FormalPara Key points

Cryotherapy is capable of clearing high-grade squamous intraepithelial lesions in HIV-positive men who have sex with men (MSM).

Side effects following cryotherapy for high-grade squamous intraepithelial lesions are mild.

Recurrences of high-grade squamous intraepithelial lesions are frequent in HIV-positive MSM treated with cryotherapy.

1 Introduction

HIV-positive men who have sex with men (MSM) are at risk for developing anal squamous cell carcinoma. Incidence rates have increased from 13–37 per 100,000 person-years in the pre-combination antiretroviral therapy (cART) era to 65–109 per 100,000 person-years in the cART era [1]. Anal intraepithelial neoplasia (AIN) is a precursor stage of anal squamous cell carcinoma and follows infection with human papillomavirus (HPV). High-resolution anoscopy (HRA)-guided biopsies are the gold standard for identifying AIN, and the prevalence of low-grade squamous intraepithelial lesions (LSILs or AIN 1) and high-grade squamous intraepithelial lesions (HSILs or AIN 2–3) is high in HIV-positive MSM: 31.3 and 29.1%, respectively [1]. Treatment of HSIL in HIV-positive MSM is highly recommended to prevent progression towards invasive disease [2]. Large randomized controlled trials are currently investigating the natural history of AIN [3,4,5].

Several treatment modalities for the treatment of HSIL have been described. Treatment varies from targeted destruction of HSILs using infrared coagulation, electrocautery, cryosurgery, or surgical excision to immunomodulation and photodynamic therapy, but with little consensus between the guidelines [6,7,8]. Electrocautery is currently the treatment of choice for intra-anal HSIL in many centers, but response rates with all HSIL treatment modalities are disappointingly low and recurrence rates are high [9, 10]. Therefore, new treatment modalities or techniques should be considered.

Cryotherapy is an old technique [11] that was first introduced in dermatology in 1899 [12] and is currently still widely used for treatment of HPV-induced pre-malignant and malignant skin disorders such as HPV-associated warts, actinic keratosis, and basal cell carcinoma [13,14,15]. Anogenital condylomata are commonly treated with liquid nitrogen because it is easily applied, is cheap, leaves no scar tissue when applied correctly, and achieves acceptable clearance rates [16].

In this retrospective study, we describe the response and recurrence rates of cryotherapy for intra- and perianal HSIL in HIV-positive MSM.

2 Materials and Methods

2.1 Patients

We studied HIV-positive MSM aged ≥30 years who underwent AIN screening with HRA and were diagnosed with intra- and/or perianal HSIL, as described by Richel et al. [17] (Fig. 1). In short, screening was performed by one of three expert anoscopists (having performed more than 200 HRAs each) and consisted of a digital rectal exam followed by peri- and intra-anal inspection with a colposcope (ZEISS opmi pico surgical microscope) after repeated application of acetic acid (3% solution) and staining with Lugol’s iodine when indicated. All lesions suspicious for AIN were biopsied and graded by one of two pathologists who specialized in squamous intraepithelial lesions. The extent of intra-anal HSIL was analyzed by describing the number of quadrants containing HSIL disease. No anal cytology was performed in any of the patients. Patients diagnosed with  intra- and/or perianal HSIL were offered standard electrocoagulation therapy and cryotherapy, with the latter introduced as a new treatment modality with unknown efficacy.

Fig. 1
figure 1

Intra-anal high-grade squamous intraepithelial lesion pre-cryotherapy treatment

Full size image

2.2 Cryotherapy

Patients scheduled for cryotherapy at the Academic Medical Center (Amsterdam, the Netherlands) between 30 December 2008 and 23 April 2015 attended a maximum of five treatment visits, each visit 4–6 weeks apart. Treatment for intra- and perianal lesions consisted of office-based HRA-guided liquid nitrogen application in three freeze–thaw cycles using a spray gun with a standard 0.56 mm aperture, spraying 10 s per lesion (Fig. 2). At each treatment visit, self-reported side effects were noted and inquiries were made regarding blood loss and pain that may have resulted from the previous treatment, followed by inspection with HRA and guided cryotherapy for remaining HSIL lesions. If no lesions were seen, treatment response was defined as a complete response (CR), no further treatments were given, and a 6-month follow-up visit was scheduled.

Treatment response was assessed 4–6 weeks after the final treatment. During this session, the intra- and perianal region was again inspected and all remaining lesions biopsied. If no lesions were seen, treatment response was defined as CR and no biopsies were taken. If histological resolution of all AIN lesions was achieved, treatment response was also defined as CR. If all HSIL lesions had regressed to at least LSIL (AIN 1), treatment response was defined as partial response (PR), and, if invasion had occurred, response was defined as progressive disease. In all other cases, the outcome was defined as stable disease (SD), which was analyzed in more detail to distinguish between persistent HSIL (in the same location) or metachronous HSIL (in a new location). Patients with CR or PR were scheduled for a follow-up HRA in 6 months. Patients with SD were again offered treatment with one of the remaining treatment modalities described, and patients with progressive disease were referred for oncologic treatment options.

Patients who received cryotherapy and completed at least one follow-up visit were included in the analyses. Patients lost to follow-up before a treatment response could be evaluated were excluded from the analyses. Patients who were lost to follow-up after the treatment response was evaluated, and patients who participated in a quadrivalent HPV vaccination trial (VACCAIN-P study) were excluded from recurrence rate analyses.

Fig. 2
figure 2

Intra-anal high-grade squamous intraepithelial lesion post-cryotherapy treatment

Full size image

2.3 Statistical Analysis

Potential risk factors for cryotherapy treatment success—nadir cluster of differentiation (CD)-4 cell count and ever smoking—were tested using univariable logistic regression analysis. Variables were considered significantly associated when the p value was <0.05. Statistical analyses were performed using SPSS software (version 23.0.3; IBM, Armonk, NY, USA). Data were expressed as medians with interquartile ranges.

3 Results

3.1 Patient Characteristics

Between 30 December 2008 and 23 April 2015, 64 patients were treated with cryotherapy and included in the present study. Their median age was 48 years, the median CD4+ T-cell count during treatment was 640 cells/μL, and 60 of 64 patients (94%) were using cART. Of the patients receiving cryotherapy, 52 (81%) received it for intra-anal HSIL only, one (2%) for perianal HSIL only, and 11 (17%) for both intra- and perianal HSIL. Baseline characteristics are listed in Table 1.

Table 1 Characteristics of HIV-positive men who have sex with men with anal high-grade squamous intraepithelial lesions treated with cryotherapy
Full size table

3.2 Treatment Response

Six of 64 patients were lost to follow-up before a treatment response could be evaluated (Fig. 3). Of 58 patients, 35 (60%) had either a CR (n = 20) or a PR (n = 15) (Table 2). Of 12 patients treated for perianal HSIL, four (33%) had either a CR (n = 1) or a PR (n = 3). Of 57 patients treated for intra-anal HSIL, 34 (60%) had either a CR (n = 19) or a PR (n = 15). In total, 27 of 40 patients (68%) with intra-anal HSIL in one quadrant, 6 of 12 (50%) with HSIL in two quadrants, and one of four (25%) with HSIL in three quadrants had either a CR or a PR. The only patient with HSIL in four quadrants had SD. Of the 23 patients with SD, 11 (40%) had persistent HSIL, seven (30%) showed only metachronous HSIL, and five (21%) showed both persistent and metachronous HSIL after treatment. None of the patients had progressive disease during the course of this study.

Fig. 3
figure 3

Flowchart of treatment intervals and treatment results in HIV-positive men who have sex with men receiving cryotherapy as treatment for anal high-grade squamous intraepithelial lesions. CR complete response, LTFU lost to follow-up, n number of study participants, PR partial response, SD stable disease, T time in weeks

Full size image
Table 2 Response to treatment of HIV-positive men who have sex with men with anal high-grade squamous intraepithelial lesions treated with cryotherapy
Full size table

Nadir CD4 cell count was not significantly associated with CR, PR, or SD. Ever smoking was not significantly associated with CR (p = 0.493) or PR (p = 0.087) but did significantly increase the odds of SD [odds ratio (OR) 0.26; 95% confidence interval (CI) 0.078–0.871; p = 0.029].

3.3 Recurrence Rate

Of the 35 patients with a CR or PR, 14 were screened and selected for the VACCAIN-P study at their first follow-up visit and were therefore excluded from further analyses. Of the 21 remaining patients, 19 (seven CR, 12 PR) were adequately followed-up, meaning they received a follow-up HRA 6 months after their initial treatment response was evaluated. Of these 19 patients, seven (37%; four CR and three PR) had an HSIL recurrence at first follow-up. Thereafter, 11 of the 12 remaining patients received a second follow-up visit in the next 6–19 months, and five (all PR) had a recurrence of HSIL. Of the remaining six patients (three CR, three PR), three had no HSIL recurrence at their third follow-up (one CR, two PR), one (CR) had recurrent HSIL and two were lost to follow-up (one CR, one PR). Overall, 13 of the 19 patients (68%) who experienced a CR or a PR after cryotherapy, were not vaccinated against HPV, and had at least one follow-up visit after 6 months, had a documented HSIL recurrence within 18 months.

3.4 Side Effects

In total, 31 of 64 patients (48%) reported one to three side effects during the cryotherapy treatment period: 19 (30%) reported one side effect, nine (14%) reported two, and three (5%) reported three. Of the 64 patients, 20 (31%) reported anal pain or tenderness during and/or following treatment, 14 (22%) reported mild blood loss for a few days, four (6%) experienced psychological problems (erectile dysfunction, stress, anxiety), and eight (13%) reported other side effects (anal fissure, constipation, temporary fecal incontinence, diarrhea, hemorrhoid). None of the treated patients required additional medical attention because of side effects. One patient with a history of anal carcinoma discontinued treatment because of anal pain and psychological stress.

4 Discussion

In this study, we investigated cryotherapy for the treatment of HSIL in HIV-positive MSM. Cryotherapy resulted in a CR or PR in 60% of patients. Half of the patients reported side effects from cryotherapy, most of which were mild pain and blood loss.

The World Health Organization recommends cryotherapy for the treatment of cervical intraepithelial neoplasia (CIN) when the loop electrosurgical excision procedure (LEEP) is not available [18]. LEEP is more effective, as it allows the treating physician to remove the entire transitional zone in one procedure; however, the transitional zone of the anal canal is anatomically unfit for LEEP. Esophageal high-grade intraepithelial neoplasia at the pectinate line (Barrett’s HG dysplasia) can be effectively and safely treated by endoscopic application of liquid nitrogen [19]. Studies showed high rates of clearance with endoscopic cryotherapy [19], but recurrence rates were higher than with photodynamic therapy or endoscopic mucosal resection [20].

Response rates for available HSIL treatment modalities (trichloroacetic acid [TCA], imiquimod, infrared coagulation, electrocautery) are similar to our rates for cryotherapy (61% [21], 61% [22], 64% [23], and 69% [9], respectively). However, if we observe the effect of cryotherapy on a lesional level in patients with SD, seven of 23 patients could be considered ‘partial responders’ because they developed metachronous lesions and all original HSIL lesions cleared; thus, the overall response rate would increase to 42 of 58 patients (72%). In this small group of patients, univariate analysis indicated that ever smoking was positively associated with SD. Future studies with larger patient groups should perform multivariate analysis to assess whether smoking is a consistent risk factor for treatment failure. Cryotherapy for HSIL in one intra-anal quadrant showed a higher treatment response (68%) than for HSIL in two quadrants (50%) or three quadrants (25%). For perianal HSIL, imiquimod treatment still shows a better treatment response than cryotherapy (91 vs. 36%), making imiquimod the more effective choice for perianal HSIL [9].

Side effects from electrocautery, imiquimod, 5-fluorouracil, and cryotherapy were reported in 76, 47, 65, and 50% of patients, respectively; however, this was not a randomized controlled study [9]. In the British Association of Dermatologists’ guidelines for the management of Bowen’s disease, Morton et al. [24] recommend cryotherapy as a generally good choice of treatment for single or multiple small (<2 cm), good healing lesions (refers to the clinician’s perceived potential for good healing at the affected site). None of the possible complications following cryotherapy described by Morton et al. [24] (poor healing and hypopigmented scarring) were observed in our study.

The recurrence rate following a CR or PR after cryotherapy for HSIL appeared to be lower than after treatment with electrocautery, but follow-up visits in our study were less stringent than in prospective studies, and recurrence rates may therefore have been underestimated [9, 25].

To our knowledge, this is the first study to describe the treatment effects of cryotherapy for intra- and perianal HSIL in HIV-positive MSM. Although the retrospective, non-comparative design is a limitation of this study, this is a first step to finding a new possible treatment modality for HSIL. The inclusion of several patients with persistent (n = 2) or recurrent (n = 22) HSIL after electrocoagulation and other treatment options (5-fluorouracil, imiquimod, or TCA) could have introduced selection bias; therefore, the reported treatment response could underestimate the true potential of cryotherapy.

5 Conclusion

Cryotherapy is capable of clearing HSIL in HIV-positive MSM. Cryotherapy is well tolerated and side effects are common and mostly mild, but HSIL recurrences were frequent. A prospective, randomized study comparing electrocautery and cryotherapy should further investigate the efficacy and tolerability of cryotherapy as a new treatment modality for intra-anal HSIL.