Abstract
Background
Detection and treatment of anal histologic high-grade squamous intraepithelial lesions (hHSIL) prevents anal cancer. However, anal hHSIL incidence among women with human immunodeficiency virus (HIV, WHIV) remains unknown. Performance of anal high-risk human papillomavirus ([hr]HPV), anal cytology (anal-cyt), and both for hHSIL detection longitudinally over 2 years also remains undetermined.
Methods
We determined 2-year incidence and cumulative risk estimates (2-y-CR) of anal hHSIL among WHIV using prevalence and incidence (per 100 person-years [py]) observations stratified by baseline hrHPV and/or anal-cyt results.
Results
In total, 229 WHIV with complete baseline data were included in the analysis; 114 women without prevalent anal hHSIL were followed with 2 annual evaluations. Median age was 51, 63% were Black, and 23% were Hispanic. Anal hrHPV or abnormal anal-cyt was associated with an increased risk of incident anal hHSIL at 2 years (18.9/100py [95% confidence interval {CI} 11.4–31.3] and 13.4/100py [95% CI 8.0–22.7], respectively) compared with no detection of anal HPV or negative cytology (2.8/100py [95% CI 1.1–7.4] and 4.2 [95% CI, 1.8–10.2]) The presence of anal hrHPV with abnormal cytology was associated with 2-y-CR of anal hHSIL of 65.6% (95% CI 55.4%–75%); negative hrHPV with negative cytology was associated with 2-y-CR of anal hHSIL of 9.2% (95% CI 7.0–16.0).
Conclusions
Detection of anal hrHPV or abnormal anal cytology are comparable predictors for 2-y-CR of anal hHSIL. The absence of anal hrHPV combined with negative cytology was predictive of a lower (but measurable) risk of developing anal hHSIL. These findings provide important data to inform anal cancer screening guidelines for WHIV.
Keywords: women with HIV, anal HSIL, anal high-risk HPV test, anal cytology, anal cancer screening
Women with human immunodeficiency virus (HIV) have a high incidence of anal high-grade squamous intraepithelial lesions (HSIL). Detection of anal human papillomavirus (HPV) and/or abnormal anal cytology predict high rates of prevalent, incident and 2-year cumulative anal HSIL.
Graphical Abstract
Graphical Abstract.
This graphical abstract is also available at Tidbit: https://tidbitapp.io/tidbits/two-year-incidence-and-cumulative-risk-and-predictors-of-anal-high-grade-squamous-intraepithelial-lesions-anal-precancer-among-women-with-hiv-84f972c8-a553-41d2-b8ba-4ba3b7a031fe/update
Women with human immunodeficiency virus (HIV, WHIV) are at a disproportionately elevated risk of developing squamous cell carcinoma of the anus (SCCA). Foundational data from the ANal Cancer/HSIL Outcomes Research (ANCHOR) trial confirm that treatment of anal high-grade squamous intraepithelial lesion (HSIL) among persons with HIV (including women) reduces SCCA risk [1]. Algorithms and guidelines similar to those for cervical cancer screening and prevention are needed for detection of anal histologic (h)HSIL, in which at-risk individuals are screened using an anal swab for cytology and/or high-risk (hr) human papillomavirus (HPV), and patients with positive screening tests are referred for diagnostic high-resolution anoscopy (HRA) with directed biopsies. If anal hHSIL is detected, then the patients are subsequently treated.
Screening tests are important for detecting not only prevalent anal hHSIL but also predicting incident (ie, future) anal hHSIL risk, particularly as triage procedures are resource intensive and screening tests that accurately predict the absence of hHSIL detection for several years may decrease the need for yearly screening and triage. Understanding the natural history of hHSIL is essential to inform SCCA prevention and guide optimal screening algorithms. Currently, anal hHSIL incidence and predictors of risk among WHIV remain undetermined. Robust estimation of anal hHSIL incidence and associated predictors from longitudinal cohorts are crucial to understand drivers of anal hHSIL that may contribute to SCCA among WHIV but remain undetermined.
The AIDS Malignancy Consortium (AMC) study, “AMC-084: Screening HIV-positive women for anal cancer precursors (AMC084),” was a multi-center national trial that was designed to provide robust estimation of performance of various screening tests for prevalent hHSIL and SCCA [2], including hHSIL prevalence [3] and incidence and associated predictors of hHSIL risk. All enrolled women underwent annual screening evaluations for 2 years including anal cytology, anal HPV tests, and concurrent HRA with biopsy (in contrast to prior studies where HRAs were only conducted in women with abnormal cytology), thus providing unbiased estimates of anal hHSIL prevalence and incidence. In this article, we describe the incidence (one of the three primary endpoints for the AMC-084 protocol) and the 2-year risk of anal hHSIL among the cohort of women enrolled in national study, AMC 084.
METHODS
Study Design and Participants
AMC-084 recruitment occurred between 2014 and 2016. Detailed methods, HSIL prevalence, and performance of screening tests are previously reported in papers presenting the two of the other primary endpoints [2, 3]. Briefly, WHIV were recruited at twelve US sites. The clinicians responsible for performing HRA at each site were certified using a standardized approach focused on quality assessments developed and implemented by the AMC HPV Working Group [3]. The study protocol was approved by the US National Cancer Institute, Cancer Therapy Evaluation Program and by institutional review boards for each participating institution. Potential participants were screened using a standardized questionnaire and medical records review. At baseline, all women were evaluated using a standard protocol. Women without hHSIL were then followed semiannually for 2 years. Follow-up was discontinued if incident anal hHSIL, the primary endpoint, was diagnosed.
Sample and Subjects
Eligible women were 18 years old or older, diagnosed with HIV, had no history of anal HSIL determined based on cytology or histology [2, 3]. Study participants without prior anal hHSIL detection and with at least 1 follow-up visit with HRA and biopsies were eligible for analysis of incident anal hHSIL.
Procedures
At the baseline visit, data was collected for HIV, HPV-related, smoking and sexual history as previously reported [2, 3]. Recent HIV viral load and CD4 count data were collected. Participants underwent a targeted physical exam, including exams of the vulva, vagina/cervix, anus and perianus for signs of HPV-related lesions. Anal specimens were collected for cytology and hrHPV analysis, digital anorectal exam (DARE) and HRA of the anal canal and perianus with biopsies as previously described [3]. Participants were seen every 6 months for 2 years. At each 6-month visit, participants underwent anal evaluations including collecting specimens for anal cytology and hrHPV as well as DARE. At the 12- and 24-month visits, participants underwent collection of specimens for cervical cytology and hrHPV testing, and HRA with at least 2 directed or random biopsies as previously described [2, 3].
Laboratory Testing
Local pathology departments processed the cervical and anal cytology specimens and evaluated the cytology using the Bethesda Classification System [4]. Abnormal cytology was defined as atypical squamous cells of undetermined significance (ASC-US) or higher grade (ASC-US+). Cervical and anal hrHPV analyses were performed at the manufacturers” laboratories (Hybrid Capture 2 (HC2), Qiagen Corporation, Gaithersburg, MD and HPV-Aptima, Hologic Inc., Marlborough, MA) and results were reported to the investigators. The methods of hrHPV testing and histology assessments have been previously reported [2]. Briefly, HPV-HC2 is a signal amplification assay that detects ≥1 picogram of HPV-DNA for a pool of 13 different high-risk HPV types (16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, 59, and 68) [5]. HPV-HC2 is hereafter referred to as HPV DNA. HPV-Aptima assay is a nucleic acid amplification test that detects the HPV E6/E7 messenger RNA (mRNA) for a pool of 14 high-risk types of HPV (16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, 59, 66 and 68) [6]. HPV detection specimens underwent further analysis with the HPV-Aptima 16 18/45 Genotype Assay [7] to identify specimens with HPV genotypes 16 and 18/45 (designated 16/18/45+). HPV-Aptima is hereafter referred to as HPV mRNA. The study outcome of interest was anal hHSIL (vs no evidence of intraepithelial lesions or low-grade squamous intraepithelial lesions) as determined by the central pathology consensus review. If central review was not available, then local review was used for analysis. Among the 64 with prevalent anal HSIL, 59 were diagnosed on central pathology review and 5 women through local pathology review only.
Statistical Analysis
The primary endpoint for this study is to estimate the incidence and the 2-year cumulative risk calculated from prevalent and incident anal HSIL cases observed.
We used the approach by Clarke et al to calculate the 2-year cumulative risk [8]:
where p(x) is the probability of prevalent anal hHSIL for those with test values x, and S(2years;x) is the hHSIL probability at 2 years for those without prevalent disease and with test values x, estimated from Cox proportional hazards model. Test value x is defined as the characteristic or test result as denoted in column 1 of Tables 1 –3. For each cumulative risk estimate, we generated 100 bootstrap samples [9] to calculate 95% confidence intervals (Cis).
Table 1.
Anal hHSIL Incidence and Incidence Rate Ratio by Demographic and Clinical History
| Characteristics | Non-Prevalent Anal hHSIL Cohort | Incident hHSIL | Person Years | Incidence per 100 py | 95% CI | Incidence Rate Ratio | 95% CI | P Value |
|---|---|---|---|---|---|---|---|---|
| N = 114 | n (%) | (Incidence per 100 py) | (Incidence Rate Ratio) | |||||
| Overall | 114 (100) | 19 (16.7) | 222.3 | 8.5 | 5.45–13.40 | |||
| Demographic characteristics | ||||||||
| Age | ||||||||
| <40 y | 14 (12) | 3 (21.4) | 26.5 | 11.3 | 3.64–35.04 | 1.4 | .38–5.07 | .61 |
| 40–49 y | 36 (32) | 6 (16.7) | 72.2 | 8.3 | 3.73–18.5 | 1.0 | .37–2.82 | .96 |
| ≥50 y | 64 (56) | 10 (15.6) | 123.5 | 8.1 | 4.35–15.04 | ref | ||
| Race/Ethnicity | ||||||||
| NH Black | 72 (63) | 16 (22.2) | 135.0 | 11.9 | 7.26–19.34 | 2.1 | .61–7.19 | .24 |
| NH White or Other | 16 (14) | 0 (0) | 34.3 | 0.0 | 0.0 | NA | NA | NA |
| Hispanic | 26 (23) | 3 (11.5) | 53.0 | 5.7 | 1.82–17.55 | ref | ||
| Smoking status | ||||||||
| Former/Current | 67 (60) | 12 (17.9) | 130.1 | 9.2 | 5.23–16.23 | 1.2 | .45–2.94 | .76 |
| Never | 45 (40) | 7 (15.5) | 87.9 | 8.0 | 3.80–16.71 | ref | ||
| Education | ||||||||
| High school diploma or less | 59 (53) | 7 (11.9) | 118.8 | 5.9 | 2.81–12.36 | 0.5 | .20–1.36 | .19 |
| Some college or higher | 52 (47) | 11 (21.1) | 98.5 | 11.2 | 6.18–20.15 | ref | ||
| Annual income | ||||||||
| <$20K | 90 (84) | 16 (17.78) | 171.7 | 9.3 | 5.71–15.21 | 1.1 | .33–3.84 | .86 |
| ≥$20K | 17 (16) | 3 (17.65) | 36.1 | 8.3 | 2.68–25.78 | ref | ||
| Marital status | ||||||||
| Married/Not married, living with someone | 26 (23) | 6 (23.1) | 49.0 | 12.2 | 5.50–27.27 | 1.6 | .62–4.30 | .32 |
| Divorced/Widowed/Single | 88 (77) | 13 (14.8) | 173.3 | 7.5 | 4.34–12.92 | ref | ||
| HIV characteristics | ||||||||
| Current CD4 T-cell count | ||||||||
| ≤200 cells/mm3 | 4 (3) | 1 (25.0) | 8.9 | 11.2 | 1.57–79.36 | 1.2 | .16–9.12 | .85 |
| 201–350 cells/mm3 | 8 (7) | 0 (0) | 17.3 | NA | NA | NA | NA | NA |
| >350 cells/mm3 | 102 (90) | 18 (17.6) | 196.1 | 9.2 | 5.78–14.57 | ref | ||
| Viral load | ||||||||
| Suppressed (≤200 copies/mm3) | 99 (88) | 17 (17.2) | 192.5 | 8.8 | 5.49–14.21 | 1.2 | .28–5.32 | .78 |
| Unsuppressed (>200 copies/mm3) | 14 (12) | 2 (14.3) | 27.8 | 7.2 | 1.79–28.73 | ref | ||
| Nadir CD4 T-cell count | ||||||||
| ≤200 cells/mm3 | 46 (42) | 7 (15.2) | 91.5 | 7.6 | 3.64–16.04 | 0.8 | .33–2.18 | .73 |
| >200 cells/mm3 | 63 (58) | 11(17.4) | 121.6 | 9.0 | 5.01–16.33 | |||
| Current cART user | ||||||||
| Yes | 107 (95) | 18 (94.7) | 207.7 | 8.7 | 5.46–13.75 | 1.1 | .15–8.21 | .93 |
| No/unsure | 6 (5) | 1 (16.7) | 12.7 | 7.9 | 1.11–56.11 | ref | ||
| Reported clinical history | ||||||||
| Lifetime male anal sex partners | ||||||||
| 0 | 48 (42) | 11 (22.92) | 90.2 | 12.2 | 6.75–22.02 | ref | ||
| 1+ | 65 (58) | 8 (12.31) | 129.9 | 6.1 | 3.1–12.3 | 0.5 | .20–1.25 | .14 |
| History of anogenital warts | ||||||||
| Yes | 22 (20) | 6 (27.3) | 41.1 | 14.6 | 6.56–32.51 | 2.1 | .80–5.71 | .13 |
| No | 89 (80) | 12 (13.5) | 176.1 | 6.8 | 3.87–12.0 | ref | ||
| History of abnormal cervical cytology | ||||||||
| Yes | 54 (48) | 11 (20.4) | 10.5.4 | 10.4 | 5.78–18.83 | 1.5 | .60–3.71 | .39 |
| No/unsure/declined | 59 (52) | 8 (13.5) | 114.5 | 7.0 | 3.49–13.97 | ref | ||
| History of sexual assault | ||||||||
| Yes | 50 (45) | 10 (20.0) | 96.0 | 10.4 | 5.60–19.35 | 1.4 | .56–3.42 | .47 |
| No/declined | 61 (55) | 9 (14.7) | 120.0 | 7.5 | 3.90–14.40 | ref |
Abbreviations: cART, combination antiretroviral therapy; CI, confidence interval; hHSIL, high-grade squamous intraepithelial lesions; HIV, human immunodeficiency virus; NH, non-hispanic; ref, reference.
Table 3.
Prevalence, Incidence, Incidence Rate Ratio and Cumulative Risk of Anal hHSIL by Baseline Anal HPV Test and/or Anal Cytology
| Study Cohort (N = 229) | Prevalent Anal hHSIL (%) (N = 64) | 95% CI Prevalence | Study Cohort Incidence N = 114 | Incidence per 100 py (95% CI) | Incidence Rate Ratio (95% CI) | P Value* | 2-y Cumulative Risk (%) (95% CI) | P Value* | |
|---|---|---|---|---|---|---|---|---|---|
| Overall | 229 | 64 (27.9) | 22.1–33.8 | 114 | 8.5 (5.4–13.4) | — | — | 38.3 (33.7–43.8) | — |
| Baseline anal screening results | |||||||||
| Anal cytology | |||||||||
| Abnormal cyrology | 135 (59) | 53 (39.3) | 31.8–48.4 | 57 (50) | 13.4 (8.0–22.7) | 3.17 (1.1–8.8) | .0268 | 52.1 (44.0–59.5) | <.001 |
| NILMa | 94 (41) | 11 (11.7) | 6.7–20.4 | 57 (50) | 4.2 (1.8–10.2) | Ref | 18.3 (13.5–25.3) | ||
| Anal HPV (DNA) | |||||||||
| HPV+ | 111 (48) | 46 (41.4) | 33.2–51.7 | 45 (39) | 18.9 (11.4–31.3) | 6.8 (2.2–20.4) | <.001 | 58.7 (51.2–70.3) | <.001 |
| HPV− | 118 (52) | 18 (15.2) | 10.0–23.3 | 69 (61) | 2.8 (1.1–7.4) | Ref | 18.7 (15.0–25.4) | ||
| Anal HPV (mRNA) | |||||||||
| HPV+ | 102 (45) | 48 (47.1) | 38.3–57.8 | 35 (31) | 18.4 (10.2–33.3) | 3.75 (1.5–9.3) | .0044 | 64.7 (56.5–73.2) | <.001 |
| HPV− | 127 (55) | 16 (12.6) | 8.0–19.9 | 79 (69) | 4.9 (2.4–9.8) | Ref | 18.5 (14.1–21.9) | ||
| Anal HPV 16+b | 30 (13.1) | 22 (73.3) | 59.1–90.9 | 5 (4.4) | 34.9 (11.2–108.3) | 4.7 (1.3–16.0) | .0143 | 87.6 (NE) | NE |
| Anal HPV 16− | 199 (86.9) | 42 (21.1) | 16.1–27.6 | 109 (95.6) | 7.4 (4.6–12.2) | Ref | 30.5 (25.3–36.9) | Ref | |
| Anal HPV (DNA) and/or cytology | |||||||||
| HPV+ and abnormal cytology | 84 (36.7) | 40 (47.6) | 38.1–59.6 | 29 (25.4) | 25.0 (14.2–44.1) | 10.8 (2.4–48.5) | .0018 | 65.6 (55.4–75.3) | <.001 |
| HPV+ and NILM | 27 (11.8) | 6 (22.2) | 11.0–45.0 | 16 (14.0) | 9.6 (3.1–29.7) | 4.1 (.7–24.8) | .1191 | 36.8 (22.0–54.0) | .001 |
| HPV− and abnormal cytology | 51 (22.3) | 13 (25.5) | 15.9–40.8 | 28 (24.6) | 3.5 (.9–14.2) | 1.5 (.2–10.9) | .6665 | 30.5 (25.0–38.7) | .03 |
| HPV− and NILM | 67 (29.3) | 5 (7.5) | 3.2–17.3 | 41 (36.0) | 2.3 (.7–9.2) | Ref | Ref | 9.2 (7.0–16.0) | Ref |
| Anal HPV (mRNA) and/or cytology | |||||||||
| HPV+ and abnormal cytology | 83 (35.8) | 44 (53.7) | 43.9–65.6 | 26 (22.8) | 21.2 (11.0–40.7) | 7.1 (1.9–26.3) | .0032 | 69.6 (58.2–79.1) | <.001 |
| HPV+ and NILM | 20 (8.7) | 4 (20.0) | 8.3–48.0 | 9 (7.9) | 11.7 (2.9–46.7) | 3.9 (.7–23.5) | .1388 | 40.2 (20.1–64.0) | .005 |
| HPV− and abnormal cytology | 53 (23.1) | 9 (17.0) | 9.4–30.8 | 31 (27.2) | 8.1 (3.3–19.4) | 2.7 (.7–11.4) | .1702 | 25.1 (17.0–34.3) | .07 |
| HPV− and NILM | 74 (32.3) | 7 (9.5) | 4.7–19.1 | 48 (42.1) | 3.0 (.9–9.2) | Ref | 12.7 (9.0–20.0) | Ref | |
| Anal HPV 16 (mRNA) and/or cytology | |||||||||
| HPV16+ and abnormal cytology | 25 (10.9) | 19 (76.0) | 60.9–94.7 | 4 (3.5) | 30.3 (7.6–121.3) | 8.8 (1.6–48.0) | .0120 | 86.5 (NE) | NE |
| HPV16+ and NILM | 5 (2.2) | 3 (60.0) | 29.3–122.7 | 1 (0.9) | 50.1 (7.1–355.9) | 14.5 (1.5–130.1) | .0166 | Model did not converge | NE |
| HPV16− and abnormal cytology | 110 (48.0) | 34 (30.9) | 23.4–40.9 | 53 (46.5) | 12.3 (7.0–21.6) | 3.6 (1.1–11.0) | .0277 | 44.4 (36.2–54.1) | <.001 |
| HPV16− and NILM | 89 (38.9) | 8 (9.0) | 4.6–17.4 | 56 (49.1) | 3.4 (1.3–9.2) | Ref | Ref | 13.9 (9.0–18.7) | Ref |
Abbreviations: CI, confidence interval; DNA, deoxyribonucleic acid; hHSIL, high-grade squamous intraepithelial lesions; HPV, human papillomavirus; NE, could not be estimated; NILM, negative for intraepithelial lesion or malignancy; Ref, reference.
aNILM: Negative for intraepithelial lesion or malignancy.
bHPV16 was only tested on those specimens where HR HPV was detected.
The analyses of incident anal hHSIL were restricted to the women whose initial histology results were classified as negative for anal hHSIL at baseline. Incidence rates were computed by dividing the total number of events observed by the total number of person-years of observed follow-up, and comparisons of the incidence rates were performed using Poisson regression [10]. Results from these incidence analyses are reported as incidence rate ratios (IRRs) and 95% CIs.
The prevalence by test value × for this cohort was defined as the ratio of the number of women diagnosed with anal hHSIL at baseline to the number of enrolled women with test values x. Log-binomial regression was used to estimate the prevalence and 95% CI [11].
All statistical analyses except bootstrapped 95% CI around cumulative risk were performed using SAS 9.4 (SAS Institute Inc., Cary, NC). Bootstrapped 95% CI around 2-year cumulative risk was generated using R (version 4.2.1). P values less than .05 were interpreted as indicating statistical significance.
RESULTS
Of the 229 WHIV enrolled in the study with complete baseline data, 64 (28%) women had prevalent anal hHSIL (by local or central pathology read), 51 (22%) were excluded because of lack of follow-up, 114 (50%) women without prevalent anal hHSIL and having at least 1 follow-up HRA were included in the analysis of incident anal HSIL (see Figure 1), and the full cohort of 229 was used for the 2-year cumulative risk calculations. The median age of WHIV in the follow-up cohort of the 114 women was 51 years (interquartile range [IQR] 44–55). Women were predominantly non-Hispanic African Americans (72 or 63%) and current or former smokers (67 or 60%). Nearly half (54 or 48%) reported a history an abnormal cervical cytology, 58% (65) reported at least 1 lifetime male anal sex partner, and nearly half (50 or 45%) reported a prior sexual assault (see Table 1). The majority had well controlled HIV infection (viral load <200 copies/mL); the median CD4 count was 691 (IQR 533–920) cells/μL, and 90% (102) of the women had CD4 counts >350 cells/μL. In total, 27% (30) had abnormal cervical/vaginal cytology, and 23% (26) tested positive for cervical/vaginal hrHPV-DNA. The 51 participants excluded from incident hHSIL analyses due to lack of follow-up had similar demographic and clinical characteristics compared with the studied cohort (data not shown).
Figure 1.
Consort diagram. Abbreviations: HSIL, high-grade squamous intraepithelial lesions; WHIV, women with human immunodeficiency virus.
The overall incidence (per 100 py) of anal hHSIL among WHIV was 8.5 per 100 person-years (py) [95% CI, 5.45–13.40]. Anal hHSIL incidence (per 100 py) did not differ by age categories (<50-year-old vs ≥50-year-old: 9.1 [95% CI, 4.74–17.51] vs 8.1 [95% CI, 4.35–15.04]) or race/ethnicity (non-Hispanic Black [11.9 {95% CI, 7.26–19.34}]) when compared to Hispanic women (5.7 [95% CI, 1.82–17.55]). Incidence estimates did not significantly differ by smoking status (former/current and never); or nadir CD4 count (≤200 and >200 cells/μL) (Table 1). Similarly, the 2-year cumulative risk did not differ by age, race, smoking status. or nadir CD4 (Supplementary Table).
Table 2 shows the relationship between cervical cytology and hrHPV test at baseline and incident anal hHSIL. Although abnormal cytology (defined as ASC-US+) and any hrHPV were not significantly associated with hHSIL, atypical squamous cells cannot rule out HSIL (ASC-H), and HSIL cytology and HPV-16 were associated with anal hHSIL (incidence rate ratio [IRR] 10.5, 95% CI, 2.39–46.27) and (IRR 8.7, 95% CI, 1.68–44.82, respectively) (Table 2). Detection of any cervical hrHPV (compared with no detection of cervical hrHPV) was associated with a higher 2-year cumulative risk of anal hHSIL (49.2%, 95% CI, 41.4%–59.2% and 33.5% CI 27.9–39.6, respectively) (Supplementary Table).
Table 2.
Anal hHSIL Incidence and Incidence Rate Ratio by Baseline Cervical/Vaginal Screening Results
| Baseline Cervical/vaginal Screening Results | Non-Prevalent Anal hHSIL Cohort | Incident hHSIL | Person Y | Incidence per 100 py | 95% CI | Incidence Rate Ratio | 95% CI | P Value |
|---|---|---|---|---|---|---|---|---|
| N = 114 | n (%) | (Incidence per 100 py) | (Incidence Rate Ratio) | |||||
| Cervical/vaginal cytology | ||||||||
| ASC-H/HSIL | 2 (2) | 2 (100) | 2.2 | 92.3 | 23.08–369.0 | 10.5 | 2.39–46.27 | <.001 |
| ASC-US/LSIL | 28 (25) | 3 (10.7) | 56.2 | 5.3 | 1.72–16.54 | 0.6 | .17–2.11 | .43 |
| NILMa | 82 (73) | 14 (17.1) | 159.5 | 8.8 | 5.20–14.82 | ref | ||
| Cervical /vaginal cytology | ||||||||
| Not NILM | 30 (27) | 5 (16.7) | 58.4 | 8.6 | 3.56–20.57 | 0.9 | .35–2.71 | .96 |
| NILM | 82 (73) | 14 (17.1) | 159.5 | 8.8 | 5.20–14.82 | ref | ||
| Cervical/vaginal HPV DNA | ||||||||
| HPV+ | 26 (23) | 6 (23.1) | 47.3 | 12.6 | 5.69–28.20 | 1.7 | .65–4.49 | .28 |
| HPV− | 88 (77) | 13 (14.7) | 175 | 7.4 | 4.31–12.80 | ref | ||
| Cervical /vaginal HPV (mRNA) | ||||||||
| HPV+ | 26 (23) | 7 (26.9) | 49.3 | 14.2 | 6.77–29.79 | 2.1 | .81–5.20 | .13 |
| HPV− | 88 (77) | 12 (13.6) | 173 | 6.9 | 3.94–12.21 | ref | ||
| Cervical /vaginal HPV positive (mRNA) | ||||||||
| HPV16+ | 2 (8) | 2 (100) | 2.2 | 92.2 | 23.08–369.01 | 8.7 | 1.68–44.82 | .0097 |
| HPV+ (not HPV16) | 24 (92) | 5 (20.8) | 47.1 | 10.6 | 4.42–25.49 | ref |
Abbreviations: ASC-US, atypical cells of undetermined significance; CI, confidence interval; DNA, Deoxyribonucleic acid; hHSIL, high-grade squamous intraepithelial lesions; HPV, human papillomavirus; LSIL, low-grade squamous intraepithelial lesion; NILM, negative for intraepithelial lesion or malignancy, ref, reference.
aNILM: Negative for intraepithelial lesion or malignancy.
The hHSIL incidence (per 100 py) among participants with abnormal anal cytology (13.4 [95% CI 8.0–22.7]) differed significantly as compared to participants having negative anal cytology (4.2 [95% CI 1.8–10.2]) (Table 3). Similarly, a statistically significant difference in hHSIL incidence was observed by anal hrHPV positivity status at baseline. HPV detection by DNA was associated with an increased risk of incident anal hHSIL (incidence of 18.9 [95% CI 11.4–31.3]) compared with no detection of anal HPV (2.8 [95% CI 1.1–7.4]); the mRNA results were similarly associated (see Table 3). Co-testing with cytology and hrHPV testing demonstrated that the incidence was 11-fold greater (IRR = 10.8; 95% CI, 2.4–48.5) higher with detection of hrHPV (DNA testing) and abnormal cytology (incidence 25.0 [95% CI 14.2–44.1]) compared to negative cytology in combination with no HPV detection (incidence 2.3 [95% CI, .7–9.2]).
The 2-year cumulative risk of anal hHSIL was similarly high for detection of anal hrHPV (58.7% [95% CI 51.2%–70.3%]) or abnormal anal cytology (52.1% [95% CI, 44.0%–59.5%]) compared with absence of anal hrHPV (18.7% [95% CI 15.0%–25.4%]) or negative cytology (18.3% [95% CI, 13.5%–5.3%]). Assessing the different combinations of co-testing (concurrent hrHPV and cytology results) demonstrated that the combination of detection of anal hrHPV and abnormal cytology was most likely to be associated with anal hHSIL at 2 years (65.6% [95% CI, 55.4%–75%]) and conversely negative hrHPV with negative cytology was least likely to be associated with anal hHSIL (9.2% [95% CI 7.0%–16.0%]) (see Table 3).
DISCUSSION
To our knowledge, this study is the first to report the 2-year incidence, 2-year cumulative risk, and predictors for anal hHSIL in a cohort of women with HIV. Anal hHSIL incidence was 8.5 per 100 py. Baseline detection of anal hrHPV (with HPV-Aptima or HC2) or abnormal anal cytology were associated with development of anal hHSIL (incidence of 18.4/100py and 13.4/100py, respectively), whereas the absence of anal hrHPV or negative cytology were predictive of a significantly lower incidence of anal hHSIL (2.8/100py and 4.2/100py, respectively). The 2-year cumulative incidence was significantly higher for detection of anal HR HPV DNA (52%) or abnormal anal cytology (59%) but remained over 18% for women with baseline absence of HR HPV DNA or negative baseline anal cytology.
Previously published prospective longitudinal studies of people with HIV included only men who have sex with men (MSM). The Study of the Prevention of Anal Cancer (SPANC) study [9] conducted in Australia included 397 HIV negative and 220 HIV positive gay and bisexual (GBM) men followed with annual HRA evaluations over 3 years to determine the natural history of anal HPV and anal HSIL (defined as composite (c) HSIL, ie, either with HSIL cytology and/or HSIL histology). The SPANC study found that for those without prevalent anal cHSIL, the incidence of anal cHSIL was 10.1/100py in HIV-negative GBM compared with 14.0/100py for GBM with HIV [12]. Risk factors for anal cHSIL incidence included age younger than 45 years, as well as diagnosed with HIV. Anal hrHPV status at the current visit was a strong predictor of concurrent anal cHSIL; cHSIL incidence was 26.2, 11.0, and 2.8 per 100py in men with HPV16, other hrHPV or no hrHPV detected, respectively. Jongen et al [13] reported a study from the Netherlands of 107 MSM with HIV without prevalent anal hHSIL who had a follow-up HRA 1–4.5 years from the first HRA and found that the incidence of anal hHSIL was 15.9 per 100py (95% CI, 10.7–23.5) and was not associated with any clinical or demographic factors. Only 107 of the 807 study participants without prevalent hHSIL had a follow-up HRA. Clark et al reported on a cohort of 135 (of the 259) MSM LWH without prevalent hHSIL who had at least 1 follow-up HRA and found that 25 had incident anal hHSIL over a median of 2.1 years [8]. In that study, predictors of incident anal hHSIL included detection of HPV (incident HSIL at 2 years HPV positive—10% and HPV negative at 3.3%) and abnormal anal cytology (ASC-US+ 10.6% and normal 3.9%). We found that the incidence of anal hHSIL in WHIV is similar to that of GBM without HIV (from the SPANC study).
Our findings also suggest that ASC-H or HSIL cervical cytology, or cervical HPV-16 detection were associated with a higher risk of anal hHSIL. Thus, because other demographic variables (including age and prior abnormal cytology) do not appear to be associated with anal hHSIL, recent cervical ASC-H/HSIL or HPV-16 results may increase the clinical suspicion for anal HSIL.
Multiple studies have reported on the prevalence and risk factors for anal hHSIL in WHIV however we are unaware of any studies reporting anal hHSIL incidence among WHIV. Liu et al [14] report prevalence data from a cohort of 381 WHIV, where women with ASC-US+ (68%) were referred for HRA; 42% of whom were found to have anal hHSIL (comparable to 37% of women with abnormal anal cytology having anal hHSIL in our cohort). In a collaborative pooled analysis by Lin et al [15] described an hHSIL prevalence of 7.7% among 1003 WHIV included in the analysis, and that detection of cervical HPV was associated with anal hHSIL. Interestingly, although we did find that cervical HPV was significant in univariate analysis for prevalent anal hHSIL, the results were not significant in multivariate analysis [3].
Unlike in cervical cancer screening, where HPV testing alone is significantly predictive of future cervical hHSIL (compared with cervical cytology), we found that anal hrHPV testing, cytology, or cotesting with cytology and HPV are all comparable predictors of future hHSIL. The lack of detection of anal hrHPV in combination with normal cytology had the lowest 2-year risk of anal hHSIL, yet co-testing would still miss 9.2% and 12.7% of hHSIL cases, respectively, using the Qiagen and Aptima tests. The number of HPV-negative anal hHSIL, especially for prevalent cases was higher than expected (when compared to negative co-testing in cervical samples). Because the hrHPV detection thresholds utilized for the Qiagen and Aptima tests are optimized for cervical specimens, the anal hrHPV testing threshold may need to be decreased in order to adjust for the relatively acellular anal specimens. A more precise (sensitive) biomarker for anal hHSIL would greatly reduce the number of patients referred for repeated HRAs in the future. Promising candidates include optimized HPV assays (using restrictive genotyping and/or alternate cycle thresholds) [16], methylation markers [17, 18] and automated dual staining [19].
The strengths of this study include the racial and ethnic diversity of the participants and that all women enrolled in the study underwent high-resolution anoscopy (HRA) and biopsy (not only women with abnormal cytology). Furthermore, all the clinicians who performed HRAs underwent a rigorous certification process before study sites were activated. In addition, our study is the first to measure the 2-year anal hHSIL cumulative risk, a methodology that has been utilized in large cervical screening cohorts, which provides a basis for comparing screening tests at the 2 anatomic sites. The limitations of the study include: no central review of the local cytology interpretations, the use of commercial non-polymerase chain reaction (PCR)-based HRHPV assays, 21.8% loss to follow-up, small number of HPV 16 cases, small number of different populations, and the small number of incident anal hHSILs, limiting the study's power, particularly to determine the additional predictive value of HPV and cytology cotesting.
CONCLUSION
WHIV have a high incidence of anal hHSIL, and detection of anal HPV and abnormal anal cytology predict prevalent and incident anal hHSIL. However, the converse, negative hrHPV and negative cytology, do not assure the absence of prevalent or incident anal hHSIL. These crucial data suggest that further improvement in anal HR HPV testing sensitivity is needed. In addition, the current study will inform anal cancer screening guidelines and suggest that annual screening with hrHPV testing or cytology may be needed, and women with a recent cervical ASC-H/HSIL cytology or cervical HPV-16 detection should be prioritized for anal cancer screening . Longer follow-up and further evaluation of additional biomarkers are needed to improve algorithms for anal cancer prevention.
Supplementary Data
Supplementary materials are available at Clinical Infectious Diseases online. Consisting of data provided by the authors to benefit the reader, the posted materials are not copyedited and are the sole responsibility of the authors, so questions or comments should be addressed to the corresponding author.
Supplementary Material
Contributor Information
Elizabeth A Stier, Department of Obstetrics and Gynecology, Boston University Chobanian & Avedisian School of Medicine, Boston, Massachusetts, USA.
Mayuri Jain, Department of Population Health Science and Policy, Icahn School of Medicine at Mount Sinai, Institute for Healthcare Delivery Science, NewYork, New York, USA; Tisch Cancer Institute, Icahn School of Medicine at Mount Sinai, NewYork, New York, USA.
Himanshu Joshi, Department of Population Health Science and Policy, Icahn School of Medicine at Mount Sinai, Institute for Healthcare Delivery Science, NewYork, New York, USA; Tisch Cancer Institute, Icahn School of Medicine at Mount Sinai, NewYork, New York, USA.
Teresa M Darragh, Department of Pathology, UCSF Mt. Zion Medical Center, SanFrancisco, California, USA.
Ashish A Deshmukh, Department of Public Health Sciences, Medical University of South Carolina, Charleston, South Carolina, USA; Hollings Cancer Center, Medical University of South Carolina, Charleston, South Carolina, USA.
Jeannette Lee, Department of Biostatistics, University of Arkansas for Medical Sciences, Little Rock, Arkansas, USA.
Mark H Einstein, Department of OB/GYN & Women's Health, Rutgers- NJMS, Newark, New Jersey, USA.
Naomi Jay, Anal Neoplasia Clinic, Research, and Education Center, University of California San Francisco, San Francisco, California, USA.
J Michael Berry-Lawhorn, Anal Neoplasia Clinic, Research, and Education Center, University of California San Francisco, San Francisco, California, USA; Division of Hematology Oncology, University of California San Francisco, San Francisco, California, USA.
Joel M Palefsky, Anal Neoplasia Clinic, Research, and Education Center, University of California San Francisco, San Francisco, California, USA; Department of Medicine, University of California San Francisco, San Francisco, California, USA.
Timothy Wilkin, Division of Infectious Diseases, Department of Medicine, Cornell University, NewYork, New York, USA.
Grant Ellsworth, Division of Infectious Diseases, Department of Medicine, Cornell University, NewYork, New York, USA.
Audrey L French, Division of Infectious Diseases, CORE Center/Stroger Hospital of Cook County, Chicago, Illinois, USA.
Luis F Barroso, Department of Internal Medicine (Infectious Diseases), Wake Forest University Health Sciences, Winston-Salem, North Carolina, USA.
Rebecca Levine, Department of Surgery, Montefiore, Bronx, NewYork, USA.
Humberto M Guiot, Department of Medicine, University of Puerto Rico School of Medicine, San Juan, PR, USA; Department of Microbiology & Medical Zoology, University of Puerto Rico School of Medicine, San Juan, Puerto Rico USA.
M Katayoon Rezaei, Department of Pathology, George Washington University, Washington, DC, USA.
Elizabeth Chiao, Department of Epidemiology, Division of Cancer Prevention, University of Texas - MD Anderson Cancer Center, Houston, Texas, USA.
Notes
Author Contributions. E. Y. C., J. L., E. A. S., and A. A. D. planned the study; E. C., H. J., M. J., and E. A. S. wrote the manuscript; E. A. S., M. H. E., N. J., J. M. B.-L., JMP, T.W., G. E., A. L. F., L. F. B., R. L., H. M. G., and E. Y. C. did acquisition of data; H. J. and M. J. designed the analysis; H. J. and M. J. analyzed the data; E. Y. C. and E. A. S. interpreted the data. All authors read and approved the final manuscript and critically revised the paper.
Financial support. This work was supported by the National Cancer Institute at the National Institutes of Health (NIH), U54 grant number U54CA242646 to J. M. P.; grant number CA163103 to E. Y. C. and E. A. S. (principal investigator [PI]: Dr Elizabeth Chiao); and AIDS Malignancy Consortium grant number UM1CA121947 to E. Y. C., E. A. S. (also reported by A. L. F., J. L., R. A. L., J. M. B.-L., and H. M. G.). (PI: Dr Joseph Sparano). H. J. was supported by the AIDS Malignancy Consortium as an employee of Mount Sinai Hospital. Hologic, Qiagen, and HPV-Aptima provided in-kind contributions for HPV testing.
References
- 1. Palefsky JM, Lee JY, Jay N, et al. Treatment of anal high-grade squamous intraepithelial lesions to prevent anal cancer. N Engl J Med 2022; 386:2273–82. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Chiao EY, Lensing SY, Wiley DJ, et al. Screening strategies for the detection of anal high-grade squamous intraepithelial lesions in women living with HIV. AIDS 2020; 34:2249–58. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Stier EA, Lensing SY, Darragh TM, et al. Prevalence of and risk factors for anal high-grade squamous intraepithelial lesions in women living with human immunodeficiency virus. Clin Infect Dis 2020; 70:1701–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Solomon D, Davey D, Kurman R, et al. The 2001 Bethesda system: terminology for reporting results of cervical cytology. JAMA 2002; 287:2114–9. [DOI] [PubMed] [Google Scholar]
- 5. https://www.qiagen.com/gb/products/diagnostics-and-clinical-research/sexual-reproductive-health/cervical-cancer-screening/digene-hc2-high-risk-hpv-dna-test-ce/. Accessed 24 July 2023.
- 6. https://www.hologic.com/package-inserts/diagnostic-products/aptima-hpv-assay. Accessed 24 July 2023.
- 7. https://www.hologic.com/package-inserts/diagnostic-products/aptima-hpv-16-1845-genotype-assay. Accessed 24 July 2023.
- 8. Clarke MA, Cheung LC, Lorey T, et al. 5-year prospective evaluation of cytology, human papillomavirus testing, and biomarkers for detection of anal precancer in human immunodeficiency virus-positive men who have sex with men. Clin Infect Dis 2019; 69:631–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Canty A, Ripley BD. Bootstrap R (S-Plus) Functions. R package version 1.3–28.1. Published 2022. Available at: https://cran.r-project.org/web/packages/boot/citation.html. Accessed 1 April 2023.
- 10. PROC GENMOD: The GENMOD Procedure: SAS/STAT(R) 9.2 User's Guide, Second Edition. Available at: https://support.sas.com/documentation/cdl/en/statug/63033/HTML/default/viewer.htm#genmod_toc.htm. Accessed 13 February 2023.
- 11. Spiegelman D, Hertzmark E. Easy SAS calculations for risk or prevalence ratios and differences. Am J Epidemiol 2005; 162:199–200. [DOI] [PubMed] [Google Scholar]
- 12. Poynten IM, Jin F, Roberts JM, et al. The natural history of anal high-grade squamous intraepithelial lesions in gay and bisexual men. Clin Infect Dis 2021; 72:853–61. [DOI] [PubMed] [Google Scholar]
- 13. Jongen VW, Richel O, Marra E, et al. Anal squamous intraepithelial lesions (SILs) in human immunodeficiency virus-positive men who have sex with men: incidence and risk factors of SIL and of progression and clearance of low-grade SILs. J Infect Dis 2020; 222:62–73. [DOI] [PubMed] [Google Scholar]
- 14. Liu Y, Weiss K, Zamudio AR, et al. Primary anal cancer screening results from 381 women living with human immunodeficiency virus. J Infect Dis 2023; 227:932–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Lin C, Slama J, Gonzalez P, et al. Cervical determinants of anal HPV infection and high-grade anal lesions in women: a collaborative pooled analysis. Lancet Infect Dis 2019; 19:880–91. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Ellsworth GB, Stier EA, Chiao EY, et al. Xpert HPV as a screening tool for anal histologic high-grade squamous intraepithelial lesions in women living with HIV. J Acquir Immune Defic Syndr 2021; 87:978–84. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. van der Zee RP, Meijer CJLM, Cuming T, et al. Characterisation of anal intraepithelial neoplasia and anal cancer in HIV-positive men by immunohistochemical markers p16, Ki-67, HPV-E4 and DNA methylation markers. Int J Cancer 2021; 149:1833–44. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Phillips S, Cassells K, Garland SM, et al. Gene methylation of CADM1 and MAL identified as a biomarker of high grade anal intraepithelial neoplasia. Sci Rep 2022; 12:3565. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Cohen CM, Wentzensen N, Lahrmann B, et al. Automated evaluation of p16/Ki-67 dual stain cytology as a biomarker for detection of anal precancer in MSM living with HIV. Clin Infect Dis 2022; 75:1565–72. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.