Women are anticipated to go through more than two rounds of cervical screening in their lifetime. Human papillomavirus (HPV) testing is increasingly used as the primary cervical cancer screening test. However, triage strategies for HPV-positive women were usually evaluated at baseline screening. We assessed the effect of sequential rounds of cervical screening on several algorithms for HPV triage. A total of 1,997 women ages 35–45 years were enrolled in 1999 in Shanxi, P.R. China and followed up three times at approximately 5-year intervals. Cervical intraepithelial neoplasia (CIN) grade 2 or worse (CIN2+) prevalence by prior HPV results and performance of 12 triage algorithms with cytology, genotyping, and prior HPV were examined among 229 HPV-positive women at the fourth round. CIN2+ prevalence varied from 56.5% (95% confidence interval, 36.8%–74.4%) following 15 years HPV persistence to 3.5% (1.2%–9.9%) with an incident HPV within 15 years. Triage with cytology (with threshold of atypical squamous cells of undetermined significance) yielded positive predictive value (PPV) of 21.4% (13.8%–29.0%), entailing immediate colposcopic referral, and negative predictive value (NPV) of 97.4% (94.6%–100%), permitting retesting at short intervals. Triage with genotyping (16/18/31/33/45/52/58) or prior HPV results showed comparable performance with cytology. Among 11 triage algorithms with similar NPV to cytology, triage with prior HPV results and reflex genotyping (16/18) achieved highest PPV of 28.9% (18.8%–39.1%) and lowest colposcopy referral of 33.2% (27.4%–39.5%). HPV persistence across rounds is an effective risk stratifier in HPV-positive women. Mainstream cytology and genotyping, with or without consideration of prior HPV results, remain effective for HPV triage at fourth round.

Prevention Relevance:

The study highlights the sustained effectiveness of mainstream HPV triage methods, such as cytology and genotyping, after sequential rounds of cervical screening. It also suggests that use of HPV persistence across rounds can improve management of HPV-positive women in cervical cancer screening.

Cervical cancer screening is shifting from primary cytology screening to primary human papillomavirus (HPV) testing worldwide, with strong evidence from randomized controlled trials on the enhanced sensitivity and greater protection against high-grade cervical intraepithelial neoplasia (CIN; refs. 1–3). In the global strategy toward eliminating cervical cancer developed by the World Health Organization, a screening coverage target of 70% with HPV testing (or a similarly high sensitivity test) is proposed (4). One of the critical concerns in roll-out of an HPV‐based screening program is the excessive colposcopy referral caused by high prevalence of transient HPV infections (5).

To mitigate the situation, several approaches potentially effective in triaging HPV-positive women have been evaluated, with cytology and HPV16/18 genotyping being the most routinely adopted methods in clinical practice (6). In most cases, triage evaluation was exclusively conducted in women attending their first round of screening. In fact, the prevalence of cervical lesions in subsequent rounds is expected to reduce because of early detection and intervention in previous rounds, especially in an HPV testing–based screening program (6, 7). It is likely that the subsequent rounds might also consist of a larger proportion of short-term HPV infections and minor abnormalities, which might impair the risk stratification capacity of triage tests (7, 8). Reevaluation of triage strategies is thus warranted to determine whether routinely adopted clinical actions for an abnormal test will be equally effective and efficient in managing HPV-positive women in later screening rounds.

Furthermore, considering the fundamental role of persistent HPV infection, rather than transient infection in cervical carcinogenesis, the risk posed by a positive HPV test could vary according to how long the HPV infection has persisted (9, 10). This suggested the rationale of incorporating prior HPV results into consideration when triaging HPV-positive women in later rounds.

Previous evidence from the POBASCAM (Population Based Screening Study Amsterdam) cohort suggested that HPV-positive women in the second HPV-based screening round can be suitably managed by cytology, HPV16/18 genotyping, and the HPV results at the previous screen (11). Currently, the screening interval adopted within HPV-based screening programs has varied from 3 to 10 years (6, 12, 13). It is possible that women would go through more than two screening rounds in their whole life. Here, we conducted a post hoc analysis within the Shanxi Province Cervical Cancer Screening Study I (SPOCCS I) in P.R. China, based on baseline exams and three subsequent rounds of screening at approximately 5 years apart, to determine the effect of sequential rounds of screening on test performance of triaging strategies to manage HPV-positive women and use of prior HPV results to allow optimized management. We used prevalent CIN2+ as the endpoint to reflect the sequential screening effect in a real-world context, noting that not all women were compliant to treatment of detected lesions in earlier rounds.

Study population

In 1999, 1,997 nonpregnant women ages 35–45 years with no history of cervical cancer or hysterectomy were enrolled in SPOCCS I via cluster sampling in Shanxi province. Three subsequent follow-up visits were conducted in 2005, 2010, and 2014. The study was approved by Human Subjects Review Board of the Cancer Institute/Hospital, Chinese Academy of Medical Sciences (CICAMS, Beijing, P.R. China) and conducted in accordance with the Ethical Principles for Biomedical Research Involving Human Subjects (Ministry of Health of the P.R. China) and the Declaration of Helsinki. All women provided written informed consent at both baseline and follow-up visits.

Screening and clinical management

Details regarding the study procedures in SPOCCS I and some other findings have been published previously (14–18). In brief, at enrollment, each woman underwent a cervical examination and testing, including high-risk HPV (hrHPV) DNA tests (hybrid capture 2, HC2), liquid-based cytology (LBC), visual inspection with acetic acid (VIA), fluorescence spectroscopy, and colposcopy examination. Final histologic diagnosis was based on four-quadrant biopsies and endocervical curettage for every woman. During the three subsequent visits, women were screened with HC2, LBC [with threshold of atypical squamous cells of undetermined significance (ASC-US)], and VIA (except VIA at fourth visit due to low sensitivity among older women; Fig. 1). Participants with any positive screening result were referred to colposcopy and underwent biopsies if necessary. All collected samples were transferred to the CICAMS (Beijing, P.R. China) central laboratory for test or diagnosis. Cervical biopsy slides were prepared and read by experienced histopathologists blinded to other screening results. For unavailable pathologic diagnoses due to nonattendance at colposcopy exams after positive screening results, a combination of screening tests was used to verify the final disease status (19). Residual cytologic specimens at baseline were not available. Residual cytologic specimens in the second, third, and fourth screening rounds were stored in the refrigerator (at −80°C). In 2016, a post hoc HPV genotyping test was performed in these specimens, which were positive by HC2, by using PCR-based Reverse Hybridization Line Probe Assay (INNO-LiPA Extra, Innogenetics) with a SPF10 Primers Set (DDL Diagnostic Laboratory; SPF10-LiPA; ref. 16). In this analysis, genotyping results only from the fourth screening round were used.

Figure 1.

Flowchart of the analytic population from SPOCCS I in four screening rounds, in P.R. China. aA total of 10 women were excluded from second round of analysis because of undetermined histology results in that year, but were eligible for inclusion in subsequent screening rounds. bA total of four women were excluded from third round of analysis because of undetermined histology results in that year, but were eligible for inclusion in subsequent screening rounds. cA total of 17 women were excluded from fourth round of analysis because of undetermined histology results in that year, but were eligible for inclusion in subsequent screening round. dHPV genotyping was performed using residual cytologic samples tested positive by HC2 in second, third, and fourth rounds of screening. In this analysis, only results from fourth screening round were used. eTwo CIN2+ cases were HPV negative.

Figure 1.

Flowchart of the analytic population from SPOCCS I in four screening rounds, in P.R. China. aA total of 10 women were excluded from second round of analysis because of undetermined histology results in that year, but were eligible for inclusion in subsequent screening rounds. bA total of four women were excluded from third round of analysis because of undetermined histology results in that year, but were eligible for inclusion in subsequent screening rounds. cA total of 17 women were excluded from fourth round of analysis because of undetermined histology results in that year, but were eligible for inclusion in subsequent screening round. dHPV genotyping was performed using residual cytologic samples tested positive by HC2 in second, third, and fourth rounds of screening. In this analysis, only results from fourth screening round were used. eTwo CIN2+ cases were HPV negative.

Close modal

Statistical analyses

We first analyzed CIN grade 2 or worse (CIN2+) prevalence by screening rounds. The diagram of inclusions and exclusions is shown in Fig. 1. A total of 1,997 women ages 35–45 years old were recruited at baseline. After excluding nonattendance and hysterectomy, 1,742, 1,455, and 1,172 women, respectively, were eligible for inclusion in the analysis of the second, third, and fourth rounds. A total of 10, four, and 17 women were additionally excluded from the second, third, and fourth rounds' analysis temporarily, because of undetermined histology results at that year, but were eligible for inclusion in subsequent screening rounds (Fig. 1). The endpoint, prevalent CIN2+, was defined as both incident CIN2+ cases at corresponding round and previous CIN2+ cases which persisted because of no treatment applied. Given that four-quadrant biopsies were performed at baseline and to facilitate comparability across rounds, we simulated the same screening strategy at baseline as in the subsequent rounds, that is, cotesting using HPV and cytology (at threshold of ASC-US), with either positive tests referred to colposcopy. Because the sensitivity of cotesting was 100% at baseline, all 86 CIN2+ cases detected at baseline were included into analysis when comparing overall CIN2+ reduction and testing trend on lesion grades at diagnosis across rounds.

A total of 1,126 women who were included in the analysis of the fourth screening round with determined histology results were then analyzed for HPV prevalence by prior HPV results. In 229 HPV-positive women in the fourth round, the positive predictive value (PPV) to detect CIN2+ by prior HPV results and test performance of 12 different triage algorithms were evaluated (Table 1). HPV history across rounds was subdivided into six categories according to HPV results in the three prior rounds to assess differences in PPV to detect CIN2+ by categories. OR was estimated using logistic regression.

Table 1.

Twelve triage algorithms analyzed at fourth screening round.

HPV triage algorithmsFirst triage testTest resultSimulative clinical action
Cytology (with ASC-US+ threshold) Cytology ASC-US+ Colposcopy referral 
   Normal — 
Genotyping (16/18) Genotyping HPV 16/18 positive Colposcopy referral 
   HPV 16/18 negative — 
Genotyping (16/18/31/33/45/52/58) Genotyping HPV 16/18/31/33/45/52/58 positive Colposcopy referral 
   HPV 16/18/31/33/45/52/58 negative — 
Genotyping (16/18) with reflex cytology Genotyping HPV 16/18 positive Colposcopy referral 
   HPV 16/18 negative Cytology reflex (with ASC-US+ threshold) 
Genotyping (16/18/31/33/45/52/58) with reflex cytology Genotyping HPV 16/18/31/33/45/52/58 positive Colposcopy referral 
   HPV 16/18/31/33/45/52/58 negative Cytology reflex (with ASC-US+ threshold) 
HPV results across rounds-a HPV results across rounds Persistent HPV within 4–15 years Colposcopy referral 
   Incident HPV within 4–15 years — 
HPV results across rounds-a with cytology reflex HPV results across rounds Persistent HPV within 4–15 years Colposcopy referral 
   Incident HPV within 4–15 years Cytology reflex (with ASC-US+ threshold) 
HPV results across rounds-a with genotyping (16/18) reflex HPV results across rounds Persistent HPV within 4–15 years Colposcopy referral 
   Incident HPV within 4–15 years Genotyping (16/18) reflex 
HPV results across rounds-a with genotyping (16/18/31/33/45/52/58) reflex HPV results across rounds Persistent HPV within 4–15 years Colposcopy referral 
   Incident HPV within 4–15 years Genotyping (16/18/31/33/45/52/58) reflex 
10 HPV results across rounds-b with cytology reflex HPV results across rounds Persistent HPV within 9–15 years Colposcopy referral 
   Persistent HPV within 4 years or Incident HPV within 4–15 years Cytology reflex (with ASC-US+ threshold) 
11 HPV results across rounds-b with Genotyping (16/18) reflex HPV results across rounds Persistent HPV within 9–15 years Colposcopy referral 
   Persistent HPV within 4 years or Incident HPV within 4–15 years Genotyping (16/18) reflex 
12 HPV results across rounds-b with Genotyping (16/18/31/33/45/52/58) reflex HPV results across rounds Persistent HPV within 9–15 years Colposcopy referral 
   Persistent HPV within 4 years or Incident HPV within 4–15 years Genotyping (16/18/31/33/45/52/58) reflex 
HPV triage algorithmsFirst triage testTest resultSimulative clinical action
Cytology (with ASC-US+ threshold) Cytology ASC-US+ Colposcopy referral 
   Normal — 
Genotyping (16/18) Genotyping HPV 16/18 positive Colposcopy referral 
   HPV 16/18 negative — 
Genotyping (16/18/31/33/45/52/58) Genotyping HPV 16/18/31/33/45/52/58 positive Colposcopy referral 
   HPV 16/18/31/33/45/52/58 negative — 
Genotyping (16/18) with reflex cytology Genotyping HPV 16/18 positive Colposcopy referral 
   HPV 16/18 negative Cytology reflex (with ASC-US+ threshold) 
Genotyping (16/18/31/33/45/52/58) with reflex cytology Genotyping HPV 16/18/31/33/45/52/58 positive Colposcopy referral 
   HPV 16/18/31/33/45/52/58 negative Cytology reflex (with ASC-US+ threshold) 
HPV results across rounds-a HPV results across rounds Persistent HPV within 4–15 years Colposcopy referral 
   Incident HPV within 4–15 years — 
HPV results across rounds-a with cytology reflex HPV results across rounds Persistent HPV within 4–15 years Colposcopy referral 
   Incident HPV within 4–15 years Cytology reflex (with ASC-US+ threshold) 
HPV results across rounds-a with genotyping (16/18) reflex HPV results across rounds Persistent HPV within 4–15 years Colposcopy referral 
   Incident HPV within 4–15 years Genotyping (16/18) reflex 
HPV results across rounds-a with genotyping (16/18/31/33/45/52/58) reflex HPV results across rounds Persistent HPV within 4–15 years Colposcopy referral 
   Incident HPV within 4–15 years Genotyping (16/18/31/33/45/52/58) reflex 
10 HPV results across rounds-b with cytology reflex HPV results across rounds Persistent HPV within 9–15 years Colposcopy referral 
   Persistent HPV within 4 years or Incident HPV within 4–15 years Cytology reflex (with ASC-US+ threshold) 
11 HPV results across rounds-b with Genotyping (16/18) reflex HPV results across rounds Persistent HPV within 9–15 years Colposcopy referral 
   Persistent HPV within 4 years or Incident HPV within 4–15 years Genotyping (16/18) reflex 
12 HPV results across rounds-b with Genotyping (16/18/31/33/45/52/58) reflex HPV results across rounds Persistent HPV within 9–15 years Colposcopy referral 
   Persistent HPV within 4 years or Incident HPV within 4–15 years Genotyping (16/18/31/33/45/52/58) reflex 

Abbreviation: ASC-US+, ASC-US or higher.

The test performance of multiple triage strategies among HPV-positive women was evaluated by calculating the sensitivity, specificity, PPV, and the negative predictive value (NPV) for CIN2+, with respective 95% confidence intervals (CI). In determining the test performance of triage algorithms and to make scientific decisions according to posttest result, a benchmarking concept of equal management to equal risk has been developed, with different thresholds and management options proposed in the United States and European countries (20–25). In this study, we adopted the intermediate benchmark risk levels proposed by Arbyn and colleagues, that is, a threshold of 20% for PPV to CIN2+, which is actually the absolute CIN2+ prevalence after positive triage test(s) in HPV-positive women entailing immediate referral to colposcopy, and 98% for NPV to CIN2+, which is one minus the absolute CIN2+ prevalence after negative triage test(s) in HPV-positive women ensuring return at routine screening intervals (24, 25). Relative sensitivity, specificity, PPV, and NPV in HPV-positive women were calculated using cytology with a threshold of ASC-US to triage HPV-positive women as the comparator for all other triage strategies. Colposcopy referral rate was evaluated among HPV-positive women, with women who are HPV positive and then positive by corresponding triage strategies as the numerator.

Relative sensitivity and specificity were calculated using Stata (version 14.0, StataCorp). The remaining statistical analyses were performed using R (version 3.6.0). P < 0.05 (two-sided) was considered to be statistically significant.

Table 2 shows the CIN2+ prevalence and distribution of lesion grades at the baseline and three subsequent rounds of screening. CIN2+ prevalence was 4.3% (95% CI, 3.5%–5.3%) in women at baseline and reduced to 1.3% (95% CI, 0.9%–2.0%), 3.3% (95% CI 2.5%–4.4%), and 2.5% (95% CI, 1.8%–3.6%), respectively, in women attending their second, third, and fourth round, which were 70%, 24%, and 43% reduction, respectively, compared with that at baseline. Cases were detected at an earlier grade at later screening rounds (P = 0.029). Among 86 CIN2+ cases detected at baseline, half of the cases were diagnosed at CIN2 and 14% were cervical cancer. In the fourth round of screening, the proportion of CIN2 increased to 72%, while proportion of cervical cancer decreased to 3.4% (Table 2).

Table 2.

CIN2+ prevalence after sequential rounds of cervical cancer screening.

CIN2+ prevalenceLesion grade distribution at diagnosis
CIN2, nCIN3, nCervical cancer, n
Calendar yearRoundNn% (95% CI)OR (95% CI)% (95% CI)% (95% CI)% (95% CI)Overall Pa
1999 Baseline 1,997 86b 4.3 (3.5–5.3) Reference 43 31 12 0.029 
      50.0 (39.5–61.3) 36.0 (25.6–47.3) 14.0 (3.5–25.2)  
2005 Second 1,732 23 1.3 (0.9–2.0) 0.30 (0.18–0.48) 16  
      69.6 (56.5–90.8) 30.4 (17.4–51.7) 0.0 (0.0–21.3)  
2010 Third 1,451 48 3.3 (2.5–4.4) 0.76 (0.52–1.10) 26 19  
      54.2 (41.7–69.7) 39.6 (27.1–55.1) 6.2 (0.0–21.8)  
2014 Fourth 1,155 29 2.5 (1.8–3.6) 0.57 (0.37–0.88) 21  
      72.4 (58.6–88.6) 24.1 (10.3–40.3) 3.4 (0.0–19.6)  
CIN2+ prevalenceLesion grade distribution at diagnosis
CIN2, nCIN3, nCervical cancer, n
Calendar yearRoundNn% (95% CI)OR (95% CI)% (95% CI)% (95% CI)% (95% CI)Overall Pa
1999 Baseline 1,997 86b 4.3 (3.5–5.3) Reference 43 31 12 0.029 
      50.0 (39.5–61.3) 36.0 (25.6–47.3) 14.0 (3.5–25.2)  
2005 Second 1,732 23 1.3 (0.9–2.0) 0.30 (0.18–0.48) 16  
      69.6 (56.5–90.8) 30.4 (17.4–51.7) 0.0 (0.0–21.3)  
2010 Third 1,451 48 3.3 (2.5–4.4) 0.76 (0.52–1.10) 26 19  
      54.2 (41.7–69.7) 39.6 (27.1–55.1) 6.2 (0.0–21.8)  
2014 Fourth 1,155 29 2.5 (1.8–3.6) 0.57 (0.37–0.88) 21  
      72.4 (58.6–88.6) 24.1 (10.3–40.3) 3.4 (0.0–19.6)  

aAsymptotic linear-by-linear association test.

bGiven that four-quadrant biopsies were performed at baseline, the same screening strategy as in the subsequent rounds was simulated at baseline, that is, cotesting using HPV and cytology (at threshold of ASC-US), with either positive tests referred to colposcopy. Because the sensitivity of cotesting was 100% at baseline, all 86 CIN2+ cases detected in 1,997 were included into analysis when comparing overall CIN2+ reduction and testing trend on lesion grades at diagnosis across rounds.

In 1,155 women attending their fourth screening round with determined histology results in 2014, HPV prevalence correlated well with HPV results in previous rounds, as shown in Table 3. Among women with consecutive positive HPV tests over 1–3 rounds, HPV prevalence was 34.9% (95% CI, 28.1%–42.5%) for women first positive 4 years ago, 50% (95% CI, 36.4%–63.6%) for women first positive 9 years ago, and 56.1% (95% CI, 41.0%–70.1%) for women first positive 15 years ago. These prevalence rates were 3.8, 7.1, and 9.1 times higher for women positive once, twice, or thrice through screening rounds compared with women with three consecutive negative HPV tests over 15 years. In 229 HPV-positive women in fourth round, PPV to detect CIN2+ by HPV history across rounds significantly increased from 3.5% (95% CI, 1.2%–9.9%) following an incident HPV within 15 years to 56.5% (95% CI, 36.8%–74.4%) following 15 years HPV persistence (Table 3).

Table 3.

HPV prevalence and PPV to detect CIN2+ in fourth screening round, stratified by HPV results across rounds.

HPV results in previous roundsPositive HPV in fourth roundPPV to detect CIN2+b
201020051999N totalnPrevalence (95% CI)OR (95% CI)Category of HPV results across roundsanPPV (95% CI)OR (95% CI)
Pos Pos Pos 41 23 56.1 (41.0–70.1) 9.09 (4.73–17.76) Persistent HPV within 15 years 13 56.5 (36.8–74.4) 35.53 (9.63–176.23) 
Pos Pos Neg 48 24 50.0 (36.4–63.6) 7.12 (3.86–13.15) Persistent HPV within 9 years 16.7 (6.7–35.9) 5.47 (1.12–29.65) 
Pos Neg Neg 166 58 34.9 (28.1–42.5) 3.82 (2.58–5.65) Persistent HPV within 4 years 12.1 (6.0–22.9) 3.75 (0.99–18.01) 
Neg Pos Pos 95 18 18.9 (12.3–28.0) 1.66 (0.93–2.86) Incident HPV within 4 years 0.0 (0.0–17.6) 0.00 (0.00–12.14) 
Neg Neg Pos 115 21 18.3 (12.3–26.3) 1.59 (0.92–2.65) Incident HPV within 9 years 0.0 (0.0–15.5) 0.00 (0.00–10.34) 
Neg Neg Neg 690 85 12.3 (10.1–15.0) Reference Incident HPV within 15 years 3.5 (1.2–9.9) Reference 
HPV results in previous roundsPositive HPV in fourth roundPPV to detect CIN2+b
201020051999N totalnPrevalence (95% CI)OR (95% CI)Category of HPV results across roundsanPPV (95% CI)OR (95% CI)
Pos Pos Pos 41 23 56.1 (41.0–70.1) 9.09 (4.73–17.76) Persistent HPV within 15 years 13 56.5 (36.8–74.4) 35.53 (9.63–176.23) 
Pos Pos Neg 48 24 50.0 (36.4–63.6) 7.12 (3.86–13.15) Persistent HPV within 9 years 16.7 (6.7–35.9) 5.47 (1.12–29.65) 
Pos Neg Neg 166 58 34.9 (28.1–42.5) 3.82 (2.58–5.65) Persistent HPV within 4 years 12.1 (6.0–22.9) 3.75 (0.99–18.01) 
Neg Pos Pos 95 18 18.9 (12.3–28.0) 1.66 (0.93–2.86) Incident HPV within 4 years 0.0 (0.0–17.6) 0.00 (0.00–12.14) 
Neg Neg Pos 115 21 18.3 (12.3–26.3) 1.59 (0.92–2.65) Incident HPV within 9 years 0.0 (0.0–15.5) 0.00 (0.00–10.34) 
Neg Neg Neg 690 85 12.3 (10.1–15.0) Reference Incident HPV within 15 years 3.5 (1.2–9.9) Reference 

Abbreviations: Neg, negative; pos, positive.

aIn women HPV positive at fourth round of screening.

bPPV of HPV primary screening at fourth round of screening.

Twelve different algorithms, with or without consideration of HPV history across rounds, were evaluated to identify the test performance in managing HPV-positive women at fourth screening round (Table 4). PPV estimate of HPV primary screening alone was 11.8% (95% CI, 8.2%–16.6%), supporting the need for a secondary triage strategy.

Table 4.

Clinical performance of different triage strategies for detection of CIN2+ in HPV-positive women attending fourth round of screening (N = 229).

Triage algorithmSensitivity % (95% CI)Relative sensitivity (95% CI)Specificity % (95% CI)Relative specificity (95% CI)PPV % (95% CI)Relative PPV (95% CI)NPV % (95% CI)Relative NPV (95% CI)Colposcopy referral % (95% CI)
Cytology (with ASC-US+ threshold) 88.9 (77.0–100) Reference 56.4 (49.6–63.2) Reference 21.4 (13.8–29.0) Reference 97.4 (94.6–100.0) Reference 48.9 (42.5–55.3) 
Genotyping (16/18) 55.6 (36.8–74.3) 0.63 (0.44–0.88) 84.2 (79.1–89.2) 1.49 (1.31–1.70) 31.9 (18.6–45.2) 1.49 (1.06–2.10) 93.4 (89.8–97.0) 0.96 (0.92–1.00) 20.5 (15.8–26.2) 
Genotyping (16/18/31/33/45/52/58) 88.9 (77–100.0) 1.00 (0.82–1.22) 48.0 (41.1–54.9) 0.85 (0.70–1.04) 18.6 (11.9–25.3) 0.87 (0.69–1.10) 97.0 (93.7–100.0) 1.00 (0.95–1.04) 56.3 (49.9–62.6) 
Genotyping (16/18) and cytology 92.6 (82.7–100.0) 1.04 (0.96–1.13) 48.5 (41.6–55.4) 0.86 (0.80–0.93) 19.4 (12.6–26.2) 0.9 (0.83–0.99) 98.0 (95.3–100.0) 1.01 (0.99–1.02) 56.3 (49.9–62.6) 
Genotyping (16/18/31/33/45/52/58) and cytology 100.0 (100.0–100.0) 1.13 (0.98–1.29) 24.8 (18.8–30.7) 0.44 (0.36–0.54) 15.1 (9.8–20.3) 0.70 (0.60–0.82) 100.0 (100.0–100.0) 1.03 (1.00–1.06) 78.2 (72.4–83.0) 
HPV results across rounds-aa 88.9 (77–100.0) 1.00 (0.82–1.22) 59.9 (53.1–66.7) 1.06 (0.90–1.25) 22.9 (14.8–30.9) 1.07 (0.84–1.35) 97.6 (94.9–100.0) 1.00 (0.96–1.04) 45.9 (39.5–52.3) 
HPV results across rounds-aa and cytology 100.0 (100.0–100.0) 1.13 (0.98–1.29) 35.1 (28.6–41.7) 0.62 (0.54–0.72) 17.1 (11.2–23.0) 0.80 (0.69–0.92) 100.0 (100.0–100.0) 1.03 (1.00–1.06) 69.0 (62.7–74.6) 
HPV results across rounds-aa and genotyping (16/18) 92.6 (82.7–100.0) 1.04 (0.87–1.25) 51.0 (44.1–57.9) 0.90 (0.76–1.08) 20.2 (13.1–27.2) 0.94 (0.76–1.17) 98.1 (95.5–100.0) 1.01 (0.97–1.05) 54.1 (47.7–60.5) 
HPV results across rounds-aa and genotyping (16/18/31/33/45/52/58) 92.6 (82.7–100.0) 1.04 (0.87–1.25) 30.7 (24.3–37.1) 0.54 (0.43–0.69) 15.2 (9.7–20.6) 0.71 (0.57–0.87) 96.9 (92.6–100.0) 0.99 (0.94–1.05) 72.1 (65.9–77.5) 
10 HPV results across rounds-bb and cytology 100.0 (100.0–100.0) 1.13 (0.98–1.29) 48.0 (41.1–54.9) 0.85 (0.79–0.92) 20.5 (13.6–27.3) 0.95 (0.84–1.08) 100.0 (100.0–100.0) 1.03 (1.00–1.06) 57.6 (51.2–63.9) 
11 HPV results across rounds-bb and genotyping (16/18) 81.5 (66.8–96.1) 0.92 (0.72–1.17) 73.3 (67.2–79.4) 1.30 (1.12–1.50) 28.9 (18.8–39.1) 1.35 (1.03–1.77) 96.7 (93.9–99.5) 0.99 (0.95–1.03) 33.2 (27.4–39.5) 
12 HPV results across rounds-bb and genotyping (16/18/31/33/45/52/58) 92.6 (82.7–100.0) 1.04 (0.87–1.25) 43.6 (36.7–50.4) 0.77 (0.63–0.95) 18.0 (11.6–24.4) 0.84 (0.67–1.05) 97.8 (94.7–100.0) 1.00 (0.96–1.05) 60.7 (54.2–66.8) 
Triage algorithmSensitivity % (95% CI)Relative sensitivity (95% CI)Specificity % (95% CI)Relative specificity (95% CI)PPV % (95% CI)Relative PPV (95% CI)NPV % (95% CI)Relative NPV (95% CI)Colposcopy referral % (95% CI)
Cytology (with ASC-US+ threshold) 88.9 (77.0–100) Reference 56.4 (49.6–63.2) Reference 21.4 (13.8–29.0) Reference 97.4 (94.6–100.0) Reference 48.9 (42.5–55.3) 
Genotyping (16/18) 55.6 (36.8–74.3) 0.63 (0.44–0.88) 84.2 (79.1–89.2) 1.49 (1.31–1.70) 31.9 (18.6–45.2) 1.49 (1.06–2.10) 93.4 (89.8–97.0) 0.96 (0.92–1.00) 20.5 (15.8–26.2) 
Genotyping (16/18/31/33/45/52/58) 88.9 (77–100.0) 1.00 (0.82–1.22) 48.0 (41.1–54.9) 0.85 (0.70–1.04) 18.6 (11.9–25.3) 0.87 (0.69–1.10) 97.0 (93.7–100.0) 1.00 (0.95–1.04) 56.3 (49.9–62.6) 
Genotyping (16/18) and cytology 92.6 (82.7–100.0) 1.04 (0.96–1.13) 48.5 (41.6–55.4) 0.86 (0.80–0.93) 19.4 (12.6–26.2) 0.9 (0.83–0.99) 98.0 (95.3–100.0) 1.01 (0.99–1.02) 56.3 (49.9–62.6) 
Genotyping (16/18/31/33/45/52/58) and cytology 100.0 (100.0–100.0) 1.13 (0.98–1.29) 24.8 (18.8–30.7) 0.44 (0.36–0.54) 15.1 (9.8–20.3) 0.70 (0.60–0.82) 100.0 (100.0–100.0) 1.03 (1.00–1.06) 78.2 (72.4–83.0) 
HPV results across rounds-aa 88.9 (77–100.0) 1.00 (0.82–1.22) 59.9 (53.1–66.7) 1.06 (0.90–1.25) 22.9 (14.8–30.9) 1.07 (0.84–1.35) 97.6 (94.9–100.0) 1.00 (0.96–1.04) 45.9 (39.5–52.3) 
HPV results across rounds-aa and cytology 100.0 (100.0–100.0) 1.13 (0.98–1.29) 35.1 (28.6–41.7) 0.62 (0.54–0.72) 17.1 (11.2–23.0) 0.80 (0.69–0.92) 100.0 (100.0–100.0) 1.03 (1.00–1.06) 69.0 (62.7–74.6) 
HPV results across rounds-aa and genotyping (16/18) 92.6 (82.7–100.0) 1.04 (0.87–1.25) 51.0 (44.1–57.9) 0.90 (0.76–1.08) 20.2 (13.1–27.2) 0.94 (0.76–1.17) 98.1 (95.5–100.0) 1.01 (0.97–1.05) 54.1 (47.7–60.5) 
HPV results across rounds-aa and genotyping (16/18/31/33/45/52/58) 92.6 (82.7–100.0) 1.04 (0.87–1.25) 30.7 (24.3–37.1) 0.54 (0.43–0.69) 15.2 (9.7–20.6) 0.71 (0.57–0.87) 96.9 (92.6–100.0) 0.99 (0.94–1.05) 72.1 (65.9–77.5) 
10 HPV results across rounds-bb and cytology 100.0 (100.0–100.0) 1.13 (0.98–1.29) 48.0 (41.1–54.9) 0.85 (0.79–0.92) 20.5 (13.6–27.3) 0.95 (0.84–1.08) 100.0 (100.0–100.0) 1.03 (1.00–1.06) 57.6 (51.2–63.9) 
11 HPV results across rounds-bb and genotyping (16/18) 81.5 (66.8–96.1) 0.92 (0.72–1.17) 73.3 (67.2–79.4) 1.30 (1.12–1.50) 28.9 (18.8–39.1) 1.35 (1.03–1.77) 96.7 (93.9–99.5) 0.99 (0.95–1.03) 33.2 (27.4–39.5) 
12 HPV results across rounds-bb and genotyping (16/18/31/33/45/52/58) 92.6 (82.7–100.0) 1.04 (0.87–1.25) 43.6 (36.7–50.4) 0.77 (0.63–0.95) 18.0 (11.6–24.4) 0.84 (0.67–1.05) 97.8 (94.7–100.0) 1.00 (0.96–1.05) 60.7 (54.2–66.8) 

aWomen with persistent HPV within 4–15 years were referred to colposcopy and women with incident HPV within 4–15 years were triaged by other tests or without any interventions.

bWomen with persistent HPV within 9–15 years were referred to colposcopy and the remaining were triaged by other tests.

Among the first five traditional strategies that did not include HPV information in previous screening rounds (strategy 1–5), cytology (strategy 1), as the reference test, demonstrated a PPV of 21.4% (95% CI, 13.8%–29.0%) for CIN2+, with the estimate higher than the colposcopy threshold of 20%, and NPV at 97.4% (95% CI, 94.6%–100%), with the estimate slightly below the routine screening threshold of 98% (Table 4). Genotyping (16/18) (strategy 2) showed a significantly higher PPV at the cost of lower NPV than cytology (relative PPV at 1.49; 95% CI, 1.06–2.10 and relative NPV at 0.96; 95% CI, 0.92–1.00). Triage with expanded genotyping (16/18/31/33/45/52/58) (strategy 3) achieved comparable performance with cytology (Table 4). Combining cytology with genotyping (16/18/31/33/45/52/58) (strategy 5) detected all CIN2+ cases identified by HPV tests, but dramatically reduced PPV (relative PPV at 0.70; 95% CI, 0.60–0.82).

Triage with HPV history alone (strategy 6) achieved comparable PPV and NPV with cytology (relative PPV at 1.07; 95% CI, 0.84–1.35 and relative NPV at 1.00; 95% CI, 0.96–1.04; Table 4). Adding reflex cytology to triage women with incident HPV (strategy 7) achieved NPV at 100%, but was also accompanied by a significantly reduced PPV (relative PPV at 0.80; 95% CI, 0.69–0.92). Adding reflex HPV16/18 genotyping (strategy 8) or reflex HPV16/18/31/33/45/52/58 genotyping (strategy 9) to triage women with incident HPV did not improve NPV and showed similar or reduced PPV compared with triage with HPV history alone (Table 4). Further triaging women with 4 years of HPV persistence (strategy 10–12) and only referring women with 9–15 years of HPV persistence to immediate colposcopy improved PPV, without significant loss on NPV (Table 4).

Colposcopy referral rates among HPV-positive women at fourth screening round are presented in Table 4. Among all strategies with comparable NPV estimates with cytology, triage algorithm which combined prior HPV results and reflex genotyping (16/18) in women following an incident HPV or persistent HPV within 4 years (strategy 11) achieved the lowest colposcopy referral rate at 33.2% (95% CI, 27.4%–39.5%). In contrast, expanded genotyping (16/18/31/33/45/52/58), either combined with cytology or prior HPV results, imposed the highest colposcopy referral rate (strategy 5, 78.2%; 95% CI, 72.4%–83.0% and strategy 9, 72.1%; 95% CI, 65.9%–77.5%).

On the basis of a longitudinal cohort, our study reported the effect of four sequential rounds of screening on CIN2+ prevalence and the test performance of different triage algorithms based on cytology, genotyping, and HPV history across rounds. At fourth screening round, we found a 43% CIN2+ reduction compared with simulated baseline screening round. We also observed the HPV prevalence and PPV to detect prevalent CIN2+ varied dramatically depending on the duration of HPV persistence. All triage algorithms presented acceptable test performance in triaging HPV-positive women at fourth screening round in a real-life context, with varied balance on screening accuracy and efficiency.

The significant reduction of CIN2+ in subsequent screening rounds, especially when HPV-based screening was adopted, has been reported in several clinical trials and organized cervical cancer screening programs (1, 6, 7). Results from ARTISTIC trial showed CIN2+ incidence rate reduced from 2.39% in round one to 0.74% in round three (1). The reduction observed in our study, using the simulated cotesting screening strategy with HPV and cytology (ASC-US) was less evident. Lack of early recalls to positive screening results after 6–12 months in our cohort limits a dramatic reduction as observed in ARTISTIC trial. Besides, we used prevalent CIN2+, rather than incident CIN2+, as the endpoint during data analysis. Although 30%–40% CIN2+ lesions diagnosed, but not treated, could regress, some CIN2+ cases might persist or progress to more severe forms and would be detected again subsequently (26). These cases were included in this analysis to reflect a real-life effectiveness after sequential screening, considering that treatment compliance is an important factor negatively impacting the effectiveness in roll-out of an organized cervical screening program. Under the global call of cervical cancer elimination, treatment of preinvasive lesions and invasive cancer of 90% has been proposed as one of the 2030 triple-intervention coverage targets (4). Recent studies on “see-and-treat” approach using thermal ablation also showed promising effect in alleviating the situation (27, 28). The trend of detecting lesions at a lower grade was observed in this cohort and supports that, with implementation of screening rounds, many more women will be diagnosed at earlier grade when treatment is less invasive and more effective, which will reduce the likelihood of death, disability, and infertility and reinforce the effectiveness of organized cervical cancer screening (29).

As observed in our study, HPV persisted in 35%–56% of women following consecutive positive HPV test for 1–3 rounds, among whom 12%–56% were detected with CIN2+. Swedescreen trial also reported similar results in women with one or three years of HPV persistence, but cytology negative at baseline (30). These results suggest that substantial proportion of infections in women with prior positive HPV results are persistent and immediate colposcopy might be efficient after a current positive HPV test without further reflex tests (31, 32). The significant low PPV in women with incident HPV within 4–15 years might suggest that either prolonged recall interval or further triage tests are both acceptable in current HPV-positive women following a recent past negative HPV test. These findings support the potential of prior HPV result as an effective risk stratifier in HPV-positive women in subsequent screening rounds, which is also considered in the 2019 revision of the American Society for Colposcopy and Cervical Pathology Risk-Based Management Consensus Guidelines (33).

A screening test or strategy is supposed to distinguish between high-grade CIN with higher probability of progressing to invasive cancer and minor lesions that probably regress, to provide informative posttest results for different clinical actions (e.g., immediate colposcopic biopsies or treatment, return for repeated screening in 6–12 months, or return for routine screening schedule; ref. 34). The concept of equal management to equal risk provides a rational benchmarking for decision-making in clinical action thresholds, especially in the context of increasing options of emerging tests or test combinations. While the threshold at which to determine immediate colposcopy and to determine the surveillance period vary between countries. In the Netherland, the decision threshold is 2% for 4-year CIN grade 3 or worse (CIN3+) risk after negative triage (1 − NPV) and 20% for immediate CIN3+ risk after positive triage (PPV; ref. 23). In Norway, the threshold of 1% and 24% for cumulative CIN3+ risk over a period of 21 months was proposed (21, 22). While in the new consensus in the United States, a more complicated scenario is utilized (20). If the immediate risk of CIN3+ is 4% or greater, immediate management via colposcopy or treatment is indicated. If the immediate risk is less than 4%, the 5-year CIN3+ risk is examined to determine whether patients should return in 1 year (≥0.55%), 3 years (0.15–0.54%), or 5 years (<0.15%). An intermediate threshold of 1% and 10% for CIN3+ prevalence and 2% and 20% for CIN2+ prevalence was proposed by Arbyn and colleagues (24, 25). Currently in China, there has not yet been a consensus on the clinical threshold to adopt. Considering the CIN2+ prevalence endpoint used in this study and according to a previous triage analysis in Chinese women, we adopted the intermediate benchmark risk levels proposed by Arbyn and colleagues (NPV at 98% and PPV at 20%) to reevaluate triage strategies in managing HPV-positive women at the fourth screening round (24, 25, 35). It is noteworthy that PPV and NPV depend on underlying prevalence in the population screened. In practice, local experts are encouraged to define the respective threshold according to the local risk levels, tolerance to overdiagnosis or overtreatment, adherence to follow-up recommendations, and also taking the resource availability into consideration when making policy decisions.

Findings of triage performance in the fourth screening round demonstrated that test performance remained acceptable on the basis of routinely adopted clinical actions for the mainstream triage algorithms incorporating cytology and genotyping, which were in line with findings from POBASCAM based on two rounds of cervical screening (11). Cytology triage alone achieved a PPV estimate higher than 20% and an NPV estimate slightly below 98% in our study, which suggest that, after sequential rounds of screening, the clinical actions of return at short intervals is still applicable after a negative cytology. Data from Kaiser Permanente Northern California concluded that routinely adopted clinical actions for abnormal cytology in HPV-positive women might need to be degraded following one or more sequential negative HPV tests, due to insufficient risks in later rounds (36). Our study is not powerful enough for such analysis as only three CIN2+ cases were detected in fourth round following an incident HPV. However, we shed light on the general impact of sequential rounds of screening on mainstream triaging algorithms when prior HPV results are not available. We also added value on the test performance of overall triage algorithms taking prior HPV results into consideration. Triage with single prior HPV results did not significantly improve the test performance compared with reference cytology or genotyping (16/18/31/33/45/52/58), but showed advantage in utilizing previous information without workload on additional triage tests based on all HPV-positive samples compared with cytology and 10% lower colposcopy referrals compared with genotyping (16/18/31/33/45/52/58). Further triaging women following a persistent HPV for 4 years using genotyping 16/18, actually made it the most effective and efficient option among all triage algorithms with acceptable NPV.

In clinical practice, although cytology is the most routinely adopted triage in national HPV-based screening program in high-income countries, one challenge is the wide variance on diagnostic performance among low-resource settings. There is also concern that when HPV primary screening is implemented in routine practice, unblinded cytologists may have some increased attention in reading HPV-positive slides, resulting in a so-called ascertainment bias (30). Capacity to undertake cytology might also be impaired because of limited resources and manpower allocated in quality control. Besides, reflex cytology requires additional tests based on all HPV-positive samples, which carries a considerable burden and cost, especially in low-resource settings. Genotyping outweighs on these points to enable the objective generalizability as a variety of commercially available HPV assays or kits, as well as the capacity to incorporate primary screening and triage into one test. The high expense of a compromise in sensitivity at fourth screening round for HPV16/18 genotyping alone was observed both in our cohort and in POBASCAM (8, 11), which has already been a concern at baseline screening (35). Nonetheless, such partial genotyping test is useful for risk stratification to refer HPV16/18-positive women to immediate colposcopy. Combination with other triage tests or simply expanding the genotypes to HPV16/18/31/33/45/52/58 would complement the lower NPV because of missed CIN2+ cases associated with other hrHPV types and reach comparable performance with cytology. Of note, expanding genotyping would require consideration according to population burden and attributable fraction in high-grade lesions in specific context, especially when younger birth cohorts enter into the screening program, for settings with HPV national immunization available now (37, 38).

A challenge for incorporating prior HPV results to manage HPV women is the availability of a linked digitalized screening registry and invitation system to enable individual linkage of results from different rounds of screening, which has not yet been established in many settings. Considering surveillance and quality assurance are important indicators in both organized screening program and postvaccination impact evaluation and also require a robust electronic health record system, the prospect of accounting for previous HPV results might be desirable (39). Furthermore, with the rapid development of technology and accumulation of evidence, management would be further improved with additional incorporation of individual demographics, vaccination, medical history, and genetic risk factors (40). This might facilitate to revolutionize prevention by recommending clinical action for either immediate referral or routine screening and minimize amount of recalls in short intervals.

The main strength of our study lies in that it is the longest follow-up cohort in P.R. China with four visits at approximately 5-year interval within 15 years. All women not reaching the CIN2+ endpoint were recalled at next screening round, which allowed us to have more detailed screening and histology data for sequential rounds of evaluation. However, the findings in this report are subject to four limitations. First, the relatively small sample size limits the possibility of using more stringent endpoint on CIN3+. Second, the threshold to determine the early or routine recall is better to be considered with cumulative risks within a given period. This research has only been able to consider the immediate prevalence of CIN2+ due to the absence of retest visits. A more comprehensive evaluation is warranted in longitudinal cohorts with larger sample size and retest visits. Third, the relatively narrow age range restricted more specified stratification by age group. While considering that the starting age of women for national cervical cancer screening program is 35 years old in P.R. China, these findings are supposed to reflect the real-life experience of current screening practice (41). Fourth, colposcopy and four-quadrant biopsies were performed at baseline, which are close to the full baseline disease ascertainment and associated with bias in comparison of CIN2+ prevalence across rounds. Notwithstanding the acknowledged limitations, this work complements the limited evidence regarding triage performance after sequential rounds of screening and clinical value of incorporating prior HPV results in subsequent screening rounds.

In conclusion, triage performance using cytology and genotyping based on routinely adopted clinical action remain suitable to manage HPV women after four rounds of screening. Taking prior HPV results into account helps to distinguish a persistent infection at higher risk of progression from transient infection that may probably clear and optimize the triaging algorithm in a more efficient way.

No disclosures were reported.

X.-Q. Xu: Conceptualization, formal analysis, visualization, writing–original draft, writing–review and editing. R. Rezhake: Writing–review and editing. S.-Y. Hu: Methodology, writing–review and editing. F. Chen: Methodology. X. Zhang: Methodology, writing–review and editing. Q.-J. Pan: Methodology, writing–review and editing. W.-H. Zhang: Methodology, writing–review and editing. J.-F. Ma: Methodology. Y.-L. Qiao: Writing–review and editing. F.-H. Zhao: Conceptualization, formal analysis, visualization, writing–original draft, writing–review and editing. M. Cruickshank: Writing–review and editing.

We acknowledge the significant contributions of all the investigators in CICAMS and local health providers for their efforts in conducting the study and women who participated in this study. This work was supported by National Natural Science Foundation of China (81761128006) and National Key R&D Program of China (2018YFC1315504).

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1.
Kitchener
HC
,
Gilham
C
,
Sargent
A
,
Bailey
A
,
Albrow
R
,
Roberts
C
, et al
A comparison of HPV DNA testing and liquid based cytology over three rounds of primary cervical screening: extended follow up in the ARTISTIC trial
.
Eur J Cancer
2011
;
47
:
864
71
.
2.
Rijkaart
DC
,
Berkhof
J
,
Rozendaal
L
,
van Kemenade
FJ
,
Bulkmans
NW
,
Heideman
DA
, et al
Human papillomavirus testing for the detection of high-grade cervical intraepithelial neoplasia and cancer: final results of the POBASCAM randomised controlled trial
.
Lancet Oncol
2012
;
13
:
78
88
.
3.
Ogilvie
GS
,
Krajden
M
,
van Niekerk
D
,
Smith
LW
,
Cook
D
,
Ceballos
K
, 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
8
.
4.
World Health Organization
. 
Draft global strategy towards the elimination of cervical cancer as a public health problem
.
Available from
: https://www.who.int/docs/default-source/cervical-cancer/cerv-cancer-elimn-strategy-16dec-12pm.pdf.
5.
Rebolj
M
,
Njor
SH
,
Lynge
E
. 
Restriction of human papillomavirus DNA testing in primary cervical screening to women above age 30: systematic review
.
Eur J Cancer Prev
2012
;
21
:
73
81
.
6.
Rebolj
M
,
Rimmer
J
,
Denton
K
,
Tidy
J
,
Mathews
C
,
Ellis
K
, et al
Primary cervical screening with high risk human papillomavirus testing: observational study
.
BMJ
2019
;
364
:
l240
.
7.
Horn
J
,
Denecke
A
,
Luyten
A
,
Rothe
B
,
Reinecke-Lüthge
A
,
Mikolajczyk
R
, et al
Reduction of cervical cancer incidence within a primary HPV screening pilot project (WOLPHSCREEN) in Wolfsburg, Germany
.
Br J Cancer
2019
;
120
:
1015
22
.
8.
Veldhuijzen
NJ
,
Polman
NJ
,
Snijders
PJF
,
Meijer
C
,
Berkhof
J
. 
Stratifying HPV-positive women for CIN3+ risk after one and two rounds of HPV-based screening
.
Int J Cancer
2017
;
141
:
1551
60
.
9.
Castle
PE
,
Rodríguez
AC
,
Burk
RD
,
Herrero
R
,
Wacholder
S
,
Alfaro
M
, et al
Short term persistence of human papillomavirus and risk of cervical precancer and cancer: population based cohort study
.
BMJ
2009
;
339
:
b2569
.
10.
Kjær
SK
,
Frederiksen
K
,
Munk
C
,
Iftner
T
. 
Long-term absolute risk of cervical intraepithelial neoplasia grade 3 or worse following human papillomavirus infection: role of persistence
.
J Natl Cancer Inst
2010
;
102
:
1478
88
.
11.
Polman
NJ
,
Veldhuijzen
NJ
,
Heideman
DAM
,
Snijders
PJF
,
Meijer
C
,
Berkhof
J
. 
Management of HPV-positive women in cervical screening using results from two consecutive screening rounds
.
Int J Cancer
2019
;
144
:
2339
46
.
12.
Dijkstra
MG
,
van Zummeren
M
,
Rozendaal
L
,
van Kemenade
FJ
,
Helmerhorst
TJ
,
Snijders
PJ
, et al
Safety of extending screening intervals beyond five years in cervical screening programmes with testing for high risk human papillomavirus: 14 year follow-up of population based randomised cohort in the Netherlands
.
BMJ
2016
;
355
:
i4924
.
13.
Cancer Council Australia Cervical Cancer Screening Guidelines Working Party
. 
National cervical screening program: guidelines for the management of screen-detected abnormalities, screening in specific populations and investigation of abnormal vaginal bleeding
.
Available from
: https://wiki.cancer.org.au/australia/Guidelines:Cervical_cancer/Screening.
14.
Zhang
L
,
Xu
XQ
,
Hu
SY
,
Chen
F
,
Zhang
X
,
Pan
QJ
, et al
Durability of clinical performance afforded by self-collected HPV testing: a 15-year cohort study in China
.
Gynecol Oncol
2018
;
151
:
221
8
.
15.
Rezhake
R
,
Hu
SY
,
Zhao
S
,
Xu
XQ
,
Zhao
XL
,
Zhang
L
, et al
Eight-type human papillomavirus E6/E7 oncoprotein detection as a novel and promising triage strategy for managing HPV-positive women
.
Int J Cancer
2019
;
144
:
34
42
.
16.
Dong
L
,
Wang
MZ
,
Zhao
XL
,
Feng
RM
,
Hu
SY
,
Zhang
Q
, et al
Human papillomavirus viral load as a useful triage tool for non-16/18 high-risk human papillomavirus positive women: a prospective screening cohort study
.
Gynecol Oncol
2018
;
148
:
103
10
.
17.
Zhang
Q
,
Dong
L
,
Hu
S
,
Feng
R
,
Zhang
X
,
Pan
Q
, et al
Risk stratification and long-term risk prediction of E6 oncoprotein in a prospective screening cohort in China
.
Int J Cancer
2017
;
141
:
1110
9
.
18.
Belinson
J
,
Qiao
YL
,
Pretorius
R
,
Zhang
WH
,
Elson
P
,
Li
L
, et al
Shanxi province cervical cancer screening study: a cross-sectional comparative trial of multiple techniques to detect cervical neoplasia
.
Gynecol Oncol
2001
;
83
:
439
44
.
19.
Holt
HK
,
Zhang
L
,
Zhao
FH
,
Hu
SY
,
Zhao
XL
,
Zhang
X
, et al
Evaluation of multiple primary and combination screening strategies in postmenopausal women for detection of cervical cancer in China
.
Int J Cancer
2017
;
140
:
544
54
.
20.
Perkins
RB
,
Guido
RS
,
Castle
PE
,
Chelmow
D
,
Einstein
MH
,
Garcia
F
, et al
2019 ASCCP risk-based management consensus guidelines for abnormal cervical cancer screening tests and cancer precursors
.
J Low Genit Tract Dis
2020
;
24
:
102
31
.
21.
Hashim
D
,
Engesæter
B
,
Baadstrand Skare
G
,
Castle
PE
,
Bjørge
T
,
Tropé
A
, et al
Real-world data on cervical cancer risk stratification by cytology and HPV genotype to inform the management of HPV-positive women in routine cervical screening
.
Br J Cancer
2020
;
122
:
1715
23
.
22.
Arbyn
M
,
Rezhake
R
,
Yuill
S
,
Canfell
K
. 
Triage of HPV-positive women in Norway using cytology, HPV16/18 genotyping and HPV persistence
.
Br J Cancer
2020
;
122
:
1577
9
.
23.
Dijkstra
MG
,
van Niekerk
D
,
Rijkaart
DC
,
van Kemenade
FJ
,
Heideman
DA
,
Snijders
PJ
, et al
Primary hrHPV DNA testing in cervical cancer screening: how to manage screen-positive women? A POBASCAM trial substudy
.
Cancer Epidemiol Biomarkers Prev
2014
;
23
:
55
63
.
24.
Arbyn
M
,
Xu
L
,
Verdoodt
F
,
Cuzick
J
,
Szarewski
A
,
Belinson
JL
, et al
Genotyping for human papillomavirus types 16 and 18 in women with minor cervical lesions: a systematic review and meta-analysis
.
Ann Intern Med
2017
;
166
:
118
27
.
25.
Arbyn
M
,
Roelens
J
,
Martin-Hirsch
P
,
Leeson
S
,
Wentzensen
N
. 
Use of HC2 to triage women with borderline and mild dyskaryosis in the UK
.
Br J Cancer
2011
;
105
:
877
80
.
26.
Ostör
AG
. 
Natural history of cervical intraepithelial neoplasia: a critical review
.
Int J Gynecol Pathol
1993
;
12
:
186
92
.
27.
Pinder
LF
,
Parham
GP
,
Basu
P
,
Muwonge
R
,
Lucas
E
,
Nyambe
N
, et al
Thermal ablation versus cryotherapy or loop excision to treat women positive for cervical precancer on visual inspection with acetic acid test: pilot phase of a randomised controlled trial
.
Lancet Oncol
2020
;
21
:
175
84
.
28.
Zhao
XL
,
Xu
XQ
,
Duan
XZ
,
Rezhake
R
,
Hu
SY
,
Wang
Y
, et al
Comparative performance evaluation of different HPV tests and triaging strategies using self-samples and feasibility assessment of thermal ablation in ‘colposcopy and treat’ approach: a population-based study in rural China
.
Int J Cancer
2020
;
147
:
1275
85
.
29.
Chen
T
,
Jansen
L
,
Gondos
A
,
Emrich
K
,
Holleczek
B
,
Luttmann
S
, et al
Survival of cervical cancer patients in Germany in the early 21st century: a period analysis by age, histology, and stage
.
Acta Oncol
2012
;
51
:
915
21
.
30.
Sahlgren
H
,
Elfström
KM
,
Lamin
H
,
Carlsten-Thor
A
,
Eklund
C
,
Dillner
J
, et al
Colposcopic and histopathologic evaluation of women with HPV persistence exiting an organized screening program
.
Am J Obstet Gynecol
2020
;
222
:
253
.
31.
Rodríguez
AC
,
Schiffman
M
,
Herrero
R
,
Wacholder
S
,
Hildesheim
A
,
Castle
PE
, et al
Rapid clearance of human papillomavirus and implications for clinical focus on persistent infections
.
J Natl Cancer Inst
2008
;
100
:
513
7
.
32.
Plummer
M
,
Schiffman
M
,
Castle
PE
,
Maucort-Boulch
D
,
Wheeler
CM
. 
A 2-year prospective study of human papillomavirus persistence among women with a cytological diagnosis of atypical squamous cells of undetermined significance or low-grade squamous intraepithelial lesion
.
J Infect Dis
2007
;
195
:
1582
9
.
33.
Schiffman
M
,
Wentzensen
N
,
Perkins
RB
,
Guido
RS
. 
An introduction to the 2019 ASCCP risk-based management consensus guidelines
.
J Low Genit Tract Dis
2020
;
24
:
87
9
.
34.
Katki
HA
,
Schiffman
M
,
Castle
PE
,
Fetterman
B
,
Poitras
NE
,
Lorey
T
, et al
Benchmarking CIN 3+ risk as the basis for incorporating HPV and Pap cotesting into cervical screening and management guidelines
.
J Low Genit Tract Dis
2013
;
17
:
S28
35
.
35.
Rezhake
R
,
Chen
F
,
Hu
SY
,
Zhao
XL
,
Zhang
X
,
Cao
J
, et al
Triage options to manage high-risk human papillomavirus-positive women: a population-based cross-sectional study from rural China
.
Int J Cancer
2020
;
147
:
2053
64
.
36.
Castle
PE
,
Kinney
WK
,
Xue
X
,
Cheung
LC
,
Gage
JC
,
Poitras
NE
, et al
Role of screening history in clinical meaning and optimal management of positive cervical screening results
.
J Natl Cancer Inst
2019
;
111
:
820
7
.
37.
de Sanjose
S
,
Quint
WG
,
Alemany
L
,
Geraets
DT
,
Klaustermeier
JE
,
Lloveras
B
, et al
Human papillomavirus genotype attribution in invasive cervical cancer: a retrospective cross-sectional worldwide study
.
Lancet Oncol
2010
;
11
:
1048
56
.
38.
Cuzick
J
,
Wheeler
C
. 
Need for expanded HPV genotyping for cervical screening
.
Papillomavirus Res
2016
;
2
:
112
5
.
39.
World Health Organization
. 
Screening and treatment of cervical pre-cancer.
Comprehensive cervical cancer control. A guide to essential practice
. 2nd ed.
Geneva, Switzerland
; 
2014
. p.
129
64
.
40.
Rothberg
MB
,
Hu
B
,
Lipold
L
,
Schramm
S
,
Jin
XW
,
Sikon
A
, et al
A risk prediction model to allow personalized screening for cervical cancer
.
Cancer Causes Control
2018
;
29
:
297
304
.
41.
Luo
XM
,
Song
L
,
Wu
JL
,
Liu
Y
,
Di
JL
,
Song
B
, et al
Analysis of the reported data of national rural cervical cancer screening project from 2012 to 2013, China
.
Zhonghua Yu Fang Yi Xue Za Zhi
2016
;
50
:
346
50
.