Abstract
Background
Persistent infection with high-risk human papillomavirus (HR-HPV) is a necessary cause of cervical cancer and its precancerous lesions. This study aimed to identify factors associated with HR-HPV persistence in Chinese women.
Methods
This study is a population-based, nationwide, multi-center prospective cohort study initiated in 2017, and a total of 10,481 women undergoing cervical cancer screening were enrolled. A subset of 1,684 women who tested HR-HPV positive at baseline were included in the analysis. The results of their HPV testing at baseline and during three follow-up visits (2018–2020) were used to assess 1-, 2-, and 3-year HR-HPV persistence. Variables with statistically significant associations in univariate analyses, expressed as odds ratios (ORs) with 95% confidence intervals (CIs), were subsequently entered into a stepwise multivariate logistic regression model to identify independent predictors of HR-HPV persistence.
Results
The mean age of 1,684 HPV-positive women at baseline was 47.6 ± 9.5 (interquartile range (IQR) = 42–54) years. The most prevalent genotypes at baseline were HPV52 (28.3%), HPV16 (20.2%), HPV58 (17.8%), HPV33 (6.7%), and HPV18 (6.2%). Overall HR-HPV persistence rates declined over time were 66.8% at 1 year, 48.3% at 2 years, and 39.9% at 3 years. Infection with HPV33 (OR = 2.40, 95% CI: 1.42–4.01), HPV52 (OR = 1.95, 95% CI: 1.45–2.64), or HPV58 (OR = 2.01, 95% CI: 1.40–2.88), as well as postmenopausal status (OR = 2.84, 95% CI: 2.17–3.72), were significantly associated with 3-year persistence, and alcohol consumption was associated with a reduced risk of persistence (OR = 0.49, 95% CI: 0.30–0.79). Furthermore, HPV16 was the most frequently detected genotype in cervical intraepithelial neoplasia grades 2 and worse (CIN2+), indicating it plays a predominant role in chronically pathogenic of high cervical disease.
Conclusion
These findings underscore the importance of genotype-specific and host-related factors in HR-HPV persistence and support the implementation of tailored cervical cancer screening strategies. Women infected with HPV33, HPV52, or HPV58, along with HPV16 may require closer long-term monitoring to prevent progression to high-grade cervical lesions.
Supplementary Information
The online version contains supplementary material available at 10.1186/s12985-025-03047-4.
Keywords: HR-HPV, Persistence, Risk factors, Cervical cancer, Cohort study, Chinese women
Background
Persistent infection with high-risk human papillomavirus (HR-HPV) is a necessary cause of cervical cancer and its precancerous lesions, representing a significant global public health burden [1]. While HPV vaccination and screening programs have substantially reduced disease incidence [2, 3], critical gaps remain— particularly in populations with incomplete vaccine coverage or limited access to healthcare services. Among women already infected with HR-HPV, understanding the factors that influence viral persistence is critical for refining clinical management strategies. Identifying these factors can help determining which individuals require closer monitoring, such as referral for colposcopy or more intensive follow-up, enabling early detection and timely intervention to prevent progression to cervical cancer.
HR-HPV persistence is influenced by a complex interplay of host, viral, and environmental factors, many of which remain poorly understood or inconclusive [4]. Previous studies have suggested that sexual behavior, immune status, co-infections, and HPV genotype play important roles in HR-HPV persistence [5–7]; however, these associations often vary by population. In particular, host factors, such as genetic susceptibility and immune response variability, as well as environmental influences, including lifestyle and healthcare access, may contribute to these discrepancies. Studies conducted in both Western and non-Western settings have highlighted regional differences in persistence patterns, suggesting that genetic and environmental factors significantly shape infection outcomes [4]. For example, research from Denmark and China has demonstrated divergent distributions of persistent HPV genotypes: HPV16, 58, and 31 are more commonly persistent in Denmark, whereas HPV52, 58, and 68 are predominant in certain regions of China [8, 9]. However, despite these insights, a comprehensive and population-specific understanding of the determinants of HR-HPV persistence in the Chinese population remains lacking.
In China, HR-HPV persistence continues to be a pressing public health concern, even as vaccination and screening efforts [10, 11]. Disparities in healthcare access, variations in screening practices, and cultural differences may contribute to HPV persistence, particularly when comparing rural and urban populations [12–14]. Moreover, existing research on HR-HPV persistence in Chinese women has largely focused on specific regions or individual genotypes, resulting in a fragmented understanding of the broader epidemiological and clinical determinants of viral persistence [15, 16]. There is a critical need for large-scale, nationwide studies to bridge these knowledge gaps and provide a more comprehensive understanding of HR-HPV dynamics in China.
This research was based on a nationwide, multi-center cervical cancer screening cohort study and aimed to identify factors associated with cervicogenital HR-HPV persistence among Chinese women. By improving our understanding of the determinants of cervicogenital HR-HPV persistence, these findings will contribute to the refinement cervical cancer prevention strategies and inform public health interventions aimed at reducing the burden of HR-HPV-related cervicogenital diseases.
Methods
Study design and participants
This study was conducted as part of a population-based, nation-wide, multi-center prospective cohort study on cervical cancer screening in China. The screening cohort was established in year 2017, with participants recruited from Sichuan, Shanxi and Henan provinces in China using a convenience sampling approach between April to October (The cohort timeline is shown in Table S3 of the supplementary materials). The inclusion criteria at recruitment were as follows: (i) women aged 21–64 years; (ii) having an intact cervix; (iii) physically and mentally competent to undergo cervical cancer screening process, including colposcopy and biopsy if necessary; (iv) possessing full legal capacity and providing signed informed consent. The exclusion criteria were: (i) previous cervical cancer diagnosis and treatment; (ii) a history of cervical surgery or pelvic radiation therapy; (iii) having undergone a hysterectomy; (iv) current or suspected pregnancy, or within 8 weeks of delivery; (v) recent sexual activity or use of vaginal topical drugs, including lotion and vaginal contraceptives, within 48 h before screening; (vi) physically and mentally incompetent to undergo cervical cancer screening or unable to provide informed consent.
Ethics approvals were obtained from the Clinical Trial Ethics Review Committee, West China Second University Hospital, Sichuan University (approval number: Q2017003); the Medical Ethics Committee of Henan Cancer Hospital (approval number: 2017009); and the Medical Ethics Committee of Changzhi Medical College (approval number: 2017002). The study was registered at the Chinese Clinical Trial Registry Center of the World Health Organization International Clinical Trials Registry Platform (Registration number ChiCRT2200055287). All participants provided written informed consent in every visits.
Baseline screening and follow-up (2018–2020) procedures
Prior to screening and after obtaining informed consent, all participants completed a structured questionnaire to assess potential risk factors, including age, marital status, education level, sexual behavior, menstrual and reproductive health history, contraceptive use, smoking and alcohol consumption (The questionnaire shown in the supplementary materials). Following this, gynecological examinations were performed by local physicians at the study clinics. Cervical and vaginal specimens were also collected by physicians for HPV and cytological testing.
Participants who tested positive for HPV16 and/or 18 at baseline were directly referred to colposcopy. Women who tested positive for other HR-HPV types or who were HPV-negative at baseline underwent cytological testing. Those with cytology results ≥ ASC-US were also referred for colposcopy. Any visible or suspicious abnormalities identified under colposcopy were biopsied. Women with biopsy-confirmed abnormality ≥ Cervical Intraepithelial Neoplasia grade 2 (≥ CIN2) were referred for treatment and clinical follow-up (Fig. 1).
Fig. 1.
The procedure of cervical cancer screening and follow-up. * Single or multiple infections were not differentiated; a total of 1,684 women were infected with any of the 14 h-HPV types. **Women diagnosed with ≥ CIN2 will be referred for treatment and followed up according to clinical guidelines. All treatment- related costs will be covered by their insurance. Other HR-HPV: HPV31, 33, 35, 39, 45, 51, 52, 56, 58, 59, 66 and 68; ASC-US: atypical squamous cells of undetermined significance; NILM: negative for intraepithelial lesion or malignancy; CIN2: cervical intraepithelial neoplasia 2
Women who were tested positive for any HR-HPV and/or had cytology results ≥ ASC-US were followed up annually from 2018 to 2020. In year 2020, all participants were recalled for follow-up screening, regardless of their baseline screening results. During follow-up, each woman underwent HPV testing and cytological examination. Women who met any of the following criteria were referred for colposcopy: ①Baseline positivity for HR-HPV types HPV16, HPV18, HPV33, HPV52, or HPV58; ②Cytological findings of ASC-US or greater at follow-up screenings. Any visible or suspicious abnormalities identified under colposcopy were biopsied, and women with biopsy-confirmed abnormality of ≥ CIN2 were referred to treatment and clinical follow-up according to clinical management guidelines. Participants with negative HPV results, normal cytology, or biopsy-confirmed lesions < CIN2 were recommended for routine cervical cancer screening (Fig. 1).
HPV testing and cytology
HPV testing was performed using the TELLGEN HPV 5 + 9® assay, which is based on fluorescent polymerase chain reaction (PCR) technology. This assay detects 14 h-HPV types. Among them, HPV16, 18, 33, 52, and 58 are individually identified, while the remaining types-HPV31, 35, 39, 45, 51, 56, 59, 66, and 68 are detected as a pooled group.
Cervical cytology was conducted using the ThinPrep Cytologic Test (TCT). Results were interpreted and reported according to the Bethesda System (TBS) classification.
Definition of followed-up and loss to follow-up
In our study, loss to follow-up at the 1-year follow-up time point was defined as participants who did not come back for the follow-up screening in 2018. For the 2-year follow-up, loss to follow-up included individuals who missed follow-up visits in either 2018 and/or 2019. For the 3-year follow-up, participants were considered lost to follow-up if they missed at least one follow-up visit between 2018 and 2020. For follow-up periods longer than one year, only participants with complete and uninterrupted follow-up across all consecutive years were classified as having completed follow-up.
Definition of HR-HPV persistence
HR-HPV persistence was defined as the repeated detection of the same HR-HPV type at follow-up visits after an initial HPV positive result at baseline. For participants infected with pooled HR-HPV types—specifically HPV31, 35, 39, 45, 51, 56, 59, 66, and 68 (hereafter referred to as “other 9 HR-HPV types”)—persistence was defined as repeated detection of any genotype within this pooled group, regardless of whether the specific type varied between visits, since individual genotypes within the pool could not be distinguished by the assay.
The type-specific HR-HPV persistence rate was calculated by dividing the number of women with persistent infection of a specific HR-HPV genotype by the total number of women who tested positive for that genotype at baseline. For the “other 9 HR-HPV types”, the persistence rate was calculated as the number of women with persistence of any genotype within the pooled group divided by the number of women who tested positive for any of these genotypes at baseline. To determine the overall HR-HPV persistence rate, the numerator was the number of women with persistence of at least one type-specific HR-HPV or any of the “other 9 HR-HPV types” at follow-ups, and denominator was the total number of women who tested positive for HR-HPV at baseline.
Additionally, participants with HR-HPV persistence were categorized as having 1-year, 2-year, or 3-year persistence, based on the length of the follow-up period.
Data collection and quality control
At each study site, two trained data entry clerks independently performed double data entry from paper-based records into a computer-based database using Microsoft Access. Upon completion, the databases were sent to Sichuan University (SCU) for validation using EpiData3.0. Any discrepancies identified by researchers from SCU between the two datasets were returned to the local clerks for adjudication until the databases agreed. As final check, one of databases was chosen and underwent a final consistency check. Logic errors (e.g. a woman reported had never been pregnant yet had more than one parity) were again reported back to the local sites. Local collaborators then contacted the respective participants using the contact information collected in the baseline risk factor questionnaire. After verification, the revised data were returned to SCU for final analysis.
Statistical analysis
Categorical variables were summarized as proportions (%). Differences in participants’ baseline characteristics and changes across the three follow-up visits were assessed using the Chi-square test or Fisher’s exact test, as appropriate. Trend analyses were conducted using the Cochran-Mantel-Haenszel test to examine linear trends across ordered groups. In univariate analysis, odds ratios (ORs) with 95% confidence intervals (CIs) were calculated. Variables that were statistically significant in the univariate analyses were entered into a multivariate logistic regression model using stepwise regression method to identify independent predictors of HR-HPV persistence, while minimizing the risk of overfitting. All statistical tests were two-sided, and a P value of less than 0.05 was considered statistically significant. Analyses were conducted using SPSS (version 27.0, IBM Corp., Armonk, NY, USA).
Results
Demographic characteristics of the cohort and baseline HR-HPV positive participants
At baseline, a total of 10,481 women were enrolled in the cohort, with a mean age of 46.51(± 9.17, interquartile range (IQR) = 41–53) years. The majority were married (98.1% (10,287/10,481)) and 78.6% (8,329/10,481) had attained an education level below senior high school. The overall HR-HPV prevalence was 16.1% (1,684/10,481). The most frequently detected genotype was HPV52 (4.5%), followed by HPV16 (3.3%), HPV58 (2.9%), HPV33 (1.1%), and HPV18 (1.0%) (Table 1).
Table 1.
Baseline characteristics of the cohort and comparison between those retained and lost to follow-up
| Baseline characteristics |
Baseline | 1-year follow-up | 2-year follow-up | 3-year follow-up | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Total (n=10481) |
HR-HPV+ (n=1684) | Followed-up (n=1319) |
Lost to follow-up (n=270) |
P | Followed-up (n=1102) |
Lost to follow-up (n=459) |
P | Followed-up (n=1003) |
Loss to follow-up (n=532) |
P | ||||
| Age, y | <0.001 | <0.001 | <0.001 | |||||||||||
| 20–29 | 563(5.4) | 88(5.2) | 53(4.0) | 33(12.2) | 36(3.3) | 49(10.7) | 29(2.9) | 55(10.3) | ||||||
| 30–39 | 1724(16.4) | 268(15.9) | 204(15.5) | 45(16.7) | 169(15.3) | 75(16.3) | 161(16.1) | 81(15.2) | ||||||
| 40–49 | 4012(38.3) | 536(31.8) | 410(31.1) | 91(33.7) | 344(31.2) | 151(32.9) | 321(32.0) | 169(31.8) | ||||||
| ≥50 | 4182(39.9) | 792(47.0) | 652 (49.4) | 101(37.4) | 553(50.2) | 184(40.1) | 492(49.1) | 227(42.7) | ||||||
| Marital status | 0.283 | 0.950 | 0.681 | |||||||||||
| Married | 10287(98.1) | 1643(97.6) | 1285(97.4) | 266(98.5) | 1075(97.5) | 448(97.6) | 980(97.7) | 518(97.4) | ||||||
| Other | 194(1.9) | 41(2.4) | 34(2.6) | 4(1.5) | 27(2.5) | 11(2.4) | 23(2.3) | 14(2.6) | ||||||
| Education | 0.110 | 0.093 | 0.019 | |||||||||||
| Primary and below | 3842(36.7) | 654(38.8) | 510(38.7) | 112(41.5) | 414(37.6) | 197(42.9) | 368(36.7) | 234(44.0) | ||||||
| Junior high school | 4397(42.0) | 699(41.5) | 563(42.7) | 95(35.2) | 476(43.2) | 167(36.4) | 438(43.7) | 191(35.9) | ||||||
| Senior high school | 1388(13.2) | 219(13.0) | 160(12.1) | 41(15.2) | 138(12.5) | 63(13.7) | 129(12.9) | 71(13.3) | ||||||
| University and above | 854(8.1) | 112(6.7) | 86(6.5) | 22(8.1) | 74(6.7) | 32(7.0) | 68(6.8) | 36(6.8) | ||||||
| Age at menarchea, y | 0.657 | 0.914 | 0.767 | |||||||||||
| <12 | 112(1.1) | 14(0.8) | 11(0.8) | 3(1.1) | 9(0.8) | 4(0.9) | 9(0.9) | 4(0.8) | ||||||
| ≥12 | 10366(98.9) | 1670(99.2) | 1308(99.2) | 267(98.9) | 1093(99.2) | 455(99.1) | 994(99.1) | 528(99.2) | ||||||
| Menopause | 0.015 | 0.019 | 0.389 | |||||||||||
| No | 6816(65.0) | 976(58.0) | 739(56.0) | 173(64.1) | 613(55.6) | 285(62.1) | 571(56.9) | 315(59.2) | ||||||
| Yes | 3665(35.0) | 708(42.0) | 580(44.0) | 97(35.9) | 489(44.4) | 174(37.9) | 432(43.1) | 217(40.8) | ||||||
| Contraceptive measurea | <0.001 | <0.001 | <0.001 | |||||||||||
| No | 3510(33.5) | 559(33.2) | 419(31.8) | 113(41.9) | 350(31.8) | 180(39.2) | 304(30.3) | 216(40.6) | ||||||
| Condom | 469(4.5) | 53(3.1) | 45(3.4) | 7(2.6) | 37(3.4) | 14(3.1) | 36(3.6) | 14(2.6) | ||||||
| Intrauterine device | 1947(18.6) | 308(18.3) | 225(17.1) | 59(21.9) | 177(16.1) | 101(22.0) | 162(16.2) | 113(21.2) | ||||||
| Sterilization | 4340(41.4) | 729(43.3) | 602(45.6) | 85(31.5) | 518(47.0) | 151(32.9) | 483(48.2) | 174(32.7) | ||||||
| Other | 209(2.0) | 35(2.1) | 28(2.1) | 6(2.2) | 20(1.8) | 13(2.8) | 18(1.8) | 15(2.8) | ||||||
| No. of sexual partners in 5 yearsa | 0.031 | 0.002 | <0.001 | |||||||||||
| 0–1 | 10339(98.8) | 1646(98.0) | 1293(98.3) | 260(96.3) | 1084(98.7) | 442(96.3) | 989(98.9) | 512(96.4) | ||||||
| ≥2 | 121(1.2) | 34(2.0) | 22(1.7) | 10(3.7) | 14(1.3) | 17(3.7) | 11(1.1) | 19(3.6) | ||||||
| No. of pregnanciesa | 0.628 | 0.208 | 0.198 | |||||||||||
| 0–2 | 4631(44.2) | 743(44.1) | 583(44.2) | 115(42.6) | 492(44.6) | 189(41.2) | 451(45.0) | 221(41.5) | ||||||
| ≥3 | 5848(55.8) | 941(55.9) | 736(55.8) | 155(57.4) | 610(55.4) | 270(58.8) | 552(55.0) | 311(58.5) | ||||||
| Age of first pregnancya, y | 0.002 | <0.001 | 0.016 | |||||||||||
| ≤20 | 2148(20.6) | 370(22.2) | 269(20.6) | 81(20.3) | 213(19.5) | 129(28.3) | 197(19.9) | 139(26.3) | ||||||
| 21–25 | 7283(70.0) | 1159(69.4) | 927(70.9) | 166(62.2) | 780(71.5) | 295(64.7) | 708(71.4) | 349(66.0) | ||||||
| ≥26 | 974(9.4) | 140(8.4) | 112(8.6) | 20(7.5) | 98(9.0) | 32(7.0) | 87(8.8) | 41(7.8) | ||||||
| No. of productionsa | 0.215 | 0.008 | 0.023 | |||||||||||
| 1–2 | 8245(79.4) | 1251(75.0) | 971(74.3) | 208(77.9) | 795(72.9) | 362(79.4) | 727(73.4) | 416(78.6) | ||||||
| ≥3 | 2137(20.6) | 417(25.0) | 336(25.7) | 59(22.1) | 295(27.1) | 94(20.6) | 264(26.6) | 113(21.4) | ||||||
| No. of live birthsa | 0.210 | 0.042 | 0.262 | |||||||||||
| 1–2 | 8844(85.2) | 1367(82.0) | 1064(81.4) | 226(84.6) | 880(80.7) | 388(85.1) | 807(81.4) | 443(83.7) | ||||||
| ≥3 | 1536(14.8) | 301(18.0) | 243(18.6) | 41(15.4) | 210(19.3) | 68(14.9) | 184(18.6) | 86(16.3) | ||||||
| Age of first deliverya, y | 0.006 | <0.001 | 0.010 | |||||||||||
| ≤20 | 1034(10.0) | 179(10.7) | 129(9.9) | 44(16.5) | 99(9.1) | 71(15.6) | 92(9.3) | 75(14.2) | ||||||
| 21–25 | 7708(74.3) | 1251(75.0) | 988(75.6) | 191(71.5) | 828(76.0) | 330(72.4) | 752(75.9) | 388(73.3) | ||||||
| ≥26 | 1639(15.8) | 238(14.3) | 190(14.5) | 32(12.0) | 163(15.0) | 55(12.1) | 147(14.8) | 66(12.5) | ||||||
| Smokinga | 0.979 | 0.914 | >0.999 | |||||||||||
| No | 10406(99.3) | 1672(99.3) | 1311(99.4) | 269(99.6) | 1095(99.4) | 457(99.6) | 997(99.4) | 529(99.4) | ||||||
| Yes | 74(0.7) | 12(0.7) | 8(0.6) | 1(0.4) | 7(0.6) | 2(0.4) | 3(0.6) | |||||||
| Alcohol consumption | 0.932 | 0.371 | 0.352 | |||||||||||
| No | 9369(89.4) | 1504(89.3) | 1175(89.1) | 29(10.7) | 987(89.6) | 404(88.0) | 898(89.5) | 468(88.0) | ||||||
| Yes | 1112(10.6) | 180(10.7) | 144(10.9) | 241(89.3) | 115(10.4) | 55(12.0) | 105(10.5) | 64(12.0) | ||||||
| HPV16 | 0.402 | 0.792 | 0.448 | |||||||||||
| No | 10140(96.7) | 1343(79.8) | 1093(82.9) | 218(80.7) | 916(83.1) | 379(82.6) | 841(83.8) | 438(82.3) | ||||||
| Yes | 341(3.3) | 341(20.2) | 226(17.1) | 52(19.3) | 186(16.9) | 80(17.4) | 162(16.2) | 94(17.7) | ||||||
| HPV18 | 0.335 | 0.487 | 0.633 | |||||||||||
| No | 10376(99.0) | 1579(93.8) | 1246(94.5) | 251(93.0) | 1040(94.4) | 429(93.5) | 945(94.2) | 498(93.6) | ||||||
| Yes | 105(1.0) | 105(6.2) | 73(5.5) | 19(7.0) | 62(5.6) | 30(6.5) | 58(5.8) | 34(6.4) | ||||||
| HPV33 | 0.821 | 0.404 | 0.216 | |||||||||||
| No | 10368(98.9) | 1571(93.3) | 1231(93.3) | 253(93.7) | 1027(93.2) | 433(94.3) | 934(93.1) | 504(94.7) | ||||||
| Yes | 113(1.1) | 113(6.7) | 88(6.7) | 17(6.3) | 75(6.8) | 26(5.7) | 69(6.9) | 28(5.3) | ||||||
| HPV52 | 0.524 | 0.561 | 0.622 | |||||||||||
| No | 10005(95.5) | 1208(71.7) | 937(71.0) | 197(73.0) | 781(70.9) | 332(72.3) | 719(71.7) | 375(70.5) | ||||||
| Yes | 476(4.5) | 476(28.3) | 382(29.0) | 73(27.0) | 321(29.1) | 127(27.7) | 284(28.3) | 157(29.5) | ||||||
| HPV58 | 0.438 | 0.283 | 0.126 | |||||||||||
| No | 10182(97.1) | 1385(82.2) | 1091(82.7) | 218(80.7) | 918 (83.3) | 372(81.0) | 840(83.7) | 429(80.6) | ||||||
| Yes | 299(2.9) | 299(17.8) | 228(17.3) | 52(19.3) | 184(16.7) | 87(19.0) | 163(16.3) | 103(19.4) | ||||||
| Other 9 HR-HPVb | 0.321 | 0.065 | 0.074 | |||||||||||
| No | 9658(92.1) | 861(51.1) | 650(49.3) | 142(52.6) | 527(47.8) | 248(53.2) | 478(47.7) | 279(52.4) | ||||||
| Yes | 823(7.9) | 823(48.9) | 669(50.7) | 128(47.4) | 575(52.2) | 218(46.8) | 525(52.3) | 253(47.6) | ||||||
The number of women who would be followed up during 1-year, 2-year, and 3-year follow-up were 1,589, 1,561 and 1,535 respectively. a There are missing values for this feature of the subjects, and the total is not 10,481: Age at menarche (3 missing), Contraceptive measures (9 missing), No. of sexual partners in 5 year (21 missing), No. of pregnancies (5 missing), Age of first pregnancy(76 missing), No. of productions(99 missing), No. of live births(104 missing), Age of first delivery(100 missing), Smoking(1 missing). bRefers to any single or multiple infection with HPV31, 35, 39, 45, 51, 56, 59, 66, and 68
P < 0.05 highlighted in bold
Among the 1,684 women who tested positive for HR-HPV at baseline, the mean age was 47.64 (± 9.52, IQR = 42–54) years. Most were married (97.6% (1,643/1,684)), and 80.3% (1,353/1,684) had an education level below senior high school. In terms of reproductive history, 0.8% (14/1,684) reported menarche at age < 12 and 42.0% (708/1,684) were postmenopausal (Table 1).
Comparison between women followed up and those lost to follow-up
Among the 1,684 women who tested positive for HR-HPV at baseline, a total of 95, 28, and 26 participants exited the cohort in 2017, 2018, and 2019, respectively, due to being diagnosed with cervical intraepithelial neoplasia grade 2 or worse (≥ CIN2), and were thus excluded from subsequent follow-ups. Ultimately, follow-up rates were 83.0% (1,319/1,589) at 1 year (2018), 70.6% (1,102/1,561) at 2 years (2018–2019), and 65.3% (1,003/1,535) at 3 years (2018–2020). Participants who completed follow-up were significantly older than those lost to follow-up (P < 0.05). Significant differences were also observed between the two groups in terms of menopausal status, contraceptive use, number of lifetime sexual partners, age at first pregnancy, and age at first delivery (P < 0.05). However, no significant differences were found in baseline HPV genotype distribution between those who completed follow-up and those lost to follow-up (Table 1).
HR-HPV prevalence and persistence
Among the 1,684 h-HPV positive women, the most prevalent genotype was HPV52 (28.3%), followed by HPV16 (20.2%), HPV58 (17.8%), HPV33 (6.7%), and HPV18 (6.2%) (Table 1). The overall persistence rates for any HR-HPV genotype were 66.6% (881/1,319) at 1 year, 48.3% (532/1,102) at 2 years, and 39.9% (400/1,003) at 3 years. For type-specific HR-HPV persistence, HPV58 showed the highest persistence at both 1 and 2 years (73.2% and 51.1%, respectively), whereas HPV18 had the lowest (57.5% and 38.7%). At the 3-year follow-up, HPV33 exhibited the highest persistence rate (44.9%), while HPV16 had the lowest (27.2%) (Fig. 2).
Fig. 2.
Type-Specific HR-HPV persistence rate at 1-year, 2-year and 3-year follow-up. Other 9 h-HPV Refers to any single or multiple infection with HPV31, 35, 39, 45, 51, 56, 59, 66, and 68
A significant decline in persistence rates was observed over time for all 14 h-HPV types, including both type-specific infections and the pooled “other 9 HR-HPV types” (P < 0.05). The declining trends were most pronounced for HPV16 and HPV58, while HPV33 showed a more gradual decline.
Factors associated with HR-HPV persistence
In univariate analysis, factors significantly associated with increased HR-HPV persistence (P < 0.05) included infection with HPV33, HPV52, or HPV58; age ≥ 50 years; postmenopausal status; marital status; age at first delivery ≥ 26 years; age at first pregnancy between 21 and 25 years; having more than two pregnancies; and having more than two live births. Conversely, a higher education level (junior high school or above), use of an intrauterine device (IUD), and alcohol consumption were associated with decreased HR-HPV persistence (Table 2).
Table 2.
Factors associated with HR-HPV 1-year, 2-year and 3-year persistence
| Baseline characteristics |
1-year persistence | 2-year persistence | 3-year persistence | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| n/N | % | OR (95%CI) | P | n/N | % | OR (95%CI) | P | n/N | % | OR (95%CI) | P | |||
| HPV16 | ||||||||||||||
| No | 725/1093 | 66.3 | Reference | 436/916 | 47.6 | Reference | 339/841 | 40.3 | Reference | |||||
| Yes | 156/226 | 69.0 | 1.13(0.83–1.54.83.54) | 0.434 | 96/186 | 51.6 | 1.17(0.86–1.61.86.61) | 0.318 | 61/162 | 37.7 | 0.89(0.63–1.27.63.27) | 0.527 | ||
| HPV18 | ||||||||||||||
| No | 831/1246 | 66.7 | Reference | 496/1040 | 47.7 | Reference | 371/945 | 39.3 | Reference | |||||
| Yes | 50/73 | 68.5 | 1.09(0.65–1.80.65.80) | 0.751 | 36/62 | 58.1 | 1.52(0.90–2.55.90.55) | 0.112 | 29/58 | 50.0 | 1.55(0.91–2.63.91.63) | 0.105 | ||
| HPV33 | ||||||||||||||
| No | 812/1231 | 66.0 | Reference | 488/1027 | 47.5 | Reference | 360/934 | 38.5 | Reference | |||||
| Yes | 69/88 | 78.4 | 1.87(1.11–3.16.11.16) | 0.017 | 44/75 | 58.7 | 1.57(0.97–2.52.97.52) | 0.062 | 40/69 | 58.0 | 2.20(1.34–3.61.34.61) | 0.001 | ||
| HPV52 | ||||||||||||||
| No | 594/937 | 63.4 | Reference | 350/781 | 44.8 | Reference | 262/719 | 36.4 | Reference | |||||
| Yes | 287/382 | 75.1 | 1.74(1.34–2.28.34.28) | <0.001 | 182/321 | 56.7 | 1.61(1.24–2.10.24.10) | <0.001 | 138/284 | 48.6 | 1.65(1.25–2.18.25.18) | <0.001 | ||
| HPV58 | ||||||||||||||
| No | 699/1091 | 64.1 | Reference | 421/918 | 45.9 | Reference | 316/840 | 37.6 | Reference | |||||
| Yes | 182/228 | 79.8 | 2.22(1.57–3.14.57.14) | <0.001 | 111/184 | 60.3 | 1.79(1.30–2.48.30.48) | <0.001 | 84/163 | 51.5 | 1.76(1.26–2.47.26.47) | <0.001 | ||
| Other 9 HR-HPVa | ||||||||||||||
| No | 438/650 | 67.1 | Reference | 254/527 | 48.2 | Reference | 193/478 | 40.4 | Reference | |||||
| Yes | 443/699 | 66.2 | 0.95(0.75–1.19.75.19) | 0.653 | 278/575 | 48.3 | 1.01(0.79–1.28.79.28) | 0.960 | 207/525 | 39.4 | 0.96(0.75–1.24.75.24) | 0.759 | ||
| Age, y | ||||||||||||||
| 20–29 | 31/53 | 58.5 | Reference | 9/36 | 25.0 | Reference | 6/29 | 20.7 | Reference | |||||
| 30–39 | 132/204 | 64.7 | 1.30(0.70–2.41.70.41) | 0.403 | 58/169 | 34.3 | 1.57(0.69–3.55.69.55) | 0.279 | 45/161 | 28.0 | 1.49(0.57–3.89.57.89) | 0.417 | ||
| 40–49 | 243/410 | 59.3 | 1.03(0.58–1.85.58.85) | 0.914 | 137/344 | 39.8 | 1.99(0.91–4.35.91.35) | 0.082 | 95/321 | 29.6 | 1.61(0.64–4.08.64.08) | 0.311 | ||
| ≥50 | 475/652 | 72.9 | 1.91(1.07–3.38.07.38) | 0.025 | 328/553 | 59.3 | 4.37(2.02–9.48.02.48) | <0.001 | 254/492 | 51.6 | 4.09(1.64–10.22.64.22) | 0.001 | ||
| Marital status | ||||||||||||||
| Married | 855/1285 | 66.5 | Reference | 513/1075 | 47.7 | Reference | 388/980 | 39.6 | Reference | |||||
| Other | 26/34 | 76.5 | 1.64(0.73–3.64.73.64) | 0.225 | 19/27 | 70.4 | 2.60(1.13–6.00.13.00) | 0.020 | 12/23 | 52.2 | 1.66(0.73–3.81.73.81) | 0.223 | ||
| Education | ||||||||||||||
| Primary and below | 342/510 | 67.1 | Reference | 220/414 | 53.1 | Reference | 166/368 | 45.1 | Reference | |||||
| Junior high school | 387/563 | 68.7 | 1.08(0.84–1.40.84.40) | 0.556 | 218/476 | 45.8 | 0.75(0.57–0.97.57.97) | 0.029 | 160/438 | 36.5 | 0.70(0.53–0.93.53.93) | 0.013 | ||
| Senior high school | 100/160 | 62.5 | 0.82(0.57–1.19.57.19) | 0.288 | 66/138 | 47.8 | 0.81(0.55–1.19.55.19) | 0.279 | 52/129 | 40.3 | 0.82(0.55–1.24.55.24) | 0.345 | ||
| University and above | 52/86 | 60.5 | 0.75(0.47–1.20.47.20) | 0.232 | 28/74 | 37.8 | 0.54(0.33–0.89.33.89) | 0.015 | 22/68 | 32.4 | 0.58(0.34–1.01.34.01) | 0.051 | ||
| Age at menarche, y | ||||||||||||||
| <12 | 7/11 | 63.6 | Reference | 4/9 | 44.4 | Reference | 2/9 | 22.2 | Reference | |||||
| ≥12 | 874/1308 | 66.8 | 1.15(0.34–3.95.34.95) | >0.999 | 528/1093 | 48.3 | 1.17(0.31–4.37.31.37) | >0.999 | 398/994 | 40.0 | 2.34(0.48–11.31.48.31) | 0.456 | ||
| Menopause | ||||||||||||||
| No | 451/739 | 61.0 | Reference | 232/613 | 37.8 | Reference | 164/571 | 28.7 | Reference | |||||
| Yes | 430/580 | 74.1 | 1.83(1.44–2.32.44.32) | <0.001 | 300/489 | 61.3 | 2.61(2.04–3.33.04.33) | <0.001 | 236/432 | 54.6 | 2.99(2.30–3.88.30.88) | <0.001 | ||
| Contraceptive measure | ||||||||||||||
| No | 285/419 | 68.0 | Reference | 176/350 | 50.3 | Reference | 137/304 | 45.1 | Reference | |||||
| Condom | 25/45 | 55.6 | 0.59(0.32–1.10.32.10) | 0.092 | 14/37 | 37.8 | 0.60(0.30–1.21.30.21) | 0.150 | 12/36 | 33.3 | 0.61(0.29–1.26.29.26) | 0.180 | ||
| Intrauterine device | 133/225 | 59.1 | 0.68(0.49–0.95.49.95) | 0.024 | 76/177 | 42.9 | 0.74(0.52–1.07.52.07) | 0.111 | 51/162 | 31.5 | 0.56(0.38–0.84.38.84) | 0.004 | ||
| Sterilization | 418/602 | 69.4 | 1.07(0.82–1.40.82.40) | 0.631 | 253/518 | 48.8 | 0.94(0.72–1.24.72.24) | 0.676 | 190/483 | 39.3 | 0.79(0.59–1.06.59.06) | 0.112 | ||
| Other | 20/28 | 71.4 | 1.18(0.51–2.74.51.74) | 0.708 | 13/20 | 65.0 | 1.84(0.72–4.71.72.71) | 0.200 | 10/18 | 55.6 | 1.52(0.59–3.97.59.97) | 0.385 | ||
| No. of sexual partners in 5 years | ||||||||||||||
| 0–1 | 864/1293 | 66.8 | Reference | 525/1084 | 48.4 | Reference | 396/989 | 40.0 | Reference | |||||
| ≥2 | 13/22 | 59.1 | 0.72(0.30–1.69.30.69) | 0.446 | 3/14 | 21.4 | 0.29(0.08–1.05.08.05) | 0.045 | 1/11 | 9.1 | 0.15(0.02–1.17.02.17) | 0.076 | ||
| No. of pregnancies | ||||||||||||||
| 0–2 | 389/583 | 66.7 | Reference | 220/492 | 44.7 | Reference | 169/451 | 37.5 | Reference | |||||
| ≥3 | 492/736 | 66.8 | 1.00(0.79–1.27.79.27) | 0.962 | 312/610 | 51.1 | 1.29(1.02–1.64.02.64) | 0.034 | 231/552 | 41.8 | 1.20(0.93–1.55.93.55) | 0.159 | ||
| Age of first pregnancy | ||||||||||||||
| ≤ 20 | 168/269 | 62.5 | Reference | 91/213 | 42.7 | Reference | 70/197 | 35.5 | Reference | |||||
| 21–25 | 631/927 | 68.1 | 1.28(0.97–1.70.97.70) | 0.085 | 397/780 | 50.9 | 1.39(1.02–1.89.02.89) | 0.034 | 295/708 | 41.7 | 1.29(0.93–1.80.93.80) | 0.121 | ||
| ≥ 26 | 77/112 | 68.8 | 1.32(0.83–2.12.83.12) | 0.243 | 42/98 | 42.9 | 1.01(0.62–1.63.62.63) | 0.982 | 33/87 | 37.9 | 1.11(0.66–1.87.66.87) | 0.698 | ||
| No. of productions | ||||||||||||||
| 1–2 | 638/971 | 65.7 | Reference | 371/795 | 46.7 | Reference | 284/727 | 39.1 | Reference | |||||
| ≥3 | 239/336 | 71,1 | 1.29(0.98–1.69.98.69) | 0.068 | 159/295 | 53.9 | 1.34(1.02–1.75.02.75) | 0.034 | 114/264 | 43.2 | 1.19(0.89–1.58.89.58) | 0.242 | ||
| No. of live births | ||||||||||||||
| 1–2 | 704/1066 | 66.2 | Reference | 419/880 | 47.6 | Reference | 321/807 | 39.8 | Reference | |||||
| ≥3 | 173/243 | 71.2 | 1.26(0.93–1.72.93.72) | 0.132 | 111/210 | 52.9 | 1.23(0.91–1.67.91.67) | 0.172 | 77/184 | 41.8 | 1.09(0.79–1.51.79.51) | 0.605 | ||
| Age of first delivery | ||||||||||||||
| ≤20 | 78/129 | 60.5 | Reference | 41/99 | 41.4 | Reference | 30/92 | 32.6 | Reference | |||||
| 21–25 | 660/988 | 66.8 | 1.29(0.88–1.88.88.88) | 0.187 | 409/828 | 49.4 | 1.38(0.91–2.11.91.11) | 0.133 | 305/752 | 40.6 | 1.41(0.89–2.23.89.23) | 0.141 | ||
| ≥26 | 139/190 | 73.2 | 1.78(1.11–2.87.11.87) | 0.017 | 80/163 | 49.1 | 1.36(0.82–2.26.82.26) | 0.228 | 63/147 | 42.9 | 1.55(0.90–2.67.90.67) | 0.114 | ||
| Smoking | ||||||||||||||
| No | 876/1311 | 66.8 | Reference | 529/1095 | 48.3 | Reference | 397/997 | 39.8 | Reference | |||||
| Yes | 5/8 | 62.5 | 0.83(0.20–3.48.20.48) | >0.999 | 3/7 | 42.9 | 0.80(0.18–3.60.18.60) | >0.999 | 3/6 | 50.0 | 1.51(0.30–7.53.30.53) | 0.612 | ||
| Alcohol consumption | ||||||||||||||
| No | 800/1175 | 68.1 | Reference | 495/987 | 50.2 | Reference | 373/898 | 41.5 | Reference | |||||
| Yes | 81/144 | 56.3 | 0.60(0.42–0.86.42.86) | 0.004 | 37/115 | 32.2 | 0.47(0.31–0.71.31.71) | <0.001 | 27/105 | 25.7 | 0.49(0.31–0.77.31.77) | 0.002 | ||
Refers to any single or multiple infection with type of HPV31, 35, 39, 45, 51, 56, 59, 66, and 68；Significant factors (P <0.05) identified by univariate analyses were included in multivariate logistic regression models for analysis
P < 0.05 highlighted in bold
In multivariate logistic regression analysis, infection with HPV33, HPV52, and HPV58, along with postmenopausal status, remained significant independent risk factors for HR-HPV persistence. In contrast, alcohol consumption was identified as a protective factor. Specifically, compared to women infected with other HR-HPV types at baseline, HPV33 was more likely to persist at both 1 year (OR = 2.13, 95% CI: 1.25–3.63) and 3 years (OR = 2.40, 95% CI: 1.42–4.01); HPV52 was associated with higher persistence at 1 year (OR = 1.96, 95% CI: 1.49–2.59), 2 years (OR = 1.73, 95% CI: 1.32–2.28), and 3 years (OR = 1.95, 95% CI: 1.45–2.64); HPV58 also showed elevated persistence rates at 1 year (OR = 2.44, 95% CI: 1.71–3.48), 2 years (OR = 1.88, 95% CI: 1.34–2.63), and 3 years (OR = 2.01, 95% CI: 1.40–2.88). Postmenopausal women had significantly higher odds of persistent HR-HPV infection at 1 year (OR = 1.69, 95% CI: 1.32–2.15), 2 years (OR = 2.48, 95% CI: 1.93–3.19), and 3 years (OR = 2.84, 95% CI: 2.17–3.72) compared to premenopausal women. In contrast, women who reported alcohol consumption had significantly lower odds of persistence at 1 year (OR = 0.60, 95% CI: 0.41–0.86), 2 years (OR = 0.47, 95% CI: 0.31–0.73), and 3 years (OR = 0.49, 95% CI: 0.30–0.79) (Table 3).
Table 3.
Multivariate logistic regression analysis of factors associated with 1-, 2-, and 3-year persistence of HR-HPV persistence
| Baseline characteristics |
1-year persistence | 2-year persistence | 3-year persistence | |||||
|---|---|---|---|---|---|---|---|---|
| OR (95%CI) | P | OR (95%CI) | P | OR (95%CI) | P | |||
| HPV33 | ||||||||
| No | Reference | - | Reference | |||||
| Yes | 2.13(1.25–3.63) | 0.006 | - | - | 2.40(1.42–4.01) | 0.001 | ||
| HPV52 | ||||||||
| No | Reference | Reference | Reference | |||||
| Yes | 1.96(1.49–2.59) | < 0.001 | 1.73(1.32–2.28) | < 0.001 | 1.95(1.45–2.64) | < 0.001 | ||
| HPV58 | ||||||||
| No | Reference | Reference | Reference | |||||
| Yes | 2.44(1.71–3.48) | < 0.001 | 1.88(1.34–2.63) | < 0.001 | 2.01(1.40–2.88) | < 0.001 | ||
| Menopause | ||||||||
| No | Reference | Reference | Reference | |||||
| Yes | 1.69(1.32–2.15) | < 0.001 | 2.48(1.93–3.19) | < 0.001 | 2.84(2.17–3.72) | < 0.001 | ||
| Alcohol consumption | ||||||||
| No | Reference | Reference | Reference | |||||
| Yes | 0.60(0.41–0.86) | 0.006 | 0.47(0.31–0.73) | < 0.001 | 0.49(0.30–0.79) | 0.003 | ||
The logistic regression model was established by the stepwise regression method to identify significant factors from those identified as significant (P < 0.05) in the univariate analyses, using an entry criterion of P < 0.05 and a removal criterion of P > 0.10 to derive the optimal model
P < 0.05 highlighted in bold
In the univariate analyses, significant factors were identified as follows: For 1-year persistence: HPV33, HPV52, HPV58, age, menopause, intrauterine device use, age at first delivery, and alcohol consumption; For 2-year persistence: HPV52, HPV58, age, marital status, education, menopause, number of pregnancies, age at first pregnancy, number of children, and alcohol consumption; For 3-year persistence: HPV33, HPV52, HPV58, age, education, age at menarche, menopause, contraceptive methods, and alcohol consumption
HPV genotype distribution in CIN2 + Lesions at baseline (2017–2020)
To facilitate accurate interpretation of HPV persistence findings, we analyzed the baseline distribution of HPV genotypes among CIN2 + cases, as individuals with prior CIN2 + diagnoses were excluded from the persistence analysis. Table 4 summarizes the annual and cumulative distribution of high-risk HPV genotypes detected in CIN2 + lesions among women who were HPV-positive at baseline from 2017 to 2020. HPV16 was the most frequently detected genotype across all years, accounting for 55.1% of cases overall. HPV52(24.7%) and HPV58(21.9%) also showed consistent prevalence, while HPV18(9.0%) was less commonly observed. The “Other 9 HR-HPV” group represented 30.9% of CIN2 + cases during the study period. Among the CIN2 + cases excluded from the 1 to 3-year persistence analysis, HPV16 infection at baseline accounted for 57.0% (85/149).
Table 4.
Baseline distribution of HPV genotypes in CIN2 + Lesions among initially HPV-Positive women (2017–2020)
| Baseline HPV subtypea | CIN2+ (2017) | CIN2+ (2018) | CIN2+ (2019) | CIN2+ (2020) | CIN2+ (2017–2020) | ||||
|---|---|---|---|---|---|---|---|---|---|
| n (%) | n (%) | n (%) | n (%) | n (%) | |||||
| HPV16 | 63(66.3) | 12(42.9) | 10(38.5) | 13(44.8) | 98(55.1) | ||||
| HPV18 | 13(13.7) | 0(0.0) | 0(0.0) | 3(10.3) | 16(9.0) | ||||
| HPV33 | 8(8.4) | 4(14.3) | 4(15.4) | 6(20.7) | 22(12.4) | ||||
| HPV52 | 21(22.1) | 7(25.0) | 7(26.9) | 9(31.0) | 44(24.7) | ||||
| HPV58 | 19(20.0) | 9(32.1) | 5(19.2) | 6(20.7) | 39(21.9) | ||||
| Other 9 h-HPV b | 26(27.3) | 6(21.4) | 13(50.0) | 10(34.5) | 55(30.9) | ||||
| Total | 95 | 28 | 26 | 29 | 178 |
aNo distinction was made between single infection and multiple infection. bRefers to any single or multiple infection with HPV31, 35, 39, 45, 51, 56, 59, 66, and 68
Discussion
This study represents the first large-scale, nationwide, multicenter cohort of Chinese women with three follow-up assessments conducted between 2017 and 2020. It provides robust data on genotype-specific persistence of high-risk human papillomavirus (HR-HPV) infections across different follow-up intervals, as well as associated factors in the general female population in China. These findings address a critical gap in the literature, particularly in the context of China’s unique demographic and cultural characteristics, where population-based longitudinal data on HR-HPV persistence have been scarce.
The HPV prevalence observed in our cohort was 16.1%, modestly exceeded that reported in a previous nationwide Chinese study (12.3%) [17], and varied international comparisons–lower in Jakarta (10.7%) [18] and Japan (10.8%) [19], but higher prevalence in Korea (27.8%) [20] and Kazakhstan (43.0%) [21]. Persistence rates declined over time, with 66.8% at 1 year, 48.3% at 2 years, and 39.9% at 3 years, reflecting the natural history of HPV infection, whereby a proportion of infections spontaneously clear as follow-up duration increases [22, 23]. Notably, the 1-year persistence rate was substantially higher than that observed in a Danish study (43.9%) [6], likely due to the inclusion of women with cytological abnormalities (≥ ASC-US) in our participants, who may have impaired immune responses or higher baseline viral loads [24]. Our 2-year persistence findings are comparable to those reported in studies from Guangdong province (47.6%) [9], Hubei province (46.5%) [15] and Italian(49.1%) [25], suggesting that similar epidemiological patterns may emerge when similar study designs and HPV detection methods are employed.
Genotype-specific differences in HR-HPV persistence were substantial. HPV58 and HPV52 exhibited the highest persistence rates at 1 and 2 years, while HPV33 showed the greatest persistence at 3 years. These genotypes may possess intrinsic biological properties that enhance viral retention, potentially influenced by host genetic susceptibility, immune response dynamics, and environmental cofactors. Baseline infection with HPV58, HPV52, or HPV33 was strongly associated with prolonged viral detectability over the follow-up period. Our results are consistent with regional data from East Asia, including Guangdong (China) and South Korea, where these genotypes are disproportionate represented in cervical cancer (CC) cases [9, 26, 27]. In contrast, Western studies frequently identify HPV16 as the most persistent genotype [6, 8, 28, 29]. This geographical divergence may reflect population-level differences in HLA allele distributions [30], behavioral patterns [31], or detection biases related to assay sensitivity and specificity [27, 32]. The regional burden of HPV52 and HPV58 in cervical carcinogenesis across East Asia [33] underscores the need for genotype-tailored vaccine strategies and region-specific screening algorithms that account for differential persistence dynamics. Variations across studies may also be attributable to differences in study design, follow-up duration, and population demographics. Harmonized methodologies and longitudinal cohorts with extended follow-up are essential to clarify the mechanisms underlying genotype-specific persistence and to inform global prevention strategies.
Previous studies [34] reported that persistent infections in older women are more likely to be prevalent rather than newly acquired, while our results indicate that age was associated with HPV persistence in univariate analysis but did not remain significant in multivariate models. Instead, menopausal status emerged as an independent predictor, potentially driven by immune senescence, diminished ovarian hormone production, and thinning of the vaginal epithelium, which collectively compromise viral clearance [35, 36]. Corroborating prior research reporting a 1.83-fold higher risk in postmenopausal women compared to their premenopausal counterparts [37]. Evidence from cohorts of women with normal cytology but positive HR-HPV DNA tests has also indicated higher persistence rates among menopausal individuals [38]. Furthermore, menopause has been identified as an independent predictor of HPV persistence following cervical conization [39]. Postmenopausal women are also prone to vaginal microbiota dysbiosis and reduced microbial diversity, which may compromise mucosal immunity and promote viral persistence [40, 41]. Given that this population typically fall outside the routine age window for HPV vaccination, access to timely and effective cervical cancer screening becomes critical. The World Health Organization currently recommends screening starting at age 30 and continuing at 5–10 years intervals [42]. Targeted strategies for postmenopausal women—such as risk-based screening algorithms or alternative triage approaches—may help reduce unnecessary colposcopies and overtreatment while improving detection and outcomes in this higher-risk group.
Although alcohol consumption is a well-established risk factor for several malignancies—including oral, esophageal, colorectal, and liver cancers [43]—our study observed an inverse association between alcohol use and HR-HPV persistence. This unexpected association may be influenced by residual confounding, behavioral clustering, or reporting bias. In our cohort, the prevalence of alcohol consumption is relatively low, participants who reported alcohol use are younger on average, and there are potential biases associated with self-reported behavioral data. It is also possible that lifestyle or socioeconomic factors correlated with alcohol use in this population—such as greater health awareness or increased healthcare access—may have indirectly contributed to reduced persistence. Further research incorporating objective biomarkers of alcohol exposure and comprehensive adjustment for confounders is warranted to clarify whether this association reflects a true biological or behavioral phenomenon.
Importantly, the observed 3-year persistence rate of HPV16 was relatively low despite its second baseline prevalence. This discrepancy likely reflects selection bias introduced by the exclusion of individuals with prior CIN2+. Among the excluded CIN2 + cases, HPV16 accounted for 57.0% (85/149) of baseline infections, followed by HPV52 and HPV58, indicating that a substantial proportion of persistent HPV16 infections may have progressed rapidly to high-grade lesions and were removed from follow-up. Therefore, while HPV33, 52, and 58 appear prominent in sustained viral presence over short- to mid-term intervals, HPV16 remains critically important in long-term oncogenic progression and cervical lesion development. These findings underscore the need for genotype-tailored vaccination strategies and region-specific screening algorithms that account for both persistence dynamics and clinical outcomes.
This study has several notable strengths, including its prospective, multicenter, population-based design, large sample size, and multiple follow-up intervals, which enable a nuanced understanding of HPV persistence dynamics. However, several limitations should be acknowledged. Excluding individuals with prior CIN2 + likely introduced selection bias in assessing genotype-specific persistence, resulting in a potential underestimation of HPV16 persistence owing to its rapid progression to lesions and early clinical management. The onset of HPV infections observed at baseline remains unknown; therefore, the actual duration of infection may be longer than recorded in our study. Differences between participants who were followed up and those lost to follow-up—particularly the older age of the former group—may have also introduced selection bias, despite statistical adjustment. The predominance of a rural population, characterized by homogeneous marital status, low smoking prevalence, and limited contraceptive diversity, may limit the external validity of the findings, especially when extrapolated to urban or more diverse populations. Reliance on self-reported data may have led to underreporting of sensitive behaviors. Additionally, the 12-month sampling interval may overestimate persistence by failing to differentiate between transient reinfections and true persistence. Finally, the number of HPV genotypes tested was limited, and variability in assay sensitivity may have resulted in misclassification.
Conclusions
This study offers important insights into the persistence of high-risk HPV infections among Chinese women, demonstrating a progressive decline in overall persistence rates over a three-year period and identifying menopause as a key risk factor. High persistence rates of HPV52, HPV58, and HPV33, along with the high representation of HPV16 in CIN2 + lesions, indicate the necessity of genotype-specific interventions for both screening and vaccination strategies. These findings support the implementation of tailored cervical cancer screening approaches for postmenopausal women, a population typically excluded from prophylactic vaccination programs. Future longitudinal studies with extended follow-up, more diverse populations, and advanced molecular diagnostics—including immune profiling and viral load quantification—are needed to deepen our understanding of HPV persistence mechanisms and to support evidence-based efforts in cervical cancer prevention globally and other HPV-associated diseases.
Supplementary Information
Acknowledgements
The authors are grateful to Henan Cancer Hospital, Changzhi Medical College for the scientific review of the protocol, the study synopsis and supporting the local field work.
Abbreviations
- HR-HPV
High-risk human papillomavirus
- HPV
Human papillomavirus
- CI
Confidence interval
- OR
Odds ratio
- CIN
Cervical intraepithelial neoplasia
- CIN2
Cervical intraepithelial neoplasia grades 2
- ≥CIN2
Greater than or equal to cervical intraepithelial neoplasia grade 2
- NILM
Negative for intraepithelial lesion or malignancy
- ASC-US
Atypical squamous cells of undetermined significance
- ≥ASC-US
Greater than or equal to atypical squamous cells of undetermined significance
- PCR
Fluorescent polymerase chain reaction
- TCT
ThinPrep Cytologic Test
- SCU
Sichuan University
- CC
Cervical cancer
Author contributions
Study design, Jing Li, Ming-Rong Xi, Guang-Dong Liao, Shao-Kai Zhang, Zhi-Fang Li, Wen Chen, You-Lin Qiao; Conceptualization, Jing Li, Chong-Bi Huang, Fan Zhao, Dan-Dan Zhao; Data curation, Meng-Yin Ao, Le-Ni Kang; Formal analysis, Chong-Bi Huang; Field work supervision and quality control, Jing Li, Ming-Rong Xi, Guang-Dong Liao, Shao-Kai Zhang, Zhi-Fang Li, Yin Liu; Pathology quality control, Lian Xu, Shu-Yuan Liu; Software, Chong-Bi Huang; Resources, Xi Zeng, Ming-Wei Yuan, Shao-Long Xue; Methodology, Jing Li, Chong-Bi Huang; Visualization, Chong-Bi Huang; Writing-original draft, Jing Li, Chong-Bi Huang, Fan Zhao, Dan-Dan Zhao; Writing-review and editing, Jing Li, Chong-Bi Huang, Fan Zhao, Dan-Dan Zhao, Guang-Dong Liao, Ming-Rong Xi. The work reported in the paper has been performed by the authors, unless clearly specified in the text. All authors contributed to the article and approved the submitted version.
Funding
This work was supported by the Natural Science Foundation of China (Grant No.82473145) and the Yunnan Fundamental Research Projects (Grant No.202201AS070059).
Data availability
The datasets analyzed during the current study are not publicly available due to privacy protection, data security, and to protect the interests involved in multi-center cooperation but are available from the corresponding author on reasonable request.
Declarations
Ethics approval and consent to participate
This study was conducted under compliance with the Declaration of Helsinki and was approved by the Clinical Trial Ethics Review Committee, West China Second University Hospital, Sichuan University (approval number: Q2017003); the Medical Ethics Committee of Henan Cancer Hospital (approval number: 2017009); and the Medical Ethics Committee of Changzhi Medical College (approval number: 2017002). The study was registered at the Chinese Clinical Trial Registry Center of the World Health Organization International Clinical Trials Registry Platform (Registration number ChiCRT2200055287). All participants provided written informed consent in every visits.
Consent for publication
Not applicable.
Competing interests
The authors declare no competing interests.
Footnotes
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Contributor Information
Ming-Rong Xi, Email: qiemingrong@scu.edu.cn.
Guang-Dong Liao, Email: liaoguangdong@scu.edu.cn.
References
- 1.Walboomers JM, Jacobs MV, Manos MM, Bosch FX, Kummer JA, Shah KV, et al. Human papillomavirus is a necessary cause of invasive cervical cancer worldwide. J Pathol. 1999;189(1):12–9. 10.1002/(sici)1096-9896(199909)189:1%3C12::Aid-path431%3E3.0.Co;2-f. [DOI] [PubMed] [Google Scholar]
- 2.Lei J, Ploner A, Elfström KM, Wang J, Roth A, Fang F, et al. HPV vaccination and the risk of invasive cervical cancer. N Engl J Med. 2020;383(14):1340–8. 10.1056/NEJMoa1917338. [DOI] [PubMed] [Google Scholar]
- 3.Ronco G, Dillner J, Elfström KM, Tunesi S, Snijders PJ, Arbyn M, et al. Efficacy of HPV-based screening for prevention of invasive cervical cancer: follow-up of four European randomised controlled trials. Lancet. 2014;383(9916):524–32. 10.1016/s0140-6736(13)62218-7. [DOI] [PubMed] [Google Scholar]
- 4.de Sanjosé S, Brotons M, Pavón MA. The natural history of human papillomavirus infection. Best Pract Res Clin Obstet Gynaecol. 2018;47:2–13. 10.1016/j.bpobgyn.2017.08.015. [DOI] [PubMed] [Google Scholar]
- 5.Schmeink CE, Melchers WJ, Siebers AG, Quint WG, Massuger LF, Bekkers RL. Human papillomavirus persistence in young unscreened women, a prospective cohort study. PLoS ONE. 2011;6(11):e27937. 10.1371/journal.pone.0027937. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Stensen S, Kjaer SK, Jensen SM, Frederiksen K, Junge J, Iftner T, et al. Factors associated with type-specific persistence of high-risk human papillomavirus infection: A population-based study. Int J Cancer. 2016;138(2):361–8. 10.1002/ijc.29719. [DOI] [PubMed] [Google Scholar]
- 7.Rosa MI, Fachel JM, Rosa DD, Medeiros LR, Igansi CN, Bozzetti MC. Persistence and clearance of human papillomavirus infection: a prospective cohort study. Am J Obstet Gynecol. 2008;199(6):617e. 10.1016/j.ajog.2008.06.033. [DOI] [PubMed] [Google Scholar]
- 8.Nielsen A, Kjaer SK, Munk C, Osler M, Iftner T. Persistence of high-risk human papillomavirus infection in a population-based cohort of Danish women. J Med Virol. 2010;82(4):616–23. 10.1002/jmv.21750. [DOI] [PubMed] [Google Scholar]
- 9.Li M, Liu T, Luo G, Sun X, Hu G, Lu Y, et al. Incidence, persistence and clearance of cervical human papillomavirus among women in Guangdong, China 2007–2018: a retrospective cohort study. J Infect Public Health. 2021;14(1):42–9. 10.1016/j.jiph.2020.11.011. [DOI] [PubMed] [Google Scholar]
- 10.Tuerxun G, Abudurexiti G, Abulizi G. Prevalence, persistence, clearance and risk factors for HPV infection in rural Uyghur women in China. BMC Womens Health. 2023;23(1):433. 10.1186/s12905-023-02558-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Liu J, Shi Y, Wang L, Wang J, Fan D, Han S, et al. Epidemiology and persistence of cervical human papillomavirus infection among outpatient women in Heilongjiang province: a retrospective cohort study. J Med Virol. 2020;92(12):3784–92. 10.1002/jmv.25899. [DOI] [PubMed] [Google Scholar]
- 12.Zhu X, Wang Y, Lv Z, Su J. Prevalence and genotype distribution of high-risk HPV infection among women in Beijing, China. J Med Virol. 2021;93(8):5103–9. 10.1002/jmv.27013. [DOI] [PubMed] [Google Scholar]
- 13.He L, He J. Distribution of high-risk HPV types among women in Sichuan province, china: a cross-sectional study. BMC Infect Dis. 2019;19(1):390. 10.1186/s12879-019-4038-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Luo LP, He P, Liu QT, Jiang YH, Zhang YN, Li QZ, et al. Prevalence and genotype distribution of HPV infection among 214,715 women from Southern China, 2012–2018: baseline measures prior to mass HPV vaccination. BMC Infect Dis. 2021;21(1):328. 10.1186/s12879-021-06019-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Pan S, Wei W, Du X, Li Z, Tuo J, Zhang M, et al. Factors associated with persistence and clearance of HPV16/18 among rural Chinese women: a cohort study in Wufeng, Hubei Province. Womens Health. 2022;62(4):276–86. 10.1080/03630242.2022.2056283. [DOI] [PubMed] [Google Scholar]
- 16.Zhang Q, Cao D, Ma Q, Li N, Cui XQ, Yang XF. Natural outcome of genital tract high-risk human papillomavirus infection and associated factors among 760 women. Zhongguo Yi Xue Ke Xue Yuan Xue Bao. 2015;37(5):534–40. 10.3881/j.issn.1000-503X.2015.05.008. [DOI] [PubMed] [Google Scholar]
- 17.Gao D, Zhao G, Wang X, Juan J, Shi Y, Xu T, et al. Association between high-risk human papillomavirus infection and cervical cytology in health check-up women – 23 PLADs, China, 2023. China CDC Wkly. 2025;7(10):327–33. 10.46234/ccdcw2025.053. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Murdiyarso LS, Kartawinata M, Jenie I, Widjajahakim G, Hidajat H, Sembiring R, et al. Single and multiple high-risk and low-risk human papillomavirus association with cervical lesions of 11,224 women in Jakarta. Cancer Causes Control: CCC. 2016;27(11):1371–9. 10.1007/s10552-016-0816-4. [DOI] [PubMed] [Google Scholar]
- 19.Takehara K, Toda T, Nishimura T, Sakane J, Kawakami Y, Mizunoe T, et al. Human papillomavirus types 52 and 58 are prevalent in uterine cervical squamous lesions from Japanese women. Pathol Res Int. 2011;2011:246936. 10.4061/2011/246936. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Nah EH, Cho S, Kim S, Cho HI. Human papillomavirus genotype distribution among 18,815 women in 13 Korean cities and relationship with cervical cytology findings. Annals Lab Med. 2017;37(5):426–33. 10.3343/alm.2017.37.5.426. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Balmagambetova SK, Tinelli A, Urazayev ON, Sakieva KZ, Koyshybaev AK, Zholmukhamedova DA, et al. HPV types distribution in general female population and in women diagnosed with cervical cancer across Western Kazakhstan. Asian Pac J Cancer Prevention: APJCP. 2019;20(4):1089–96. 10.20517/2394-4722.2023.54. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Rositch AF, Koshiol J, Hudgens MG, Razzaghi H, Backes DM, Pimenta JM, et al. Patterns of persistent genital human papillomavirus infection among women worldwide: a literature review and meta-analysis. Int J Cancer. 2013;133(6):1271–85. 10.1002/ijc.27828. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Zhao M, Zhou D, Zhang M, Kang P, Cui M, Zhu L, et al. Characteristic of persistent human papillomavirus infection in women worldwide: a meta-analysis. PeerJ. 2023;11:e16247. 10.7717/peerj.16247. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Maucort-Boulch D, Plummer M, Castle PE, Demuth F, Safaeian M, Wheeler CM, et al. Predictors of human papillomavirus persistence among women with equivocal or mildly abnormal cytology. Int J Cancer. 2010;126(3):684–91. 10.1002/ijc.24752. [DOI] [PubMed] [Google Scholar]
- 25.Sammarco ML, Del Riccio I, Tamburro M, Grasso GM, Ripabelli G. Type-specific persistence and associated risk factors of human papillomavirus infections in women living in central Italy. Eur J Obstet Gynecol Reprod Biol. 2013;168(2):222–6. 10.1016/j.ejogrb.2013.01.012. [DOI] [PubMed] [Google Scholar]
- 26.Ingabire C, Lim MK, Won YJ, Oh JK. Human papillomavirus genotype-specific persistence and potential risk factors among Korean women: results from a 2-year follow-up study. Cancer Res Treat. 2018;50(3):813–22. 10.4143/crt.2017.340. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Chan PK, Cheung TH, Tam AO, Lo KW, Yim SF, Yu MM, et al. Biases in human papillomavirus genotype prevalence assessment associated with commonly used consensus primers. Int J Cancer. 2006;118(1):243–5. 10.1002/ijc.21299. [DOI] [PubMed] [Google Scholar]
- 28.Huica I, Iancu IV, Botezatu A, Plesa A, Socolov D, Teleman S, et al. Factors associated with persistence of Hpv genital infection in a small cohort of Romanian women. Acta Clin Croatica. 2019;58(3):410–6. 10.20471/acc.2019.58.03.02. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Mollers M, Boot Hein J, Vriend Henrike J, King Audrey J, van den Broek Ingrid VF, van Bergen Jan EA, et al. Prevalence, incidence and persistence of genital HPV infections in a large cohort of sexually active young women in the Netherlands. Vaccine. 2013;31(2):394–401. 10.1016/j.vaccine.2012.10.087. [DOI] [PubMed] [Google Scholar]
- 30.Zoodsma M, Nolte IM, Schipper M, Oosterom E, van der Steege G, de Vries EG, et al. Analysis of the entire HLA region in susceptibility for cervical cancer: a comprehensive study. J Med Genet. 2005;42(8):e49. 10.1136/jmg.2005.031351. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Almonte M, Albero G, Molano M, Carcamo C, García PJ, Pérez G. Risk factors for human papillomavirus exposure and co-factors for cervical cancer in Latin America and the Caribbean. Vaccine. 2008;26(Suppl 11):L16–36. 10.1016/j.vaccine.2008.06.008. [DOI] [PubMed] [Google Scholar]
- 32.de Sanjosé S, Diaz M, Castellsagué X, Clifford G, Bruni L, Muñoz N, et al. Worldwide prevalence and genotype distribution of cervical human papillomavirus DNA in women with normal cytology: a meta-analysis. Lancet Infect Dis. 2007;7(7):453–9. 10.1016/s1473-3099(07)70158-5. [DOI] [PubMed] [Google Scholar]
- 33.Guan P, Howell-Jones R, Li N, Bruni L, de Sanjosé S, Franceschi S, et al. Human papillomavirus types in 115,789 HPV-positive women: a meta-analysis from cervical infection to cancer. Int J Cancer. 2012;131(10):2349–59. 10.1002/ijc.27485. [DOI] [PubMed] [Google Scholar]
- 34.Rodriguez AC, Schiffman M, Herrero R, Hildesheim A, Bratti C, Sherman ME, et al. Longitudinal study of human papillomavirus persistence and cervical intraepithelial neoplasia grade 2/3: critical role of duration of infection. J Natl Cancer Inst. 2010;102(5):315–24. 10.1093/jnci/djq001. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Davis SR, Pinkerton J, Santoro N, Simoncini T. Menopause-biology, consequences, supportive care, and therapeutic options. Cell. 2023;186(19):4038–58. 10.1016/j.cell.2023.08.016. [DOI] [PubMed] [Google Scholar]
- 36.González P, Hildesheim A, Rodríguez AC, Schiffman M, Porras C, Wacholder S, et al. Behavioral/lifestyle and immunologic factors associated with HPV infection among women older than 45 years. Cancer Epidemiol Biomarkers Prev. 2010;19(12):3044–54. 10.1158/1055-9965.Epi-10-0645. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Sui S, Niyazi M, Wang L, Lu P, Zhu K, Jiao Z. A nested case-control study on type-specific persistent HPV infection in Uyghur women. J Public Health (Germany). 2016;24(3):245–51. [Google Scholar]
- 38.Tifaoui N, Maudelonde T, Combecal J, Vallo R, Doutre S, Didelot MN, et al. High-risk HPV detection and associated cervical lesions in a population of French menopausal women. J Clin Virol. 2018;108:12–8. 10.1016/j.jcv.2018.08.010. [DOI] [PubMed] [Google Scholar]
- 39.Zhang Y, Ni Z, Wei T, Liu Q. Persistent HPV infection after conization of cervical intraepithelial neoplasia– a systematic review and meta-analysis. BMC Womens Health. 2023;23(1):216. 10.1186/s12905-023-02360-w. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.Ritu W, Enqi W, Zheng S, Wang J, Ling Y, Wang Y. Evaluation of the associations between cervical microbiota and HPV infection, clearance, and persistence in cytologically normal women. Cancer Prev Res (Phila). 2019;12(1):43–56. 10.1158/1940-6207.Capr-18-0233. [DOI] [PubMed] [Google Scholar]
- 41.So KA, Hur SY, Cho CH, Lee JK, Seong SJ, Jeong DH, et al. Menopausal status induces vaginal dysbiosis in women with human papillomavirus infection. Sci Rep. 2024;14(1):7092. 10.1038/s41598-024-56314-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.World Health Organization. WHO guideline for screening and treatment of cervical pre-cancer lesions for cervical cancer prevention, second edition. 2022. https://www.who.int/publications/i/item/9789240040434. Accessed 28 Dec 2024. [PubMed]
- 43.Alcohol consumption and ethyl carbamate. IARC Monogr Eval Carcinog Risks Hum. 2010;96:3–1383. [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The datasets analyzed during the current study are not publicly available due to privacy protection, data security, and to protect the interests involved in multi-center cooperation but are available from the corresponding author on reasonable request.