Healthcare decision makers are keen to implement cervical cancer (CC) screening strategies that produce better clinical outcomes, while controlling the cost. These choices are complex and evidence often incomplete. This modelling study facilitates this process by comparing clinical benefits and costs of replacing; 1) cytology with pooled HPV triage (current practice), with 2) cobas HPV genotyping and reflex CINtec PLUS Cytology triage (comparator) in the national CC screening program of Norway.
Sensitivity limitations of cytology and moderate specificity of pooled HPV testing, may require more frequent follow-ups or unnecessary invasive procedures. The cobas HPV test identifies women at increased risk for CC, by detecting the two highest-risk genotypes 16/18, which cause 70% of all CC. The CINtec PLUS confirms a transforming HPV infection by detecting cervical cells where HPV has disrupted cellular control (p16/Ki-67+). CINtec PLUS triage of HPV genotype 16/18+ predicts which women most likely have pre-cancerous cervical lesions and therefore benefit from an immediate colposcopy.
The Markov model compares clinical impacts and annual costs of the screening strategies. A hypothetical cohort of 836,000 Norwegian 25-65 year old women goes through two screening cycles in the model. Natural progression/regression of the disease is also modelled. In the current practice; cytology negative women return to routine screening, a reflex HPV test is done for ASCUS and LSIL results, and women with HSIL undergo a colposcopy. In the comparator strategy; HPV negative women return to routine screening, a reflex CINtec PLUS done for HPV genotypes other than 16/18, and women with the high risk genotypes 16/18 undergo a colposcopy. The screening intervals are 3 years in the current and 5 years in the comparator strategy. Test sensitivity and specificity data are from the ATHENA study1. Other inputs include prevalence of HPV, HPV genotypes 16/18, abnormal cytology, CIN and CC in Norway. Screening, diagnosis and CC treatment costs are calculated from a healthcare provider’s perspective in 2015.
The comparator strategy increases the detection of CC cases from 50.7% to 90.2% and reduces annual incidence of CC in the screened population from 154.8 to 71.9. Importantly, it reduces annual costs by 4%, from 145 to 139 million NOK, which is driven by the longer screening interval and averted CC treatment costs.
The results suggest that replacing cytology with pooled HPV triage, with cobas HPV primary screening with genotyping and reflex CINtec PLUS Cytology triage, in the national cervical cancer screening program of Norway produces better clinical outcomes and saves costs.
1. Castle, P., Stoler, M., Wright, T., Sharma, A., Wright, T., & Behrens, C. (2011). Performance of carcinogenic human papillomavirus (HPV) testing and HPV16 or HPV18 genotyping for cervical cancer screening of women aged 25 years and older: a subanalysis of the ATHENA study. The Lancet Oncology, 12(9), 880-890. http://dx.doi.org/10.1016/s1470-2045(11)70188-7