TL;DR
This study investigates the potential of alpha-fetoprotein (AFP) and soluble intercellular adhesion molecule-1 (sICAM-1) as biomarkers for predicting prognosis and response to immunotherapy in hepatocellular carcinoma (HCC). The findings suggest that combining these markers offers improved diagnostic and prognostic accuracy, and that low levels of AFP and sICAM-1 may predict better outcomes with immune checkpoint blockade therapy.
What Problem Does This Paper Address?
Hepatocellular carcinoma (HCC) remains a significant global health challenge due to its high mortality rate, often linked to difficulties in early diagnosis and effective treatment. While alpha-fetoprotein (AFP) and soluble intercellular adhesion molecule-1 (sICAM-1) are known diagnostic markers for HCC, their full potential in predicting patient prognosis, particularly their response to emerging immunotherapies, is not well understood. This paper aims to bridge this knowledge gap by exploring how AFP and sICAM-1 levels, both individually and in combination, can predict tumor recurrence, survival, and response to immune checkpoint blockade (ICB) therapy in HCC patients.
Methods
The study employed a multi-faceted approach combining clinical data analysis with bioinformatics techniques.
- Clinical Analysis: 87 HCC patients undergoing surgical resection had their pre- and post-operative serum levels of AFP and sICAM-1 measured. These levels were analyzed in relation to tumor recurrence and overall survival (OS) rates. Patients were grouped based on their marker levels (normal vs. elevated) and changes in levels (e.g., normalized post-surgery).
- Bioinformatics Analysis: Publicly available RNA-seq and clinical data from The Cancer Genome Atlas (TCGA) Liver Hepatocellular Carcinoma (LIHC) dataset (371 HCC samples and 50 normal samples) were used to:
- Evaluate the diagnostic and prognostic accuracy of AFP and sICAM-1 using Receiver Operating Characteristic (ROC) curves and logistic regression models.
- Analyze the relationship between AFP/sICAM-1 expression and immune cell infiltration using the xCell algorithm.
- Assess the potential of AFP and sICAM-1 in predicting immunotherapy response using the Tumor Immune Dysfunction and Exclusion (TIDE) algorithm and the ImmuCellAI tool.
- Investigate the correlation between AFP/sICAM-1 expression and immune checkpoint-relevant genes.
Datasets
- Clinical Dataset: Serum samples from 87 patients with primary HCC treated at Xijing Hospital, China, between May 2017 and December 2017. Follow-up data was collected for 18-66 months.
- TCGA-LIHC Dataset: RNA sequencing and clinical data from 371 HCC samples and 50 normal samples, accessed via the Genomic Data Commons (GDC) data portal.
Key Insights and Ideas
- Synergistic Prognostic Power: While AFP and sICAM-1 have individual diagnostic value, their combined assessment appears to offer a more robust prediction of prognosis than either marker alone.
- Normalization is Key: Patients whose AFP and sICAM-1 levels normalize after surgery, or remain normal, show significantly better outcomes (lower recurrence, better survival) compared to those with persistently elevated levels. This highlights the importance of post-treatment marker dynamics.
- Immune Microenvironment Link: AFP and sICAM-1 expression levels are significantly correlated with the infiltration of various immune cells in HCC tumors, but not in normal tissues. This suggests a role in shaping the tumor microenvironment (TME).
- Predicting Immunotherapy Response: Low expression of AFP and sICAM-1 is associated with a more favorable immune microenvironment and potentially better response to immune checkpoint blockade (ICB) therapy. Conversely, high expression may indicate poorer response.
Key Results
- Clinical Cohort: Patients with normal pre-operative AFP (< 20 µg/L) or sICAM-1 (< 1000 µg/L), or those whose levels normalized post-surgery, exhibited lower recurrence rates and better OS compared to patients with elevated levels.
- TCGA Data:
- The combined application of AFP and sICAM-1 showed higher accuracy in both diagnosis (AUC: 0.783) and prognostic prediction (AUC: 0.659 at 1 year, 0.562 at 5 years) compared to individual markers.
- AFP and sICAM-1 were significantly correlated with numerous immune cells (e.g., macrophages, monocytes, T cells) in tumor samples, but not in normal samples.
- Patients with low AFP or low sICAM-1 expression had higher immune and stromal scores, which are generally associated with a favorable prognosis.
- Analysis using TIDE and ImmuCellAI suggested that patients with high AFP or high sICAM-1 levels had higher TIDE scores, indicating a poorer response to ICB therapy.
Limitations and Future Work
- Sample Size: While the study used both a clinical cohort and TCGA data, the authors acknowledge that further exploration with expanded sample volumes is needed to solidify the clinical utility of AFP and sICAM-1 for prognosis evaluation.
- Inconsistent Cutoffs: The study notes that cutoff values for AFP and sICAM-1 can be inconsistent across studies, possibly due to variations in sample size.
- Mechanism Exploration: While the study identifies correlations between AFP/sICAM-1 and immune cells, further research is needed to fully elucidate the underlying mechanisms.
- Clinical Practice Integration: The authors suggest that while AFP and sICAM-1 show promise as prognostic predictors and potential indicators for immunotherapy response, their integration into routine clinical practice requires further validation. They also highlight the potential role of HAVCR2 (encoding TIM-3) as a target for ICB therapy, particularly in patients with low AFP and ICAM-1.
Summary updated