The other side of screening: predictors of treatment and follow-up for anal precancers in a large health system : AIDS

Secondary Logo

Journal Logo


The other side of screening: predictors of treatment and follow-up for anal precancers in a large health system

Silvera, Richarda; Martinson, Tylerb; Gaisa, Michael M.a; Liu, Yuxinc; Deshmukh, Ashish A.e; Sigel, Keithd

Author Information
doi: 10.1097/QAD.0000000000002948
  • Free



Incidence and mortality rates of human papillomavirus (HPV)-associated squamous cell carcinoma of the anus (SCCA) have been rising in the United States [1,2]. People with HIV (PWH) are disproportionately affected with an annual incidence rate of 50 per 100 000; a 19-fold elevated risk compared with the general population [3]. The risk further increases among MSM with HIV with an annual incidence of 89 per 100 000 and standardized incidence ratio of 39 [3–9]. The incidence of SCCA is also elevated among HIV-uninfected MSM, with an annual incidence of 19 per 100 000 person-years [10,11].

Like cervical cancer, SCCA is preceded by high-grade squamous intraepithelial lesions (HSILs). Given the markedly elevated SCCA risk among PWH and emulating the successes of cervical cancer screening, most experts advocate for anal HSIL screening and treatment among PWH [12]. Whether treating anal HSIL ultimately leads to decreased anal cancer rates is currently under investigation; an ongoing large prospective randomized study (ANCHOR) is comparing HSIL treatment via ablation to observation to determine if surveillance and ablation is related to anal cancer incidence [13]. Several retrospective studies suggest that treating anal HSIL may prevent at least some anal cancers [14,15]. Among PWH, the prevalence of anal HSIL is high, a 2012 meta-analysis of HSIL among men with HIV found a pooled prevalence of 29.1% (22.8–35.4%) and among HIV-uninfected men, a pooled prevalence of 21.5% (9.2–14.9%) [16]. Out of concern for scarring, stricture, and mechanical compromise, anal HSIL treatment relies on targeted destruction. The tradeoff to this approach is substantial post-treatment recurrence; our cohort noted at a median of 12.2 months that 45% of patients had local recurrence and 60% had recurrence of any type [15]. These high rates of recurrence make ongoing surveillance following diagnosis and treatment all the more important [15,17–20].

Data on patient adherence to anal HSIL treatment and surveillance is scarce but crucial as utility, performance, and effectiveness of anal cancer screening continue to be debated. Using data from a large, longitudinal HRA research database, we evaluated point-prevalence and predictors of adherence to treatment and surveillance following diagnosis of anal HSIL.


The Mount Sinai Anal Dysplasia Program serves a large urban population of PWH and HIV-uninfected MSM. Patients are offered annual anal cytology screening. Cytological diagnoses of atypical cells of undetermined significance (ASCUS) or higher grade abnormalities trigger a referral for HRA, which is performed following previously described techniques [21]. Patients with histologically confirmed HSIL are offered a return visit for electrocautery ablation, preferably within 3 months of diagnosis. If disease burden is extensive and/or if patients cannot tolerate an office-based procedure, they are referred for surgical fulguration. Patients who undergo HSIL treatment by any modality are advised to return for surveillance HRA 6 months after treatment. Patients who decline initial treatment are also offered follow-up HRA 6 months after their index HSIL diagnosis. As adherence to these rigorous follow-up time intervals for treatment and surveillance HRA is challenging, we expanded the allowable timelines for the purpose of this analysis as described below.

From a longitudinal HRA research database, we identified patients who attended an initial HRA visit between April 2009 and December 2018. Data for this analysis was collected through 30 June 2020. Data were abstracted on demographics, insurance status, HIV clinical variables, HRA results, and retention in care. Approximately 70% of race/ethnicity data were self-reported and the remaining 30% were determined by a published probabilistic approach [22]. We identified patients who were diagnosed with HSIL on index HRA and measured the following primary outcomes: return for HSIL treatment within 6 months of diagnosis; follow-up HRA within 18 months of index HSIL diagnosis; and follow-up HRA within 18 months of index HSIL diagnosis for untreated patients. For this analysis, we defined ‘lost to follow-up’ as no documented HRA at our testing program during the 18 months after initial HSIL diagnosis. We also captured incident anal cancer diagnoses for patients who completed treatment within 6 months of HSIL diagnosis and repeat HRA within 18 months of HSIL diagnosis as well as among patients who did not complete those visits for whom data was available. We then compared the proportion of baseline characteristics by completion of treatment as well as completion or repeat examination within 18 months of diagnosis, testing for differences using χ2 tests. Prior to our analysis, we also selected several of our predictors to fit adjusted logistic regression models predicting the primary outcomes. Statistical analysis was performed using SAS 9.4.

Ethical approval and waiver of written consent for this retrospective analysis was obtained from the IRB at the Icahn School of Medicine at Mount Sinai.


We identified 3369 patients who underwent at least one HRA. On initial HRA, 1179 (35%) unique individuals were diagnosed with anal HSIL, whereas 14 (0.4%) were found to have anal cancer (excluded from further analysis). Among those diagnosed with HSIL, at birth 1054 (90%) were men (of whom 99% self-identified as MSM and 1% as heterosexual) and 114 (10%) were women. Sex and gender data for 11 patients was missing. HSIL patients were racially and ethnically diverse: 36% were white, 23% black, and 25% Hispanic. Among the HSIL cohort, 91% were PWH, and the majority had public insurance (51%).

Among 1179 patients diagnosed with HSIL (Fig. 1), 684 (58%) returned for treatment within 6 months of HSIL diagnosis, either receiving office-based electrocautery ablation (n = 478, 70%) or fulguration in the operating room by colorectal surgery (n = 206, 30%). The median time to treatment following HSIL diagnosis was 56 days. The remaining 495 (42%) did not return for treatment of HSIL within 6 months of initial diagnosis; the range of time to treatment following HSIL diagnosis for all patients including those who did return within 6 months was 0–517 days. In the treatment group, 174 (25%) underwent surveillance HRA within 18 months of diagnosis of index HSIL, whereas 510 (75%) did not complete a repeat HRA within 18 months of initial HSIL diagnosis and were considered lost to follow-up after treatment.

Fig. 1:
Study flowchart.

The majority of untreated HSIL patients (n = 448, 91%) did not return for surveillance within 18 months of index diagnosis. A small subset (n = 47, 9%) returned for surveillance HRA within 18 months of their initial HSIL diagnosis.

In total, there were eight incident anal cancer diagnoses in the cohort. The majority of cancers (5) occurred in untreated HSIL patients who did not receive a surveillance HRA within 18 months of initial HSIL diagnosis. Three incident anal cancers occurred in patients treated for anal HSIL; all but one cancer arose in those who did not complete a surveillance examination within 18 months of initial HSIL diagnosis. Among all patients found to have anal cancer, the median time from index HSIL to cancer diagnosis was 31.7 months (range 6.6--37.5)

In unadjusted analyses (Table 1), being Black [odds ratio (OR) 0.47, 95% confidence interval (CI) 0.35–0.65), being of other/unknown race/ethnicity (OR 0.69, 95% CI 0.49 –0.98), current cigarette smoking (OR 0.61, 95% CI 0.46 – 0.81), HIV infection without viremia (OR 0.25, 95% CI 0.14 –0.43), and HIV infection with viremia (OR 0.20, 95% CI 0.11–0.36) were all significantly associated with a lower likelihood to return for treatment within 6 months of HSIL diagnosis while identifying as MSM (OR 1.6, 95% CI 1.12–2.28) and having private/commercial (versus public) insurance was associated with greater likelihood of returning for HSIL treatment within 6 months of HSIL diagnosis (OR 2.23, 95% CI 1.73–2.87). In multivariable analyses, Black race [adjusted odds ratio (AOR) 0.61, 95% CI 0.43–0.85), other/unknown race (AOR 0.63, 95% CI 0.43– 0.92) as well as HIV infection without viremia (AOR 0.28, 95% CI 0.16–0.51) and HIV infection with viremia (AOR 0.23, 95% CI 0.12–0.43) were independently associated with not returning for HSIL treatment within 6 months of HSIL diagnosis while private insurance remained a predictor of receiving treatment within 6 months of HSIL diagnosis (AOR 1.83, 95% 1.39–2.41).

Table 1 - New York City (2009–2018) cohort characteristics and proportions treated for high-grade squamous intraepithelial lesion, proportions presenting for surveillance within 18 months of treatment of high-grade squamous intraepithelial lesion.
Outcome: treatment
Characteristic (n = 1179) Treatment (n = 684) No treatment (n = 495) Unadjusted ORa (95% CI) Adjusted ORa (95% CI)
 Mean (SD): 42.3 (11.5) 41.9 (11.4) 42.9 (11.5) 0.87 (0.68–1.12) 0.89 (0.68–1.17)
Sexual practices
 Heterosexuald (n = 136, 11.5%) 65 (47.8%) 71 (52.2%) 1.60 (1.12–2.28) 1.15 (0.78–1.70)
 MSM (n = 1043, 88.5%) 619 (59.4%) 424 (40.7%)
 Whited (n = 425, 36.0%) 278 (65.4%) 147 (34.6%) 1 1
 Black (n = 269, 22.8%) 127 (47.2%) 142 (52.8%) 0.47 (0.35–0.65) 0.61 (0.43–0.85)
 Hispanic (n = 291, 24.7%) 169 (58.1%) 122 (41.9%) 0.73 (0.54–1.00) 0.91 (0.65–1.27)
 Other/unknown (n = 194, 16.5%) 110 (56.7%) 84 (43.3%) 0.69 (049–0.98) 0.63 (0.43– 0.92)
 Neverd (n = 561, 49.1%) 349 (62.2%) 212 (37.8%) 1 1
 Former (n = 264, 23.1%) 157 (59.5%) 107 (40.5%) 0.89 (0.66–1.20) 0.94 (0.68–1.29)
 Current (n = 317, 27.8%) 159 (50.2%) 158 (49.8%) 0.61 (0.46–0.81) 0.76 (0.57–1.02)
 Negatived (n = 101, 8.6%) 85 (84.2%) 16 (15.8%) 1 1
 Positive (n = 756, 64.3%) 430 (56.9%) 326 (43.1%) 0.25 (0.14–0.43) 0.28 (0.57–1.02)
 Positive/viremic (n = 318, 27.1%) 165 (51.9%) 153 (48.1%) 0.20 (0.11–0.36) 0.23 (0.12–0.43)
CD4+ (cells/μl)
 Median 201 202 164
Insurance status
 None or public (n = 763, 64.7%) 392 (51.4%) 371 (48.6%) 2.23 (1.73– 2.87) 1.83 (1.39–2.41)
 Private (n = 416, 35.3%) 292 (70.2%) 124 (29.8%)
Outcome: surveillance after treatment
Characteristic (n = 684) No surveillance w/in 18 months (n = 510) Surveillance w/in 18 months (n = 174) ORc (95% CI) Adjusted ORc (95% CI)
 Mean (SD) 41.9 (11.4) 42.2 (11.2) 41.4 (12.2) 0.96 (0.66–1.38) 1.09 (0.68–1.50)
Sexual practices
 Heterosexualc (n = 65, 9.5%) 48 (73.9%) 17 (26.2%) 1.10 (0.62–1.97) 1.11 (0.59–2.10)
 MSM (n = 619, 90.5%) 445 (71.9%) 159 (28.1%)
 Whitec (n = 278, 40.6%) 204 (73.4%) 74 (26.6%) 1 1
 Black (n = 127, 18.6%) 99 (77.2%) 29 (22.8%) 0.82 (0.49–1.34) 0.90 (0.52–1.54)
 Hispanic (n = 169, 24.7%) 120 (71.0%) 49 (29.0%) 1.13 (0.74–1.72) 1.14 (0.72–1.79)
 Other/unknown (n = 110, 16.1%) 71 (64.6%) 39 (34.5%) 1.51 (0.94–2.43) 1.30 (0.79–2.16)
 Neverc (n = 349, 52.5%) 247 (70.8%) 102 (29.2%) 1 1
 Former (n = 157, 23.6%) 118 (75.2%) 39 (24.8%) 0.80 (0.52–1.23) 0.82 (0.53–1.28)
 Current (n = 159, 23.9%) 116 (73.0%) 43 (27.0%) 0.90 (0.59–1.37) 1.0 (0.65–1.54)
 Negativec (n = 85, 12.5%) 53 (62.4%) 32 (37.7%) 1 1
 Positive (n = 430, 63.2%) 309 (71.9%) 121 (28.1%) 0.65 (0.40–1.06) 0.68 (0.40–1.16)
 Viremic (n = 165, 24.3%) 128 (77.6%) 37 (22.4%) 0.48 (0.27–0.85) 0.52 (0.28–0.96)
CD4+ (cells/μl)
 Median 202 203 201
Insurance status
 None or publicc (n = 392, 57.3%) 287 (73.2.%) 105 (26.8%) 1.14 (0.82–1.60) 1.12 (0.77–1.63)
 Private (n = 292, 42.7%) 206 (70.6%) 86 (29.5%)
CI, confidence interval; OR, odds ratio.
aOdds ratio for HSIL treatment receipt.
bAge less then median compared with age greater then median.
cOdds ratio for surveillance within 18 months of high-grade squamous intraepithelial lesion (HSIL) diagnosis.
dReference category.

For patients treated for HSIL, the only significant predictor of failure of surveillance HRA within 18 months of initial HSIL diagnosis was HIV infection with viremia in both unvariate (OR 0.48, 95% 0.27–0.85) and multivariable analyses (AOR 0.52, 95% CI 0.28–0.96).


In a large cohort at risk for anal cancer, we found that adherence to treatment and surveillance following a histological diagnosis of anal HSIL was poor. Only 55% received treatment within 6 months of HSIL diagnosis and 75% of treated patients did not return for repeat HRA within 18 months of HSIL diagnosis. Among untreated HSIL patients, an astounding 91% never engaged in surveillance within 18 months of diagnosis. Factors negatively impacting retention of anal HSIL patients in care included Black race, HIV with detectable viremia, and low income noted by public insurance. These factors likely identify patients who experience a greater burden of stress and discrimination as well as diminished social and material resources. The clinical and public health implications of our findings are significant as poor adherence to treatment and surveillance of anal HSIL is likely to attenuate potential benefit of screening. These factors may also contribute to health disparities in anal cancer screening, diagnosis, and treatment.

Previous research has reported mixed adherence to treatment and follow-up for anal HSIL. A San Francisco-based study of 246 patients with anal HSIL (79% PWH or with other immunocompromise) found that less than 20% underwent treatment from 1996 to 2006 [23]. A slightly higher number (50 of 155 or 32%) of anal HSIL patients (72% of whom were PWH) from a Boston cohort followed up within 6 months of diagnosis [24]. Another New York-based cohort study of MSM with and without HIV found that although 90% of anal HSIL patients received treatment within 65 days of HRA, 35% were lost to follow-up within 1 year [25]. Our findings add to the heterogeneous patterns of patient engagement in HSIL care and highlight the challenges of engaging patients at elevated risk for anal cancer in appropriate screening and follow-up.

Poor adherence to HSIL surveillance and treatment is of particular concern as 1% of anal HSIL patients in our cohort who received neither treatment or surveillance developed anal cancer. Although three patients diagnosed with HSIL who were initially treated progressed, most cancer diagnoses occurred in patients who did not complete surveillance after treatment. This is consistent with other studies suggesting that active enrollment in an anal cancer screening program may protect against some but not all anal cancers [14,26].

Recently published data show a worrisome trend in anal cancer incidence. Compared with adults born circa 1946, Black men born circa 1986 had a nearly five-fold higher cancer risk [2]. Against this backdrop, our finding that Black HSIL patients are significantly less likely to return for HSIL treatment than their nonblack counterparts is particularly troublesome. The cause of nonadherence to the rigorous visit schedule currently recommended for anal cancer screening is likely multifactorial; additional research to indentify barriers to adherence and interventions to overcome them are needed.

Identified barriers to engagement in anal HSIL follow-up include at the patient, provider, and systems factors. Patient demographics, comorbid burden, beliefs about HPV-related disease or HRA, and stigma have all been identified as barriers to anal cancer screening [27–29]. Provider-level knowledge and expertise, communication skills, and relationship-building with patients have also been described as predictors of adherence to HRA follow-up [27]. Structural and systemic factors also likely influence engagement with HSIL care, including insurance barriers and healthcare system inefficiencies. Studies have also noted that receiving care at an academic medical institution, with difficult scheduling, has impeded HRA follow-up [27,30]. Further research is needed to understand how these factors facilitate or hinder appropriate screening, treatment, and ongoing surveillance for anal precancers.

There are several limitations of this study. Routine clinical and administrative data were used, which may have been irregulary collected. Additionally, we only report on care within our health system and cannot determine if patients sought care elsewhere. Duration of time between HRA and follow-up was likely influenced by several not routinely collected factors, which were therefore absent from this analysis. Finally, missing data on race/ethnicity were determined using a probabilistic methodology, which may have misidentified or oversimplified patients’ race/ethnicity. Major strengths of this study are its large size, nearly 10 years of longitudinal data collection, and the diverse cohort at increased anal cancer risk.

In this study, we found low adherence to treatment and surveillance for anal precancerous lesions among a cohort at high risk for anal cancer. This highlights important health disparities with those at highest risk least likely to receive treatment or posttreatment surveillance. Evidence-based interventions to improve patient participation in and adherence to anal cancer screening are needed.


Funding sources: This project was supported by the National Cancer Institute (R01CA232888).

Conflicts of interest

There are no conflicts of interest.


1. Deshmukh AA, Suk R, Shiels MS, Damgacioglu H, Lin YY, Stier EA, et al. Incidence trends and burden of human papillomavirus-associated cancers among women in the United States, 2001-2017. J Natl Cancer Inst 2020; djaa128doi: 10.1093/jnci/djaa128. Online ahead of print.
2. Deshmukh AA, Suk R, Shiels MS, Sonawane K, Nyitray AG, Liu Y, et al. Recent trends in squamous cell carcinoma of the anus incidence and mortality in the United States, 2001-2015. J Natl Cancer Inst 2020; 112:829–838.
3. Colón-López V, Shiels MS, Machin M, Ortiz AP, Strickler H, Castle PE, et al. Anal cancer risk among people with HIV infection in the United States. J Clin Oncol 2018; 36:68–75.
4. Vanhaesebrouck A, Pernot S, Pavie J, Lucas ML, Collias L, Péré H, et al. Factors associated with anal cancer screening uptake in men who have sex with men living with HIV: a cross-sectional study. Eur J Cancer Prev 2020; 29:1–6.
5. Bouvier A-M, Belot A, Manfredi S, Jooste V S Uhry##Z, Faivre J, et al. French network of cancer registries FRANCIM. Trends of incidence and survival in squamous-cell carcinoma of the anal canal in France: a population-based study. Eur J Cancer Prev 2016; 25:182–187.
6. Wells JS, Holstad MM, Thomas T, Bruner DW. An integrative review of guidelines for anal cancer screening in HIV-infected persons. AIDS patient care and STDs 2014; 28:350–357.
7. Wilkinson J, Morris E, Downing A, Finan PJ, Aravani A, Thomas JD, Sebag-Montefiore D. The rising incidence of anal cancer in E ngland 1990-2010: a population-based study. Colorect Dis 2014; 16:O234–O239.
8. Machalek DA, Poynten M, Jin F, Fairley CK, Farnsworth A, Garland SM, et al. Anal human papillomavirus infection and associated neoplastic lesions in men who have sex with men: a systematic review and meta-analysis. Lancet Oncol 2012; 13:487–500.
9. Shiels MS, Cole SR, Kirk GD, Poole C. A meta-analysis of the incidence of non-AIDS cancers in HIV-infected individuals. J Acquir Immune Defic Syndr 2009; 52:611–622.
10. Clifford GM, Georges D, Shiels MS, Engels EA, Albuquerque A, Poynten IM, et al. A meta-analysis of anal cancer incidence by risk group: Toward a unified anal cancer risk scale. Int J Cancer 2021; 148:38–47.
11. van der Zee RP, Richel O, De Vries H, Prins JM. The increasing incidence of anal cancer: can it be explained by trends in risk groups. Neth J Med 2013; 71:401–411.
12. Fontham ET, Wolf AM, Church TR, Etzioni R, Flowers CR, Herzig A, et al. Cervical cancer screening for individuals at average risk: 2020 guideline update from the American Cancer Society. CA Cancer J Clin 2020; 70:321–346.
13. Palefsky JM. Screening to prevent anal cancer: current thinking and future directions. Cancer Cytopathol 2015; 123:509–510.
14. Revollo B, Videla S, Llibre JM, Paredes R, Piñol M, García-Cuyàs F, et al. Routine screening of anal cytology in persons with human immunodeficiency virus and the impact on invasive anal cancer: a prospective cohort study. Clin Infect Dis 2019; 71:390–399.
15. Gaisa MM, Liu Y, Deshmukh AA, Stone KL, Sigel KM. Electrocautery ablation of anal high-grade squamous intraepithelial lesions: Effectiveness and key factors associated with outcomes. Cancer 2020; 126:1470–1479.
16. Machalek DA, Jin F, Poynten IM, Hillman RJ, Templeton DJ, Law C, et al. SPANC study team. Prevalence and risk factors associated with high-grade anal squamous intraepithelial lesions (HSIL)-AIN2 and HSIL-AIN3 in homosexual men. Papillomavirus Res 2016; 2:97–105.
17. Goldstone SE, Johnstone AA, Moshier EL. Long-term outcome of ablation of anal high-grade squamous intraepithelial lesions: recurrence and incidence of cancer. Dis Colon Rectum 2014; 57:316–323.
18. Silvera RJ, Smith CK, Swedish KA, Goldstone SE. Anal condyloma treatment and recurrence in HIV-negative men who have sex with men. Dis Colon Rectum 2014; 57:752–761.
19. Stier EA, Abbasi W, Agyemang AF, Valle Alvarez EA, Chiao EY, Deshmukh AA. Brief report: recurrence of anal high-grade squamous intraepithelial lesions among women living with HIV. J Acquir Immune Defic Syndr 2020; 84:66–69.
20. Long KC, Menon R, Bastawrous A, Billingham R. Screening, surveillance, and treatment of anal intraepithelial neoplasia. Clin Colon Rectal Surg 2016; 29:57–64.
21. Gaisa M, Sigel K, Hand J, Goldstone S. High rates of anal dysplasia in HIV-infected men who have sex with men, women, and heterosexual men. AIDS 2014; 28:215–222.
22. Fiscella K, Fremont AM. Use of geocoding and surname analysis to estimate race and ethnicity. Health Serv Res 2006; 41 (4 pt 1):1482–1500.
23. Pineda CE, Berry JM, Jay N, Palefsky JM, Welton ML. High-resolution anoscopy targeted surgical destruction of anal high-grade squamous intraepithelial lesions: a ten-year experience. Dis Colon Rectum 2008; 51:829–837.
24. Apaydin KZ, Nguyen A, Borba CP, Shtasel DL, Ulery S, Mayer KH, Keuroghlian AS. Factors associated with anal cancer screening follow-up by high-resolution anoscopy. Sex Transm Infect 2019; 95:83–86.
25. Goldstone RN, Goldstone AB, Russ J, Goldstone SE. Long-term follow-up of infrared coagulator ablation of anal high-grade dysplasia in men who have sex with men. Dis Colon Rectum 2011; 54:1284–1292.
26. Arens Y, Gaisa M, Goldstone S, Liu Y, Wisnivesky J, Sigel CS, et al. Risk of invasive anal cancer in HIV infected patients with high grade anal dysplasia: a population-based cohort study. Dis Colon Rectum 2019; 62:934.
27. Apaydin KZ, Nguyen A, Panther L, Shtasel DL, Dale SK, Borba CPC, et al. Facilitators of and barriers to high-resolution anoscopy adherence among men who have sex with men: a qualitative study. Sex Health 2018; 15:431–440.
28. Schiphorst AH, Verweij NM, Pronk A, Hamaker ME. Age-related guideline adherence and outcome in low rectal cancer. Dis Colon Rectum 2014; 57:967–975.
29. Newman PA, Roberts KJ, Masongsong E, Wiley D. Anal cancer screening: barriers and facilitators among ethnically diverse gay, bisexual, transgender, and other men who have sex with men. J Gay Lesbian Soc Serv 2008; 20:328–353.
30. Kole AJ, Stahl JM, Park HS, Khan SA, Johung KL. Predictors of nonadherence to NCCN guideline recommendations for the management of stage I anal canal cancer. J Natl Compr Canc Netw 2017; 15:355–362.

adherence; anal cancer; high-resolution anoscopy; human papillomavirus; screening

Copyright © 2021 Wolters Kluwer Health, Inc. All rights reserved.