Caroline Hallas Hemmingsen

Combined estrogen–progestin oral contraceptives are classified as hepatic carcinogens based largely on older studies of outdated formulations, creating uncertainty about the risks of contemporary hormonal contraceptives. We conducted a Danish nationwide registry-based cohort study including 1 957 490 women aged 15-49 years followed from 1995 to 2021 to examine current hormonal contraceptive use and liver cancer risk. Use was assessed as a time-dependent exposure using prescription data, and cancer diagnoses were identified through the Danish Cancer Registry. Adjusted incidence rate ratios (IRRs) were estimated using Poisson regression. During 24.5 million person-years of follow-up, 130 women developed liver cancer. Ever use of hormonal contraception was not associated with liver cancer risk (IRR = 0.76, 95% CI = 0.50 to 1.15), nor were current use, product type, duration, or time since last use. Findings suggest that contemporary hormonal contraceptives are not associated with increased liver cancer risk and support reassessment of existing carcinogenic classifications.

Background: Sex hormones have been implicated in leukemogenesis, but evidence regarding hormonal contraceptive use and leukemia risk remains limited and primarily based on older formulations. Given the widespread use of contemporary hormonal contraceptives, clarification of this potential association is needed. This study examines the association between contemporary hormonal contraceptives and leukemia risk. Methods and findings: In a nationwide cohort design, we assessed associations between the use of contemporary hormonal contraceptives and the risk of leukemia based on a cohort of all women aged 15–49 years residing in Denmark from 1995 to 2021 with no previous cancer, hysterectomy, oophorectomy, or sterilization. Information on hormonal contraception use, leukemia diagnoses, and potential confounders (age, calendar year, education) was obtained from nationwide registries. Adjusted incidence rate ratios (IRRs) and 95% confidence intervals [CIs] were estimated for any leukemia, and specific types of leukemia, associated with any hormonal contraceptive use, current and recent use, and previous use, type of product used, duration, and time since last use. Among 1,957,490 pre-menopausal women followed for 24.5 million person-years (median 12.5 years, interquartile range: 5.9,20.5), 671 were diagnosed with leukemia. The incidence rate for leukemia among current and recent users was similar to that among women who had never used hormonal contraception: IRR 0.95 (95% CI [0.78,1.16]; p = 0.62). No association with different durations of use was found: 0–5 years; IRR 0.93 (95% CI [0.75,1.14]; p = 0.48), >5–10 years; IRR 1.16 (95% CI [0.84,1.61]; p = 0.37), >10 years; IRR 0.67 (95% CI [0.33,1.37]; p = 0.27); nor for time since last use: 0–5 years; IRR 1.01 (95% CI [0.78,1.29]; p = 0.96), >5–10 years; IRR 1.05 (95% CI [0.76,1.45]; p = 0.75), >10 years; IRR 0.88 (95% CI [0.60,1.29]; p = 0.52). Also, the IRRs for leukemia with use of different hormonal contraceptive types (e.g., combined products; IRR 0.91 (95% CI [0.73,1.14]; p = 0.42) and progestin-only products; IRR 1.05 (95% CI [0.78,1.40]; p = 0.75)), as well as for product-specific durations of use, were for the majority close to 1. The IRRs were similar for different types of leukemia. Main study limitations include small case numbers in some analyses; therefore, additional large-scale studies are warranted to reliably exclude weak associations. Conclusions: Contemporary hormonal contraceptives were not associated with leukemia, independent of product used, duration of use, time since last use, and type of leukemia. While estimates were imprecise for some subgroups, the overall findings do not support an association.

Background: Lymphomas occur less frequently in women than men, but the causes remain unclear. Given the immunomodulatory potential of female hormones, hormonal contraceptive use could influence lymphoma risk, yet existing research is inconclusive. Methods: We investigated the relationship between contemporary hormonal contraceptive use and lymphoma risk in a nationwide Danish cohort of women aged 15-49 years (1995-2021). Women with prior cancer, hysterectomy, oophorectomy, or sterilization were excluded. Data on hormonal contraceptive use, lymphomas, and confounders were retrieved from national registries. Incidence rate ratios with 95% confidence intervals (CIs) were estimated for any hormonal contraceptive use and lymphoma risk, including lymphoma subtypes, contraceptive types, duration of use, and time since last use. All statistical tests were 2-sided. Results: During 24.5 million person-years of follow-up, 1777 lymphomas (686 Hodgkin, 1091 non-Hodgkin) occurred among 1 957 490 women. Ever-using hormonal contraception was associated with reduced lymphoma risk (incidence rate ratio = 0.84, 95% CI = 0.75 to 0.94), with similar incidence rate ratios across lymphoma types. This was driven by current or recent use of both combined products (incidence rate ratio = 0.80, 95% CI = 0.70 to 0.91) and progestin-only products (incidence rate ratio = 0.81, 95% CI = 0.67 to 0.97), especially oral combined and nonoral progestin-only products. Risk decreased with longer use, with the lowest risk for use more than 10 years (incidence rate ratio = 0.53, 95% CI = 0.33 to 0.85), but no long-term sustained protective effect was seen after cessation. Conclusion: The findings indicate that hormonal contraceptive use may have a role in lymphoma etiology and confer protective effects, underscoring the need to explore hormonal pathways in a preventive and therapeutic context.

An association between maternal hormonal contraception use and childhood central nervous system (CNS) tumors has been suggested, but findings are inconclusive. This population‐based cohort study includes Scandinavian nationwide registry data on liveborn children (1996–2018). Children were followed from birth until CNS tumor (<20 years) or censoring (other cancer, emigration, death, 20th birthday, or end of follow‐up in 2017–2020). Cox proportional hazards models estimated hazard ratios (HRs) and 95% confidence intervals (95% CIs) for the association between maternal hormonal contraception use (any type, type‐specific) and CNS tumor risk (any, any malignant, type‐specific). Maternal use was categorized as “recent use” (0–3 months before or during pregnancy, except for non‐oral progestin‐only types), “previous use” (before recent use), and “no use”. A total of 3,183,316 children were followed for 29,455,528 person‐years, during which time 1384 children developed a CNS tumor (610 malignant). Compared with no use, maternal previous or recent use of any hormonal contraception (HR 0.93, 95% CI 0.82–1.05; HR 0.99, 95% CI 0.83–1.19), combined and progestin‐only types (oral, non‐oral), were not associated with childhood CNS tumor risk. However, maternal recent progestin‐only injection use was associated with malignant childhood CNS tumors (HR 3.95, 95CI % 1.46–10.68), compared with no use (number needed to harm: 1 per 14,577 person‐years). In conclusion, no association was found between maternal use of common types of hormonal contraception and CNS tumors in children. The rarely used progestin‐only injections (medroxyprogesterone acetate) were associated with malignant CNS tumor risk in children, though based on few children.

Background: In Denmark, a girls-only human papillomavirus (HPV) vaccination program was initiated in 2008–2009. The study aim was to assess the HPV prevalence and type distribution in younger men prior to HPV vaccination in men. Methods: The study population was younger men who attended information days regarding military service. At random days (2019–2020), 280 men were included. We collected questionnaire data regarding risk factors for HPV infection and a penile swab for HPV testing. We compared results in this study with those from a previous study of young men (2006–2007). Results: The majority of participants (94%) were 18–20 years old. The median number of lifetime sexual partners was 4. Altogether, 130 men (46.4%) were HPV positive. No infections with HPV types 6, 11, 16, 18, 31, and 45 were detected. The most frequent type was HPV-51 (detected in 11.1%). Comparison showed that the odds of high-risk HPV type infection were higher in 2019–2020 (prevalence odds ratio [POR], 1.7 [95% confidence interval {CI}, 1.1–2.7]) compared with 2006–2007. In contrast, the odds were lower (POR, 0.3 [95% CI, .1–.6]) for HPV types targeted by the 9-valent HPV vaccine. Conclusions: The multicohort girls-only vaccination program has to a large degree protected young men against the HPV types included in the licensed vaccines. This does not speak against gender-neutral vaccination as the HPV prevalence is still high, although consisting largely of less carcinogenic HPV types.

Background: Maternal hormonal contraception use has been associated with an increased risk of leukaemia and central nervous system (CNS) tumours in children. However, little is known about the association with other, less common childhood cancer types. We assessed associations between maternal hormonal contraceptive use and all main types of childhood cancer, excluding leukaemia and CNS tumours.Methods: In this nationwide PECH cohort study, we followed all Danish children liveborn in 1996–2014 (n = 1 185 063) from birth until 2018. Registry information was extracted on maternal filled prescriptions of hormonal contraception and childhood cancer diagnoses (age at diagnosis <20 years). The exposure of interest was maternal ‘recent use’ of hormonal contraception (use ≤3 months before or during pregnancy, except for injections, implants and intrauterine devices with a different timeframe), compared to no use. Cox proportional hazard models were used to estimated hazard ratios (HRs) and 95% confidence intervals (CIs) for childhood cancer according to any use, regiment (combined/progestin‐only) and administration route (oral/non‐oral).Results: During 15 335 990 person‐years of follow‐up, 1384 children were diagnosed with a rare cancer type. Compared to children of mothers with no use of hormonal contraception, maternal recent use of hormonal contraception was not statistically significantly associated with any of the childhood cancers assessed. Except for maternal recent use of non‐oral hormonal contraception, where a hazard ratio (HR) of 6.90 (95% CI: 1.31, 36.47) for hepatic tumours was found. However, this finding was based on a small number of children and if causal, this would result in one additional case per about 99 500 person‐years of follow‐up.Conclusions: Maternal use of hormonal contraception up to or during pregnancy was not associated with nearly every rare childhood cancer type assessed, compared to no use.

International Journal of Epidemiology, dyab219, https://doi.org/10.1093/ije/dyab219 Published online: 13 October 2021

Background: Studies have suggested increased risks of childhood leukaemia after prenatal exposure to antibiotics, particularly nitrofurantoin. However, these findings may be related to the underlying maternal infection. This multinational study aimed to investigate the association between prenatal nitrofurantoin exposure and childhood leukaemia while accounting for maternal infection. Methods: In a population-based cohort study of children born in Denmark, Finland, Norway or Sweden from 1997 to 2013, prenatal exposure to nitrofurantoin or pivmecillinam (active comparator) was ascertained from national Prescription Registries. Childhood leukaemia was identified by linkage to national Cancer Registries. Poisson regression was used to estimate incidence rate ratios (IRRs) and incidence rate differences (IRDs) with inverse probability of treatment weights applied to account for confounding. Results: We included 44 091 children prenatally exposed to nitrofurantoin and 247 306 children prenatally exposed to pivmecillinam. The children were followed for 9.3 years on average (standard deviation 4.1). There were 161 cases of childhood leukaemia. The weighted IRR for prenatal nitrofurantoin exposure when compared with pivmecillinam was 1.34 (95% confidence interval 0.88, 2.06), corresponding to an IRD of 15 per million person-years. Higher point estimates were seen for first- and third-trimester exposure. There was no evidence of a dose–response relationship. Conclusions: Prenatal exposure to nitrofurantoin was not substantially associated with childhood leukaemia, although a slightly elevated IRR with confidence intervals including the null was observed, corresponding to a small absolute risk. The lack of a dose–response relationship and a clear biological mechanism to explain the findings suggests against a causal association.