What are the risk factors for CIN1 and CIN2/3 in women with a low-grade smear? Results from a case-control study nested in the UK Trial Of Management of Borderline and Other Low-grade Abnormal smears (TOMBOLA)


Year:

Session type:

Anne-Elie Carsin1, Seonaidh Cotton2, Linda Sharp1, Rashmi Seth3

1National Cancer Registry, Cork, Ireland, 2University of Aberdeen, Aberdeen, UK, 3Queens Medical Centre, Nottingham, UK

Abstract

Objectives

The objectives were to investigate (1) risk factors for cervical intraepithelial neoplasia (CIN) grades 1 and 2/3; and (2) which risk factors persist after accounting for high-risk HPV (hrHPV) status.

Method

A case-control study was conducted within the colposcopy arm of TOMBOLA. Subjects were 20-59 years, with a recent low-grade smear. At recruitment, women completed a lifestyle questionnaire and were tested for hrHPV using PCR-EIA and GP5+/6+ primers. At colposcopy, women with an abnormal transformation zone underwent biopsies and/or loop excision. Controls had a normal colposcopy or histology showing no CIN (n=1197); cases had histologically-confirmed CIN1 (n=285) or CIN2/3 (n=347). Analysis was by polytomous regression.

Results

33% of controls were hrHPV positive compared to 50% of CIN1 and 73% of CIN2/3 cases. After adjusting for age and recruitment smear, HPV infection was more strongly associated with CIN2/3 (odds ratio (OR)=3.48; 95% CI 2.62-4.62) than with CIN1 (OR=1.53; 95% CI 1.15-2.02). After adjusting for hrHPV and smear status, risks of both CIN1 and CIN2/3 declined with increasing age. Compared to a BNA recruitment smear, a mild recruitment smear was associated with double the disease risk; the risk estimates were similar for CIN1 and CIN2/3. After adjusting for hrHPV, age and recruitment smear, having had children and having been pregnant before age 20 were associated with significantly raised risks of CIN1 and CIN2/3; current smokers had significantly increased risk of CIN2/3. Analyses stratified by hrHPV status will be presented.

Conclusions

Analyses of this type help elucidate the aetiology of CIN.

Objectives

The objectives were to investigate (1) risk factors for cervical intraepithelial neoplasia (CIN) grades 1 and 2/3; and (2) which risk factors persist after accounting for high-risk HPV (hrHPV) status.

Method

A case-control study was conducted within the colposcopy arm of TOMBOLA. Subjects were 20-59 years, with a recent low-grade smear. At recruitment, women completed a lifestyle questionnaire and were tested for hrHPV using PCR-EIA and GP5+/6+ primers. At colposcopy, women with an abnormal transformation zone underwent biopsies and/or loop excision. Controls had a normal colposcopy or histology showing no CIN (n=1197); cases had histologically-confirmed CIN1 (n=285) or CIN2/3 (n=347). Analysis was by polytomous regression.

Results

33% of controls were hrHPV positive compared to 50% of CIN1 and 73% of CIN2/3 cases. After adjusting for age and recruitment smear, HPV infection was more strongly associated with CIN2/3 (odds ratio (OR)=3.48; 95% CI 2.62-4.62) than with CIN1 (OR=1.53; 95% CI 1.15-2.02). After adjusting for hrHPV and smear status, risks of both CIN1 and CIN2/3 declined with increasing age. Compared to a BNA recruitment smear, a mild recruitment smear was associated with double the disease risk; the risk estimates were similar for CIN1 and CIN2/3. After adjusting for hrHPV, age and recruitment smear, having had children and having been pregnant before age 20 were associated with significantly raised risks of CIN1 and CIN2/3; current smokers had significantly increased risk of CIN2/3. Analyses stratified by hrHPV status will be presented.

Conclusions

Analyses of this type help elucidate the aetiology of CIN.