No excess liver cancer risk seen with oral contraceptive use

Two large prospective cohorts show no evidence of an association between the use of oral contraceptives and the risk of liver cancer.

Compared with never use, ever use of oral contraceptives posed no increased risk of liver cancer in either the Million Women Study (MWS) (hazard ratio [HR], 1.05, 95 percent confidence interval [CI], 0.97–1.13; p=0.27) or the UK Biobank (HR, 1.08, 95 percent CI, 0.76–1.55; p=0.66). [Lancet Oncol 2025;doi:10.1016/S1470-2045(25)00222-0]

In an analysis by histological subtype, oral contraceptive had a null association with hepatocellular carcinoma (MWS: HR, 1.04, 95 percent CI, 0.90–1.21; UK Biobank: HR, 1.23, 95 percent CI, 0.61–2.48) and intrahepatic cholangiocarcinoma (MWS: HR, 1.03, 95 percent CI, 0.93–1.15; UK Biobank: HR, 1.08, 95 percent CI, 0.70–1.66).

These findings held true in a meta-analysis of 23 observational studies, including the MWS and the UK Biobank. There was no evidence to support a link between ever use of oral contraceptives and liver cancer (relative risk [RR], 1.04, 95 percent CI, 0.98–1.11; I²=45.9 percent, p=0.0080).

A slight risk increase was seen among women who used oral contraceptives for longer. Each 5 years of use of oral contraceptives raised the risk of liver cancer by 6 percent (RR, 1.06, 1.02–1.10; I²=63.9 percent, p<0.0001) compared with never use. But this association was not observed when explored by subtypes of liver cancer (p=0.82).

“The slight increase in the risk of liver cancer with increasing duration of use of oral contraceptives was largely driven by the largest cohorts (MWS and UK Biobank). However, neither of these studies showed a statistically significant trend with increasing duration of use,” the investigators noted.

“Although we carefully adjusted for all measured confounders at recruitment, the possibility of residual confounding or unmeasured confounding variables, particularly the hepatitis C or hepatitis B virus status of participants, cannot be ruled out. Women might have also developed diabetes over the study period, or they might have underestimated their weight, which are important risk factors for liver cancer,” they acknowledged. [Am J Gastroenterol 2018;113:1494-1505; Cancer Res 2016;76:6076-6083]

Taken together, the data presented in this study stand in contrast with the position on oral contraceptive use in the most recent IARC monograph, wherein the expert review panel supported the findings of an association between oral contraceptives and liver cancer, according to the investigators. [https://publications.iarc.fr/Book-And-Report-Series/Iarc-Monographs-On-The-Identification-Of-Carcinogenic-Hazards-To-Humans/Pharmaceuticals-2012]

Need for further research

In a linked commentary, Dr Lisa Iversen from the Institute of Applied Health Sciences, University of Aberdeen, Aberdeen, UK, highlighted a need for large, long-term studies that can more accurately examine patterns of oral contraceptive use, such as current use, how recently it was used, and the total duration, to better understand the true risk. [Lancet Oncol 2025;doi:10.1016/S1470-2045(25)00222-0]

“Examining oral contraceptives and liver cancer risk is complex, because many women use different contraceptives throughout their reproductive lives. Over time, the formulations of combined pills have also changed,” Iversen wrote.

She mentioned data linkage as a promising research method, pointing out that linking existing healthcare records from across an entire country or multiple countries could be a cheaper alternative to traditional prospective cohort studies, with minimal loss to follow-up.

“The challenge with such an approach is whether, in addition to detailed prescribing records, information is also routinely gathered about important confounders, including hepatitis B and C viruses, type 2 diabetes, and lifestyle, in particular alcohol consumption, tobacco use, and BMI,” she added.

“While gathering of further robust evidence continues … it is important to highlight that if the relationship between longer duration of oral contraceptive use and liver cancer was causal, the number of extra cases of liver cancer attributable to oral contraceptives would be very small. This small increased risk needs to be balanced against the other known lifetime cancer risks and benefits of oral contraceptives,” Iversen concluded.

Study details

A total of 1,305,024 women from the MWS and 253,408 women from the UK Biobank were included in the analysis. Liver cancer occurred in 0.21 percent of the MWS participants and in 0.08 percent of the UK Biobank participants over a median follow-up of 21.4 and 12.6 years, respectively.

Compared with participants who had never used oral contraceptives, those who had ever used oral contraceptives in both MWS and UK Biobank cohorts were generally younger; more likely to consume more alcohol, be current smokers, use menopausal hormone therapy; and less likely to have diabetes.

A sensitivity analysis stratified by age at recruitment (≤55 and >55 years) to try to account for women who might have used older formulations of oral contraceptives yielded results similar to that of the main analysis. No evidence of associations was found between oral contraceptive use and liver cancer risk across the age subgroups in either the MSW or UK Biobank cohorts.