412P - Multiplex digital spatial profiling (DSP) of protein reveals distinct immune and molecular phenotypes in hepatocellular carcinoma

Background

HCC is one of the most prevalent cancers with high death rate in Asia. To understand tumoral heterogeneity, differences between tumor and adjacent liver tissue at molecular level and tumor microenvironment (TME) change after multi-targeted kinase inhibitor sorafenib (Sora) treatment (Tx), DSP was used for protein profiling in pre- and post- Sora Tx HCC formalin-fixed, paraffin-embedded (FFPE) samples.

Methods

Total 25 FFPE slides from 13 HCC patients were analyzed by GeoMx^TM DSP protein next generation sequencing, including 9 patients with paired pre- and post- Sora Tx HCC samples. Using EGFR and MET immunohistochemistry (IHC) expression level as references, total 332 regions of interest (ROIs) were identified from tumor tissue and adjacent non-tumor liver tissue for DSP protein profiling. Besides Human Protein Core, 8 more panels were selected to cover key cancer signaling and immune-related proteins in the experiment. Multivariate linear regression was used to investigate the difference at different tissue type, Tx status and individual level.

Results

EGFR abundance measured by DSP was significantly higher in EGFR IHC-high (IHC score = 3+) compared with IHC-low (IHC score ≤ 1+) group (p = 4.96e-15). Similar trend observed in MET expression between DSP and IHC further warranted the performance of DSP protein profiling. Hierarchical clustering revealed expression patterns of ROIs from different cells of origin. Known IO targets such as PD-1 (p = 1.17e-10) was with higher abundance in ROIs of liver tumor region comparing to those of adjacent hepatocytes. When focusing on liver tumor region, most markers such as Phospho-AKT1 and Phospho-ERK1/2 upregulated in post- Sora ROIs which reflects the cell proliferation signal at disease progression after Sora Tx. IO targets under clinical development such as GITR and Tim-3 also showed upregulation in post-Tx ROIs.

Conclusions

This study, to our knowledge, represents the first DSP analysis of HCC and shows high concordance with IHC results. Immune and molecular pattern discovered in different tissue type and Tx stage improved our understanding of HCC TME and provided insights for future HCC clinical development.

Clinical trial identification

Editorial acknowledgement

Legal entity responsible for the study

Janssen Research & Development.

Funding

Janssen Research & Development.

Disclosure

C.J. Yen: Non-Financial Interests, Personal, Principal Investigator: National Cheng Kung University college of medicine. M. Qing, Q. Wu, X. Lyu, M. Xia, F. Yang, H. Xu: Financial Interests, Personal, Full or part-time Employment: Janssen research& development. L. Zhou: Financial Interests, Institutional, Full or part-time Employment: Janssen Research & Development.