Purpose/Objective(s): A number of chromosomal alterations have been identified as important factors in the initiation and progression of cervical cancer. These data, however, are mainly derived from surgical specimens of early stage tumors. The purpose of our study was to analyze the whole genome profile of locally advanced cervical cancers treated with radiotherapy and concurrent cisplatin (RTCT) and to identify recurrent DNA copy number alterations associated with the malignant phenotype.

Materials/Methods: Tumor biopsies of locally advanced cervical cancers were obtained at the time of examination under anesthesia, prior to RTCT. All biopsies were reviewed by a gynecologic pathologist. Macro-dissected tumor samples were co-isolated using the Qiagen All-Prep DNA/RNA/miRNA universal kit. DNA samples were submitted for shallow whole genome sequencing where reads were aligned to the hg19 reference genome with an average sequencing coverage of 0.38x. Copy-number was estimated using QDNAseq and ichorCNA by segmenting the genome into genomic bins of 50kb or 1000kb. Age, histology, grade, stage and lymph node status were recorded. Patients were treated with external radiotherapy (45-50 Gy) and concurrent weekly cisplatin (40 mg/m²), followed by pulsed-dose-rate (PDR) brachytherapy (40 Gy). Median follow up was 8.4 (range: 0.2-16.5) years.

Results: 121 patients were enrolled in the study from October 1999 to March 2012. They were representative of the cervical cancer population treated with RTCT. The median age was 48.5 (range: 26-84) years. FIGO stage was 1B in 31%, IIA-IIB in 40%, and IIIA-IIIB in 29%. The median percent genome altered (PGA) was 13.2% (range: 0-47.5%). Recurrent copy number amplification events were consistent with previously reported data and include: 11q22.1 (YAP1, BIRC2, BIRC3), 3q26.2 (MECOM, TERC), 1q23.3, 8q24.23 (MYC, PVT2), 20q11.21 (BCL2L1), and 9p24.1 (CD274, PDCDILG2). In contrast to previous reports, locus 7p11.2 coding EGFR was not amplified. Recurrent copy number deletions included: 2q37.1, 2q36.3, 11q23.3, 3p12.3, 19p13.3, and 13q14.11. Previously identified deletions at loci 4q35.2 (FAT1) and 10q23.31 (PTEN) were not seen in our cohort. There were no associations between PGA and any of the baseline clinical factors. There was no association between PGA and disease free survival.

Conclusion: In summary, this study improves our understanding of the genomic landscape in locally advanced cervical cancer. While our findings are similar to those reported previously for earlier stage disease, we also identified potentially important differences that warrant further evaluation. Future analyses will explore relationships between specific chromosomal alterations and gene expression profiles as well as their integrated effects on treatment response and survival. The ultimate goal is to identify genomics biomarkers to aid in the subclassification of disease, prediction of prognosis, and selection of patients who may benefit from treatment intensification.