Introduction Human Papilloma Viruses (HPV) infection may occur in 40% of rectal adenocarcinomas (RCa). HPV positivity (mainly HPV16) at other tumour sites is a predictor for treatment response to chemo-radiotherapy (CRT). Here, we performed a ‘proof of concept’ study in patients with RCa testing the hypothesis that HPV16 tumour positivity is a treatment predictive biomarker for complete response (CR) to CRT.

Method The study was a case-control design in 118 patients undergoing CRT for RCa (CR, 30: non-CR, 88) from two UK cancer centres (2008 to 2013). DNA was extracted from pre-treatment paraffin-embedded blocks with histologically verified carcinoma. High-sensitivity multiplex PCR for HPVs was performed using two sets of primers for each HPV to identify genotypes: HPV6, 16, 18, 33. The probability of a CR was expressed as odd ratios (ORs) using logistic regression models.

Results Any HPV positivity was noted in 25% of tumours. Individual genotype frequencies were: HPV16, 16%; HPV18, 6%; HPV33, 3%; HPV6, 3%. In multivariate models that included age, gender, pre-treatment T-size and N status, there was no association between either any HPV positivity (OR = 1.095, 95% CIs: 0.394, 3.044) or HPV16 positivity (OR = 1.072, 95% CIs: 0.332, 3.465), and complete response to CRT.

Conclusion In this stage 2 biomarker discovery study using high-sensitivity HPV multi-valent assays, we found tumour HPV positivity rates lower than those reported elsewhere in the literature. We found no clear ‘signal’ to take forward HPV genotyping for validation as a predictive biomarker for chemo-radiotherapy response in patients with rectal cancer.

This study was generously supported by the BDRF.

Disclosure of interest None Declared.

References

1. Baricevic-Jones, et al. High-sensitivity HPV genotyping reveals near universal positivity in anal squamous cell carcinoma: different implications for vaccine prevention and prognosis. Eur J Cancer [in press]