See editorial on page 1339.
Anal intraepithelial neoplasia (AIN) III, a dysplastic condition of squamous tissue that develops from contact with oncogenic human papillomavirus (HPV) strains, is considered a premalignant stage of anal squamous cell carcinoma (SCC).1 Although progression of AIN III to anal SCC parallels the pathway of cervical dysplasia to cervical cancer, AIN III is thought to be more persistent, rarely regressing spontaneously.2,3 Human papillomavirus (HPV) infections tend to be asymptomatic, making surveillance for AIN and anal cancer critical.1
Adding urgency to the discussions on anal cancer surveillance is the concerning statistic that rates of anal SCC are increasing by ≈2% per year,4 and incidence of AIN is increasing at an even higher rate, estimated at 11% per year in a population-based cohort from San Francisco, California.5 Despite the increasing prevalence of disease, the risk of anal SCC after a previous diagnosis of AIN III is unclear. Most of the literature include single-institution studies that cite a 0.4% to 2.1% per-year risk of anal cancer after a diagnosis of AIN III,6–12 although 1 small study estimated a 6.1% annual risk of anal cancer in HIV-positive men who have sex with men with high-grade AIN.13 These studies are difficult to interpret, because sample sizes were small and patients varied in terms of HIV status, sexual practices, degree of dysplasia of initial AIN, and AIN treatment.
These limited data have led to uncertainty about the true risk of progression to anal cancer and the appropriateness of surveillance for anal dysplasia. A recent survey of colorectal surgeons found that only 48% screen for anal dysplasia, despite knowingly managing high-risk patients in their practices.14 Although high-resolution anoscopy is effective and recommended for AIN screening,6,15 most colorectal surgeons are not formally trained, and there is substantial variability in technique.14 Therefore, the primary aim of this study was to quantify the risk of subsequent anal SCC diagnosis in patients with previous AIN III using a national data set. The secondary aim was to identify predictors for malignant transformation from AIN III to anal SCC.
PATIENTS AND METHODS
Study Design
We conducted a retrospective study of a cohort of patients with AIN III diagnosed between 1973 and 2014 in the Surveillance, Epidemiology, and End Results (SEER) Program SEER*Stat Database: Incidence – SEER 18 Regs Research Data + Hurricane Katrina Impacted Louisiana Cases (November 2016 submission). The SEER registry, organized by the National Cancer Institute, collects cancer incidence, prevalence, and survival data from population-based cancer registries covering ≈28% of the US population. SEER*Stat version 8.3.4 was used to link separate tumors based on patient identification number.
For patients with multiple diagnoses of AIN III, the first recorded diagnosis was identified as the index diagnosis for analysis. Likewise, for patients with multiple diagnoses of anal cancer, the first diagnosis was chosen as the index case. Any tumors diagnosed within ≤2 months were considered synchronous. Patients who had a diagnosis of anal cancer before an AIN III diagnosis were excluded from the analysis. Our Institutional Review Board approved this study (protocol No. 2017P001718).
Variable Definitions
Basic demographic characteristics were obtained, including age, sex, race, marital status, and US region (based on US Census Bureau designations). Surgery type for initial AIN III was categorized as follows: no procedure, ablative surgery (thermal ablation, electrocautery, fulguration, local tumor destruction, photodynamic therapy, or laser), excisional surgery (local tumor excision, polypectomy, excisional biopsy, or laser excision), and surgery not otherwise specified. The rationale for why a patient did not receive surgery was also recorded. Tumor size was defined as the largest dimension of the primary tumor (in centimeters). Squamous cell anal cancers were assigned T stage by tumor size (tumors ≤2 cm were T1, tumors >2 cm and ≤5 cm were T2, and tumors >5 cm were T3), based on the American Joint Commission on Cancer 8th edition.
Diagnoses of AIN III and anal SCC were confirmed based on a primary site of anus, anal canal, and anorectum and specific International Classification of Diseases for Oncology, Third Edition, codes based on histopathologic specimens (Appendix, Supplemental Digital Content, https://links.lww.com/DCR/A793). Histology codes of SCC in situ were included as diagnoses of AIN III. HPV-related tumors were defined as any anal, cervical, oropharyngeal (tongue, tonsil, or oropharynx), penile, vaginal, or vulvar neoplasm. AIDS-defining malignancies were defined as Kaposi sarcoma, non-Hodgkin’s lymphoma, and cervical cancer. Tumors could be recorded as both HPV-related and AIDS-defining malignancies. The primary outcomes were subsequent anal cancer diagnosis and time (years) between initial AIN III diagnosis and subsequent anal cancer diagnosis.
Statistical Analysis
Variables were summarized as median (interquartile range (IQR)) or count (percentage, 95% CI) as appropriate. Categorical variables were compared with the Fisher exact test, continuous variables were compared with the 2-sample t test, and multivariable logistic regression was used to adjust for potential confounders. Time-to-event analysis was performed with Kaplan–Meier curves and multivariable Cox proportional hazards models. All of the statistical analyses were performed using Stata software, version SE 14.0 (Stata Corp, College Station, TX). All of the tests were 2 sided, and statistical significance was accepted at the p < 0.05 level.
RESULTS
Overall Cohort
We identified 2129 patients with a diagnosis of AIN III between 1973 and 2014. Of those, 55 patients had a diagnosis of anal cancer before AIN III diagnosis and were therefore excluded from the analysis. Thus, our final cohort included 2074 patients.
Of the cohort, 60.3% were men and 84.3% were white, with a median age of 52 years (IQR, 44–61 y; Table 1). In terms of marital status, 47.2% were single, 22.8% were married or with a domestic partner, and 13.0% were divorced, separated, or widowed. For their initial AIN III, 14.3% of patients underwent ablative surgery, 55.5% underwent excisional surgery, and 28.5% had no procedures. The median follow-up time was 4.0 years (IQR, 1.8–6.7 y).
;)
TABLE 1.: Demographics of 2074 patients with diagnosis of AIN III
Other Primary Malignancies
Of the overall AIN III cohort, 935 patients (45.1% (95% CI, 43.0%–47.2%)) had a primary malignancy before their first AIN III diagnosis, whereas 912 patients (44.0% (95% CI, 41.9%–46.1%)) had a primary malignancy afterward (Table 2). Approximately 323 patients (15.6% (95% CI, 14.1%–17.2%)) went on to develop an additional HPV-related neoplasm, including 171 (8.2% (95% CI, 7.1%–9.5%)) who went on to develop anal SCC and 56 (2.7% (95% CI, 2.1%–3.5%)) who went on to develop an AIDS-defining malignancy. In addition, 317 patients (15.3% (95% CI, 13.8%–16.9%)) developed recurrent AIN III.
TABLE 2.: Patients with AIN III who were diagnosed with other primary neoplasms
Of the 171 patients who were diagnosed with anal SCC after an initial diagnosis of AIN III, 73.1% were men, 80.7% were white, 53.2% were single, and the median age was 50 years (IQR, 44–60 y). For their initial AIN III treatment, 3.5% underwent ablative surgery, 71.4% underwent excisional surgery, and 22.8% had no procedure. Median time from AIN III diagnosis to anal SCC diagnosis was 2.7 years (IQR, 1.1–4.5 y). At diagnosis, 32.9% of anal SCCs were stage T1, 22.4% were stage T2, 8.2% were stage T3, and the stage was unknown or not recorded for 36.5% (Table 3). By the end of the follow-up period, 14.3% of patients with T1 anal cancers, 21.1% of those with T2 tumors, and 42.9% of those with T3 tumors had died (p = 0.06).
TABLE 3.: Demographics of 171 patients who developed anal squamous cell carcinoma after previous diagnosis of AIN III
Predictors of Subsequent Diagnosis of Anal SCC
Comparing the demographics of patients who were or were not subsequently diagnosed with anal SCC revealed significant differences on univariate analysis (Table 4). Patients who were subsequently diagnosed with anal SCC tended to be 41 to 50 years old (40.9% vs 29.3%; p = 0.02) and men (73.1% vs 59.1%; p < 0.001). In terms of marital status, more single people (53.2% vs 46.6%) tended to be diagnosed with anal cancer, as opposed to divorced people (5.9% vs 13.7%; p = 0.02). There were no differences in race or region of the United States. Patients who were subsequently diagnosed with anal cancer tended to undergo excisional surgery for their initial AIN III (71.4% vs 54.0%; p < 0.001), not ablative surgery (3.5% vs 15.2%; p < 0.001), as compared with patients who did not develop anal cancer.
TABLE 4.: Univariate and multivariable analyses of predictors of developing subsequent anal squamous cell carcinoma in patients with previous AIN III
These trends remained significant on multivariable analysis (Table 4). Independent predictors of subsequent diagnosis of anal carcinoma in patients with previous AIN III included being 41 to 50 years old (OR = 1.69 (95% CI, 1.01–2.83); p = 0.047) and male (OR = 2.18 (95% CI, 1.47–3.24); p < 0.001). Patients who were divorced, separated, or widowed were less likely to be diagnosed with anal SCC (OR = 0.45 (95% CI, 0.22–0.93); p = 0.03) than single patients. In terms of treatment for AIN III, patients who underwent excisional surgery were more likely to be diagnosed with anal SCC (OR = 1.82 (95% CI, 1.23–2.70); p = 0.003), whereas ablative surgery was associated with decreased likelihood of a subsequent diagnosis of anal cancer (OR = 0.27 (95% CI, 0.11–0.66); p = 0.004).
Factors Associated With Shorter Time Between AIN III and Anal SCC Diagnoses
In time-to-event analysis of the overall cohort, we examined anal SCC–free survival, with subsequent anal cancer diagnosis as the primary end point. We found that 9.5% of patients developed anal SCC within 5 years of AIN III diagnosis (Fig. 1A). Separating patients by treatment of initial AIN III, 2.0% of patients who underwent ablation, 8.0% of patients who received no procedure, and 12.2% of patients who underwent excision were diagnosed with anal cancer within 5 years (p < 0.001; Fig. 1B). There was no significant correlation between time from AIN III diagnosis to anal cancer diagnosis and T stage of the subsequent anal cancer. The median times from AIN III to anal cancer diagnosis were 2.8 (IQR, 1.0–4.8 y), 3.1 (IQR, 1.1–4.4 y), and 3.4 years (IQR, 1.2–4.6 y) for T1, T2, and T3 cancers (p > 0.05).
FIGURE 1.: Kaplan–Meier curves depicting anal squamous cell carcinoma–free survival in patients with previous anal intraepithelial neoplasia (AIN) III in (A) the overall cohort and (B) separated by type of surgical intervention for previous AIN III.
Multivariable time-to-event analysis revealed that age 41 to 50 years old (HR = 1.74 (95% CI, 1.05–2.87); p = 0.03) and male sex (HR = 1.84 (95% CI, 1.26–2.71); p = 0.002) were associated with earlier time to development of anal cancer (Table 5). Patients who were divorced, separated, or widowed tended to develop anal cancer later than single patients (HR = 0.49 (95% CI, 0.24–0.98); p = 0.04). Race and region of the United States were not associated with time to development of anal cancer. Compared with no procedure for initial AIN III, ablations were associated with significantly later development of anal cancer (HR = 0.31 (95% CI, 0.13–0.74); p = 0.01), whereas excisions were associated with significantly earlier development of anal carcinoma (HR = 1.51 (95% CI, 1.04–2.19); p = 0.03).
TABLE 5.: Multivariable time-to-event analysis of predictors of early development of subsequent anal squamous cell carcinoma in patients with previous AIN III
DISCUSSION
This study represents the largest published cohort of patients with AIN III. In this group of 2074 patients, the 5-year incidence of anal SCC after previous AIN III diagnosis was 9.5% or ≈1.9% per year. This cohort, composed of patients drawn from population-based national registries, is likely more representative of the true risk of malignancy than previous single-institution studies, most of which involved small groups of patients who received variable treatments.6–12
Undergoing ablative surgery for AIN III was significantly associated with decreased risk of subsequent anal cancer, whereas excisional procedures for AIN III were significantly associated with increased risk of subsequent anal SCC. This difference may have been seen because surgeons excised larger, visible AIN lesions that may have had more malignant potential. Although high-resolution anoscopy is not recorded in SEER, it is also possible that surgeons who performed ablations also used high-resolution anoscopy more often and thus obtained better visualization of the areas of anal dysplasia, therefore enabling more complete eradication of disease.16 It is also possible that patients who underwent ablations were followed more closely with surveillance examinations than those whose lesions were excised and that the correlation between treatment type and rate of subsequent anal cancer was attributed in part to surveillance regimen; however, it is impossible to ascertain with this data set.
AIN is becoming widely recognized as a multifocal disease of the anal squamocolumnar junction. In patients at high risk for anal dysplasia, random biopsies of the anal canal in areas thought to be free of disease have later been found to have high-grade dysplasia.17 Furthermore, in a small prospective trial, patients with high-grade dysplasia who underwent circumferential ablation of the anal squamocolumnar junction had no evidence of recurrent or persistent disease after 2 treatments, suggesting that AIN often occurs at multiple locations throughout the anal canal, not just in visible abnormalities on screening examinations.18 Our findings support this idea that AIN is multifocal, therefore simply excising visible or palpable lesions is inadequate for removing all disease and preventing progression to anal cancer. Patients who underwent excision for AIN III may have had remaining disease that went unidentified and therefore untreated, which may have then progressed to cancer.
This study also found that patients who had no procedure for AIN III were less likely to develop subsequent anal SCC than patients whose AIN III was excised. This difference appears more pronounced as more time lapses after AIN III diagnosis (Fig. 1B). This interesting finding is likely attributed to a combination of factors, including AIN III lesion size (patients with larger lesions were probably more likely to be excised) and surveillance schedule (patients with no procedure for AIN III may have been more closely followed, whereas patients who were excised may have only returned to care after developing symptoms). Other potential contributing factors include patients moving to states not captured by the SEER registry, physicians using topical treatments not recorded in the database, differences in immunosuppressed status, and spontaneous regression of anal dysplasia. Nevertheless, compared with patients with no recorded procedure for AIN III, patients who underwent ablations were significantly less likely to develop anal cancer. These differences again emphasize that AIN is a multifocal disease, requiring close surveillance for recurrence or progression and that simply excising a visible lesion is unlikely to completely eradicate the disease.
In this study, other independent predictors of a subsequent diagnosis of anal cancer were male sex and single status. This likely represents the subset of patients who were men who have sex with men, who are significantly more likely to have anal-receptive intercourse, HIV infection,19 and a high number of sexual partners.20 Also, men have been shown to be less compliant with follow-up appointments, making it possible that they missed more screening opportunities.21
Of note, despite a previous diagnosis of AIN III, one third of anal cancers were diagnosed at stage T2 or higher. This demonstrates the importance of emphasizing close surveillance with both providers and patients. An interesting survey of members of the American Society of Colon and Rectal Surgeons found that, although nearly all responders recognized risk factors for AIN and 92% treated patients with anal dysplasia, less than half conducted anal dysplasia screening.14 Only 31% of those surgeons used acetic acid with some form of magnification to evaluate for anal dysplasia, and only 35% performed high-resolution anoscopy, of whom 46% were formally trained. Of the providers who did not screen, 52% stated that they were never formally trained, whereas 23% did not think it was a priority. In light of these data demonstrating a ≈10% incidence of anal cancer within 5 years of AIN III diagnosis and a protective association with ablation of AIN III lesions, colorectal surgeons must not only make ablation a key part of their training and practice but must also maintain close surveillance of their patients with AIN III, regardless of the treatment performed.
Stronger screening guidelines for patients at high risk of anal cancer are needed to ensure a standardization of the most appropriate practices.22,23 In our cohort, nearly one third of anal cancers were diagnosed at stage T2 or higher, suggesting that anal cancers were missed at early stages. The Practice Parameters for Anal Squamous Neoplasms published by Steele et al24 in 2012 lay an excellent foundation, and several prospective trials examining screening strategies are underway, including the Study of the Prevention of Anal Cancer and the Anal Cancer High-Grade Squamous Intraepithelial Lesion Outcomes Research study.25 Strong guidelines from colorectal surgery societies would increase awareness among patients and providers about the high incidence of malignancy and the need for standardized surveillance in patients with AIN III.
This study has a number of limitations. The SEER registry does not record HIV status, sexual preference, number of sexual partners, or immunosuppressed status. It also does not document the use of high-resolution anoscopy, nonprocedural treatments for AIN, frequency of surveillance, number of treatments, how the anal lesion was detected, or specialty of the provider performing the procedure for AIN. We were also unable to classify lesions as high- or low-grade squamous intraepithelial lesions, which have become the new consensus terminology for classifying anal dysplasia, instead using the former terminology of AIN III. In addition, although it is possible to link neoplasms in the same patient, it is impossible to determine whether an AIN III lesion progressed to a subsequent anal cancer. Furthermore, only patients with a neoplasm diagnosed between 2004 and 2014 are entered into this registry, and their previous neoplasms are then recorded, leading to a potential for selection bias in this study. It is also possible that the SEER registry is not fully representative of the US population, because it captures data from specific states and intentionally overrepresents ethnic minorities and underserved populations. However, the population-based estimates provided by SEER are more likely to be generalizable to the practicing clinician across the country than single-institution studies.
CONCLUSION
This study highlights the high rate of malignancy in patients with AIN III and the need for effective therapies and close surveillance. Nearly 10% of patients with AIN III are diagnosed with anal cancer within 5 years, and nearly one third of those anal cancers are diagnosed at stage T2 or higher. There is an urgent need for prospective trials to identify the most effective treatments for AIN III and provide the basis for stronger guidelines for AIN III and anal cancer surveillance and management.
REFERENCES
1. Ricciardi R. Steele SR, Hull TL, Read TE, Saclarides TJ, Senagore AJ, Whitlow CB. Anal intraepithelial neoplasia. In: The ASCRS Textbook of Colon and Rectal Surgery. 2016:1. 3 ed. Arlington Heights, IL: Springer; 343353.
2. Palefsky JM, Holly EA, Ralston ML, et al. Anal squamous intraepithelial lesions in HIV-positive and HIV-negative homosexual and bisexual men: prevalence and risk factors. J Acquir Immune Defic Syndr Hum Retrovirol. 1998;17:320326.
3. Saleem AM, Paulus JK, Shapter AP, Baxter NN, Roberts PL, Ricciardi R. Risk of anal cancer in a cohort with human papillomavirus-related gynecologic neoplasm. Obstet Gynecol. 2011;117:643649.
4. American Cancer Society. Cancer Facts & Figures 2017. Special Section: Rare Cancers in Adults. 2017.Atlanta, GA: American Cancer Society;
5. Simard EP, Watson M, Saraiya M, Clarke CA, Palefsky JM, Jemal A. Trends in the occurrence of high-grade anal intraepithelial neoplasia in San Francisco: 2000-2009. Cancer. 2013;119:35393545.
6. 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;57:316323.
7. Fazendin EA, Crean AJ, Fazendin JM, et al. Condyloma acuminatum, anal intraepithelial neoplasia, and anal cancer in the setting of HIV: do we really understand the risk? Dis Colon Rectum. 2017;60:10781082.
8. Sobhani I, Walker F, Roudot-Thoraval F, et al. Anal carcinoma: incidence and effect of cumulative infections. AIDS. 2004;18:15611569.
9. Crawshaw BP, Russ AJ, Stein SL, et al. High-resolution anoscopy or expectant management for anal intraepithelial neoplasia for the prevention of anal cancer: is there really a difference? Dis Colon Rectum. 2015;58:5359.
10. 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;51:829835.
11. Weis SE, Vecino I, Pogoda JM, Susa JS. Treatment of high-grade anal intraepithelial neoplasia with infrared coagulation in a primary care population of HIV-infected men and women. Dis Colon Rectum. 2012;55:12361243.
12. Tong WW, Jin F, McHugh LC, et al. Progression to and spontaneous regression of high-grade anal squamous intraepithelial lesions in HIV-infected and uninfected men. AIDS. 2013;27:22332243.
13. Tinmouth J, Peeva V, Amare H, et al. Progression from perianal high-grade anal intraepithelial neoplasia to anal cancer in HIV-positive men who have sex with men. Dis Colon Rectum. 2016;59:836842.
14. Factor SH, Cooperstein A, Pereira GA, Goldstone SE. Are colon and rectal surgeons ready to screen for anal dysplasia? Results of a survey on attitudes and practice. Sex Transm Dis. 2014;41:246253.
15. Hillman RJ, Cuming T, Darragh T, et al. 2016 IANS International Guidelines for Practice Standards in the Detection of Anal Cancer Precursors. J Low Genit Tract Dis. 2016;20:283291.
16. Swedish KA, Lee EQ, Goldstone SE. The changing picture of high-grade anal intraepithelial neoplasia in men who have sex with men: the effects of 10 years of experience performing high-resolution anoscopy. Dis Colon Rectum. 2011;54:10031007.
17. Silvera R, Gaisa MM, Goldstone SE. Random biopsy during high-resolution anoscopy increases diagnosis of anal high-grade squamous intraepithelial lesions. J Acquir Immune Defic Syndr. 2014;65:6571.
18. Goldstone RN, Hasan SR, Goldstone SE. Brief report: radiofrequency ablation therapy for anal intraepithelial neoplasia: results from a single-center prospective pilot study in HIV+ participants. J Acquir Immune Defic Syndr. 2017;76:e93e97.
19. Centers for Disease Control and Prevention (CDC). HIV infections attributed to male-to-male sexual contact: metropolitan statistical areas, United States and Puerto Rico, 2010. MMWR Morb Mortal Wkly Rep. 2012;61:962966.
20. Glick SN, Morris M, Foxman B, et al. A comparison of sexual behavior patterns among men who have sex with men and heterosexual men and women. J Acquir Immune Defic Syndr. 2012;60:8390.
21. Zelle BA, Buttacavoli FA, Shroff JB, Stirton JB. Loss of follow-up in orthopaedic trauma: who is getting lost to follow-up? J Orthop Trauma. 2015;29:510515.
22. Leeds IL, Fang SH. Anal cancer and intraepithelial neoplasia screening: a review. World J Gastrointest Surg. 2016;8:4151.
23. Ong JJ, Chen M, Grulich AE, Fairley CK. Regional and national guideline recommendations for digital ano-rectal examination as a means for anal cancer screening in HIV positive men who have sex with men: a systematic review. BMC Cancer. 2014;14:557.
24. Steele SR, Varma MG, Melton GB, Ross HM, Rafferty JF, Buie WD; Standards Practice Task Force of the American Society of Colon and Rectal Surgeons. Practice parameters for anal squamous neoplasms. Dis Colon Rectum. 2012;55:735749.
25. Machalek DA, Grulich AE, Hillman RJ, et al.; SPANC Study Team. The Study of the Prevention of Anal Cancer (SPANC): design and methods of a three-year prospective cohort study. BMC Public Health. 2013;13:946.
