Preview

PDF, English Download (1MB) | Terms of use

Abstract

Cervical cancer (CxCa) is the second most common cancer among women world-wide. DNA of 14 high-risk human papillomavirus (hrHPV) types is found in almost all cervical cancers (>96.3%), with HPV16 being the most prevalent type. Cervical cells can be screened for abnormalities using the cytological Papanicolaou (Pap) test and for the presence of HPV DNA and HPV E6/E7 fulllength mRNA. However, all of the available tests show considerable drawbacks exhibiting either poor clinical sensitivity or specificity. Reduced clinical specificity leads to over-treatment, additional costs and enormous anxiety for women concerned. A highly specific identification of high-grade lesions is desirable since those require surgical therapy contrary to early lesions that are likely to regress spontaneously. Recently described HPV16 RNA patterns comprising the upregulated transcripts E6*II and/or E1C, and the downregulated transcripts E1^E4 and/or L1 in HPV16-transformed cervical cells showed potential to substantially improve HPV-based cervical precancer screening: in a small cytology-based pilot study these patterns identified severe HPV16- induced cervical lesions with a clinical sensitivity of 74% and a specificity of 84%. However, singleplex nucleic acid sequence-based amplification (NASBA) assays used in the previous work was labour, time and cost intensive. In this PhD thesis, novel quantitative multiplex high-throughput reverse transcription PCR were developed for the detection of the HPV16 RNA patterns with detection limits below 100 transcript copies per PCR. Cross-reactivity with unspliced HPV16 sequences and cellular background DNA/RNA was excluded. The comparison of the newly developed assay and the established singleplex NASBA assays using RNA from 32 HPV16-positive fresh-frozen oropharyngeal squamous cell carcinomas indicated a good quantitative correlation. The concordance of HPV16 RNA patterns between both methods was 100%. In order to validate HPV16 RNA patterns as diagnostic marker for cervical cancer and its severe precursor lesions 165 single HPV16 DNApositive cervical cell samples were analysed. The sensitivity for CIN3 and CxCa was 88% and the specificity 84%, even in archived specimens with low RNA quality. Several HPV16 RNA patterns “false-positive” CIN1 lesions in follow-up had progressed to CIN2 or CIN3 lesions, raising the possibility that HPV RNA patterns could be an earlier marker for developing severe lesions than histology. Poor quality of RNA extracted from formalin-fixed paraffin embedded tissues strongly limits HPV16 RNA patterns analysis. To extend the HPV RNA patterns to the further seven most frequent hrHPV types 18, 31, 33, 35, 45, 52 and 58, unknown splice junctions were identified. For HPV18 singleplex assays were developed. The detection limit ranged between 101 E6*I and E1C and 100 E1^E4 copies per PCR and will be improved in future experiments.