Demonstrating a Statistically Significant Association Between Anal HSIL and Positive OncoE6 Anal Test Result in MSMLWH

Abstract

Objectives –

To determine whether a positive OncoE6 Anal Test result has statistically significant higher odds of being associated with HSIL, and to calculate sensitivity and specificity of this test for predicting HSIL in adult men who have sex with men and are living with HIV (MSMLWH).

Methods –

MSMLWH ≥18 years having ≥ASCUS-grade anal cytology results were eligible to enroll in this cross-sectional study. Anal samples were collected just prior to the HRA procedure. OncoE6 anal test results were compared to histology, the reference standard. Sensitivity, specificity and odds ratio were calculated using HSIL as the threshold.

Results –

277 consented MSMLWH were enrolled between 06/2017 and 01/2022. Of these, 219 (79.1%) had biopsies obtained and histology performed; 81/219 (37%) participants had ≥1 biopsies with HSIL results while the remaining 138/219 (63%) had only LSIL or were negative for dysplasia. Anal samples from 7 participants (8.6%; 7/81) with HSIL and 3 (2.2%; 3/138) with LSIL had positive OncoE6 Anal Test results. Odds of having HSIL were 4.26 times higher among participants testing positive for HPV16/HPV18 E6 oncoprotein(s) (OR 4.26, 95% CI 1.07–16.95, p = 0.04). The OncoE6 Anal Test demonstrated excellent specificity; 97.83% (93.78–99.55) but poor sensitivity; 8.64% (3.55–17.0).

Conclusions –

In this highest-risk population for anal cancer, one could combine the OncoE6 Anal Test, having excellent specificity, with the anal Pap test, having higher sensitivity. Patients found having both an abnormal anal Pap and positive OncoE6 Anal Test result could be triaged for rapid scheduling of their HRA.

Précis:

Found a statistically significant association between HSIL and positive OncoE6 anal test results, and excellent specificity for the OncoE6 test relative to histology.

Introduction

Persistent infection with a high-risk human papillomavirus (HR-HPV) genotype can lead to integration of viral genome into the host genome¹. This can lead to an upregulation of oncoprotein expression, one of the necessary events for anal cancer progression^2,3. Integration of HR-HPV genome can result in high-grade squamous intraepithelial neoplasia, a known precursor to anal cancer⁴.

Although anal cancer is rare in most populations, its incidence in men who have sex with men, and who are living with HIV (MSMLWH) is about 3-fold greater than that of the highest worldwide incidences of cervical cancer^5–7. Incidence rates for anal cancer in the MSMLWH population increase to over 100/100,000 person-years in those ≥60 years⁸. Of the ~50,865 new cases of anal cancer diagnosed worldwide in 2020⁹, 9,440 were diagnosed in the U.S. Approximately 90% of these cases are estimated to be caused by HR-HPV, with ~88% of them being associated with HPV16 or HPV18^5,9.

To date, evidence-based guidelines for anal cancer screening in high-risk populations are evolving rapidly^10–12. Liquid-based anal cytology (anal Pap test) or digital anal rectal examination (DARE) along with HR-HPV DNA testing are currently used by some practitioners for primary screening. If the anal Pap or DARE results are abnormal, high resolution anoscopy (HRA) is recommended to the patient so biopsies can be obtained if lesions are suspicious for HSIL during the procedure. These biopsies are sent for histologic evaluation; the reference standard test for diagnosing anal cancer.

Screening algorithms incorporating anal cytology and HR-HPV DNA testing demonstrate good sensitivity but poor specificity for predicting high-grade squamous intraepithelial lesions (HSIL) in MSMLWH^11–15. Among this high-risk population, where prevalence of HR-HPV infection is high (~88%) and transient in nature, a more specific companion diagnostic to the anal Pap test is urgently needed to improve a triage algorithm that could be used for rapid scheduling of HRA for the most at-risk patients. This practice could also help reduce unnecessary use of invasive HRA and biopsy procedures, using HRA judiciously when needed, to better predict high-grade anal lesions¹⁶. To that end, we evaluated the OncoE6 Anal Test (Arbor Vita Corp., Fremont, CA), a lateral flow assay detecting HPV16 and HPV18 E6 oncoproteins from anal swab-based specimens to predict HSIL lesions in a population of MSMLWH.

Methods

Study Participants and Sample Collection

Approval was granted by the Committee on Human Research, Institutional Review Board for experimentation with human subjects for this study. All participants provided written informed consent. Anal swab-based specimens were collected between June 2017 and January 2022 on 277 eligible study participants recruited from a clinic where they were scheduled to undergo HRA guided biopsy procedures because of a history of abnormal anal cytology (≥atypical squamous cells of undetermined significance; ≥ASCUS) or DARE results.

The inclusion criteria included: being born male, of at least 18 years of age, having a history of sex with a male partner or someone who self-identifies as gay or bisexual, being a person living with HIV, and having ≥ASCUS grade cytology or abnormal DARE results. Individuals with a history of anal cancer were excluded from this study. Of these 277 eligible and consented study participants, 219 participants had anal biopsies obtained during their HRA procedure.

Swab-based anal specimens were collected by the attending physician, just prior to performing the HRA procedure. A pre-moistened FloqSwab (Cat. #502CS01, Copan Diagnostics, Murrieta, CA) was used to collect cells from the anal canal. After collection, the swab was immediately rinsed in 20 mL of PreservCyt fluid (Hologic, Inc., Marlborough, MA) to release the cells. These specimen vials were then transported to the laboratory on cold packs either same day, or after overnight 4°C storage at the clinic. Once in the lab, the specimens continued to be stored at 4°C prior to performing the OncoE6 anal test (Arbor Vita Corporation, Fremont, CA). Testing was performed within approximately two weeks of sample receipt.

OncoE6 Anal Test

The OncoE6 Anal Test consists of a processing kit (PN# 2000047) and rinse solution (PN# 2002010) (Arbor Vita Corporation, Fremont, CA). It is a lateral flow test that captures HPV16 and HPV18 E6 oncoproteins from the processed anal specimens via monoclonal antibodies. The complex is visualized using alkaline phosphatase-conjugated monoclonal antibody and specific substrate.

The swab-based anal samples in PreservCyt media were tested for E6 oncoproteins according to the manufacturer’s protocol. Briefly, 5 mL of PreservCyt fluid was centrifuged; 5 minutes (min) at 500 x g and the resulting pellets rinsed in 1 mL of rinse solution, vortexed, and centrifuged; 5 min at 1,000 x g. Solution A (750 uL) was added to the pellet, after which sample was rotated; 10 min at 8 rpm. Solution B (182 uL) was added next to tube, vortexed and rotated again; 15 min at 8 rpm before centrifuging; 15 sec at 10,000 x g. Resulting lysate was transferred to a clean 1.5 mL tube and Solution C (87 uL) was added, vortexed, and rotated; 15 min at 8 rpm. Tube was centrifuged; 15 seconds at 10,000 x g. The resulting sample (200 uL) was analyzed for HPV16 and HPV18 E6 oncoprotein using an immunochromatographic test strip via capillary action; 55 min at ambient temperature and across test area into absorbent pad. Test strip was transferred into Solution E for 12 min, and then to Solution F for 20–30 min depending upon temperature.

Results were read within 3 min of completing the testing and interpreted for presence of visible bands at HPV16, HPV18 and/or internal control locations on the test strip. Two independent readers, blinded to both clinical information and histology results, examined and interpreted test strip results. Discordant interpretations were resolved between the two readers before generating results. Photo images of each test strip was electronically recorded. Lack of a visible internal control line indicated an invalid result, prompting retesting. Each new lot of test strips underwent quality control before use with study participant samples, which consisted of testing cultured HeLa (ATCC^® CCL-2), and Caski (ATCC^® CRL-1550) cells that had been cultured and then stored in PreservCyt media at 4°C.

Statistical Analyses

Statistical analyses were conducted using R version 4.2.0. Odds ratios (95% CI) for HSIL versus ≤LSIL histology results based on the presence or absence of HPV16/HPV18 E6 oncoproteins was calculated and tested for association using a Fisher’s Exact test. A p value of <0.05 was considered statistically significant. Sensitivity and specificity (95% CI) were calculated using the statistical calculator at https://www.medcalc.org/calc/diagnostic_test.php.

Sample size calculation was based on detecting a minimum acceptable specificity of 85%, and was derived from prior HRA clinic volumes using a 15% prevalence rate for anal HSIL among a population of MSMLWH. The estimated total sample size needed for minimum acceptable specificity (range: 0.85–0.95) of the OncoE6 Anal Test for predicting HSIL in MSMLWH was 230; this assumed a prevalence of HSIL among MSMLWH receiving HRA in the clinic sample was between 10–30%, with an alpha = 0.05, 95% CI.

Results

Participants with a positive OncoE6 Anal Test result had a statistically significant higher odds of having HSIL histology results from their anal biopsies.

Seventy percent (7/10) of participants with a positive OncoE6 anal test result had one or more anal biopsies with HSIL lesions compared to 35.4% (74/209) of those with OncoE6 anal test negative results (Table 1). The odds of having HSIL histology results were 4.26 times higher among patients with detectable OncoE6 results. A Fisher’s Exact test was used to look for an association and found to have a statistically significant p value (p = 0.04) (Table 1).

Table 1.

Anal Histology Grade by HPV16/HPV18 E6 Oncoprotein Result

Group	HSIL Histology	≤LSIL Histology	Total
OncoE6 Positive	7	3	10
OncoE6 Negative	74	135	209
Total	81	138	219

OR (95% CI) = 4.26 (1.07–16.95)

Fisher’s test p value = 0.04

Sensitivity (95% CI) = 8.64% (3.55–17.00)

Specificity (95% CI) = 97.83% (93.78–99.55)

OncoE6 Anal Test demonstrated high specificity for predicting anal HSIL histology.

Of the 219 participants in which biopsies were obtained and histology results provided, 81/219 (37%) were found to have one or more HSIL reported. The remaining 138/219 (63%) participants had only ≤LSIL reported. In all, 10/219 (4.6%) participants had a positive OncoE6 anal test result. Seven of those positive results were from the HSIL group (8.6%; 7/81). Of these 7 OncoE6 positive samples, 2 were positive for HPV16 E6 oncoprotein, 4 were positive for HPV18 oncoprotein and 1 was positive for both HPV16 and HPV18 E6 oncoproteins. Three participants with OncoE6 positive results were from the ≤LSIL group (2.2%; 3/138: all 3 participants had LSIL). Of these 3 OncoE6 positive samples, 2 were positive for HPV16 E6 oncoprotein and 1 was positive for HPV18 E6 oncoprotein. OncoE6 demonstrated excellent specificity; 97.82% (95% CI 93.78–99.55%) but poor sensitivity; 8.64% (95% CI 3.55–17.0%) compared to histology (Table 1).

Discussion

E6 oncoprotein expression is key to HR-HPV-induced cellular transformation including high-grade squamous intraepithelial neoplasia, a known precursor to anal cancer^2–4. Previous studies have described the feasibility of detecting HPV16/HPV18 E6 oncoproteins when screening for cervical cancer. All of these published studies screened cervical samples using the OncoE6 Cervical Test; no anal samples were tested, nor was the OncoE6 Anal Test evaluated. These publications include a study by Chibwesha et al. where 200 women living with HIV in Zambia were screened, the OncoE6 Cervical Test illustrates excellent specificity (99%) but low sensitivity (31%)¹⁷. Another study by Ndizeye et al. testing 51 HIV-positive women from Burundi who had biopsies obtained for histology demonstrates high specificity (CIN2+; 98% or CIN3+; 96.1%) but low sensitivity (CIN2+; 42.1% or CIN3+; 58.3%) for the OncoE6 Cervical Test in detecting cervical precancerous lesions¹⁸. Lastly, in a prospective cohort study by Dong et al., the OncoE6 Cervical Test shows a specificity of 86.5% and sensitivity of 57.1% for detecting CIN3+ in HPV16 positive women¹⁹.

In this study, we evaluated the OncoE6 Anal Test on anal samples collected from 219 MSMLWH study participants, all of whom had abnormal anal cytology or DARE results and biopsies collected with anal histology data. In all, 37% (81/219) of participants had biopsies with evidence of high-grade lesions (HSIL), while only 4.6% (10/219) of these anal samples tested positive for detectable HPV16/HPV18 E6 oncoprotein(s).

This study describes the performance of the OncoE6 Anal Test to predict high-grade anal lesions from anal samples. These results demonstrated high specificity but poor sensitivity for the OncoE6 Anal Test; findings that are comparable to those described in previously published studies where cervical samples had been screened.

One strength of this study was the study population: individual living with HIV with the highest risk for developing anal cancer participated. Secondly, we had a significant percentage of individuals who underwent HRA and had biopsies collected for histologic evaluation that could be used in this evaluation.

A limitation of this study is that the conclusions drawn here may not be applicable to other groups of individuals who are at a lower risk of developing anal cancer. Secondly, although ~88% of anal cancers arise from HPV16 or HPV18, other HR-HPV genotypes associated with the remaining cases of anal cancer cannot be detected using this current version of the OncoE6 Anal Test²⁰. Incorporating other HR-HPV types into this testing platform would be helpful.

The New York State Department of Health AIDS Institute recommends anal cytology screening in all MSMLWH²¹. In this population at highest risk for anal cancer, where HR-HPV prevalence is high, one could consider incorporating the OncoE6 Anal Test, with its excellent specificity, as a companion diagnostic to the anal Pap test, with its higher sensitivity (69%−83%), but where false negative cytology results in the MSMLWH can be as high as 45%^22,23. In the MSMLWH population, the OncoE6 anal test proved to be highly specific, 97.83% (93.78–99.55) and had 4.26-fold higher odds of having a positive HPV OncoE6 Anal test result among participants with HSIL compared to those with ≤LSIL grade histology results (p=0.04). Combined testing for anal cancer screening incorporating the anal Pap test and the OncoE6 Anal Test could lead to identifying those MSMLWH at greatest need for HRA procedure; specifically, those patients having both an abnormal anal Pap result and a positive OncoE6 anal test result. This information could be used to triage these individuals for rapid HRA scheduling.

Conclusions

From this study conducted in a population at highest risk for anal cancer, we found a statistically significant association between having high-grade histology results and a positive OncoE6 anal test result. We also demonstrated excellent specificity of the OncoE6 anal test relative to histology.

There are a very limited number of HRA providers throughout the U.S. This situation results in clinics often having waiting times of at least three months or longer for the next available anoscopy appointment. While a negative OncoE6 Anal Test would not eliminate the need for HRA, the combination of an abnormal anal pap and a positive OncoE6 Anal Test could be used to triage patients for rapid scheduling of their HRA. Additionally, when practicing in a setting with a high lost to follow up rate, if the OncoE6 anal test result is positive, knowing that this test has a high specificity for HSIL, rather than waiting for the pathology report on the biopsy, the provider could biopsy and ablate the lesion(s) in a single visit.

In conclusion, in this high-risk population one could consider using the OncoE6 Anal Test, with its excellent specificity, in combination with the anal Pap, with its higher sensitivity. Future studies are urgently needed to determine whether this algorithm would improve prediction of HSIL, and perhaps, result in fewer unnecessary HRA procedures being performed.

List of all abbreviations and acronyms

ASCUS

Atypical squamous cells of undetermined significance

ATCC

American Type Culture Collection

°C

Degrees Centigrade

CI

Confidence interval

DARE

Digital anal rectal examination

DNA

Deoxyribonucleic acid

g

Relative centrifugal force

HIV

Human immunodeficiency virus

HPV

Human papillomavirus

HR-HPV

High-risk human papillomavirus

HRA

High resolution anoscopy

HSIL

High-grade squamous intraepithelial lesions

LSIL

Low-grade squamous intraepithelial lesions

min

Minutes

mL

Milliliter

MSM

Men who have sex with men

MSMLWH

Men who have sex with men and are living with HIV

OR

Odds ratio

rpm

Rotations per minute

sec

Seconds

uL

Microliter