Revisiting unmet needs in clinical research on direct-acting antiviral therapy for HCC patients: Correspondence to letter to the editor on “Direct-acting antiviral therapy for patients with HCV-related hepatocellular carcinoma: A nationwide cohort study”

Dear Editor,

We sincerely appreciate the thoughtful letter from Dr. Wang and colleagues [1] and are grateful for their insightful comments and kind suggestions regarding our recent nationwide cohort study [2]. In our study, we demonstrated that achieving a sustained virological response (SVR) through direct-acting antiviral (DAA) therapy for chronic hepatitis C virus infection significantly improves survival in patients with hepatocellular carcinoma (HCC), across various stages of the disease. In their letter, Dr. Wang et al. raised several important concerns about the limitations of our study, including the lack of stratification based on the biological heterogeneity of HCC, the assessment of liver functional reserve, the optimal timing of DAA therapy, and the potential immunological effects of DAA treatment. They also suggested the application of competing risk analysis in the Cox subdistribution hazards model. We address their concerns as follows:

First, while certain molecular or clinical parameters, such as poor tumor differentiation, microvascular invasion, or specific gene mutations, may indeed offer additional prognostic value or help predict lower SVR rates within the same HCC stage [3], these variables are not routinely available in real-world clinical settings, including within our nationwide database. Currently, the Barcelona Clinic Liver Cancer (BCLC) staging system remains the most validated and widely used framework for prognostication and clinical decision-making in HCC [4]. Furthermore, although tumor aggressiveness may be associated with reduced SVR rates [5], the introduction of DAAs, particularly pangenotypic DAA regimens, appears to have mitigated much of this effect [6-8]. As the data provided in our previous correspondence to the Editorial [9], SVR rates reached as high as 94.9% even among patients with advanced HCC features, including portal vein thrombosis and extrahepatic metastasis (BCLC stage C). Nonetheless, we agree that future investigations focusing on distinct molecular or clinical subgroups, particularly patients with limited life expectancy, are warranted to further refine treatment strategies.

Second, Child-Turcotte-Pugh (CTP) classification could be the most appropriate system for evaluating liver decompensation in our study. Although alternative models may offer additional insights into liver functional reserve, they may not directly assess the clinical manifestations of liver decompensation. For example, albumin-bilirubin grade 2 may include patients with a CTP score of 5–6 (Class A) [10]. Furthermore, the Model for End-Stage Liver Disease is primarily applied in patients with end-stage liver disease and is therefore unsuitable for our study population, as patients with CTP Class C (BCLC stage D) were excluded [2]. While data on transient elastography were unavailable in our nationwide database, we evaluated liver fibrosis using the fibrosis-4 index. As shown in Table 1, we chose not to include these data in the main manuscript due to a lack of statistically significant findings. However, we concur with Dr. Wang et al. that the dynamic interplay between SVR, liver function preservation, and HCC progression deserves further exploration to better understand the mechanisms underlying SVR-associated survival benefits.

Table 1.

Additional data from the logistic regression model for evaluating risk factors of SVR in HCC patients receiving DAA therapy

Variable		Number	SVR, number (%)	Crude
				OR (95% CI)	P-value
AST (IU/L)	≤80	1,413	1,372 (97.1)	1
	>80	792	757 (95.6)	0.65 (0.41–1.02)	0.063
ALT (IU/L)	≤80	1,410	1,368 (97.0)	1
	>80	795	761 (95.7)	0.69 (0.43–1.09)	0.110
Fibrosis-4 index	≤3.25	559	539 (96.4)	1
	>3.25	1,646	1,590 (96.6)	1.05 (0.63–1.77)	0.844
Liver cirrhosis	No	800	778 (97.3)	1
	Yes	1,405	1,351 (96.2)	0.71 (0.43–1.17)	0.178
eGFR (mL/min/1.73m2)	≥60	1,715	1,652 (96.3)	1
	<60	490	477 (97.3)	1.40 (0.76–2.56)	0.278

ALT, alanine aminotransferase; AST, aspartate aminotransferase; eGFR, estimated glomerular filtration rate; HCC, hepatocellular carcinoma; OR, odds ratio; SVR: sustained virological response.

Third, the optimal timing of DAA therapy relative to HCC treatment remains uncertain. For those in BCLC stage 0/A, HCC is potentially curable, and guidelines recommend initiating DAA therapy after curative intervention [4,11]. Although early concerns emerged regarding increased HCC occurrence following DAA therapy, subsequent robust evidence has largely dispelled these fears [12]. DAA treatment has been shown to reduce the risks of cirrhosis-related complications, HCC development, liver-related mortality, and overall mortality, thus supporting its role in early-stage HCC [4,11]. However, whether initiating DAAs before curative treatment confers added benefit remains to be determined. For patients with BCLC stage B/C disease, who typically receive palliative care, the sequencing of DAA therapy relative to HCC treatment is poorly studied [6,9,13]. Our findings indicate that achieving SVR is associated with improved overall survival even in this group, highlighting the need for further research on the optimal integration of DAA therapy into the treatment algorithm for intermediate to advanced HCC.

Fourth, the role of DAA therapy in patients receiving immunotherapy is currently unexplored. As previously stated [9], immunotherapeutic agents were not reimbursed under Taiwan’s National Health Insurance system during our study period (December 2013 to December 2020), which precluded evaluation of this potential interaction. Nevertheless, in the contemporary era of immunotherapy for unresectable HCC, studies examining DAA use in this context are urgently needed. The inclusion of immune-related biomarkers in such studies would be particularly valuable for elucidating the immunological consequences of DAA therapy.

Finally, we considered the application of competing risk analysis but ultimately determined that it was not necessary, as overall survival (OS) was our primary endpoint. Competing risks become relevant when the event of interest may be precluded by other events, such as death occurring before tumor development or progression [14,15]. However, when OS is used as the endpoint, death is the event of interest itself, eliminating the need for competing risk models. Moreover, distinguishing between liver failure–related death and HCC-related death is notoriously difficult in clinical settings, even in prospective trials [16], due to the shared contributions of cirrhosis and malignancy. For these reasons, OS remains the most objective and reliable outcome measure to assess survival benefits associated with therapeutic interventions.

In conclusion, we thank Dr. Wang and colleagues for the opportunity to engage in this meaningful academic dialogue. While our study provides compelling evidence supporting the survival benefit of DAA therapy in patients with HCC, we acknowledge the limitations raised and agree that further research is essential to optimize treatment strategies and personalize care for this complex patient population.

Abbreviations

ALT

alanine aminotransferase

AST

aspartate aminotransferase

BCLC

Barcelona Clinic Liver Cancer

CTP

Child-Turcotte-Pugh

DAA

direct-acting antiviral

eGFR

estimated glomerular filtration rate

HCC

hepatocellular carcinoma

N

numbers

OR

odds ratio

OS

overall survival

SVR

sustained virological response