ABSTRACT
Objective
Cervical cancer poses a significant burden globally, particularly in developing countries, with China accounting for 20% of the global incidence. This study aimed to examine temporal changes in the pathological distribution of cervical cancer, particularly squamous cell carcinoma (SCC) and adenocarcinoma (ADC), in China from 2013 to 2022.
Methods
A nationwide, multicenter, retrospective study analyzed 6645 cervical cancer cases diagnosed between 2013 and 2022 across seven tertiary hospitals in China. The study analyzed demographic, obstetric, and reproductive factors to identify trends in the proportions of SCC and ADC. Multivariable logistic regression was further performed among women with SCC or ADC to identify factors independently associated with histological type.
Results
SCC accounted for 86.9% and ADC for 10.4% of invasive cervical cancer (ICC) cases. Although no significant change in SCC proportion was observed overall, declines were observed in specific subgroups—unemployed women, premenopausal women, and those with a history of reproductive system disease (p trend = 0.012, 0.021, and 0.011, respectively). In multivariable analysis, compared with Central China, the odds of ADC relative to SCC were significantly higher in North China, Southwest China, East China, and South China, whereas menopausal women, Uyghur women, and rural residents showed lower odds of ADC relative to SCC.
Conclusions
Although the overall proportion of SCC among ICC cases remained statistically stable during 2013–2022, SCC showed a decreasing trend in several high‐risk subpopulations. Compared with our 2005–2010 series, the present 2013–2022 study showed a higher ADC proportion and a lower SCC proportion. These findings suggest a gradual shift in the pathological landscape of ICC in China and provide up‐to‐date nationwide multicenter data that may help refine future screening and prevention strategies.
Keywords: cervical adenocarcinoma, cervical cancer, human papillomavirus (HPV), proportion, squamous cell carcinoma, temporal change
1. Introduction
Globally, more than 85% of cervical cancer cases occur in developing countries [1]. China, in particular, bears a substantial burden, accounting for approximately 20% of newly diagnosed cervical cancer cases worldwide—an estimated 150,700 annually. The country also reports about 55,700 cervical cancer‐related deaths each year, representing 15% of the global total [2]. While the incidence of cervical cancer has been declining in many Western nations, it continues to rise in China [3, 4, 5, 6, 7]. Persistent infection with high‐risk human papillomavirus (HR‐HPV) is the primary cause of cervical precancerous lesions and the development of cervical cancer.
Squamous cell carcinoma (SCC) and adenocarcinoma (ADC) are the two dominant pathological types of invasive cervical cancer (ICC), accounting for 97.1% of all ICC cases. SCC comprises 91.9% of these, while ADC represents only 5.2% [8]. Notably, approximately 90% of SCC cases are attributed to HPV infection, whereas only 62% of ADC cases are linked to HPV. This difference highlights the superior effectiveness of the HPV vaccine in preventing SCC. Furthermore, cytological and HPV screening are more effective in detecting SCC and its precancerous lesions than ADC and its precursors, which are relatively more challenging to identify.
Both domestic and international studies have shown that the implementation of vaccines and screening technologies increases the likelihood of early prevention or diagnosis of SCC and its precancerous stages, leading to a shift in the proportions of SCC and ADC in ICC. Therefore, in many countries, the proportion of SCC cases has been decreasing while the proportion of ADC cases has been rising [9]. Given the considerable difference in HPV‐related risk between SCC and ADC, as the proportion of ADC increases, the overall HPV‐attributable risk in ICC may decrease. This shift will inevitably impact the current effectiveness of HPV vaccination and screening programs aimed at early diagnosis and prevention of ICC. Therefore, understanding the changing trends of ADC in ICC in China will provide a theoretical foundation for adjusting vaccination and screening strategies, ultimately helping further optimize these strategies to improve survival and quality of life for female patients.
Several domestic studies have shown that the proportion of ADC varies significantly, ranging from 3% to 20% [10, 11, 12, 13]. However, the discrepancies across studies are substantial, as many of these studies are single‐center investigations that lack representative data on the temporal trends of ADC proportion.
To further refine the research and investigate how different risk factors, population origins, and other variables affect the proportions of SCC and ADC in ICC, we conducted a nationwide multicenter retrospective study. This study reviewed the pathological types of cervical cancer over the past decade, selecting confirmed cases from tertiary hospitals to minimize bias caused by variations in physicians' diagnostic capabilities. The aim of this study was to clarify recent trends in the histological distribution of SCC and ADC in invasive cervical cancer in China and to discuss their potential implications in the context of established differences in HPV prevalence and genotype attribution between these histological types.
2. Method and Materials
2.1. Study Design
2.1.1. Selection of Regions and Hospitals
With a cervical cancer incidence rate of 17.7 per 100,000 in China [5], the number of cases identified through population screening is insufficient to meet the sample size requirements. Consequently, this study included cervical cancer patients diagnosed between 2013 and 2022 from seven tertiary hospitals representing seven regions of China: North China, Northeast, Northwest, Central China, South China, East China, and Southwest. These hospitals are renowned for their high diagnostic standards, ensuring consistency and reliability in pathology, thereby minimizing potential diagnostic variability. The selected institutions include Liaoning Cancer Hospital, the Senior Department of Obstetrics & Gynecology of Chinese People's Liberation Army General Hospital, the Affiliated Cancer Hospital of Xinjiang Medical University, Henan Cancer Hospital, Zhejiang Cancer Hospital, Yunnan Cancer Hospital, and the First Affiliated Hospital of Jinan University.
2.1.2. Study Population
Before launching the main data collection, we conducted a small feasibility assessment in centers where basic statistical functions were available in the hospital information systems, in order to assess the availability and basic distribution of cervical cancer cases over the calendar year. In this assessment, local staff used hospital information systems and pathology registries to identify records with a diagnosis of ICC and to verify that such cases occurred in all months of the year, without abstracting detailed clinicopathological variables or histological subtypes. This preliminary work confirmed that cervical cancer cases were diagnosed throughout the year and that it was technically possible to identify all cervical cancer cases within a predefined sampling month at each center.
By contrast, the full study required manual retrieval of individual records and extraction of detailed demographic, reproductive, and pathological information for every eligible case. To balance representativeness with these workload and resource constraints, we therefore adopted a systematic time‐sampling strategy and defined March of each year from 2013 to 2022 as the common enrolment month in all centers.
The inclusion criteria were as follows: (1) initial diagnosis of cervical cancer between 2013 and 2022; (2) female patients aged 18 years or older. Exclusion criteria were as follows: (1) duplicate records identified during data cleaning; (2) records with irreconcilable logical inconsistencies, such as a first‐visit date occurring after the discharge date or clear disagreement between discharge and clinical diagnoses, for which no additional information was available to correct the error. (3) records with missing key variables (including age at diagnosis, year of diagnosis, hospital region, or histological subtype).
2.2. Data Management and Statistical Analysis
Data management across all project sites followed standardized procedures to ensure consistency and comparability, with strict adherence to Standard Operating Procedures (SOPs) and privacy regulations. After data entry and verification of key variables, data were anonymized. A robust data control mechanism was established, including data cleaning, logical checks, validation, and review processes. Variables were sourced from the Hospital Information System (HIS) of each center, with data entry carried out by at least two trained data stewards. Statistical analysis categorized continuous variables like Body Mass Index (BMI) based on recognized standards and applied Restricted Cubic Splines (RCS) to improve model prediction. Multivariable logistic regression models were then used to estimate the association between independent variables and cervical cancer pathology (SCC vs. ADC), providing adjusted odds ratios (ORs) and 95% confidence intervals (CIs). Categorical predictors were coded as sets of indicator variables so that odds ratios could be estimated for each subgroup within a given variable. Temporal trends in the proportion of SCC and ADC among all invasive cervical cancer cases were evaluated using chi‐square tests for heterogeneity across calendar years and logistic regression models with year entered as a continuous variable, from which p values for linear trend (p trend) were derived.
2.3. Ethics Statement
This study was approved by the Institutional Review Board (IRB) by the Declaration of Helsinki (amended October 2013) and the Measures for the Ethical Review of Biomedical Research Involving Humans (Order No. 11). As this is a retrospective non‐interventional study, ethical approval was granted with an exemption from informed consent by the IRB/Ethics Committee. Approval was granted by the Ethics Committee of Biomedical Research at the Chinese Academy of Medical Sciences and Peking Union Medical College (Approval No. CAMS&PUMC‐2022‐060).
3. Results
3.1. Basic Information of Study Subjects
Grouping of continuous variables based on RCS curves is shown in Figure S1. A total of 6752 cervical cancer cases met the inclusion criteria for this study. After excluding 20 duplicate cases, 25 cases with logical errors, and 62 cases with missing key variables, a total of 6645 cases were included in the statistical analysis. Of the 62 records with missing key variables, 34 (0.5% of all extracted records) lacked a coded histological subtype in the hospital information system, which were scattered across all seven participating regions (range 1–9 cases per region), while the remaining records were missing diagnosis date, age, or cervical cancer diagnostic information. The inclusion and exclusion process is shown in Figure 1.
FIGURE 1.
Flowchart of case identification, exclusions, and final analytic sample (China, 2013–2022).
Among the 6645 included cases, 5772 were SCC (86.9%, 95% CI: 86.0%–87.6%) and 690 were ADC (10.4%, 95% CI: 9.7%–11.1%). In addition, 96 cases (1.4%) were adenosquamous carcinoma (ASC) and 87 cases (1.3%) were other rare histological subtypes, including neuroendocrine carcinoma, carcinosarcoma, basal cell carcinoma, mixed tumors, melanocytic tumors and a few other specified types (Figure 2). At a significance level of 0.05, the study achieved a statistical power (1‐β) of 97.85%, indicating robust statistical efficacy.
FIGURE 2.
Distribution of different pathological types of cervical cancer. ADC, adenocarcinoma; ASC, adenosquamous carcinoma; PNET, primitive neuroectodermal tumor; MPNST, malignant peripheral nerve sheath tumor; SCC, squamous cell carcinoma. The left panel shows the overall proportions of SCC, ADC, ASC, and other rare histological subtypes. The right panel shows the annual distribution of rare histological subtypes categorized as “other” from 2013 to 2022, including carcinosarcoma, basal cell carcinoma, mixed tumors, melanocytic tumors, neuroendocrine carcinoma, PNET, and MPNST.
The demographic, obstetric, and gynecological characteristics of the study population, refined based on RCS curves, are presented in Table 1. Most participants were diagnosed with cervical cancer at age ≥ 51 (54.0%), while only 2.4% were diagnosed at age ≤ 30, aligning with the typical age‐related pathogenesis of cervical cancer in Chinese women. The mean age at diagnosis was 51.8 years (95% CI: 51.5–52.0). Han Chinese comprised the majority of the cohort (87.9%), with representation from 33 ethnic minority groups, as shown in Figure S2. The mean age at menarche was 15.1 years (95% CI: 14.9–15.2), with 69.0% experiencing menarche at ≤ 15 years. The mean age at first coitus was 21.6 years (95% CI: 21.3–21.8), with 39.6% reporting first sexual intercourse at < 18 years, 48.8% between 18 and 22 years, and 11.6% at > 22 years.
TABLE 1.
Participant demographics.
| Characteristic | Number of cases | Proportion (%) | Characteristic | Number of cases | Proportion (%) |
|---|---|---|---|---|---|
| Age at first cervical cancer diagnosis (years) | Place of residence | ||||
| ≤ 30 | 161 | 2.4 | Urban | 3064 | 46.8 |
| 31–50 | 2895 | 43.6 | Rural | 3478 | 53.1 |
| ≥ 51 | 3589 | 54.0 | Other | 7 | 0.1 |
| Total | 6645 | 100.0 | Total | 6549 | 100.0 |
| Ethnicity | Region of origin | ||||
| Han | 5832 | 87.9 | North China | 996 | 15.0 |
| Uyghur | 398 | 6.0 | Central China | 988 | 14.9 |
| Manchu | 69 | 1.0 | Northeast China | 980 | 14.7 |
| Yi | 57 | 0.9 | Northwest China | 1004 | 15.1 |
| Hui | 56 | 0.8 | Southwest China | 1013 | 15.2 |
| Other | 224 | 3.4 | East China | 1076 | 16.2 |
| Total | 6636 | 100.0 | South China | 588 | 8.8 |
| Total | 6645 | 100.0 | |||
| Marital status | Occupation | ||||
| Married | 6074 | 93.5 | Unemployed | 1370 | 25.9 |
| Unmarried | 70 | 1.1 | Self‐employed | 160 | 3.1 |
| Widowed | 198 | 3.0 | Worker | 270 | 5.1 |
| Divorced | 154 | 2.4 | Farmer | 2246 | 41.9 |
| Total | 6496 | 100.0 | Enterprise/Institution employee | 531 | 10.0 |
| BMI | Private business owner | 127 | 2.5 | ||
| ≤ 23.6 | 2903 | 50.8 | Retiree | 496 | 9.5 |
| > 23.6 | 2811 | 49.2 | Other | 108 | 2.1 |
| Total | 5714 | 100 | Total | 5418 | 100.0 |
| Menopausal status | Age at menarche | ||||
| No | 2865 | 45.1 | ≤ 15 years | 4074 | 69.0 |
| Yes | 3485 | 54.9 | > 15 years | 1829 | 31.0 |
| Total | 6350 | 100.0 | Total | 5903 | 100.0 |
| Age at first coitus | History of reproductive diseases | ||||
| > 22 years | 79 | 11.6 | No | 3577 | 77.6 |
| 18–22 years | 331 | 48.8 | Yes | 1031 | 22.4 |
| < 18 years | 269 | 39.6 | Total | 4608 | 100.0 |
| Total | 679 | 100.0 | |||
| Contraceptive methods | Age at first childbirth | ||||
| Condom | 123 | 2.8 | < 23 years | 840 | 40.3 |
| None | 2609 | 58.8 | ≥ 23 years | 1242 | 59.7 |
| Oral contraceptives | 28 | 0.6 | Total | 2082 | 100.0 |
| Intrauterine device | 708 | 16.0 | Gravidity | ||
| Sterilization | 942 | 21.2 | 0 | 62 | 1.0 |
| Other | 26 | 0.6 | 1–3 | 3589 | 57.3 |
| Total | 4436 | 100 | > 3 | 2609 | 41.7 |
| Total | 6260 | 100.0 | |||
3.2. Trends in the Proportions of SCC and ADC
The proportion of SCC among ICC cases during 2013–2022 showed no statistically significant change over time (χ 2 = 13.507, p = 0.141, p trend = 0.218) as illustrated in Figure 3. Within the same period, the annual proportion of ADC among ICC cases fluctuated around 10%, ranging from 8.7% to 13.4% (8.7% in 2017 and 13.4% in 2022), without evidence of a significant linear temporal trend (p trend = 0.195; Figure 3B). Year‐to‐year changes in ADC tended to mirror reciprocal changes in SCC, consistent with the fact that SCC and ADC together account for the vast majority of ICC cases while other histological subtypes remain rare and relatively stable over time. When stratified by region, no significant linear decrease in the proportion of SCC or ADC was observed in any of the seven regions (all p trend > 0.05 for all; Table S1).
FIGURE 3.
Comparison of Annual SCC and ADC Percentage (2005–2010 vs. 2013–2022) with annual data for 2013–2022.
3.3. SCC and ADC Risk Groups
We performed a multivariable logistic regression analysis restricted to women with SCC or ADC, with histological type as the dependent variable (ADC = 1, SCC = 0). Univariable associations for all candidate variables are presented in Table 2 and Table S2, whereas the multivariable model was limited to variables with relatively complete data and adequate category sizes in order to preserve sample size and model stability (Table 3). After adjustment, significant associations with histological type were observed for region, menopausal status, ethnicity, and place of residence.
TABLE 2.
Univariable associations between clinicodemographic factors and adenocarcinoma (ADC).
| Category | ADC | Total | Proportion (%) a | OR | 95% CI | p | p χ 2 |
|---|---|---|---|---|---|---|---|
| Region | |||||||
| Central China | 71 | 988 | 7.2% | Reference | — | — | < 0.001* |
| North China | 124 | 996 | 12.4% | 1.837 | (1.352, 2.494) | < 0.001* | |
| Northeast China | 87 | 980 | 8.9% | 1.258 | (0.908, 1.745) | 0.168 | |
| Northwest China | 82 | 1004 | 8.2% | 1.149 | (0.825, 1.599) | 0.411 | |
| Southwest China | 140 | 1013 | 13.8% | 2.071 | (1.534, 2.796) | < 0.001* | |
| East China | 119 | 1076 | 11.1% | 1.606 | (1.181, 2.184) | 0.003* | |
| South China | 67 | 588 | 11.4% | 1.661 | (1.17, 2.359) | 0.005* | |
| Occupation | |||||||
| Unemployed | 120 | 1401 | 8.6% | Reference | — | — | < 0.001* |
| Self‐employed | 11 | 166 | 6.6% | 0.758 | (0.400, 1.436) | 0.395 | |
| Workers | 35 | 276 | 12.7% | 1.55 | (1.038, 2.315) | 0.032* | |
| Farmers | 220 | 2271 | 9.7% | 1.145 | (0.907, 1.446) | 0.255 | |
| Enterprise/Institution employee | 85 | 540 | 15.7% | 1.994 | (1.481, 2.686) | < 0.001* | |
| Private business owners | 103 | 134 | 76.9% | 1.873 | (1.124, 3.121) | 0.016* | |
| Retirees | 59 | 516 | 11.4% | 1.378 | (0.991, 1.916) | 0.056 | |
| Others | 9 | 114 | 7.9% | 0.915 | (0.452, 1.854) | 0.805 | |
| Ethnicity | |||||||
| Han | 620 | 5832 | 10.6% | Reference | — | — | 0.005* |
| Hui | 8 | 56 | 14.3% | 1.401 | (0.660, 2.975) | 0.380 | |
| Manchu | 7 | 69 | 10.1% | 0.949 | (0.433, 2.083) | 0.896 | |
| Uyghur | 18 | 398 | 4.5% | 0.398 | (0.246, 0.644) | < 0.001* | |
| Yi | 8 | 57 | 14.0% | 1.372 | (0.647, 2.911) | 0.409 | |
| Other | 28 | 224 | 12.5% | 1.201 | (0.801, 1.800) | 0.375 | |
| Age at diagnosis | |||||||
| ≤ 30 years | 22 | 161 | 13.7% | Reference | — | — | < 0.001* |
| 31–50 years | 357 | 2895 | 12.3% | 0.889 | (0.559, 1.412) | 0.618 | |
| ≥ 51 years | 311 | 3589 | 8.7% | 0.599 | (0.377, 0.954) | 0.031* | |
| Place of residence | |||||||
| Urban | 355 | 3064 | 11.6% | Reference | — | — | 0.016* |
| Rural | 327 | 3478 | 9.4% | 0.792 | (0.676, 0.928) | 0.004* | |
| Other | 0 | 7 | 0.0% | — | — | — | |
| BMI | |||||||
| ≤ 23.6 | 305 | 2903 | 10.5% | Reference | — | — | — |
| > 23.6 | 298 | 2811 | 10.6% | 1.01 | (0.853, 1.196) | 0.907 | |
| Menopausal status | |||||||
| No | 373 | 2865 | 13.0% | Reference | — | — | — |
| Yes | 287 | 3485 | 8.2% | 0.600 | (0.510, 0.705) | < 0.001* | |
| Age at menarche | |||||||
| ≤ 15 years | 438 | 4074 | 10.8% | Reference | — | — | — |
| > 15 years | 168 | 1829 | 9.2% | 0.840 | (0.696, 1.012) | 0.067 | |
| Age at first coitus | |||||||
| > 22 years | 33 | 269 | 12.3% | Reference | — | — | 0.001* |
| 18–22 years | 16 | 331 | 4.8% | 0.363 | (0.195, 0.676) | 0.001* | |
| < 18 years | 2 | 79 | 2.5% | 0.186 | (0.044, 0.792) | 0.023* | |
| Contraceptive methods | |||||||
| Condom | 24 | 123 | 19.5% | Reference | — | — | 0.006* |
| None | 235 | 2609 | 9.0% | 0.408 | (0.256, 0.650) | < 0.001* | |
| Oral contraceptives | 3 | 28 | 10.7% | 0.495 | (0.138, 1.777) | 0.281 | |
| Intrauterine device | 81 | 708 | 11.4% | 0.533 | (0.322, 0.881) | 0.014* | |
| Sterilization | 97 | 942 | 10.3% | 0.474 | (0.289, 0.775) | 0.003* | |
| Other | 2 | 26 | 7.7% | 0.344 | (0.076, 1.556) | 0.166 | |
| Gravidity | |||||||
| 0 | 17 | 62 | 27.4% | Reference | — | — | < 0.001* |
| 1–3 | 402 | 3589 | 11.2% | 0.334 | (0.189, 0.589) | < 0.001* | |
| > 3 | 236 | 2609 | 9.0% | 0.263 | (0.148, 0.467) | < 0.001* | |
| Age at first childbirth | |||||||
| ≥ 23 years | 166 | 1242 | 13.4% | reference | — | — | — |
| < 23 years | 77 | 840 | 9.2% | 0.654 | (0.492, 0.87) | 0.004* | |
| Number of abortions | |||||||
| Continuous variable | 597 | 5717 | 10.4% | 1.01 | (0.946, 1.078) | 0.769 | — |
Note: “Reference” indicates the control group in the comparison; the OR values for other subgroups are obtained by comparing each subgroup to the reference group. For variables that are dichotomized or continuous, the OR values and corresponding p values are sufficient to indicate differences. The overall Wald chi‐square value and its p value are not provided, and “—” is used to represent this.
Abbreviations: ADC, adenocarcinoma; CI, confidence interval; OR, odds ratio.
Represents the ADC incidence proportion, calculated as ADC cases/total cases.
Indicates p < 0.05, indicating statistical significance.
TABLE 3.
Multivariable logistic regression analysis of factors associated with ADC versus SCC.
| Variable | Category | SCC n (%) | ADC n (%) | Adjusted OR (95% CI) | p |
|---|---|---|---|---|---|
| Region | Central China | 870 (16.1) | 70 (10.8) | Reference | — |
| North China | 715 (13.2) | 109 (16.7) | 1.70 (1.22–2.36) | 0.002* | |
| Northeast China | 800 (14.8) | 84 (12.9) | 1.24 (0.89–1.75) | 0.21 | |
| Northwest China | 890 (16.4) | 82 (12.6) | 1.41 (0.98–2.04) | 0.065 | |
| Southwest China | 827 (15.3) | 133 (20.4) | 1.87 (1.38–2.57) | < 0.001* | |
| East China | 856 (15.8) | 111 (17.1) | 1.66 (1.22–2.29) | 0.002* | |
| South China | 460 (8.5) | 62 (9.5) | 1.66 (1.15–2.38) | 0.006* | |
| Age at diagnosis | ≤ 30 years | 126 (2.3) | 21 (3.2) | Reference | — |
| 31–50 years | 2275 (42.0) | 337 (51.8) | 0.94 (0.59–1.55) | 0.787 | |
| ≥ 51 years | 3017 (55.7) | 293 (45.0) | 0.89 (0.53–1.54) | 0.66 | |
| Menopausal status | No | 2350 (43.4) | 368 (56.5) | Reference | — |
| Yes | 3068 (56.6) | 283 (43.5) | 0.64 (0.49–0.83) | < 0.001* | |
| Ethnicity | Han | 4721 (87.1) | 584 (89.7) | Reference | — |
| Uyghur | 371 (6.8) | 18 (2.8) | 0.45 (0.25–0.77) | 0.005* | |
| Other minority | 326 (6.0) | 49 (7.5) | 1.11 (0.79–1.53) | 0.535 | |
| Residence | Urban | 2497 (46.1) | 341 (52.4) | Reference | — |
| Rural | 2921 (53.9) | 310 (47.6) | 0.82 (0.68–0.98) | 0.026* |
Note: Table 2 shows univariable analyses for all variables of interest. For the multivariable model (Table 3), only variables with relatively complete data and sufficient category sizes were retained to avoid substantial loss of observations under complete‐case analysis and to improve model stability. Several reproductive variables (including age at first sexual intercourse, contraceptive use, pregnancy number, and age at first birth) were therefore not entered into the final multivariable model because of extensive missingness.
Abbreviations: ADC, adenocarcinoma; CI, confidence interval; OR, odds ratio; SCC, squamous cell carcinoma.
Indicates p < 0.05, indicating statistical significance.
Compared with Central China, the odds of ADC relative to SCC were significantly higher in North China (aOR 1.70, 95% CI 1.22–2.36), Southwest China (aOR 1.87, 95% CI 1.38–2.57), East China (aOR 1.66, 95% CI 1.22–2.29), and South China (aOR 1.66, 95% CI 1.15–2.38). By contrast, menopausal women showed lower odds of ADC relative to SCC than non‐menopausal women (aOR 0.64, 95% CI 0.49–0.83). Uyghur women also had lower odds of ADC relative to SCC than Han women (aOR 0.45, 95% CI 0.25–0.77), and rural residents had lower odds of ADC relative to SCC than urban residents (aOR 0.82, 95% CI 0.68–0.98). Age group was not independently associated with histological type after adjustment. Detailed results are presented in Table 3.
3.4. Trends in the Proportions of Subgroup Populations
Subgroup analyses were conducted by categorizing the enrolled women based on demographic and obstetric/gynecological information. Trends in SCC incidence over time were observed in subgroups defined by occupation, ethnicity, menopausal status, age at first sexual intercourse, number of pregnancies, and history of reproductive system diseases. Statistically significant decreasing trends in SCC risk (OR < 1, p < 0.05) were found among unemployed or self‐employed women, women aged 31–50 years, women from minority groups labeled as “other,” premenopausal women at diagnosis, women whose first sexual intercourse occurred between ages 18 and 22, those with no pregnancies, and women with a history of reproductive system diseases (Table 4).
TABLE 4.
Regression results of year variable for SCC risk among different populations.
| Category | OR | 95% CI | p | Category | OR | 95% CI | p | ||
|---|---|---|---|---|---|---|---|---|---|
| Lower | Upper | Lower | Upper | ||||||
| Occupation | Ethnicity | ||||||||
| Unemployed/Self‐employed | 0.929 | 0.877 | 0.984 | 0.012* | Han | 0.989 | 0.963 | 1.016 | 0.418 |
| Workers | 1.035 | 0.919 | 1.166 | 0.570 | Hui | 1.198 | 0.903 | 1.588 | 0.210 |
| Farmers | 1.026 | 0.977 | 1.077 | 0.311 | Manchu | 0.955 | 0.742 | 1.229 | 0.719 |
| Enterprise/Institution employees | 0.968 | 0.900 | 1.042 | 0.385 | Uyghur | 1.104 | 0.952 | 1.281 | 0.191 |
| Private business owners | 1.072 | 0.923 | 1.245 | 0.365 | Yi | 0.964 | 0.752 | 1.236 | 0.773 |
| Retirees | 1.020 | 0.934 | 1.114 | 0.661 | Others | 0.831 | 0.721 | 0.957 | 0.010* |
| Others | 0.790 | 0.615 | 1.015 | 0.065 | |||||
| Age | Place of residence | ||||||||
| ≤ 30 years | 1.081 | 0.934 | 1.251 | 0.295 | Urban | 0.980 | 0.947 | 1.015 | 0.263 |
| 31–50 years | 0.961 | 0.928 | 0.996 | 0.031* | Rural | 1.003 | 0.966 | 1.043 | 0.864 |
| ≥ 51 years | 0.983 | 0.946 | 1.022 | 0.386 | |||||
| BMI | Menopausal status | ||||||||
| ≤ 23.6 | 0.995 | 0.958 | 1.033 | 0.795 | No | 0.959 | 0.926 | 0.994 | 0.021* |
| > 23.6 | 0.974 | 0.937 | 1.013 | 0.186 | Yes | 0.995 | 0.957 | 1.034 | 0.787 |
| Age at menarche | Age at first childbirth | ||||||||
| ≤ 15 years | 0.978 | 0.947 | 1.010 | 0.168 | ≥ 23 years | 0.965 | 0.913 | 1.021 | 0.220 |
| > 15 years | 0.986 | 0.937 | 1.037 | 0.588 | < 23 years | 0.969 | 0.895 | 1.049 | 0.436 |
| Age at first coitus | Gravidity | ||||||||
| ≥ 22 years | 0.988 | 0.877 | 1.113 | 0.846 | 0 | 0.706 | 0.549 | 0.908 | 0.007* |
| 18–21 years | 0.826 | 0.717 | 0.952 | 0.008* | 1–3 | 1.007 | 0.973 | 1.042 | 0.696 |
| < 18 years | 1.105 | 0.808 | 1.511 | 0.533 | > 3 | 0.976 | 0.934 | 1.019 | 0.267 |
| Contraceptive methods | Abortion history | ||||||||
| Condom | 0.964 | 0.821 | 1.131 | 0.652 | None | 0.982 | 0.933 | 1.033 | 0.474 |
| None | 0.992 | 0.949 | 1.037 | 0.735 | Yes | 0.994 | 0.962 | 1.027 | 0.721 |
| Oral contraceptives | 1.027 | 0.691 | 1.526 | 0.897 | |||||
| Intrauterine device | 0.978 | 0.908 | 1.053 | 0.551 | History of reproductive diseases | ||||
| Sterilization | 0.973 | 0.905 | 1.047 | 0.470 | No | 0.992 | 0.957 | 1.028 | 0.655 |
| Others | 1.111 | 0.608 | 2.031 | 0.732 | Yes | 0.925 | 0.87 | 0.983 | 0.012* |
p < 0.05, indicating statistically significant differences.
Yearly analysis revealed that for these subgroups, the proportion of SCC cases decreased significantly. For instance, SCC proportions decreased from 92.5% to 84.9% among unemployed/self‐employed women, from 100% to 68.2% in the “other” minority group, from 86.1% to 77.1% among premenopausal women, and from 97.8% to 80.0% for those with first intercourse between ages 18 and 22 (Table 5). Although fluctuations were more pronounced in smaller subgroups, particularly in the “other” minority and no‐pregnancy groups, the data met Chi‐square analysis assumptions. Beyond that, the largest SCC decrease occurred among women with first intercourse between ages 18 and 22 (17.8%), while the smallest decrease was in unemployed/self‐employed women (7.6%) (Table 5, Figure S3).
TABLE 5.
Trends in SCC incidence over time among different populations.
| Year | SCC | Total | Proportion (%) | SCC | Total | Proportion (%) |
|---|---|---|---|---|---|---|
| Unemployed/Self‐employed | Other minority groups | |||||
| 2013 | 132 | 143 | 92.3% | 16 | 16 | 100.0% |
| 2014 | 135 | 146 | 92.5% | 15 | 16 | 93.8% |
| 2015 | 135 | 150 | 90.0% | 23 | 25 | 92.0% |
| 2016 | 107 | 117 | 91.5% | 25 | 31 | 80.6% |
| 2017 | 117 | 129 | 90.7% | 14 | 15 | 93.3% |
| 2018 | 166 | 191 | 86.9% | 32 | 38 | 84.2% |
| 2019 | 185 | 212 | 87.3% | 21 | 26 | 80.8% |
| 2020 | 146 | 168 | 86.9% | 9 | 13 | 69.2% |
| 2021 | 143 | 159 | 89.9% | 20 | 22 | 90.9% |
| 2022 | 129 | 152 | 84.9% | 15 | 22 | 68.2% |
| p trend = 0.012* | p trend = 0.009* | |||||
| Premenopausal women | Age at first coitus (18–22 years) | |||||
| 2013 | 267 | 312 | 85.6% | 54 | 57 | 94.7% |
| 2014 | 290 | 346 | 83.8% | 44 | 45 | 97.8% |
| 2015 | 277 | 326 | 85.0% | 32 | 34 | 94.1% |
| 2016 | 262 | 317 | 82.6% | 24 | 25 | 96.0% |
| 2017 | 267 | 310 | 86.1% | 28 | 32 | 87.5% |
| 2018 | 279 | 329 | 84.8% | 38 | 40 | 95.0% |
| 2019 | 205 | 251 | 81.7% | 33 | 40 | 82.5% |
| 2020 | 191 | 231 | 82.7% | 20 | 24 | 83.3% |
| 2021 | 181 | 220 | 82.3% | 21 | 24 | 87.5% |
| 2022 | 172 | 223 | 77.1% | 8 | 10 | 80.0% |
| p trend = 0.021* | p trend = 0.007* | |||||
| Women with no pregnancies | History of reproductive system diseases | |||||
| 2013 | 5 | 5 | 100.0% | 73 | 84 | 86.9% |
| 2014 | 2 | 3 | 66.7% | 88 | 97 | 90.7% |
| 2015 | 3 | 4 | 75.0% | 111 | 133 | 83.5% |
| 2016 | 7 | 8 | 87.5% | 95 | 116 | 81.9% |
| 2017 | 2 | 2 | 100.0% | 100 | 120 | 83.3% |
| 2018 | 9 | 11 | 81.8% | 117 | 138 | 84.8% |
| 2019 | 4 | 6 | 66.7% | 84 | 102 | 82.4% |
| 2020 | 6 | 10 | 60.0% | 55 | 69 | 79.7% |
| 2021 | 2 | 4 | 50.0% | 71 | 84 | 84.5% |
| 2022 | 3 | 9 | 33.3% | 64 | 88 | 72.7% |
| p trend = 0.004* | p trend = 0.011* | |||||
p < 0.05, indicating statistically significant differences.
4. Discussion
4.1. Proportion and Significance of Different Pathological Types in ICC
This study represents the first multicenter investigation in China to examine shifts in different pathological types in ICC proportions over the past decade, utilizing data from seven tertiary hospitals across diverse regions. We found that the overall proportion of SCC among ICC cases remained statistically stable during 2013–2022. In our previous multicenter study of ICC cases diagnosed between 2005 and 2010, SCC and ADC accounted for 92.9% and 7.1% of cases, respectively [14]. In the present 2013–2022 study, SCC and ADC accounted for 86.9% and 10.4% of cases (ADC: 7.1% vs. 10.4%; SCC: 92.9% vs. 86.9%; both p < 0.001). Although these figures come from two separate cohorts rather than from a pooled raw‐data analysis, this descriptive comparison suggests a gradual shift in the pathological spectrum of ICC in China, with a relative decrease in SCC and an increase in ADC. In our previous multicenter case series conducted between 2005 and 2010, ICC cases were recruited from nine tertiary hospitals distributed across seven major geographic regions of China (Northeast, North, Northwest, Central, Southwest, South and East). The present 2013–2022 multicenter case series also includes tertiary cancer hospitals from these same macroregions, providing broadly comparable geographical and institutional coverage. This suggests that the observed differences in the proportions of SCC and ADC between 2005–2010 and 2013–2022 are unlikely to be explained solely by changes in the regional distribution or hospital level of the participating centers. However, because the two series were assembled in different time periods, individual‐level data from the earlier series were not reanalyzed jointly with the current data, and only a limited set of basic characteristics (such as age and reproductive history) were collected in a comparable manner in 2005–2010, we cannot exclude the possibility that part of the observed changes reflects shifts in patient characteristics or referral patterns over time. Therefore, the cross‐period comparison should be interpreted as descriptive.
SCC and ADC are the most common pathological types of cervical cancer, accounting for 95%–97% of ICC cases in previous studies [8, 15]. Rare pathological types, such as ASC, neuroendocrine carcinoma, and melanoma, comprise only 2.7% of cases in this study, with fluctuations between 2.0% and 3.4% across years. Given that SCC and ADC together account for most ICC cases and that other histological subtypes remain rare and stable, any decline in the proportion of SCC necessarily corresponds to a relatively higher proportion of ADC within ICC, as a compositional effect.
Global multicenter study shows SCC and ADC constitute 91.9% and 5.2% of ICC cases [8], respectively, with a meta‐analysis in Chinese women reporting ADC at 9.0% (95% CI: 7.7%–10.3%). This study aligns with these findings, with SCC at 86.9% and ADC at 10.4%. Current Screening and HPV vaccination programs, which are more effective in detecting SCC, have led to a decline in SCC and a rise in ADC in some regions [16, 17, 18, 19]. In developed countries, increased HPV vaccination and widespread screening have led to a significant decline in cervical cancer incidence, including both SCC and other types. In contrast, in China, with less coverage of screening and preventive measures, as well as higher HPV infection rates, the incidence of SCC, ADC, and cervical cancer overall remains relatively high [20].
4.2. Characteristics of the Population With a Decline in SCC Proportion
This study found that the percentage of ADC was higher in South, East, and Southwest China compared to Central China. This may reflect the more advanced screening tools and higher coverage in South and East China, which likely reduce SCC incidence. In contrast, the limited economic and healthcare resources in Southwest China, along with a higher prevalence of early marriage and early childbearing, may be associated with a higher proportion of ADC, potentially through correlated factors such as early sexual activity onset and lower contraception use [21, 22, 23, 24]. SCC risk was notably higher among Uyghur women, women aged ≥ 51, rural residents, menopausal women, those who had sexual debut by age ≤ 22, those not using contraception or using an IUD, and those with pregnancies or first birth at < 23 years. However, SCC proportion showed a decline over time only in specific subgroups: patients aged 31–50, those with first intercourse at 18–22, non‐menopausal women, certain minority groups, those with no pregnancy history, and those with a history of reproductive system disease. These findings highlight the need to strengthen prevention efforts for both high‐risk ADC and SCC populations.
4.3. Changing Histological Profile of ICC and Its Impact on HPV‐Attributable Burden
While the percentage of SCCs in this study showed no statistically significant change from 2013 to 2022, a significant decrease was observed compared to the 2005–2010 period, indicating a relative decline in SCC and suggesting progress in cervical cancer prevention. According to the latest data from the World Health Organization (WHO), cervical cancer incidence in China is 13.8 per 100,000, with a mortality rate of 4.5 per 100,000. Compared to 2020 (incidence of 10.7 and mortality of 5.3 per 100,000) [2, 25], the incidence has risen slightly. A 2017 report from China's National Cancer Registry showed an annual growth rate in cervical cancer incidence of 10.5% [26], with both incidence and mortality increasing from 2000 to 2016 [27].
HPV prevalence and genotype attribution differ between SCC and ADC; therefore, changes in the SCC‐to‐ADC ratio in ICC cases could influence HPV's contribution to ICC. Previous epidemiological studies and pooled analyses have shown that both SCC and adenocarcinoma are strongly associated with HPV infection, but that the overall HPV DNA positivity is consistently higher in SCC than in ADC. In a large worldwide study based on tumor tissue specimens and sensitive HPV DNA assays, de Sanjosé et al. [8] reported HPV DNA in 87% of SCC cases and 62% of ADC, indicating a substantially larger HPV‐negative or non‐attributable fraction among ADC. Similarly, a large Chinese multicenter study found HPV prevalence of 97.6% in SCC but only 74.5% in ADC [14]. More recent systematic reviews and meta‐analyses of ADC report pooled HPV DNA positivity for Asia is 72.5%, which, while still indicating that most ADCs are HPV‐driven, remains consistently lower than that observed for SCC in comparable datasets [28]. In other words, the fraction of cervical cancers that cannot be attributed to detectable HPV infection appears to be higher in ADC than in SCC.
In the context of primary HPV‐based screening and prophylactic vaccination, which are designed to prevent or detect HPV‐driven disease, an increasing proportion of ADC therefore implies that a slightly larger share of the overall ICC burden may fall outside the direct reach of HPV targeted prevention and screening, even if coverage and test performance are optimal. Our observation of a rising ADC proportion, together with a decreasing trend in SCC in several subgroups, thus highlights the need to monitor HPV‐type specific and HPV attributable disease burden and to ensure that screening strategies retain adequate sensitivity for glandular lesions, including those that may be HR‐HPV negative or more challenging to detect. The complete series from 2013 to 2022 lacks standardized individual‐level HPV genotyping data, which prevents direct analysis of HPV subtype distributions across different histological types in this study. Nevertheless, drawing on the consistent findings from large, well‐characterized external studies allows our inferences to remain scientifically grounded and to provide important contextual information and guidance for future research and policy.
Additionally, the greater diagnostic difficulty of adenoepithelial lesions compared to squamous lesions may further impact the effectiveness of current screening methods, not only HPV testing but also cytologic testing. Given the significant burden of cervical cancer in China, sustained efforts to meet WHO's goal of eliminating cervical cancer are crucial, requiring enhanced prevention, control measures, and high‐quality healthcare services for women.
4.4. Clinicodemographic Correlates of ADC Relative to SCC
Beyond its potential implications for HPV‐attributable burden and screening performance, the changing histological profile of ICC also raises the question of which patient groups are more likely to contribute to the relative increase in ADC. In the multivariable logistic regression analysis, region, menopausal status, ethnicity, and place of residence remained independently associated with histological type. Compared with Central China, the relative odds of ADC were higher in North, Southwest, East, and South China, suggesting that the proportion of glandular tumors is not evenly distributed across China. This geographic heterogeneity may reflect regional differences in healthcare access, screening practices, reproductive patterns, or other unmeasured contextual factors. In contrast, menopausal women, Uyghur women, and rural residents showed lower odds of ADC relative to SCC, indicating that SCC remains disproportionately represented in these populations. These findings suggest that the relative burden of ADC and SCC is shaped not only by temporal change but also by demographic and regional heterogeneity.
Several additional variables, including occupation, age at first sexual intercourse, contraceptive use, number of pregnancies, and age at first birth, showed potentially informative patterns in the univariable analysis, but were not retained in the final multivariable model because of substantial missingness and the resulting loss of sample size and model stability. Therefore, these associations should be interpreted cautiously as exploratory rather than definitive. Nevertheless, they suggest that reproductive and behavioral factors may still contribute to histological variation in cervical cancer and deserve further study in datasets with more complete exposure information.
5. Conclusion
The overall proportion of SCC among ICC cases in China remained statistically stable during 2013–2022. However, ADC proportion has increased markedly compared to our 2005–2010 multicenter study. Although SCC has shown a decreasing trend among certain high‐risk subgroups, further improvement in cervical cancer prevention and control remains necessary. The rising proportion of ADC may affect the prevalence and type distribution of HPV in ICC cases and impacts the effectiveness of current screening methods (HPV tests and cytology). Screening programs should be further developed to closely monitor shifts in ICC pathology types, thereby enhancing screening and diagnostic approaches to improve early detection and treatment outcomes in populations with declining SCC proportion or high ADC prevalence.
Author Contributions
Li Li: investigation. Runhua Li: investigation. Hongping Zhang: investigation. Jia You: investigation. Rufei Duan: investigation. Hong Wang: investigation. Tingting Zuo: investigation. Lingbin Du: investigation. Shuai Yuan: investigation. Weizhao Lin: investigation. Jian Yin: conceptualization, methodology, software, data curation, investigation, validation, formal analysis, supervision, funding acquisition, visualization, project administration, resources, writing – original draft, writing – review and editing. Wen Yang: investigation. Yufei Li: conceptualization, methodology, software, data curation, investigation, validation, formal analysis, supervision, visualization, project administration, resources, writing – original draft, writing – review and editing. Xiaofen Gu: investigation. Zhe Wang: investigation. Huan Yang: writing – review and editing, resources, supervision. Youlin Qiao: conceptualization, methodology, software, data curation, investigation, validation, formal analysis, supervision, funding acquisition, visualization, project administration, resources, writing – original draft, writing – review and editing. Bo Zhu: investigation. Shaokai Zhang: investigation.
Funding
This work was supported by the Chinese Academy of Medical Sciences (CIFMS 2021‐I2M‐1‐004) and in part by a research grant from the Investigator‐Initiated Studies Program of MSD (MISP#100948); CAMS Innovation Fund for Medical Sciences(CIFMS) 2026‐I2M‐SXZX10‐04.
Conflicts of Interest
The authors declare no conflicts of interest.
Supporting information
Table S1: Trends in proportion of ICC across regions from 2013 to 2022.
Table S2: Univariable associations between clinicodemographic factors and cervical squamous cell carcinoma (SCC).
Figure S1: RCS curves of age at first diagnosis, age, BMI, age at first childbirth, age at menarche, and gravidity.
Figure S2: Population distribution by ethnic group.
Figure S3: Time trend of SCC proportion in each subgroup.
Contributor Information
Huan Yang, Email: yanghuan_0722@163.com.
Youlin Qiao, Email: qiaoy@cicams.ac.cn.
Data Availability Statement
The data presented in this study are fully included in the manuscript and Appendix S1, with no additional data required for replication of the findings. Raw data generated during the study are available from the corresponding author upon reasonable request.
References
- 1. WHO Guidelines Approved by the Guidelines Review Committee , WHO Guideline for Screening and Treatment of Cervical Pre‐Cancer Lesions for Cervical Cancer Prevention (World Health Organization, 2021). [PubMed] [Google Scholar]
- 2. IARC , “GLOBOCAN 2022,” https://gco.iarc.who.int/media/globocan/factsheets/populations/160‐china‐fact‐sheet.pdf.
- 3. Lönnberg S., Hansen B. T., Haldorsen T., Campbell S., Schee K., and Nygård M., “Cervical Cancer Prevented by Screening: Long‐Term Incidence Trends by Morphology in Norway,” International Journal of Cancer 137, no. 7 (2015): 1758–1764, 10.1002/ijc.29541. [DOI] [PubMed] [Google Scholar]
- 4. Siwei Z., Kexin S., Rongshou Z., et al., “Cancer Incidence and Mortality in China, 2015,” Journal of the National Cancer Center 1, no. 1 (2021): 2–11, 10.1016/j.jncc.2020.12.001. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Zheng R., Zhang S., Zeng H., et al., “Cancer Incidence and Mortality in China,” Journal of the National Cancer Center 2, no. 1 (2016): 1–9, 10.1016/j.jncc.2022.02.002. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Gu X., Sun G., Zheng R., et al., “Incidence and Mortality of Cervical Cancer in China in 2015,” Journal of the National Cancer Center 2, no. 2 (2015): 70–77, 10.1016/j.jncc.2022.01.002. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Shelton J., Zotow E., Smith L., et al., “25Year Trends in Cancer Incidence and Mortality Among Adults Aged 35–69 Years in the UK, 1993–2018: Retrospective Secondary Analysis,” BMJ 384 (2024): e076962, 10.1136/bmj-2023-076962. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. de Sanjose S., Quint W. G., Alemany L., et al., “Human Papillomavirus Genotype Attribution in Invasive Cervical Cancer: A Retrospective Cross‐Sectional Worldwide Study,” Lancet Oncology 11, no. 11 (2010): 1048–1056, 10.1016/s1470-2045(10)70230-8. [DOI] [PubMed] [Google Scholar]
- 9. Woodman C. B., Collins S. I., and Young L. S., “The Natural History of Cervical HPV Infection: Unresolved Issues,” Nature Reviews. Cancer 7, no. 1 (2007): 11–22, 10.1038/nrc2050. [DOI] [PubMed] [Google Scholar]
- 10. Su F. L. and Kong W. M., “Analysis of the Characteristics and Trend of Cervical Cancer Incidence in Beijing Obstetrics and Gynecology Hospital, Capital Medical University Over the Past 40 Years,” Chinese Journal of Family Planning & Gynecotokology 13, no. 2 (2021): 60–63, 10.3969/j.issn.1674-4020.2021.02.17. [DOI] [Google Scholar]
- 11. Ye J., Dang Y. W., Liu Q. D., et al., “Prevalence of Human Papillomavirus in Women With Cervical Cancer in Guangxi Region,” Shandong Medical Journal 50, no. 46 (2010): 20–22, 10.3969/j.issn.1002-266X.2010.46.012. [DOI] [Google Scholar]
- 12. Li J., Zhang D., Zhang Y., Wang X., Lin Y., and Hu L., “Prevalence and Genotype Distribution of Human Papillomavirus in Women With Cervical Cancer or High‐Grade Precancerous Lesions in Chengdu, Western China,” International Journal of Gynaecology and Obstetrics 112, no. 2 (2011): 131–134, 10.1016/j.ijgo.2010.08.010. [DOI] [PubMed] [Google Scholar]
- 13. Zheng B., Li Z., Griffith C. C., et al., “Prior High‐Risk HPV Testing and Pap Test Results for 427 Invasive Cervical Cancers in China's Largest CAP‐Certified Laboratory,” Cancer Cytopathology 123, no. 7 (2015): 428–434, 10.1002/cncy.21557. [DOI] [PubMed] [Google Scholar]
- 14. Chen W., Sun H., Molijn A., et al., “The Variable Characteristics of Human Papillomavirus in Squamous Cell Carcinoma and Adenocarcinoma of Cervix in China,” Journal of Lower Genital Tract Disease 22, no. 4 (2018): 355–361, 10.1097/lgt.0000000000000408. [DOI] [PubMed] [Google Scholar]
- 15. Yifan M., Tian C., Shitong L., et al., “Clinicopathological Characteristics and Prognosis of Cervical Cancer With Different Histological Types: A Population‐Based Cohort Study,” Gynecologic Oncology 163, no. 3 (2021): 545–551, 10.1016/j.ygyno.2021.10.007. [DOI] [PubMed] [Google Scholar]
- 16. Crosbie E. J., Einstein M. H., Franceschi S., and Kitchener H. C., “Human Papillomavirus and Cervical Cancer,” Lancet 382, no. 9895 (2013): 889–899, 10.1016/s0140-6736(13)60022-7. [DOI] [PubMed] [Google Scholar]
- 17. Bray F., Carstensen B., Møller H., et al., “Incidence Trends of Adenocarcinoma of the Cervix in 13 European Countries,” Cancer Epidemiology, Biomarkers & Prevention 14, no. 9 (2005): 2191–2199, 10.1158/1055-9965.Epi-05-0231. [DOI] [PubMed] [Google Scholar]
- 18. Peter S. and Joanna A., “Changing Rates of Adenocarcinoma and Adenosquamous Carcinoma of the Cervix in England,” Lancet 357, no. 9267 (2001): 1490–1493, 10.1016/S0140-6736(00)04646-8. [DOI] [PubMed] [Google Scholar]
- 19. Vizcaino A. P., Moreno V., Bosch F. X., Muñoz N., Barros‐Dios X. M., and Parkin D. M., “International Trends in the Incidence of Cervical Cancer: I. Adenocarcinoma and Adenosquamous Cell Carcinomas,” International Journal of Cancer 75, no. 4 (1998): 536–545. [DOI] [PubMed] [Google Scholar]
- 20. Bruni L. A. G., Serrano B., Serrano B., et al., “Human Papillomavirus and Related Diseases in the World. Summary Report 17 June 2019,” accessed March 20, 2024, https://hpvcentre.net/statistics/reports/XWX.pdf.
- 21. “Cervical Cancer Screening Rates Among Chinese Women—China, 2015,” accessed September 5, 2026, https://weekly.chinacdc.cn//article/id/2993c668‐204a‐4862‐b5e0‐d7e2d2bbf9dd. [DOI] [PMC free article] [PubMed]
- 22. Xavier‐Júnior J. C., Dufloth R. M., Vale D. B., Lima M. T., and Zeferino L. C., “Early Age at First Sexual Intercourse Is Associated With Higher Prevalence of High‐Grade Squamous Intraepithelial Lesions (HSIL),” Revista Brasileira de Ginecologia e Obstetrícia 39, no. 2 (2017): 80–85. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Zhang R., Shi T. Y., Ren Y., et al., “Risk Factors for Human Papillomavirus Infection in Shanghai Suburbs: A Population‐Based Study With 10,000 Women,” Journal of Clinical Virology 58, no. 1 (2013): 144–148, 10.1016/j.jcv.2013.06.012. [DOI] [PubMed] [Google Scholar]
- 24. Benard V. B., Watson M., Castle P. E., and Saraiya M., “Cervical Carcinoma Rates Among Young Females in the United States,” Obstetrics and Gynecology 120, no. 5 (2012): 1117–1123, 10.1097/aog.0b013e31826e4609. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Sung H., Ferlay J., Siegel R. L., et al., “Global Cancer Statistics 2020: GLOBOCAN Estimates of Incidence and Mortality Worldwide for 36 Cancers in 185 Countries,” CA: A Cancer Journal for Clinicians 71, no. 3 (2021): 209–249, 10.3322/caac.21660. [DOI] [PubMed] [Google Scholar]
- 26. Li X., Zheng R., Li X., et al., “Trends of Incidence Rate and Age at Diagnosis for Cervical Cancer in China, From 2000 to 2014,” Chinese Journal of Cancer Research 29, no. 6 (2017): 477–486, 10.21147/j.issn.1000-9604.2017.06.02. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27. Chen H., Xia C. F., You T. T., Qiao Y. L., and Zhao F. H., “Causes and Countermeasures of the Rapidly Rising Burden on Cervical Cancer in Chinese Women,” Chinese Journal of Epidemiology 5 (2022): 761–765, 10.3760/cma.j.cn112338-20,211,004-00765. [DOI] [PubMed] [Google Scholar]
- 28. Reynders C., Lerho T., Goebel E. A., et al., “Prevalence and Genotype Distribution of Human Papillomavirus in Cervical Adenocarcinoma (Usual Type and Variants): A Systematic Review and Meta‐Analysis,” Journal of Medical Virology 95, no. 10 (2023): e29190, 10.1002/jmv.29190. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Table S1: Trends in proportion of ICC across regions from 2013 to 2022.
Table S2: Univariable associations between clinicodemographic factors and cervical squamous cell carcinoma (SCC).
Figure S1: RCS curves of age at first diagnosis, age, BMI, age at first childbirth, age at menarche, and gravidity.
Figure S2: Population distribution by ethnic group.
Figure S3: Time trend of SCC proportion in each subgroup.
Data Availability Statement
The data presented in this study are fully included in the manuscript and Appendix S1, with no additional data required for replication of the findings. Raw data generated during the study are available from the corresponding author upon reasonable request.