Post-menopausal hormonal therapy and gallbladder cancer risk

Authors


Abstract

The relation between post-menopausal hormone replacement therapy (HRT) and gallbladder cancer was analyzed in women above age 45 years, using data of a case-control study conducted in Italy between 1985 and 1997, on 31 incident, histologically confirmed cases and 3,702 controls in hospital for acute, non-neoplastic conditions. The multivariate odds ratio (OR) was 3.2 (95% confidence interval: 1.1–9.3) for those who had ever used HRT and the OR tended to rise with longer duration. Although based on small numbers, due to the rarity of the disease, these findings provide the first direct epidemiological evidence of an association between HRT and gallbladder cancer. © 2002 Wiley-Liss, Inc.

Gallbladder cancer, together with thyroid cancer, is the only non-sex-related neoplasm that is found more frequently in females than in males, with male to female sex ratios being around 0.5–0.8 in most countries.1, 2 This may be related to gallstones, which are the best recognized risk factor for gallbladder cancer and are more frequent in women than in men.3–5

It is known that female hormones may change the composition of bile acids and the motility of the biliary tract. Some studies6–9 have reported a direct relation between parity and gallbladder cancer, but the role of other hormonal and menstrual factors, including age at menarche and menopause and oral contraceptive use, remains controversial.10–13

Likewise, the possible relation between hormone replacement therapy (HRT) in menopause and gallbladder cancer risk has not been adequately addressed and the issue was not considered in the IARC Monograph on hormonal contraception and post-menopausal hormonal therapy.13 This is likely due to the rarity of the disease and the consequent difficulties in obtaining valid information. We analyzed therefore data from a case-control study of gallbladder cancer conducted in northern Italy.10

SUBJECTS AND METHODS

The data were derived from a case-control study of digestive tract cancers conducted in the Greater Milan area between 1985 and 1997, whose general design has already been described.10, 14

For the present analysis, the cases were 31 women aged 45–79 years (median age 64 years) with incident, histologically confirmed gallbladder cancer. The comparison group included 3,702 women aged 45–79 years (median age 60 years) who were admitted to the same network of hospitals for acute, non-neoplastic, non hormone-related conditions (30% traumas, 29% other orthopaedic conditions, 16% acute surgical conditions and 25% other miscellaneous diseases) and who had not undergone cholecystectomy. Less than 5% of the cases and controls that were approached refused to participate.

Trained interviewers used a structured questionnaire to obtain information on sociodemographic factors, personal characteristics and habits, a problem-oriented medical history and use of selected drugs, including HRT. The age and duration of each episode of HRT use were recorded.

Odds ratios (OR) and the corresponding 95% confidence intervals (CI) were derived from multiple logistic regression equations,15 including terms for age, year of interview, education, tobacco smoking, alcohol drinking, body mass index (BMI), parity and type and age at menopause.

RESULTS

Table I shows the association between gallbladder cancer risk and HRT use. A total of 5 (16%) cases and 222 (6%) controls had ever used HRT. The corresponding OR was 3.2 (95% CI: 1.1–9.3). ORs were 2.3 (95% CI: 0.5–10.5) for duration of use <2 years and 3.0 (95% CI: 0.6–15.0) for ≥2 years; they were 1.7 (95% CI: 0.2–14.0) for use stopped since <10 years and 3.2 (95% CI: 0.8–12.9) for use stopped since ≥10 years.

Table I. Frequencies and Odds Ratio of Gallbladder Cancer in Relation to Post-Menopausal Hormone Replacement Therapy (HRT) use; Italy, 1985-97
HRT useGallbladder cancerControlsOdds ratio1 (95% confidence interval)
  • 1

    Odds ratio derived from multiple logistic regression equations including terms for age, year of interview, education, tobacco smoking, alcohol drinking, body mass index, parity, type and age at menopause.

  • 2

    Reference category.

Never263,48012
Ever52223.2 (1.1–9.3)

The OR for women who had ever used HRT, conditioned for age and year of interview and adjusted for education, tobacco smoking, alcohol drinking, BMI, parity, type and age at menopause, was 3.8 (95% CI: 1.0–14.2). Four of 31 cases had clinical history of gallstones; none of these had ever used HRT.

DISCUSSION

Although a relation between HRT and the risk of gallbladder cancer is plausible,10, 16–18 analytical studies to date have provided no information on the issue.5, 11, 13 Thus, the findings from the present study, of a significant excess gallbladder cancer risk in HRT users, are of substantial interest since they provide direct evidence of a role of HRT on gallbladder carcinogenesis, despite the small number of cases, due to the rarity of the disease and the relatively low frequency of HRT users in this Italian population.19

Although this is a hospital-based case-control study, it is unlikely that its results are explained by selection, information bias or confounding, since the catchment area of cases and controls were comparable, participation was almost complete, there was no reason to suspect differential recall of HRT use by gallbladder cancer cases and controls and allowance for several potential confounding factors did not appreciably modify the relative risk estimates. HRT may reduce the risk of trauma and other orthopaedic conditions. However, the OR was 3.4 (95% CI: 1.1–10.7) when comparison was made with other diagnostic categories of controls only.

Gallbladder cancer is associated with gallstones,5 and gallstones are associated with HRT.3, 4 Colelithiasis is a possible pathogenic link of the association between HRT and gallbladder cancer. Furthermore, HRT stimulates gallbladder motility and can make the gallstones symptomatic. Consequently, this could determine an early diagnosis of gallbladder cancer in subjects with cholelitiasis. The subclinical (ultrasounds) prevalence of gallstones however is undefined in this population. These limitations notwithstanding, the multivariate OR for those who have ever used HRT, including controls who underwent cholecystectomy, after further adjustment for colelithiasis, was 3.1 (95% CI: 1.0–10.1). Finally, a companion study of hepatocellular carcinoma, based on the same study population and methods, showed a significant inverse relation with HRT.14

Given the rarity of the disease and the size of the excess risk, the present findings have in any case limited public health implications, but provide additional relevant information for any global risk-benefit assessment of HRT use.

Acknowledgements

The authors thank Ms. I. Garimoldi for editorial assistance. S.G. was supported by a Monzino Foundation and C.B. by an AIRC/FIRC fellowship.

Ancillary