High-risk HPV distribution and importance of continuing cervical cancer screening of women aged 65 years and older: a study based on 2 152 766 women in China

Introduction

Cervical cancer is one of the most common malignancies among women globally, posing a serious threat to women’s health. Although the incidence is higher among younger women, the ageing population has led to a rising trend in cervical cancer among women aged≥65.1 2 According to the latest statistics of the WHO,3 there were 157 182 new cases of cervical cancer in this age group worldwide, with an age-standardised incidence rate of 37.2 per 100 000, ranking 10th among all cancers. The number of deaths was 124 269, with an age-standardised mortality rate of 28.3 per 100 000, ranking ninth. The impact of cervical cancer in women aged≥65 has reached concerning levels, yet this population’s health concerns remain underappreciated by society.

Currently, most screening guidelines suggest that women aged≥65 may cease cervical cancer screening if they meet specific criteria4 5: having at least three Pap tests or two human papillomavirus (HPV) tests in the past 10 years with normal results, no history of cervical precancer or having undergone a total hysterectomy for non-cancerous conditions such as fibroids. WHO guidelines recommend starting screening at age 30 and stopping at age 49, but suggest screening women aged 50–65 who have never been screened if resources allow.6 These guidelines account for the low benefits of continued screening in women aged≥65 compared with potential harms like discomfort during sampling, false positives and unnecessary treatments.7 However, with age-related physiological and immune changes, the risk and duration of HPV infection may increase in older women, with early symptoms of cervical cancer often being subtle and easily overlooked. Studies indicate that women aged≥65 have higher incidence rates and lower survival rates for cervical cancer.8 9 And research on HPV prevalence in this age group is insufficient, and the importance of screening women aged≥65 is not widely recognised. Therefore, the appropriate age to stop screening warrants further investigation.

In this context, our study aims to analyse cervical cancer screening data to elucidate the current status and characteristics of high-risk HPV (hr-HPV) infection among older women, including infection rates and genotype distribution, as well as the prevalence of cervical intraepithelial neoplasia grade 2 or worse (CIN2+) and their association. By integrating existing clinical and public health research, we will assess the importance and necessity of cervical cancer screening in older women. We hope the findings will raise awareness of the health needs of older women, prompting policy adjustments and resource allocation to ensure timely and appropriate health services for this group.

Methods

Results

Study population and data characteristics

This study collected 2 580 829 records initially, and 2 152 766 complete records were included based on criteria, yielding a data validity rate of 83.4% (see figure 1). Among these, 57.47% (n=1 237 231) of the women underwent screening at hospitals. Demographic characteristics and clinical histories are detailed in table 1. The average age of participants was 39.85±9.22 years, with 90.56% aged between 25 and 54 years. Only 2.25% had received the HPV vaccine. The results of the hr-HPV test showed that 91.88% were negative for hr-HPV genotypes, 2.26% were positive for HPV16/18, and 5.86% were infected with non-HPV16/18 genotypes. The referral rate for colposcopy was 4.14% (n=89 148), with an abnormal rate of 41.97% (n=37 418) among these. Histopathological diagnoses were distributed as follows: 67.65% (n=25 427) normal/benign, 23.02% (n=8613) LSIL, 8.44% (n=3157) HSIL and 0.59% (n=221) cancer.

Figure 1

Request permissions

The flowchart of this study. Note: The primary screening result was based on a co-testing approach combining HPV and cytology. HPV16/18 genotypes results were considered positive. For non-HPV16/18 genotypes, cases with cytology showing atypical squamous cells of undetermined significance or worse (ASC-US+) were classified as positive, while the rest were considered negative. HPV, human papillomavirus; hr-HPV, high-risk HPV; HSIL, high-grade squamous intraepithelial lesion; LSIL, low-grade squamous intraepithelial lesion.

Table 1

View inline•Open as popup

Characteristics of the study population (N (%))

Age-specific prevalence of inclusive hr-HPV infections and their proportion of CIN2+

Figure 2 illustrates the prevalence of inclusive single and double hr-HPV infections, along with the proportion of CIN2+ in histological results. The total number of hr-HPV infections was 174 819, with an infection rate of 8.12% and a CIN2+ proportion of 1.83% (n=3198). Among women aged≥65, there were 2382 (13.67%) HPV infections, among whom the CIN2+ detection rate was 2.23% (n=53). For those<65, the infection rate was 8.08% (n=172 437), among whom the CIN2+ detection rate was 1.82% (n=3145). Hr-HPV infection showed a statistical difference between women aged<65 and ≥65 (p<0.001), but there was no difference in CIN2+ detection rate. Detailed results are in online supplemental table S2 and S3.

Figure 2

Request permissions

The prevalence of inclusive high-risk HPV infection and its proportion of CIN2+. (A) All ages; (B) Age<64; (C) Age≥65. (a) The prevalence of inclusive double high-risk HPV infection. This prevalence equals the number of cases with inclusive double-specific high-risk HPV infection divided by the total number of all-inclusive double-infection cases. (b) The prevalence of inclusive single high-risk HPV infection. This prevalence equals the number of cases with inclusive single-specific high-risk HPV infection divided by the total number of all-inclusive single infection cases. (c) The proportion of CIN2+ detection in cases of inclusive double high-risk HPV infection. The proportion of CIN2+ is calculated by dividing the number of CIN2+ cases with inclusive double high-risk HPV infection by the total number of all-inclusive double infection cases. (d) The proportion of CIN2+ detection in cases of inclusive single high-risk HPV infection. The proportion of CIN2+ is calculated by dividing the number of CIN2+ cases with inclusive single high-risk HPV infection by the total number of all-inclusive single infection cases. Note: Inclusive double hr-HPV infection: This term refers to cases where at least two specific high-risk HPV genotypes are present, that is, additional HPV genotypes are also allowed; similar considerations apply to single high-risk HPV infection. CIN2+, cervical intraepithelial neoplasia grade 2 or worse; HPV, human papillomavirus.

In the hr-HPV infections, the five most common genotypes for those<65 were HPV52, 16, 58, 51 and 18, with infection rates of 22.72%, 16.95%, 10.85%, 7.66% and 6.84%, respectively, as detailed in figure 2B. However, among those≥65, the most prevalent genotypes were HPV52, 16, 58, 56 and 68, with rates of 23.11%, 19.87%, 13.27%, 6.99% and 5.67%, as shown in figure 2C. For CIN2+ detection, the leading hr-HPV genotypes in women aged<65 were HPV16, 33, 31, 18 and 58, with detection rates of 4.58%, 2.80%, 2.26%, 2.17% and 2.12%, respectively. In those≥65, the top CIN2+ detecting genotypes were HPV18, 16, 33, 35 and 58, with rates of 5.11%, 4.19%, 4.13%, 2.50%, and 2.42%. Detailed results are provided in online supplemental table S4.

In the inclusive double hr-HPV infections, 46 186 pairs were detected, with 45 172 pairs (97.80%) in women aged<65. Among these, the most common double-genotype infections were HPV52/58 (n=2423, 5.43%), 16/52 (n=2154, 4.78%), 51/52 (n=1704, 3.75%), 16/18 (n=1464, 3.24%) and 52/68 (n=1453, 3.23%), as shown in figure 2B. In those≥65, 1014 pairs were detected, with the most frequent genotypes being HPV52/58 (n=84, 8.28%), 16/52 (n=53, 5.23%), 52/56 (n=43, 4.24%), 52/68 (n=36, 3.55%) and 16/18 (n=33, 3.25%), as detailed in figure 2C. The CIN2+ rate from double hr-HPV infections in the<65 group was 2.58% (1164/45 172), with the highest detection rates for combinations HPV16/33 (48/505, 9.50%), 16/68 (42/680, 6.18%), 16/31 (23/416, 5.53%), 18/31 (11/203, 5.42%) and 16/58 (64/1,248, 5.13%). In those≥65, the CIN2+ rate was 2.86% (29/1014), with leading combinations being HPV33/39 (1/3, 33.33%), 35/31 (1/3, 33.33%), 18/39 (1/4, 25.00%), 16/39 (2/8, 25.00%) and 18/33 (2/9, 22.22%). Detailed results are provided in online supplemental table S5 and S6.

Age-specific prevalence of different numbers of high-risk HPV infections and their proportion of CIN2+

The number of hr-HPV infections increased from single to triple infections, while the proportion of CIN2+ increased, as detailed in online supplemental table S2 and S3. Hr-HPV infections were predominantly single (6.79%), with multiple infections less common (1.33%). In the≥65 group, the detection numbers of single, double, triple and multiple hr-HPV infections were 1839 (10.56%), 404 (2.32%), 99 (0.57%) and 543 (3.12%), respectively, with CIN2+ detection rates of 2.01% (n=37), 2.73% (n=11), 4.04% (n=4) and 2.95% (n=16). In those<65, the numbers for single, double, triple and multiple infections were 1 144 301 (6.76%), 22 650 (1.06%), 4341 (0.20%) and 28 136 (1.32%), respectively, with CIN2+ detection rates of 1.69% (n=2440), 2.45% (n=554), 2.49% (n=108), and 2.51% (n=705).

The distribution of differents numbers of hr-HPV infection and its CIN2+ detection rates are illustrated in figure 3. Figure 3A shows the proportion of single hr-HPV infections, while figure 3E presents the proportion of CIN2+ cases. This study included 146 140 cases of single hr-HPV infections, with 1839 (1.26%) in women aged≥65. In the<65 group, the five most prevalent hr-HPV genotypes were HPV52, 16, 58, 51 and 18, with infection rates of 24.77%, 19.11%, 10.72%, 7.12% and 6.92%, respectively. The highest CIN2+ detection rates in this group were for HPV16, 33, 58, 31 and 18, with the proportions of 4.53%, 2.27%, 1.84%, 1.78% and 1.59%, respectively. Among those≥65, the most common hr-HPV genotypes were HPV52, 16, 58, 56 and 18, with infection rates of 24.90%, 24.42%, 13.05%, 6.42% and 5.38%, respectively. The highest CIN2+ detection rates in this group were for HPV18, 16, 33, 51 and 58, with the proportions of 5.05%, 4.45%, 2.90%, 1.67% and 1.35%, respectively. Detailed results are provided in online supplemental table S7.

Figure 3

Request permissions

The prevalence of different numbers of high-risk HPV infections and their proportion for CIN2+*. (A–D) The prevalence of different numbers of high-risk HPV infections. Outer ring: age≥65; middle ring: age<64; inner ring: all ages. (E–H) The proportion of CIN2+detected among high-risk HPV infections (the top five hr-HPV genotypes). (A, E) Single high-risk HPV infection; (B, F) Double high-risk HPV infection. (C, G) Triple high-risk HPV infection; (D, H) Multiple high-risk HPV infection. *Note: Double infection refers to the proportion of a specific high-risk HPV genotype in combination with one of the other 13 high-risk HPV genotypes, as well as the proportion of CIN2+ detected within these specific combinations; similar considerations apply to triple and multiple infections. CIN2+, cervical intraepithelial neoplasia grade 2 or worse; HPV, human papillomavirus; HSIL, high-grade squamous intraepithelial lesion.

Figure 3B,F illustrate the proportions of double hr-HPV infections and their corresponding CIN2+ detection rates. In these figures, the specific proportions of double infections formed by particular hr-HPV genotypes combined with one of the other 13 hr-HPV genotypes are illustrated, along with the proportions of CIN2+ detected within these combinations. The study involved 23 054 pairs of infections, with 404 (1.75%) women aged≥65. In the<65 group, the most frequent co-infections involved HPV52 (19.06%), HPV16 (12.67%), HPV58 (11.33%), HPV51 (8.89%) and HPV68 (7.40%). The highest CIN2+ detection rate was for women with HPV16 co-infections at 4.81%, followed by HPV18 (3.69%), HPV33 (3.46%), HPV58 (2.86%) and HPV31 (2.55%). Among those≥65, the highest proportions of co-infections were with HPV52 (20.92%), HPV16 (15.72%), HPV58 (13.74%), HPV68 (7.18%) and HPV56 (6.93%). The highest CIN2+ detection rate was for those with HPV31 co-infections at 5.26%, followed by HPV18 (4.55%), HPV33 (3.70%), HPV58 (3.60%) and HPV35 (3.57%). Detailed results are provided in online supplemental table S8.

This study presents the distribution and CIN2+ detection rates for the 15 most common double hr-HPV infections, as detailed in online supplemental table S9. In women<65, the top five hr-HPV double infections were HPV52/58 (6.53%), 16/52 (5.79%), 16/18 (5.03%), 51/52 (4.48%) and 52/68 (3.82%), and the highest CIN2+proportions were HPV16/33 (9.38%), 18/35 (7.41%), 16/35 (6.60%), 35/45 (5.88%) and 16/58 (5.87%). In the≥65 age group, the most frequent double infections were HPV52/58 (9.90%), 16/52 (7.18%), 16/18 (5.20%), 16/58 (4.46%) and 51/52 (3.96%), and the highest CIN2+proportions were with HPV31/35 (100%), 33/58 (50%), 16/39 (25%), 18/52 (25%) and HPV16/31 (9.09%). The patterns of triple hr-HPV infections, multiple hr-HPV infections and their corresponding CIN2+ detection rates are similar to those observed for single and double infections. These patterns are illustrated in figure 3C and G for triple infections and figure 3D and H for multiple infections, respectively. Detailed results are provided in online supplemental table S10.

Discussion

This study analysed data from 2 152 766 women undergoing cervical cancer screening. Results indicated that the hr-HPV infection rate and the CIN2+ detection rate were higher in women aged≥65 compared with those<65. There was a dose-response relationship between the number of hr-HPV infections and the risk of CIN2+ in the ≥65 women. These findings suggest a higher risk of cervical cancer in older women. In women aged≥65, the distribution of single and double hr-HPV infections differed in their pathogenic potential to cause CIN2+, and the distribution was not the same as that in those<64. These findings highlight unique characteristics of HPV infection in older women. Therefore, the importance and necessity of cervical cancer prevention and control in older women should be recognised.

Cervical cancer remains one of the most common cancers among women worldwide. Despite declines in incidence due to widespread screening and vaccination, older women continue to face significant cervical cancer risks. WHO statistics showed that women aged≥65 accounted for 23.7% (157 182/662 301) of new cervical cancer cases and 35.6% (124 269/348 874) of deaths.3 These data indicate that women≥65 are a high-risk group for cervical cancer incidence and mortality, necessitating urgent attention from countries worldwide. Currently, most countries recommend ceasing screening at age 64 or 65,12–16 including China, with only a few extending to ages 6917–19 or 74.20 21 Most guidelines suggest stopping screening for those with adequate primary screening and no high-risk factors, particularly for women under 65. However, the situation differs for those≥65, who may not have been vaccinated or thoroughly screened. With increasing life expectancy, the risk of cervical cancer in this demographic is significantly heightened. Research by Malagón et al 22 indicates that cervical cancer prevention in later life largely relies on screening before age 55. Current screening policies fail to adequately protect this high-risk group (women aged≥65), exacerbating their disease burden and demanding urgent reconsideration.

To eliminate cervical cancer, defined as reducing incidence to below 4 per 100 000,23 effective measures must be implemented to prevent new cases and promptly screen and treat existing high-risk populations. Although many countries initiated cervical cancer prevention efforts early on, the lack of comprehensive preventive strategies initially limited their effectiveness. For instance, China began cervical cancer screening in 2009, yet the incidence and mortality rates remain high. Global attention to cervical cancer prevention significantly increased following the WHO’s 2018 call for action to eliminate the disease.1 In 2020, the WHO launched a global strategy to accelerate the elimination of cervical cancer, aiming for a ‘90-70-90’ target by 2030,24 leveraging advances in HPV vaccination and cervical screening technologies. The WHO currently recommends prioritising cervical cancer screening for the 30–49 age group if universal screening for women aged 30 and above is not feasible,6 as this is a high-risk period for cervical cancer.22 This targeted approach is cost-effective in resource-limited settings. While some models suggest lower cost-effectiveness for screening women aged≥65,5 25 this group bears a significant disease burden.7 26 Therefore, countries should tailor strategies to their realities, aiming to reduce incidence and mortality in older women.

As population ageing intensifies, the risks of hr-HPV infection and cervical cancer among older women become increasingly prominent, necessitating targeted public health measures. This study shows that hr-HPV infection rate in women aged≥65 is significantly higher than in those<65 (13.67% vs 8.08%). The elderly declining immunity and postmenopausal hormonal changes may raise hr-HPV susceptibility and cervical cancer risk, especially if they missed early screening. A systematic review by Osmani et al 27 reports hr-HPV prevalence at 6.04% for ages 50–54, dropping to 4.61% at ages 60–64 but rising again to 6.33% at ages 65–69, with regional variations. Hammer et al 28 found the highest cervical cancer incidence at ages 75–79 (29.4 per 100 000 person-years). This study also indicates that the detection rate of CIN2+ in women aged≥65 is higher than in<65 (3.33‰ vs 1.55‰), suggesting potential gaps in early screening and intervention for older women. Typically, ageing is associated with decreased hormone levels, menopausal status and increased likelihood of epithelial and cervical atrophy, as well as vaginal narrowing, raising the probability of transformation zone 3,29 30 and thus the risk of missed colposcopy diagnoses.31 32 Studies indicate that cervical cancer mortality significantly rises with age,26 33 particularly in women aged≥65 for whom routine screening is not recommended.33 Therefore, given the high infection and lesion detection rates in women aged≥65, public health policies should reconsider cervical cancer screening strategies to protect this high-risk group better.

Our study showed a strong association between hr-HPV positivity and CIN2+ risk (OR=203.181) by logistic regression analysis, consistent with existing studies.34 35 The study revealed significant OR differences between the<65 and ≥65 age groups. Notably, the impact of hr-HPV infection was more pronounced in women<65 years (OR=206.704) compared with those≥65 years (OR=59.528). This discrepancy may stem from age-related differences in immune response, hormones and cervical cell biology. Younger women often clear hr-HPV more efficiently, leading to a relatively higher risk of lesions despite the same infection. The study also found a dose-response relationship between the number of hr-HPV infections and CIN2+risk, aligning with most current research.15 35–38 This result suggests potential synergistic effects among different genotypes, accelerating cervical lesion progression. However, some studies argue that multiple HPV infections show no significant difference in the risk of cervical precancerous lesions.39–41 In this study, the OR for triple hr-HPV in women aged≥65 was 85.447, which, though lower than that for women<65 (OR=330.598), still showed an upward trend. Notably, the incidence of single, double and triple hr-HPV infections, as well as the proportion causing CIN2+, was higher in women aged≥65 compared with those<65. This finding underscores the clinical importance of recognising multiple hr-HPV infections, suggesting screening and treatment strategies should account for infection genotype and number. Therefore, screening policies could be personalised based on hr-HPV infection risk across different age groups.

This study highlights the distribution characteristics of hr-HPV and hr-HPV genotypes associated with CIN2+in older women, uncovering the distribution of hr-HPV infection differs from pathogenic genotypes. Among women aged≥65, HPV52, HPV16 and HPV58 were the most prevalent hr-HPV genotypes, while HPV18, HPV16 and HPV33 showed the highest CIN2+ detection rates. This finding indicates that although HPV52 is highly prevalent in this population, its capacity to cause CIN2+ is lower than HPV18 and HPV16. Most studies report similar overall trends in hr-HPV distribution despite some regional variations.42–45 Additionally, inclusive double hr-HPV infections, such as HPV52/58, HPV16/52 and HPV52/56, were common in this age group. The combinations HPV33/39, HPV35/31 and HPV18/39 were associated with the highest CIN2+ rates. Co-infections may contribute to more complex clinical presentations, underscoring the importance of considering such risks in cervical screening strategies for older women.

The complexity of diagnosing and treating cervical cancer is heightened in older women who often suffer from comorbidity, further threatening their quality of life. Unique challenges in disease management include lower health awareness and reluctance or inability to undergo regular screening. Currently, there is a significant global gap in cervical cancer prevention for older women, and urgent action is needed. First, screening and early diagnosis for women aged≥65 should be strengthened, including affordable screening services and age-appropriate technologies to detect and treat precancerous lesions. Additionally, community engagement, health education and media campaigns can raise awareness of cervical cancer risks and prevention among older women, encouraging active participation in screening programmes. Healthcare providers also need specialised training to address the unique needs of older women, offering personalised treatment and psychological support. Furthermore, governments and health organisations must increase policy support and funding to ensure adequate prevention and treatment services resources. In conclusion, effective cervical cancer prevention for women≥65 requires collaborative efforts and comprehensive measures to reduce incidence, improve survival rates and ultimately enhance the quality of life for this population.

This study used high-quality, large-sample, real-world cervical cancer screening data collected over 7 years in Shenzhen, providing strong evidence for the importance of cervical cancer screening in women aged≥65. However, there are limitations to this study. First, the data is from a screening population, with women≥65 not included in the national target screening population. Those who participated may have had symptoms or concerns, introducing potential bias. Second, the low number of hr-HPV infections in this age group led to some results trending towards extremes, compromising stability. Third, the data comes from one region in China, limiting its applicability to other populations. Fourth, this study was unable to obtain specific information on the types of cytological detection products and HPV genotyping products used. Variations in the standards and accuracy of different LBC products and HPV genotyping products may have impacted the precision and comparability of the results. Lastly, while HPV vaccination may influence the prevalence of high-risk HPV types, the low vaccination rate (2.25%) in the screening population likely had minimal impact on our findings. Future research should focus on the biological mechanisms of hr-HPV infection in women≥65 and explore its specific relationship with cervical cancer incidence, providing a scientific basis for more targeted prevention strategies.

Conclusion

This study shows that the prevalence of hr-HPV and its association with CIN2+ are higher in women aged≥65, with a dose-response relationship between hr-HPV infection and CIN2+ risk. These findings highlight the importance of regular screening in this population. Additionally, the distribution and pathogenic types of HPV in women aged≥65 differ from those in younger women, suggesting a unique cervical cancer burden in this group. These findings emphasise the urgent need for more effective screening and intervention strategies for women aged≥65. Therefore, continued cervical cancer screening is necessary for this age group, although national policies should be adapted to local conditions.