Advertisement

Long-term cervical precancer outcomes after a negative DNA- or RNA-based human papillomavirus test result

Open AccessPublished:May 31, 2021DOI:https://doi.org/10.1016/j.ajog.2021.05.038

      Background

      Cervical cancer, a preventable disease associated with the human papillomavirus, is responsible for significant morbidity and mortality globally. Primary human papillomavirus testing is more sensitive in detecting precancerous cervical lesions than cytologic screening and can be conducted using either DNA- or RNA-based assays. Screening programs must select the most appropriate assay from several available assays for their population. It is not yet known whether these assays perform equivalently in the long term, particularly among women with a negative human papillomavirus test result. This study aims to compare long-term safety after a negative human papillomavirus test result across both DNA- and RNA-based testing assays.

      Objective

      This study aimed to compare long-term high-grade cervical intraepithelial neoplasia (grade 2 or higher and grade 3 or higher) outcomes of 2 DNA-based assays (Digene Hybrid Capture 2 High-Risk HPV DNA Test and cobas 4800 HPV Test) and 1 messenger RNA-based assay (Aptima HPV Assay) using data from the Human Papillomavirus For Cervical Cancer Trial-DECADEl (FOCAL-DECADE) cohort, by first comparing the positive and negative rates between the assays and then investigating the cumulative incidence of cervical intraepithelial neoplasia grade 2 and higher and grade 3 or higher detection among participants in the FOCAL DECADE cohort over follow-up according to human papillomavirus testing assays.

      Study Design

      The FOCAL Trial was a randomized controlled trial that evaluated human papillomavirus testing for primary cervical cancer screening. The FOCAL-DECADE cohort subsequently followed FOCAL Trial participants passively through the British Columbia Cervix Screening Program Database for approximately 10 years after the FOCAL Trial study exit to examine the rates of cervical intraepithelial neoplasia grade 2 or higher and grade 3 or higher. For this study, eligible participants had baseline human papillomavirus–negative results from at least 1 assay and had 1 or more cytologic screens after baseline (9509 participants for DNA-based and 3473 participants for DNA- vs RNA-based assay comparisons). We constructed cumulative incidence curves and compared the hazard ratios for cervical intraepithelial neoplasia grade 2 or higher and grade 3 or higher detection according to the assays.

      Results

      Over 10 years of follow-up, the cumulative incidence of cervical intraepithelial neoplasia grade 2 or higher and grade 3 or higher did not significantly differ between the DNA-based assays (hazard ratio, 0.95; 95% confidence interval, 0.84–1.06; P=.35 and hazard ratio, 0.82; 95% confidence interval, 0.66–1.01; P=.06 for cervical intraepithelial neoplasia grade 2 or higher and cervical intraepithelial neoplasia grade 3 or higher, respectively) or between the DNA- and RNA-based assays (hazard ratio, 0.97; 95% confidence interval, 0.87–1.06; P=.48 and hazard ratio, 0.94; 95% confidence interval, 0.79–1.13; P=.52 for cervical intraepithelial neoplasia grade 2 or higher and cervical intraepithelial neoplasia grade 3 or higher, respectively).

      Conclusion

      Among participants who tested negative for human papillomavirus at baseline, the long-term risk of cervical intraepithelial neoplasia grade 2 or higher and grade 3 or higher did not significantly differ regardless of whether DNA- or RNA-based human papillomavirus testing assays were used. Screening program decision makers can be confident that for women who test negative for human papillomavirus, DNA- and RNA-based assays exhibit similar cervical intraepithelial neoplasia grade 2 or higher outcomes over several years.

      Key words

      Introduction

      Cervical cancer, an almost entirely preventable disease, is responsible for significant morbidity and mortality globally, particularly in low- and middle-income countries and among systemically disadvantaged populations in high-income countries.
      World Health Organization
      A Global Strategy for elimination of cervical cancer.
      Existing cytologic screening approaches have significantly decreased cervical cancer rates in well-screened populations in high-resource settings.
      World Health Organization
      A Global Strategy for elimination of cervical cancer.
      The World Health Organization has called for the elimination of cervical cancer by the end of this century.
      World Health Organization
      A Global Strategy for elimination of cervical cancer.
      As a persistent infection with a high-risk strain of the human papillomavirus (HPV) is necessary for the development of cervical cancer,
      • Walboomers J.M.M.
      • Jacobs M.V.
      • Manos M.M.
      • et al.
      Human papillomavirus is a necessary cause of invasive cervical cancer worldwide.
      ,
      • Bosch F.X.
      • Lorincz A.
      • Muñoz N.
      • Meijer C.J.L.M.
      • Shah K.V.
      The causal relation between human papillomavirus and cervical cancer.
      the use of screening methods that include HPV testing can help countries meet this goal.

       Why was this study conducted?

      This study aimed to compare long-term high-grade cervical intraepithelial neoplasia (grade 2 or higher [CIN2+] and grade 3 or higher [CIN3+]) risk after a baseline negative result by 2 DNA- and 1 RNA-based human papillomavirus (HPV) assays, using data from the Human Papillomavirus For Cervical Cancer Trial (HPV FOCAL)-DECADE cohort.

       Key findings

      Long-term risk of CIN2+ and CIN3+ did not significantly differ regardless of whether DNA- or RNA-based HPV testing assays were used at baseline.

       What does this add to what is known?

      This study extends previous findings by demonstrating that long-term CIN2+ outcomes remain similar over 10 years’ follow-up.
      Multiple randomized controlled trials have found that HPV testing is more sensitive in detecting precancerous lesions than cytologic screening, thus demonstrating the potential of HPV testing to replace cytologic screening in cervical screening programs.
      • Canfell K.
      • Caruana M.
      • Gebski V.
      • et al.
      Cervical screening with primary HPV testing or cytology in a population of women in which those aged 33 years or younger had previously been offered HPV vaccination: results of the Compass pilot randomised trial.
      • Ogilvie G.S.
      • Van Niekerk D.
      • Krajden M.
      • et al.
      Effect of screening with primary cervical HPV testing vs cytology testing on high-grade cervical intraepithelial neoplasia at 48 months: the HPV FOCAL randomized clinical trial.
      • Leinonen M.K.
      • Nieminen P.
      • Lönnberg S.
      • et al.
      Detection rates of precancerous and cancerous cervical lesions within one screening round of primary human papillomavirus DNA testing: prospective randomised trial in Finland.
      • Ronco G.
      • Giorgi-Rossi P.
      • Carozzi F.
      • et al.
      Efficacy of human papillomavirus testing for the detection of invasive cervical cancers and cervical intraepithelial neoplasia: a randomised controlled trial.
      • Mayrand M.H.
      • Duarte-Franco E.
      • Rodrigues I.
      • et al.
      Human papillomavirus DNA versus Papanicolaou screening tests for cervical cancer.
      However, HPV testing is less specific than cytologic evaluation and detects persistent infections, which may lead to cervical cancer, and transient infections, which will spontaneously clear without treatment.
      • Isidean S.D.
      • Mayrand M.H.
      • Ramanakumar A.V.
      • et al.
      Human papillomavirus testing versus cytology in primary cervical cancer screening: end-of-study and extended follow-up results from the Canadian cervical cancer screening trial.
      • Gage J.C.
      • Schiffman M.
      • Katki H.A.
      • et al.
      Reassurance Against future risk of precancer and cancer conferred by a negative human papillomavirus test.
      • Castle P.E.
      • Kinney W.K.
      • Xue X.
      • et al.
      Role of screening history in clinical meaning and optimal management of positive cervical screening results.
      This leads to potential overdiagnosis and overtreatment, which have many associated adverse effects.
      • Arbyn M.
      • Anttila A.
      • Jordan J.
      • et al.
      European Guidelines for Quality Assurance in Cervical Cancer Screening--summary document.
      Extended intervals between testing and more specific HPV assays (ie, those that better differentiate between progressive and transient lesions) have the potential to alleviate such adverse effects.
      Currently, both DNA- and RNA-based HPV assays are available. DNA-based HPV assays, such as Digene Hybrid Capture 2 High-Risk HPV DNA test (Qiagen; Germantown, MD) (HC2) and cobas 4800 HPV test (CB) (Roche; Indianapolis, IN),
      Roche. Cobas
      HPV Test Solut Publ.
      detect a structural gene of HPV (L1) that indicates the presence of HPV but does not distinguish between transient infections and infections at a higher risk of resulting in precancerous lesions or cervical cancer.
      • Cuschieri K.
      • Wentzensen N.
      Human papillomavirus mRNA and p16 detection as biomarkers for the improved diagnosis of cervical neoplasia.
      On the contrary, the RNA-based Aptima HPV Assay (AHPV) (Hologic; San Diego, CA)
      Hologic. APTIMA
      APTIMA HPV tests | Hologic. Published. 2019.
      detects transcripts from 2 viral oncogenes: E6 and E7. When expressed, their resulting oncoproteins disrupt normal cell regulation and tumor suppression; therefore, their detection might be a better indicator of a higher risk of progression to cervical cancer.
      • Cuschieri K.
      • Wentzensen N.
      Human papillomavirus mRNA and p16 detection as biomarkers for the improved diagnosis of cervical neoplasia.
      The HPV For Cervical Cancer Screening (HPV FOCAL) trial was conducted in British Columbia, Canada, from 2008 to 2016 with the primary goal of comparing the efficacy of HPV-based assays with cytologic screening for cervical cancer detection.
      • Ogilvie G.S.
      • Van Niekerk D.
      • Krajden M.
      • et al.
      Effect of screening with primary cervical HPV testing vs cytology testing on high-grade cervical intraepithelial neoplasia at 48 months: the HPV FOCAL randomized clinical trial.
      ,
      • Ogilvie G.S.
      • Van Niekerk D.J.
      • Krajden M.
      • et al.
      Primary cervical cancer screening with HPV testing compared with liquid-based cytology: results of round 1 of a randomised controlled trial-the HPV FOCAL Study.
      • Ogilvie G.S.
      • van Niekerk D.J.
      • Krajden M.
      • et al.
      A randomized controlled trial of human papillomavirus (HPV) testing for cervical cancer screening: trial design and preliminary results (HPV FOCAL Trial).
      • Ogilvie G.S.
      • Krajden M.
      • van Niekerk D.
      • et al.
      HPV for cervical cancer screening (HPV FOCAL): complete Round 1 results of a randomized trial comparing HPV-based primary screening to liquid-based cytology for cervical cancer.
      The HPV FOCAL-DECADE cohort (FOCAL-DECADE) followed FOCAL participants for approximately 10 years after study exit to assess the long-term risk of future precancerous lesions after 1 or 2 rounds of HPV testing.
      • Gottschlich A.
      • van Niekerk D.
      • Smith L.W.
      • et al.
      Assessing 10-year safety of a single negative HPV test for cervical cancer screening: evidence from FOCAL-DECADE cohort.
      The FOCAL trial used 3 HPV assays for screening: 2 DNA-based assays (HC2 and CB),
      Roche. Cobas
      HPV Test Solut Publ.
      and 1 RNA-based assay (AHPV);
      Hologic. APTIMA
      APTIMA HPV tests | Hologic. Published. 2019.
      however, the participants were managed based on only HC2 results before exit testing if either a positive HC2 result or the detection of HPV16, 18, or 45 by either CB or AHPV referred to colposcopy. In this study, the long-term risk of detecting precancerous lesions after 1 negative baseline HPV test result using each of the 3 specific HPV assays was assessed to confirm that a negative HPV test result from any of these assays will predict a similar future risk of cervical intraepithelial neoplasia grade 2 or higher and grade 3 or higher (CIN2+ and CIN3+, respectively) detection. We compared the relative performance of the 2 DNA-based assays and of the DNA-based assays vs the RNA-based assay by first comparing the positive and negative rates between the assays and then investigating the cumulative incidence of CIN2+ and CIN3+ detection over follow-up among baseline HPV-negative participants in the FOCAL-DECADE cohort according to HPV testing assays.

      Materials and Methods

       Study design and timing

      The FOCAL trial has been previously described in detail.
      • Ogilvie G.S.
      • Van Niekerk D.
      • Krajden M.
      • et al.
      Effect of screening with primary cervical HPV testing vs cytology testing on high-grade cervical intraepithelial neoplasia at 48 months: the HPV FOCAL randomized clinical trial.
      ,
      • Ogilvie G.S.
      • Van Niekerk D.J.
      • Krajden M.
      • et al.
      Primary cervical cancer screening with HPV testing compared with liquid-based cytology: results of round 1 of a randomised controlled trial-the HPV FOCAL Study.
      • Ogilvie G.S.
      • van Niekerk D.J.
      • Krajden M.
      • et al.
      A randomized controlled trial of human papillomavirus (HPV) testing for cervical cancer screening: trial design and preliminary results (HPV FOCAL Trial).
      • Ogilvie G.S.
      • Krajden M.
      • van Niekerk D.
      • et al.
      HPV for cervical cancer screening (HPV FOCAL): complete Round 1 results of a randomized trial comparing HPV-based primary screening to liquid-based cytology for cervical cancer.
      Briefly, the trial (N=25,223) was conducted from 2008 to 2016, with participants enrolled for a maximum of 4 years, and consisted of 3 arms: the control arm (n=9457), which received liquid-based cytology (LBC) at baseline at 24 months and co-testing (LBC and HPV testing) at 48 months; the intervention arm (n=9552), which received HPV testing at baseline and co-testing at 48 months; and the safety arm (n=6214), which received HPV testing at baseline and LBC at 24 months. HPV samples from the participants randomized to the intervention arm before December 2010 were tested with 2 DNA-based assays (HC2 and CB), whereas those randomized in the beginning of December 2010, when Hologic joined the trial, had their samples additionally tested with AHPV, as well as HC2 and CB. At baseline, the participants were managed based on the results of the HC2 assay alone, whereas, at exit, they were managed based on a combination of their exit results (ie, those who were HC2-positive or HC2-negative with HPV 16, HPV 18, or HPV 45 detection on either CB or AHPV were referred to colposcopy, whereas those who were negative were returned to provincial screening). The CB and AHPV assays that were conducted before the trial exit, that is, at baseline, were performed blinded, and the results were unblinded at the conclusion of the trial. After the trial, the participants returned to the standard-of-care conventional cytologic screening through the provincial screening program (British Columbia Cancer’s Cervix Screening Program [BC CSP]).
      The BC Cancer Agency
      BC cancer cervix Screening 2018 program results.
      Because HPV testing is not standard of care in British Columbia, there was no specific guidance on management made available to clinicians; however, all providers were offered attendance at a results webinar/presentation and access to the JAMA 2018 paper.
      • Ogilvie G.S.
      • Van Niekerk D.J.
      • Krajden M.
      • et al.
      Primary cervical cancer screening with HPV testing compared with liquid-based cytology: results of round 1 of a randomised controlled trial-the HPV FOCAL Study.
      The FOCAL-DECADE cohort, also previously described,
      • Gottschlich A.
      • van Niekerk D.
      • Smith L.W.
      • et al.
      Assessing 10-year safety of a single negative HPV test for cervical cancer screening: evidence from FOCAL-DECADE cohort.
      was created by linking the participants’ data from the FOCAL trial to their corresponding data from BC CSP for approximately 10 years after the trial exit. The BC CSP is a province-wide, publicly funded program that maintains standardized screening guidelines and includes a registry of results of all screens, colposcopies, and associated histologic examinations provided in British Columbia.
      The BC Cancer Agency
      BC cancer cervix Screening 2018 program results.
      ,
      The BC Agency
      BC cancer cervix Screening 2018 program results.
      From the screening program data, we were able to identify all detections of CIN2+ and CIN3+ lesions that occurred in the FOCAL-DECADE cohort over the follow-up period. At the time of this analysis, the BC CSP recommended routine cervical screening every 3 years, and the data were available through mid-2019.
      This study aimed to investigate the relative performance of 3 HPV testing assays by comparing the long-term risks of detecting precancerous lesions after a negative HPV test result. The analysis was carried out using: (1) one of the 2 DNA-based assays (HC2 or CB) and (2) the 2 DNA-based assays (HC2 or CB) vs an RNA-based assay (AHPV) assays by investigating the positive and negative testing rates and the 10-year risk of CIN2+ and CIN3+ among participants in the FOCAL-DECADE cohort with a baseline negative HPV test result using 1 of the 3 assays.

       Variable creation

      To be eligible for this study, the participants were required to have undergone baseline HPV testing in the FOCAL trial and have at least 1 subsequent cervical screen after baseline. The study was restricted to the intervention arm because only those participants had their samples tested with all 3 HPV assays. Data from eligible participants (N=9509) were used to calculate positive and negative testing rates, and those who were baseline HPV negative (HC2, N=8321; CB, N=8274) were selected to compare long-term outcomes after a negative HPV test result using HC2 vs CB. Data from a subset of these participants (N=3473) who had their sample additionally tested with AHPV were used to compare positive and negative test rates between the DNA- and RNA-based assays, and data from those who were baseline HPV negative (HC2, N=2996; CB, N=2973; AHPV, 3038) were used to compare HC2 and CB with AHPV. The cases were ascertained either through a referral to colposcopy after a positive test result at trial exit or through routine screening after reentering the provincial screening program.
      The main exposure variable of interest was defined as a negative HPV test result from 1 of the 3 assays at baseline, and the outcome variables of interest were defined as the detection of CIN2+ and CIN3+ from trial baseline to 120 months post exit.

       Statistical analyses

      The overall agreement and Cohen’s kappa were calculated for HC2 and CB for DNA-based assay comparison and for HC2, CB, and AHPV for DNA- vs RNA-based assay comparison. The cumulative incidence of CIN2+ and CIN3+ detection among the subset of baseline HPV-negative participants was estimated according to the HPV assays. The maximum follow-up time was 120 months, with time to event calculated as the difference between the dates of CIN2+ or CIN3+ detection and the FOCAL trial baseline screening. Time to censor was calculated as the difference between the date of the most recent cytologic screening result and the baseline date for participants without detected CIN2+ or CIN3+. The Cox proportional hazard models were run and adjusted for repeated measures, and hazard ratios (HRs) were calculated to compare the cumulative incidence rates of each assay.

       Ethical approval

      Written informed consent was obtained from all participants, and ethical approval was obtained from the University of British Columbia Clinical Research Ethics Board (FOCAL: H06-04032, FOCAL-DECADE: H18-02063).

      Results

       DNA-based assay comparison

      Of the 9552 participants enrolled in the HPV FOCAL intervention arm, 9509 were included in this comparison. The participants were aged 25–65 years and had valid test results for HC2 and CB (5 participants were excluded owing to the failure to meet the FOCAL eligibility criteria,
      • Ogilvie G.S.
      • Van Niekerk D.
      • Krajden M.
      • et al.
      Effect of screening with primary cervical HPV testing vs cytology testing on high-grade cervical intraepithelial neoplasia at 48 months: the HPV FOCAL randomized clinical trial.
      and 38 participants were excluded owing to missing HC2 and/or CB results). The overall agreement between HC2 and CB was 96.3%, and the Cohen’s kappa was 0.76, demonstrating substantial agreement between the assays (Table 1).
      • McHugh M.L.
      Interrater reliability: the kappa statistic.
      The positive rate was 8.0% (95% confidence interval [CI], 7.5–8.6) and 8.6% (95% CI, 8.0–9.1) for HC2 and CB, respectively, indicating no statistical difference between the assays (P=.16).
      Table 1Assay comparison results at baseline, percent agreement, and Cohen’s kappa
      % agreementCohen’s kappa
      CB
      NegativePositive96.300.76
      HC2Negative8545202
      Positive149613
      HC2
      NegativePositive96.500.76
      AHPVNegative314282
      Positive40209
      CB
      NegativePositive95.500.7
      AHPVNegative3111113
      Positive44205
      AHPV, Aptima HPV Assay; CB, Roche cobas 4800 HPV test; HC2, Digene Hybrid Capture 2 High-Risk HPV DNA test.
      Strang et al. Comparison of DNA- and RNA-based human papillomavirus assays. Am J Obstet Gynecol 2021.
      The cumulative incidence of CIN2+ and CIN3+ over 120 months of follow-up among participants who were HPV negative at baseline according to the testing assays is depicted in Figure 1. As there were no tests conducted in the first 48 months after baseline testing for those who were HPV negative, the cumulative incidence remains zero during this period. The median follow-up time was 6.6 years. No significant difference in CIN2+ or CIN3+ detection was observed for CB compared with HC2 (Cox proportional HR, 0.95; 95% CI, 0.84–1.06; P=.35 and HR, 0.82; 95% CI, 0.66–1.01; P=.06 for CIN2+ and CIN3+, respectively) (Table 2).
      Figure thumbnail gr1
      Figure 1DNA-based assay comparison
      Cumulative incidence of CIN2+ and CIN3+ among HPV-negative participants at baseline by HPV testing assays.
      CB, Roche cobas 4800 HPV test; CIN2+ cervical intraepithelial neoplasia grade 2 or higher, CIN3+, cervical intraepithelial neoplasia grade 3 or higher; HC2, Digene Hybrid Capture 2 High-Risk HPV DNA test; HPV, human papillomavirus.
      Strang et al. Comparison of DNA- and RNA-based human papillomavirus assays. Am J Obstet Gynecol 2021.
      Table 2Cox proportional hazards comparison of cumulative incidence of CIN2+ and CIN3+ detection
      HC2 and CBHC2, CB, and AHPV
      CIN2+0.95 (95% CI, 0.84–1.06), P=.350.97 (95% CI, 0.87–1.06), P=.48
      CIN3+0.82 (95% CI, 0.66–1.01), P=.060.94 (95% CI, 0.79–1.13), P=.52
      AHPV, Aptima HPV Assay; CB, Roche cobas 4800 HPV test; CI, confidence interval; CIN2+, cervical intraepithelial neoplasia grade 2 or higher; CIN3+, cervical intraepithelial neoplasia grade 3 or higher; HC2, Digene Hybrid Capture 2 High-Risk HPV DNA test.
      Strang et al. Comparison of DNA- and RNA-based human papillomavirus assays. Am J Obstet Gynecol 2021.

       DNA- vs RNA-based assay comparison

      Of the 9509 participants included in the previous comparison, 3473 were included in this comparison. The exclusion criteria were as described above, with the additional exclusion of participants who did not undergo the AHPV testing assay (N=6036).
      Among this population, the overall agreement between HC2 and AHPV at baseline was 96.5%, and the Cohen’s kappa was 0.76, demonstrating substantial agreement (Table 1).
      • McHugh M.L.
      Interrater reliability: the kappa statistic.
      The overall agreement between CB and AHPV at baseline was 95.5% (95% CI), and the Cohen’s kappa was 0.70, demonstrating substantial agreement (Table 1).
      • McHugh M.L.
      Interrater reliability: the kappa statistic.
      The percent positivity rates for HC2, CB, and AHPV were 8.4% (95% CI, 7.5–9.3), 9.2% (95% CI, 8.2–10.1), and 7.2% (95% CI, 6.3–8.0), respectively (Table 3). There was no significant difference in the percentage of positive AHPV results at baseline compared with HC2 (P=.06); however, there were significantly fewer positive AHPV results at baseline compared with CB (P=.002).
      Table 3Positivity rates of results of DNA-based assays compared with an RNA-based assay
      N% positive95% CIP value
      HC22918.4[7.5–9.3].06
      CB3189.2[8.2–10.1].002
      AHPV2497.0[6.3–8.0]Referent
      AHPV, Aptima HPV Assay; CB, Roche cobas 4800 HPV test; CI, confidence interval; HC2, Digene Hybrid Capture 2 High-Risk HPV DNA test.
      Strang et al. Comparison of DNA- and RNA-based human papillomavirus assays. Am J Obstet Gynecol 2021.
      The cumulative incidence of CIN2+ and CIN3+ detection over 120 months of follow-up among participants who were HPV negative at baseline stratified by the testing assays is depicted in Figure 2. No testing was conducted in the first 48 months after baseline testing for those who were HPV negative; therefore, the cumulative incidence was zero. The median follow-up time was 6.2 years. No significant difference in detection between those tested with CB and AHPV compared with HC2 was identified (Cox proportional HR, 0.97; 95% CI, 0.87–1.06; P=.48 and HR, 0.94; 95% CI, 0.79–1.13; P=.52 for CIN2+ and CIN3+, respectively) (Table 2).
      Figure thumbnail gr2
      Figure 2DNA- vs RNA-based assay comparison
      Cumulative incidence of CIN2+ and CIN3+ among HPV-negative participants at baseline by HPV testing assay assays.
      AHPV, Aptima HPV Assay; CB, Roche cobas 4800 HPV test; CIN2+ cervical intraepithelial neoplasia grade 2 or higher, CIN3+, cervical intraepithelial neoplasia grade 3 or higher; HC2, Digene Hybrid Capture 2 High-Risk HPV DNA test; HPV, human papillomavirus.
      Strang et al. Comparison of DNA- and RNA-based human papillomavirus assays. Am J Obstet Gynecol 2021.

      Discussion

       Principal findings

      Our study found that a negative baseline HPV test result obtained by any 1 of the 3 assays used in the HPV FOCAL trial (HC2, CB, or AHPV) resulted in statistically similar CIN2+ and CIN3+ detection over 10 years’ follow-up among this population.

       Results

      These findings are generally consistent with existing comparisons that have shown similar HC2 vs CB sensitivity
      • Rao A.
      • Young S.
      • Erlich H.
      • et al.
      Development and characterization of the cobas human papillomavirus test.
      ,
      • Cook D.A.
      • Mei W.
      • Smith L.W.
      • et al.
      Comparison of the Roche cobas® 4800 and Digene Hybrid Capture® 2 HPV tests for primary cervical cancer screening in the HPV FOCAL trial.
      ; however, some studies have shown slightly higher specificity for CB.
      • Phillips S.
      • Garland S.M.
      • Tan J.H.
      • Quinn M.A.
      • Tabrizi S.N.
      Comparison of the Roche Cobas® 4800 HPV assay to Digene Hybrid Capture 2, Roche Linear Array and Roche Amplicor for Detection of High-Risk Human Papillomavirus Genotypes in Women undergoing treatment for cervical dysplasia.
      ,
      • Lloveras B.
      • Gomez S.
      • Alameda F.
      • et al.
      HPV Testing by cobas HPV Test in a Population from Catalonia.
      This study, one of the first of its kind with up to 10 years’ follow-up after initial HPV testing with multiple assays, extends the previous findings to demonstrate that performance remains similar over long-term follow-up.

       Clinical implications

      Literature comparing DNA-based HPV assays with RNA-based HPV assays is limited. We found that the percent positive at baseline did not significantly differ. Statistically significantly fewer baseline AHPV positive test results were obtained, but similar incidence of CIN2+ and CIN3+ detection was observed over long-term follow-up compared with CB, which indicates that AHPV may be slightly better at distinguishing between high-risk and transient cervical lesions during the first round of HPV testing. This is generally consistent with the limited existing research that compares HC2 and CB with AHPV over short-term follow-up. Previous comparisons of AHPV with HC2 and CB during testing or over short-term follow-up have reported similar sensitivity and higher specificity for AHPV.
      • Ratnam S.
      • Coutlee F.
      • Fontaine D.
      • et al.
      APTIMA HPV E6/E7 mRNA test is as sensitive as hybrid capture 2 assay but more specific at detecting cervical precancer and cancer.
      • Iftner T.
      • Becker S.
      • Neis K.J.
      • et al.
      Head-to-head comparison of the RNA-based APTIMA human papillomavirus (HPV) assay and the DNA-based hybrid capture 2 HPV test in a routine screening population of women aged 30 to 60 years in Germany.
      • Castle P.E.
      • Eaton B.
      • Reid J.
      • Getman D.
      • Dockter J.
      Comparison of human papillomavirus detection by APTIMA HPV and cobas HPV tests in a population of women referred for colposcopy following detection of atypical squamous cells of undetermined significance by Pap cytology.
      • Coldman A.J.
      • Phillips N.
      • van Niekerk D.
      • et al.
      Projected impact of HPV and LBC primary testing on rates of referral for colposcopy in a Canadian Cervical Cancer Screening Program.
      • Cook D.A.
      • Smith L.W.
      • Law J.
      • et al.
      APTIMA HPV Assay versus Hybrid Capture® 2 HPV test for primary cervical cancer screening in the HPV FOCAL trial.
      Another study found that AHPV detected significantly fewer transient HPV infections compared with CB.
      • Chorny J.A.
      • Frye T.C.
      • Fisher B.L.
      • Remmers C.L.
      Human papillomavirus detection with genotyping by the cobas and APTIMA assays: significant differences in HPV 16 detection?.
      However, earlier studies have found that this benefit disappears during subsequent rounds of HPV testing.
      • Cook D.A.
      • Smith L.W.
      • Law J.H.
      • et al.
      Comparative performance of human papillomavirus messenger RNA versus DNA screening tests at baseline and 48 months in the HPV FOCAL trial.
      This study extends such previous findings to demonstrate that CIN2+ and CIN3+ incidence remains similar between DNA- and RNA-based HPV testing assays over up to 10 years’ follow-up. The cumulative rates of CIN2+ incidence found during the FOCAL-DECADE follow-up were similar to those observed in other studies.
      • Isidean S.D.
      • Mayrand M.H.
      • Ramanakumar A.V.
      • et al.
      Human papillomavirus testing versus cytology in primary cervical cancer screening: end-of-study and extended follow-up results from the Canadian cervical cancer screening trial.
      ,
      • Gage J.C.
      • Schiffman M.
      • Katki H.A.
      • et al.
      Reassurance Against future risk of precancer and cancer conferred by a negative human papillomavirus test.
      ,
      • Dillner J.
      • Rebolj M.
      • Birembaut P.
      • et al.
      Long term predictive values of cytology and human papillomavirus testing in cervical cancer screening: joint European cohort study.
      We reported a 5-year cumulative risk of 0.35% and 0.20% for CIN2+ and CIN3+, respectively, after a negative HPV test result in the FOCAL-DECADE cohort, whereas the Kaiser Permanente Northern California cohort
      • Gage J.C.
      • Schiffman M.
      • Katki H.A.
      • et al.
      Reassurance Against future risk of precancer and cancer conferred by a negative human papillomavirus test.
      reported a risk of 0.40% and 0.14% for CIN2+ and CIN3+, respectively. The Canadian Cervical Cancer Screening Trial (CCCaST)
      • Isidean S.D.
      • Mayrand M.H.
      • Ramanakumar A.V.
      • et al.
      Human papillomavirus testing versus cytology in primary cervical cancer screening: end-of-study and extended follow-up results from the Canadian cervical cancer screening trial.
      reported 0.04% CIN2+ risk, and the joint European cohort study
      • Dillner J.
      • Rebolj M.
      • Birembaut P.
      • et al.
      Long term predictive values of cytology and human papillomavirus testing in cervical cancer screening: joint European cohort study.
      reported a 0.20% CIN3+ risk. Although the CCCaST 5-year CIN2+ risk is much lower than that found in other studies (potentially owing to differing age-inclusion criteria), the 10-year risk (1.3%) is very similar to that found in this analysis (1.2%).

       Strengths and limitations

      This study has several strengths, including the observation of participants over a long period of follow-up. In addition, using 3 HPV assays in the same population throughout a randomized controlled trial allows for more direct comparisons to be drawn. However, the results of this study should be interpreted considering its limitations. Cervical cancer has low incidence in British Columbia,
      • Ogilvie G.S.
      • Van Niekerk D.
      • Krajden M.
      • et al.
      Effect of screening with primary cervical HPV testing vs cytology testing on high-grade cervical intraepithelial neoplasia at 48 months: the HPV FOCAL randomized clinical trial.
      and there were few cases of CIN2+ or CIN3+ detected over follow-up even with a large study sample size. This makes differences between the assays difficult to ascertain owing to wide CIs. Another limitation is the exclusion of participants for whom we had no information regarding follow-up screening. It is possible that the participants excluded for this reason were different than participants with completed follow-up screening. For example, if women who were lost to follow-up were at a higher risk than those who had follow-up screening, we may have underestimated the rate of precancerous lesion detection in the population. However, it is unlikely that the assays performed differently among those who were lost to follow-up. Furthermore, despite participants being selected from a population-based screening program, there is possible selection bias. The FOCAL participants were primarily highly educated and were from 2 urban regions of British Columbia and were possibly more likely to be well screened before enrollment in the FOCAL trial.
      • Ogilvie G.S.
      • Van Niekerk D.
      • Krajden M.
      • et al.
      Effect of screening with primary cervical HPV testing vs cytology testing on high-grade cervical intraepithelial neoplasia at 48 months: the HPV FOCAL randomized clinical trial.
      There was limited representation of remote and other populations that may exhibit poorer uptake of screening and are, therefore, at a higher risk of developing cervical cancer.

       Research implications and conclusions

      Future investigations into the performance of DNA- and RNA-based HPV testing assays should include the comparison of CIN2+ and CIN3+ detection over long-term follow-up according to specific HPV strains identified using CB; AHPV; and different triage strategies, including cytologic evaluation, partial genotype testing, and combination approaches.
      In conclusion, the long-term risk of CIN2+ and CIN3+ among participants who were HPV negative at baseline appears to be similar and quite low using either HC2, CB, or AHPV. Similar performance of these DNA- and RNA-based HPV testing assays indicates that the safety of defining screening intervals for individuals who test negative in HPV-based cervical cancer screening programs is also similar among these assay types. This is an important finding because long-term safety provided by a negative result from a given assay is considered along with other criteria such as the ability of an assay to genotype and the cost of the assay when decision makers compare the advantages and disadvantages of DNA- vs RNA-based assays during the development of HPV-based cervical cancer screening programs. As HPV-based screening programs increasingly rely on triage by genotyping, which HC2 does not offer, it is important to recognize the overall equivalent performance of other HPV testing assays.

      References

        • World Health Organization
        A Global Strategy for elimination of cervical cancer.
        (Available at:)
        • Walboomers J.M.M.
        • Jacobs M.V.
        • Manos M.M.
        • et al.
        Human papillomavirus is a necessary cause of invasive cervical cancer worldwide.
        J Pathol. 1999; 189: 12-19
        • Bosch F.X.
        • Lorincz A.
        • Muñoz N.
        • Meijer C.J.L.M.
        • Shah K.V.
        The causal relation between human papillomavirus and cervical cancer.
        J Clin Pathol. 2002; 55: 244-265
        • Canfell K.
        • Caruana M.
        • Gebski V.
        • et al.
        Cervical screening with primary HPV testing or cytology in a population of women in which those aged 33 years or younger had previously been offered HPV vaccination: results of the Compass pilot randomised trial.
        PLoS Med. 2017; 14e1002388
        • Ogilvie G.S.
        • Van Niekerk D.
        • Krajden M.
        • et al.
        Effect of screening with primary cervical HPV testing vs cytology testing on high-grade cervical intraepithelial neoplasia at 48 months: the HPV FOCAL randomized clinical trial.
        JAMA. 2018; 320: 43-52
        • Leinonen M.K.
        • Nieminen P.
        • Lönnberg S.
        • et al.
        Detection rates of precancerous and cancerous cervical lesions within one screening round of primary human papillomavirus DNA testing: prospective randomised trial in Finland.
        BMJ. 2012; 345: e7789
        • Ronco G.
        • Giorgi-Rossi P.
        • Carozzi F.
        • et al.
        Efficacy of human papillomavirus testing for the detection of invasive cervical cancers and cervical intraepithelial neoplasia: a randomised controlled trial.
        Lancet Oncol. 2010; 11: 249-257
        • Mayrand M.H.
        • Duarte-Franco E.
        • Rodrigues I.
        • et al.
        Human papillomavirus DNA versus Papanicolaou screening tests for cervical cancer.
        N Engl J Med. 2007; 357: 1579-1588
        • Castle P.E.
        • Kinney W.K.
        • Xue X.
        • et al.
        Role of screening history in clinical meaning and optimal management of positive cervical screening results.
        J Natl Cancer Inst. 2019; 111: 820-827
        • Isidean S.D.
        • Mayrand M.H.
        • Ramanakumar A.V.
        • et al.
        Human papillomavirus testing versus cytology in primary cervical cancer screening: end-of-study and extended follow-up results from the Canadian cervical cancer screening trial.
        Int J Cancer. 2016; 139: 2456-2466
        • Gage J.C.
        • Schiffman M.
        • Katki H.A.
        • et al.
        Reassurance Against future risk of precancer and cancer conferred by a negative human papillomavirus test.
        J Natl Cancer Inst. 2014; 106: dju153
        • Arbyn M.
        • Anttila A.
        • Jordan J.
        • et al.
        European Guidelines for Quality Assurance in Cervical Cancer Screening--summary document.
        Ann Oncol. 2010; 21: 448-458
        • Qiagen. Digene
        Digene HC2 HPV DNA test.
        (Available at:)
        • Roche. Cobas
        HPV Test Solut Publ.
        (Available at:)
        • Cuschieri K.
        • Wentzensen N.
        Human papillomavirus mRNA and p16 detection as biomarkers for the improved diagnosis of cervical neoplasia.
        Cancer Epidemiol Biomarkers Prev. 2008; 17: 2536-2545
        • Hologic. APTIMA
        APTIMA HPV tests | Hologic. Published. 2019.
        (Available at:)
        • Ogilvie G.S.
        • Van Niekerk D.J.
        • Krajden M.
        • et al.
        Primary cervical cancer screening with HPV testing compared with liquid-based cytology: results of round 1 of a randomised controlled trial-the HPV FOCAL Study.
        JAMA. 2012; 107: 1917-1924
        • Ogilvie G.S.
        • van Niekerk D.J.
        • Krajden M.
        • et al.
        A randomized controlled trial of human papillomavirus (HPV) testing for cervical cancer screening: trial design and preliminary results (HPV FOCAL Trial).
        BMC Cancer. 2010; 10: 111
        • Ogilvie G.S.
        • Krajden M.
        • van Niekerk D.
        • et al.
        HPV for cervical cancer screening (HPV FOCAL): complete Round 1 results of a randomized trial comparing HPV-based primary screening to liquid-based cytology for cervical cancer.
        Int J Cancer. 2017; 140: 440-448
        • Gottschlich A.
        • van Niekerk D.
        • Smith L.W.
        • et al.
        Assessing 10-year safety of a single negative HPV test for cervical cancer screening: evidence from FOCAL-DECADE cohort.
        Cancer Epidemiol Biomarkers Prev. 2021; 30: 22-29
        • The BC Cancer Agency
        BC cancer cervix Screening 2018 program results.
        (Available at:) (Accessed June 15, 2021)
        • The BC Agency
        BC cancer cervix Screening 2018 program results.
        (Available at:)
        • McHugh M.L.
        Interrater reliability: the kappa statistic.
        Biochem Med. 2012; 22: 276-282
        • Rao A.
        • Young S.
        • Erlich H.
        • et al.
        Development and characterization of the cobas human papillomavirus test.
        J Clin Microbiol. 2013; 51: 1478-1484
        • Cook D.A.
        • Mei W.
        • Smith L.W.
        • et al.
        Comparison of the Roche cobas® 4800 and Digene Hybrid Capture® 2 HPV tests for primary cervical cancer screening in the HPV FOCAL trial.
        BMC Cancer. 2015; 15: 968
        • Phillips S.
        • Garland S.M.
        • Tan J.H.
        • Quinn M.A.
        • Tabrizi S.N.
        Comparison of the Roche Cobas® 4800 HPV assay to Digene Hybrid Capture 2, Roche Linear Array and Roche Amplicor for Detection of High-Risk Human Papillomavirus Genotypes in Women undergoing treatment for cervical dysplasia.
        J Clin Virol. 2015; 62: 63-65
        • Lloveras B.
        • Gomez S.
        • Alameda F.
        • et al.
        HPV Testing by cobas HPV Test in a Population from Catalonia.
        PLoS One. 2013; 8: e58153
        • Ratnam S.
        • Coutlee F.
        • Fontaine D.
        • et al.
        APTIMA HPV E6/E7 mRNA test is as sensitive as hybrid capture 2 assay but more specific at detecting cervical precancer and cancer.
        J Clin Microbiol. 2011; 49: 557-564
        • Iftner T.
        • Becker S.
        • Neis K.J.
        • et al.
        Head-to-head comparison of the RNA-based APTIMA human papillomavirus (HPV) assay and the DNA-based hybrid capture 2 HPV test in a routine screening population of women aged 30 to 60 years in Germany.
        J Clin Microbiol. 2015; 53: 2509-2516
        • Castle P.E.
        • Eaton B.
        • Reid J.
        • Getman D.
        • Dockter J.
        Comparison of human papillomavirus detection by APTIMA HPV and cobas HPV tests in a population of women referred for colposcopy following detection of atypical squamous cells of undetermined significance by Pap cytology.
        J Clin Microbiol. 2015; 53: 1277-1281
        • Coldman A.J.
        • Phillips N.
        • van Niekerk D.
        • et al.
        Projected impact of HPV and LBC primary testing on rates of referral for colposcopy in a Canadian Cervical Cancer Screening Program.
        J Obstet Gynaecol Can. 2015; 37: 412-420
        • Cook D.A.
        • Smith L.W.
        • Law J.
        • et al.
        APTIMA HPV Assay versus Hybrid Capture® 2 HPV test for primary cervical cancer screening in the HPV FOCAL trial.
        J Clin Virol. 2017; 87: 23-29
        • Chorny J.A.
        • Frye T.C.
        • Fisher B.L.
        • Remmers C.L.
        Human papillomavirus detection with genotyping by the cobas and APTIMA assays: significant differences in HPV 16 detection?.
        Diagn Cytopathol. 2018; 46: 568-571
        • Cook D.A.
        • Smith L.W.
        • Law J.H.
        • et al.
        Comparative performance of human papillomavirus messenger RNA versus DNA screening tests at baseline and 48 months in the HPV FOCAL trial.
        J Clin Virol. 2018; 108: 32-37
        • Dillner J.
        • Rebolj M.
        • Birembaut P.
        • et al.
        Long term predictive values of cytology and human papillomavirus testing in cervical cancer screening: joint European cohort study.
        BMJ. 2008; 337: a1754