INTRODUCTION
Hepatocellular carcinoma (HCC) is the fifth most common malignancy and fourth most common cause of cancer related death globally[1]. Data from the National Cancer Institute’s Surveillance Epidemiology and End Results database suggest a bleak outcome for patients with HCC in the United States, with a 5-year overall survival (OS) of 19.6%[1]. The OS rates depend on the stage at which the disease is diagnosed and for early disease may be over 70% at 5-years. However, as the disease progresses to more advanced stages, there is a sharp decline in OS, with the 5-year survival rates for metastatic HCC being just 2.5%.
There are multiple reasons why HCC is associated with such a poor prognosis. Delayed diagnosis, cachexia, and association with liver cirrhosis all play a role in determination of the outcomes in such patients[2,3]. While an early presentation provides the physician the chance to offer therapies with curative intent such as resection, ablation or liver transplantation (LT), patients with advanced disease often present with large tumor size (beyond Milan criteria), portal vein invasion, poor liver function or performance status which limits the therapeutic options to locoregional therapies (LRT) [transarterial chemo- or radioembolization, stereotactic body radiation therapy (SBRT)] or systemic therapy [tyrosine kinase inhibitors (TKI) or immune checkpoint inhibitors][4]. LRTs play a role in the management of 50%-60% patients with HCC and transarterial chemoembolization (TACE) is presently the standard therapy for intermediate stage HCC and is associated with a median survival of 25-30 months[5].
PATHOPHYSIOLOGY AND EVOLUTION OF COMBINATION THERAPY IN HCC
While LRTs have been available for over 15 years in clinical practice, they have been shown to improve progression-free survival (PFS) rather than OS. Smaller cohort studies on newer LRTs, such as transarterial radioembolization or SBRT, have shown good results in HCC but lack robust evidence to replace conventional LRTs. TKIs have also failed to show an improvement on OS and are commonly associated with adverse effects (hand-foot syndrome, oral ulcers), which significantly impair the patients' quality of life, limiting their use[6]. The recent IMBRAVE 150 and HIMALAYA trials brought new hope to patients with advanced HCC by demonstrating improvement in OS as compared to therapy with TKI[7,8]. The use of a drug combination such as atezolizumab (PD-L1 antibody) and bevacizumab [Bev, vascular endothelial growth factor (VEGF) antibody] was based on a potential therapeutic synergism of both drugs. Similar synergism may exist with TKI as well, although it is yet to be demonstrated clinically. VEGF promotes T-cell immunosuppression and hinders the differentiation and activation of dendritic cells in the tumor microenvironment, thereby promoting local immunosuppression. By blocking VEGF function, these inhibitors modulate the tumor microenvironment, normalize tumor vasculature, and promotes infiltration of T-cells, thereby enhancing the response to ICI[9]. Similarly, TKIs have been shown to overcome the tumor intrinsic resistance to immune checkpoint therapy by blocking multiple tyrosine kinase-based downstream signaling pathways (MAPK, PTEN, WNT-β-catenin, etc.)[10]. This allows greater infiltration by T-cells and mature dendritic cells and increased interferon signaling, resulting in increased responsiveness to immune checkpoint therapy.
In the recent edition of the World Journal of Gastroenterology, Han et al[11], present the results of an interesting observational data comparing either lenvatinib (Len) or Bev with sintilimab along with interventional therapy in intermediate-advanced unresectable HCC. Sintilimab (Sin) is a PD-1 immune checkpoint inhibitor that has been studied extensively in China and has been shown to improve OS and PFS in HCC when combined with a Bev biosimilar[12], and increase recurrence-free survival in patients with a high-risk of recurrence post resection of HCC[13]. There is evolving data on the combination of Sin with TKI and Sin with Bev and radiotherapy in HCC[14,15].
REVIEW OF PRESENT STUDY
In this retrospective observational, single-center study from China, the authors compared patients receiving a combination of Bev-Sin or Len-Sin along with intervention therapy (IT) in a 1:2 propensity-score-matched cohort. The rationale of combining IT with drug therapy is supported by evidence that tissue necrosis induced by IT releases tumoral neoantigens, which then recruit and activate dendritic cells in the tumor microenvironment leading to the conversion of an “immunosuppressive” microenvironment to a more “immune supportive” setting which is conducive to the action of ICI[5]. Patients receiving either modality of therapy from February 2020 to July 2021 were identified through a review of medical records and were considered for inclusion if they were > 18 years of age, with BCLC-B/C disease, ECOG-PS-0/1, at least a single lesion on imaging and optimal liver function, who had received BeSiIT or LeSiIT as first line therapy. Optimal organ function was defined as absolute neutrophil count of ≥ 1.2 × 109/L, platelet count of ≥ 60 × 109/L, albumin concentration of ≥ 30 g/L, total bilirubin concentration of < 30 μmol/L, aspartate and alanine transaminase ≤ 5 × upper limit of the normal range, creatinine clearance rate of ≤ 1.5 × upper limit of the normal range, and left ventricular ejection ≥ 45%. Patients belonging to Child-Pugh Class C, unmeasured hepatic lesions, non-HCC tumors, incomplete data or lost to follow-up were excluded.
Both Bev and Sin were given as infusions at 3 weekly intervals, while patients in the Lev-Sin group received Sin infusions at 3 weeks with daily oral Len. The first dose of Sin was given within 7 days of the initiation of Len. Modalities for IT were decided by two radiologists and were either TACE or hepatic artery infusion therapy (HAIC) of 5-fluorouracil, leucovorin and cisplatin. Drug therapy was provided within 3 days prior or after IT. Baseline imaging was performed 7 days prior to initiation of therapy and 2-3 times monthly, subsequently. The study endpoint was PFS. Secondary endpoints included OS, objective response rate (ORR), disease control rates (DCR), and adverse events (AE).
Two hundred and eight patients were screened of whom 127 patients were assessed for eligibility and 78 patients were matched in a 1:2 ratio (28 receiving BeSiIT and 50 receiving LeSiIT). In the unmatched cohort, a greater number of patients in the LeSiIT cohort had more than 3 tumors and vascular invasion as compared to BeSiIT, which was addressed by propensity score matching. Sixteen (57.1%) patients in the BeSiIT cohort could be downstaged as compared to 13 (26%) in the LeSiIT cohort (P = 0.04). Despite this, however, there was no improvement in OS (less than median OS in both groups) or PFS [11.0 months (95%CI: 6.0-NA) in the BeSiIT group compared to 12.0 months (95%CI: 7.0-NA) in the LeSiIT group, P = 0.68]. ORR and DCR were similar in both groups. BeSiIT therapy was better tolerated with patients on LeSiIT therapy having greater dyspeptic complaints, diarrhea, and fatigue. Patients on LeSiIT also had a higher incidence of Grade 3-4 events, which would presumably have led to changes in the dosing regimen or drug discontinuation. On multivariable analysis, higher serum alpha fetoprotein (AFP) levels were associated with poorer PFS.
DISCUSSION OF RESULTS
The study provides interesting results in the striking number of patients who could be down staged to ablation or resection, along with safety data on both regimens. As it was a retrospective analysis, there was limited data on indications, patient selection and details of technical/clinical success. However, an important exploratory outcome here would be the number of patients who became amenable for LT after downstaging. Timing of LT after downstaging with ICI is a major area of ongoing studies and current consensus is to wait for a washout period of at least 50 days to reduce rejection rates[16]. The patient selection is reflective of a real-world scenario where the majority of patients are of Child-Pugh class A/B. This also influences disease outcomes as recent observational studies have shown that HCC outcomes are influenced by both Child-Pugh class as well as the etiology of liver disease, but not AFP levels alone (although higher AFP values > 1000 ng/mL is suggestive of metastatic disease with poor outcomes)[17-19]. In terms of treatment options for unresectable HCC, TACE is more commonly used than HAIC, which is more commonly used in China. There is limited data on comparison between the two interventions, which may contribute to a certain degree of heterogeneity in the comparison. Because of the widespread use of TACE, quality metrics, parameters of clinical/technical success are better defined, along with complication profiles such as post-TACE syndrome. Multiple ongoing trials also explore the roles of pre-emptive medical therapy such as use of N-acetylcysteine and dexamethasone in preventing post TACE decompensation. Such data is limited for HAIC and is an area for further exploration. The AE profile for the two groups was provided, but further granularity regarding the grade of AE and its potential impact on further treatment and quality of life remain desirable. Expectedly, proteinuria was the major complication with BeSiIT, while fatigue, hyperbilirubinemia and gastrointestinal symptoms were major complications with LeSiIT. A recent meta-analysis compared the various Grade 3-4 AEs in studies pertaining to ICI use in HCC[18]. Proteinuria (cumulative pooled incidence 25.2%), hypertension (27%) and fatigue (25.1%) were the major AEs reported across 23 studies with the cumulative incidence of any AE being 82.7%. Grade 3-4 AE was seen in largely 4%-5% of patients. There are certain limitations to this study that would require it to be clarified in larger prospective trials. The authors have used propensity score matching to enhance comparability and eliminate bias, however, that may not completely address the limitations of a retrospective, observational design and future randomized trials are required for validation of findings. Further, while propensity score matching balanced some factors, a subgroup analysis of unmatched cohort data, particularly regarding tumor numbers and vascular invasion, would have enhanced the robustness of the comparison. It is intriguing to find a lack of improvement in OS despite such a large number of patients being successfully down staged. While patients were matched for tumor size, AFP and number, the metastatic load in patients may have been different and a PET scan at baseline would have been helpful in distinguishing this. Ancillary information such as etiology, suppression of viral load, abstinence, coexistent diabetes etc., would be desirable. Similarly, while there are reports of good efficacy regarding HAIC, there are no randomized trials comparing the efficacy of HAIC and TACE. An additional subgroup analysis may help identify differences in outcomes between the two groups, as for adequate comparison of OS/PFS, both IT modalities must be equivalent or at least non-inferior in efficacy. In addition, HAIC may also have influenced the severity of the gastrointestinal adverse effects noted in this study. The overall numbers in both groups are small and the risk of sparse data bias along with overfitting in the multivariable analysis mandate cautious interpretation and application of this data[20,21]. Future studies should also look to incorporate data on biomarkers of HCC progression and response to ICI therapy to provide better understanding of the disease pathogenesis.
CONCLUSION
To conclude, combination therapy in patients with unresectable, intermediate-advanced HCC can provide clinical benefit in the short- and long-term outcomes without increasing severe AEs. However, more clarity is needed regarding the optimal regimen, potential beneficiary, and latent AEs. Further large-scale, randomized trials will be helpful in better clarifying the role of the combination modality for unresectable, intermediate-advanced HCC.
Provenance and peer review: Invited article; Externally peer reviewed.
Peer-review model: Single blind
Specialty type: Gastroenterology and hepatology
Country of origin: India
Peer-review report’s classification
Scientific Quality: Grade A, Grade A, Grade B
Novelty: Grade A, Grade A, Grade B
Creativity or Innovation: Grade A, Grade A, Grade B
Scientific Significance: Grade A, Grade A, Grade B
P-Reviewer: Li JT; Lin JY S-Editor: Qu XL L-Editor: Filipodia P-Editor: Zheng XM
