4026 Background: Hepatocellular carcinoma (HCC) has rising incidence and mortality rates. Tumor hypoxia is important in HCC pathogenesis but has not been effectively translated into practice. We studied whether an RNA expression-based hypoxia score (HS) can serve as a prognostic and predictive biomarker in HCC. Methods: Solid tumors across a range of tissues (N=91516) were tested at Caris Life Sciences (Phoenix, AZ) with NextGen Sequencing of DNA (592-gene or whole exome) and RNA (whole transcriptome), including 432 HCC. Mutation prevalence (Mt) was calculated for pathogenic SNVs/indels. PD-L1 expression (SP142; +: ≥2+, ≥5%) was tested by IHC. HS based on RNA expression of 15 genes and normalized across a range of solid tumors was implemented as previously described (Bhandari et al, 2019). Tumors were defined as HS high (-H), medium (-M) and low (-L) by hierarchical clustering. A transcriptomic signature associated with immunotherapy response (T-cell inflamed score) was applied. Fisher’s Exact/χ 2 tests were applied as appropriate with p-values adjusted for multiple comparisons ( p < 0.05). Real-world overall survival (OS) data was obtained from insurance claims and log-rank estimates were calculated for molecularly defined subpopulations. Results: Of 432 HCC, 5% were classified as HS-H, 36% HS-M and 59% HS-L. Median age was 67 years (HS-H:70, -M:67, -L:67, p = 0.45), and 77% were male (HS-H: 68%, -M: 75, -L: 79, p= 0.43). The proportion of metastatic tumors did not significantly differ across the three HS groups (%; -H: 36, -M: 29, -L: 23, p= 0.16) nor did the score vary across metastatic sites (% HS-L; Adrenal gland: 46, Bone: 66, Lymph: 52, Misc.: 46, p= 0.49). Prevalence of TP53-Mt in HS-H (72.7%) was higher than in -M (39.5%) and -L (29.1%, p< 0.01). HS-H had higher prevalence of PD-L1+ and T cell-inflamed tumors as compared to -M and -L (PD-L1+ (%): 28.6 v 6.2 v 3.9, p< 0.01; T cell-inflamed (%): 86 v 46 v 17, p< 0.001). HS-H tumors had significantly shorter OS from date of collection (166 days [D]) v -M (396.5 D) and -L (513 D, p< 0.001). Yet, no significant difference in survival since start of treatment (SSOT) was observed in the patient subset that received Atezolizumab (Atezo) (-H: 210 D, -M: 355 D, -L: 386 D, p= 0.54). In HS-M tumors alone, SSOT was similar across investigated therapeutics (Atezo 322 D, Lenvatinib (Lenvat) 366.5 D, Sorafenib (Soraf) 289 D, p= 0.34). In HS-L tumors, patients treated with Soraf (1341 D) had significantly longer SSOT than with Lenvat (444 D) or with Atezo (281.5 D, p= 0.002). Conclusions: In HCC, RNA expression-based HS is associated with TP53 mutations and an immunosuppressive microenvironment. HS-H tumors has worse OS that may be improved with Atezo, whereas HS-L tumors may respond better to Soraf. HS is a potential prognostic and predictive biomarker in HCC that deserves validation in orthogonal data sets and prospective studies.
No takes yet. Share an insight, caveat, or question.
Connor et al. (2024) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: