In this present publication, only recommendations for the use of HPV mRNA testing are presented. For other recommendations, please refer to the July 2021 publication of the WHO guideline for screening and treatment of cervical pre-cancer lesions for cervical cancer prevention, second edition (12), or to Annex 5 of this guideline.
In the general population of women, HPV DNA is the recommended primary screening test, but HPV mRNA detection may also be used.
When providing HPV mRNA testing, WHO suggests:
- ▪
providing it with or without triage;
- ▪
using samples taken by the health-care provider; and
- ▪
5-year screening intervals.
This is a conditional recommendation, based on low-certainty evidence.
Remarks:
- ▪
HPV DNA is the recommended screening test. Choosing the alternative option of HPV mRNA testing implies having the capacity to provide follow-up screening at 5-year intervals.
RECOMMENDATION:
WHO suggests that HPV mRNA detection using samples taken by the health-care provider may be used as a primary screening test, either with or without triage, to prevent cervical cancer in the general population of women with regular screening every 5 years. 
Note: No recommendation was made for using HPV mRNA in women living with HIV because evidence on the outcomes of using HPV mRNA detection applicable to this population was not identified.
For the detailed algorithms (clinical flowcharts) based on this recommendation, see Annex 4.
Justification
Despite the similar cross-sectional sensitivity and specificity of HPV mRNA testing compared with HPV DNA testing, a conditional recommendation was made for the use of HPV mRNA as a primary screening test because the longitudinal evidence on HPV mRNA test performance is uncertain. Modelling data suggest that there may be similar reductions in cervical cancer cases and deaths when using HPV mRNA testing with or without triage compared with HPV DNA testing with or without triage. In addition, there may be fewer treatments for pre-cancerous lesions when using HPV mRNA testing. However, the evidence from the mathematical model is uncertain, as the predicted reductions in cases and deaths when using HPV mRNA testing overlap with the uncertainty intervals for those with HPV DNA testing, and the model validation was performed against limited longitudinal data. Some longitudinal data with follow-up of more than five years and a model trial validation exercise (based on follow-up at 4–7 years) suggest that the incidence of CIN3+ may be higher in women who were negative for HPV mRNA compared with those who were negative for HPV DNA. There also do not appear to be other reasons related to feasibility or resources in favour of selecting HPV mRNA testing rather than HPV DNA testing.
The evidence available did not include women living with HIV, and data from the general population of women was not applicable to that population. Therefore, no recommendation was made for women living with HIV.
Summary of the evidence
The GDG considered evidence from cross-sectional and longitudinal studies, and from mathematical modelling of long-term outcomes. Well designed and implemented cross-sectional studies consistently found that compared with HPV DNA testing, the sensitivity of HPV mRNA is likely slightly lower and the specificity slightly higher at baseline: relative sensitivity and specificity for CIN2+ are 0.97 (95% CI: 0.95–1.00) and 1.03 (95% CI: 1.02–1.05), respectively, and for CIN3+ are 0.98 (95% CI: 0.95–1.02) and 1.03 (95% CI: 1.01–1.06), respectively (based on moderate-certainty evidence) (21). Few studies measured the longitudinal performance and performance over repeat rounds of screening with HPV mRNA tests. Some of the available long-term data suggest that women who test negative for HPV mRNA may have a higher subsequent incidence of CIN3+ than those who test negative for HPV DNA, especially when using longer screening intervals (5+ years), but the data are sparse and the findings are also inconsistent across studies (low-certainty evidence) (see Web annex: EtD framework for mRNA testing for HPV).
The model used data extracted from the cross-sectional studies in the systematic review on sensitivity and specificity and was validated against the available longitudinal evidence. The findings – based on low-certainty evidence – suggest that when comparing the long-term effects of repeated rounds of HPV mRNA testing with HPV DNA testing, when implemented programmatically at 5-year screening intervals, there may be up to 8–12% higher relative cervical cancer incidence and 6–8% higher cervical cancer mortality for those screened with HPV mRNA compared with HPV DNA tests. However, after considering uncertainties in observed HPV mRNA test performance in a sensitivity analysis, the predicted reductions in cases and deaths with HPV mRNA testing overlap with the ranges of uncertainty for HPV DNA testing (for detailed results, see Web annex: EtD framework for mRNA testing for HPV). The modelling evidence also suggests that the number of treatments for pre-cancer lesions may be lower than with HPV DNA testing strategies (27–33% fewer pre-cancer treatments), leading to lower costs (6–10% lower), although there are also uncertainties in these predictions, which are contingent on the available data.
When compared with VIA or cytology screening, HPV mRNA screening may result in greater reductions in cervical cancer incidence and mortality. There is very low-certainty evidence comparing samples taken by health-care providers with self-collected samples for HPV mRNA testing. Available data suggest that HPV mRNA testing performs worse with self-collected samples, and therefore the recommendation suggests samples taken by health-care providers (see Web annex: EtD framework for mRNA testing for HPV).
None of the studies reviewed contained information specifically for women living with HIV, and a separate analysis could not be performed. The GDG agreed that the agreed recommendation presented here does not apply to women living with HIV.
The GDG also considered whether there are advantages to using HPV mRNA testing over HPV DNA testing, and could not identify any advantages related to feasibility, acceptability, equity or resources, since implementation of HPV mRNA is similar to HPV DNA testing. The findings of the survey on values and preferences performed for the first phase of the guideline update were considered when formulating this recommendation. However, because collection of samples by health-care providers rather than self-sampling is suggested for HPV mRNA testing, the feasibility and acceptability of HPV mRNA testing may be different. The Aptima™ mRNA assay is the only HPV mRNA test that has been widely assessed to date. The prices of both HPV mRNA and HPV DNA testing are generally in the same range in high-income countries and both have similar equipment and training needs.
