Long-Term Outcomes of Co-Testing with 3-Type HPV mRNA (16/18/45) and Cytology in Women Under 40: A Real-World Cohort from Northern Norway (8–10 Years of Follow-Up)

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

I congratulate the authors on their work – a presentation is sometimes 8-10 years old; the work is well-structured, written in correct English, and easily read by the reader. I recommend publishing the work after a minor review and necessary restructuring of the Discussion section.

 

1.     The authors describe HPV 16, 18, and 45 as the most common genotypes. Please provide information on whether this is the most common genotype in Northern Norway. Literature data from other countries indicate a completely different distribution of the most common genotypes. In Poland, for example, it is 16, 31, and 45.

2.     The Introduction section provides an interesting historical context for Norway – please also address the global situation in other countries. Please also remove the last sentences from the introduction, which refer to the conclusions of the publication – they should not be included in this section.

3.      Please clarify the inclusion and exclusion criteria – have all women not been vaccinated against HPV? Has this been verified? Furthermore, there is no information about whether any women had a history of HPV infection or precancerous cervical conditions.

4.      Please complete the captions and expand the abbreviations under Figures.

5.      What does ASC-US+ mean? Please expand on the acronym.

6.      Please correct the tables – the data is scattered and completely unreadable.

7.     It would be helpful to include an additional table with the percentage of individual HPV genotypes. What is the rate of multiple infections?

8.     The discussion needs to be completely reviewed. Although the results should be presented briefly, the bulleted results in this section are confusing and should be moved to the Results section. There are no references to the literature or results from other researchers – this must be presented. Please add more information about other molecular tests, including self-sampling or expression of E4 protein. Przybylski M, Pruski D, Millert-Kalińska S, Krzyżaniak M, de Mezer M, Frydrychowicz M, Jach R, Żurawski J. Expression of E4 Protein and HPV Major Capsid Protein (L1) as A Novel Combination in Squamous Intraepithelial Lesions. Biomedicines. 2023 Jan 16;11(1):225. doi: 10.3390/biomedicines11010225. PMID: 36672733; PMCID: PMC9855969.

Author Response

Reviewer 1

I congratulate the authors on their work – a presentation is sometimes 8-10 years old; the work is well-structured, written in correct English, and easily read by the reader. I recommend publishing the work after a minor review and necessary restructuring of the Discussion section.

Comments 1: The authors describe HPV 16, 18, and 45 as the most common genotypes. Please provide information on whether this is the most common genotype in Northern Norway. Literature data from other countries indicate a completely different distribution of the most common genotypes. In Poland, for example, it is 16, 31, and 45.

Response 1: Thank you for pointing this out. HPV types 16, 18 and 45 are not the three most common HPV types in screening populations, but they are the three most common HPV types in women with cervical cancer, both in Northern Norway and in Europe. We have now clarified this in the manuscript and added the following references: Arbyn 2014, Tjalma 2013 and Sørbye 2023.

Comments 2: The Introduction section provides an interesting historical context for Norway – please also address the global situation in other countries. Please also remove the last sentences from the introduction, which refer to the conclusions of the publication – they should not be included in this section.

Response 2: We agree with the reviewer. We have revised the Introduction to briefly describe the global situation in other countries and have removed the final sentences that anticipated the study’s conclusions.

Comments 3: Please clarify the inclusion and exclusion criteria – have all women not been vaccinated against HPV? Has this been verified? Furthermore, there is no information about whether any women had a history of HPV infection or precancerous cervical conditions.

Response 3: Thank you for this important comment. Women in this cohort attended the organized cervical cancer screening program in Northern Norway between 2013 and 2015. In Norway, HPV vaccination was introduced for 12-year-old girls in 2009, and a national catch-up program was not implemented until 2016–2019. Thus, women invited to screening in 2013–2015 were born before the vaccine-eligible birth cohorts and were not targeted by the school-based program. Individual HPV vaccination status is not recorded in the screening database used for this study, but opportunistic HPV vaccination outside the childhood program was very uncommon in this age group during the study period because the vaccine was expensive and not actively promoted. We therefore consider the study cohort to be virtually unvaccinated.

All women under 40 years of age with a cervical screening test in 2013–2015 in Northern Norway were eligible, irrespective of any previous HPV infection, cytological abnormalities, or histologically confirmed precancerous lesions. The only exclusions were technically invalid screening samples (HPV mRNA test or cytology); these women were advised to have a repeat test, and the valid repeat sample was then included in the study. We have now clarified the inclusion and exclusion criteria, and the absence of individual vaccination data, in the Methods section.

Comments 4: Please complete the captions and expand the abbreviations under Figures.

Response 4: We have revised all figure captions to provide complete descriptions and have expanded all abbreviations in the figure legends throughout the manuscript.

Comments 5: What does ASC-US+ mean? Please expand on the acronym.

Response 5: In the manuscript, ASC-US+ denotes all abnormal cytology according to the Bethesda classification (“ASC-US or more severe cytological abnormalities”), i.e., ASC-US, LSIL, ASC-H, HSIL, AIS and cytological cervical cancer. We have now clarified this at first mention in the text and added it to the list of abbreviations.

Comments 6: Please correct the tables – the data is scattered and completely unreadable.

Response 6: We agree with the reviewer. The poor layout resulted from technical formatting problems when converting our original Word tables into the MDPI Targets template. In the revised manuscript, we have rebuilt and reformatted all tables to ensure that the data are as clearly aligned and readable as possible. If any residual layout issues remain after conversion to the journal’s production system, we trust that the MDPI Targets production team will assist in finalizing the table formatting. We apologize for the inconvenience caused by the previous version.

Comments 7: It would be helpful to include an additional table with the percentage of individual HPV genotypes. What is the rate of multiple infections?

Response 7: Thank you for this helpful suggestion. We have now added a new table (Table 3) presenting HPV mRNA positivity by age group, including the percentage of each individual genotype (HPV16, HPV18 and HPV45) and the proportion of women with multiple infections. These results are described in a new subsection of the Results (Section 3.10).

Comments 8: The discussion needs to be completely reviewed. Although the results should be presented briefly, the bulleted results in this section are confusing and should be moved to the Results section. There are no references to the literature or results from other researchers – this must be presented. Please add more information about other molecular tests, including self-sampling or expression of E4 protein. Przybylski M, Pruski D, Millert-Kalińska S, Krzyżaniak M, de Mezer M, Frydrychowicz M, Jach R, Żurawski J. Expression of E4 Protein and HPV Major Capsid Protein (L1) as A Novel Combination in Squamous Intraepithelial Lesions. Biomedicines. 2023 Jan 16;11(1):225. doi: 10.3390/biomedicines11010225. PMID: 36672733; PMCID: PMC9855969.

Response 8: Thank you for these valuable comments. We have substantially revised the Discussion section. The previous bulleted list of results has been removed, and the Discussion has been restructured into continuous prose with a clearer focus on interpretation rather than repetition of detailed numbers. Key numerical findings are now primarily presented in the Results section (including new text and a new table), and in the Discussion we only briefly summarize the main outcomes. We have also incorporated multiple references to prior Norwegian and international studies using HPV mRNA testing and co-testing and now explicitly relate our findings to these data (Sections 4.2–4.6). In addition, we have added a new subsection (Section 4.7, “Other Molecular Approaches, Including Self-Sampling and E4/L1 Biomarkers”) where we summarize evidence on HPV DNA testing of self-collected samples and discuss tissue biomarkers, including combined E4 and L1 expression, citing Przybylski et al. (Biomedicines 2023;11:225).

 

Reviewer 2 Report

Comments and Suggestions for Authors

The study followed 11,395 women screened in northern Norway for 8–10 years. Among women under 40 years of age, combined testing of type 3 HPV mRNA and cytology increased the sensitivity of CIN2+ detection and optimized risk stratification compared to cytology alone. This study has significant clinical value; however, the following issues need to be addressed before publication.

 

1. What is the molecular mechanism underlying the higher CIN2+ risk in the HPV mRNA positive/normal cytology group compared to the HPV mRNA negative/ASC-US group in the combined test?
2. Is the difference in PPV among HPV16, 18, and 45 directly related to the carcinogenic potential of these viral types?
3. What is the missed diagnosis rate of CIN2+ caused by other high-risk HPV types not included in this study?
4. Is the diagnostic performance of this combined test consistent among women under 40 years of age who have received the HPV vaccine?
5. Are there more data supporting the long-term (>10 years) safety of the double-negative group with no cervical cancer cases?
6. Is there predictive data on the clinical impact of the 7.3 percentage point increase in sensitivity to cervical cancer mortality?
7. How does the cost-effectiveness of this combined test compare with that of expanded HPV DNA genotyping?
8. How should the optimal follow-up interval and monitoring method be determined for women with HPV mRNA-positive/normal cytology?
9. The authors need to rewrite the Discussion section.
10. The quality of the Figure is poor.

Author Response

Reviewer 2

The study followed 11,395 women screened in northern Norway for 8–10 years. Among women under 40 years of age, combined testing of type 3 HPV mRNA and cytology increased the sensitivity of CIN2+ detection and optimized risk stratification compared to cytology alone. This study has significant clinical value; however, the following issues need to be addressed before publication.

Comments 1: What is the molecular mechanism underlying the higher CIN2+ risk in the HPV mRNA positive/normal cytology group compared to the HPV mRNA negative/ASC-US group in the combined test?

Response 1: Thank you for this important point. HPV mRNA positivity in women with normal cytology indicates transcriptionally active expression of the E6/E7 oncogenes from HPV16/18/45, which drive cell-cycle deregulation and clonal expansion before overt morphological atypia becomes apparent on cytology. In contrast, ASC-US cytology in HPV mRNA–negative women frequently reflects reactive or borderline changes without an underlying transforming HPV infection (e.g., inflammation, metaplasia, atrophy) or infections with non-detected types, and thus carries a lower biological risk of progression. In other words, detection of E6/E7 mRNA is a more direct molecular marker of transforming HPV infection than equivocal cytological atypia alone, which explains why the HPV mRNA–positive/normal cytology group has a higher subsequent CIN2+ risk than the HPV mRNA–negative/ASC-US group. We have now added references supporting this mechanistic interpretation in the Discussion (Section 4.3).

Comments 2: Is the difference in PPV among HPV16, 18, and 45 directly related to the carcinogenic potential of these viral types?

Response 2: Thank you for this pertinent question. In our cohort, the higher PPV for CIN2+ in HPV16 mRNA–positive women compared with HPV18- and HPV45-positive women likely reflects underlying genotype-specific differences in carcinogenic potential and lesion detectability. HPV16 is consistently associated with the highest risk of CIN3+ and cervical cancer, whereas HPV18 and HPV45, although clearly carcinogenic, are more often linked to adenocarcinoma and may have a shorter or less easily detectable CIN2/3 phase in the squamous epithelium. As a result, HPV16-positive infections more frequently present with histologically confirmed CIN2+ during follow-up, yielding a higher PPV, while HPV18- and HPV45-positive infections may be underrepresented in CIN2+ endpoints relative to their contribution to invasive cancer.

Our findings are in line with previous genotype-specific mRNA triage data from HPV DNA–positive women, where HPV16 mRNA positivity was associated with the highest CIN2+ risk, followed by HPV18 and other high-risk types (Sørbye et al. 2025). We have now clarified this interpretation and cited the relevant literature in the Discussion (Section 4.4).

Comments 3: What is the missed diagnosis rate of CIN2+ caused by other high-risk HPV types not included in this study?

Response 3: Thank you for this question. In our co-testing cohort, 262 women developed CIN2+ during 8–10 years of follow-up. Of these, 131/262 (50.0%) were HPV mRNA–positive (HPV16/18/45) at baseline and 131/262 (50.0%) were HPV mRNA–negative, corresponding to a baseline sensitivity of 50.0% for CIN2+ over the entire follow-up period when using the 3-type HPV mRNA test alone (Table 1). This 50% should be interpreted as an upper bound for CIN2+ potentially caused by other HPV types not targeted by the assay, because we did not have extended genotyping and cannot distinguish between lesions caused by non-16/18/45 types already present at baseline and lesions arising from new infections acquired during the 8–10-year follow-up in this young, sexually active population.

For comparison, in the population-based study by Katki et al. using a 13-type HPV DNA test (Hybrid Capture 2), the baseline HPV DNA test was positive in 79.6% (1838/2310) of women who developed CIN2+ and in 69% (60/87) of women who developed cervical cancer over 5 years of follow-up [Katki et al., Lancet Oncol. 2011].

Importantly, our study was not designed to use the 3-type HPV mRNA assay as a stand-alone primary screening test. In the co-testing strategy, many CIN2+ lesions in mRNA-negative women were still detected through abnormal cytology, so they were not “missed” by the combined approach. The main purpose of the targeted mRNA test in this setting is to enrich for women at particularly high risk (especially HPV16-positive) rather than to capture all CIN2+ caused by any high-risk HPV type. We have clarified this interpretation and the limitation regarding non-16/18/45 types in the Discussion (Section 4.2/4.3).

Comments 4: Is the diagnostic performance of this combined test consistent among women under 40 years of age who have received the HPV vaccine?

Response 4: Thank you for this important question. Our study was conducted in a cohort of women who were essentially unvaccinated against HPV, and the diagnostic performance estimates for the combined 3-type HPV mRNA/cytology strategy therefore apply to pre-vaccination conditions. In cohorts with high uptake of the bivalent or quadrivalent vaccine, the prevalence of HPV16/18 infections and the proportion of CIN2+ attributable to these types will be markedly reduced. Consequently, although the type-specific risk associated with HPV16/18 mRNA positivity in an infected woman is likely to remain high, the overall sensitivity of a 3-type HPV16/18/45 mRNA assay for all-type CIN2+ in the screened population will decrease, and its incremental value as a triage tool will be smaller.

In settings with widespread use of the nonavalent vaccine (Gardasil 9), the prevalence of lesions caused by HPV16/18/31/33/45/52/58 is expected to become very low, and modelling studies from Norway indicate that both the number and intensity of screening rounds must be reduced to remain cost-effective in highly vaccinated cohorts [Portnoy et al. 2022; Pedersen et al. 2018]. This implies that the role of a focused 3-type mRNA assay in such populations will differ from that observed in our study and may primarily be relevant for quality assurance or targeted triage in selected subgroups. We have now clarified in the Discussion that our performance estimates cannot be directly extrapolated to fully vaccinated cohorts and that dedicated evaluations in vaccinated birth cohorts are needed.

Comments 5: Are there more data supporting the long-term (>10 years) safety of the double-negative group with no cervical cancer cases?

Response 5: Thank you for this important comment. In our cohort, no cervical cancers occurred among women with double-negative co-tests (HPV mRNA–negative and normal cytology) during 8–10 years of follow-up. Given that, in women under 40, most cervical cancers are caused by HPV16, 18 and 45, this absence of cancer is biologically plausible. Tjalma et al. reported that the average time from CIN3 to invasive cancer was approximately 15 years for HPV16, 9 years for HPV18 and only 1 year for HPV45, whereas the corresponding times for HPV31 and HPV33 were 23.5 and 20 years, and around 17 years for other types. Thus, an 8–10-year follow-up window is sufficient to capture the vast majority of cancers arising from HPV16/18/45-related CIN3 present at baseline, particularly in this relatively young population.

Our findings are also consistent with long-term data from other HPV-based screening cohorts. In the 12-year follow-up study by Rad et al., using a 5-type HPV mRNA assay (HPV16/18/31/33/45) in women aged 25–69 years, the incidence of cervical cancer among HPV mRNA–negative women was 2 per 100,000 woman-years, comparable to the incidence observed after a negative 13-type HPV DNA test in large European and U.S. studies. In the meta-analysis of four European trials by Ronco et al., the cumulative incidence of cervical cancer after a negative HPV DNA test was approximately 2 per 100,000 woman-years at 6.5 years of follow-up, and in the study by Katki et al. the incidence among HPV DNA–negative women was 3.8 per 100,000 woman-years. In the Netherlands, HPV DNA–based screening is offered to women at ages 30, 35, 40, 50 and 60, with screening intervals extended from 5 to 10 years for older women with previous negative HPV tests. Together, these external data support our observation that double-negative women have a very low long-term risk of cervical cancer and reinforce the notion that negative HPV-based tests (including targeted mRNA assays) provide strong reassurance over extended intervals. We have added this context and the relevant references to the Discussion (Sections 4.5 and 4.6).

Comments 6: Is there predictive data on the clinical impact of the 7.3 percentage point increase in sensitivity to cervical cancer mortality?

Response 6: Thank you for this important comment. We did not directly model cervical cancer mortality in this study, but the observed 7.3 percentage point increase in CIN2+ sensitivity with co-testing is clinically relevant because even small improvements in detection of high-grade precursors have been shown to translate into fewer interval cancers and, ultimately, reduced mortality in HPV-based screening programs. International data indicate that false-negative cytology is a key driver of preventable cancers: rescreening of archival “normal” smears in women who later develop cervical cancer typically reveals previously missed abnormalities (ASC-US+) in 30–40% of cases, and recent Swedish data show that the increased cancer incidence after normal cytology has persisted over time, underscoring limitations of cytology alone and the need for improved triage and quality assurance [Wang et al. 2024; Nanda et al. 2000].

Our findings are consistent with these observations and with prior Norwegian work on HPV mRNA–based quality assurance. In an independent UNN cohort, Strand et al. reported that among 61,635 women with normal cytology, 752 (1.2%) had a positive 3-type HPV mRNA test (HPV16/18/45); during follow-up, CIN2+ was diagnosed in 34% and cervical cancer in 0.7%, and four of five cancers had their index cytology reclassified from normal to ASC-US+ after HPV mRNA–triggered review [Strand et al. 2024]. This illustrates how targeted mRNA testing can uncover misclassified cytology and prevent progression to invasive disease. Moreover, large randomized trials and cohort studies of HPV-based primary screening have shown that increased sensitivity for CIN2+/CIN3+ leads to substantially lower rates of invasive cervical cancer after a negative HPV test compared with cytology-based screening [Ronco et al. 2014; Gage et al. 2014]. Taken together, these data support the view that the 7.3 percentage point gain in CIN2+ sensitivity observed with co-testing in our study is likely to have a meaningful impact on future cancer incidence and mortality, even though this was not quantified directly in our analysis.

 

Comments 7: How does the cost-effectiveness of this combined test compare with that of expanded HPV DNA genotyping?

Response 7: Thank you for raising this important issue. Our study did not include a formal cost-effectiveness analysis, and we therefore cannot directly compare the cost-effectiveness of 3-type HPV mRNA co-testing with expanded HPV DNA genotyping. Conceptually, several cost drivers differ between these approaches: the unit cost of the screening test itself (including laboratory and personnel costs), the overall test positivity rate, the number of follow-up tests, colposcopies and biopsies, treatment of CIN2+, and the downstream costs of missed cancers and cancer treatment.

Co-testing with a targeted 3-type HPV mRNA assay plus cytology will generally be more expensive per screening round than HPV DNA testing alone because of the additional cytology component, but it yields a lower overall positivity rate and a more selective identification of women at highest short-term risk (particularly HPV16-positive), which may translate into fewer follow-up visits and colposcopies per screen. In contrast, expanded HPV DNA genotyping typically has a higher positivity rate, especially in younger women, and many HPV-positive women will be followed more closely than cytology-negative, mRNA-negative women—even when DNA positivity reflects less oncogenic types—potentially increasing downstream costs and overmanagement.

One plausible future strategy, suggested by modelling work from Norway, is to use more intensive or combined approaches (e.g., cytology plus targeted mRNA) in the first screening rounds and then extend intervals or transition to HPV DNA testing with genotyping at older ages and in vaccinated cohorts, thereby reducing the total number of lifetime screens while maintaining safety. However, the relative cost-effectiveness of such strategies compared with expanded HPV DNA genotyping needs to be addressed in dedicated modelling studies; we now emphasize in the Discussion that our data provide clinical performance inputs but do not replace formal economic evaluations.

Comments 8: How should the optimal follow-up interval and monitoring method be determined for women with HPV mRNA-positive/normal cytology?

Response 8: In our study, women with HPV mRNA–positive/normal cytology were managed with repeat co-testing at 12-month intervals. If either cytology became abnormal (ASC-US+) or HPV mRNA positivity persisted, women were referred for colposcopy and biopsy. In cases with persistent HPV mRNA positivity and negative or non-diagnostic biopsies, a third round of biopsy or a diagnostic LEEP was considered to detect potential occult CIN2+.

Given the high PPV for CIN2+ among HPV mRNA–positive women, including those with normal cytology, an alternative—and arguably more efficient—approach would be to refer all women ≥25 years with HPV16/18/45 mRNA positivity directly to colposcopy at baseline, irrespective of cytology, reserving repeated co-testing for younger women (e.g., <25 years) or for selected low-risk constellations. Optimal follow-up intervals and monitoring strategies for HPV mRNA–positive/normal cytology should be defined in future guideline and modelling work, balancing cancer prevention against over-referral and overtreatment.

Comments 9: The authors need to rewrite the Discussion section.

Response 9: Thank you for this comment. We have completely revised and reorganized the Discussion section, removed the previous bullet-point structure, and moved descriptive numerical results back to the Results section. The revised Discussion now focuses on interpretation, places our findings in the context of previous Norwegian and international studies, and includes additional information on other molecular tests and self-sampling approaches, as requested by the reviewers.

Comments 10: The quality of the Figure is poor.

Response 10: We agree with the reviewer. Figures 2 and 3 have been completely redrawn at higher resolution with simplified layouts, clearer labelling, and consistent colour schemes to improve readability and print quality.

 

Reviewer 3 Report

Comments and Suggestions for Authors

Dear Authors,

Thank you for submitting this important manuscript evaluating long-term outcomes of co-testing with a 3-type HPV mRNA assay (16/18/45) and cytology in women under 40. The study addresses a clinically relevant question with a large, population-based cohort and extended follow-up, which strengthens its contribution to the literature. The findings—particularly the incremental sensitivity gain, the strong reassurance in double-negative women, and the identification of high-risk subgroups missed by cytology alone—are valuable for informing quality assurance and triage strategies in cervical cancer screening.

That said, several areas require improvement before the manuscript can reach publication standard:

Major comments: 
Methods: 
1. Lines 102-105: Provide additional information on how verification bias was addressed, given that not all women underwent biopsy.
2. Lines 121-128: The study design is appropriate, but the description of bias minimization and handling of missing data should be more detailed. 
3. Results: Results are comprehensive, but presentation requires improvement: 
3.1. Tables: Current formatting does not follow MDPI guidelines. Avoid merged cells, and ensure consistent headings.
3.2. Figures: Flow diagram (Figure 1) should follow STROBE-style conventions; risk diagrams (Figures 2–3) need standardized axes, consistent scales, and clearer legends. Ensure figures are self-explanatory without requiring extensive reference to the text.
4. Lines 309-316: Future research directions (e.g., cost-effectiveness, prospective validation in vaccinated cohorts, head-to-head comparisons with extended DNA genotyping) should be highlighted more prominently.
5. Lines 327-333: The conclusions are supported by the data, but claims about policy implications should be tempered. The cohort is largely unvaccinated and calendar-based, limiting generalizability to current vaccinated populations and primary HPV DNA screening frameworks.

Minor comments:
6. Reporting Guidelines: Please consider submitting the STROBE or STARD checklists with the revised manuscript.
6.1. As this is a retrospective cohort study, adherence to STROBE (Strengthening the Reporting of Observational Studies in Epidemiology) is recommended.
6.2. Given the diagnostic accuracy focus, inclusion of a STARD (Standards for Reporting Diagnostic Accuracy Studies) checklist would further enhance transparency and reproducibility.
7. The English is generally clear, but several sentences are overly complex. Streamlining phrasing and improving flow will enhance readability.

With these revisions, the paper will make a valuable contribution to the literature on cervical cancer screening in younger women.

Comments on the Quality of English Language

The manuscript is generally well written, with clear organization and appropriate scientific terminology, and the English language is understandable throughout; however, several areas would benefit from minor editing to improve fluency and readability, as some sentences are overly long and could be shortened for clarity, occasional grammatical inconsistencies such as verb tense shifts and article usage should be corrected, and technical terms should be used consistently across sections (e.g., “voice therapy program,” “follow-up period”); in addition, tables and figure legends are clear but would benefit from minor stylistic polishing to enhance precision, so overall the quality of English is adequate for publication after minor language editing.

Author Response

Reviewer 3

Thank you for submitting this important manuscript evaluating long-term outcomes of co-testing with a 3-type HPV mRNA assay (16/18/45) and cytology in women under 40. The study addresses a clinically relevant question with a large, population-based cohort and extended follow-up, which strengthens its contribution to the literature. The findings—particularly the incremental sensitivity gain, the strong reassurance in double-negative women, and the identification of high-risk subgroups missed by cytology alone—are valuable for informing quality assurance and triage strategies in cervical cancer screening.

Comments 1: Lines 102-105: Provide additional information on how verification bias was addressed, given that not all women underwent biopsy.

Response 1: Thank you for highlighting the issue of verification bias. In Norway, cervical screening is organized nationally with three-year intervals for women aged 25–69 years, and our department receives essentially all screening samples from Troms and Finnmark (≈5% of the Norwegian target population). For this study, all women with screening tests at UNN in 2013–2015 were passively followed for 8–10 years through our laboratory information system (SymPathy), which prospectively records all cervical cytology, HPV tests, cervical biopsies, LEEP specimens and hysterectomy samples using a unique personal identifier. In addition to the national call–recall system, our laboratory applies local follow-up routines with reminders for missing control tests; for women who moved to other regions, we retrieved subsequent histology and screening results via the national registry (KREMT). Non-attenders with persistent abnormalities could be offered self-sampling for HPV testing. As a result, most women with abnormal findings underwent guideline-based colposcopy and biopsy, and the majority of women either cleared infection or received treatment (LEEP) during follow-up. Nevertheless, because not all women had histology, we cannot fully exclude work-up (verification) bias; in our primary analyses, women without histology and without a CIN2+ diagnosis during 8–10 years of follow-up were classified as non-CIN2+, and we now clarify this approach and its limitations in Section 2.7.

Comments 2: Lines 121-128: The study design is appropriate, but the description of bias minimization and handling of missing data should be more detailed. 

Response 2: Thank you for this comment. We have substantially expanded the description of bias minimization and handling of missing data in Section 2.7. The revised text now explains that all women were followed for 8–10 years within the organized Norwegian screening program through the laboratory information system (SymPathy), which captures all cytology, HPV tests, biopsies, LEEP specimens and hysterectomy samples from Troms and Finnmark. We clarify that, in addition to the national call–recall system, local reminder routines and HPV self-sampling were used to reduce loss to follow-up. We explicitly state that women without histology and without a CIN2+ diagnosis during follow-up were classified as non-CIN2+, and that pre-specified sensitivity analyses restricted to women with ≥1 follow-up test and stratified by baseline cytology (NILM vs abnormal) were planned to assess potential verification bias. We also specify that records with invalid mRNA ISC or unsatisfactory cytology were excluded a priori and that missing co-variates (e.g., vaccination status, parity) were not included in primary models.

 

Comments 3.1: Tables: Current formatting does not follow MDPI guidelines. Avoid merged cells, and ensure consistent headings.

Response 3.1: Thank you for this comment. We have now reformatted all tables to comply with the MDPI Targets guidelines: merged cells have been removed, column structures have been harmonized across tables, and headings and footnotes have been standardized and aligned with the journal’s requirements. Any remaining minor layout issues will be addressed by the journal production team at the proof stage.

Comments 3.2: Figures: Flow diagram (Figure 1) should follow STROBE-style conventions; risk diagrams (Figures 2–3) need standardized axes, consistent scales, and clearer legends. Ensure figures are self-explanatory without requiring extensive reference to the text.

Response 3.2: Thank you for these helpful suggestions. Figures 2 and 3 have been redesigned with standardized axes, consistent scales, and simplified, clearer legends. Both risk diagrams now include explicit labels for all strata and risks, and the figure captions have been expanded so that each figure is self-explanatory without requiring extensive reference to the main text.

Comments 4: Lines 309-316: Future research directions (e.g., cost-effectiveness, prospective validation in vaccinated cohorts, head-to-head comparisons with extended DNA genotyping) should be highlighted more prominently.

Response 4: We agree with the reviewer. We have now expanded and rephrased the “Future Research” section to more prominently highlight key next steps, including cost-effectiveness analyses, prospective validation in vaccinated cohorts, and head-to-head comparisons with extended HPV DNA genotyping. We also briefly reiterate these priorities in the Conclusions section.

Comments 5: Lines 327-333: The conclusions are supported by the data, but claims about policy implications should be tempered. The cohort is largely unvaccinated and calendar-based, limiting generalizability to current vaccinated populations and primary HPV DNA screening frameworks.

Response 5: We agree with the reviewer that our policy statements should be more cautious given the calendar-based design and largely unvaccinated cohort. We have therefore tempered the Conclusions to emphasize the historical context and to limit generalizations to contemporary vaccinated, primary HPV DNA–based programs.

Comments 6.1: As this is a retrospective cohort study, adherence to STROBE (Strengthening the Reporting of Observational Studies in Epidemiology) is recommended.

Response 6.1: Thank you for this important remark. We have now reviewed the manuscript against the STROBE checklist for cohort studies and revised the text to improve transparency in study design, setting, participants, variables, bias minimization, and follow-up procedures (particularly in Sections 2.1–2.7 and 4.7). A completed STROBE checklist has been prepared and will be submitted as supplementary material.

Comments 6.2: Given the diagnostic accuracy focus, inclusion of a STARD (Standards for Reporting Diagnostic Accuracy Studies) checklist would further enhance transparency and reproducibility.

Response 6.2: We agree that adherence to STARD recommendations improves transparency in diagnostic accuracy research. We have now completed the STARD checklist for our study and will submit it as supplementary material together with the revised manuscript.

Comments 7: The English is generally clear, but several sentences are overly complex. Streamlining phrasing and improving flow will enhance readability.

Response 7: Thank you for this comment. We have carefully revised the manuscript for language, shortening long and complex sentences, simplifying phrasing, and improving paragraph flow throughout the Introduction, Methods, Results, Discussion, and Conclusions. The revised version has also been copy-edited using a professional language tool, and all changes have been reviewed and approved by the authors.

Comments 8: With these revisions, the paper will make a valuable contribution to the literature on cervical cancer screening in younger women.

Response 8: We appreciate the reviewer’s positive assessment and constructive feedback. We have implemented the suggested revisions to strengthen the methodological transparency, contextualization within the literature, and clarity of the Discussion and Conclusions. We hope that the revised manuscript will now be suitable for publication and contribute meaningfully to the evidence base on cervical cancer screening in younger women.

 

Round 2

Reviewer 2 Report

Comments and Suggestions for Authors

The author provided a thorough response, revised the manuscript, and recommended it for publication.

Author Response

Reviewer 2

Comments 1: The author provided a thorough response, revised the manuscript, and recommended it for publication.

Response 1: We thank Reviewer 2 for the positive evaluation and recommendation for publication. No further changes were deemed necessary in response to this comment.

Reviewer 3 Report

Comments and Suggestions for Authors

The revised manuscript has been substantially improved and now presents a clear, rigorous, and well‑structured study. The authors have addressed the major concerns raised in the previous review, particularly in clarifying methodology, strengthening data presentation, and situating the findings within the Norwegian and international screening context. The long‑term follow‑up and large cohort size add significant value to the evidence base for HPV mRNA co‑testing in younger women. Overall, the manuscript is scientifically robust, and  I appreciate your efforts and contribution. 

Minor comment:

1. Tables 1-3: A few captions and table layouts contain minor typographic inconsistencies (spacing, alignment). These should be standardized.

 

Author Response

Reviewer 3: 

The revised manuscript has been substantially improved and now presents a clear, rigorous, and well‑structured study. The authors have addressed the major concerns raised in the previous review, particularly in clarifying methodology, strengthening data presentation, and situating the findings within the Norwegian and international screening context. The long‑term follow‑up and large cohort size add significant value to the evidence base for HPV mRNA co‑testing in younger women. Overall, the manuscript is scientifically robust, and  I appreciate your efforts and contribution. 

Comments 1: Tables 1-3: A few captions and table layouts contain minor typographic inconsistencies (spacing, alignment). These should be standardized.

Response 1: We thank Reviewer 3 for the positive and encouraging assessment of our work. As suggested, we have carefully reviewed Tables 1–3 and standardized the captions and layouts, correcting minor typographic inconsistencies in spacing and alignment.