The ATHENA HPV study underrepresents other high-risk HPV genotypes when compared with a diverse New York City population

Transcription

1 Accepted: 6 March 2017 DOI: /cyt ORIGINAL ARTICLE The ATHENA HPV study underrepresents other high-risk HPV genotypes when compared with a diverse New York City population G. Ramos Rivera a S. N. Khader a S. Lajara K. Schlesinger D. Y. Goldstein R. C. Naeem M. J. Suhrland A. S. Fox Pathology Department, Montefiore Medical Center, Bronx, NY, USA Correspondence Gloria Ramos Rivera, Montefiore Medical Center, Bronx, NY, USA. s: Gloria.ramosriv@gmail.com and gramosr@montefiore.org Objective: Persistent infection with oncogenic high risk HPV (hrhpv) types causes virtually all cases of cervical cancer. HPV 16 and 18 have been targeted for individual genotyping and vaccination because of their presence in 71% of invasive cervical cancers worldwide. Montefiore Medical Center, Bronx, New York serves a population known for ethnic and racial diversity. Given this diversity it is possible that HPV genotypes not individually detected by current testing are causing significant disease. Methods: We conducted a retrospective analysis of liquid based cervicovaginal cytology and Cobas HPV results reported between October 5, 2015 and March 30, This included samples from patients aged (average age 42), with racial distribution including: African-American 32.4%, Other (includes denied, unknown, mixed, Hispanic) 52.1%, Caucasian 14.5%, Asian 0.7%, American Indian/Alaskan Native 0.3%. In all, samples (72.9%) were submitted for clinically requested COBAS 4800 HPV testing, which separately reports HPV 16, 18 and a pool of 12 other hrhpv. Results: A total of 3180 (21.5%) tested hrhpv positive. The percentage of patients with cytologic diagnosis of HSIL (high-grade squamous intraepithelial lesion) that were positive only for HPV 16 was 19.4% vs 1.8% for all cytologic diagnoses. However, only one of the HSIL cases was HPV 18 positive along with other hrhpv (OHR). Surprisingly, a majority (64.5%) was positive for only OHR. Conclusions: Further evaluation is needed to determine if this pool of other hrhpv includes individual genotypes that in our population carry a higher risk of persistence and progression to cancer. KEYWORDS HPV, Papanicolaou test, papillomaviridae, uterine cervical neoplasms 1 INTRODUCTION Among the several hrhpv genotypes associated with the development of cervical cancer, HPV 16 and 18 have been shown to be of a These authors contributed equally to this work. primary importance. In two studies, which included a meta-analysis of 1 million women, HPV16 and 18 were the most common hrhpv types detected in patients with both normal and abnormal cytological findings. 1,2 A prospective study has shown a significantly higher cumulative incidence of HSIL in women infected with HPV 16 and/ or Retrospective analyses of cervical cancer biopsies showed Cytopathology. 2017;28: wileyonlinelibrary.com/journal/cyt 2017 John Wiley & Sons Ltd 413

2 414 RAMOS RIVERA ET AL. that HPV 16 and 18 were associated with up to 70% of invasive cancer cases 2 and that HPV 16 was associated with 52.4% of invasive cancer cases. 4 As a result of the prevalence and increased risk of progression to cancer, HPV 16 and 18 were the first genotypes to be targeted by HPV vaccines, Cervarix (GlaxoSmithKline, Brentford, London, UK) and Gardasil (Merck & Co, Kenilworth, NJ, USA), the latter also covering the low-risk HPV types 6 and 11, the aetiologic agents of warts and mucosal papillomatous lesions. In terms of detection, HPV 16 and 18 were also the first to be separately identified in a clinical assay for hrhpv. The Cobas HPV test (Roche Diagnostics, Basel, Switzerland) approved by the Food and Drug Administration (FDA) in 2011, provides individual detection of HPV 16 and HPV 18 from a pool of 12 other hrhpv types. 5 HPV 16 and 18 genotype testing are now incorporated into current American Society for Colposcopy and Cervical Pathology (ASCCP) guidelines for cervical cancer screening. 6,7 While HPV types 16 and 18 have already been a primary clinical concern in HPV testing for some time, ever-evolving guidelines and advancing diagnostic technologies are constantly adding to our knowledge of HPV leading to an understanding broader than simply HPV 16 and 18 alone. Several studies have demonstrated regional and population differences in the distribution of hrhpv genotypes. A study in Ghana found that the most common hrhpv genotypes were HPV 18, 59 and In a study of Southeast Asian women, only one patient tested positive for HPV 16, and the most common hrhpv genotypes were HPV 51, 52, 56, and At Mount Sinai Hospital, New York, HPV 59, 16 and 51 were the most common hrhpv types in women with abnormal cervical cytology. 10 Recognition and understanding of regional variation are important to ensure appropriate molecular detection. Although the latest FDA approved vaccine, Gardasil 9, includes additional hrhpv types: 31, 33, 45, 52 and 58, in addition to 6, 11, 16 and 18, it does not cover 51, 56, or Montefiore Medical Center (MMC), Bronx, New York, serves a population of over 1.4 million individuals. These patients are some of the most underserved in the City of New York as well as in the entire region. 12 The Bronx is known for its ethnic and racial diversity; almost one-third of its residents are immigrants, and over 110 languages are spoken by MMC patients. Bronx census data from 2015 reveals race distribution as, 45.4% White, 43.4% African Americans, 4.5% Asian, 2.9% American Indian and Alaska native, 0.4% Native Hawaiian/Pacific Islander and 3.4% others. 12 It is important to note that 55.1% of the population identified as Hispanics and are distributed among other race categories, whereas 10.1% identified as only White (not Hispanic or latino). 12 In 2015, 4% (1678/42081) of the Pap tests performed at MMC were diagnosed as low-grade squamous intraepithelial lesions (LSIL), and 0.6% (234/42081) were diagnosed as HSIL, indicating a significantly higher risk of developing cervical cancer, when compared with the overall United States population. Based on the Laboratory Percentile-Reporting Rate statistical records from the College of American Pathologists (CAP), our rates for LSIL falls between the 75th and 90th percentile and HSIL is at the 75th percentile. 13 In recognition of the population diversity and increased risk of disease, part of the Pap quality control programme involves senior cytotechnologist rescreening of all negative Paps from patients with a high-risk history (history of abnormal Paps or gynecological histological abnormalities, patients with multiple sex partners, history of sexually transmitted infections or history consistent with high-risk sexual behaviour) This high-risk group comprises 20% of all Paps evaluated. In October 2015, MMC Division of Cytology transitioned their testing platform from Hybrid Capture II (HCII) hrhpv test (Digene Corporation, Gaithersburg, MD, USA) to the Cobas HPV test, to separately identify the presence of hrhpv 16 and 18 types in addition to the other 12 types. This new system at MMC allowed for the first time to consider and examine whether the diverse population served by MMC in the Bronx might present with different distributions of high-risk viral genotypes. Additionally, utilising the Cobas assay allowed for a direct comparison of data presented in the ATHENA study, 14 with its more homogenous populations, to the broad population diversity served by our medical centre. The ATHENA study, a large clinical trial to assess the ability of advanced HPV assays to triage women with abnormal findings on cytopathological examination, had an ancillary objective of looking at the ability of advanced HPV assays being utilised as primary firstline test for cervical cancer screening. 14 ATHENA prospectively analysed data from patients from 61 clinical centres across the United States in which the Cobas HPV test was performed and correlated with cytology and subsequent histology METHODS A retrospective collection of cytology results, along with the hrhpv results were retrieved and analysed on female patients of all ages seen from 5 October 2015 through to 30 March Data collection was performed by querying the in-house pathology information system (SoftPathologyDx ) for all patients having complete data regarding Pap testing. Patient s cytology cases are linked to histology cases by the patient s medical record number in the electronic medical record system. In the case of all HSILs cases diagnosed on Pap tests, an additional query was performed to find the histological diagnosis, if available. Sample collection and cytopathological preparations were performed utilising liquid-based cytology (SurePathâ Pap Test) and a slide preparation system according to manufacturers recommendations. Slides produced underwent screening, quality control rescreening as needed, and cytopathologist review as needed. Negative Pap test cytology diagnoses are released by cytotechnologists, and all abnormal Pap diagnoses (atypical or greater) are reviewed by cytopathologists. Patients were tested for HPV as follows: atypical cells of undetermined significance (ASC-US) cytological diagnosis at any age, patients over 30 years old with a negative cytological diagnosis, and previous positive HPV result at any age. HPV testing was performed on the Cobas HPV test as per New York State Department of Health guidelines for FDA

3 RAMOS RIVERA ET AL. 415 TABLE 1 Pap patients age, racial and diagnosis distribution, MMC vs ATHENA 14 TABLE 2 ATHENA 14 Pap diagnosis distribution by age group, MMC vs MMC ATHENA No. of Pts. Percentage No. of Pts. Percentage Total % % Age distribution % a a % % % % % % % % Race distribution African American % % American Indian/ Alaskan native % % Asian % % Caucasian % % Native Hawaiian/ Pacific Islander 6 0.0% % Other/missing b % % Pap diagnosis distribution NILM % % ASC-US % % ASC-H % % LSIL % % HSIL % % SCC 1 0.0% 2 0.0% AGC % % AGC, favour neoplastic % 5 0.0% EMC> % - - Unsatisfactory % - - Invalid c % AGC, atypical glandular cells; EMC>45, endometrial cells in women over 45 years of age. a Age range not examined by the ATHENA study. b Other category includes Hispanics, which is reflective of the 52.1% from the overall MMC population. c Invalid was the term in the ATHENA study used for endometrial cells older than 40 years of age, inadequate cellularity, or no sample tested. modified laboratory developed test. 15 This is a qualitative multiplex assay that detects HPV genotypes 16, 18, and other high-risk genotypes: 31, 33, 35, 39, 45, 51, 52, 56, 58, 59, 66 and 68. While this system allows individual detection of HPV genotypes 16 and 18 it is of importance to note that all additional genotypes utilise a single detection channel and cannot be differentiated. Internal cellularity controls utilise an amplified portion of the beta globin gene. The race distribution at MMC is 32.4% (n=6638) African American, 0.3% (n=69) American Indian/ Alaskan native, 0.7% (n=131) ATHENA Age group Cytological diagnosis MMC patients (n=20 483) study patients (n=46 887) <30 ASCUS and LSIL % % ASC-H andhsil % % Total % % 30 ASCUS and LSIL % % ASC-H and HSIL % % Total % % Asian, 14.5% (n=2965) Caucasian, 0.0% (n=6) Native Hawaiian/Pacific Islander and 52.1% (n=10 674) other race. Data from the ATHENA study were culled from publically available databases and literature. 3 RESULTS At the MMC Division of Cytology, SurePath Pap tests were received from 5 October 2015, through to 30 March 2016, of patients ranging from 16 to 95 years (mean age 42 years) as shown in Table 1. Compared with the patient population in the ATHENA study, the age distributions differ somewhat in that the percentage of patients with 50 years or older is greater, and the percentage of patients with years is lower in our study population. Only 14.5% of the patients identified themselves as Caucasian at MMC, vs 83.0% of the patients in the ATHENA study. Half of the population identified themselves as Hispanic (included in the Other group) and about one-third as African American demonstrating that the MMC population is much more racially diverse compared with the ATHENA study. As shown in Table 1, the MMC population has an increased rate of gynecologic abnormality compared to ATHENA. In the ATHENA study, 90.9% of the Pap tests were diagnosed as negative for intraepithelial lesion or malignancy (NILM); in the MMC population, the NILM percentage is 83.9%. The percentages of Pap tests in all abnormal diagnostic categories are higher in our study population. Patients with a cytology diagnosis of atypical cells of undetermined significance, cannot exclude HSIL (ASC-H) or HSIL is higher in patients 30 years and older compared with the younger population. Furthermore, it is slightly higher in MMC patients (0.5%) when compared with the ATHENA study (0.3%) as seen in Table 2. Of the study patients, a total of (72.9%) underwent Cobas HPV testing. Of those tested, 3180 (21.5%) samples were positive for hrhpv. Samples tested included 1770 (12%) with a cytological diagnosis of ASC-US (reflex testing for ASCUS at MMC); the remaining 88% represented hrhpv co-testing, most commonly in patients older than 30 years with normal cytology.

4 416 RAMOS RIVERA ET AL. TABLE 3 hrhpv testing results by patient age group, MMC vs ATHENA 14 MMC patients ATHENA study patients HPV 16 HPV 18 HPV 16 positive HPV 18 Age Patient tested OHR positive positive positive Patient tested OHR positive positive < % % % 9 1.8% % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % 7 0.2% Total % % % % % % % % TABLE 4 HPV results by cytology diagnosis categories a HPV genotype EMC>45 NILM ASC-US LSIL ASC-H HSIL Total % % % % % % % % 16 and OHR 1 0.3% % % % 5 8.2% 3 9.7% % 16 and <0.1% 4 0.2% % % 16, 18 and OHR % 9 0.5% 3 1.0% % % % % 8 2.7% 2 3.3% % 18 and OHR 1 0.3% % % % 2 3.3% 1 3.2% % Total 16 and/or 18 with or without OHR 6 2.0% % % % % % % OHR only % % % % % % % Negative % % % % % 1 3.2% % Grand total % % % % % % % a Failed and invalid HPV results, and glandular abnormalities were excluded (149 cases of ). Table 3 shows the distribution of positive COBAS HPV test results for HPV 16 and 18 (with or without other genotypes), and OHR genotypes only by age group. Similar to the ATHENA study, OHR genotypes are the most common type in all age groups, with the decreasing prevalence with age from 45.8% in patients under 25, to 12.8% in patients over 60 years. In all age groups, the percentage of MMC patients who test positive for OHR genotypes is higher than in the ATHENA study. The difference is most marked in patients under 30 years, likely because testing within this group is performed on samples with abnormal cytology at MMC, whereas all patients in the ATHENA study underwent HPV testing. With regard to HPV 16 and 18, the percentage of MMC patients who test positive for these HPV genotypes is slightly lower compared with ATHENA in the group <30 years old. In patients over 30 years old, the percentage of MMC patients who test positive for HPV 16 and 18 is higher than in ATHENA study, reflecting the fact that this is a high-risk population. The correlation of cytology diagnoses and Cobas HPV test results is shown in Table 4. The percentages shown are the percentage within each diagnostic category with a given HPV result. As expected, the percentage of hrhpv positive results, both overall and for individual genotypes, increases with increasing level of cytologic abnormality. HPV 16 only, for example, is present in 1.7% of cases diagnosed as NILM and increases to 19.4% in cases diagnosed as HSIL. OHR genotypes only are present in 13.7% of NILM and 64.5% (20 of 31 cases) of HSIL. Moreover, OHR genotypes were the only type identified in the one case that proved to show invasive cancer on biopsy. Of the 20 OHR only positive cases, 18 have had histological follow-up thus far, of which 12 (60%) showed HSIL or worse, including one squamous cell carcinoma (SCC). Of the HPV 16 positive cases, three out of six (50%) with histologic follow-up showed HSIL diagnosis on cytology. The numbers are small, and cyto-histological correlation may be limited by biopsy sampling error, but, based on limited data, there appears to be no marked difference between the HPV 16 and OHR genotypes in histological follow-up of cases showing HSIL on cytology. A majority of the HSIL patients had a history of prior abnormal cytology (ASC-US or greater) and or histology, which not infrequently, extended over years. Twenty patients had prior abnormality: 12 of them (60%) had OHR genotype, three had HPV 16, two had HPV 16 and OHR, and one had HPV 18 and OHR. One patient had a prior negative history and the remaining patients either had no prior Pap tests performed at MMC or had a gap in the follow-up of

5 RAMOS RIVERA ET AL. 417 at least 5 years preceding the current Pap. This appears to be true regardless of HPV type. 4 CONCLUSIONS The ATHENA trial provided early data which ultimately led to the FDA approval of the Roche COBAS HPV assay. It additionally demonstrated genotypic data from a large multicentre cohort from the United States. While the undertaking of the trial was of enormous proportions, analysis of the study populations revealed that concerns regarding the homogeneity of the population tested may have an impact on their results. With respect to race 83% of the ATHENA population identified as Caucasian. This is a remarkable contrast to the population at our institution where only a small proportion of the patient population is categorised as such. There is strong data suggesting ethnic and regional variations in HPV infection rates as well as genotypes. The population in our study mirrors more directly the population of the county of the Bronx. This diverse population presents with its own distinct challenges in providing health care to several underserved communities. Our results indicate that in nearly every age group the positivity of OHR HPV genotypes is greater in our diverse population vs the ATHENA group; however, the proportion of HPV 16 and 18 are not significantly different. This may indicate that while the burden of HPV 16 and 18 are similar in both populations, our more diverse population may demonstrate a wider variety of viral genotypes. Additionally, no conclusion can be drawn regarding viral genotypes not assayed by the COBAS HPV test. Given the geographic, ethnic and socioeconomic variation of HPV types, it is not without merit to suggest that this diverse population may harbour additional high-risk types, but are simply not detected by the assay. This consideration may very well become more of a concern as vaccination programmes directed toward HPV 16 and 18, may shift the disease burden away from these viral genotypes. The MMC study population is very different from the ATHENA study group. The MMC population closely resembled the overall Bronx population, where Hispanics and African Americans compose more than 80% of the population. In the ATHENA population, 83% were Caucasians. This substantial difference in population diversity is reflected in the contrasting results between both groups. The data from our study population revealed that OHR genotypes are the most common present in all age groups and are associated with the majority of HSILs. It also was associated with the only invasive cancer case in the study, which seems contrary to prior reports that HPV 16 and 18 account for a majority of invasive cancers. Interestingly, the prevalence of hrhpv within every age category is higher in our study population compared with the ATHENA study population. This unexpected finding highlights that the current ASCCP guidelines do not take into consideration the unfavourable outcomes OHR genotypes have and their risk of developing highgrade squamous lesions. The present results suggest that OHR genotypes should be treated the same way as HPV genotypes 16 and 18. Based on these results, we recommend expanding hrhpv testing for cervical cancer screening to include OHR genotypes and warrant the same care and follow-up currently applied for HPV genotypes 16 and 18. Given the racial and ethnic diversity of the Bronx population, and in light of prior studies demonstrating regional differences in prevalent genotypes, further genotyping of MMC patients with abnormal cervical cytology may reveal additional genotypes that should be targeted for developing vaccines and diagnostic technology. Increased attention may need to be placed on prevention through extending coverage of HPV vaccines to include other hrhpv genotypes as we begin to understand the nuances of HPV genotype distributions among varying populations. While this study identifies a greater prevalence of other high-risk HPV types in a diverse population, our data raise additional questions that require further studies. What other genotypes are causative in disease progression other than the ones included in the diagnosis platform? What is the natural history of the OHR genotypes? What is the role of co-infection among OHR genotypes and other sexually transmitted infections? These questions will require additional study as we begin to expand our understanding of HPVrelated cervical disease. Prospective studies within the diverse population in the Bronx will further allow for elucidation of some of these questions. REFERENCES 1. Bruni L, Diaz M, Castellsague X, Ferrer E, Bosch FX, de Sanjose S. Cervical human papillomavirus prevalence in 5 continents: meta-analysis of 1 million women with normal cytological findings. J Infect Dis. 2010;202: Clifford G, Franceschi S, Diaz M, Mu~noz N, Villa LL. Chapter 3: HPV type-distribution in women with and without cervical neoplastic diseases. Vaccine. 2006;24(Suppl 3): Khan MJ, Castle PE, Lorincz AT, et al. The elevated 10-year risk of cervical precancer and cancer in women with human Papillomavirus (HPV) type 16 or 18 and the possible utility of type-specific HPV testing in clinical practice. J Natl Cancer Inst. 2005;97: Bao YP, Li N, Smith JS, Qiao YL. Human papillomavirus type distribution in women from Asia: a meta-analysis. Int J Gynecol Cancer. 2008;18: Rao A, Young S, Erlich H, et al. Development and characterization of the cobas human papillomavirus test. J Clin Microbiol May;51: Saslow D, Solomon D, Lawson HW, et al. American Cancer Society, American Society for Colposcopy and Cervical Pathology, and American Society for Clinical Pathology Screening Guidelines for the Prevention and Early Detection of Cervical Cancer. CA Cancer J Clin. 2012;62: Massad LS, Einstein MH, Huh WK, et al updated consensus guidelines for the management of abnormal cervical cancer screening tests and cancer precursors. J Low Genit Tract Dis. 2013;17(Suppl 1): Awua AK, Sackey ST, Osei YD, Asmah RH, Wiredu EK. Prevalence of human papillomavirus genotypes among women with cervical cancer in Ghana. Infect Agent Cancer. 2016;11:4.

6 418 RAMOS RIVERA ET AL. 9. Robison K, Hosmer J, Murray A, et al. Prevalence of HPV and type distribution in Southeast Asian women residing in Rhode Island. Austin J Obstet Gynecol. 2014;1: Ye F, Chan N, Feng T, et al. High prevalence of HPV59 in cytologically abnormal cervical samples. Exp Mol Pathol. 2015;99: Braaten KP, Laufer MR. Human papillomavirus (HPV), HPV-related disease, and the HPV vaccine. Rev Obstet Gynecol. 2008;1: Population Estimates, July 1, 2015, (V2015). Census.gov. N.p., Web Accessed June 30, College of American Pathologists (CAP). All Common Checklist, Cytopathology Checklist. July 28, Available at: p.org. Accessed November 16, Wright TC, Stoler MH, Behrens CM, Apple R, Derion T, Wright TL. The ATHENA human papillomavirus study: design, methods, and baseline results. Am J Obstet Gynecol. 2012;206: Naeem RC, Goldstein DY, Ramos Rivera G, et al. SurePath specimens versus ThinPrep specimen types on the COBAS 4800 platform: high-risk HPV status and cytology correlation in an ethnically diverse Bronx population. Lab Med. 2017; [Epub ahead of print]. How to cite this article: Ramos Rivera G, Khader SN, Lajara S, et al. The ATHENA HPV study underrepresents other high-risk HPV genotypes when compared with a diverse New York City population. Cytopathology. 2017;28: