3120 - A Nested Case-Control Study to Identify Candidate Biomarkers that Predict Radioresistance in Cervical Cancer
Presenter(s)
T. Oike1, K. Tomizawa1, S. Sakamoto1, Y. Miyasaka1, H. Hirata2, Y. Yoshimoto3, Y. Sasaki4, T. Tokino5, K. Ando1, and T. Ohno1; 1Department of Radiation Oncology, Gunma University Graduate School of Medicine, Maebashi, Japan, 2Department of Radiation Oncology, National Cancer Center Hospital East, Kashiwa, Japan, 3Department of Radiation Oncology, Fukushima Medical University School of Medicine, Fukushima, Japan, 4Center for Medical Education, Sapporo Medical University, Sapporo, Japan, 5Research Institute for Frontier Medicine, Sapporo Medical University, Sapporo, Japan
Purpose/Objective(s): Despite technological advances in radiotherapy, including image-guided brachytherapy, a subset of uterine cervical cancers recurs locally after definitive treatment, highlighting the need to establish biomarkers that predict radioresistance and facilitate treatment personalization. However, low local recurrence rates post-radiotherapy constitute a barrier to research progress; i.e., omics analyses that enable unbiased exploration of candidate genes across an entire cohort with sufficient sample size to achieve adequate statistical power is difficult and costly. Indeed, previous studies with this research aim focused primarily on investigating given genes of interest using targeted approaches such as qPCR or immunohistochemistry. To overcome these obstacles, we tested the hypothesis that candidate biomarkers that predict radioresistance of uterine cervical cancer can be identified by comprehensive omics analyses using a nested case-control cohort approach.
Materials/Methods: From a series of 192 patients with squamous cell carcinoma of the uterine cervix treated with definitive radiotherapy, a discovery cohort (n = 8, with available pretreatment frozen tumor samples) was created by 1:1 random case-control matching on pelvic recurrence (PR), and a validation cohort (n = 54, with available pretreatment formalin-fixed and paraffin-embedded tumor samples) was created by 1:2 propensity score matching on PR. Tumor samples from the discovery cohort were analyzed by RNA sequencing, followed by Gene Set Enriched Analysis (GSEA). Samples from the validation cohort were analyzed by a panel-based DNA sequencing. PR-free survival (PRFS) was analyzed using the Kaplan–Meier method, the log-rank test, and cause-specific proportional hazards models.
Results: GSEA of the discovery cohort revealed that genes targeted by MYC were upregulated in patients with PR (MYC_targets_v1, normalized enrichment score [NES] =1.6, p < 0.001; MYC_targets_v2, NES =1.8, p < 0.001). In the validation cohort, the MYC amplification was significantly more prevalent in patients with PR than in those without (72% vs. 36%, respectively; p = 0.020). Patients with MYC amplification had significantly worse PRFS than those without (p = 0.020), with 5-year PRFS rates of 47% (95% confidence interval [CI]: 27%–65%) and 81% (95% CI: 59%–91%), respectively. Furthermore, multivariate analysis identified MYC amplification as the only significant predictor of PR (hazard ratio: 3.1, 95% CI: 1.1–9.8) among the parameters examined (i.e., MYC amplification, pretreatment tumor volume, pelvic lymph node status, and concurrent chemotherapy).
Conclusion: Comprehensive omics analyses using a nested case-control cohort approach identified MYC amplification as a candidate biomarker that predicts radioresistance of uterine cervical cancer.