Abstract
Current American Cancer Society (ACS) guidelines estimated that screening starting at the age of 25 years with Pap and/or human papillomavirus (HPV) testing is sufficient to prevent cervical cancer (CC). The effect of having HPV infections without Pap-based care until age 25 on the prevalence of higher grades of cervical intraepithelial neoplasia (≥ CIN 2) and their determinants are largely unknown. The objectives of the study were to document the potential effects of age-based changes in screening guidelines on the identification of ≥ CIN 2 and their determinants. The study included 1584 women diagnosed with abnormal Pap and tested for HPVs and histological diagnoses of cervical lesions. The association between demographic/lifestyle factors and HPV status and risk of being diagnosed with ≥ CIN 2 among younger (21-<25 years) or older (≥ 25 year) women was tested using unconditional multiple logistic regression models. We observed that younger women who are not screened have a similar or higher risk of developing specific high risk (HR)-HPV genotype-associated ≥ CIN 2 lesions compared to older women who are screened according to the current guidelines. In addition, younger women who reported live births, smoking, contraceptive use and a higher number of sexual partners were significantly at higher risk of being diagnosed with ≥ CIN 2. Targeted screening of younger women at risk for developing ≥ CIN 2 will address the concern of overtreatment while providing the recommended care to those who require such care to prevent the development of CC.
Keywords: Screening, age, cervical cancer, risk factors
Introduction
For over 50 years, the Papanicolaou (Pap) test has been the gold standard for cervical cancer (CC) screening because of its effect on lowering CC mortality in developed nations that adopted regular screening programs based on this test1–3 Theincidence and mortality from CC have declined markedly in the United States since the mid-20th century, largely due to widespread Pap-based screening practices that were initiated in the 1950s4 Nevertheless, in the US, an estimated 14,100 cases of CC will be diagnosed and an estimated 4282 deaths from this cancer will occur in 2022 with disparities mainly by race/ethnicity of the women affected.5 Recommendations for CC screening have evolved over the years, influenced by a greater understanding of the causal role of infection with high-risk human papillomavirus (HR-HPV) genotypes, the causative agents for developing higher grades of cervical intraepithelial neoplasia (≥ CIN 2), the precursor lesions for developing CC. Currently, molecular-based detection of HPVs that rely on fully automated procedures with less human involvement are more sensitive and reproducible than the Pap test have also become cost-effective. Therefore, the World Health Organization (WHO) recommends implementation of primary HPV screening in countries where Pap-based screening programs do not exist or are ineffective6 However, the Pap test is still considered the most useful tool for diagnosing HPV status in many developing counties. Even though currently, HPV DNA test combined with Pap test, known as “co-testing,” is also approved for screening as combining these tests results in improved detection of cervical precancerous lesions and CC7, this approach is cost-prohibitive for many resource-limited settings.
The previous guidelines of the American Cancer Society (ACS), issued in 2012, recommended initiation of cervical cancer screening at the age 21 years8 Since then, as prophylactic HPV vaccines are recommended for the prevention of CC, newer guidelines of the ACS estimated that the change to primary screening starting at the age of 25 years with Pap and/or HPV testing is sufficient to prevent CC9 However, this approach disregards the fact that current vaccines are only effective in women never infected with HR-HPVs while millions of women are already infected with HR-HPVs and vaccines only prevent lesions that are caused by a limited number of HR-HPV genotypes that are included in vaccines. Further, the US HPV vaccine coverage is well-below the Healthy People 2020 targets with frustrating disparities across states10 Another complication of the HPV vaccine approach is that the Advisory Committee on Immunization Practices (ACIP) has no recommendation for additional 9-valent HPV vaccine (9v) doses for women who started the series with a quadrivalent HPV vaccine (qv)11 even though 9v provides protection against additional HR-HPV genotypes. Even though these recommendations do not take away the access for screening at a younger age, screening of women ≥ 25 years is currently widely accepted as the standard care. However, it is important to consider that HPV infects up to 80% of sexually active females, often in adolescence and in early adulthood.12 The effect of having this infection without Pap-based care until age 25 on the prevalence of ≥ CIN 2 and their determinants are largely unknown. Based on this background, the objectives of the current study were to document the potential effects of age-based changes in screening guidelines on the identification of ≥ CIN 2 and their determinants using a population of 21-50-year-old women who were diagnosed with varying severity of abnormal Pap diagnoses and therefore were referred for colposcopy-directed and histology-based diagnoses of cervical lesions. We focused on ≥ CIN 2 that includes CIN 2 and CIN 3 as a group and CIN 3 separately since ≥ CIN 2 lesions quality for surgical removal in general and especially in older women and removal of CIN 3, regardless of the age of the woman. All women were also tested for the status of 37 HPV genotype and characterized for demographic factors (age, race, education), lifestyle-related factors (smoking, body mass index (BMI), parity, contraceptive use, and the lifetime number of sexual partners. To our knowledge, no previous studies have assessed the consequences of current screening guidelines in order to document their advantages and drawbacks in CC preventive care in women diagnosed with abnormal pap and/or exposed to HPV infections.
Materials and Methods
Description of the Study Population
The study population consisted of 1584 pre-menopausal women between the ages of 21-50 years who were diagnosed with abnormal Pap by the clinics of the health departments in Jefferson County, Birmingham, Alabama who were referred to UAB for further examination by colposcopy and biopsy between 2004-2013 and were enrolled in studies funded by the National Cancer Institute (R01 CA102489 and R01 CA105448). Of the 1584 women, 863 (54%) were between the age 21-<25 years and 721 (46%) were ≥ 25 years. All women included in this study were tested for 37 HPV genotypes that included 13 HR-HPV genotypes (HPV 16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, 59 and 68) 24 LR-HPV genotypes [6, 11, 26, 40, 42, 53, 54, 55, 61, 62, 64, 66, 67, 69, 70, 71, 72, 73 (MM9), 81, 82 (MM4), 83 (MM7), 84 (MM8), IS39 and CP6108] using the Roche Diagnostic Linear Array13 In order to generate results that are relevant to currently used HPV vaccines, women were categorized based on the presence of any HR-HPV genotype of the qv (positive for HPV 16 or 18) while positive or negative (+/−) for any other HR-HPV, n=533, negative for HPV 16 and 18 but positive for any other HR-HPV, n=714 and based on the presence of any HR-HPV genotype of the 9v (HPVs 16, 18, 31, 33, 45, 52, 58) while +/− for any other HR-HPV, n=957, negative for 9v HR-HPVs and positive for any other HR-HPV, n=290 and negative for any HR-HPV, n=337. Of the 1584 women, 373 were diagnosed with CIN 2 + (cases, that included CIN 2 [n =217] and CIN 3 [n = 156], and 1211 women were diagnosed with ≤ CIN 1 (non-cases, that included normal cervical epithelium [n = 85], HPV cytopathic effect [n = 166], reactive nuclear enlargement [n = 157], and CIN 1 [n = 803]).
Data and sample collection
Demographic information (age, race, and the level of education) and lifestyle information (smoking status, parity, use of hormonal contraceptives and the number of lifetime sexual partners) were obtained using a validated risk factor questionnaire administered by the study personnel. Pelvic examinations and collection of cervical cells and biopsies were carried out in accordance with the protocols of the colposcopy clinic. The study protocol and procedures were approved by the UAB Institutional Review Board in accordance with the ethical standards laid down in the 1964 Declaration of Helsinki and its later amendments. All women gave their signed informed consent prior to their inclusion in the study (approval numbers, IRB- F060511015 and F040126002).
Statistical methods
Descriptive statistics were used to describe the study population based on the case status in two categories of age, 21-<25 years (younger) or ≥ 25 years (older). Differences in the proportions between cases and non-cases were tested using Pearson’s chi-square test. The association between independent variables, namely, demographic (race as African American vs Caucasian American), education level (high school or higher vs less than high school), BMI (< 25 vs ≥25), parity (0 live birth vs ≥1 live births) and lifestyle factors (smoking status as never vs ever smoker since ever smoker category captures both current as well previous smokers whose smoking-related cancer risk is higher than that of non-smokers14), lifetime number of partners (<5 vs ≥5 partners or undisclosed/unknown) and hormone contraceptive use (non-users vs users) and the status of HPV as negative for any HR-HPV vs positive for any qv or 9v HR-HPV genotype and +/− for any other HR-HPV or negative for qv or 9v HR-HPVs but positive for any other HR-HPV genotype. The dependent variable was categorized as ≥ CIN 2 or CIN 3. Unconditional multiple logistic regression models stratified by two age categories were simultaneously adjusted for the above stated variables. The referent category for each variable is shown in tables as 1.00. A P-value of <0.05 was considered a statistically significant association. All analyses were performed using JMP® Pro 16, RRID:SCR-014242 (SAS Institute Inc. Cary, NC, USA).
Data availability statement
The data generated in this study are available upon request from the corresponding author.
Results
Descriptive results of the study population based on the CIN status (≤ CIN 1 vs ≥ CIN 2) in the two age group categories are shown in Table 1. A higher percentage of younger women diagnosed with ≥ CIN 2 was more likely to be Caucasian Americans (CAs), smokers, reported ≥5 lifetime number of partners and used hormonal contraceptives compared to those diagnosed with ≤ CIN 1 (P=0.0056, 0.0003, 0.0465 and 0.0016, respectively). We did not observe significant differences in the proportions of those variables in older (≥25 years) women. In both age categories, a higher percentage of women with ≥ CIN 2 diagnosis had parity ≥1 live births compared to women with ≤ CIN 1 (P=0.0026 and 0.0335, in the younger and older age groups, respectively). In both age categories, a higher percentage of women diagnosed with ≥ CIN 2 was positive for HR-HPV genotypes of the qv (33% and 36%, respectively) or the 9v (31% and 29%, respectively) (P<0.0001 for all comparisons).
Table 1.
Demographic, lifestyle factors and the status of HR-HPV genotypes of qv or 9v HPV vaccine by the case status (≤ CIN 1 vs ≥ CIN 2) among younger (21-< 25 years) and older (≥25 years) women, N=1584
| Variables | 21-< 25 years (N=863) | ≥ 25 years (N=721) | ||||
|---|---|---|---|---|---|---|
| ≤ CIN 1 (n=659) (76%) | ≥ CIN 2 (n=204) (24%) | P-value | ≤ CIN 1 (n=552) (77%) | ≥ CIN 2 (n=169) (23%) | P-value | |
| Race | ||||||
| African American | 367 (80%) | 91 (20%) | 0.0056 | 326 (77%) | 98 (23%) | 0.8047 |
| Caucasian American | 292 (72%) | 113 (28%) | 226 (76%) | 71 (24%) | ||
| Educational level | ||||||
| Less than high school education | 533 (76%) | 164 (24%) | 0.8772 | 461 (77%) | 141(23%) | 0.9798 |
| High school education or higher | 126 (76%) | 40 (24%) | 91 (76%) | 28 (24%) | ||
| BMI | ||||||
| < 25 | 257 (73%) | 95 (27%) | 0.0577 | 175 (73%) | 66 (27%) | 0.0814 |
| ≥ 25 | 397 (79%) | 108 (21%) | 375 (78%) | 103 (22%) | ||
| Parity | ||||||
| 0 live birth | 271 (82%) | 60 (18%) | 0.0026 | 112 (84%) | 22 (16%) | 0.0335 |
| ≥ 1 live birth(s) | 388 (73%) | 144 (27%) | 440 (75%) | 147(25%) | ||
| Smoking status | ||||||
| Ever | 301 (71%) | 123 (29%) | 0.0003 | 271 (76%) | 86 (24%) | 0.6833 |
| Never | 358 (82%) | 81 (18%) | 281 (77%) | 83 (23%) | ||
| Lifetime number of sexual partners | ||||||
| < 5 | 228 (81%) | 54 (19%) | 0.0465 | 126 (72%) | 48 (28%) | 0.1982 |
| ≥ 5 | 294 (73%) | 110 (27%) | 285 (77%) | 87 (23%) | ||
| Unknown/Undisclosed | 137 (77%) | 40 (23%) | 141 (81%) | 34 (20%) | ||
| Hormone contraceptive use | ||||||
| Users | 123 (87%) | 19 (13%) | 0.0016 | 147 (80%) | 36 (20%) | 0.1567 |
| Non-users | 536 (74%) | 185 (26%) | 403 (75%) | 133 (25%) | ||
| HPV status of women | ||||||
| Positive for HPV 16 or 18 and +/−* for any other HR-HPV (n=533) | 218 (67%) | 109 (33%) | <0.0001 | 132 (64%) | 74 (36%) | <0.0001 |
| Negative for HPV 16 and 18 but positive for any other HR-HPV (n=714) | 283 (79%) | 75 (21%) | 274 (92%) | 82 (8%) | ||
| Negative for any HR-HPV (n=337) | 158 (89%) | 20 (11%) | 146 (92%) | 13 (8%) | ||
| Positive for any 9v HR-HPV and +/−* for other HR-HPV n=957) | 359 (69%) | 160 (31%) | <0.0001 | 309 (71%) | 129 (29%) | <0.0001 |
| Negative for 9v HR-HPV and positive for any other HR-HPV (n=290) | 142 (86%) | 24 (14%) | 97 (78%) | 27 (12%) | ||
| Negative for any HR-HPV (n=337) | 158 (89%) | 20 (11%) | 146 (92%) | 13 (8%) | ||
Positive or negative (+/−)
Descriptive results of the population based on ≤ CIN 1 vs. CIN 3 after excluding those diagnosed with CIN 2 in the two age categories are shown in Table 2. A higher percentage of younger women diagnosed with CIN 3 were more likely to be CAs, smokers, users of hormonal contraceptives compared to those diagnosed with ≤ CIN 1 (P=0.0006, 0.0098 and 0.0071 respectively). We did not observe significant differences in the proportion of those variables in older women by CIN status. In both age groups, a higher percentage of women with CIN 3 had parity ≥1live births compared to women diagnosed with ≤ CIN 1(P=0.0317 and 0.0185 in younger and older age categories, respectively). In both age groups, a higher percentage of women diagnosed with CIN 3 were positive for HR-HPV genotypes of the qv vaccine (P<0.0001 for both comparisons) or 9v vaccine (P<0.0001 and P=0.0008, in younger and older age categories, respectively) compared to women diagnosed with ≤ CIN 1.
Table 2.
Demographic, lifestyle factors and the status of HR-HPV genotypes of qv or 9v HPV vaccine by the case status (≤ CIN 1 vs CIN 3) among younger (21-< 25 years) and older (≥25 years) women, N=1367
| Variables | < 25 years (N=745) | ≥ 25 years (N=622) | ||||
|---|---|---|---|---|---|---|
| ≤ CIN 1 (n=659) (88%) | CIN 3 (n=86) (12%) | P-value | ≤ CIN 1 (n=552) (89%) | CIN 3 (n=70) (11%) | P-value | |
| Race | ||||||
| African American | 367 (92%) | 31 (8%) | 0.0006 | 326 (91%) | 33 (9%) | 0.0573 |
| Caucasian American | 292 (84%) | 55 (16%) | 226 (86%) | 37 (14%) | ||
| Educational level | ||||||
| Less than high school education | 533 (88%) | 69 (11%) | 0.8860 | 461 (89%) | 54 (11%) | 0.1833 |
| High school education or higher | 126 (88%) | 17 (12%) | 91 (85%) | 16 (15%) | ||
| BMI | ||||||
| < 25 | 257 (87%) | 39 (13%) | 0.2740 | 175 (86%) | 28 (14%) | 0.1695 |
| ≥ 25 | 397 (89%) | 47 (11%) | 375 (90%) | 42 (10%) | ||
| Parity | ||||||
| 0 live birth | 271 (91%) | 25 (8%) | 0.0317 | 112 (94%) | 6 (5%) | 0.0185 |
| ≥ 1 live birth(s) | 388 (86%) | 61 (14%) | 440 (87%) | 64 (13%) | ||
| Smoking status | ||||||
| Ever | 301 (85%) | 52 (15%) | 0.0098 | 271 (86%) | 43 (14%) | 0.0518 |
| Never | 358 (91%) | 34 (9%) | 281 (91%) | 27 (9%) | ||
| Lifetime number of sexual partners | ||||||
| < 5 | 228 (92%) | 19 (8%) | 0.0570 | 126 (86%) | 20 (14%) | 0.5218 |
| ≥ 5 | 293 (86%) | 48 (14%) | 285 (90%) | 32 (10%) | ||
| Unknown/ Undisclosed | 137 (87%) | 20 (13%) | 141 (89%) | 18 (11%) | ||
| Hormone contraceptive use | ||||||
| Users | 123 (95%) | 6 (5%) | 0.0071 | 147(91%) | 14 (9%) | 0.2297 |
| Non-users | 536 (87%) | 80 (13%) | 404 (88%) | 56 (12%) | ||
| HPV status of women | ||||||
| Positive for HPV 16 or 18 and +/−* for other HR-HPV (n=451) | 218 (78%) | 61 (22%) | <0.0001 | 132 (77%) | 40 (23%) | <0.0001 |
| Negative for HPV16 and 18 and positive for any other HR-HPV (n=605) | 283 (92%) | 23 (8%) | 274 (92%) | 24 (8%) | ||
| Negative for any HR-HPV (n=314) | 158 (98%) | 4 (2%) | 146 (96%) | 6 (4%) | ||
| Positive for any 9v HR-HPV and +/−* for other HR-HPV(n=804) | 359 (82%) | 80 (18%) | <0.0001 | 309 (85%) | 55 (15%) | 0.0008 |
| Negative for 9v HR-HPVs and positive for any other HR-HPV (n=252) |
142 (98%) | 4 (2%) | 97 (92%) | 9 (8%) | ||
| Negative for any HR-HPV(n=314) | 158 (98%) | 4 (2%) | 146 (96%) | 6 (4%) | ||
Positive or negative (+/−)
The results of the logistic regression models testing the association between the status of HR-HPV genotypes of the qv vaccine and risk of being diagnosed with ≥ CIN 2 after adjusting for demographic and lifestyle factors among younger and older women are shown in Table 3. We observed that younger women who were positive for any qv HR-HPV, irrespective of whether they are +/− for any other HR-HPV were 4.19 times more likely to be diagnosed with ≥ CIN 2 (P<0.0001). Similarly, women who were negative for any qv HR-HPVs but +/− for any other HR-HPV were 2.43 times more likely to be diagnosed with ≥ CIN 2 (P<0.0016) compared to women who were negative for any HR-HPV. Similarly, older women who were positive for any qv HR-HPV and +/− for any other HR-HPV and women who were negative for any qv HR-HPVs and +/− for any other HR-HPV were more likely to be diagnosed with ≥ CIN 2 (6.90 times and 3.58 times, respectively) compared to women who were negative for any HR-HPV (P<0.0001 for both). In both age categories, women with a higher parity were more likely to be diagnosed with ≥ CIN 2 (P=0.0029 and 0.0499, respectively). Younger women who had ≥5 lifetime sexual partners were more likely to be diagnosed with ≥ CIN 2 compared to those with <5 lifetime sexual partners (P=0.0407) while older women who did not disclose or did not respond with regard to number of lifetime sexual partners were less likely to be diagnosed with ≥ CIN 2 compared to those who had <5 lifetime sexual partners (P=0.0257). We also observed that only younger women who ever smoked or used hormonal contraceptives were more likely to be diagnosed with ≥ CIN 2 compared to their respective referent group (P=0.0029 and 0.0052, respectively).
Table 3.
Associations between demographic lifestyle factors and the status of HR-HPV genotypes of the qv HPV vaccine and risk of being diagnosed with CIN 2+ among younger (21-< 25 years) and older (≥25 years) women, N=1584
| Variables | 21-< 25 years (N=863) | ≥ 25 years (N=721) | ||
|---|---|---|---|---|
| ≤ CIN 1 vs CIN 2+ | ≤ CIN 1 vs CIN 2+ | |||
| OR (95%CI) | P-value | OR (95%CI) | P-value | |
| Race | ||||
| African American | 1.00 | 0.4046 | 1.00 | 0.3424 |
| Caucasian American | 1.17 (0.81-1.68) | 0.82 (0.54-1.24) | ||
| Educational level | ||||
| High school education or higher | 1.00 | 0.1969 | 1.00 | 0.7649 |
| Less than high school education | 0.75 (0.49-1.16) | 1.08 (0.65-1.78) | ||
| BMI | ||||
| < 25 | 1.00 | 0.3544 | 1.00 | 0.1025 |
| ≥ 25 | 0.85 (0.61-1.19) | 0.73 (0.50-1.07) | ||
| Parity | ||||
| 0 live birth | 1.00 | 0.0029 | 1.00 | 0.0499 |
| ≥ 1 live birth(s) | 1.74 (1.21-2.50) | 1.67 (1.00-2.79) | ||
| Smoking status | ||||
| Never | 1.00 | 0.0307 | 1.00 | 0.5903 |
| Ever | 1.50(1.04-2.18) | 1.12 (0.75-1.66) | ||
| Lifetime number of sexual partners | ||||
| < 5 | 1.00 | 1.00 | ||
| ≥ 5 | 1.51 (1.02-2.23) | 0.0407 | 0.70 (0.46-1.09) | 0.1137 |
| Unknown/Undisclosed | 1.09 (0.67-1.78) | 0.7296 | 0.54 (0.32-0.93) | 0.0257 |
| Hormone contraceptive use | ||||
| Non-users | 1.00 | 0.0052 | 1.00 | 0.2800 |
| Users | 2.13 (1.25-3.61) | 1.27 (0.82-1.97) | ||
| HPV status of women | ||||
| Negative for any HR-HPV | 1.00 | 1.00 | ||
| Positive for any qv HR-HPV and +/−* for any other HR-HPV | 4.19 (2.44-7.17) | <0.0001 | 6.90 (3.61-13.22) | <0.0001 |
| Negative for qv HR-HPVs and positive for any other HR-HPV | 2.43 (1.40-4.22) | 0.0016 | 3.58 (1.92-6.70) | <0.0001 |
Positive or negative (+/−)
The results from the logistic regression models that tested the association between the status of HR-HPV genotypes of the 9v vaccine and risk of being diagnosed with ≥ CIN 2 after adjusting for demographic and lifestyle factors among younger and older women are shown in Table 4. We observed that younger women who were positive for any 9v HR-HPV and +/− for any other HR-HPV were 3.89 times more likely to be diagnosed with ≥ CIN 2 compared to women who were negative for any HR-HPV (P<0.0001). Among older women, those positive for any 9v HR-HPV and +/− for any other HR-HPV and those negative for any 9v HR-HPVs and +/− for any other HR-HPV were more likely (5.00 and 3.38 times, respectively) to be diagnosed with ≥ CIN 2 compared to women who were negative for any HR-HPV (P<0.0001 and P=0.0009). In both age categories, women with a higher parity were more likely to be diagnosed with ≥ CIN 2 (P=0.0039 and 0.0456 in younger and older women, respectively). Among younger women, those who had ≥5 lifetime sexual partners were more likely to be diagnosed with ≥ CIN 2 compared to those with <5 lifetime sexual partners (P=0.0560) while among older women, those who did not disclose or did not respond to that question were less likely to be diagnosed with ≥ CIN 2 compared to those who had <5 lifetime sexual partners (P=0.0256). We also observed that only among younger women, ever smokers or users of hormonal contraceptives were more likely to be diagnosed with ≥ CIN 2 compared to the respective referent group (P=0.0298 and 0.0036, respectively).
Table 4.
Associations between demographic lifestyle factors and the status of HR-HPV genotypes of the 9v HPV vaccine and risk of being diagnosed with CIN 2+ among younger (21-< 25 years) and older (≥25 years) women, N=1584
| Variables | 21-< 25 years (N=863) | ≥ 25 years (N=721) | ||
|---|---|---|---|---|
| ≤ CIN 1 vs CIN 2+ | ≤ CIN 1 vs CIN 2+ | |||
| OR (95%CI) | P-value | OR (95%CI) | P-value | |
| Race | ||||
| African American | 1.00 | 0.2408 | 1.00 | 0.7980 |
| Caucasian American | 1.24 (0.87-1.78) | 0.95 (0.63-1.43) | ||
| Educational level | ||||
| High school education or higher | 1.00 | 0.2095 | 1.00 | 0.5999 |
| Less than high school education | 0.76 (0.49-1.17) | 0.88 (0.53-1.44) | ||
| BMI | ||||
| < 25 | 1.00 | 0.4718 | 1.00 | 0.1052 |
| ≥ 25 | 0.88 (0.63- 1.24) | 0.73 (0.50-1.07) | ||
| Parity | ||||
| 0 live birth | 1.00 | 0.0039 | 1.00 | 0.0456 |
| ≥ 1 live birth(s) | 1.71 (1.19-2.47) | 1.68 (1.01-2.80) | ||
| Smoking status | ||||
| Never | 1.00 | 0.0298 | 1.00 | 0.6106 |
| Ever | 1.51 (1.04-2.19) | 1.11 (0.75-1.64) | ||
| Lifetime number of sexual partners | ||||
| < 5 | 1.00 | 1.00 | ||
| ≥ 5 | 1.47 (0.99-2.18) | 0.0560 | 0.72 (0.46-1.10) | 0.1288 |
| Unknown/Undisclosed | 1.07 (0.66-1.76) | 0.7689 | 0.54 (0.32-0.93) | 0.0256 |
| Hormone contraceptive use | ||||
| Non-users | 1.00 | 0.0036 | 1.00 | 0.3330 |
| Users | 2.20 (1.29-3.73) | 1.24 (0.80-1.91) | ||
| HPV status of women | ||||
| Negative for any HR-HPV | 1.00 | 1.00 | ||
| Positive for any 9v HR-HPV and +/−* for other HR-HPV | 3.89 (2.31-6.55) | <0.0001 | 5.00 (2.71-9.22) | <0.0001 |
| Negative 9v HR-HPVs and positive for any other HR-HPV | 1.54 (0.80-2.97) | 0.1937 | 3.38 (1.65-6.92) | 0.0009 |
Positive or negative (+/−)
The results of the unconditional multiple logistic regression models testing the association between the status of HR-HPV genotypes of the qv vaccine and risk of being diagnosed with CIN 3 after adjusting for demographic and lifestyle factors among younger and older women are shown in Table 5. We observed that younger women who were positive for any qv HR-HPV and +/− for any other HR-HPV and those women who were negative for qv HR-HPVs and positive for any other HR-HPV were more likely (10.89 and 3.61 times, respectively) to be diagnosed with CIN 3 compared to women who were negative for any HR-HPV (P<0.0001 and 0.0209, in younger and older women, respectively). Older women who were positive for any qv HR-HPV and +/− for any other HR-HPV were 7.30 times more likely to be diagnosed with CIN 3 compared to women who were negative for any HR-HPV (P<0.0001). Younger women who had ≥5 lifetime sexual partners and users of hormonal contraceptives were more likely to be diagnosed with CIN 3 compared to the respective referent group (P=0.0227 and 0.0122, respectively). None of the demographic and lifestyle variables were associated with the risk of being diagnosed with CIN 3 among older women.
Table 5.
Associations between demographic lifestyle factors and the status of HR-HPV genotypes of the qv HPV vaccine and risk of being diagnosed with CIN 3 among younger (21-< 25 years) and older (≥25 years) women, N=1367
| Variables | 21-< 25 years (N=745) | ≥ 25 years (N=622) | ||
|---|---|---|---|---|
| ≤ CIN 1 vs CIN 3 | ≤ CIN 1 vs CIN 3 | |||
| OR (95%CI) | P-value | OR (95%CI) | P-value | |
| Race | ||||
| African American | 1.00 | 0.0706 | 1.00 | 0.9135 |
| Caucasian American | 1.63 (0.96-2.78) | 1.03 (0.58-1.85) | ||
| Educational level | ||||
| High school education or higher | 1.00 | 0.4583 | 1.00 | 0.4145 |
| Less than high school education | 0.79 (0.42-1.47) | 1.34(0.68-2.59) | ||
| BMI | ||||
| < 25 | 1.00 | 0.8995 | 1.00 | 0.2935 |
| ≥ 25 | 0.97 (0.59-1.58) | 0.75 (0.43-1.29) | ||
| Parity | ||||
| 0 live birth | 1.00 | 0.0557 | 1.00 | 0.0544 |
| ≥ 1 live birth(s) | 1.68 (0.99-2.85) | 2.38 (0.98-5.78) | ||
| Smoking status | ||||
| Never | 1.00 | 0.4935 | 1.00 | 0.2171 |
| Ever | 1.21 (0.71-2.06) | 1.44 (0.81-2.58) | ||
| Lifetime number of sexual partners | ||||
| < 5 | 1.00 | 1.00 | ||
| ≥ 5 | 2.00 (1.10-3.62) | 0.0227 | 0.62 (0.32-1.17) | 0.1373 |
| Unknown/ Undisclosed | 1.50 (0.74-3.06) | 0.2647 | 0.74 (0.35-1.58) | 0.4413 |
| Hormone contraceptive use | ||||
| Non-users | 1.00 | 0.0122 | 1.00 | 0.1859 |
| Users | 3.07 (1.28-7.40) | 1.55 (0.81-2.99) | ||
| HR-HPV status of women | ||||
| Negative for any HR-HPV | 1.00 | 1.00 | ||
| Positive for qv HR-HPVs and +/−* for other HR-HPV | 10.89 (3.84-30.87) | <0.0001 | 7.30 (2.94-18.10) | <0.0001 |
| Negative for qv HR-HPVs and positive for any other HR-HPV | 3.61 (1.21-10.74) | 0.0209 | 2.15 (0.85-5.43) | 0.1063 |
Positive or negative (+/−)
The results of the unconditional multiple logistic regression models testing the association between the status of HR-HPV genotypes of 9v vaccine and risk of being diagnosed with CIN 3 after adjusting for demographic and lifestyle factors among younger and older women are shown in Table 6. We observed that younger women who were positive for any 9v HR-HPV and +/− for any other HR-HPV were 9.20 times more likely to be diagnosed with CIN 3 compared to women who were negative for any HR-HPV (P<0.0001). The older women who were positive for any 9v HR-HPV and +/− for any other HR-HPV were 4.23 times more likely to be diagnosed with CIN 3 compared to women who were negative for any HR-HPV (P=0.0013). Among younger women, CAs, those who had partners ≥5 and users of hormonal contraceptives were more likely to be diagnosed with CIN 3 compared to the respective referent group (P=0.0280, 0.0417 and 0.0103, respectively). None of the demographic and lifestyle variables were associated with the risk of being diagnosed with CIN 3 among older women.
Table 6.
Associations between demographic lifestyle factors and the status of HR-HPV genotypes of the 9v HPV vaccine and risk of being diagnosed with CIN 2+ among younger (21-< 25 years) and older (≥25 years) women, N=1367
| Variables | < 25 years (N=745) | ≥ 25 years (N=622) | ||
|---|---|---|---|---|
| ≤ CIN 1 vs CIN 3 | ≤ CIN 1 vs CIN 3 | |||
| OR (95%CI) | P-value | OR (95%CI) | P-value | |
| Race | ||||
| African American | 1.00 | 0.0280 | 1.00 | 0.5215 |
| Caucasian American | 1.80 (1.07-3.05) | 1.20 (0.68-2.11) | ||
| Educational level | ||||
| High school education or higher | 1.00 | 0.5607 | 1.00 | 0.5927 |
| Less than high school education | 0.83 (0.45-1.55) | 1.19 (0.62-2.29) | ||
| BMI | ||||
| < 25 | 1.00 | 0.9547 | 1.00 | 0.3626 |
| ≥ 25 | 1.01 (0.62-1.66) | 0.78 (0.46-1.33) | ||
| Parity | ||||
| 0 live birth | 1.00 | 0.0818 | 1.00 | 0.0420 |
| ≥ 1 live birth(s) | 1.60 (0.94-2.72) | 2.48 (1.03-5.95) | ||
| Smoking status | ||||
| Never | 1.00 | 0.4112 | 1.00 | 0.1683 |
| Ever | 1.25 (0.73-2.13) | 1.50 (0.84-2.66) | ||
| Lifetime number of sexual partners | ||||
| < 5 | 1.00 | 1.00 | ||
| ≥ 5 | 1.86 (1.02-3.37) | 0.0417 | 0.63 (0.34-1.19) | 0.1541 |
| Unknown/Undisclosed | 1.47 (0.72 – 3.01) | 0.2926 | 0.73 (0.35-1.54) | 0.4080 |
| Hormone contraceptive use | ||||
| Non-users | 1.00 | 0.0103 | 1.00 | 0.2595 |
| Users | 3.15 (1.31-7.58) | 1.45 (0.76-2.75) | ||
| HPV status of women | ||||
| Negative for any HR-HPV | 1.00 | |||
| Positive for any 9v HR-HPV and +/−* for other HR-HPV | 9.20 (3.28-25.77) | <0.0001 | 4.23 (1.76-10.18) | 0.0013 |
| Negative 9v HR-HPVs and positive for any other HR-HPV | 1.25 (0.30-5.13) | 0.7583 | 2.33 (0.80-6.84) | 0.1228 |
Positive or negative (+/−)
Discussion
Although it is possible that HPV infections and their pre-cancerous lesions may clear up on their own, there is a risk for any women regardless of the age for HPV infections to become chronic and develop pre-cancerous lesions that progress to invasive CC. In women with a “normal” immune system, it may take 15-20 years to develop CC but in women at increased risk such as those with a weakened immune system, it may only take 5-10 years to develop CC9 Studies have also shown that invasive CC develops in 5-10 years in 20%-30% of patients with CC precursors15 Further, studies have also shown that there is a recent increase in CC in women aged 20-24 years and there is concern whether this change is linked to the withdrawal of cervical screening in women aged 20-24 according to the revised screening guidelines16 In this study based in England, Scotland and Wales where CC screening of 20-24 year old women was phased out between 2004 and 2009, an increase in CC rate in in this age group was observed from 2013 onwards. It is also important to note that in these countries, an even more dramatic increase in CC was observed in 25-year-old women following the change in age-based screening guidelines. Previous studies have estimated that screening from age 25 years could lead to an increase of CC compared with screening from age 20, based on an annual CIN 3 progression rate of 0.2% to CC. This excess risk was attributable to the failure to treat ≥ CIN 2 prior to those becoming CC at the age 25 years17
Even though infection with human immunodeficiency virus (HIV) is widely known as a cause of immunosuppression that increases the risk of HPV-associated ≥ CIN 2 and CCs, several other conditions such as poor diet and insulin resistance/diabetes mellitus may also result in immunodeficiency and increase the risk of HPV infections. The risk of progression or time for HR-HPV infections to progress toward ≥ CIN 2 or CC under those conditions are largely unknown. We have previously reported that approximately 80% of our study population has unhealthy dietary patterns18 and insulin resistance due to unhealthy dietary pattern-related lower intake levels of important micronutrients19, conditions that may lower their immune response leading to the persistence of HR-HPV infections, a key factor for developing ≥ CIN 2 or CC. Several of our previous studies have also documented that the circulating status of specific “cancer-protective” micronutrients alter the risk of acquisition, persistence, clearance20 and methylation status of HR-HPVs21 and the development of ≥ CIN 2 lesions associated with those micronutrients22
We now report that women younger than 25 years of age in this population who would not have received screening until the age of 25 according to current guidelines have a similar prevalence of CIN 2 and 3 combined (≥ CIN 2) as well as CIN 3 compared to women ≥ 25 years who would have received screening according to the current guidelines. Since the study demonstrated that independent of their specific HPV genotype status, younger women who reported live births, smoking, contraceptive use, and a higher number of lifetime sexual partners were significantly at higher risk of being diagnosed with ≥ CIN 2, targeted screening of younger women based on those factors will reduce their risk of developing CC. We specially recommend screening of younger women based on a higher number of lifetime sexual partners and contraceptive use since those factors were significantly associated with developing CIN 3. However, we recognize that the borderline significance of the association between parity and CIN 3, a well-documented risk factor for CC23 could be due to the smaller number of CIN 3 than CIN 2 identified in the study. Therefore, this observation needs to be verified in a study with a larger sample size of CIN 3 cases. We also noted that younger Caucasian American women were significantly at higher risk for developing CIN 3 when adjusted for 9v HPV genotypes. This indicates that race-specific screening to identify and treat those lesions of younger women, especially when they have additional risk factors discussed above will benefit from such targeted screening to prevent the progression of those lesions to CC.
As expected, irrespective of the age, women infected with HPV 16 or 18 were at higher risk for developing ≥ CIN 2 and CIN 3 compared to women who were not infected with any HR-HPV indicating a protective effect of the qv HPV vaccine. However, the observation that women negative for HPV 16 or 18 but positive for other HR-HPVs that are not included in qv were also at significantly higher risk for ≥ CIN 2, emphasize that qv only provides partial protection against CC risk. A 4-fold higher risk of CIN 3 lesions associated with HR-HPVs other than HPV 16 or 18 among only younger women especially emphasizes the fact that qv has limited use to prevent CC in this age group. This is concerning since the ACIP does not recommend providing 9v for women who started the series with a qv. Our study also showed that being infected with 9v HPV genotypes increases the risk of CIN 3 especially in younger women. Being positive for HR-HPVs other than 9v HPV genotypes did not increase the risk of CIN 3 in younger or older women indicating the effectiveness of preventing CC risk in those who are vaccinated with this vaccine. It is unclear why the HR-HPV genotypes not included in 9v increases the risk of ≥ CIN 2 only in older women.
In this study, we document that the prevalence of CIN lesions that need medical care is similar among both younger and older women suggesting that the current screening guidelines that will not allow treatment of ≥ CIN 2 lesions may increase the risk of developing CC in younger women over time. Therefore, we believe that younger women will benefit from targeted screening based on their race, smoking habits, contraceptive use, parity, and risky sexual behavior. The success rate of tailored screening will depend on identification of those high-risk younger females by pediatricians, family medicine physicians or primary care physicians. Studies have documented challenges in providing confidential care to adolescents because of limitations in physician time alone with an adolescent and their concerns about privacy to discuss factors that increase their exposure to HR-HPVs via risky sexual behavior and other factors such as contraceptive use, smoking and poor diet that facilitate transformation of HR-HPV infections to abnormal Pap and ≥ CIN 224-27 Systemic changes are needed to enhance the quality and delivery of such confidential services for adolescents. Several studies have indicated that limitations in providing needed care for adolescents remains a serious concern28–30 Some studies indicated that clinicians are committed to offer time alone during preventive care visits but not frequently at other times.31
In conclusion, younger women who are not screened have a similar or higher risk of developing specific HR-HPV genotype-associated ≥ CIN 2 lesions compared to older women who are screened according to the current guidelines. Targeted screening of younger women at risk for developing ≥ CIN 2 will address the concern of overtreatment32 while providing the recommended care to those who require such care to prevent and control the development of CC. Replication of results observed in this study in other women populations at risk for developing CC will increase the scientific credibility and generalizability of our findings.
Prevention Relevance.
This study documents the concerns of the age-based changes in screening guidelines on the identification of higher grades of cervical intraepithelial neoplasia and their determinants in women diagnosed with abnormal Pap smear and emphasize the need for targeted screening of younger women to prevent cervical cancer.
Acknowledgments
Authors thank Nuzhat Rahman Siddiqui and Michelle M Chambers for their excellent technical support.
This study was supported by R01 105448 (CJ Piyathilake), R01 102489 (CJ Piyathilake), funded by the National Cancer Institute and T37-MD001448 (PE Jolly), funded by the Minority Health Research Training Grant, National Institute on Minority Health and Health Disparities.
Footnotes
Conflict of interest: The authors declare no potential conflicts of interest
References
- 1.Quinn M, Babb P, Jones J, Allen E. Effect of screening on incidence of and mortality from cancer of cervix in England: evaluation based on routinely collected statistics. BMJ 1999; 318:904–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Sasieni P, Adams J. Effect of screening on cervical cancer mortality in England and Wales: analysis of trends with an age period cohort model. BMJ 1999; 318:1244–5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.van der Aa MA, Pukkala E, Coebergh JW,Anttila A, Siesling S. Mass screening programmes and trends in cervical cancer in Finland and the Netherlands. Int J Cancer 2008; 122:1854–8. [DOI] [PubMed] [Google Scholar]
- 4.Rust G, Satcher D, Fryer GE, Levine RS, Blumenthal DS. Triangulating on success: innovation, public health, medical care, and cause-specific US mortality rates over a half century (1950--2000). Am J Public Health 2010;100: S95–104. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Key Statistics for Cervical Cancer. Cancer Facts & Figures 2022. Atlanta, Ga: American Cancer Society; 2022, https://www.cancer.org/research/cancer-facts-statistics/all-cancer-facts-figures/cancer-facts-figures-2022.html. [Google Scholar]
- 6.Pan American Health Organization/World Health Organization/Centers for Disease Control and Prevention. Integrating HPV testing in cervical cancer screening programs: a manual for program managers, 2016. [cited 2017 Sep 27]. http://iris.paho.org/xmlui/handle/123456789/31393, pages 1–57.
- 7.Fontham ETH, Wolf AMD, Church TR, Etzioni R, Flowers CR, Herzig A, et al. Cervical cancer screening for individuals at average risk: 2020 guideline update from the American Cancer Society. CA Cancer J Clin 2020;70(5):321–346. [DOI] [PubMed] [Google Scholar]
- 8.Saslow D, Solomon D, Lawson HW, Killackey M, Kulasingam SL, Cain J et al. American Cancer Society, American Society for Colposcopy and Cervical Pathology, and American Society for Clinical Pathology Screening Guidelines for the Prevention and Early Detection of Cervical Cancer. CA Cancer J Clin 2012; 62:147–172. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Fontham ETH, Wolf AMD, Church TR, Etzioni R, Flowers CR, Herzig A et al. Cervical cancer screening for Individuals at average Risk: 2020 guideline update from the American Cancer Society. CA Cancer J Clin 2020;70: 321–346. [DOI] [PubMed] [Google Scholar]
- 10.Chen S, Huybrechts K, Bateman B, Hernández-Díaz S. Trends in human papillomavirus vaccination in commercially insured children in the United States. Pediatrics 2020. Oct;146(4): e20193557. [DOI] [PubMed] [Google Scholar]
- 11.Supplemental Information and guidance for vaccination providers regarding use of 9-valent HPV vaccine (cdc.gov), CS HCVG-15-PTT-107 11/29/2016 2016, https://www.cdc.gov/hpv/downloads/9vhpv-guidance.pdf.
- 12.Chesson HW, Dunne EF, Hariri S, Markowitz LE. The Estimated lifetime probability of acquiring human papillomavirus in the United States. Sex Transm Dis 2014; 41(11): 660–664. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Schmid M, Schalasta G, Speicher A, Enders G. Linear array human papillomavirus genotyping test amenable to automation by implementation of GT-Blot 48-A pilot study. The open infectious diseases journal, 2008, 2, 40–44. [Google Scholar]
- 14.Tindle HA, Duncan MS, Greevy RA, Vasan RS, Kundu S, Massion PP, et al. Lifetime smoking history and risk of lung cancer: results from the Framingham heart study,JNCI: Journal of the National Cancer Institute, volume 110, issue 11, November 2018, Pages 1201–1207. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Schiffman M, Castle PE, Jeronimo J, Rodriguez AC, Wacholder S. Human papillomavirus and cervical cancer. Lancet 2007;370(9590): 890–907. [DOI] [PubMed] [Google Scholar]
- 16.Castanon A, Sasieni P. Is the recent increase in cervical cancer in women aged 20-24 years in England a cause for concern? Prev Med. 2018. Feb; 107: 21–28. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Landy R, Birke H, Castanon A, Sasieni P. Benefits and harms of cervical screening from age 20 years compared with screening from age 25 years. Br J Cancer 2014;110, 1841–1846. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Piyathilake CJ, Badiga S, Kabagambe EK, Azuero A, Alvarez RD, Johanning GL, et al. A Dietary Pattern Associated with LINE-1 Methylation Alters the Risk of Developing Cervical Intraepithelial Neoplasia. Cancer Prev Res (Phila) 2012; 3: 385–92. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Piyathilake CJ, Badiga S, Hernandez A, Brill IK, Jolly PE. The consumption of micronutrients in relation to calorie intake and risk of insulin resistance. Nutr Metab Cardiovasc Dis 2022; Jun;32(6):1385–1391. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Piyathilake CJ, Henao OL, Macaluso M, Cornwell PE, Meleth S, Heimburger DC, et al. Folate is Associated with the Natural History of High-Risk Human Papillomaviruses. Cancer Res 2004. Dec 1; 64 (23): 8788–93. [DOI] [PubMed] [Google Scholar]
- 21.Piyathilake CJ,Macaluso M,Chambers MM,Badiga S,Siddiqui NR,Bell WC, et al. Folate and vitamin B12 may play a critical role in lowering the HPV 16 methylation associated risk of developing higher grades of CIN. Cancer Prevent Res 2014; 7(11):1128–37. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Piyathilake CJ, Macaluso M, Brill I, Heimburger DC, Partridge EE. Lower red blood cell folate enhances the HPV 16-associated risk of cervical intraepithelial neoplasia. Nutrition 2007: 23(3):203–10. [DOI] [PubMed] [Google Scholar]
- 23.Tekalegn Y,Sahiledengle B,Woldeyohannes D,Atlaw D,Degno S,Desta F, et al. High parity is associated with increased risk of cervical cancer: Systematic review and meta-analysis of case-control studies. Womens Health (Lond) 2022; 18:17455065221075904. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Klein J, Wilson K, McNulty M, Kapphahn C, Collins KS. Access to medical care for adolescents: results from the 1997 Commonwealth Fund Survey of the Health of Adolescent Girls. J Adolesc Health 1999;25(2):120–130. [DOI] [PubMed] [Google Scholar]
- 25.Ford CA, Bearman P, Moody J. Foregone health care among adolescents. JAMA 1999;282(23):2227–2234. [DOI] [PubMed] [Google Scholar]
- 26.Lehrer JA, Pantell R, Tebb K, Shafer MA. Forgone health care among US adolescents: associations between risk characteristics and confi dentiality concern. J Adolesc Health 2007;40(3):218–226. [DOI] [PubMed] [Google Scholar]
- 27.Ford C, Millstein S, Halpern-Felsher B, Irwin CJ. Infl uence of physician confi dentiality assurances on adolescents’ willingness to disclose information and seek future health care. A randomized controlled trial. JAMA 1997;278(12):1029–1034. [PubMed] [Google Scholar]
- 28.Akinbami LJ, Gandhi H, Cheng TL. Availability of adolescent health services and confi dentiality in primary care practices. Pediatrics 2003;111(2):394–401. [DOI] [PubMed] [Google Scholar]
- 29.McKee MD, Fletcher J, Schechter CB. Predictors of timely initiation of gynecologic care among urban adolescent girls. J Adolesc Health 2006;39(2):183–191. [DOI] [PubMed] [Google Scholar]
- 30.Fairbrother G, Scheinmann R, Osthimer B, Dutton MJ, Newell K-N, Fuld J, et al. Factors that infl uence adolescent reports of counseling by physicians on risky behavior. J Adolesc Health 2005;37(6):467–476. [DOI] [PubMed] [Google Scholar]
- 31.McKee MD, Rubin SE, Campos G, O’Sullivan LF. Challenges of Providing Confidential Care to Adolescents in Urban Primary Care: Clinician Perspectives. Annals of family medicine 2011, vol 9 (1): 37–43. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Benard VB, Watson M, Castle PE, Saraiya M. Cervical Carcinoma Rates Among Young Females in the United States. Obstet Gynecol 2012; 120(5): 1117–1123. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
The data generated in this study are available upon request from the corresponding author.