November 16, 2007 — The use of oral contraceptives is associated with an increase in the risk for invasive cervical cancer, researchers from the International Collaboration of Epidemiological Studies of Cervical Cancer report. An analysis of pooled data from 24 studies, published in the November 10 issue of The Lancet, found that for current users, the relative risk for cancer rises with increasing duration of oral contraceptive use.

In an accompanying editorial, Peter Sasieni, PhD, of the Wolfson Institute of Preventive Medicine, Queen Mary University of London, United Kingdom, notes that because the association between cervical cancer and oral contraceptives is limited to current and recent former users of contraceptives, this suggests that the causative role of oral contraception is probably linked to malignant progression.

"However, the epidemiological observation that the association with invasive cervical cancer is no stronger than is the association with CIN3 [cervical intraepithelial neoplasia grade 3] suggests that there is no further role for exogenous hormones after the development of CIN3," he writes, and also points out that the study is reassuring in that the association becomes weaker immediately after usage stops and is very weak 10 years after last use.

In the collaboration, Jane Green, MD, an epidemiologist from the Cancer Epidemiology Unit, University of Oxford, United Kingdom, and colleagues, reanalyzed data pooled from 24 studies, to investigate the association between cervical carcinoma and patterns of oral contraceptive use. The total combined cohort included 16,573 women with cervical cancer and 35,509 controls.

"This paper essentially adds another piece to the jigsaw when considering pill risks and benefits overall," Dr. Green told Medscape Oncology. "It is reassuring that the increased risk while women are taking the pill goes away, and that therefore, over a woman's lifetime, the absolute increase in risk of cervix cancer from pill use is small."

They found that among women currently using oral contraceptives, the risk for invasive cervical cancer increased with increasing duration. Women who had been using oral contraceptives for 5 or more years had approximately twice the risk vs women who had never used them.

The researchers also estimated that use of oral contraceptives for a period of 10 years, from approximately age 20 to 30 years, raises the cumulative incidence of invasive cervical cancer by age 50 years from 3.8 to 4.5 per 1000 women in industrialized countries and 7.3 to 8.3 per 1000 women in developing countries.

The risk declined after the use of oral contraceptives was stopped and by 10 or more years had returned to the same rate at that of never-users. A similar pattern of risk was seen both for invasive and in situ cancer and in women who tested positive for high-risk human papillomavirus (HPV). Information on HPV status was available in 13 studies, but analyses were restricted to high-risk strains only.

"The relation between oral contraceptive use and HPV exposure or infection is complicated and is very variable by time, place, culture," said Dr. Green. "We allowed for number of sexual partners in our analyses, which is the main determinant of HPV exposure or infection, and where possible also for condom use. In our dataset, women who used oral contraceptives were not significantly more likely to be high-risk HPV positive, and this is consistent with other work. So we do not think that oral contraceptive use is simply correlated with HPV infection."

It is important to bear in mind that this small increase in the risk for cervical cancer and a similar small increase for breast cancer as well are outweighed by the benefits of oral contraceptives, she explained. This includes a reduced risk for cancer of the ovaries and endometrium, as well as the contraceptive benefits.

"For some women, if oral contraceptive use means they start their family later and have fewer children, then this will itself reduce cervical cancer risk," said Dr. Green.

In his editorial, Dr. Sasieni writes, "This thorough meta-analysis of the association between hormonal contraceptives and cervical cancer should both lead scientists to a better understanding of the cofactors affecting HPV infection and cervical neoplasia, and reassure women that fear of cervical cancer should not be a reason to avoid use of oral contraception."

This study was funded by Cancer Research United Kingdom; the International Agency for Research on Cancer; and the United Nations Development Programme/United Nations Population Fund/World Health Organization/World Bank Special Program of Research, Development and Research Training in Human Reproduction.

Dr. Sasieni has received funds from Cancer Research United Kingdom.

Lancet. 2007;370:1591-1592, 1609-1621.

Clinical Context

Study Highlights

• The PubMed database, reviews, and other sources were searched, and epidemiologic studies were identified that had an outcome of invasive cervical cancer and CIN3/carcinoma in situ with information on duration and type of hormonal contraceptive used.
• For case-control studies, at least 100 cases of invasive cancer or 200 cases of CIN3/carcinoma in situ were required.
• For prospective cohort studies, at least 30 cases of cervical cancer were included.
• Women aged 18 to 89 years were included, and matched controls were identified for each study.
• Excluded were women who had undergone a hysterectomy, those who never had sexual partners, and those with HIV infection.
• Hormonal contraception was classified as oral or injectable, combined estrogen-progestagen, or progestagen-only.
• Histologic cervical cancers were classified with use of World Health Organization criteria.
• Stratification of data was by study center, age at first intercourse, number of term pregnancies and sexual partners, smoking, and cervical cancer screening status.
• Cumulative incidence rates of cervical cancer with different durations of use were calculated for less developed and more developed countries.
• Among the 24 studies available were 3 international multicenter studies, and half the studies were from less developed countries.
• The studies included 16,573 women with cervical cancer (11,170 with invasive and 5403 with CIN3/carcinoma in situ) and 35,509 women without cervical cancer (controls).
• Median age at diagnosis was 45 years for invasive cancers and 35 years for CIN3/carcinoma in situ.
• 85% of cancers were squamous cell carcinomas and 15% were adenocarcinomas.
• Use of combined oral contraceptives was reported by 32% of women with invasive cancer, 57% with CIN3/carcinoma in situ, and 34% of controls.
• Current use of contraceptives for 5 or more years was reported by 33% of women with invasive cancer and 61% with CIN3/carcinoma in situ.
• Mean duration of use was 7 years for women with invasive cancer, 7 years for those with CIN3/carcinoma in situ, and 6 years for controls.
• Mean age at first use was 24 years for women with invasive cancer, 21 years for those with in situ carcinoma, and 23 years for controls.
• In the control group, ever-use of combined oral contraceptives was associated with previous Papanicolaou tests, more sexual partners, lower age at first intercourse, smoking, and higher parity.
• For duration of use of 5 years or less, there was no increase in the relative risk for cervical cancer.
• In those with use of more than 5 years (mean, 9.8 years) the relative risk in current users was 1.90 vs never-users of combined contraceptives.
• Each year of use increased the relative risk by a factor of 1.07 equivalent to a change in relative risk per 5 years of 1.38.
• The relative risk declined with increasing time since last use (P < .0001) and by 10 or more years since last use, approached that of never-users.
• Timing of use such as age at first use and time since first use did not independently influence the risk for cervical cancer.
• A small increase in the risk for invasive cancer was seen with progestagen-only injectable use for 5 years or more.
• Use of 10 years of oral contraceptives from ages 20 to 30 years is associated with an increase from 7.3 to 8.3 per 1000 women in cervical cancer in less developed countries and an increase from 3.8 to 4.5 per 1000 women in industrialized countries.

Pearls for Practice

• Use of oral contraceptives for 5 or more years is associated with the increased risk for invasive cervical cancer and CIN3/carcinoma in situ, and the risk increases with duration of use.
• Risk for cervical cancer associated with use of oral contraceptives diminishes with years from last use and reaches the level of never-users by 10 years.

CME/CE Test