Introduction

Shifting from cytology to HPV cervix screening will initially raise colposcopy referral rates. The anticipated impact on health systems has been a barrier to implementing this shift. It is unclear if increased referrals persist past initial HPV screens or revert to new lower baselines due to earlier detection and treatment of precancer.

Methods

Participants of the HPV FOCAL trial received one (HPV1, N = 6204) or two (HPV2, N = 9540) rounds of HPV screening. After exit, they returned to the BC cytology screening program. A comparison cohort from the BC screening population (BCS, N = 1,140,745) was extracted, mirroring trial inclusion criteria. All participants were followed for up to 10 years through the provincial screening registry. Trial and post-trial referral rates per 1000 screens (totals– HPV1: 27,341; HPV2: 36,982; BCS: 5,076,312) were calculated for each group under two HPV screening scenarios: (1) all HPV+ referred to colposcopy; (2) HPV+ to cytology triage with ASCUS+ referred to colposcopy. A multivariate flexible survival regression model compared rates throughout follow-up.

Results

Scenario 2 referral rates were higher after an HPV screen versus cytology screening (HPV1: 28 per 1000 women, HPV2: 32, BCS: 8). However, post-trial rates in HPV1 and HPV2 were significantly less than those in BCS during post-trial follow-up. Cumulative rates in HPV1 and HPV2 approached the cumulative rate in BCS by the end of follow-up (HPV1: 11 per 1000 women, HPV2: 16, BCS: 11). Adjusted HRs for referral in HPV1 and HPV2 compared to BCS were <1 beginning 24 months post-final HPV screen.

Conclusions

Reduced colposcopy referral rates are demonstrated after HPV screening implementation reached a steady state. After initial rounds of HPV screening, referral rates dropped below the current rates seen in a centralized cytology program. An expected increase in referrals could be moderated by program implementation strategies.

Introduction

Cervical cancer screening aims to detect precancers that can be effectively treated to prevent progression to cancer. However, prevalent cancers can also be detected, particularly in under-screened populations. Here, cervical cancers detected by screening efforts in ESTAMPA are described.

Methods

From 2012 to 2021, 42,502 women aged 30-64 years from 9 Latin American countries were screened with HPV-testing and cytology which included visualisation of the cervix during a speculum exam. Those positive for any test were referred to colposcopy with biopsy and treatment as needed. Women without high-grade disease (≤CIN1) were followed-up at 18 months with HPV-testing; those positive were referred again to colposcopy for disease ascertainment.

Results

55 cancers were detected, 47 at the initial screen and 8 at follow-up (mean age 43.8±9; 47 SCC and 8 ADC; 39, 5, 7 and 3 with FIGO stage I to IV; Figure 1). All cases were HPV-positive except for one SCC suspected of cancer at speculum exam (FIGO stage III, Figure 1-P55-45y). Cytology was abnormal for 32 cases (21 HSIL+, 6 LSIL and 5 ASC-US) and speculum exam for 24 (including 8 cancer suspicions). Initial colposcopy was positive for 51 cases, including 21 cancer suspicions, 16 major colposcopic changes (MCC) and 14 minor colposcopic changes (mCC); all histologically confirmed except for 4 mCC (3 CIN1 and 1 negative biopsy). Among those 8 cancers not initially detected (4 normal colposcopies and 4 mCC) follow-up colposcopy was positive for 7 (1 cancer suspicion, 3 MCC and 3 mCC; all histologically confirmed). One SCC was finally detected by LLETZ at 26 months (Figure 1-P50-36y).

Conclusions

Cancer detection by HPV-testing (98.2%) was superior to cytology (66.7%). Colposcopy with biopsy initially failed to detect 8 cancers (14.5%, assumed present at enrolment). Neither normal colposcopy nor negative biopsy provides absolute reassurance against cervical cancer, highlighting the need for follow-up.

Introduction

As Malaysia phases in HPV-based screening, it is critical to evaluate the management pathway for women who are screened positive. This study aims to evaluate the performance of the current triage strategy (combined HPV16/18 genotyping and cytology) and explore alternative triage options for the detection of high-grade precancers among HPV-positive women.

Methods

Malaysian women aged 30-65 years old who were screened positive for primary HPV testing and attended to colposcopy clinic, University Malaya Medical Centre from July 2018 to June 2022 were recruited. Liquid based cytology (LBC) and biopsy samples were collected under colposcopic guidance. A repeat HPV testing was performed using an aliquot of LBC. Immediate CIN2+ and CIN3+ risks were estimated and the triage strategies for colposcopy referral were evaluated using sensitivity, number of colposcopy needed to detect 1 CIN2+/CIN3+ case and referral rate.

Results

This study recruited 477 eligible HPV-positive women but 45 of them were excluded due to invalid test outcomes. The median days to follow up was 49 days (IQR: 33 – 71). Using 20% CIN2+ risk threshold, all HPV16/18 positive women and non-HPV16/18 positive women with LBC ASCUS+ (current strategy) would be referred to colposcopy (Figure 1A). It reduced the referral rate to 42.6% compared to absence of triage but would miss 20% of CIN2+ cases (sensitivity: 79.5%, CI: 68.4 – 88.0). Compared to the current strategy, triage using HPV16/18 genotyping and repeat HPV testing at 4% CIN3+ risk threshold (Figure 1B) had a higher sensitivity (91.8%, CI: 83.0 − 96.9) but resulted in 1 additional colposcopy needed to detect 1 case of CIN3+.

Figure 1. Pre-test-post-test probability plot associated with two triage strategies using 20% CIN2+ risk (A) and 4% CIN3+ risk (B) as the threshold for colposcopy referral.

Conclusions

The current strategy is 80% sensitive in detecting high-grade precancers. Repeat HPV-testing can be a potential triage strategy among non-HPV16/18 positive women.

Introduction

The rarity of invasive cervical cancer (ICC) means that most studies are not large enough to study cancer as an outcome and are obliged to report CIN2+ or CIN3+ as a surrogate. Long-term follow-up of large cohorts are therefore valuable for completing our understanding of risks following HPV infection.

Methods

Approximately 72,000 women were recruited to one of two large research studies when attending routine cervical screening in Greater Manchester, UK: 1987-93 for the Manchester Cohort (MC) and 2001-03 for the ARTISTIC Trial Cohort (AC). All women in the AC were tested for HPV using Hybrid Capture 2 and a random sample of stored samples taken from the MC were tested using HPV L1 MY09/MY11 consensus primers. Both cohorts were followed through national registration for cancer incidence and mortality to 2020.

Results

In 30 years of follow-up in the MC, 1152 women were diagnosed with CIN3 and 144 with ICC. A further 427 CIN3s and 32 ICC were diagnosed over the 17 years of follow-up of the AC. Risk patterns following HPV infection differed for CIN3 and ICC. Risk of ICC in the MC rises for 30 years following a single positive HPV test, reaching 2.4% (95%CI:1.3%-4.4%). A similar pattern was seen in the AC, but the risks of cancer were approximately halved. CIN3 was diagnosed much sooner in the AC due to more efficient cytology. More sensitive HPV testing was able to better predict future risk.

Conclusions

The sensitivity of HPV testing and cytology influence the CIN3 detection rate. Sensitive HPV testing enables effective risk stratification. Increased risk of ICC is observed 15-30 years after HPV infection. Women testing HPV+ should be followed until their infection clears. Discharging women from screening programmes whilst they remain HPV+ may not be safe, even if cytology and colposcopy tests are normal.

Introduction

In Argentina, HPV-testing was introduced in 2012-2014 through the Jujuy Demonstration Project (JDP) for women aged 30+. HPV-negative women were recommended re-screening in three years. In 2015 the screening protocol changed, and HPV-negative women were recommended re-screening in five years. Conventional cytology was used as triage. We evaluated implementation of the second round of screening in the JDP. We report preliminary results.

Methods

Retrospective cohort study. From the National Screening Information System (SITAM), we obtained data for all HPV-negative women tested in Jujuy from January 1, 2012 until December 31, 2014 (follow-up period January 2015-December 2021). We used the reach, effectiveness, adoption, implementation, and maintenance (RE-AIM) framework to evaluate implementation

Results

Of 42,350 previously HPV-negative women, 47.9% were re-screened in a second round (76.1% with clinician-collected tests and 23.9% with self-collected tests) three years after the first negative HPV-test or later (Average=5.4 years, range: 2.6-8.7). The HPV-test was HC2. Re-screening was opportunistic. 9959 (23.5%) of re-screened women were HPV-tested between 3-5 years after the first HPV-negative test. 83.3% of health centers provided second-round HPV-testing. Of the clinician-collected tests, 1884 were HPV-positive (8.8%), of which 395 (29.3%) were positive at cytology triage. Of the self-collected tests, 535 were HPV-positive (11.1%), of which 423 (79.1%) had triage Paps. 100 (24%) of self-collected tests were positive at cytology triage. 72.9% of the HPV-positive/abnormal Pap women had colposcopy. There were 80 CIN2+ (65 among clinician-collected tests; 15 among self-collected tests). CIN2+ detection was 0.40%. 85% of women with CIN2+ were treated. Re-screening of HPV-negative women first tested during 2015-2017 was 15.4%.

Conclusions

Re-screening of HPV-negative women at 3-5 years was relatively low, with most women being re-screened after 5 years. Adherence to follow-up was adequate, and most women received treatment. As CIN2+ detection was relatively high, strategies to increase women participation in second round should be implemented.

Introduction

In most high-resource populations in the Americas and Europe, HPV prevalence peaks shortly after sexual initiation and declines with age, with a moderate rebound around menopause often attributed to immune senescence. We hypothesized that, if viral latency is common, a general decline in cell-mediated immunity (CMI) around menopause would lead to increased appearance of multiple HPV types acquired cumulatively in prior years.

Methods

Using longitudinal data from the Guanacaste Natural History Study and the ASCUS-LSIL Triage Study, we identified women who were HPV-negative for all types at an index visit and tested HPV-positive for one or more HPV types at the subsequent visit (i.e., appearance). We grouped women based on age at appearance (<25; 25-34; 35-49; and ≥50 years) and number of HPV infections concurrently detected (1, 2, 3, ≥4 types). We calculated the ratio of multiple versus single types. We also calculated ratios after stratifying by recent and new sex partners, and lifetime sex partners (LTSP).

Results

The analytic population included 2146 women with at least one newly appearing HPV infection. Given new HPV appearance, the ratio of multiple versus single types was highest in the youngest age group and declined with age, contrary to the immune senescence hypothesis. This age-related decline held regardless of number of recent partners and LTSP, although multiple concurrently appearing infections were slightly more common among those with new partners and those with ≥4 LTSP at all ages (too few infections among women with new partners at ≥50 years to assess).

Conclusions

The ratio of newly appearing HPV infections that are detected concurrently with multiple types declines with age, regardless of past and recent sexual behavior. Epidemiologic evidence did not support our hypothesis that general CMI senescence leads to increased appearance of multiple infections, or increased detectability of ‘latent’ infections, with age.

Introduction

In Ethiopia, cervical cancer is the second leading cause of morbidity and mortality from all cancers in women. We determined the population‐based prevalence of HPV infection and genotype distribution, their persistence and clearance rates within 2 years of follow up.

Methods

A total of 893 rural women aged 30‐49 years in Butajira, south‐central Ethiopia were tested at baseline using self‐sampling device (Evalyn Brush®, Rovers, Oss, The Netherlands). HPV testing was done using multiplexed genotyping (MPG) by BSGP5+/6+ PCR with Luminex read out. All the hr-HPV positive women at baseline were invited for follow-up testing at 6 and 24 months. Cervical examination using VIA, cytology, and colposcopy (if indicated) was done during the follow up visits.

Results

HPV positivity rate at baseline testing was 23.2%. Of these 20.5% and 10.3% women were hr‐ and lr‐ HPV positive, respectively. Age‐specific hr‐HPV infection peaked in the age‐group 30‐34 years old (58.6%) and decreased in 35‐39, 40‐44, and 45‐49 years to 20.4%, 4.5% and 3.8% respectively. The top five prevalent hr‐HPV genotypes were HPV16 (57.1%), 35 (20.3%), 52 (15.8%), 31 (14.1%), and 45 (9.6%) in the Butajira district. The loss to follow-ups were 47 (30%) at the 6 months and 63 (40.1 %) at the 24 months. hr-HPV infection clearance was observed in 70 women (73.7%) within 6 months and among 77 women (84.6%) within 2 years. HPV68, 82, 53, 52, 56 were the most persisted genotypes with 100%, 75%, 42.9%, 31%, and 25% persistence rates respectively after 24 months.

Conclusions

This study provided new data on the overall prevalence of HPV infection and distribution of specific HPV types in rural Ethiopia. Most of the hr-HPV infections among rural Ethiopian women were cleared within 2 years.

Introduction

Australia is predicted to achieve cervical cancer elimination (<4 new cases per 100,000 women) at the national level within the next decade, but we have previously estimated this will take 20-25 years longer in Aboriginal and Torres Strait Islander women (hereafter respectfully referred to as Indigenous). Cervical cancer incidence among Indigenous women is around twice the national rate, driven by longstanding inequity in screening coverage. A policy change enabling universal access to self-sampling in Australia could increase screening participation. We aimed to predict if increased screening coverage and/ or vaccination uptake could expedite elimination of cervical cancer among Indigenous women.

Methods

We used an existing dynamic model of HPV transmission and vaccination and linked model of cervical screening and cancer in Australia (Policy1-Cervix) that has been adapted to reflect data on Indigenous women for HPV vaccination and screening coverage, and disease outcomes. We modelled a scenario that increased HPV vaccination from current levels (83%) to 90%, and several scenarios for screening improvements, including increased: i) uptake (ie reduce never-screened); ii) timeliness (reduce under-screening); and iii) attendance for follow-up tests in Indigenous women. The timing of cervical cancer elimination was estimated for each of these scenarios and compared with a scenario where coverage does not change.

Results

Improving vaccination coverage brought forward elimination in Indigenous women by one year (from 2049 to 2048). Increasing screening uptake, on-time attendance, and attendance for follow-up tests to all be consistent with national rates would expedite elimination of cervical cancer in Indigenous women by about eight years (to 2041). Improving screening uptake had the largest impact (three years, compared to two for follow-up attendance and two for routine screening timeliness) on reaching elimination.

Conclusions

Urgent action is required to close the gap in screening to ensure Indigenous women are not left behind in the goal to achieve elimination.

Introduction

The Food and Drug Administration supports innovations to facilitate new indications for high-risk human papillomavirus (HPV) testing. A new indication for Onclarity using PreservCyt liquid-based cytology (LBC) was evaluated.

Methods

The New Mexico HPV Pap Registry, a state-wide surveillance program was used to select an age- and cytology-stratified random sample of 19,879 women, ages 21-65 years, undergoing opportunistic cervical screening and follow-up in routine clinical practice. A subset of existing PreservCyt-based cervical specimens from these women (n=4,674) was utilized for comparison between Onclarity and cobas HPV assays. Point estimate differences and ratios were calculated for cervical disease detection and sensitivity, respectively, with 95% confidence intervals (95% CI). The cumulative risk of ≥CIN2 or ≥CIN3, with up to five -years of follow-up, was estimated for Onclarity by Kaplan-Meier.

Results

Negative HPV results (by either Onclarity or cobas) provided much lower CIN3+ risks (0.15%) than NILM cytology overall (0.44%), offering greater safety (Figure 1). Five-year cumulative ≥CIN3 detection was 5.55% and 4.59% for positive Onclarity and cobas, respectively (estimate difference=0.96% [0.47% to 1.51%]). Sensitivity for ≥CIN3 detected within <1 year for Onclarity and cobas was 95.3% and 94.5%, respectively (estimate ratio=1.01 [.98 to 1.06]) and sensitivity for Onclarity and cobas was also similar for ≥CIN3 detected up to five years. For ≥CIN3, positive agreement between Onclarity and cobas for HPV16 and HPV18, respectively, was 100% (95%CI=95.0% to 100%) and 90.9% (95%CI=62.3% to 98.4%). HPV16 carried the highest ≥CIN2/3 risk, followed by HPV18/31/33/58/52/45, and then HPV35/56/59/51/56/59/66.

Conclusions

Onclarity and cobas show equivalent performance using PreservCyt LBC media, and Onclarity genotyping provides effective ≥CIN2 and ≥CIN3 risk stratification. This real-world evidence study involved a unique, population-based design and a rapid, cost-effective approach to support a new HPV-device indication that would offer many additional healthcare providers in the United States a risk-based approach for management through extended genotyping.