Transarterial chemoembolization with/without immune checkpoint inhibitors plus tyrosine kinase inhibitors for unresectable hepatocellular carcinoma: a single center, propensity score matching real-world study

Liver cancer (LC) is one of the most prevalent causes of cancer-related death globally, especially hepatocellular carcinoma (HCC) associated with hepatitis B virus in Asia. [1] Although increasing rates of screening and surveillance for LC across most countries, more than 70% of patients are still diagnosed as intermediate or advanced-stage disease, leading to poor prognosis. [2‐4]. TACE is widely used in intermediate HCC globally and in intermediate-advanced HCC domestically. [5‐8] However, Transarterial chemoembolization (TACE) has not been universally successful in controlling tumor growth, recurrence and metastasis due to the high rate of incomplete embolization and embolization-induced changes in the tumor microenvironment (TME). [9, 10]

There are some rationales about TACE combining with immune checkpoint inhibitors (ICIs) and tyrosine kinase inhibitors (TKIs). Embolization-induced hypoxia after TACE can lead to upregulation of vascular endothelial growth factor (VEGF), which might contribute to revascularization, making it the theoretical basis of using anti-angiogenesis in the combination therapy. [11, 12] The efficacy of TACE in combination with anti-angiogenesis drugs in HCC has been proven in preclinical and clinical trials. [13‐15] In addition, TACE results in necrosis of the tumor tissue and releases tumor antigens which may promote tumor-specific immune responses. [16] It is reported that intra-tumoral exhausted effector cells, such as CD8 + /PD-1 + T cells and T regulatory cells (CD4 + /FOXP3 +) would become lower after TACE, which may transform an immuno-suppressive microenvironment into an immune supportive setting to enhance the response of ICIs. [17] Therefore, TACE combined with TKI and ICI is theoretically reasonable. To our knowledge, few data of prospective clinical trials are published about this regimen. Herein, more research is needed on the efficacy and safety of this triple combination therapy.

Therefore, we aimed to further validate the better efficacy with acceptable safety of TACE with ICI plus TKI than TACE monotherapy in this retrospective, propensity score matching (PSM) cohort of HCC patients in the real-world study.

Our study showed that TACE in combination with TKIs and ICIs achieved the higher mOS (21.9 months, 95% CI 17.2–26.6 months) and mPFS (8.3 months, 95% CI 6.0–7.4 months) when compared to TACE alone. Subgroup analyses showed that the mOS was 19.2 months (95% CI 14.3–24.1 months) in the “before TACE group” and 26.8 months (95% CI 15.7–37.9 months) in the “after TACE group” (P = 0.011). No worse AEs were observed in the present study compared with previously reported clinical trials. [18‐20] Most AEs were easily managed and controlled with mild-to-moderate severity, consistent with the results of other previous studies. This combination regimen promises to be a potential first-line option for patients with advanced HCC.

Currently, immunotherapy has approved for the treatment in HCC. “A + T” has been recommended as the first-line treatment in andvanced HCC after IMbrave150 trial published. Compared to updated data of IMbrave150 and Oriental-32, our study seemed to delivery more promising mOS (21.9 vs 19.2 compared IMbrave150, 21.9 vs 10.4 compared Orient-32) and mPFS (8.3 vs 6.9 compared IMbrave150, 8.3 vs 6.9 compared Orient-32). [21, 22] It probably means that combining TACE with TKIs + ICIs improves mOS and mPFS of patients with intermediate-advanced HCC. This has also been validated in real-world studies. Median OS and PFS of combination group were 19.2 months [95%CI (16.1–27.3)] and 9.5 months [95%CI (8.4–11.0)] in CHANCE001 trial, 24.1 months [95%CI 20.0-NR] and 13.5 months [95%CI 9.0–18.0] in CHANCE2211 trial, respectively. [23, 24] Our study demonstrated considerable efficacy, especially in “after TACE” group. However, the ORR of immunotherapy in HCC patients is 15–30%. Its efficacy is affected by a variety of factors, such as liver function status, BCLC staging, baseline tumor burden, etiology, etc. [25, 26] How to improve the efficacy of immunotherapy is still a problem for clinicians. The OS benefit with atezolizumab plus bevacizumab vs. sorafenib was generally consistent across the patient subgroups in IMbrave150 trial, except in patients who had non-viral etiology of HCC (HR for death 1.05; 95% CI 0.68–1.63). [25, 27] Immune therapy did not improve survival in patients with non-viral HCC reported in a meta-analysis, which showed that non-viral HCC, might be less responsive to immunotherapy. That was visibly observed in HCC caused by Non-Alcoholic SteatoHeaptitis(NASH-HCC) probably owing to NASH-related aberrant T cell activation causing tissue damage and impaired immune surveillance. [25, 28]While the synergistic antitumor effects underlying TACE plus ICIs and TKIs was verified and could possibly be explained as: (1) TACE causes necrosis and disintegration of the tumor cells, which results in releasing a large number of tumor antigens, increasing the expression of programmed cell death protein-1 (PD-1) and programmed cell death protein ligand-1 (PD-L1), and improving tumor recognition [12, 17]; (2) TACE could alter the TME of HCC by producing local inflammation, promoting lymphocyte infiltration, and activating the immune system, which may improve the overall anticancer effect of ICIs [16, 29]; (3) TKIs could antagonize the effect of TACE-induced angiogenesis by blocking the VEGF signaling pathway and enhance the efficacy of ICIs by reversing VEGF-mediated immunosuppression [30‐32]. These proposed mechanisms suggest that the combination of TACE, TKIs and ICIs might produce synergistic anti-HCC effects and therefore improve the OS and PFS of the patients.

In the subgroup analysis, “After TACE group” delivered more promising result of OS compared “Before TACE group”. One possible explanation for prolong mOS is that the efficacy of TACE, tumor antigens releasing and TME altering, are time-consuming [33], possibly due to the worsened hypoxia and immunosuppressive TME of residual tumor tissue following TACE [9, 32]. At this consuming moment, TKIs are needed to alleviate hypoxia and immunosuppressive TME through vascular normalization so that the inflammatory environment suitable for T-cell responses could be maintained, which improves the therapeutic effectiveness of ICIs [31]. In addition, previous studies have found a significant infiltration of CD8 + T cells after TACE, indicating an immunological enhancement in the body [12, 34, 35]. LAUCH trial displayed better benefits of TACE plus lenvatinib than TACE monotherapy, which means TKIs administration after TACE procedure may improve the efficacy of TACE. The rationale is clear to conduct ICIs treatment on the basis of inflammatory response and enhanced immunity after TACE.

Compared with monotherapy group, worse liver function, infusion-related reaction, proteinuria, pain and Hand-foot syndrome were observed in combination group. Fortunately, all adverse events were controlled and manageable with dosage reduction. Lower incidence rate of grade 3 or higher AEs occurred than those reported in IMbrave150, ORIENT-32 and CARES-310 trials (21.6% vs 63%, 21.6% vs 56% and 21.6% vs 71%). [21, 36, 37]Grade 3 or higher AEs in our study is similar to other real word studies (16.7% in CHANCE2211 and 15.8% in CHANCE2201). [23, 24] Safety is one of challenges in combination therapy. Collectively, TACE + TKIs + ICIs therapy had acceptable safety and improved the OS and PFS of patients.

There were several limitations in this study. Firstly, this was a retrospectively designed study, which may lead to potential selection bias. Besides, the sample size of the study is limited. The conclusions drawn from this study should be verified in better-designed prospective studies. Thirdly, a group of patients taking TKIs before TACE but accepting ICIs therapy after TACE was lacking, since several studies found that pre-treatment with TKIs before TACE can normalize tumor vessels and suppress VEGF, which may lead to a homogeneous distribution of lipiodol mixed anticancer drugs in the tumors [15, 36, 37]. This might be an issue to be further explored. Additionally, another limitation of our study was a dropout rate of approximately 10%, but without different significance between the two groups. We had tried our best to make that within a statistically reasonable range, but we need research with lower or even without dropout to improve the reliability of the results in the future. Finally, although we used PSM to minimize bias as much as possible, the diversity of TKIs and ICIs can still lead to biased results. It makes sense that similar results in TACE plus mono-TKI and mono-ICI but more rigorously prospective studies needed to confirm this.

In conclusion, our retrospective data indicated that TACE + TKIs + ICIs seemed to have considerable efficacy in patients with unresectable HCC compared to TACE monotherapy. Specially, “TKIs + ICIs” co-treatment within 3 months after the first TACE procedure might be a better medication strategy to achieve better clinical outcomes.