can also offer clinical benefits to patients with Ad-BTCs [21, 22]. A recent phase 2 clinical study indicated that HAIC with oxaliplatin and 5-fluorouracil for advanced pCCA can effectively control local tumour growth and provide an encouraging survival benefit [23]. Tumours that cannot be completely ablated and lesions outside the biliary tract can be simultaneously controlled by compensating HAIC. Intra-RFA combined with HAIC with oxaliplatin and 5-fluorouracil may be a promising strategy for Ad-BTCs with BO. To our knowledge, there is no report regarding this new treatment modality. Therefore, we conducted a multicentre, retrospective study to evaluate the efficacy and safety of stent placement + intra-RFA+HAIC for Ad-BTCs with BO.

In the present study of patients with Ad-BTCs and BO, we found that SPT and OS were markedly longer in patients treated by stent placement + intra-RFA+HAIC than in those treated by stent placement alone. The absence of metastasis, a stricture of < 4 cm, and the level of BO in Bismuth types I–II were independent predictors of longer SPT and OS. All AEs could be tolerated or relieved after active symptomatic treatments.

In cases of BO caused by BTCs or other malignancies, relieving obstructive jaundice and maintaining stent patency are extremely pivotal to prolong survival and improve the quality of life. In our study, SPT and OS were significantly longer in the combination group than in the control group, with significantly higher survival rates at 6, 9, and 12 months in the former group. These encouraging outcomes can be attributed to the use of intra-RFA+HAIC for restricting local disease progression. The ablation electrode can generate high thermal energy, which induces coagulative necrosis of tumour cells in the bile duct cavity. Monga confirmed the disappearance of tumour blood vessels and lumen enlargement after intra-RFA [25]. Several studies have reported the clinical efficacy, safety, and survival benefit of intra-RFA for MBO [13, 15, 26]. In a prospective, randomised trial comparing intra-RFA + stent placement with stent placement alone for unresectable extrahepatic cholangiocarcinoma with BO, the combined treatment significantly improved the overall mean survival time and mean stent patency period (6.8 vs. 3.4 months; p = 0.02; 13.2 ± 0.6 vs. 8.3 ± 0.5 months; p < 0.001) [27]. However, intra-RFA reportedly exhibits several limitations; in particular, it cannot ablate lesions outside the biliary tract or destroy a tumour mass completely [27, 28]. If other treatment measures are not implemented in time, the residual tumour will grow rapidly, eventually leading to stent stenosis.

Most BTCs are hypovascular tumours and visualisation of tumour staining and tumour vascularity is difficult. Therefore, the therapeutic efficacies of transarterial chemoembolisation and transarterial radioembolisation are largely limited in patients with BTCs, particularly patients with extrahepatic cholangiocarcinoma. Continuous low-dose infusion of drugs can achieve a high local drug concentration; moreover, the toxicity of the drugs can be greatly reduced by hepatic first-pass metabolism. In addition, the synergistic effect of oxaliplatin and 5-fluorouracil is more effective for killing tumour cells. Oxaliplatin can effectively inhibit tumour cell growth by inducing ribosome biogenesis stress [29], while continuous arterial infusion of 5-fluorouracil may effectively kill tumour cells that recover from the mitotic block produced by oxaliplatin as they progress into the S phase [30]. Furthermore, hepatic arterial infusion of oxaliplatin can induce high local concentration ratios between peripheral tissues and tumours [31]. Because HAIC is not restricted to a single session, repeated treatments and multiple sessions can be performed to increase tumour destruction and prolong OS. HAIC is an effective treatment for various malignancies, including BTCs [22, 32, 33]. A phase 2 study reported that overall response and disease control rates of 67.6% and 89.2%, respectively, could be achieved with the use of HAIC with oxaliplatin and 5-fluorouracil for pCCA [23]. Therefore, intra-RFA+HAIC can induce potent local anti-tumour effects.

In our study, the median OS in the combination group was 13.2 months, which was slightly longer than that in a phase 3 clinical study (11.7 months) where patients with Ad-BTCs received systemic chemotherapy with cisplatin and gemcitabine [34]. The median OS (13.4 months) in a recent randomised clinical trial of gemcitabine plus S-1 therapy for Ad-BTCs was similar to that in our study [35]. Notably, the proportion of patients with BO was 44.7% and 41% in the phase 3 study and randomised trial, respectively, while it was 100% in the present study. Therefore, the survival benefit of intra-RFA + HAIC may be noteworthy. The results of HAIC therapy alone or combined systemic therapies have been reported in a few studies till date, although the outcomes are inconsistent. The median survival time reported here is similar to that reported by Kasai et al (median 14.6 months) [21] but lower than that reported by Cercek et al (median 25.0 months) [22] and Konstantinidis et al (median 30.8 months) [36]. In this study, the patients did not receive systemic chemotherapy and exhibited concurrent BO, which may have resulted in the different prognosis. In addition, BO was caused by various types of tumours; therefore, an accurate comparison of effects between different studies is difficult. We also compared the results for our combination therapy with those for intra-RFA + stent placement in previous studies by Yang et al [27] and Inoue et al [14] and found that SPT and OS in our combination group were similar to those in the former study, while the median SPT was similar to that in the latter study (7.6 months). However, SPT in our combination group was obviously longer than that in a meta-analysis of intra-RFA + stent placement, where the pooled weighted mean difference in stent patency was only 1.7 months [16]. Therefore, our future study will focus on comparing the efficacy of stent placement + intra-RFA+HAIC with that of stent placement + intra-RFA for the treatment of Ad-BTCs with BO.

The absence of metastatic diseases, a stricture of < 4 cm, and the level of BO in Bismuth types I–II were independent predictors of longer SPT and OS, probably because the ablation extent and heat penetration depth are limited in patients with larger tumour burdens and complicated anatomies. For example, in pCCA cases, intra-RFA may not achieve complete coagulation necrosis in intraluminal masses in the hilar region [37]. This limitation can be overcome by step-by-step ablation in cases with stenoses longer than 2 cm; however, complete ablation remains impossible. The higher the residual tumour burden after intra-RFA, the harder it is to control by compensating HAIC. From the perspective of biological tumour behaviour, the absence of distant metastasis may imply relatively slow growth of tumour cells and relatively less invasion [38]. In summary, these could be the reasons why the three factors were independent predictors of longer SPT and OS in this study. We also found that a history of previous treatments was associated with shorter SPT and OS. There are several reasons for this. First, the biliary structures were reconstructed by surgery, leading to a more complex anatomy and limiting the effects of intra-RFA [39]. Furthermore, chemoradiotherapy can change the biological tumour behaviour, leading to reduced sensitivity to the drugs used in HAIC. Because the therapeutic effect of intra-RFA+HAIC was restricted, SPT and OS were shorter in patients with a history of previous treatments than in those with no previous history of surgery, radiotherapy, or chemotherapy.

The common AEs caused by biliary tract operation procedures were similar to previously reported AEs, including pain, biliary infection, haemorrhage, and pancreatitis. The incidences and grades of AEs showed no significant differences between the two groups. Patients with serious biliary infection were cured with injections of potent antibiotics based on the findings of blood bacterial cultures and drug sensitivity tests. HAIC-related AEs can be divided into two categories, those caused by the arterial catheter and those caused by the toxicity of oxaliplatin and fluorouracil. The incidence of AEs in the former category was extremely low, and timely diagnosis and effective treatment can avoid their aggravation. AEs in the second category were managed by treatment interruption, dose modification, or symptomatic treatments.

To the best of our knowledge, this study is the first to report the efficacy and safety of stent placement + intra-RFA+HAIC and the associated prognostic factors in patients with Ad-BTCs and BO. However, the study has several limitations. First, it was a retrospective study, which may have resulted in unexpected deviation during data extraction and analysis. In particular, treatment heterogeneity occurred because of disease variation, patient preferences, and economic situations. Second, it is difficult to distinguish between intra-RFA and HAIC in terms of their contribution to the longer OS and SPT. Third, because the metal stent interfered with radiological response evaluation, the local tumour control rate for intra-RFA and HAIC was not explored. Finally, SPT was not accurately evaluated for patients who were lost to follow-up or did not exhibit BO recurrence during their lifespan. In the future, well-controlled, prospective studies with larger samples should investigate the effect of these two novel therapeutic options on survival benefit and stent patency.

In conclusion, our findings suggest that stent placement combined with intra-RFA and HAIC is a feasible and safe option for the palliative treatment of Ad-BTCs with BO. Further prospective, multicentre, randomised controlled studies with more samples are necessary to further validate these findings.