Letter to the Editor: AADN score predicts overall and progression-free survival in triple therapy for hepatocellular carcinoma

Abstract

The integration of transarterial chemoembolization with immune checkpoint inhibitors and tyrosine kinase inhibitors has reshaped the therapeutic landscape of unresectable hepatocellular carcinoma. Nonetheless, outcome heterogeneity remains substantial, underscoring the need for simple and biologically grounded prognostic tools. In this recent issue of the World Journal of Gastroenterology, Zhang et al propose the AADN score, a laboratory-based model incorporating alpha-fetoprotein, alkaline phosphatase, direct bilirubin, and neutrophil-to-lymphocyte ratio. Developed using multivariable regression and validated in an independent cohort, the AADN score demonstrated good discrimination for overall and progression-free survival. This article contextualizes the score within existing prognostic systems, discusses its biological plausibility and methodological considerations, and critically appraises its current level of evidence. While the AADN score shows promise for risk stratification and hypothesis generation, further prospective and multicenter validation is required before it can inform routine clinical decision-making.

Key Words: Hepatocellular carcinoma; Oncology; Interventional radiology; Overall survival; Liver cancer

Core Tip: The AADN score, derived from alpha-fetoprotein, alkaline phosphatase, direct bilirubin, and neutrophil-to-lymphocyte ratio, provides a simple yet powerful tool to predict overall and progression-free survival in patients with unresectable hepatocellular carcinoma treated with transarterial chemoembolization, sintilimab, and lenvatinib. By integrating markers of tumor biology, hepatic function, and systemic inflammation, this model enables individualized risk stratification and supports precision-based decision-making in triple therapy for hepatocellular carcinoma.

TO THE EDITOR

Hepatocellular carcinoma (HCC) remains a leading cause of cancer-related mortality worldwide despite advances in surveillance and treatment[1]. For patients with intermediate or advanced-stage disease, curative options are frequently unavailable, and multimodal strategies combining locoregional and systemic therapies have become central to management within contemporary Barcelona clinic liver cancer recommendations[2]. The combination of transarterial chemoembolization (TACE) with immune checkpoint inhibitors (ICI) and tyrosine kinase inhibitors (TKI) has shown synergistic antitumor activity and survival benefit in selected populations[3,4]. However, marked interpatient variability in outcomes persists, highlighting the unmet need for reliable, accessible prognostic tools to support risk stratification in this complex therapeutic setting. Against this background, I read with interest the study by Zhang et al[5], published in the recent issue of the World Journal of Gastroenterology, which report a multicenter retrospective study introducing the AADN score as a prognostic model for unresectable HCC (uHCC) patients treated with TACE, sintilimab, and lenvatinib. The AADN score is based on four routinely available laboratory parameters: Alpha-fetoprotein (AFP), alkaline phosphatase (ALP), direct bilirubin (DBIL), and the neutrophil-to-lymphocyte ratio (NLR). Each variable captures a distinct biological domain relevant to treatment response. AFP reflects tumor burden and aggressive tumor biology; ALP is associated with biliary involvement and intrahepatic tumor spread; DBIL provides a sensitive measure of hepatic excretory function and cholestasis; and NLR represents systemic inflammation and immune dysregulation, which are increasingly recognized as key determinants of immunotherapy efficacy. By integrating tumor biology, liver function, and host inflammatory status, the AADN score conceptually aligns with the multifactorial nature of outcomes in TACE-ICI-TKI therapy. From a methodological standpoint, the authors identified AFP ≥ 100 ng/mL, ALP > 120 U/L, DBIL > 7.3 μmol/L, and NLR > 2.5 as independent predictors of overall survival in a cohort of 188 patients.

The score was derived from regression coefficients and applied to stratify patients into low-, intermediate-, and high-risk groups, which demonstrated clearly separated survival curves in both the training and validation cohorts. The reported area under the curve values exceeding 0.82 indicate good discriminatory performance. For practical interpretation, it is important to clarify that the AADN score is a weighted model based on regression coefficients rather than a simple unweighted summation of variables. Risk stratification thresholds were subsequently defined according to the total score distribution, enabling categorization into prognostic groups. Explicit description of this calculation framework is essential for readers seeking to understand how the score may be applied or reproduced in future studies. When compared with existing prognostic systems such as the CRAFITY[6] and TAE[7] scores, the AADN score demonstrated higher reported area under the curve values. However, these comparisons are based on literature-derived performance metrics rather than direct head-to-head testing within the same patient cohort. Differences in development populations, treatment regimens, and endpoint definitions should therefore be acknowledged, and claims of superiority should be interpreted cautiously. The potential incremental value of AADN may plausibly relate to the inclusion of DBIL as a more precise marker of hepatic functional reserve and to the combined assessment of inflammation and cholestasis, but these hypotheses require formal comparative validation. Despite its strengths, several methodological considerations warrant explicit discussion. The rationale for selected cutoff values, including AFP ≥ 100 ng/mL and NLR > 2.5, is not fully elucidated and may reflect data-driven optimization within a relatively modest sample size. This raises the possibility of overfitting, particularly given the limited number of events relative to predictors. In addition, details regarding handling of missing data, model calibration, and decision-analytic performance were not extensively reported, limiting assessment of clinical reliability beyond discrimination alone. These factors should be addressed in future studies to strengthen confidence in the model. The clinical implications of the AADN score should therefore be framed as exploratory rather than definitive. Rather than supporting immediate routine use, the score may serve as a tool for hypothesis generation, trial stratification, or exploratory risk enrichment.

Importantly, all parameters are derived from routine laboratory testing, enabling dynamic longitudinal assessment during treatment follow-up. From a mechanistic perspective, the strength of the AADN score rests on its foundation in inflammation-immunity-liver function interplay, aligning with current understanding of the tumor-immune microenvironment in HCC[8].

Locoregional ischemia induced by TACE releases tumor antigens, while ICIs restore exhausted T-cell function and TKIs modulate the vascular and immunologic milieu. In this complex setting, systemic inflammatory markers such as NLR and hepatic function indicators like DBIL or ALP become not merely prognostic correlates but reflections of therapeutic biology[9,10]. The AADN score thus operationalizes these interdependent processes into a practical quantitative tool.

Conceptually, low-risk patients might be considered suitable candidates for aggressive triple therapy, intermediate-risk patients for close monitoring and early response assessment, and high-risk patients for alternative strategies or intensified supportive care; however, such pathways remain hypothetical and require prospective testing.

In summary, Zhang et al[5] present a biologically plausible and accessible prognostic model for patients with uHCC undergoing TACE-ICI-TKI therapy. The AADN score illustrates how routinely available laboratory data can be integrated into a structured risk stratification framework. While promising, its role should currently be viewed as investigational, pending prospective validation in larger, ethnically diverse, and treatment-heterogeneous cohorts. Such efforts will be critical to determine whether the AADN score can ultimately contribute to precision-based decision-making in advanced HCC.