Dataset Information

Molecular predictors of prevention of recurrence in hepatocellular carcinoma with sorafenib as adjuvant treatment in the phase 3 STORM trial

ABSTRACT: BACKGROUND & AIMS: Sorafenib is the standard systemic therapy for advanced hepatocellular carcinoma (HCC). In early-stage HCC, survival benefits of resection/local ablation are compromised by high recurrence rates (70% at 5-years). In this setting, the phase-3 STORM-trial did not achieve its primary end-point of improving recurrence-free-survival (RFS) comparing sorafenib to placebo as adjuvant treatment. Herein we present results of the biomarker companion study BIOSTORM in which we define predictors of recurrence prevention with sorafenib and prognostic factors with B-level of evidence. METHODS: Tumor tissue from 188 patients randomized to receive sorafenib (83) or placebo (105) in the STORM trial was collected. Analyses included gene-expression profiling, targeted exome sequencing (18 known onco-drivers), immunohistochemistry (pERK, pVEGFR2, Ki67), FISH (VEGFA) and immunome. A gene signature capturing improved RFS in sorafenib-treated patients was generated. All 70 RFS events were recurrences, thus time-to-recurrence equaled RFS. Predictive and prognostic value was assessed using Cox regression models and interaction test. RESULTS: BIOSTORM recapitulates clinical and pathological characteristics of STORM patients. None of the biomarkers tested that are known to be related to angiogenesis and proliferation, or previously proposed gene-signatures, or mutations predicted sorafenib benefit. A newly generated 146-gene signature identifying 30% of patients captured benefit to sorafenib in terms of RFS (p-of-interaction=0.04). These sorafenib responders were significantly enriched in CD4+ T, B, and cytolytic Natural Killer cells, and showed absence of activated macrophages and other components of adaptive immunity. Hepatocytic pERK (HR=2.41; p=0.012) and microvascular invasion (HR=2.09; p=0.017) were independent prognostic factors. CONCLUSIONS: In BIOSTORM, hepatocytic pERK and microvascular invasion predicted poor RFS. No mutation, gene amplification or previously proposed gene-signatures predicted sorafenib benefit. A newly generated multi-gene signature associated with improved RFS on sorafenib warrants further validation. KEYWORDS: sorafenib; tyrosine-kinase inhibitor; trial enrichment; stratification.

ORGANISM(S): Homo sapiens

PROVIDER: GSE109211 | GEO | 2018/08/21

REPOSITORIES: GEO

ACCESS DATA

Similar Datasets

Project description:Purpose: To determine the 8-week disease control rate (DCR) of sorafenib monotherapy in patients with advanced non-small-cell lung cancer (NSCLC) in the BATTLE trial. Methods: Patients with pre-treated NSCLC consented to baseline biopsies for pre-specified biomarkers assessment and biomarker discovery. Sorafenib was given at 400 mg orally twice daily until tumor progression or unacceptable toxicity. Outcomes by pre-specified biomarkers were analyzed and a Sorafenib sensitivity signature developed using high-throughput gene expression profiles of NSCLC cell lines and baseline biopsies. Results: 105 patients were eligible and 98 patients were evaluable. Median age was 62 (range 34-81) years, 51% of patients were male, 75% were former/current smokers, and 89% had an ECOG performance status of 0-1. Median prior chemotherapies for stage IV NSCLC were two. Median follow-up was 9.4 (range: 1.3-32.2) months. Overall, 8-week DCR was 58.2%. Patients with EGFR mutations had significantly lower 8-week DCR compared to patients with wild-type tumors (23.1% vs. 64.2%, P=0.0119), and patients with K-RAS mutations had the highest 8-week DCR (67%). Most commonly reported treatment-related adverse events include hand-foot syndrome (59.6%), fatigue (42.3%), rash (40.4%), diarrhea (38.5%), and weight loss (38.5%). Sorafenib sensitivity signature developed in cell lines was associated with an improved outcome. Conclusion: Patients with wild-type EGFR, including those with K-RAS mutation, may benefit from sorafenib as opposed to patients with EGFR mutation. We identify a gene expression signature associated with an improved outcome in patients with wild-type EGFR treated with sorafenib. BATTLE-2 trial is ongoing to validate those results. ClinicalTrials.gov number, NCT00411671. Gene expression profiles were measured in 37 core biopsies from patients with refractory non-small cell lung cancer treated with sorafenib in the Biomarker-integrated Approaches of Targeted Therapy for Lung Cancer Elimination (BATTLE) trial. We report a Sorafenib sensitivity gene expression signature trained in vitro (separate GEO submission), and tested in baseline biopsies collected in the BATTLE trial.

Project description:Introduction: The I-SPY 1 TRIAL was designed to evaluate complete pathologic response and tumor volume change, measured by magnetic resonance imaging (MRI), stratified by molecular subtypes and link response to 3-year recurrence free survival (RFS).Methods: Eligible patients had T =3 cm and received neoadjuvant chemotherapy with AC plus optional taxane. Serial MRI, biopsy, and blood draws were conducted over the course of treatment, and a database of multiple molecular profiles was assembled.Results: A total of 221 patients were eligible for analysis: median tumor size was 6.0 cm and median age was 49 years. The I-SPY 1 TRIAL patients had tumors with aggressive biology: 45% were estrogen receptor (ER) negative, 31% Her2+; and 91% high risk by the Netherlands Cancer Institute (NKI) 70-gene profile. After a median of 3.9 years, RFS was 77%. HR/HER2 status improved predictability of RFS with the greatest difference between HR+/HER2- and triple-negative disease (hazard ratio 0.39). Wound healing signature activation, p53 mutation, and Risk of Relapse (ROR) score were highly correlated and also significantly improved the accuracy of RFS predictions when added to stage. The rate of pathologic complete response (pCR) varied considerably from a low of 0-2% (NKI low risk, luminal A) to a high of 43-61% (17q Amplified, HER2 enriched, HR-/HER2+). Both pCR and the more refined residual cancer burden (RCB) were significant predictors of RFS for all patients and even more predictive when analyzed within biomarker subsets. Good risk (NKI low, ROR-S low risk, Wound Healing quiescent, p53 wild type) signatures were associated with significantly higher 3-year RFS than poor risk expression signatures (ROR-S high risk, Wound Healing Activated, p53 mutation, NKI high risk).Conclusion: Importantly, in this set of biologically poor prognosis tumors, pCR predicts for better outcome, especially when analyzed within breast cancer subsets. Keywords: reference x sample reference x sample

Dataset Information

Molecular predictors of prevention of recurrence in hepatocellular carcinoma with sorafenib as adjuvant treatment in the phase 3 STORM trial

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure