Prognostic Value of Hypoxia Gene Expression Signature in Early-Stage Non-Small Cell Lung Cancer Patients Treated with Stereotactic Body Radiation Therapy

Abstract

Stereotactic body radiation therapy (SBRT) is an effective alternative to surgery in the treatment of early-stage non-small cell lung cancer (NSCLC). However, disease recurrence occurs in approximately 20-30% of patients. Hypoxia is a well-known factor promoting solid tumor progression, radiation resistance, and immune evasion. In this study, we utilize a previously published hypoxia gene expression signature (Buffa) and measure its association with clinical outcomes in patients with NSCLC treated with SBRT. Our study focused on patients with localized, node-negative NSCLC treated with SBRT, and we identified 92 patients treated at our institution between 2008 and 2018 with available gene expression data. Total RNA from formalin-fixed paraffin-embedded archival biopsy specimens (pre-therapy) was isolated and mRNA expression was analyzed using an assay for human samples. For each gene in the Buffa signature, the top 50% of mRNA abundance values were given a score of +1, and the bottom 50% were given a score of -1 to generate a tumor hypoxia score. High scores suggest a hypoxic tumor and low scores imply normoxia. Kaplan-Meier curves were used to assess overall survival, disease-free survival, local recurrence, regional nodal recurrence, and distant recurrence. Cox proportional hazards were performed to account for confounding due to age, T stage, ECOG performance status, biologically effective dose, patient sex, and histology (squamous vs non-squamous). The patient's gene expression data were dichotomized based on their median hypoxia score (Table 1). Median follow-up was 23.9 months (95% CI 20.6 - 26.6). A high hypoxia score was associated with squamous histology (p = 0.003). On univariate analysis, a high hypoxia score was significantly associated with local recurrence (hazard ratio [HR] = 5.63; 95% CI 1.03 - 30.78; p = 0.046) and distant recurrence (HR = 1.85; 95% CI 1.14 - 3.02; p = 0.013). There was no significant difference in overall survival (HR = 1.93; 95% CI 0.98 - 3.82; p = 0.058), disease-free survival (HR = 2.13; 95% CI 0.93 - 4.89; p = 0.074), or regional nodal recurrence (HR = 1.45; 95% CI 0.41 - 5.14; p = 0.57). On multivariate analysis, a high hypoxia score was associated with worse overall survival (HR = 2.73; 95% CI 1.23 - 6.07; p = 0.014), local recurrence (HR = 12.85; 95% CI 1.08 - 152.49; p = 0.043) and distant recurrence (HR = 1.91; 95% CI 1.13 - 3.22; p = 0.016). Our findings demonstrate that a hypoxia gene signature is significantly associated with worse survival and increased local and distant recurrence after SBRT for localized NSCLC. This hypoxia signature may be useful in prognostication, selecting patients for surgery versus radiation, or selecting patients for treatment intensification with radiation or other systemic agents, and should be prospectively validated.