Incidence of Cancer among Women Using Long versus Monthly Cycle Hormonal Replacement Therapy, Finland 1994-1997

Objective: This study was initiated as a consequence of a clinical observation in one hospital of several endometrial carcinomas among users of a fixed combination of a 3-month cycle oral hormonal replacement therapy (HRT), which has been marketed in Finland since 1990. We studied whether the use of 3-month ("long") cycle HRT is accompanied by a higher risk of endometrial cancer than the use of monthly cycle HRT. Methods: A nationwide cohort of 15,956 long cycle and 78,549 monthly cycle HRT users since January 1994 was extracted from the files of the national medical reimbursement register and followed up for cancer incidence through the Finnish Cancer Registry up to the end of 1997. Results: There were 61 cases of endometrial cancer among long cycle HRT users which significantly exceeds the average incidence in the Finnish population (standardized incidence ratio (SIRlong) 2.0, 95% confidence interval (CI) 1.6-2.6). The SIR among users of monthly cycle HRT products (SIRmonthy) was 1.3 (1.1-1.6) and the ratio SIRlong/SIRmonthlythus 1.5 (95% CI 1.1-2.1). Endometrial cancers among long cycle HRT users occur more often in early stages than in the general population and are more often highly differentiated. A survey of a sample of endometrial cancer patients in the long cycle HRT group revealed that all of them also had a history of other HRT. Users of both long and monthly cycle HRT had a similar statistically significant 30% excess of breast cancer in comparison to national incidence rates, while the incidence of colon cancer was decreased in both groups. There was no difference between the HRT groups in overall cancer morbidity. Conclusions: Our results imply that all HRT users have an increased risk of endometrial cancer, and long cycle HRT carries a higher relative risk than monthly cycle HRT. However, in this non-randomized setting it is impossible to judge whether this excess is attributable to the type of HRT or to patient selection. In-depth studies are needed to find out possible dose-response associations and to evaluate the role of potential confounders in detail.