AUTHOR=Li Yongjiang , Pan Yangxun , Lin Ximeng , Hou Jingyu , Hu Zili , Xu Li , Zhou Zhongguo , Zhang Yaojun , Chen Minshan , Hu Dandan 

TITLE=Development and Validation of a Prognostic Score for Hepatocellular Carcinoma Patients in Immune Checkpoint Inhibitors Therapies: The Hepatocellular Carcinoma Modified Gustave Roussy Immune Score

JOURNAL=Frontiers in Pharmacology

VOLUME=12

YEAR=2022

URL=https://www.frontiersin.org/journals/pharmacology/articles/10.3389/fphar.2021.819985

DOI=10.3389/fphar.2021.819985

ISSN=1663-9812

ABSTRACT=<p><bold>Background:</bold> There is not yet an effective marker in predicting the efficacy of immune checkpoint inhibitors (ICIs) in treating hepatocellular carcinoma (HCC) patients. The Gustave Roussy Immune Score (GRIm-Score) based on three objective variables, namely, neutrophil-to-lymphocyte ratio (NLR), serum albumin level (ALB), and lactate dehydrogenase (LDH), was developed as feasible prognostic indication in lung cancer patients receiving ICIs therapies. Our study aimed to adapt the GRIm-Score (HCC-GRIm-Score) in HCC patients who received ICIs therapies and thus improving the predictive ability.</p><p><bold>Methods:</bold> From January 2018 to September 2020, 261 patients who received ICIs therapy were retrospectively included and divided into training and validation groups. After determining the factors for HCC-GRIm-Score by multivariable analysis from training group, the optimized HCC-GRIm-Score was validated and compared to the original GRIm-Score and the Barcelona clinic liver cancer (BCLC) staging system.</p><p><bold>Results:</bold> One hundred sixty-one and 80 patients were assigned into the training and validation groups, respectively. Two more factors, aspartate transaminase-to-alanine transaminase ratio [hazard ratio (HR), 1.51; 95% confidence interval (CI), 0.94–2.42] and total bilirubin (HR, 1.76; 95% CI, 1.07–2.88), were identified as independent prognostic factors for overall survival (OS) and integrated in the HCC-GRIm-Score system according to the multivariable analysis. A risk score based on the HCC-GRIm-Score indicated that patients presenting high score (&gt;2) suffered from significantly shorter median OS of 10.3 months compared to those with a low score (not reached; HR, 2.99; 95% CI, 1.89–4.75; <italic>p</italic> &lt; 0.001). In the validation group of 80 patients, the patients presenting a high score showed an inferior OS (HR 5.62, 95% CI, 1.25–25.24; <italic>p</italic> = 0.024). HCC-GRIm-Score had the highest area under curve of 0.719 (95% CI, 0.661–0.773) compared to original GRIm-Score and BCLC staging system.</p><p><bold>Conclusion:</bold> The present study confirmed that the modified HCC-GRIm-Score system provided superior predictive ability in identifying the HCC patients potentially benefit from ICIs therapies, compared to the original GRIm-Score and the BCLC staging system.</p>