The Treatment of Anal Squamous Cell Carcinoma in a High HIV Prevalence Population

Anal Squamous Cell Carcinoma (ASCC) is a rare cancer that has a rapidly increasing incidence in areas with highly developed economies. ASCC is strongly associated with HIV and there appears to be increasing numbers of younger male persons living with HIV (PLWH) diagnosed with ASCC in Greater London. This is a retrospective cohort study of HIV positive and HIV negative patients diagnosed with primary ASCC between January 2000 and January 2020 in a demographic group with high prevalence rates of HIV. 176 patients were included, and clinical data was retrieved from multiple, prospective databases. A clinical subgroup was identied in this cohort of younger HIV positive males who were more likely to have had a prior diagnosis of Anal Intraepithelial Neoplasia (AIN). Gender and HIV status had no effect on staging or disease-free survival. PLWH were more likely to develop a recurrence (p < 0.000) but had a longer time to recurrence than HIV negative patients, however this was not statistically signicant (46.1 months vs. 17.5 months; p = 0.077). Patients known to have a previous diagnosis of AIN were more likely to have earlier staging and local tumour excision but there was no identiable survival benet in this cohort.


Introduction
Anal Squamous Cell Carcinoma (ASCC) is considered a rare cancer, with an incidence rate that is frequently quoted as between 1-2 cases per 100000 per year worldwide [1]. Nevertheless, the incidence of ASCC is rapidly increasing; Islami et al (2017) demonstrated a percentage incidence increase of between 2.8-6.8% per year in men and 1.5-10% per year in woman in Western developed countries between 1971-2007. [2] The known risk factors for the development of ASCC include immunosuppression, smoking, receptive anal intercourse, sexually transmitted infections, previous or current Human Papillomavirus (HPV) related dysplasia and malignancies and Human Immunode ciency Virus (HIV). [3] HIV is a signi cant risk factor in the development of ASCC, and ASCC is the third most common cancer in people living with HIV (PLWH). [4] It is likely that that the increasing incidence of ASCC related to the higher number of PLWH in developed countries where, in the era of Highly Active Antiretroviral Therapy (HAART), PLWH are able to achieve a near normal life expectancy and go on to develop long-term sequelae of HIV infection. [5] Despite improving outcomes of HIV prognosis with HAART, the use of HAART does not reduce the risk of ASCC in PLWH, [4] therefore it is likely that the incidence of ASCC in PLWH will continue to rise. Page 3/17 ASCC is associated with persistent oncogenic Human Papillomavirus (HPV) infections and has a known HPV related dysplastic precursor; Anal Intraepithelial Neoplasia (AIN). With a similar natural history to Cervical Intraepithelial Neoplasia (CIN), it hypothesised that like cervical cancer, ASCC could become a largely preventable cancer with HPV vaccination, ASCC screening programmes and appropriate treatment for AIN when identi ed.
Nevertheless, the screening for, and treatment of, AIN remains controversial in clinical guidelines. [6,7] Although clinical guidelines identify PLWH as a high-risk group for ASCC, there is signi cant discrepancy in the literature regarding the progression rates of high-grade AIN to ASCC [8][9][10][11] and, unlike cervical cancer screening programmes which have a good evidence base in preventing malignancy, corresponding ASCC screening programmes have not been shown to be as effective. There is also a lack of evidence that treating AIN once it has been identi ed can prevent ASCC. [12] Despite the controversy, many HIV centres provide local screening services for PLWH. In our clinical centre, a high resolution anoscopy and cytology screening programme is offered to PLWH and in particular men who have receptive anal intercourse (MSM).
Historically, the majority of ASCC diagnosed are in women in their 6th or 7th decade who are likely to have had a previous high-risk HPV infection and the published literature focuses mainly on this patient demographic. [7,13,14] Although, research involving the screening and treatment of AIN is centred around the newly increasing PLWH demographic, there is less clinical evidence about the outcomes of the PLWH subgroup once they are diagnosed with ASCC.
We present our experience with managing patients with ASCC over two decades between 2000 to 2020 with emphasis on whether there exists a separate subpopulation of ASCC patients with HIV within the Trust and whether this affects clinical prognosis.

Methods
We undertook a retrospective cohort study following the "strengthening the reporting of observational studies in epidemiology (STROBE)" statement. [15] Consecutive patients undergoing treatment for Anal Squamous Cell Carcinoma between January 2000 and January 2020 were identi ed from prospectively maintained clinical databases. The departments of sexual health, surgery and HIV kept separate clinical databases across two clinical sites. Data entries were linked by hospital number and date of birth and then anonymised for analysis. The study protocol was reviewed and approved by the London-Westminster Research and Ethics Committee prior to data collection.

Inclusion Criteria
The study included all adult patients over 18 years of age diagnosed with primary histological proven Anal Squamous Cell Carcinoma (ASCC) between January 2000 and January 2020.

Exclusion Criteria
Patients under 18 years of age. Anal cancers with a histopathology other than Squamous Cell Carcinoma were excluded from this analysis.

Data Collection
For each patient included in the study, the authors retrieved data including demographics, HIV outcomes, attendance at screening, histopathology, staging, and treatment received (Table 1) from separate clinical databases and combined it to form one data record per patient. Staging and treatment were commonly repeated across clinical databases, and any discrepancies were independently checked by two independent clinical reviewers (DRLB and CCK).

Gender
Men were more likely to have HIV (p < 0.001, Chi Squared), and to have had previous treatment for AIN (p < 0.001, Fisher's Exact Test). A higher proportion of male patients presented with an early cancer (Stage 1 and 2: 56% compared to 47%) but this was not statistically signi cant (Fig. 2). There was no difference in gender proportions when comparing treatment modalities, recurrence rates and disease-free survival.  patients underwent an Abdominoperineal Resection as salvage surgery after their recurrence, no immediate post-operative complications were identi ed. There was no statistical association between recurrence rate and age, sex, hepatitis status, previous AIN treatment or previous cervical or vulval dysplasia. PLWH were more likely to develop a recurrence compared to HIV negative and non-tested patients (Table 5). However, PLWH also took longer to develop recurrent disease when compared to HIV negative patients (46.1 months compared to 17.5 months) but this was not statistically signi cant; (p = 0.077, ANOVA) (Fig. 2). Interestingly, as they are demographically very similar (see Table 3) if we group untested and HIV negative patients together this statistic becomes signi cant (46.1 months vs. 18.1 months; p = 0.022, ANOVA).
PLWH that recurred were more likely to have received chemoradiotherapy (p = 0.031, Fisher's Exact Test) and have a moderately or poorly differentiated malignancy (Table 6).    There was no statistical association between disease free survival and HIV status, gender, and previous AIN or non-anal intraepithelial dysplasias and malignancies.
50% of patients who were given a stoma for symptomatic control were eventually reversed. 33% patients died with a stoma in situ.

Discussion
ASCC is a cancer of increasing incidence in the western world, [2] which may in part be due to the increasing prevalence of HIV. This cohort study was undertaken to examine patients with ASCC at a specialist HIV referral centre with emphasis on whether PLWH have different clinical outcomes to their HIV negative counterparts.
Within this cohort, we identi ed a clinical subgroup of younger male PLWH who were more likely to have a previous diagnosis of AIN. The cohort also showed a difference in recurrence rates and HIV status, although HIV negative patients on average recurred earlier than PLWH there was no difference in mortality. Taking this into consideration, it is possible that as PLWH are statistically younger at the time of ASCC diagnosis, that PLWH with good viral control have a longer expected life span when compared to older HIV negative patients and are therefore they more likely to develop a recurrence in later life. Our data suggests that long term follow up, beyond the routine 5 years after ASCC treatment, may be advisable for some PLWH as they are more likely to recur later than patients without a positive HIV test (p = 0.022, ANOVA).
The use of High Resolution Anoscopy (HRA) as a screening tool remains controversial. In recent clinical guidelines, The Association of Coloproctology in Great Britain and Ireland could not recommend screening practices such as HRA for routine clinical practice. [7] Similarly, the American Society of Colon and Rectal Surgeons limited its recommendations for the use of HRA to experienced practitioners performing screening on high risk individuals only. [6] These recommendations are based on a lack of evidence that HRA is a screening method that prevents cancer. A recent systematic review also did not identify any evidence based treatments that could de nitely prevent the onset of ASCC if high-grade AIN were detected. [12] As yet there are very few studies with long enough follow up times to be able to identify a risk reduction.
Interestingly, PLWH within our cohort who also had a prior diagnosis of AIN were more likely to present with early-stage tumours (< 5 cm). We believe that this could be an important nding as the earlier detection of disease is an important criterion for a successful screening programme. Due to the retrospective nature of this cohort study, we cannot be certain however whether this nding is a result of cancer being detected earlier by HRA or whether a selection bias exists as patients who comply with a voluntary screening programme are more likely to be informed about the risks of ASCC and present earlier. There was no associated survival bene t identi ed but this is unsurprising as unfortunately the size of the cohort is underpowered to do so. Nevertheless, there is likely to be a signi cant bene t to quality of life as patients with smaller tumours are less likely to recur or require permanent colostomy. Only larger datasets involving PLWH screening programmes will be able to identify any bene ts with certainty.
Recently, there have been encouraging developments in establishing datasets evaluating AIN management and ASCC prevention. The Multinational Anal Squamous Cell Carcinoma: Audit and Registry (mASCARA) started recruiting patients in 2019, [3] and pursues a collaborative approach combining sexual health, gynaecology, HIV and oncology outcomes. The dataset is being used to investigate the screening, prevention, and early diagnosis of ASCC. Moreover, the Anal Cancer HSIL Outcomes Research (ANCHOR) study (NCT02135419) is a multicentre randomised controlled trial based in the USA, which aims to determine whether the treatment of AIN does prevent ASCC. The use of datasets could improve patient outcomes and open the era of adaptive therapy for cancer patients, ideally suiting multicentric registries and international collaboratives. [3,16,17] Strengths and Limitations of study This is a large cohort of patients with a rare cancer, in a specialist centre for the management of HIV and ASCC. The data available permitted the review of a 20-year experience of treating ASCC in a demographic where HIV is highly prevalent. There is little previous data comparing the outcomes between HIV positive and HIV negative patient groups.
Notwithstanding the large cohort for a rare cancer; we are limited by recall bias and a high number of patients being lost to follow up. The data collection for PLWH was also more detailed in the rst few years of the cohort when compared to HIV negative patients.
Interestingly, our experience did not identify any relationship between Hepatitis B and C infections that have been shown to be a risk of ASCC in other studies. [18,19] and unfortunately the data available did not permit an analysis based on previous anal warts, sexual history and sexually transmitted diseases in PLWH. McCloskey et al (2017) identi es a strong increased risk of AIN with syphilis, genital herpes simplex virus and gonorrhoea infections as well as with increasing numbers of sexual partners [20] it is possible that this is a hidden confounding factor.
In the past, ASCC has been associated with a higher incidence in women aged 60-70.  (Fig. 3).

Conclusion
We present a 20-year experience with ASCC in a HIV tertiary referral centre. We identi ed a patient subset of younger male PLWH developing ASCC. Patients known to have a previous diagnosis of AIN were more likely to have earlier disease but there was not any identi able survival bene t.
There was no statistical difference in staging and disease-free survival and HIV status. However, PLWH were more likely to recur and when this occurred, it took longer to develop recurrent disease.
Five-year Disease-Free Survival was associated with tumour size and the absence of nodal or metastatic disease.

Con icts of Interest
None to declare DATA AVAILABILITY The complete datasets generated and analysed in this study are not publicly available without ethical approval as they contain personally identi able data from hospital databases. Anonymised supplementary information can be requested from the corresponding author if the request is reasonable and meets the conditions of the ethical approval granted to undertake this study.

ETHICAL APPROVAL
The authors con rm that the study has been approved by the London-Westminster Research and Ethics Committee and certify that the study was performed in accordance with the ethical standards as laid down in the 1964 Declaration of Helsinki and its later amendments or comparable ethical standards.

CODE AVAILABILITY
Not applicable

AUTHORS CONTRIBUTIONS
All listed authors con rm that they had made substantial contribution to either the conception, design, acquisition, analysis, and interpretation of the data and either drafted the original article or provided critical revisions or prior drafts. All listed authors agree to the version that is published and are accountable for its accuracy and integrity.