Introduction

We conducted an integrated population-based analysis of subtype-specific cervical cancer incidence, survival, and incidence-based mortality by race and ethnicity, with correction for hysterectomy prevalence.

Methods

Using the Surveillance, Epidemiology and End Results (SEER) 21 and 18 registries, we evaluated age-adjusted incidence rates of primary malignant cervical cancer cases among women >15 years of age between 2000- 2018 (SEER21) and incidence-based mortality rates among deaths from 2005-2018 (SEER18), per 100,000 person-years, respectively. Rates were stratified by histologic subtype and race and ethnicity (incidence and mortality), as well as by stage and age at diagnosis (incidence only). Rates were corrected for hysterectomy using data from the Behavioral Risk Factor Surveillance System. We estimated five-year relative survival by histologic subtype and stratified by stage at diagnosis.

Results

Black and Hispanic women had the highest rates of cervical squamous cell carcinoma (SCC) respectively, while rates of cervical adenocarcinoma (ADC) were highest among Hispanic and White women, respectively, particularly for localized ADC. Incidence rates of both SCC and ADC peaked at ages 35-44 years for White women; however, rates of SCC and ADC in Black and SCC in Hispanic women continued to increase with age, peaking at 65-74 years. Although Black women had the lowest incidence of ADC, they had the highest overall mortality-rates and lowest five-year relative survival, for both SCC and ADC, irrespective of stage. Disparities in survival were particularly pronounced for Black women with regional and distant ADC, compared to other racial and ethnic groups.

Conclusions

Although Black women are less likely to be diagnosed with ADC compared to all other racial and ethnic groups, they experience the highest mortality rates for this subtype, likely attributed to the poor survival observed for Black women diagnosed with regional and distant ADC.

Introduction

Dual-stained cytology (DS) for p16/Ki67 is potentially a more effective triage than liquid-based cytology (LBC) after primary HPV screening, but data from HPV-vaccinated populations are limited. Compass is a major RCT in Australia - primary randomization is to HPV vs LBC screening; recruitment is stratified by birth-cohort according to whether offered HPV vaccination (born</>= 1980), with vaccination coverage of 50-70%+ in catch-up cohorts vaccinated in 2007-9 to 26 years.

Methods

In the main trial, 75,875 women aged 25-74 were recruited from 2015-2019; 50,732 were randomized to HPV screening, of these 43,693 were routine screeners; 576 (1.3%) had HPV16/18 and 3,396 (7.8%) had another high-risk HPV infection (HPVOHR). Women with HPVOHR were prospectively secondarily-randomized to DS vs LBC triage; women with HPV16/18 were referred to colposcopy with concealed DS testing. Triage-negative women had 12-month follow-up with HPV and a further 6 months follow-up for colposcopy/biopsy data; DS and LBC sensitivity and specificity were estimated using worst-case histology over the period.

Results

For routine screeners, for both CIN2+ and CIN3+, DS had high absolute sensitivity (77-88%) for both vaccinated/unvaccinated populations and across type-groups (HPV16/18 and HPVOHR), but specificity was higher for HPVOHR(68-74%) vs HPV16/18(50-60%) (Table 1). Findings were similar for all primary-randomized women (including those not in routine screening). In the prospective secondary-randomized comparison for HPVOHR vs. LBC, DS had higher relative CIN2+ sensitivity (1.63[95%CI:1.38–1.92]) but lower specificity (0.75[95%CI:0.73–0.77]). For CIN3+, relative sensitivity/specificity were 1.52[95%CI:1.24-1.87] and 0.73 [95%CI:0.71-0.75], respectively (Table 2).

Conclusions

DS triage performance was consistent across vaccinated/unvaccinated cohorts. The specificity of DS appears higher in HPVOHR. Relative to LBC, DS has higher sensitivity for CIN3+, indicating a useful role in programs with partial genotyping where DS+ HPVOHR+ can be immediately referred alongside HPV16/18+. DS has potential for automation which should facilitate implementation at scale in high-income countries.

Introduction

A fraction of HPV-positive women initially diagnosed with ≤CIN1 are subsequently diagnosed with CIN3 or cancer (CIN3+). We aimed to explore the role of triage tests and women’s baseline characteristics as potential markers for detection of CIN3+ during follow-up in HPV-positive women without cervical disease at entry in ESTAMPA.

Methods

In ESTAMPA, women aged 30-64 years were screened with HPV and cytology and referred to colposcopy with biopsy and treatment as needed. Those without evidence of cervical precancer (CIN1, negative biopsy, or negative colposcopy) were recalled to a follow-up visit at ~18 months for further disease ascertainment. Odds ratios adjusted by clinical and sexual and reproductive characteristics were used to estimate the risk of subsequent CIN3+ among HPV positive women initially diagnosed with ≤CIN1. The association of clinical and sexual and reproductive characteristics was also explored by triage result.

Results

3,161 HPV-positive women with ≤CIN1 at entry were included in the analysis. 2.6% (95%CI 2.1-3.2) had CIN3+ at the follow-up visit (73 CIN3, 3 AIS, 6 cancers). Women with positive triage results had at least 3-fold the risk of subsequent CIN3+ compared to those with negative triage (Table 1). Among women with ≥ASC-US, having ≥4 sexual partners and ≥2 years since the last cytology increased the risk of subsequent CIN3+ compared to those with less sexual partners and screened within 2 years (Table 2). Among women with non-HPV16/18 infection, having ≥2 years since the last cytology was also associated to subsequent CIN3+ (Table 2).

Conclusions

HPV-positive women with no evidence of precancer at screening but with positive triage results, particularly HSIL+ cytology, HC2-RLU≥10, and HPV16/18 are at higher risk of subsequent detection of precancer compared to those with negative triage and should be prioritised for follow-up. Additionally, sexual behaviour characteristics and screening history may help guide the clinical management of HPV-positive women.

Introduction

Oral oncogenic HPV infection is the cause of HPV-related oropharyngeal cancer. Unfortunately, little is known regarding the rate of acquisition of these infections or the natural history. The aim of this study was to estimate incidence of oncogenic oral HPV overall, and by country and age group, in the multi-national Human Papillomavirus Infection in Men (HIM) Study.

Methods

HIM Study participants (n=3,137) ages 18-70, who provided at least two oral gargle samples, every 6 months for 48 months, were included in this analysis. HPV was genotyped using the HPV SPF10 PCR-DEIA-LiPA25, (DDL Diagnostic Laboratory, Netherlands) line probe assay. Incidence rates were calculated using Poisson models and differences in oncogenic oral HPV acquisition by country and age were estimated using the Kaplan-Meier (KM) method.

Results

Among men from the USA (881), Brazil (1235), and Mexico (1021), 337 acquired an oral oncogenic HPV infection. Incidence was 2.4 per 1000 person-months with incidence significantly higher among men from the USA compared to men in Brazil and Mexico (3.8 vs 1.9 and 1.7 per 1000 person-months, respectively). HPV types 52 and 16 were the most common among all men, at 0.8 and 0.6 per 1000 person-months respectively. Men acquired new oral oncogenic HPV evenly over time (2.4, 2.8, 2.6 and 2.4 per 1000 person-months at 12, 24, 36 and 48 months respectively). Probability of new oral oncogenic infections was higher among men ages 18-24 years (HR; 95% CI 1.44 (1.01-2.05) compared to older age groups, with the effect driven by a higher incidence in younger men in Brazil.

Conclusions

Our data suggest that men are at risk of acquiring oral oncogenic HPV across the lifespan with incidence higher among men from USA than those from Mexico and Brazil. To better inform development of preventative interventions, future studies should elucidate acquisition of oncogenic oral HPV by age.

Introduction

The Canadian Cancer Registry does not collect demographic data beyond age and sex, making it hard to monitor socioeconomic inequalities in cancer incidence in Canada, a country with public healthcare and a large number of immigrants. We used data linkage to compare cervical and anal cancer incidence rates by race and visible minority group.

Methods

We used data from the CanCHEC 2006 and 2011 cohorts, which are population-based probabilistically linked datasets of 5.9 million respondents of the 2006 Canadian long-form census and 6.5 million respondents of the 2011 National Household Survey. Respondents’ Indigenous identity and visible minority group identity were self-reported. Respondent data were linked with the Canadian Cancer Registry up to 2015. We calculated age-standardized incidence rate ratios (ASIRR), comparing group-specific rates to the overall population rate with bootstrapped 95% confidence intervals (95%CI). We used negative binomial regressions to adjust rates for socioeconomic variables.

Results

Age-standardized cervical cancer incidence rates were significantly lower in Chinese (ASIRR 0.64, 95%CI 0.50-0.78) and South Asian (ASIRR 0.66, 95%CI 0.49-0.86) women, and were significantly higher in women not belonging to a visible minority group (ASIRR 1.04, 95%CI 1.02-1.06) and Indigenous women (ASIRR 1.61, 95%CI 1.38-1.85) compared to the overall population. Differences in incidence rates by race and visible minority group persisted even after adjusting for household income, education, rural residence, and immigration status. Similar results were observed for anal cancer, with significantly lower incidence rates in Chinese and South Asian populations than in the overall population.

Conclusions

Although cervical cancer screening rates are known to be lower in visible minority women in Canada, many visible minority women had lower cervical cancer incidence rates than women who do not identify as visible minorities. Differences are potentially attributable to variations in human papillomavirus infection prevalence, as well as a strong healthy immigrant effect resulting from immigration selection processes.

Introduction

The only population-level evaluation of South Africa’s cervical cancer (CxCa) screening programme is a biased estimate of screening coverage based on an aggregated count of screens divided by the population eligible for screening each year. There is no routine reporting of the linkage of referrals to colposcopy clinics or to CxCa treatment.

Methods

In the public sector of the Western Cape, a unique health identifier is used in electronic record keeping in healthcare facilities. This identifier enables linkage across several databases, and we use this data to estimate screening coverage, screening intervals, and linkage to treatment, by HIV status.

Results

We estimate 10-yearly screening coverage among HIV-negative women or women with unknown HIV status to be 59.8% in 2019/20. Coverage of 3-yearly screening among women with HIV is estimated to be 53.6% in 2019/20. The screening schedule of once in 10 years for HIV-negative women and once in 3 years for women with HIV is not adhered to – 1 in 5 HIV-negative women return for another routine screen within 3 years and only half return within 10 years. The median time between screens for women with HIV is 5.1 years. Around 40% of HIV-negative women and 50% of women with HIV who were referred to treatment in 2018 had no evidence of attending colposcopy services.

Conclusions

CxCa prevention in the public sector of the Western Cape does not meet the 90-70-90 targets as proposed by the WHO’s CxCa elimination strategy. Although around 90% of pre-adolescent girls who attend public schools get vaccinated, less than 60% of women get screened at the appropriate times and less than 50% of women had evidence of accessing treatment. This study shows the crucial importance of a unique health identifier to improve patient-level care and to monitor progress toward the 90-70-90 elimination targets.

Introduction

Cervical HPV prevalence is high among women living with HIV (WLWH) and transient infections in liquid-based cytology (LBC) samples are frequent. HPV testing on cervical histopathology samples may help identify persistent genotypes. We explored HPV genotype agreement in paired LBC and histopathology samples among WLWH in Zambia.

Methods

We included WLWH aged 18-65 years who participated in a cervical cancer screening test accuracy study and returned for follow-up after 30-36 months. We obtained LBC samples and ≥2 cervical biopsies, which were formalin-fixed and paraffin-embedded, from all participants. Both samples were tested for 28 HPV genotypes (Anyplex II HPV28, Seegene, Seoul). We calculated type-specific HPV prevalence and used Cohen's kappa to assess agreement beyond chance between HPV on cytology and histopathology.

Results

From February to September 2022, 146 WLWH (133 <CIN2, 13 CIN2+) had valid paired HPV results. Prevalence of any HPV was similar on LBC (46%, n=67) and histopathology (36%, n=53). Multiple-type infections were more common in LBC (19%, n=28) than histopathological samples (9%, n=13). HPV-16 was the most frequent genotype: 9% (n=13) on LBC, 14% (n=21) on histopathology. Overall agreement for any HPV was 58.9% (Kappa=0.16, 95% confidence interval [CI] 0.00-0.32), with frequent discordance between sample types (Figure). Sixteen women (15 <CIN2, 1 CIN2+) had HPV-16 detected on histopathology but not on LBC (Kappa 0.21, 95% CI 0.00-0.42), of whom four had been treated for CIN2+ diagnosed at the initial study visit; one woman had persistent CIN2+.

Conclusions

Our data from an ongoing study of WLWH with mostly low-grade cervical lesions in Zambia show more multiple-type infections on LBC than histopathology and limited HPV genotype agreement between sample types. This may be partly explained by more frequent transient HPV infections in LBC. The substantial proportion of WLWH with HPV-16 on histopathology but not on LBC warrants further investigation.

Introduction

The CATCH trial’s interim analysis (June 2017, n=277) demonstrated a 36% protective effect of carrageenan. We now report the final analysis of efficacy and safety of a carrageenan-based gel in reducing incidence and prevalence of genital HPV among sexually-active young women.

Methods

The CATCH trial was conducted in Montreal, Canada. Its primary outcomes were HPV type-specific incident and clearance of prevalent infections. Women were randomized (1:1) to a carrageenan-based or placebo gel to be applied every other day for the first month and before, during, and after intercourse as needed. At each study visit (months 0, 0.5, 1, 3, 6, 9, 12), participants provided questionnaire data and a self-collected vaginal sample, which was tested for 36 HPV types (Linear Array). Intention-to-treat analyses were conducted using Cox proportional hazards regression models with participant-level and HPV-level (unit of observation being HPV type) data. Trial registration: ISRCTN96104919.

Results

461 enrolled participants (2013-2020) were randomized to the carrageenan (n=227) or placebo (n=234) arm. The last follow-up visit occurred in October 2021. Incidence, clearance, and safety analyses included 427, 239, and 441 participants, respectively. We restricted analyses to having valid HPV results for ≥2 visits. Considering 427 participants with valid baseline HPV data, we found a 37% reduction in the risk of incident infection among participants in the carrageenan compared to those in placebo arm (hazard ratio [HR]=0.63, 95% confidence interval [CI]:0.49-0.80). Similar results were obtained with HPV-level data. Of 241 participants positive for HPV at baseline, clearance of infections (two consecutive HPV-negative visits following ≥1 positive visit) was comparable between groups (HR=1.16, 95% CI:0.73-1.84). A total of 37% (79/212, carrageenan) and 41.0% (94/229, placebo) reported an adverse event (p-value=0.42).

Conclusions

Consistent with results of its interim analysis, the CATCH trial found that use of a carrageenan-based gel reduced the risk of incident genital HPV infections in women.

Introduction

In El Salvador, cervical cancer prevention is a mixture of cytology and primary HPV screening. The latter consists of a first visit for HPV testing, followed by a second visit to receive results and ablation treatment for HPV positive (HPV+) women. There is a 30-day wait time between visits for test results to be ready. During the pandemic, non-emergency services were interrupted, personnel were shifted to COVID-related tasks, and fear of contagion resulted in an estimated 5,000 screen-positive women who were not able to receive follow-up care. In partnership with the Ministry of Health (MoH), we devised a rapid response approach to provide care to these women.

Methods

From April 2021 date, we used MoH records to identify women who were either HPV+ or had abnormal cytology but no follow-up. A team of 8 colposcopists and 8 nurses were contracted to provide colposcopy and biopsy to all women to determine management. Rates of cervical intraepithelial neoplasia grade 2 or higher (CIN2+) and invasive cervical cancer (ICC) were compared with available historical data from a similar population of participants in the HPV program.

Results

Out of 3,861 women identified, 2,853 (74%) underwent colposcopy and biopsy, 702 (18%) had been treated privately, 210 (5.5%) declined care, and 96 were lost to follow-up (2.5%). Median days between screening test and colposcopy/biopsy were 112 (IQR = 210.75).

Biopsy results were 1,976 (69.3%) normal, 648 (22.7%) CIN2+, and 26 (.01%) ICC, while 203 (7.1%) were missing. In contrast, data from 2013-2014 show lower CIN2+ (56/385 [14.5%]) and ICC (1/385 [.003%]) rates.

Conclusions

Even a two-year interruption in cervical cancer prevention programming had noticeable effects in a vulnerable population. Efforts to strengthen such programs and to increase access to screening and treatment must be priorities in cervical cancer prevention in low and middle-income countries.