O‐1 Results from TOMBOLA (trial of management of borderline and other low‐grade abnormal smears)

Abstract

Annually in the UK around 250 000 cervical smears show low‐grade abnormalities. Alternative management policies following a low‐grade smear are cytological surveillance or referral for colposcopy. Their effectiveness and cost‐effectiveness, and the potential for human papillomavirus (HPV) testing to triage women to either management, has been debated. Trial of management of borderline and other low‐grade abnormal smears (TOMBOLA) (a large RCT) addresses these uncertainties, considering clinical, psychosocial and economic outcomes. 4439 women aged 20–59, resident in Grampian, Tayside or Nottingham with a low‐grade smear were randomised to cytological surveillance (six‐monthly smears in primary care) or hospital‐based colposcopy. At colposcopy, women with visible abnormality were randomised to immediate treatment or biopsy and recall for treatment if necessary. Recruitment HPV status was assessed using PCR techniques. Women were followed for three years to an exit colposcopy. Cumulative incidence of CIN2 or more severe disease (CIN2+) in the colposcopy arm was 7.9% per year, higher than in cytological surveillance (5.8%; OR = 1.43, 95% CI 1.23–1.67). This difference was less marked for CIN3+ (OR = 1.27, 1.04–1.55), suggesting spontaneous regression of some CIN2, and that initial colposcopy can lead to over‐treatment. There was little difference in psychosocial outcomes between arms. In comparison of biopsy and recall versus immediate treatment, there was no difference in cumulative incidence of CIN2+ or psychosocial outcomes. There was over‐treatment and increased frequency/duration of bleeding with immediate treatment. There was no compelling economic reason to favour any one management method. Testing for HPV does not appear to be effective in triage. Based on these findings, we make management recommendations for women with low‐grade smears.