Authors
E. GÖKCE (1), J. VAN DORPE (2), A. HOORENS (2), J. PELGROM (3), F. VAN WANZEELE (3), L. VANDEKERCKHOVE (3), S. CALLENS (3), D. DE LOOZE (4) / [1] Ghent University, Ghent, Belgium, Faculty of Medicine and Health Sciences, [2] Ghent University Hospital, Ghent, Belgium, Pathology, [3] Ghent University Hospital, Ghent, Belgium, Internal Medicine, [4] Ghent University Hospital, Ghent, Belgium, Gastroenterology
Introduction
Although anal cancer is rare in the overall population, its incidence is increasing in the last decades. Especially HIV-positive men are at increased risk for developing anal squamous cell carcinoma (SCC), mainly because of the high prevalence of high-grade anal intraepithelial neoplasia (HGAIN; AIN2/3) among those patients. No studies have examined the prevalence and the natural history of anal intraepithelial neoplasia (AIN) in HIV-positive men in Belgium.
Aim
Our study aims to identify the prevalence of AIN in HIV-patients, who underwent anal screening in the Ghent University Hospital. Furthermore, we tried to determine the natural history of HGAIN. In addition, we compared the results of the cytology to those obtained from high-resolution anoscopy (HRA).
Methods
Forty-eight HIV-positive men [95,8% men who have sex with men (MSM)], who underwent ≥2 screening procedures for anal dysplasia, were enrolled in this study. All patients received digital rectal examination, anal cytology and HRA at each visit. Biopsies were taken if suspicious lesions were revealed by HRA. All follow-up visits were included, giving a total of 166 cytology samples and HRA procedures performed between 18/08/2010 – 15/10/2019. The swab samples have been grouped in ‘HSIL’ [high-grade squamous intra-epithelial lesions (HSIL) or atypical squamous cells-cannot exclude HSIL (ASC-H)] and ‘non HSIL’ [Negative for Intraepithelial Lesion or Malignancy (NILM), atypical squamous cells of undetermined significance (ASC-US) or low-grade squamous intra-epithelial lesions (LSIL)].
Results
During a median follow-up period of 44 months the median number of screening procedures was 3 per patient (median age 51 years). 160 of 166 cytology samples were suitable for analysis, 47 (29.4%) of which showed ‘HSIL’ while the remaining 113 (70.6%) swabs showed ‘non HSIL’. Among the participants: the prevalence of HSIL was 50% on the first visit, 23.3% on the second and 16.7% on the third. We identified a regression of 30.2% from ‘HSIL’ to ‘non HSIL’ between the first and second visit (p= 0.001); a regression of 36.7% between the first and third visit (p < 0,001); and a regression of 24.0% between the second and third visit (p = 0,388). Only two patients received coagulation therapy for intra-anal warts; the other patients regressed spontaneously. Throughout the entire follow-up duration, we found one patient, who progressed to carcinoma in situ. Furthermore, a statistically significant association is found between detecting clinically suspicious lesions on HRA and having an anal swab that shows ‘HSIL’. Anoscopic findings suggestive of dysplasia were found during 50 of 160 (31.1%) HRA procedures. If a diagnosis of ‘HSIL’ was detected on cytology, the prevalence of visualizing a suspicious lesion on concurrent HRA was 53.2% (24 of 47), while this percentage was 22.1% (25 of 113) when ‘non HSIL’ was found on anal cytology.
Conclusions
We found a high percentage of spontaneous regression from ‘HSIL’ to ‘non HSIL’ among HIV-positive participants. Further studies with a larger study population and longer median follow-up duration are required to determine the sustainability of these regressions and to generalize our findings to the entire HIV-population. Despite the high regression grade, one patient progressed to carcinoma in situ. This shows the importance of regular screening to prevent invasive SCC. Based on the findings of new studies that are trying to determine biomarkers and risk factors that predict whether HSIL will regress or progress, it may be possible in the future to create a screening program wherein patients with an increased risk of progression can be monitored closely, while patients with a high probability of regression can be screened at larger intervals.
