A randomized controlled trial to study the effectiveness of laser ablation versus observation to prevent anal cancer in men with human immunodeficiency virus who have high-grade anal intraepithelial neoplasia

Design: HIV-positive MSM with biopsy-proven AIN 2/3 are randomized 1:1 into a treatment and observation arm. Each arm is stratified for high- (3-4 quadrant) and low-volume (1-2 quadrant) AIN 2/3. Within 6 weeks of randomisation, participants will have the first clinic visit. Those in arm 1 have first treatment, and those in arm 2 undergo first observation. An anal Pap smear and HPV test is taken from each participant. Participants are then followed on a 6-monthly basis until month 72. At follow-up, all participants undergo standard examination using high resolution anoscopy (HRA), digital rectal examination (DRE) and biopsy of any new (metachronous) or persistent AIN2/3. If AIN 2/3 is in >1 quadrant or more than one lesion, multiple biopsies are taken. At treatment visits, participants also undergo HRA and DRE but no biopsy is taken. At post-treatment visits HPV-typing will be done. If, on follow-up, persistent or new AIN2/3 is detected in arm 1 patients, they will have further laser treatment within 6 weeks of follow-up (up to 4 treatments in total). All participants receive HPV test, anal Pap smear and biopsy at their last visit. HTA: Lasers emit electromagnetic radiation and are used in medical surgery. The diode laser is best suited to treat anal canal lesions because it can be inserted into the anal canal through a fibre close to lesions and has an ample penetration depth (10mm). It is more precise, safe and relatively inexpensive. Standard care in the NHS is a watch and wait approach. The value against that is that laser can clear AIN 2/3 and hence may prevent cancer. Excision such as used for cervical disease will cause severe morbidity. Unlike the scalpel, lasers seal the surgical wound and effect haemostasis. Electrocautery of extensive AIN2/3 requires a staged approach with multiple operating theatre-based procedures and has high rates of local recurrence, as does infrared coagulation or topical agents,especially in high-volume disease. This is unlike laser ablation which has good clearance and lower recurrence rates. The healing process is also rapid with minimal or no scarring. Measurement: The cost-effectiveness of laser therapy to prevent anal cancer will be assessed by two separate methods: 1) an economic evaluation alongside the clinical trial to directly estimate the incremental cost of each case of anal cancer prevented, and 2) use of data on anal neoplasia grading to parameterize a model of AIN progression, regression and clearance as a result of treatment. This will provide a separate estimate of the effect of treatment on anal cancer prevention. The advantage of the second approach is that it allows to validate the first approach, to increase the statistical power of the data (since intermediate endpoints such as AIN will be used in addition to anal cancer), to elucidate the natural history of AIN, and to compare our findings to other natural history studies of AIN. For this approach, we expand an existing model of anal neoplasia recently developed by the applicants [32]. The economic evaluation will be conducted based on assumptions for the NICE reference case [35]. Costs will be assessed from the perspective of the NHS, a lifetime time horizon will be used, and outcomes will be calculated in terms of cost per quality adjusted life year (QALY) gained. Cost per clinical outcome, e.g. anal cancer case prevented, will be estimated, and one-way and probabilistic sensitivity analyses will be conducted to show the uncertainty around cost-effectiveness estimates. The cost of clinic interventions will be estimated by observing staff time (clinician, nurse, admin) taken for a sample of study visits, using standard sources to cost staff-time [36]. The cost of hospital care for anal cancer will be estimated by updating an analysis on the cost of cancer treatment conducted at the HPA, using the most recent data from Hospital Episode Statistics and the Office for National Statistics [37]. To measure the impact of treatment on Quality of Life (QoL), patients in both arms are asked to complete questionnaires at baseline and a month after the 6, 24 and 60 month clinic visits. These will include validated measures [38-42], and new items on AIN and anal cancer are assessed with the think aloud technique for clarity and acceptability [43]. The EuroQol EQ-5D is included to enable economic evaluation conforming to the NICE reference case. Patients receiving treatment are asked to report their experience of each treatment episode and post-treatment symptoms using measures from the TOMBOLA trial [44] and from studies of anal microsurgery [45]. Timetable: Recruitment: 0-9 month: Identification of eligible patients; consenting; recruitment of 75 cases 15 month: 165 recruited 24 month: Reach recruitment target of primary cohort (400), for 4/5year follow-up 24-36 month: Continue recruitment of further participants (~260), for 3-4 years follow-up 36 month: All participants (~660) are recruited into study Follow-up: Every 6 months from recruitment, until 4-5 years for primary cohort and 3-4 years for other recruits Data analysis: 12 month: Annual data monitoring. 24 month: Annual data monitoring. 30 month: Interim analysis of quality of life and cost-effectiveness. 36 month: Annual data monitoring. 48 month: Annual data monitoring. 60 month: Annual data monitoring. 72-8 month: Final analysis. Setting: Treatment, follow-up and observation take place in an outpatient-based specialist clinic for AIN at the Homerton University Hospital. Case finding, recruitment and follow-up will be extended up to 5 other centres based on the availability of expertise and infrastructure. Population: The target population of the study is the anal cancer high-risk group of HIV-positive MSM. Inclusion criteria are a minimum of 18 years old, a CD4 count of over 350 or if less on antiretroviral treatment for over 3 months. Exclusion criteria include previous treatment for AIN2/3 within 6 months of recruitment or previous diagnosis of anal cancer. Sample size: The sample size (400+260)is based on published progression rates (11%)over 5 years. We consider incidence rates in arm 2 of 1.0%/yr, 2.0%/yr and 2.5%/yr and treatment effects of 90%,80%,65% and 50% and allow 5% loss from each visit. For instance, at 5.5 years with 2%/yr in observation arm and 0.4%/yr in the treatment arm we have 93% power, whereas with 1.0%/yr vs 0.1/yr we have 85% power.