Anal Dysplasia Screening Among Women Living With HIV in Metropolitan Miami
Isabella Rosa Cuhna1, Alex Sanchez Covarrubias1, Patricia Jeudin1, Faith Doyle1, Karla McGuire1, Carlos Briones Chavez1, Daniel M O'Shea1, Thomas Iglesias1, Johnny Galli1, JoNell Potter1, and Lunthita Duthely1. 1University of Miami
Objective: To evaluate anal cytology and histology results for women living with HIV, attending an anal dysplasia clinic in Miami, FL.
Methods: Retrospective chart review of women living with HIV who attended the anal dysplasia clinic at Jackson Memorial Hospital in Miami, FL between September 2016 and September 2018 and underwent high resolution anoscopy (HRA) with anal cytology and histology of visible lesions-captured the most severe histologic result per patient. Data were analyzed using SAS-UE-2018.
Results: A total of 215 women were identified. Median age was 53 (IQR=11), 95% were on combination antiretroviral therapy (cART), 79% had CD4>200/mm3 and 92% were virologically suppressed (<200 copies/mL). Overall, 81% had biopsy-proven anal dysplasia. Anal cytology results included: 9 Negative, 101 ASCUS, 80 LSIL and 25 HSIL. Histology findings were: 40 benign, 102 LSIL, 73 HSIL. Correlation of anal cytology and histology is demonstrated in Table 1. An inter-test agreement was assessed between anal cytology and anal histology, with an unweighted kappa of 0.141 (95% CI [0.054, 0.23])—suggesting low agreement over what was expected due to chance.
Conclusions: Preliminary analyses suggests that women living with HIV in Miami have high rates of anal dysplasia, including high grade dysplasia, the anal cancer precursor. Anal cytology results did not predict anal histologic results—specifically, we found only 14% agreement over expected results. Findings support the need for further comprehensive studies to guide patient care, particularly screening recommendations for anal cytology and HRA for this high-risk population.
High-Risk HPV-Negative Anal HSILs: Potential Causes and Clinical Implications
Yuxin Liu1, Keith Sigel1, Wenxin Zheng2, and Michael Gaisa1. 1The Mount Sinai Hospital, 2UT Southwestern Medical Center
Objective: High-grade squamous intraepithelial lesions (HSIL) in the lower anogenital tract are virtually always caused by high-risk human papillomavirus (HPV) infection. High-risk HPV-negative HSILs occur occasionally in clinical practice and constitute an unexpected departure from that rule, leading to diagnostic and therapeutic challenges. Using a cohort of patients with biopsy-proven anal HSIL but negative HPV results, we aimed to investigate the potential causes for these discrepant cases and their clinical implications.
Methods: We searched the pathology database between 2012 to 2018 for patients with biopsy diagnosis of anal HSIL and concurrent negative HPV in anal cytological sample according to the Cobas HPV DNA test (16,18, and 12 other high-risk HPV types). Demographic characteristics and cytological diagnoses were extracted from medical records. Biopsy samples containing HSIL were tested for HPV DNA using the BioPerfectus Multiplex Real Time (BMRT) HPV assay to detect 3 low-risk types (6, 11 and 81) and 18 high-risk types (16,18, 26, 31, 33, 35, 39, 45, 51, 52, 53, 56, 58, 59, 66, 68, 73 and 82).
Results: Among 1,088 anal cytological samples collected from patients with anal HSIL, 43 (4%) tested negative for high-risk HPV types by Cobas assay. Median age was 50 years (range 23-72). Subjects were HIV-infected (91%), men who have sex with men (88%) and women (12%). Cytological diagnoses were unsatisfactory (12%), benign (16%), ASCUS (49%), LSIL (14%), ASC-H (5%), and HSIL (4%). Compared to HPV-positive samples, HPV-negative ones had a higher percentage of unsatisfactory samples (12% vs. 5%) and patients with only one or two HSILs (95% vs. 77%). There was no significant difference between the two groups regarding age, gender, race, smoking history, HIV status, CD4+ T cell count, and HIV viral load. For HPV-negative cases, BMRT HPV assay on biopsy specimens revealed that 12 lesions were negative for HPV and 2 were positive for low-risk HPV 6. The remaining 26 lesions were positive for high-risk HPV. Half of HPV positive lesions harbored types included in the Cobas assay (16, 18, 31, 33, 35, 45, 51, 52, 58 and 59), while the other half harbored hrHPV types not included in the assay (53, 67, 69, 73, and 82). The most common type detected in HSIL but not included in the Cobas assay was HPV 67 (n=6).
Conclusions: As with cervical HSIL cases, outlying HPV-negative results occur at a low frequency among patients with biopsy-proven anal HSIL. Our results should help clinical management for these discrepant cases by providing potential causes such as unsatisfactory samples, limited disease, and most importantly, rare high-risk HPV types that are not routinely screened.
Recurrence After Treatment of Anal High-Grade Intraepithelial Lesions (HSIL) in HIV-Infected (HIV+) Women
Amma Agyemang1, Wafaa Abbasi1, Eduardo Valle1, and Elizabeth Stier1. 1Boston Medical Center
Objective: HIV+ women have a high risk of invasive anal cancer. Prevention of anal cancer with early detection and treatment of anal HSIL, similar to cervical cancer prevention, is recommended. However, there is limited data on the efficacy of anal HSIL treatment in HIV+ women. This study investigated the rate of recurrent anal HSIL after treatment in HIV+ women.
Methods: We conducted a retrospective study of HIV+ women with follow-up after treatment for anal HSIL from January 1st, 2004 – December 31st, 2017 at an urban tertiary care hospital.
Results: 48 HIV+ women with follow-up after treatment for anal HSIL were identified. The mean age at diagnosis was 51.8 years (range 33-66 years). 62.6% were African American, 27% were Hispanic/Latino, 68.7% were current or past smokers, and 50% had past cervical- and 24% past vulvar-intraepithelial neoplasia. At 2 years post-treatment, 30 (62.5%) women had recurrent anal HSIL. Significant risk factors for recurrence at 2 years included mean age at diagnosis of anal HSIL (47.4 ± 7.7 versus 42.2 ± 8.4 years, p<0.05) and duration of known HIV infection (15.6 ± 8.2 versus 10.1 ± 7.3 years, p<0.05) for patients with recurrence compared with those without recurrence respectively (see Table 1).
Conclusions: HIV+ women with anal HSIL have high recurrence rates post-treatment (similar to rates of recurrence of cervical HSIL in HIV+ women). Our data contributes to the information needed to develop effective anal cancer prevention guidelines in HIV+ women.
Acceptability of Novel Treatments for Cervical Pre-Cancer for Use in Low and Middle-Income Countries
Miriam Cremer1, Montserrat Soler2, Raul Murillo3, Ana Celia Uriarte4, Suhui Wu5, Mauricio Maza2, Rachel Masch2, and Karla Alfaro2. 1Cleveland Clinic, 2Basic Health International, 3Hospital Universitario San Ignacio, 4Instituto Salvadoreño del Seguro Social, 5Shanxi Dayi Hospital
Objective: To assess patient pain and side-effects of treatment alternatives for cervical pre-cancer in low and middle income countries (LMICs).
Methods: Data was collected from 167 women enrolled in an ongoing study in El Salvador, Colombia, and China. Women were randomized to three arms: double-freeze CO2 cryotherapy, single-freeze CryoPen® (non-gas cryotherapy), and thermoablation. Pain during treatment was assessed using the Wong-Baker FACES scale (0 = no pain, 10 = most pain). ANOVAs were used to compare pain score means. Percentages and chi square tests were used to describe and compare the proportion of women reporting side-effects at a 6-week follow-up visit.
Results: There were no significant differences among the three treatment arms in mean pain during treatment (cryotherapy = 2.91±2.91, CryoPen 3.78± 2.98, and thermoablation = 3.76±3.12, F(2,163) = 1.53, p= 0.22). At the 6-week visit, most participants (96%) reported experiencing at least one side effect. The most common of these were mild to moderate fluid discharge (94.4%) and cramps (56.4%). Less than a third of patients reported malodorous discharge (30.2%) and spotting or bleeding (26.2.%). There were no significant differences between treatment group and side effects, whether it was any side effect at all X2= 1.35 (p = .509) or specific side effects across groups, X2=6.16, p=0.629. Most women (90.8%) reported they would recommend the treatment they received to a friend.
Conclusions: CryoPen® and thermoablation do not present more side effects and are as acceptable as conventional gas-based cryotherapy to most women.
Long Term Follow-Up of Women With AIS Diagnosed in Cervical Cone Biopsy
Swikrity Upadhyay Baskota1, and Chengquan Zhao1. 1UPMC Magee Women's Hospital
Objective: To evaluate the risk of residual/recurrent disease for women with cervical adenocarcinoma in situ (AIS) after conization and follow-up results.
Methods: This is a retrospective analysis of HPV, cytologic and histologic status of 208 women with AIS and LEEP/cone biopsy from 2009- 2018.
Results: The median age of the patients was 36 years (range=22-65 years). 90 patients (43.3%) also had CIN2/3) in addition to AIS. Out of 208 AIS cases with the LEEP/cone biopsy, 64 cases (30.8%) had positive margins. Rate of residual AIS (52.8%) and invasive adenocarcinoma (5.6%) was very high in hysterectomy specimens with positive margins compared to hysterectomy specimens with negative margins: 6.9% for AIS and 1.7% for invasive adenocarcinoma of cervix respectively. In women treated with conservative follow-up, residual AIS was seen in 43% (n=7) of cases of women with original positive cone margin and in 4.9% (n=2) in negative margin group. There were total of 16 HPV positive cases (7.7%) in the follow-up period. Five patients had margin positive cone biopsy and were treated with follow-up cervical biopsy. Two had residual AIS. One of the margin negative cases with HPV positive result had residual AIS in follow-up biopsy.
Conclusions: Our study demonstrates the positive margin on the conization is an important risk factor for the residual AIS with both hysterectomy and conservative follow-up. Pap with concurrent HPV test is sensitive to detect the recurrent disease.
Medical Chart Review of 376 Women With Invasive Cervical Cancer: Case Investigation of Cervical Cancer Study, 2013–2016
Virginia Senkomago1, Vicki Benard1, Warner Huh1, Aprl Greek3, J. Elizabeth Jackson3, Cheryll Thomas1, and Lisa Richardson1. 1CDC, 2University of Alabama at Birmingham, 3Battelle
Objective: To examine medical chart abstraction of the 5 years of a woman’s medical history up to and including the date of diagnosis with invasive cervical cancer to determine screening history.
Methods: The study included women with cervical cancer diagnoses reported in state cancer registries in Louisiana, Michigan and New Jersey (2013-2016) who consented to have their medical chart data reviewed and for whom medical records could be obtained. We defined a valid screening test as one that was conducted 6 months prior to the date of diagnosis. We analyzed data from 376 women, assessing Pap or human papillomavirus (HPV) testing in the 5 years prior to diagnosis.
Results: A majority of the women (n=311; 83%) had a Pap or HPV test recorded in their medical charts in the 5 years prior to their cervical cancer diagnosis. However, the majority (n=178; 57%) of these women had only one test confirmed in their record. Of those with 1 Pap or HPV test for whom testing date was available (n=161), most (n=149; 92%) were collected within 6 months of the cancer diagnosis. Sixty % of the women diagnosed with invasive cervical cancer, based on our definition, were not screened in the 5 years prior to diagnosis.
Conclusions: This is the only population-based study to measure screening history in the 5-years prior to cervical cancer diagnosis. Medical chart abstraction combined with a mailed survey of behaviors provided additional insight on how to address increasing regular and timely cervical cancer screening.
Optimization of a Thermoablator Device Adapted for Use in LMICs
Miriam Cremer1, Montserrat Soler2, Karla Alfaro2, Rachel Masch2, Mauricio Maza2, and Stephan Hilgers3. 1Cleveland Clinic, 2Basic Health International, 3WiSAP Medical Technologies
Objective: There is an urgent need for portable cervical precancer treatments that do not require gas. Thermoablation is a potentially game-changing option that uses electricity or a battery. However, the most commonly used device is not well-suited for remote or low-income settings where the need is greatest. We have collaborated with the manufacturer of the most widely used thermoablation device to develop a new prototype uniquely suited for the needs of LMICs.
Methods: The optimized handheld thermoablator is lightweight (approximately 2 pounds) and features an external rechargeable battery than can treat up to 100 patients per charge. Integrated side lighting and an elongated probe shaft improve visualization of the cervix. The probe includes a safety cover that prevents vaginal burns. Importantly, probes are interchangeable and feature a variety of sizes and shapes.
Results: The optimized device will be utilized in an upcoming 3-arm non-inferiority randomized clinical trial designed to compare two thermoablation treatment protocols against conventional double-freeze gas-based cryotherapy. One protocol will utilize a single 16mm tip for multiple exocervical applications of 20-seconds. The second protocol will replicate the most-widely used treatment consisting of a 20 second endocervical application with a 12mm tip followed by 20 second exocervical applications with a 16mm tip as needed. Approximately 1,282 women with biopsy-confirmed CIN2+ will be enrolled and treated. We estimate that 80% (n = 924) will return at 12-months post-treatment to assess cure rates.
Conclusions: If effective, the optimized device can help standardize thermoablation guidelines and increase access to cervical precancer treatment for thousands of women.
Racial/Ethnic Differences in Age, Disease Grade, and Time-to-Treatment Among Incident CIN2+ Cases in Alameda County, CA
Deanna Fink1, Kayla Saadeh1, Erin Whitney1, and Ina Park2. 1California Emerging Infections Program, 2University of California San Francisco
Objective: In the United States, cervical cancer incidence is higher among Black and Hispanic women compared to women of other races/ethnicities. We analyzed median age, disease grade, and time-to-treatment for incident high-grade cervical intraepithelial neoplasia (CIN2+) by race/ethnicity.
Methods: Data were analyzed from population-based surveillance of CIN2+ in Alameda County, CA (HPV-Impact) from 2008-2016. We evaluated median age at CIN2+ diagnosis, disease grade (CIN2 or CIN2/3 vs CIN3 or adenocarcinoma in situ (AIS)), and median days from screening test to diagnostic biopsy and from biopsy to treatment by race/ethnicity (Asian, White, Black, and Hispanic), with pairwise comparisons, using Kruskal-Wallis tests for median age and time-to-treatment, and chi-square for disease grade.
Results: Among 5,295 women with CIN2+, median age at diagnosis among Asian women was higher compared to other races/ethnicities (all p<0.0001) (Table 1). Black women were less frequently diagnosed with CIN3 or AIS compared to other races/ethnicities (all p<0.005). Compared to White women, Black and Hispanic women had significantly more time between screening and diagnosis, and between diagnosis and treatment (all p<0.005). One-quarter of Black women had ≥111 days between abnormal screen and diagnostic biopsy.
Conclusions: Delays in care experienced by Black and Hispanic women with CIN2+ may not be clinically significant but may indicate barriers to healthcare access and/or utilization. Later age at CIN2+ diagnosis among Asian women may be indicative of later initiation or less frequent cervical cancer screening compared to other races/ethnicities. Further investigation is warranted to identify underlying reasons for racial/ethnic differences in cervical pre-cancer diagnosis and management.
Scaling Up Use of Human Papillomavirus (HPV) Testing in Central America
Silvia de Sanjose1, Francesca Holme1, Karla Alfaro2, Claudia Camel3, Francisco Maldonado4, Manuel Sandoval5, Rose Slavkovsky1, and Jose Jeronimo6. 1PATH, 2Basic Health International, 3Ministerio de Salud Publica y Asistencia Social, 4Fundacion Movimiento Contra el Cancer (Movicancer-Nicaragua), 5Asociacion Hondurena de planificacion de familia (ASHONPLAFA), 6Global Coalition Against Cervical Cancer
Objective: To improve quality of cervical cancer screening in Guatemala, Honduras, Nicaragua, and El Salvador by introducing HPV screening tests in predefined areas and providing technical assistance to ministries of health (MOHs). We present the Scale-Up results for 2015–2018.
Methods: PATH and local partners engaged key stakeholders at various levels of MOHs to define target areas, algorithms, laboratory locations, and outreach strategies. Free-of-charge screening was offered to women older than 30 years of age using careHPV. Active community-mobilization strategies were implemented. Self-sampling was offered in Guatemala and Nicaragua and in a subset of the target areas in Honduras, but not in El Salvador. HPV tests were processed locally. MOHs routinely compiled data on women screened, screening test used, use of self-sampling, age, time elapsed since last screening, and number of eligible women treated. Data was gathered by local NGOs and jointly analyzed with PATH.
Results: From 2015–2018, 249,089 women were HPV screened, with positivity ranging from 11.9%–14.3%. First-time screening was highest in Nicaragua at 47.3% where self-sampling was also most widely used, at 95.8% of women screened. Among HPV-positive women with a positive result in a triage test, 84.8%, 39.5%, and 64.6% were treated in Guatemala, Honduras, and Nicaragua respectively; in El Salvador, where all HPV-positive women are eligible for treatment, 76.6% were treated.
Conclusions: MOHs have overcome challenges to implement HPV testing, reaching many unscreened women through self-sampling and in some cases achieving high treatment completion. Local leadership, political, systemic, and policy issues are key in determining screening-program performance.
SOX1 Methylation as a Promising Marker in Cervical Adenocarcinoma
Jing Wang1, Nayiyuan Wu1, Zitong Zhao1, and Peilin Ouyang2. 1Hunan Cancer Hospital, 2Hunan Cancer Hospital, The Affiliated Cancer Hospital of Xiangya School of Medicine, Central South University
Objective: This study aimed to evaluate association between methylation status of SOX1 and histologically confirmed cervical adenocarcinomas (CAs).
Methods: The methylation status of SOX1 was assessed by methylation-specific polymerase chain reaction. The status of SOX1m were compared between patients with ACs and normal cervix and associated with clinicopathological factors using Chi-square test. Overall and disease-free survival (OS and DFS) rates were estimated and compared by Kaplan-Meier method. A Cox regression model was used to assess the prognostic significance of SOX1m in cervical adenocarcinoma patients. Sensitivity and specificity for the the DNA methylated level were analyzed.
Results: In total, 186 ACs and 30 normal control cervices between December 2012- April 2015 were enrolled in this study. The mean methylation (M)-indices of SOX1 was significantly higher than noncancer tissues (P <0.01). SOX1 methylation statue was significantly correlated with lymph node metastasis (P <0.001). We found a significantly shorter overall survival in SOX1m-negative cervical adenocarcinoma patients (P=0.038), especially in small tumor size patients with SOX1m -negative (P<0.01).
Conclusions: These findings indicate that low methylation statue of the SOX1 are associated with aggressive progression and poor prognosis of cervical adenocarcinoma.
The Effect of HPV Testing in the Detection of Adenocarcinoma In Situ and Adenocarcinoma of the Cervix
Ran Zhang1, Nujsaubnusi Vue1, Lisa Obasi1, Rachel Isaksson Vogel1, Amar Subramanian2, Mahmoud Khalifa1, Buvana Reddy2, and Britt Erickson1. 1University of Minnesota, 2HealthPartners
Objective: Evaluate if method of diagnosis (HPV testing alone, cytology alone, or HPV and cytology) of AIS or occult adenocarcinoma affects clinical outcome and management.
Methods: This was a multicenter retrospective cohort study. All sites provided IRB approval. Patients with AIS or occult carcinoma detected by routine screening between 2007-2017 were divided into 3 cohorts based on which test led to their diagnosis: (1) HPV positive with normal cytology (HPV+/Pap-), (2) HPV negative with abnormal cytology (HPV-/Pap+), (3) HPV positive with abnormal cytology (HPV+/Pap+). Clinical demographics, treatment course, and surgical pathology were reviewed.
Results: 130 patients were diagnosed with adenocarcinoma in situ (n=108) or adenocarcinoma (n=22) based on abnormal screening tests. Sixteen subjects were HPV+/Pap- (12%), 11 were HPV-/Pap+ (8%), and 103 were HPV+/Pap+ (79%). Demographics were similar between groups, though HPV+/Pap- patients were slightly older and more likely to have used an IUD for contraception compared to the other two groups. Rates of positive margins were highest in the HPV-/Pap+ group (63.6%) and similar in the HPV+/Pap+ (26.5%) and HPV+/Pap- (28.6%) groups (p=0.05). Similar percentages of women in each group were offered fertility preservation (50.0-61.3%, p=0.79). After adjusting for differences in age and margin status, rates of definitive hysterectomy were similar between groups (52.1-81.3%, p=0.59).
Conclusions: We did not observe differences in clinical outcomes of lesions detected by abnormal HPV testing alone versus abnormal co-testing. Notably, a significant proportion of adenocarcinoma in situ and adenocarcinoma was detected by both HPV alone and pap alone, arguing for continued HPV and pap co-testing and against the options of HPV testing alone or cytology alone for cervical cancer screening.
Trends in HPV Type-Specific Cervical Precancer Rates Among Young Women in the United States, 2008-2014
Nancy McClung1, Julia Gargano1, Rayleen Lewis1, Marie Griffin2, Nancy Bennett3, Erin Whitney4, Nasreen Abdullah5, Monica Brackney6, Angela Cleveland1, Troy Querec1, Elizabeth Unger1, and Lauri Markowitz1. 1Centers for Disease Control, 2Vanderbilt University Medical Center, 3University of Rochester, 4CA Department of Public Health, 5Oregon Health Authority, 6Yale University
Objective: Declines in cervical precancer incidence rates in young women have been attributed to HPV vaccination, although changes in cervical screening and management guidelines could also have contributed to observed trends. To further evaluate vaccine impact, we described trends in HPV type-specific cervical precancer rates among women aged 18-24 years, 2008-2014.
Methods: We analyzed data on cases of cervical intraepithelial neoplasia grades 2-3 and adenocarcinoma in situ (CIN2+) from a 5-site population-based laboratory surveillance system. HPV DNA typing was performed on all available archived diagnostic specimens. HPV types were grouped hierarchically: 16/18, 31/33/45/52/58, other high-risk (OHR) types, and any other types. HPV type-specific CIN2+ incidence per 100,000 women screened for cervical cancer was estimated using inverse probability weights based on diagnosis, year, and site. We estimated average annual percent changes (AAPCs) using log-linear models.
Results: HPV typing results were available on 2,383 (69.0%) cases. From 2008-2014, HPV 16/18-positive CIN2+ rates decreased from 397 to 133 per 100,000 screened 18-20 year-olds (AAPC= -20.7%, 95% CI [-29, -11.3]), and from 667 to 340 per 100,000 screened 21-24 year-olds (AAPC= -11.6%, 95% CI [-14.6, -8.4]). No significant declines were observed in rates of CIN2+ with 31/33/45/52/58, OHR, or other types in both age groups (Table 1).
Conclusions: Declines in rates for vaccine-type, but not other HPV type-associated CIN2+ further supports attributing decreasing cervical precancer rates in young women to vaccination. Continued surveillance of cervical lesions with HPV typing is important to monitor the impact of changes in vaccination policy and recommendations for cervical screening and management.
Correlation of Universal Endocervical Sampling and Presence of High Grade Endocervical Disease at Time of Treatment
Danielle Chiavetta1, Christine Holschneider1, Chisa Aoyama1, and Lisa Garcia1. 1Olive View-UCLA Medical Center
Objective: Using a population of patients with high-grade cervical dysplasia (CIN2+) who underwent universal endocervical sampling (ECS) at colposcopy, we sought to determine the rate of high-grade ECS at time of colposcopy and presence of endocervical disease at time of treatment in two patient cohorts: patients for whom ECS at colposcopy is preferred by ASCCP (pECS) versus patients for whom ASCCP denotes performance of ECS acceptable (aECS).
Methods: Retrospective review of consecutive cohort of 390 women with biopsy-proven CIN2+ for which excisional treatment was performed between 2005 and 2015. Endocervical disease at time of excision was defined as CIN2+ in top hat loop electrosurgical excision procedure specimen, CIN2+ at endocervical margin of cold knife cone, or ECC with CIN2+ at time of excision. Fisher’s exact and student’s t-tests were used to examine clinical outcomes.
Results: The rate of positive ECS at time of colposcopy was lower in aECS compared to pECS (42.4% v 53.1% respectively, p=0.048), but there was no difference in the rate of high-grade endocervical disease at time of excision between cohorts (28.5% v 28.6% respectively, p=0.987). Of patients whose indication for treatment was having only high-grade ECS (no high-grade ectocervical disease), there was no difference in the rate of high-grade endocervical disease at time of excision between cohorts (29.4% v 30.4% respectively, p=0.933).
Conclusions: The presence of high-grade endocervical disease at time of excisional procedure correlates with high-grade ECS at time of colposcopy regardless of whether ECS collection was deemed preferred or acceptable by ASCCP, suggesting the value of universal ECS at time of colposcopy.
Is Risk-Based Colposcopy Practice Recommended by ASCCP Feasible in China?
Jia Peng1, and Hui Bi1. 1Peking University First Hospital
Objective: To evaluate whether colposcopy practice based on screening cytology, HPV testing, and colposcopic impression could improve detection of cervical precancer in Chinese women.
Methods: 1671 women aged 19-83 referred to colposcopy because of abnormal cervical cancer screening results between Jan 2017 and June 2018 in Peking University first Hospital were included in the study. Multiple biopsies, at least 2 and up to 4, were taken during colposcopy. Cervical cytology, HPV genotyping and colposcopy impression were used to establish up to 24 different risk strata. Cervical precancers, which included HSIL or worse that were detected by colposcopy-guided biopsy during the colposcopy visit were analyzed.
Results: The risk of detecting precancer ranged from 0% to 93% across 24 strata based on colposcopy impression (High grade; low grade; normal), cytology (HSIL; LSIL or ASC-US; normal), and HPV genotyping (HPV16/18+; other 12 high-risk+; negative). Women in the highest risk strata had risks of precancer in agreement with immediate treatment. Women with a low previous risk and a completely normal colposcopy impression had a very low risk of cervical precancer (10%) and thus only required surveillance. In other risk strata, multiple biopsies were suggested to improve detection of cervical precancer.
Conclusions: Risk assessment at the colposcopy visit demonstrates more effective and efficient detection of cervical precancer, and is practicable for Chinese women.
Utilization of Colposcopy Standards: Low-Volume Versus High-Volume Colposcopists
Alyson Banegas1, Christine Conageski1, and Jeanelle Sheeder1. 1University of Colorado
Objective: The question of minimal numbers for proficiency has yet to be addressed. Our goal was to utilize the 2017 ASCCP US-Based Colposcopy Standards as a framework to evaluate the performance between low-volume and high-volume colposcopists.
Methods: A retrospective chart review was performed at the University of Colorado Hospital and its affiliations between 2013 and 2016. Colposcopies were identified by ICD-10 code. Demographic data of providers and data that pertained to the Colposcopy Standards were abstracted. Standard statistical analyses were performed using SPSS. The study was approved by COMIRB.
Results: Fifty providers performed 1418 colposcopies. The majority of providers were OB-GYNs (48%). We defined those providers who performed more than 20 procedures as the “high volume colposcopists,” assuming, they performed at least one procedure per month. No providers met 100% of the 13 basic standards. High volume colposcopists documented a median number of 8 items (range 1-11), compared to low volume colposcopists who documented a median 7 items (range 1-11). See Table 1 for statistically significant values. Both high and low volume colposcopists were equally as likely to take biopsies when indicated.
Conclusions: High-volume versus low-volume colposcopists document the performance of colposcopy differently. Those providers who do more colposcopies, document a more thorough exam. Both sets of providers are equally as likely to take biopsies when appropriate.
Correlates of Anal and Cervical Dysplasia in Women Living with HIV, Attending an Anal Dysplasia Clinic in an Urban Area
Patricia Jeudin1, Isabella Rosa Cunha1, Alex Sanchez Covarrubias1, faith doyle1, Karla McGuire1, Daniel M O'Shea1, Carlos Briones Chavez1, Thomas Iglesias1, Johnny Galli1, JoNell Potter1, and Lunthita Duthely1. 1University of Miami
Objective: To evaluate the rates and correlation of anal and cervical dysplasia among women living with HIV (WLWH), attending an anal dysplasia clinic in Miami, FL.
Methods: A retrospective chart review was conducted on WLWH, attending an anal dysplasia clinic in Miami, FL, and had high resolution anoscopy (HRA) with anal cytology and histology of visible lesions. Records from June 2018 and October 2018 were reviewed. Patients with colposcopy results were included. HRA was correlated with colposcopy histology results. Statistical analyses were conducted with SAS-UE-2018.
Results: Of the 254 WLWH who underwent HRA, a total of 70 underwent a colposcopic exam as well. Median age was 53 (IQR 12); majority of the patients were on combination antiretroviral therapy (cART) (99%), had CD4>200/mm3 (89%) and were virologically suppressed (HIV VL <75 copies/mL) (89%). Approximately half of the women (33) had biopsy-proven high grade (HSIL) anal dysplasia, 43% had low grade dysplasia (LSIL). Of those with anal HSIL, only 3 had cervical HSIL, nearly half (18) had cervical LSIL. We found a conditional probability of 0.429 (not significant) of having an abnormal cervical histology result, given a normal anal histology result. Table 1 summarizes the correlation of anal and cervical histologic results.
Conclusions: Currently, there is no consensus regarding the optimal anal cancer screening strategy among individuals living with HIV, particularly among women. In our study, WLWH had high rates of anal HSIL—more frequent than cervical HSIL Further studies are needed to determine the prevalence and risk factors for anal HSIL, and screening strategies for WLWH.
Malassezia Globosa and Malassezia Restricta Are Associated With High-Risk HPV Infections in the Anogenital Tract
Josefina Romaguera1, Filipa Godoy-Vitorino1, Gilmary Ortiz-Morales1, Frances Vázquez-Sánchez1, Anelisse Dominicci-Maura1, and Ana P. Ortiz1. 1University of Puerto Rico
Objective: At the forefront of anogenital microbiome studies are extensive bacterial surveys associating bacteria with disease phenotypes. As fungi account for a relatively small percentage of the human microbiome compared to their bacterial counterpart, much less has been revealed of their diversity. We hypothesized that the anogenital mycobiome changes according to the risk-types of HPV infections.
Methods: We recruited 62 patients coming for gynecology and colposcopy evaluation at the UPR and San Juan City clinics (Puerto Rico) and collected sterile cotton swabs from the anus, introitus and cervix for DNA extraction, HPV typing and mycobiome analyses using ITS2 genes. Patients were divided into low-risk and high-risk HPV group. Data analyses were done in QIIME, MicrobiomeAnalyst and taxonomic assignment was done with BLASTn against NCBI’s nt reference.
Results: Analyses of ~4,003,005 reads and 501 OTUs, revealed that alpha diversity was higher in samples from patients with low-risk cervical HPV infections regardless of the body site. Malassezia globosa and Malassezia restricta were significantly associated with high-risk infections irrespective of the sample site (p-value= 0.016). Candida albicans and Saccharomyces cerevisiae were more prevalent in the anus, but in high-risk and low-risk HPV infections respectively.
Conclusions: We found conserved fungal composition in the anogenital tract however markers of high-risk cervical HPV infections were found to be Malassezia restricta and Malassezia globosa. We are now beginning to grasp the significant role that these fungal variations play in HPV-related pathologies, which could reveal new interventions based on the microbiome to treat or prevent anogenital disease.
Monitoring Burden of Juvenile-Onset Recurrent Respiratory Papillomatosis (JORRP) in the United States
Raiza Amiling1, Laura Stone2, Troy Querec1, Vidisha Singh1, Krystle Love1, Florence George2, Elizabeth Unger1, Craig Derkay2, Lauri Markowitz1, and Elissa Meites1. 1Centers for Disease Control and Prevention, 2Eastern Virginia Medical School
Objective: Juvenile-onset recurrent respiratory papillomatosis (JORRP) is a rare but serious disease most often caused by human papillomavirus (HPV) types 6 or 11, presumably acquired through vertical transmission. In the United States, HPV vaccination is routinely recommended for adolescents; quadrivalent and 9-valent HPV vaccines, introduced in 2006 and 2015, respectively, protect against these and other types. We are monitoring JORRP burden to evaluate potential impact of HPV vaccination.
Methods: Since January 2015, we have enrolled patients aged <18 years with JORRP from U.S. pediatric otolaryngology clinics. Clinical history is abstracted from medical records, patient demographics and maternal characteristics are reported by mothers, and tissue and brush biopsies from papillomas are tested for HPV DNA (37 types). We calculated descriptive statistics including interquartile ranges (IQRs) and report maternal vaccination status.
Results: Through June 2018, 143 prevalent JORRP cases were reported from 20 participating clinics. Median diagnosis age was 4 years (IQR:2–6), birth year ranged from 1997 to 2015. The majority were first-born children (94, 65.7%) and were delivered vaginally (130, 90.9%). Median maternal age at delivery was 22 years (IQR:19–26); none reported HPV vaccination before delivery. Among 117 case-patients with available typing, HPV 6 was detected in 93 (79.5%), HPV 11 in 14 (12.0%); none had only non-vaccine HPV types.
Conclusions: JORRP case-patients were commonly first-born children delivered vaginally by unvaccinated young mothers. HPV types 6 or 11 were detected in nearly all case-patients. Increasing vaccine uptake in the target age group could reduce and even eliminate JORRP.
HPV Screening and Management
A Rapid, High-Volume, See-and-Treat Screening Program using HPV Self-Sampling, Rapid DNA Hybrid, and Digital Colposcopy
Andrew Goldstein1, Lena Goldstein2, Roberta Lipson3, Sarah Bedell4, Jue Wang3, Sarah Stamper5, Gal Brenner6, Gail Goldstein7, Karen O'Keefe8, Casey O'Keefe9, McKenna O'Keefe10, Tierney O'Keefe8, and Anna Zhao3. 1Center for Vulvovaginal Disease, 2Yale, 3United Family Hospitals, 4New Age Women's Health, 5Johns Hopkins, 6MobileODT, 7Annapolis Dermatology Center, 8Bellingham Bay Family Medicine, 9Pacific Northwest Urology Specialists, 10Univeristy of California, Berkeley
Objective: Currently, China does not have national screening guidelines or services for cervical cancer. Only 20.7% of Chinese women reported having ever been screened.
Methods: 500-1000 women/day self-swabbed for high-risk HPV (hrHPV+). Four careHPVTM machines (Qiagen) were run simultaneously to test the specimens. All +hrHPV patients were contacted and digital colposcopy (DC) was performed with the EVA system (MobileODT). Digital images were obtained, and all suspected lesions were biopsied. Suspected CIN1 lesions were treated with cryotherapy and suspected CIN2+ lesions were treated with LEEP.
Results: 3600 women were screened/treated by 4 physicians in 6 days. Mean age of women screened was 50.2 yo. 216 women (6%) were hrHPV+. 168 underwent same-day colposcopy (23 CIN1, 17 CIN2+). 48 followed up at a regional hospital.
Conclusions: Logistical and economic issues make traditional cytology-based cervical cancer screening challenging in China. This study illustrates a cost-effective, rapid, and practical strategy that can be used to screen-and-treat 400+ million Chinese women for cervical cancer. The first component of this strategy is HPV self-sampling. The second component is digital colposcopy on hrHPV+ women with a portable digital colposcope. While the WHO conditionally recommends VIA over DC, the colposcope used in this study may improve sensitivity and specificity over VIA as the high-resolution digital images can be greatly magnified to see changes consistent with CIN2+ lesions. DC adds little to the time and cost of screening, and produces permanent images that can be used for documentation, quality control, and expert consultation via a cloud-based portal.
Application of Internet-Assisted Community Participatory Cervical Cancer Screening Model
Hui Du1, Xinfeng Qu2, Yun Chen1, Juan Min3, Zhaohui Wang3, Chun Wang1, Hongxue Luo4, Xia Huang1, and Ruifang Wu1. 1Peking University Shenzhen Hospital, 2Preventive Onclology International.Inc, 3Pingshan People's Hospital, 4Peking University People's Hospital
Objective: To evaluate the application of an internet-assisted community participatory cervical cancer screening model using self-sampling HPV tests as the primary screening.
Methods: A population based cervical cancer screening program was conducted in suburb Shenzhen, China, from July 2014 to July 2017. Women 25-60 years of age with no pregnancy were eligible for participation. Participants could register for screening by logging in www.mcareu.com with the assistance of local medical providers. Name, ID, telephone number, and address were the required personal identification information for registration. Samplers were shipped to registered participants and the self-collected vaginal specimens were shipped via commercial logistic service to the lab for HPV test. Testing results were reported on the website and phone calls were given to all the positives to instruct them for management. Positive women could visit any hospital for management or be managed locally by a team of doctors assigned by Peking University Shenzhen Hospital every 3 month.
Results: 12,699 women registered for screening at 46 primary medical services. 10,792 samplers were shipped out and 10011 women shipped their samples to the lab. 99.5% (9960/10011) of the personal identification information were correct (retractable). No reverse event was reported. 1.1% specimens were unqualified for tests.
Conclusions: Internet assisted facilitates screening registration through enabling multiple inputs with minimum information errors. Which greatly When self-sampling HPV testing is used as the primary testing, internet assisted community participatory screening model facilitates more participation of community women, therefor potentially benefits the screening coverage.
Comparison of Multiplex PCR for High-Risk HPV With a Commercially Available Test - 8 Years of Clinical Follow-Up
Laurie Griesinger1, Amanda Kitson1, Heather Walline1, Lili Zhao1, and Richard Lieberman1. 1University of Michigan
Objective: As guidelines move towards primary HPV screening for cervical cancer, critical evaluation of the methodology used is pivotal. The Hybrid Capture II (HC2) high-risk HPV (hrHPV) test detects 13 carcinogenic HPV types, with results in the positive, negative, and indeterminate/equivocal ranges. Reported sensitivity and specificity of HC2 in detecting high-grade dysplasia (CIN2+) and hrHPV has varied, and few studies have analyzed the breakdown of HPV types in the different HC2 categories.
Methods: Pap smears with a concurrent HC2 test from a 4-month period in 2010 were selected for the study (n=378). Multiplex PCR for 15 HPV types was performed. Review of clinical records through 2018 was performed for follow-up.
Results: The sensitivity and specificity of HC2 for identifying hrHPV (excluding two types not tested by HC2) were 93% and 64% respectively, and in predicting a subsequent CIN2+ biopsy were 100% and 6%, respectively. Multiplex PCR had decreased sensitivity (82%), but improved specificity (22%), in detecting CIN2+. No HPV type was more strongly associated with high-grade dysplasia than others. A subsequent CIN2+ biopsy was equally likely in women with HC2 results of equivocal or positive.
Conclusions: Though HC2 fails to identify some cases of hrHPV positivity, its ability to identify CIN2+ lesions in this study is even better than PCR. The specificity, however, is low, leading to an increase in potentially unnecessary follow-up. In this cohort, the rates of subsequent CIN2+ were not significantly different in patients with an equivocal/indeterminate versus a positive HC2 result, suggesting that equivocal/indeterminate results should be treated as positive.
Distribution of Human Papillomavirus Genotypes in 533 Women With Cervical Lesions in Chongqing, Southwest China
Dongling Zou1, Dong Wang1, and Qi Zhou1. 1Chongqing Cancer Hospital
Objective: Human papillomavirus (HPV) has been recognized as the most important risk factor for the development of cervical cancer. However, the HPV genotypes distribution in Chongqing, Southwest China was rarely reported, which is important to evaluate therapeutic effect of HPV vaccine and to prevent cervical cancer regionally. The aim of this study is to investigate the distribution of HPV genotypes in women with cervical lesions in Chongqing, Southwest China.
Methods: Samples from 351 normal tissues, 58 low-grade squamous intraepithelial lesions (LSIL), 43 high-grade squamous intraepithelial lesions (HSIL), 73 squamous cell carcinoma and 8 adenocarcinoma samples were collected. HPV genotypes were determined by real-time polymerase chain reaction (Real-Time PCR).
Results: In all, 83/351 normal tissues (23.65%) resulted positive for HPV. The most prevalent HPV genotypes were HPV-16 (24.10%), HPV-52 (16.87%) and HPV-58 (13.25%). Moreover, 46/58 LSIL subjects (79.31%) were detected as HPV positive and the most prevalent HPV genotypes were HPV-58 (23.91%), HPV-52 (23.91%), HPV-51 (19.57%) and HPV-16 (19.57%). While, The most prevalent HPV genotypes in HSIL samples were HPV-16 (51.16%) and HPV-52 (27.91%), respectively. Furthermore, 100% cervical cancer samples were detected as HPV positive and the most prevalent HPV genotypes were HPV-16 (57.53%) in squamous cell carcinoma samples and HPV-18 (75%) in adenocarcinoma samples.
Conclusions: The HPV genotypes distribution in Chongqing is slightly different from the studies in China. However, our results also suggest that the risk of the progression of cervical cancer is closely related to HPV infection and HPV genotypes. The most prevalent HPV genotypes were HPV-16, HPV-58, HPV-52 and HPV-18. Even corroboration from further studies are needed, our work suggests that current commercially available HPV vaccine would be effective and necessary for prevention of cervical cancer in Chongqing, Southwest China.
Evaluation of Cobas HPV and SeqHPV NGS Assays in the Chinese Multi-Center Screening Trial (CHIMUST)
Ruifang Wu1, Hui Du1, Jerome L Belinson2, Chun Wang1, Xinfeng Qu1, Lihui Wei3, Duan Xianzhi4, Liu yan5, Shi bin6, and Wei Zhang1. 1Peking University Shenzhen Hospital, 2Preventive Oncology International Inc., and the Cleveland Clinic, 3Peking University People's Hospital, 4Capital Medical University Beijing Tongren Hospital, 5Fudan University Huanshan Hospital North, 6Hebei Medical University the Second Hospital
Objective: Test two PCR-based high-risk HPV assays with self(S) and direct(D) cervical samples in a population based cervical cancer screening trial.
Methods: 10885 women ages 30-59y women from 15 sites in 7 provinces with no cervical cancer screening for 3 years were eligible. All participating women contributed one self-collected sample(S) and one physician-collected endocervical sample(D). The self-collected sample was first applied to the POI solid media transport card then the brush placed in 2.5ml ThinPrep Solution. All samples were tested with Cobas 4800 and SeqHPV HR-HPV assays. Patients HPV positive (self or direct) were recalled for colposcopy and biopsies.
Results: 10403 women had complete data. The mean age was 43.9 years. 1.3% (141/10403) had CIN2+ and 0.6%(66/10403) had CIN3+. In the liquid specimens the overall HPV infection rates were 10.7% for Cobas and 10.9% for SeqHPV in D-sample, and 13.8% for Cobas and 11.6% for SeqHPV in S-sample, respectively. The sensitivity of Cobas-D, Cobas-S, SeqHPV-D and SeqHPV-S for CIN2+ was 95.04%, 95.04%, 94.33%, and 96.45%, respectively. The specificity of Cobas-D, Cobas-S, SeqHPV-D and SeqHPV-S for CIN2+ was 90.38%, 87.35%, 90.21%, and 89.52% respectively. There were no differences in sensitivity or specificity when applying the two assays to both self and direct collected samples in liquid transport media. (P>0.05).
Conclusions: Cobas 4800 and SeqHPV have similar sensitivities and specificities for diagnosing CIN 2+ in both self and direct collected liquid specimens.
Evaluation of the POI Card for the Detection of HR-HPV in Both Self and Direct Collected Specimens
Hui Du1, Jerome L Belinson2, Chun Wang1, Xinfeng Qu1, and Ruifang Wu1. 1Peking University Shenzhen Hospital, 2Preventive Oncology International Inc., and the Cleveland Clinic
Objective: To evaluate the POI solid media transport card to detect HR-HPV in self-samples compared to physician-collected samples.
Methods: The POI card is a novel solid media transport card with a small size that allows fast, easy DNA extraction from a variety of biological samples. 10885 women between the ages of 30-59 with no screening for 3 years were enrolled. The self-collected sample was first applied to the POI card (POI) then the brush placed in 2.5 ml ThinPrep liquid (SELF). Then a physician collected a direct endocervical specimen into ThinPrep (DOC). Samples were tested with Cobas 4800 and the SeqHPV NGS assay for HR-HPV. Patients positive on any test were recalled for colposcopy and biopsy.
Results: 10403 participants had complete HPV data. The mean age was 43.9years. CIN 2+ rates were 1.3% (141/10403). The overall agreements of HrHPV detection were 95.1%(Cobas SELF vs DOC), 96.2%(Cobas POI vs DOC), 96.9%(Cobas POI vs SELF), 98.1%(SeqHPV SELF vs DOC), 98.2%( SeqHPV POI vs DOC), 99.5%( SeqHPV POI vs SELF). For both HR-HPV assays, the sensitivities were similar for the two self-sample media (SELF, POI). Tested with Cobas, the sensitivity for CIN 2+ of self-POI was 95.04% (95%CI 89.65-97.81), and for the physician-collected sample was 95.04%(95%CI 89.65-98.26), Tested with SeqHPV, the sensitivity for CIN2+ of self-POI was 95.74%(95%CI 90.57-98.26), and for physician -collected sample was 94.33%(95%CI 88.76-97.33).
Conclusions: The POI card when used with self-collected vaginal samples is as accurate as physician-collected direct endocervical samples assayed by two different PCR based HR-HPV tests.
Patient Satisfaction With HPV Self-Sampling as Part of a High-Volume Cervical Cancer Screening Program in Yunnan, China
Andrew Goldstein1, Sarah Stamper2, Benjamin Plafker3, Lena Goldstein4, Roberta Lipson3, Sarah Bedell5, Jue Wang3, Gal Brenner6, Gail Goldstein7, Karen O'Keefe8, Casey O'Keefe9, Tierney O'Keefe8, McKenna O'Keefe10, and Anna Zhao3. 1Center for Vulvovaginal Disease, 2Johns Hopkins, 3United Family Hospitals, 4Yale University, 5New Age Women's Health, 6MobileODT, 7Annapolis Dermatology Center, 8Bellingham Bay Family Medicine, 9Pacific Northwest Urology Specialists, 10University of California, Berkeley
Objective: Rural Yunnan Province is one of the most ethnically, culturally, and religiously diverse regions in China and the majority of its women have received little formal education. It is not known whether women in this region feel comfortable self-swabbing for HPV.
Methods: In a 6-day period, 3600 women were taught the role of HPV in cervical cancer. They were then given self-swabbing instructions. After obtaining their specimens, 600 women were interviewed about their opinions regarding HPV self-testing. The women were of the Han, Yi, Hui, and Dai ethnicities.
Results: 588/600 (98%) of the women surveyed understood the self-sampling instructions and 584/600 (97%) felt comfortable self-sampling. 389/600 (65%) preferred self-sampling to provider testing. Of the women who preferred self-sampling, 80/389 (21%) preferred this method because they felt capable of obtaining the specimens and 79/389 (20%) felt it was the most convenient way to get tested. Of the women who preferred provider obtained specimens, the majority felt that the testing would be more accurate. Only 27/600 (4.5%) felt embarrassed obtaining the samples.
Conclusions: Currently, China does not have national screening services for cervical cancer. It has been estimated that it would take as many at 50,000 health care providers to screen the 400 million Chinese women who have never been screened. However, as HPV self-testing is at least 85% as sensitive as provider testing, this method can significantly decrease the provider burden. Despite the diversity of the women screened, the vast majority felt comfortable self-sampling and the majority preferred this method to provider testing.
Risk of Cervical and Vaginal Dysplasia After Surgery for Vulvar Intraepithelial Neoplasia or Cancer: a 6-year Follow Up
Abigail Zamorano1, Tommy Buchanan1, L. Stewart Massad1, and Lindsay Kuroki1. 1Washington University/Barnes-Jewish Hospital
Objective: To evaluate the utility of cervicovaginal cytology after surgical management for high- grade vulvar intraepithelial neoplasia or cancer (VIN3+) by estimating subsequent risk of high-grade cervical or vaginal intraepithelial neoplasia (CIN2+/VAIN2+) or cancer.
Methods: Women who underwent surgery for high-grade VIN or vulvar cancer between 2006 and 2014 were identified retrospectively. Univariate and multivariate logistic regression analyses were used to identify clinical correlates of later abnormal cytology.
Results: During our 8-year study period, 302 women were followed for a median of 72 months (range 56-237). One hundred (33%) had abnormal cytology: 69 (23%) low-grade, 28 (9%) high-grade, and 2 (0.7%) carcinoma. Prior hysterectomy was not associated with lower risk of intraepithelial neoplasia or cancer [9/99 (9%) VAIN2+ risk vs. 15/203 (7%) CIN2+ risk, p = 0.48], nor with abnormal cytology (OR 0.87, 95% CI 0.5-1.6, p = 0.64). Correlates of high-grade cytology following treatment for VIN/vulvar cancer included non-white race [odds radio (OR) 3.6, 95% confidence interval (CI) 1.7-7.8, p=0.001], immunodeficiency (patients with human immunodeficiency virus or on immunosuppressive medications) (OR 3.4, 95% CI 1.3-8.8, p=0.01), and prior abnormal cytology (OR 3.5, 95% CI 1.6-7.6) on multivariable analysis. (p=0.001) (Table 1).
Conclusions: Women treated surgically for VIN3+/vulvar cancer have a 10% risk of at least high-grade cytology on surveillance screening. Prior hysterectomy does not mitigate the risk. We recommend surveillance cytology screening beginning 6-12 months after treatment, especially in women of non-white race and those with a history of immunosuppression or prior abnormal cytology.
Risk-Based Management Algorithms Incorporating Extended Genotyping - Results From the Onclarity HPV Clinical Trial
Thomas Wright1, and Mark Stoler2. 1Columbia University, 2University of Virginia Health System
Objective: U.S. primary HPV cervical cancer screening guidelines recommend partial genotyping (16/18) with cytology triage for other HPV genotypes. The availability of extended genotyping HPV assays opens the possibility of developing alternative, risk-based, management strategies.
Methods: 29,513 subjects, ≥25 years, had complete baseline data and 5,415 with ≥ASC-US and/or HPV+ results plus a random 5% subset of negative women underwent colposcopy. The BD Onclarity™ HPV Assay was used for extended genotyping. At one year, repeat cytology and HPV testing was done in the colposcopy group with repeat colposcopy following abnormal cytology. Performance of risk-based management strategies was determined using cumulative, 1 year detection of ≥CIN3. Strategies varied as to which genotypes resulted in immediate colposcopy, how women with other genotypes were managed, and whether genotype specific HPV persistence at 1 year was the basis for colposcopy referral.
Results: 194 cumulative cases of ≥CIN3 were detected. A number of risk-based primary screening strategies were identified that resulted in minimal or no reduction in detection of ≥CIN3, while at the same time reducing referral to colposcopy by up to 10%. For example, a strategy incorporating immediate colposcopy for HPV 16 positive women and cytology triage for other genotypes but limiting colposcopy to ASCUS/LSIL cases with HPV 31, 18, 33/58, 52, 45 reduced the colposcopy rate from 9.7% (current strategy with 16/18 genotyping) to 8.7%, without impacting ≥CIN3 detection.
Conclusions: Risk-based extended genotyping strategies may reduce the number of colposcopies needed to detect the same number of ≥CIN3 cases as current screening strategies.
The Relationship Between Cervical Lesions and Type-Specific Viral Load of HPV Reflected by the Ct Value of Cobas4800
Ruifang Wu1, Lvfang Duan1, Hui Du1, Ai MIN Xiao1, Chun Wang1, and Xia Huang1. 1Peking University Shen Zhen Hospital
Objective: To explore the Ct values of Cobas4800 HR-HPV testing in reflecting viral load and its relationship with cervical lesions.
Methods: From August 2016 to September 2017, 7,000 women from Shenzhen, China, were recruited for cervical cancer screening with Cobas4800 HPV testing and cytology co-test. Colpo/biopsies were performed on women who were positive of HPV16/18, positive of HPV types other than 16/18 with cytology(≥ASCUS), or HPV negative but abnormal of cytology (≥LSIL). The Ct values of HPV16, 18 and all combined other types coming from Cobas4800 tests were used as an indicator of viral load to analyze the relationship between type-specific HPV load and the cervical lesions.
Results: The HPV16-Ct values is inversely correlated with the upgrading of the lesions (r=-0.606, P=0.000), significant different to the Ct-value of other types in all levels of cervical lesions (P<0.05). There was no correlation between HPV18-Ct and cervical lesions. The Ct value of other 12 HPV types was statistically difference between CIN1+ and non-CIN (P <0.05), but not difference among CIN2+ (P >0.05).
Conclusions: The Ct-value of HPV16 in Cobas4800 testing as an indicator of virus load obviously correlates with different grades of cervical lesions, therefor can be a reference of cervical lesion existence and an indicator of lesion prognosis.
Value of High Risk HPV Viral Load in Predicting Cervical Lesions and Triaging for ASCUS
Min Wang1, and Fanli Shi1. 1Shengjing Hospital affiliated to China Medical University
Objective: To evaluate the value of high risk HPV (HR-HPV)viral load in predicting cervical lesions and triaging for ASCUS.
Methods: A cross-sectional study came from a multicenter cervical cancer screening program.1183 patients were screened for cervical cancer by HR-HPV genotyping, HR-HPV viral load (HPV16/18 and other 12 HR-HPV type), liquid based cytology test (TCT) detection. The women with abnormal screening results underwent colposcope and histopathological examination further. SPSS22.0 statistical software was used to analyze results.
Results: With the increase of the degree of cervical lesions HPV16/18 viral load showed a trend of increase (P = 0.0002, r = 0.273), there was no rank correlation between other 12 HR-HPV type viral load and the degree of cervical lesions (P > 0.05), the differences of medians of other 12 HR-HPV type viral load had significance in groups (P < 0.05). The diagnostic efficiency (AUC) of HPV 16/18 viral load is 0.676, the best cut-off predict CINII+ lesions is 38.06 pg/ml, the sensitivity and speciality are 44.6% and 86.1% respectively. In patients with ASCUS cytology diagnose, the differences of HPV16/18 and other 12 HR-HPV type viral load in groups of chronic cervicitis and CINI lesions and CINII+ lesions both are significant (P < 0.05). In the groups of chronic cervicitis and CINI lesions and CINII+ lesions, the differences of pregnant times and postcoital bleeding are statistically significant (P < 0.05).
Conclusions: The risk of cervical cancer and precancerous lesions increases with the increasing HR-HPV viral load. HPV16/18 viral load can predict the degree of cervical lesions, HR - HPV viral load is expected to become a clinical triage diagnostic indicator in ASCUS, therefore help to allocate the health resources more effectively.
The Association of Risk Perception, Response Efficacy and HPV Vaccine Decision Making among Adolescent Parents/Guardians
Deanna Teoh1, Tiaj Xiong1, Alicia Myhre1, and Rachel Isaksson Vogel1. 1University of Minnesota
Objective: To evaluate the association of risk perception and response efficacy (belief a behavior will result in the intended outcome) among parents/guardians with adolescent HPV vaccination decisions.
Methods: A cross-sectional survey of adolescent parents/guardians was conducted over a 1-week period at the Minnesota State Fair. To measure risk perception, participants ranked HPV and vaccine risks against diseases/side-effects for which numerical risks were provided. To measure response efficacy, participants indicated (scale 0-100) their ability to prevent HPV infection. Chi-squared and Fisher's exact tests compared risk perception and response efficacy of those who vaccinated/planned to vaccinate against HPV ("vaccinators") with those with no plan to vaccinate ("non-vaccinators").
Results: Of 405 participants, 74% were vaccinators, 13% were non-vaccinators. Non-vaccinators were more likely to under-estimate risk of HPV-related cancers (p<0.05) and over-estimate risk of vaccine-related side-effects (p<0.05); perceived risk of genital warts did not differ by vaccination decison (p=0.5). Non-vaccinators were more confident in their ability to prevent HPV infection without vaccination (mean score 64.1±23.9 vs. 30.0±29.2; p<0.0001), less confident in their ability to prevent HPV infection with vaccination (51.7±30.5 vs. 74.8±25.5; p<0.0001). The primary healthcare information source for non-vaccinators was more likely to be online (16% vs. 4%), and less likely to be a healthcare provider (53% vs. 87%; p<0.0001).
Conclusions: Low HPV risk perception and high response efficacy of non-vaccination HPV prevention behaviors are associated with the decision not to vaccinate. Multi-modal education about HPV prevention is needed, as non-vaccinators are also less likely to get health information from a healthcare provider.
Vaccine Impact on HPV Prevalence among Young Adult Men who have Sex with Men — United States, 2016—2017
Elissa Meites1, Rachel Winer2, Michael Newcomb3, Pamina Gorbach4, Richard Crosby5, Troy Querec1, Jessica Rudd1, John Lin2, Thomas Remble3, Fred Swanson2, Justin Franz3, Janell Moore4, Ethan Cardwell5, Tom Collins5, Brian Mustanski3, Matthew Golden2, Elizabeth Unger1, and Lauri Markowitz1. 1Centers for Disease Control, 2University of Washington, 3Northwestern University, 4University of California at Los Angeles, 5University of Kentucky
Objective: Men who have sex with men (MSM) are at high risk for human papillomavirus (HPV) infection and HPV-related diseases. Since 2011, vaccination has been recommended for U.S. MSM through age 26 years. We evaluated oral and anal prevalence of HPV vaccine types among young adult MSM by vaccination history.
Methods: The Vaccine Impact in Men (VIM) study enrolled MSM and transgender women aged 18–26 years in Seattle, Chicago, and Los Angeles, during 2017. Participants self-reported demographic characteristics, sexual behaviors, and vaccination status, and submitted self-collected anal swab and oral rinse specimens. Type-specific DNA testing was conducted for 37 HPV types by L1-consensus PCR assay. We compared 4vHPV-type prevalence (any HPV6/11/16/18) among vaccinated and unvaccinated participants, and determined prevalence ratio adjusting for age, race/ethnicity, and HIV status (aPR).
Results: Among 693 participants, 71 (10.2%) disclosed being HIV-positive. By self-report, 276 (39.8%) had received any HPV vaccine [140 (50.7%) with 3 doses], 274 (39.5%) were unvaccinated, and 143 (20.6%) were unsure. Median lifetime number of sex partners was 25 among vaccinated and 20 among unvaccinated participants; median age at vaccination (20 years) was older than at first sex (17 years). 4vHPV-type prevalence was 27.0% (187/693) in anal specimens, 2.3% (16/693) in oral specimens, and 27.6% (191/693) in either. Prevalence in either specimen was 21.7% (60/276) in vaccinated compared with 34.7% (95/274) in unvaccinated participants (aPR:0.64, 95% confidence interval:0.49-0.83).
Conclusions: Findings suggest effectiveness of HPV vaccine among MSM. Ongoing monitoring will allow refinements in vaccine effectiveness estimates in this high-risk population.
The Value of PathCINp16INK4a for Early Diagnosis of Cervical Cancer
Xinfeng Qu4, Ruifang Wu1, Hui Du1, Lvfang Duan1, Chun Wang1, Peisha Yan2, Bin Yang3, and Jerome Belinson4. 1Peking University Shenzhen Hospital, 2University of Texas, 3Cleveland Clinic, 4Preventive Oncology International Inc.
Objective: To explore the use of PathCINp16INK4a immuno-stained cytology as the primary screen for cervical cancer prevention.
Methods: 5,504 women were co-tested for cervical cancer screening (Cobas4800 HPV/cytology). Among them, 902 from one site all had cytology slides immuno-stained using PathCINp16INK4a antibody from Shenzhen Senying Bio-tech, Shenzhen, China. All remaining cases positive for HPV16/18, HPV-other plus cytology (≥ASCUS), or Cytology ≥LSIL were also immuno-stained. Colpo/biopsies were performed using the POI biopsy protocol of directed and random biopsies for all women protocol positive. Evaluation of PathCINp16INK4afor screening was based on detection of CIN2+.
Results: 392 (7.1%) were referred by protocol to colposcopy and 51 were diagnosed as CIN2+. The sensitivities to detect CIN2+ of PathCINp16INK4a, Cytology (≥LSIL), and HPV were 90.2%, 76.5%, and 96.1% respectively (P>0.05); and the specificities were 90.9%, 91.9%, and 82.5% respectively (P<0.05). The concordance of the 2 cytopathologists in interpreting cytology and PathCINp16INK4a were 96.3% and 99.4% respectively (P<0.05).
Conclusions: PathCINp16INK4a cytology for primary screening is equally sensitive and specific to traditional cytology but may be less reliant on a cytopathologist’s expertise. PathCINp16INK4a is equally sensitive and more specific compared to primary HPV screening. Therefore, PathCINp16INK4a may be applicable as a single screening technology.
Underserved Populations (Transgender, Homeless, Native American, etc.)
Abnormal Cervical Cancer Screening Results in U.S. Justice-Involved Women Compared to a County Hospital Population
Brittany Manobianco1, Dominique Jodry1, Sophia Yang2, George Birdsong1, Megha Ramaswamy3, and Lisa Flowers1. 1Emory University School of Medicine, 2Santa Clara Valley Medical Center, 3University of Kansas Medical Center
Objective: Justice -involved women (JIW) have disparate rates of cervical cancer in the United States, but little is known about the factors that drive that risk. Our objective was to investigate the cervical cancer screening results of JIW compared to a county hospital sample in the Southeastern United States.
Methods: A retrospective medical records review was conducted of the Pap screening results of women with a history of correctional custody at a large county hospital between 2014 and 2018 (N = 191). An age-matched cohort of subjects with no recorded correctional involvement from the same hospital was used as a control population (N = 394).
Results: Rates of abnormal Paps were high in both JIW and controls (40.83% and 17.97%, respectively). Compared to controls, JIW were at increased risk of experiencing high-grade abnormalities on Pap smear (OR 4.25, 95% CI 1.94-9.36, P<0.005). Risk factors for abnormal Pap results, including HIV (OR 3.19; 95% CI 1.45-7.02, P<.005), and smoking (OR 2.29, 95% CI 1.26-4.16, P <.05), were more prevalent in JIW. Inadequate follow-up of abnormal Paps was high in both groups, but was not significantly different between JIW vs. controls (65.2% vs. 60.5% respectively; OR 0.98, 95% CI 0.52-1.90, P=.97).
Conclusions: JIW represent a high-risk subpopulation for high-grade abnormalities on Pap smear, reflecting a confluence of risk factors and gaps in the pipeline from detection to treatment. These abnormalities parallel the disparate rates of cervical cancer in these women, underscoring the need for careful attention to screening programs and follow-up interventions.
Assessment of Colposcopy Adherence in a Low Income Patient Population
Katherine Hyon1, Teresa Wilson1, Kathleen Groesch1, Allen Ghareeb1, Paula Diaz-Sylvester1, Lindsay Kuroki2, Kristin Delfino1, L. Stewart Massad2, Yahia Zeino1, Assaad Semaan1, and Anne Martin1. 1Southern Illinois University School of Medicine, 2Washington University
Objective: To identify unmet basic needs amongst women with an abnormal cervical cancer screening test who are referred to colposcopy and to determine if unmet needs are a factor in adherence to follow-up.
Methods: A prospective, pilot study was conducted from 9/2017-10/2018 at an academic colposcopy center serving primarily rural patients on federally or state-funded insurance plans. In Phase 1, basic needs (safety, housing, family, financial, transportation, childcare) were assessed via a phone survey prior to their colposcopy appointment. In Phase 2, patients screening positive for ≥ 1 basic need(s) were contacted by a life navigator with resources to help mitigate these needs.
Results: To date, 44 patients were enrolled (Phase 1=25; Phase 2=19), the majority of our subjects were Caucasian (64%), mean age was 36 (range: 21-72 years), 64% had ≥1 child at home and 55% reported at least one basic need. There was no significant difference in attendance of colposcopy appointments between Phase 1 and Phase 2 (64% vs. 68%, respectively). However, prior to initiation of this study, the percentage of patients who presented to their initial colposcopy appointment in our clinic was 52%. Yet, during the timeframe of this study, adherence to one’s initial colposcopy appointment increased to 61%, although this increase was not statistically significant.
Conclusions: Patient’s adherence to their initial colposcopy appointment showed little to no improvement upon initiation of the study or the introduction of the intervention for unmet basic needs in our largely rural population. Assessment of other means to increase awareness/adherence are being investigated.
Prevalence of Vulvodynia in Transmasculine Individuals
Lauren Abern1, Karla Maguire2, Julia Seay2, and Jose A. "Tony" Carugno2. 1Harvard Vanguard Medical Associates, 2University of Miami
Objective: The prevalence of vulvodynia has not been well studied in individuals on the transmasculine spectrum. The goal of this study is to determine the prevalence of vulvodynia and its relationship to hormone therapy in transmasculine individuals.
Methods: Transmasculine individuals ages 18-64 participated in an online survey that included questions about demographics, medical history, hormone usage, and vulvodynia symptoms. Descriptive statistics were used to analyze the data. This is an update to preliminary data presented at ASCCP 2018.
Results: 781 individuals completed the survey. The mean age was 27 (SD 8.6). The majority was white (663, 85%). 344 (44%) had minimum of a college degree. 530 had private insurance (68%), and 413 (53%) made less than $24,000/year. 468 (60%) were on hormone therapy. 455 (60%) were sexually active. 68 (10%) of individuals have experienced itching, burning or knife-like pain in the vulvo-vaginal area for more than 3 months. For those on testosterone, 42 (9%) had symptoms of vulvodynia. Of all participants experiencing symptoms, 42/68 (62%) were on systemic testosterone.
Conclusions: The prevalence of vulvodynia in transmasculine individuals appears to be similar to that of the general population. Only a small percentage of individuals on testosterone have symptoms consistent with vulvodynia. Testosterone use does not seem to increase the risk. However, out of those individuals experiencing symptoms, a larger percent were on testosterone. Further research is warranted to clarify the impact of vulvodynia in transmasculine individuals.
Vaginal Intraepithelial Lesions and Neoplasia
Alterations of Vaginal Microbiota and Discordance of Pap Test and Cervical Biopsy Correlations
Rachel Jester1, Iris Martin1, Kathryn Lindsey1, Jack Yang1, and Olga Chajewski1. 1Medical University of SC
Objective: Alterations in vaginal microbiota, including bacterial vaginosis (BV), candida species, and trichomonas vaginalis have been shown to promote the persistence of HPV infection. On cervical cytology reactive changes must be differentiated from dysplasia, and identifiable causes of reaction (i.e. inflammation and infection) may increase the threshold for atypia. We hypothesized that this increased threshold in paps with BV, candida or trichomonas, along with the potential for increased persistence of HPV in women with these conditions could result in more significant discordant diagnoses between the Pap test and cervical biopsy specimens.
Methods: Alterations in vaginal microbiota, including bacterial vaginosis (BV), candida species, and trichomonas vaginalis have been shown to promote the persistence of HPV infection. On cervical cytology reactive changes must be differentiated from dysplasia, and identifiable causes of reaction (i.e. inflammation and infection) may increase the threshold for atypia. We hypothesized that this increased threshold in paps with BV, candida or trichomonas, along with the potential for increased persistence of HPV in women with these conditions could result in more significant discordant diagnoses between the Pap test and cervical biopsy specimens.
Results: Of 1285 cytology-histology correlations, there were 416 (32.4%) discordances. Excluding ASCUS cytology, 157 (12.2%) cases with significant discordance remained. Of those, 29 (18.5%), 10 (6.3%), and 7 (4.5%) had evidence of BV, candida, and trichomonas respectively on cytology. Overall, there was not a statistically significant difference in cytological-histologic discordance based on the presence of BV (p-value 0.207) or candida (p-value 0.074); however, the presence of trichomonas did lead to a statistically significant discordance rate (p=0.008).
Conclusions: Additional study of these common cervical infections and their relationship to HPV persistence is warranted. Furthermore, the presence of trichomonas may be a pitfall in the diagnosis of squamous dysplasia.
Vulvar Intraepithelial Lesions and Neoplasia
Effect of An Alternative Strategy For Measuring Depth of Invasion in Early Vulvar Squamous Cell Carcinoma
Stephanie Skala1, and Richard Lieberman1. 1University of Michigan
Objective: Measurement of invasion of vulvar squamous cell carcinoma (SCC) is sometimes challenging. Poor interobserver reproducibilty has been documented regarding presence of and depth of invasion. Tumors with ≥1 mm invasion trigger sentinel biopsy and/or groin lymph node dissection. A previous study of stage IB vulvar SCC (n=69) demonstrated no change in outcome when depth was measured to the nearest lowest dysplastic rete peg rather than the American Joint Committee on Cancer-recommended nearest highest dermal papilla. We report our institutional experience with early vulvar SCC, including clinical follow-up and different methods for measuring depth of invasion.
Methods: 75 cases of pT1 invasive vulvar SCC resected from 1998-2018 with lymph node sampling were identified from the surgical pathology database. Pathologic staging data and clinical follow-up were recorded. Each invasive SCC was measured from nearest highest dermal papilla (conventional method) and from the adjacent dysplastic rete peg (alternative method).
Results: Fifteen cases of invasive SCC were reclassified as pT1a when measured using the alternative method (conventional pT1b). All reclassified cases were pN0 and lacked lymphovascular or perineural invasion. 3/15 patients (20%) reclassified as pT1a had recurrent invasive carcinoma, compared to 15/58 (26%) that remained pT1b. Of the 58 cases that remained pT1b, 5 (9%) had lymph node involvement, 12 (21%) had lymphovascular invasion, and 8 (14%) had perineural invasion. 2 cases staged as pT1a regardless of method were pN0 without lymphovascular invasion, perineural invasion, or recurrent invasive carcinoma.
Conclusions: In 15 cases, the depth of invasion was <1 mm by the alternative method. None of these cases demonstrated nodal involvement or lymphovascular or perineural invasion. These results suggest that this alternative method could safely allow some conventional stage IB vulvar SCC patients to avoid groin surgery, reducing treatment-related morbidity. It remains unclear which factors in early vulvar SCC are most associated with nodal metastasis and aggressive clinical course.
Morphologic Assessment and Clinical Follow-Up of Differentiated Vulvar Intraepithelial Neoplasia and Lichen Sclerosus
Grace Wang1, Stephanie Skala1, Shula Schechter1, May Chan1, and Richard Lieberman1. 1University of Michigan
Objective: Differentiated vulvar intraepithelial neoplasia (dVIN) is a human papillomavirus-independent lesion with potential for rapid progression to invasive squamous cell carcinoma (SCC). Because dVIN usually does not show diffuse architectural or cytologic atypia, it is often underdiagnosed in cases showing concomitant lichen sclerosus (LS), or misdiagnosed as lichen simplex chronicus (LSC). We assess morphologic features of a large cohort of dVIN and LS cases in order to identify useful differentiating features between dVIN and LS without dVIN which would not require vulvar resection.
Methods: Biopsies diagnosed as dVIN and LS (2011-2018) were retrieved. The included LS cases lacked prior, concurrent, or subsequent dVIN or SCC based on follow-up ≥2 years; a subset of these cases had superimposed features of LSC. Morphologic features were compared using Fisher’s exact tests. In the dVIN group, features associated with concurrent or subsequent progression to invasive SCC were identified.
Results: The majority (93%) of dVIN cases had associated LS. Differentiating features between dVIN and LS without dVIN are listed below: Presence of moderate-marked lichenoid inflammation did not show significant association with either group. 37 (54%) cases of dVIN were associated with SCC. Keratin pearl formation was the only discriminator between dVIN associated with SCC (17/37, 46%) and dVIN without progression (7/32, 22%) (p=0.045).
Conclusions: The strongest discriminators between dVIN and LS without dVIN were dyskeratosis, paradoxical maturation, nuclear enlargement, large nucleoli, acanthosis, elongated and anastomosing rete ridges, basal pleomorphism, diffuse parakeratosis, keratin pearl formation, basal layer disarray, and mitotic figures above the basal layer. Keratin pearl formation was more frequently seen in dVIN associated with SCC, and may be viewed as an alarming sign of early progression.
Risk of HPV16-Related Vulvar Neoplasia Associated With HLA Class I/II Alleles
Margaret Madeleine1, Mollie McWhorter2, David Doody1, Denise Galloway1, and Steve Schwartz1. 1Fred Hutchinson Cancer Research, 2University of Washington
Objective: Strong associations have been confirmed between specific human leukocyte antigen (HLA) alleles and risk of cervical cancer. We found that HLA alleles and risk of HPV16 and HPV18-specific cervical cancer were not similar. Here we compare risk of HPV16-specific cervical and vulvar intraepithelial neoplasia (VIN3)/vulvar cancer (VC) to a population control group.
Methods: Participants were 18-74 year old women residing in Seattle, WA. HLA alleles were assayed by high-resolution molecular methods to identify Class I (A, B, C) and Class II (DRB1, DQB1) polymorphisms. HPV DNA from tumor blocks was typed using MY09/11 L1 primers and HPV16/18 E6 type-specific primers. We restricted this analysis to 238 women with HPV16+ VIN3/VC, 238 HPV16+ cervical cancer, and 637 population-based controls. We estimated odds ratios using polytomous regression models.
Results: Risk of VIN3/VC was increased for seven (A*2902, B*0702, B*4403, B*5501, C*0602, C*0702, and C*1601) and decreased for two (B*1501 and B*3501) class I alleles. Likewise, there were elevated risks of VIN3/VC associated with six (DRB1*0701, DRB1*1101, DRB1*1501, DQB1*0202, DQB1*0301, DQB1*0602) and decreased risks associated with six (DRB1*0101, DRB1*0301, DRB1*1301, DQB1*0201, DQB1*0501, DQB1*0603) class II alleles. Only B*5501 and DRB1*1101 were significantly different for vulvar vs. cervical cancer (p value for equal ORs: 0.046 and 0.028, respectively).
Conclusions: These data strongly suggest that associations between HLA alleles and HPV16+ cancers of the vulva and cervix are largely similar, supporting an underlying uniformity in HPV16 antigen recognition across anatomic sites. T helper and cytotoxic T cells are important cofactors with HPV for both vulvar and cervical cancers. These results may be useful for development of therapeutic vaccines.