Long-acting, progestin-based contraceptives and risk of breast, gynecological, and other cancers

Karen M Tuesley; Katrina Spilsbury; Sallie-Anne Pearson; Peter Donovan; Andreas Obermair; Michael D Coory; Sitwat Ali; Nirmala Pandeya; Louise Stewart; Susan J Jordan; Penelope M Webb

doi:10.1093/jnci/djae282

. 2025 Jan 14;117(5):1046–1055. doi: 10.1093/jnci/djae282

Long-acting, progestin-based contraceptives and risk of breast, gynecological, and other cancers

Karen M Tuesley ^1,^2,^✉, Katrina Spilsbury ³, Sallie-Anne Pearson ^4,⁵, Peter Donovan ^6,⁷, Andreas Obermair ^8,⁹, Michael D Coory ¹⁰, Sitwat Ali ^11,¹², Nirmala Pandeya ^13,¹⁴, Louise Stewart ¹⁵, Susan J Jordan ^16,^17,^b, Penelope M Webb ^18,^19,^b

¹ School of Public Health, University of Queensland, Brisbane, QLD 4006, Australia

² Population Health Program, QIMR Berghofer Medical Research Institute, Brisbane, QLD 4006, Australia

³ Institute for Health Research, The University of Notre Dame Australia, Fremantle, WA 6160, Australia

⁴ School of Population Health, University of New South Wales, Sydney, NSW 2052, Australia

⁵ Centre of Research Excellence in Medicines Intelligence, University of New South Wales, Sydney, NSW 2052, Australia

⁶ Clinical Pharmacology Department, Royal Brisbane and Women’s Hospital, Brisbane, QLD 4006, Australia

⁷ Faculty of Medicine, University of Queensland, Brisbane, QLD 4006, Australia

⁸ Queensland Centre for Gynaecological Cancer, Royal Brisbane and Women’s Hospital, Brisbane, QLD 4006, Australia

⁹ Centre for Clinical Research, University of Queensland, Brisbane, QLD 4006, Australia

¹⁰ Mater Research Institute, University of Queensland, Brisbane, QLD 4101, Australia

¹¹ School of Public Health, University of Queensland, Brisbane, QLD 4006, Australia

¹² Population Health Program, QIMR Berghofer Medical Research Institute, Brisbane, QLD 4006, Australia

¹³ School of Public Health, University of Queensland, Brisbane, QLD 4006, Australia

¹⁴ Population Health Program, QIMR Berghofer Medical Research Institute, Brisbane, QLD 4006, Australia

¹⁵ School of Population and Global Health, The University of Western Australia, Perth, WA 6009, Australia

¹⁶ School of Public Health, University of Queensland, Brisbane, QLD 4006, Australia

¹⁷ Population Health Program, QIMR Berghofer Medical Research Institute, Brisbane, QLD 4006, Australia

¹⁸ School of Public Health, University of Queensland, Brisbane, QLD 4006, Australia

¹⁹ Population Health Program, QIMR Berghofer Medical Research Institute, Brisbane, QLD 4006, Australia

^✉

Corresponding author: Karen M. Tuesley, PhD, School of Public Health, The University of Queensland, 266 Herston Road, Herston, QLD 4006, Australia (k.tuesley@uq.edu.au).

S. J. Jordan and P. M. Webb are considered co–senior authors of this work.

Roles

Karen M Tuesley: PhD, Formal analysis, Investigation, Methodology, Writing - original draft, Writing - review & editing

Katrina Spilsbury: PhD, Formal analysis, Funding acquisition, Investigation, Methodology, Writing - review & editing

Sallie-Anne Pearson: PhD, Funding acquisition, Investigation, Writing - review & editing

Peter Donovan: MSc, Funding acquisition, Investigation, Writing - review & editing

Andreas Obermair: MD, Investigation, Writing - review & editing

Michael D Coory: PhD, Funding acquisition, Investigation, Writing - review & editing

Sitwat Ali: MPH, Investigation, Writing - review & editing

Nirmala Pandeya: PhD, Funding acquisition, Investigation, Writing - review & editing

Louise Stewart: PhD, Funding acquisition, Investigation, Writing - review & editing

Susan J Jordan: PhD, Conceptualization, Formal analysis, Funding acquisition, Investigation, Methodology, Supervision, Writing - review & editing

Penelope M Webb: DPhil, Conceptualization, Formal analysis, Funding acquisition, Investigation, Methodology, Supervision, Writing - review & editing

PMCID: PMC12058256 PMID: 39805314

Abstract

Background

Use of long-acting, reversible contraceptives has increased over the past 20 years, but an understanding of how they could influence cancer risk is limited.

Methods

We conducted a nested case-control study among a national cohort of Australian women (n = 176 601 diagnosed with cancer between 2004 and 2013; 882 999 matched control individuals) to investigate the associations between the levonorgestrel intrauterine system, etonogestrel implants, depot-medroxyprogesterone acetate and cancer risk and compared these results with the oral contraceptive pill. We used conditional logistic regression to estimate odds ratios (OR) and 95% confidence intervals (CI).

Results

Levonorgestrel intrauterine system and etonogestrel implant use was associated with breast cancer risk (OR = 1.26, 95% CI = 1.21 to 1.31, and OR = 1.24, 95% CI = 1.17 to 1.32, respectively), but depot-medroxyprogesterone acetate was not, except when used for 5 or more years (OR = 1.23, 95% CI = 0.95 to 1.59). Reduced risks were seen for levonorgestrel intrauterine system (≥1 years of use) in endometrial cancer (OR = 0.80, 95% CI = 0.65 to 0.99), ovarian cancer (OR = 0.71, 95% CI = 0.57 to 0.88), and cervical cancer (OR = 0.62, 95% CI = 0.51 to 0.75); for etonogestrel implant in endometrial cancer (OR = 0.21, 95% CI = 0.13 to 0.34) and ovarian cancer (OR = 0.76, 95% CI = 0.57 to 1.02); and for depot-medroxyprogesterone acetate in endometrial cancer (OR = 0.21, 95% CI = 0.13 to 0.34). Although levonorgestrel intrauterine system, etonogestrel implant and depot-medroxyprogesterone acetate were all associated with increased cancer risk overall, for etonogestrel implant, the risk returned to baseline after cessation, similar to the oral contraceptive pill. We were unable to adjust for all potential confounders, but sensitivity analyses suggested that adjusting for parity, smoking, and obesity would not have materially changed our findings.

Conclusion

Long-acting, reversible contraceptives have similar cancer associations to the oral contraceptive pill (reduced endometrial and ovarian cancer risks and short-term increased breast cancer risk). This information may be helpful to women and their physicians when discussing contraception options.

Introduction

Long-acting, reversible contraceptives, particularly progestin-based long-acting, reversible contraceptives, provide convenient, long-lasting, and highly effective female contraception.¹ They include intrauterine devices, subdermal implants, and depot-medroxyprogesterone acetate injections, and their use has increased substantially over the past 20 years.^2-5 Their advantages include long durations of action, ranging from a few months (injections) up to 7 years (intrauterine devices), and use when taking estrogen is contraindicated or undesirable.

Long-term combined oral contraceptive pill use substantially reduces endometrial and ovarian cancer risks,^6-9 but current or recent users have slightly increased risks of breast and cervical cancers.^10-12 As long-acting, reversible contraceptive use increases, it is important that we understand the potential effects on future cancer rates. There are currently few data on the effects of long-acting, reversible contraceptives on cancer risk, despite increasing use of these agents.

A 2023 meta-analysis reported 20% to 30% increased risks of breast cancer for premenopausal women who used progestogen-only contraceptives, regardless of type, that did not return to baseline until 10 or more years after last use.¹³ Two studies specifically investigating use of the levonorgestrel intrauterine system and ovarian cancer¹⁴^,¹⁵ found a 40% to 50% reduction in risk. Similarly, a pooled analysis reported that use of depot-medroxyprogesterone acetate was associated with a 35% reduction in ovarian cancer risk.¹⁶ Studies of other cancers have been limited, although there are suggestions that the levonorgestrel intrauterine system may also protect against endometrial¹⁷ and cervical cancers.¹⁸

We investigated the associations between use of the levonorgestrel intrauterine system, 68-mg etonogestrel implants, and depot-medroxyprogesterone acetate and overall risk of developing cancer as well as risks of breast, endometrial, ovarian, cervical, and other individual cancers, using linked administrative health data. We compared our results with those for the oral contraceptive pill.

Methods

Study design and participants

We conducted a nested case-control study in a cohort of all Australian women (female sex assigned at birth) enrolled in Medicare, Australia’s universal health insurance scheme, on July 1, 2002.¹⁹ Women entered the cohort on the later of July 1, 2004 (yielding ≥2 years of dispensing history before cancer diagnosis) or their 20th birthday. Women younger than 55 years of age on July 1, 2002 (we were interested in premenopausal hormone use) with no prior cancer were eligible. Their records were linked to dispensing claims from the Pharmaceutical Benefits Scheme (PBS), the Australian Cancer Database, and the National Death Index (Figure S1).

Case and control definition

In total, 176 601 women aged 20 years or older had a first cancer diagnosis between July 1, 2004, and December 31, 2013. Using the cancer diagnosis date as the index date, we matched up to 5 randomly selected control individuals to each case by birth year (±1), state of residence, Socio-Economic Indexes for Areas quintile, and remoteness category (Supplementary Methods).²⁰^,²¹ Women could be selected as a control for more than 1 case, and women with a cancer diagnosis could be selected as controls before their diagnosis. There were 882 999 matched control individuals (Figure 1).

Figure 1. — Flowchart of the selection of case and control individuals for the study. Abbreviation: PBS = Pharmaceutical Benefits Scheme.

Concessional subcohort

We identified a subcohort of women who had been PBS concessional beneficiaries (who receive discounted medicines) for at least 2 years to assess depot-medroxyprogesterone acetate and oral contraceptive pill use because these low-cost medicines were not captured in the PBS for general beneficiaries before 2012 (see Supplementary Materials for details). Women diagnosed with cancer while eligible for the concessional subcohort (n = 47 645) were individually matched to a maximum of 5 control individuals (n = 238 107) who were also eligible for the subcohort at the index date, using the same criteria as noted plus concessional eligibility date (±2 years) (Figure 1).

Exposure variables

We classified women as levonorgestrel intrauterine system, etonogestrel implant, or depot-medroxyprogesterone acetate users (Table S3) at the date of their first dispensing record. Levonorgestrel implants are not available in Australia. We classified women as oral contraceptive pill users at the date of their second dispensing record. Etonogestrel implant and levonorgestrel intrauterine system have been available in Australia only since 2001 and 2002, respectively, so PBS records capture most use; however, we will have misclassified women who only used depot-medroxyprogesterone acetate or oral contraceptive pills before July 2002 as nonusers. We estimated duration of levonorgestrel intrauterine system and etonogestrel implant use as the period from date of first supply until the earlier of the index date or 5 or 3 years (based on clinical guidelines)²²^,²³ after the last dispensing, respectively, recognizing that this would overestimate duration if devices were removed earlier (studies suggest discontinuation rates of 15% to 25% at 1 year²⁴^,²⁵). We similarly estimated duration of depot-medroxyprogesterone acetate and oral contraceptive pill use from first use until the earlier of the index date or 3 months after the last dispensing, recognizing that we would underestimate duration of use before July 2002.

We assigned categories of use by combining duration and recency. Women were defined as recent users if their estimated end of use date was within 2 years of their index date and as former users when it was more than 2 years before the index date. We then categorized recent users as short term (<12 months), medium term (1 to <5 years), and long term (≥5 years). Because we assumed that levonorgestrel intrauterine systems/etonogestrel implants were removed after 5 and 3 years, respectively, women classified as former levonorgestrel intrauterine system/etonogestrel implant users had at least 5 or 3 years of use, respectively. For depot-medroxyprogesterone acetate and oral contraceptive pills, we categorized them separately as short-term former use (<12 months).

Covariates

We used PBS records to define women as users of menopausal hormone therapy if, before their index date, they were dispensed hormone therapy at least twice within an 18-month period. To adjust for potential confounding by comorbidity, we used the weighted Rx-Risk Comorbidity Score²⁶ (Table S5).

Statistical analyses

We used conditional logistic regression to estimate odds ratios (ORs) and 95% confidence intervals (CI) for the associations between long-acting, reversible contraceptives use and the risk of cancer overall and by type: breast, endometrial, cervical, and epithelial ovarian cancers plus colorectal, lung, and thyroid cancers and melanoma for secondary analyses (Table S1); all other cancers were grouped as “other” (Table S2). We initially assessed associations for the levonorgestrel intrauterine system and etonogestrel implants separately in the full cohort, and then included both in the same model. As the levonorgestrel intrauterine system is often used to treat heavy menstrual bleeding,²⁷ a potential symptom of gynecological cancer, we additionally assessed the associations with use for more than 1 year. All models were adjusted for the weighted Rx-Risk Comorbidity Score.

We repeated these models in the concessional subcohort, where we also evaluated depot-medroxyprogesterone acetate and oral contraceptive pills and assessed the effects of adjusting for use of these contraceptives and of menopausal hormone therapy. To evaluate whether the overall risk-benefit profile of long-acting, reversible contraceptives varied with age, we conducted additional analyses of cancer risk overall, stratified by age at diagnosis in 10-year groups. Analyses were performed in SAS, version 9.4, statistical software (SAS Institute Inc).

Sensitivity analyses

We did not have information about potential confounders such as obesity, smoking, and parity for the national dataset; therefore, we conducted sensitivity analyses to assess the likely effects on our results. We first restricted analyses to women in Western Australia, where we had additional hospital morbidity information (99% of case individuals and 98% of control individuals had a hospital record) and could exclude those individuals who had undergone hysterectomy or oophorectomy (where necessary) and additionally adjusted for smoking and parity (Table S4). We then used PBS data to predict obesity²⁸ to assess confounding in the concessional subcohort (Supplementary Methods). We used quantitative bias analysis²⁹ to quantify the potential bias for underreporting of smoking status in hospital records.

Because we did not have information about depot-medroxyprogesterone acetate or oral contraceptive pill use before July 2002, we also conducted sensitivity analyses for breast cancer, first excluding women aged 45 to 54 years on July 1, 2002, because they would be most likely to be former contraceptive users and second, commencing follow-up in 2008 to ensure that we had a minimum of 6 years of prescribing history for all women. In a final sensitivity analysis, we compared exclusive use of each contraceptive with no use of any long-acting, reversible contraceptives or oral contraceptive pill as the reference group.

Results

Characteristics of the full cohort are presented in Table S6. As we excluded women older than 55 years of age in 2002, the oldest women in the study were 67 years of age. Most cancers were diagnosed between the ages of 50 and 59 years (43%) or 40 and 49 years (27%), with only 4% of women diagnosed before they were 30 years of age. Compared with the full cohort, women with cancer in the concessional subcohort (Table S7) tended to live in more disadvantaged areas and be older.

Results for the concessional subcohort for levonorgestrel intrauterine system and etonogestrel implant use were generally consistent with the full cohort, so we focused on the results for the levonorgestrel intrauterine system and etonogestrel implants from the full cohort and depot-medroxyprogesterone acetate from the concessional subcohort.

All cancer

Overall, levonorgestrel intrauterine system use was associated with a 25% increased risk of being diagnosed with cancer (Table 1, Figure 2). This result was partly because of the strong association seen for recent short-term use (OR = 1.69), a likely artefact (see below). When we excluded users of less than 1 year, the association attenuated (OR = 1.16, 95% CI = 1.13 to 1.19). Recent medium-term and long-term use was associated with a 14% to 18% increased risk, with an 11% increase seen for former users (with median time since last use of 3 years).

Table 1.

Association between long-acting, progestin-based contraceptives and all, breast, endometrial, epithelial ovarian, and cervical cancers

Medication used^a	Cases	Controls	OR (95% CI)^b	Cases	Controls	OR (95% CI)^b	Cases	Controls	OR (95% CI)^b	Cases	Controls	OR (95% CI)^b	Cases	Controls	OR (95% CI)^b
	All cancer			Breast cancer			Endometrial cancer			Epithelial ovarian cancer			Cervical cancer
Full cohort, No.	176 601	882 999		67 470	337 347		7290	36 450		4752	23 760		4755	23 775
Levonorgestrel intrauterine system, %
No use	95.5	96.4	(Referent)	95.2	96.2	(Referent)	92.3	97.5	(Referent)	97.1	96.8	(Referent)	95.7	95.3	(Referent)
Any use	4.5	3.6	1.25 (1.21 to 1.28)	4.8	3.8	1.26 (1.21 to 1.31)	7.7	2.5	3.58 (3.19 to 4.02)	2.9	3.2	0.90 (0.75 to 1.08)	4.3	4.7	0.89 (0.77 to 1.04)
Any use (≥1 y)	3.5	3.0	1.16 (1.13 to 1.19)	4.1	3.2	1.28 (1.22 to 1.34)	1.5	2.1	0.80 (0.65 to 0.99)	2.0	2.8	0.71 (0.57 to 0.88)	2.4	3.8	0.62 (0.51 to 0.75)
Use category and duration
Recent: short term	1.0	0.6	1.69 (1.60 to 1.79)	0.7	0.6	1.16 (1.04 to 1.28)	6.2	0.4	21.2 (17.1 to 26.3)	0.9	0.4	2.31 (1.61 to 3.33)	1.9	0.9	2.11 (1.63 to 2.71)
Recent: medium-term	2.3	2.0	1.18 (1.14 to 1.22)	2.6	2.1	1.27 (1.20 to 1.34)	1.0	1.3	0.92 (0.72 to 1.18)	1.4	1.8	0.76 (0.58 to 0.98)	1.8	2.7	0.65 (0.52 to 0.82)
Recent: long term	0.9	0.8	1.14 (1.08 to 1.21)	1.1	0.9	1.33 (1.22 to 1.44)	0.4	0.5	0.82 (0.54 to 1.23)	0.4	0.8	0.53 (0.34 to 0.83)	0.4	0.9	0.46 (0.29 to 0.73)
Former^c	0.3	0.3	1.11 (1.01 to 1.22)	0.4	0.3	1.26 (1.10 to 1.44)	0.1	0.3	0.30 (0.13 to 0.68)	0.2	0.2	0.96 (0.48 to 1.90)	0.2	0.2	0.86 (0.42 to 1.77)
Etonogestrel implant, %
No use	97.6	97.8	(Referent)	97.8	98.2	(Referent)	99.8	99	(Referent)	98.9	98.5	(Referent)	94.7	95.2	(Referent)
Any use	2.4	2.2	1.09 (1.06 to 1.13)	2.2	1.8	1.24 (1.17 to 1.32)	0.2	1.0	0.21 (0.13 to 0.34)	1.1	1.5	0.76 (0.57 to 1.02)	5.3	4.8	1.12 (0.97 to 1.29)
Use category and duration
Recent: short term	0.2	0.2	1.17 (1.05 to 1.31)	0.2	0.1	1.32 (1.07 to 1.63)	—	0.1	—^d	<0.1	<0.2	0.81 (0.24 to 2.78)	0.6	0.4	1.25 (0.82 to 1.92)
Recent: medium term	1.0	0.9	1.11 (1.05 to 1.17)	0.9	0.7	1.30 (1.18 to 1.42)	<0.1	<0.5	0.13 (0.05 to 0.36)	<0.5	<0.8	0.63 (0.40 to 0.98)	2.2	2.2	1.00 (0.80 to 1.24)
Recent: long term	0.4	0.3	1.17 (1.07 to 1.28)	0.4	0.3	1.40 (1.22 to 1.61)	<0.1	<0.3	0.14 (0.03 to 0.64)	<0.2	<0.3	0.82 (0.39 to 1.74)	0.8	0.6	1.37 (0.95 to 1.96)
Former^c	0.8	0.8	1.03 (0.97 to 1.09)	0.8	0.7	1.12 (1.02 to 1.23)	0.2	0.4	0.29 (0.15 to 0.58)	0.4	0.5	0.94 (0.58 to 1.52)	1.7	1.6	1.14 (0.89 to 1.46)
Concessional subcohort, No.	47 645	238 107		15 391	76 912		2206	11 024		1273	6365		1702	8510
Depot-medroxyprogesterone acetate, %
No use	96.2	96.6	(Referent)	97	97	(Referent	99	98.3	(Referent)	97.9	97.4	(Referent)	90.1	92.6	(Referent)
Any use	3.8^e	3.4^e	1.11 (1.05 to 1.17)	3.1	3.0	1.01 (0.91 to 1.12)	1.0	1.7	0.52 (0.32 to 0.84)	2.1	2.6	0.82 (0.54 to 1.25)	9.9	7.4	1.38 (1.15 to 1.67)
Use category and duration
Recent: short term	0.6	0.5	1.10 (0.96 to 1.26)	0.3	0.4	0.89 (0.66 to 1.19)	0.4	0.2	1.61 (0.67 to 3.88)	<0.5	<0.5	0.76 (0.22 to 2.67)	1.8	1.1	1.59 (1.05 to 2.43)
Recent: medium term	0.7	0.6	1.16 (1.03 to 1.31)	0.5	0.5	1.00 (0.78 to 1.28)	<0.3	<0.3	0.54 (0.16 to 1.85)	0.8	0.5	1.53 (0.74 to 3.15)	2.1	1.6	1.33 (0.91 to 1.95)
Recent: long term	0.6	0.5	1.28 (1.12 to 1.47)	0.5	0.4	1.23 (0.95 to 1.59)	<0.3	<0.3	0.25 (0.03 to 1.87)	<0.5	<0.5	0.71 (0.20 to 2.50)	1.5	0.9	1.74 (1.09 to 2.78)
Former: short term	1.2	1.2	1.02 (0.93 to 1.12)	1.1	1.1	0.98 (0.83 to 1.16)	0.3	0.6	0.66 (0.29 to 1.51)	0.5	1.0	0.54 (0.23 to 1.26)	2.8	2.5	1.20 (0.87 to 1.66)
Former: ≥1 y^c	0.8	0.7	1.13 (1.01 to 1.27)	0.7	0.6	1.08 (0.87 to 1.34)	<0.3	<0.5	0.26 (0.06 to 1.11)	<0.5	<0.5	0.80 (0.31 to 2.07)	1.7	1.3	1.33 (0.87 to 2.03)
Oral contraceptive pill
No use	86.8	87.3	(Referent)	86.5	88.5	(Referent)	96.6	93.4	(Referent)	92.2	89.9	(Referent)	73.4	74.7	(Referent)
Any use	13.2^e	12.7^e	1.06 (1.02 to 1.09)	13.5	11.5	1.24 (1.17 to 1.31)	3.4	6.6	0.44 (0.34 to 0.57)	7.8	10.2	0.72 (0.56 to 0.91)	26.6	25.3	1.09 (0.96 to 1.23)
Use category and duration
Recent: short-term	1.2	1.0	1.14 (1.03 to 1.25)	1.0	0.8	1.26 (1.05 to 1.51)	0.6	0.4	1.02 (0.49 to 2.11)	0.8	0.7	1.02 (0.50 to 2.07)	2.5	2.0	1.25 (0.87 to 1.78)
Recent: medium term	3.6	3.1	1.19 (1.13 to 1.26)	3.9	2.7	1.51 (1.37 to 1.67)	0.6	1.4	0.30 (0.16 to 0.56)	2.0	2.6	0.76 (0.49 to 1.18)	7.0	6.0	1.21 (0.97 to 1.50)
Recent: long-term	2.3	1.9	1.27 (1.18 to 1.36)	2.7	1.7	1.70 (1.51 to 1.92)	0.3	0.8	0.24 (0.10 to 0.60)	1.4	1.4	0.89 (0.52 to 1.53)	6.6	4.4	1.51 (1.19 to 1.91)
Former: short-term	2.8	3.0	0.94 (0.88 to 1.00)	2.6	2.8	0.98 (0.88 to 1.10)	1.1	1.7	0.53 (0.33 to 0.85)	2.0	2.5	0.76 (0.49 to 1.18)	4.5	5.8	0.82 (0.63 to 1.05)
Former: ≥1 y^c	3.4	3.7	0.93 (0.88 to 0.98)	3.4	3.5	1.04 (0.94 to 1.14)	0.8	2.3	0.33 (0.20 to 0.55)	1.6	3.0	0.51 (0.31 to 0.82)	6.0	7.1	0.89 (0.71 to 1.12)

Open in a new tab

Abbreviation: OR = odds ratio.

Short-term use: <12 mo; medium-term use: 1 to <5 y; long-term use: ≥5 y.

All models were adjusted for weighted Rx-Risk score at index date; matched by age, Socio-Economic Indexes for Areas, remoteness, and registered state; and included other contraceptives in the same model. Models using the concessional subcohort additionally adjusted for menopausal hormone therapy use.

For former users (≥1 y for depot-medroxyprogesterone acetate and oral contraceptive pill), median (IQR) time in years since last use, levonorgestrel intrauterine system: 3.0 (2.5-3.8); etonogestrel implant: 4.3 (3.0-5.8); depot-medroxyprogesterone acetate: 4.2 (3.0-6.0); oral contraceptive pill: 4.2 (3.0-5.9).

No cases in this category.

Our data includes medicine use only from 2002; therefore, we will have misclassified women who only used depot-medroxyprogesterone acetate or the oral contraceptive pill before July 2002 as nonusers; therefore, women who used it may be underascertained.

Figure 2. — Summary of associations between recent and former contraceptive use and risk of cancer. ^a Possible uncontrolled confounding or detection bias.

Any use of the etonogestrel implant was associated with a 9% increased cancer risk, but this increase was restricted to recent use, regardless of duration (eg, long-term use: OR = 1.17, 95% CI = 1.07 to 1.28), and there was no association with former use. Any depot-medroxyprogesterone acetate use was associated with an 11% increased risk of cancer, with higher risks for individuals with longer-term use (recent long-term use: OR = 1.28, 95% CI = 1.12 to 1.47). The increased risk persisted for former users with more than 12 months of use. Similarly, recent long-term oral contraceptive pill use (≥5 years) was associated with a 27% increase in cancer risk, but former users had no increased risk. There was no consistent variation by age (Table S8).

Breast cancer

Ever use of levonorgestrel intrauterine system or etonogestrel implant was associated with a 26% and 24% increased risk of breast cancer, respectively, compared with no use (Table 1). In the full study population, the risk associated with levonorgestrel intrauterine system was higher for longer duration of use and persisted among former users who, by definition, had at least 5 years of use but at least 2 years of nonuse. Recent etonogestrel implant use was associated with a 30% to 40% increased risk, regardless of duration. The risk was lower but still statistically significantly elevated 2 or more years since stopping (OR = 1.12, 95% CI = 1.02 to 1.23).

In contrast, former depot-medroxyprogesterone acetate use or recent short-term and medium-term use was not associated with risk of breast cancer, but there was the suggestion of an association for recent long-term use (OR = 1.23, 95% CI = 0.95 to 1.59). The results were essentially unaltered when we restricted the cohort to women younger than 45 years of age on July 1, 2002, or women diagnosed after 2008, suggesting that they were not biased by the lack of information about depot-medroxyprogesterone acetate use before 2002 (Table S9). Our sensitivity analysis among women in Western Australia showed that adjustment for smoking attenuated the overall estimate toward the null (Table S10).

We saw the expected associations for the oral contraceptive pill with recent long-term use associated with a 70% increased risk that returned to baseline for women who had stopped use for at least 2 years. Sensitivity analyses did not materially change these results (Tables S9 and S10).

Endometrial cancer

Women who commenced levonorgestrel intrauterine system use within 12 months of the index date had a high risk of being diagnosed with endometrial cancer compared with never users (OR = 21.2, 95% CI = 17.1 to 26.3) (Table 1), likely reflecting use for treatment during work-up for endometrial cancer rather than a causal effect. Use of more than 1 year was associated with reduced risk (OR = 0.80, 95% CI = 0.65 to 0.99), with a 70% reduced risk for former users (7 years since last dispensing) (OR = 0.30, 95% CI = 0.13 to 0.68).

Etonogestrel implant and depot-medroxyprogesterone acetate use was associated with substantially reduced risks of endometrial cancer compared with never use (OR = 0.21, 95% CI = 0.13 to 0.34, and OR = 0.52, 95% CI = 0.32 to 0.84, respectively) that persisted among former users. These associations were similar to those seen for the oral contraceptive pill.

Epithelial ovarian cancer

As for endometrial cancer, women who commenced levonorgestrel intrauterine system use within 12 months of the index date had an increased risk of epithelial ovarian cancer compared with never users (OR = 2.31, 95% CI = 1.61 to 3.33), but long-term use was associated with a 47% reduced risk (Table 1). Recent use of the etonogestrel implant was also associated with lower risk (eg, medium-term use: OR = 0.63, 95% CI = 0.40 to 0.98). For both the levonorgestrel intrauterine system and the etonogestrel implant, risk returned close to baseline among former users. We did not find any clear associations with depot-medroxyprogesterone acetate, but most risk estimates were below 1.0 (Table 2). Oral contraceptive pill use was associated with a reduced risk that persisted after cessation.

Table 2.

Association between long-acting, progestin-based contraceptives and colorectal, lung, melanoma, thyroid, and other cancers.

Medication used^a	Cases	Controls	OR (95% CI)^b	Cases	Controls	OR (95% CI)^b	Cases	Controls	OR (95% CI)^b	Cases	Controls	OR (95% CI)^b	Cases	Controls	OR (95% CI)^b
	Colorectal cancer			Lung cancer			Melanoma			Thyroid cancer			Other cancers
Full cohort, No.	14 558	72 790		8505	42 521		23 171	115 853		9578	47 890		36 523	182 614
Levonorgestrel intrauterine system, %
No use	96.4	96.9	(Referent)	98.1	97.5	(Referent)	95.0	95.9	(Referent)	93.9	95.5	(Referent)	96.3	96.7	(Referent)
Any use	3.6	3.1	1.17 (1.06 to 1.29)	1.9	2.5	0.73 (0.61 to 0.86)	5.0	4.1	1.23 (1.15 to 1.32)	6.1	4.5	1.38 (1.25 to 1.52)	3.7	3.3	1.11 (1.04 to 1.18)
Use category and duration
Recent: short term	0.6	0.5	1.16 (0.92 to 1.48)	0.3	0.3	0.74 (0.46 to 1.16)	0.9	0.7	1.20 (1.03 to 1.40)	1.1	0.8	1.30 (1.04 to 1.61)	0.8	0.5	1.44 (1.26 to 1.64)
Recent: medium term	2.0	1.6	1.24 (1.09 to 1.42)	1.0	1.3	0.74 (0.59 to 0.94)	2.9	2.2	1.31 (1.20 to 1.43)	3.3	2.4	1.45 (1.28 to 1.65)	1.9	1.8	1.03 (0.95 to 1.12)
Recent: long-term	0.7	0.7	1.05 (0.85 to 1.31)	0.5	0.6	0.74 (0.53 to 1.04)	1.0	0.9	1.15 (0.99 to 1.33)	1.3	1.0	1.38 (1.13 to 1.68)	0.7	0.7	1.02 (0.89 to 1.17)
Former	0.3	0.3	1.11 (0.81 to 1.52)	0.2	0.3	0.58 (0.33 to 1.01)	0.3	0.3	1.07 (0.83 to 1.38)	0.4	0.3	1.09 (0.75 to 1.59)	0.3	0.3	1.18 (0.97 to 1.45)
Etonogestrel implant, %
No use	98.2	98.2	(Referent)	99.0	99.0	(Referent)	96.5	96.7	(Referent)	96.1	96.5	(Referent)	97.7	97.8	(Referent)
Any use	1.8	1.8	1.04 (0.91 to 1.19)	1.0	1.0	0.97 (0.76 to 1.25)	3.5	3.3	1.07 (0.99 to 1.16)	3.9	3.5	1.09 (0.97 to 1.22)	2.3	2.2	1.04 (0.97 to 1.13)
Use category and duration
Recent: short term	0.1	0.2	0.72 (0.44 to 1.18)	0.1	0.1	1.46 (0.59 to 3.63)	0.4	0.3	1.27 (1.00 to 1.62)	0.4	0.3	1.47 (1.04 to 2.07)	0.2	0.2	1.02 (0.80 to 1.31)
Recent: medium term	0.8	0.7	1.16 (0.94 to 1.42)	0.3	0.4	0.84 (0.56 to 1.26)	1.5	1.5	1.03 (0.92 to 1.16)	1.8	1.5	1.21 (1.02 to 1.43)	1.0	0.9	1.08 (0.96 to 1.21)
Recent: long term	0.3	0.2	1.34 (0.97 to 1.86)	0.1	0.2	0.47 (0.22 to 1.02)	0.5	0.5	1.17 (0.96 to 1.43)	0.4	0.5	0.81 (0.58 to 1.14)	0.4	0.3	1.09 (0.90 to 1.32)
Former	0.7	0.6	0.89 (0.70 to 1.12)	0.5	0.4	1.31 (0.91 to 1.87)	1.1	1.1	1.02 (0.89 to 1.17)	1.2	1.2	0.96 (0.78 to 1.17)	0.7	0.7	0.99 (0.86 to 1.13)
Concessional subcohort	4124	20 618		3525	17 617		5346	26 713		2454	12 260		11 412	57 046
Depot-medroxyprogesterone acetate, %
No use	96.9	97.7	(Referent)	97.2	98.0	(Referent)	94.6	94.8	(Referent)	94.4	94.9	(Referent)	95.8	96.5	(Referent)
Any use	3.1	2.3	1.34 (1.09 to 1.65)	2.8	2.0	1.36 (1.08 to 1.73)	5.4	5.2	1.04 (0.91 to 1.19)	5.6	5.1	1.06 (0.87 to 1.29)	4.2	3.5	1.20 (1.08 to 1.33)
Use category and duration
Recent: short term	0.3	0.3	0.92 (0.48 to 1.76)	0.3	0.2	1.98 (0.99 to 3.96)	0.8	0.9	0.89 (0.64 to 1.24)	0.9	0.9	1.01 (0.64 to 1.59)	0.7	0.5	1.34 (1.04 to 1.73)
Recent: medium term	0.5	0.4	1.07 (0.65 to 1.76)	0.3	0.3	1.04 (0.55 to 1.97)	1.2	1.1	1.17 (0.89 to 1.54)	1.1	0.9	1.25 (0.82 to 1.92)	0.9	0.7	1.33 (1.06 to 1.65)
Recent: long-term	0.5	0.3	1.75 (1.07 to 2.86)	0.6	0.3	1.63 (0.96 to 2.76)	0.6	0.6	1.04 (0.72 to 1.52)	0.8	0.7	1.18 (0.71 to 1.96)	0.6	0.4	1.41 (1.08 to 1.84)
Former: short term	0.9	0.5	1.18 (0.82 to 1.69)	0.9	0.7	1.39 (0.93 to 2.08)	1.6	1.7	0.90 (0.71 to 1.15)	1.9	1.7	1.07 (0.77 to 1.49)	0.7	0.6	1.08 (0.90 to 1.29)
Former: ≥1 y	0.9	0.8	1.98 (1.33 to 2.94)	0.6	0.5	1.11 (0.67 to 1.80)	1.2	0.9	1.34 (1.01 to 1.78)	0.9	1.0	0.83 (0.52 to 1.34)	1.3	1.2	1.04 (0.81 to 1.33)
Oral contraceptive pill, %
No use	90.3	90.5	(Referent)	93.8	92.4	(Referent)	78.4	81.5	(Referent)	81.3	81.0	(Referent)	88.2	87.4	(Referent)
Any use	9.7	9.5	1.04 (0.92 to 1.18)	6.2	7.6	0.80 (0.68 to 0.93)	21.6	18.5	1.27 (1.17 to 1.37)	18.7	19.0	0.97 (0.86 to 1.10)	11.8	12.6	0.91 (0.85 to 0.98)
Use category and duration
Recent: short term	0.6	0.8	0.81 (0.53 to 1.23)	0.6	0.4	1.33 (0.80 to 2.21)	2.2	1.9	1.24 (1.00 to 1.53)	1.8	1.7	1.07 (0.77 to 1.49)	1.1	1.1	1.03 (0.84 to 1.25)
Recent: medium term	2.6	2.1	1.22 (0.98 to 1.53)	1.6	1.6	1.00 (0.74 to 1.36)	5.9	5.0	1.29 (1.13 to 1.47)	5.1	4.7	1.09 (0.88 to 1.34)	3.1	3.2	0.98 (0.87 to 1.11)
Recent: long term	1.6	1.4	1.17 (0.88 to 1.55)	0.8	1.0	0.77 (0.51 to 1.16)	3.3	2.6	1.39 (1.17 to 1.66)	2.7	2.9	0.93 (0.71 to 1.23)	1.9	1.8	1.01 (0.87 to 1.18)
Former: short term	2.2	2.2	1.05 (0.83 to 1.32)	1.2	2.0	0.63 (0.46 to 0.86)	4.7	4.2	1.22 (1.05 to 1.41)	4.3	4.6	0.93 (0.75 to 1.15)	2.6	2.9	0.87 (0.76 to 0.99)
Former: ≥1 y	2.7	3.0	0.93 (0.75 to 1.15)	1.9	2.7	0.74 (0.57 to 0.96)	5.5	4.8	1.26 (1.10 to 1.44)	4.7	5.1	0.91 (0.73 to 1.12)	3.0	3.6	0.82 (0.73 to 0.92)

Open in a new tab

Abbreviation: OR = odds ratio.

Short-term use: <12 mo; medium-term use: 1 to <5 y; long-term use: ≥5 y.

All models were adjusted for weighted Rx-Risk score at index date; matched by age, Socio-Economic Indexes for Areas, remoteness, and registered state; and included other contraceptives in the same model. Models using the concessional subcohort were additionally adjusted for menopausal hormone therapy use.

Cervical cancer

Recent commencement of levonorgestrel intrauterine system use was also associated with an increased risk of cervical cancer (OR = 2.11, 95% CI = 1.63 to 2.71) (Table 1). After 12 months of use, however, the risk was reduced (OR = 0.62, 95% CI = 0.51 to 0.75). For the etonogestrel implant, ever use was associated with a small but not statistically significant increased risk (OR = 1.12, 95% CI = 0.97 to 1.29), with a stronger association seen for depot-medroxyprogesterone acetate (OR = 1.38, 95% CI = 1.15 to 1.67). The sensitivity analysis among women in Western Australia showed that adjustment for smoking attenuated the estimate by 0.05 for the etonogestrel implant and 0.1 to 0.2 for depot-medroxyprogesterone acetate. Recent long-term but not former oral contraceptive pill use was also associated with an increased risk.

Other cancers

To understand the full potential effects of long-acting, reversible contraceptives, we also evaluated the associations for other common cancers (Table 2), with the results for the full cohort shown for the levonorgestrel intrauterine system and the etonogestrel implant and for the concessional subcohort for depot-medroxyprogesterone acetate and the oral contraceptive pill. Ever use of the levonorgestrel intrauterine system was associated with an increased risk of colorectal cancer, melanoma, thyroid cancer, and all other noncategorized cancers combined but a reduced risk of lung cancer.

Overall, use of the etonogestrel implant was not associated with other cancers, although recent short-term use was associated with an increased risk of melanoma, and recent short-term and medium-term use was associated with increased thyroid cancer risk. Use of depot-medroxyprogesterone acetate was associated with statistically significantly increased risks of colorectal cancer, lung cancer, and all other cancers combined. Our sensitivity analysis using women in Western Australia showed an attenuation toward the null for lung cancer when we adjusted for smoking (Table S10). Use of the oral contraceptive pill was associated with an increased risk of melanoma and a reduced risk of lung cancer and all other cancers combined, particularly for former users.

Other sensitivity analyses

Unless otherwise noted above, our sensitivity analyses adjusting for smoking, parity, and hysterectomy and oophorectomy and for predicted obesity did not materially change the results (Tables S10 and S11). For the Western Australia cohort, 35% had smoking recorded in a hospital record (33% for controls only). The proportion of Australian adult women who ever smoked ranged from 45% in 2002 to 37% in 2013.³⁰ Quantitative bias analysis assuming a 20% underreporting of smoking and the known associations between smoking and each cancer did not alter the smoking-adjusted estimates by more than 0.01. Our sensitivity analysis with no long-acting, reversible contraceptive/oral contraceptive pill as the reference (Table S12) produced estimates consistent with the concessional subcohort analysis results for breast cancer (Table S9).

Discussion

Our study shows that all long-acting, reversible contraceptives, particularly recent use, are associated with overall increased cancer risk that is of similar or greater magnitude to the oral contraceptive pill. These associations were primarily driven by the increased breast cancer risks among recent users, with some attenuation from the reduced risks observed for gynecological cancers. Associations differed between the types of long-acting, reversible contraceptives for specific cancers and when duration and timing of use were considered.

Recent levonorgestrel intrauterine system and etonogestrel implant use was associated with 20% to 40% increased breast cancer risk, and use of 12 months or longer was associated with decreased endometrial and epithelial ovarian cancer risk. The high risk of endometrial cancer for women with less than 12 months of levonorgestrel intrauterine system use probably reflects the use of the levonorgestrel intrauterine system to treat hyperplasia and symptoms of early-stage endometrial cancer.²⁷^,³¹ Similar phenomena may also explain the more modest short-term increases in ovarian and cervical cancer risk; for example, cervical cancer screening is often performed before levonorgestrel intrauterine system insertion.³² Recent (>1 year) levonorgestrel intrauterine system use but not etonogestrel implant or depot-medroxyprogesterone acetate use was associated with decreased cervical cancer risk. Recent levonorgestrel intrauterine system and etonogestrel implant use but not depot-medroxyprogesterone acetate use was also associated with increased risk of thyroid cancer and melanoma. Both cancers are subject to overdiagnosis,³³ so observed associations could reflect increased testing among women who choose these contraceptives. Some of this variation may be caused by confounding by factors that increase cancer risk and the type of long-acting, reversible contraceptive used. Except for breast cancer, there was no increased risk of cancer among former levonorgestrel intrauterine system or etonogestrel implant users, and substantial reductions in risk were seen for endometrial cancer. Consequently, there was no association between former etonogestrel implant users (>2 years since last use) and cancer risk overall; however, we did not see the same risk reduction as we did for former oral contraceptive pill users. Former levonorgestrel intrauterine system users still had an 11% increased risk of developing cancer, but longer follow-up is needed to determine whether this association will weaken further with longer periods since stopping use.

In contrast, depot-medroxyprogesterone acetate was not associated with breast cancer but was associated with a reduced risk of endometrial cancer and increased risks of cervical, colorectal, lung, and the combined group of “other” cancers, although it is likely that the apparent associations with lung cancer (and perhaps other cancers) were at least partially due to uncontrolled confounding by smoking. One explanation for the breast cancer association is that depot-medroxyprogesterone acetate suppresses ovarian function and circulating estrogen to a greater extent than the levonorgestrel intrauterine system and etonogestrel implant,^34-36 which may mean that synergistic effects of estrogen and progestin on breast epithelial cell proliferation may be reduced during depot-medroxyprogesterone acetate use.³⁷

Our findings are consistent with prior studies showing associations between long-acting, reversible contraceptive use and increased breast cancer¹³^,³⁸ and decreased ovarian cancer risk^14-16; however, we also investigated other cancer types as well as duration and timing of use. We compared the associations with long-acting, reversible contraceptives with those for the oral contraceptive pill; these associations were consistent with prior studies showing a short-term increase for breast cancer but long-term reductions in risk for endometrial and epithelial ovarian cancers.⁸^,¹⁰ With the exception of melanoma, there was no increased risk of other specific cancers among former users and, as a result, former oral contraceptive pill use was associated with a 7% reduced risk of all cancers combined. A previous study of oral contraceptive pill use and melanoma noted that sunburn and use of sunbeds were higher among oral contraceptive pill users,³⁹ suggesting that this association could be the result of residual confounding.

To our knowledge, this is the first study to investigate the timing and duration of levonorgestrel intrauterine system, etonogestrel implant, and depot-medroxyprogesterone acetate use as well as associations with overall cancer risk and a wide range of individual cancer types. We had a large sample size and accurate medication dispensing histories. Our findings highlighted that timing and duration of use are important when considering associations between long-acting, reversible contraceptive use and cancer risk.

The major limitations of our study are the lack of data for depot-medroxyprogesterone acetate and oral contraceptive pill use before July 2002 and the limited information about potential confounders. In the early years of follow-up, this limitation will have led to some long-term depot-medroxyprogesterone acetate and oral contraceptive pill users being misclassified as short-term users and former users as never users. Our sensitivity analyses among subgroups where misclassification would be less likely did not materially alter our results, suggesting that they were not seriously biased. We could not determine whether the levonorgestrel intrauterine system or etonogestrel implant were removed early, so we may have overestimated durations of use, hence underestimating the strength of associations. Confounding by indication for each long-acting, reversible contraceptive is possible, and a woman’s reason for using a specific long-acting, reversible contraceptive may be related to her cancer risk. Our sensitivity analyses among subgroups for whom we had information about smoking, parity, and obesity, however, showed that adjusting for these factors in our main analysis would have been unlikely to materially change our findings. Although underreporting of smoking status from hospital records may have affected our ability to adjust for smoking status, our quantitative bias analysis suggests that this was not a major issue. Multiple comparisons were made in our analyses; therefore, it is possible that some associations arose by chance. We were unable to investigate associations for subtypes of hormonally driven cancer due to data limitations (breast cancer) or small numbers (endometrial and ovarian cancers). Finally, the relatively recent introduction of the levonorgestrel intrauterine system and etonogestrel implant means that we did not have information for women older than 70 years of age at cancer diagnosis and so could not explore the potential for further long-term reductions in gynecological cancer risk. Follow-up beyond 10 years from initial use is required before conclusions can be drawn about long-term harms and benefits.

Our study suggests that, like the oral contraceptive pill, long-acting, reversible contraceptives—particularly the levonorgestrel intrauterine system and etonogestrel implant—are associated with reduced endometrial and ovarian cancer risk and increased breast cancer risk. This finding suggests that temporal changes in contraceptive use are unlikely to have a substantial impact on rates of cancer affected by contraceptive use. This information may be helpful to women and their physicians when discussing contraception options.

Supplementary Material

djae282_Supplementary_Data

djae282_supplementary_data.pdf^{(488.6KB, pdf)}

Acknowledgments

The authors thank the staff from the Australian Institute of Health and Welfare for data linkage and custodians of the Medicare Enrolments File, PBS, Australian Cancer Database, National Death Index, and state-based cancer registries. We also thank the staff at the Western Australian Data Linkage Branch and staff and custodians of the Electoral Roll, Emergency Department Data Collection, Hospital Morbidity Data Collection, Midwives Notification System, and Birth Registrations. The funder had no role in the design of the study; the collection, analysis and interpretation of the data; the writing of the manuscript; or the decision to submit the manuscript for publication.

Contributor Information

Karen M Tuesley, School of Public Health, University of Queensland, Brisbane, QLD 4006, Australia; Population Health Program, QIMR Berghofer Medical Research Institute, Brisbane, QLD 4006, Australia.

Katrina Spilsbury, Institute for Health Research, The University of Notre Dame Australia, Fremantle, WA 6160, Australia.

Sallie-Anne Pearson, School of Population Health, University of New South Wales, Sydney, NSW 2052, Australia; Centre of Research Excellence in Medicines Intelligence, University of New South Wales, Sydney, NSW 2052, Australia.

Peter Donovan, Clinical Pharmacology Department, Royal Brisbane and Women’s Hospital, Brisbane, QLD 4006, Australia; Faculty of Medicine, University of Queensland, Brisbane, QLD 4006, Australia.

Andreas Obermair, Queensland Centre for Gynaecological Cancer, Royal Brisbane and Women’s Hospital, Brisbane, QLD 4006, Australia; Centre for Clinical Research, University of Queensland, Brisbane, QLD 4006, Australia.

Michael D Coory, Mater Research Institute, University of Queensland, Brisbane, QLD 4101, Australia.

Sitwat Ali, School of Public Health, University of Queensland, Brisbane, QLD 4006, Australia; Population Health Program, QIMR Berghofer Medical Research Institute, Brisbane, QLD 4006, Australia.

Nirmala Pandeya, School of Public Health, University of Queensland, Brisbane, QLD 4006, Australia; Population Health Program, QIMR Berghofer Medical Research Institute, Brisbane, QLD 4006, Australia.

Louise Stewart, School of Population and Global Health, The University of Western Australia, Perth, WA 6009, Australia.

Susan J Jordan, School of Public Health, University of Queensland, Brisbane, QLD 4006, Australia; Population Health Program, QIMR Berghofer Medical Research Institute, Brisbane, QLD 4006, Australia.

Penelope M Webb, School of Public Health, University of Queensland, Brisbane, QLD 4006, Australia; Population Health Program, QIMR Berghofer Medical Research Institute, Brisbane, QLD 4006, Australia.

Author contributions

Karen M. Tuesley, PhD (Formal analysis; Investigation; Methodology; Writing – original draft; Writing – review & editing), Katrina Spilsbury, PhD (Formal analysis; Funding acquisition; Investigation; Methodology; Writing – review & editing), Sallie-Anne Pearson, PhD (Funding acquisition; Investigation; Writing – review & editing), Peter Donovan, MSc (Funding acquisition; Investigation; Writing – review & editing), Andreas Obermair, MD (Investigation; Writing – review & editing), Michael Coory, PhD (Funding acquisition; Investigation; Writing – review & editing), Sitwat Ali, PhD (Investigation; Writing – review & editing), Nirmala Pandeya, PhD (Funding acquisition; Investigation; Writing – review & editing), Louise Stewart, PhD (Funding acquisition; Investigation; Writing – review & editing), Susan Jordan, PhD (Conceptualization; Formal analysis; Funding acquisition; Investigation; Methodology; Supervision; Writing – review & editing), Penelope Webb, DPhil (Conceptualization; Formal analysis; Funding acquisition; Investigation; Methodology; Supervision; Writing – review & editing).

Supplementary material

Supplementary material is available at JNCI: Journal of the National Cancer Institute online.

Funding

This work was supported by the Australian National Health and Medical Research Council (GNT1121151, GNT1173346 to P.M.W.).

Conflicts of interest

P.M.W. received grant funding from AstraZeneca for an unrelated study of ovarian cancer and a speaker’s fee from AstraZeneca (November 2021). All remaining authors have declared no conflicts of interest.

Data availability

The data underlying this article cannot be shared for the privacy of individuals who participated in the study. Summary-level data can be shared on request to the corresponding author, K.M.T. Individual-level data can be accessed only from within Australia by named project staff with data custodian and Human Research Ethics Committee approval. Please contact K.M.T. for further information.

Ethical approval

The study was approved by the following Human Research Ethics Committees: the University of Queensland; the QIMR Berghofer Medical Research Institute; the Australian Institute of Health and Welfare; Australian Capital Territory Health on April 3, 2017; New South Wales Population and Health Services; Western Australia Department of Health; Queensland Department of Health; University of Tasmania; and Northern Territory Department of Health.

References

1. Bahamondes L, Fernandes A, Monteiro I, Bahamondes MV.. Long-acting reversible contraceptive (LARCs) methods. Best Pract Res Clin Obstet Gynaecol. 2020;66:28-40. [DOI] [PubMed] [Google Scholar]
2. Grzeskowiak LE, Calabretto H, Amos N, Mazza D, Ilomaki J.. Changes in use of hormonal long-acting reversible contraceptive methods in Australia between 2006 and 2018: a population-based study. Aust N Z J Obstet Gynaecol. 2021;61:128-134. [DOI] [PubMed] [Google Scholar]
3. Joshi R, Khadilkar S, Patel M.. Global trends in use of long-acting reversible and permanent methods of contraception: seeking a balance. Int J Gynaecol Obstet. 2015;131(suppl 1):S60-S63. [DOI] [PubMed] [Google Scholar]
4. Eeckhaut MCW, Rendall MS, Zvavitch P.. Women's use of long-acting reversible contraception for birth timing and birth stopping. Demography. 2021;58:1327-1346. [DOI] [PMC free article] [PubMed] [Google Scholar]
5. Hellstrom A, Gemzell Danielsson K, Kopp Kallner H.. Trends in use and attitudes towards contraception in Sweden: results of a nationwide survey. Eur J Contracept Reprod Health Care. 2019;24:154-160. [DOI] [PubMed] [Google Scholar]
6. Iversen L, Sivasubramaniam S, Lee AJ, Fielding S, Hannaford PC.. Lifetime cancer risk and combined oral contraceptives: the Royal College of General Practitioners' Oral Contraception Study. Am J Obstet Gynecol. 2017;216:580.e1-e9. [DOI] [PubMed] [Google Scholar]
7. Wilson LF, Antonsson A, Green AC, et al. How many cancer cases and deaths are potentially preventable? Estimates for Australia in 2013. Int J Cancer. 2018;142:691-701. [DOI] [PubMed] [Google Scholar]
8. Beral V, Doll R, Hermon C, Peto R, Reeves G; Collaborative Group on Epidemiological Studies of Ovarian Cancer. Ovarian cancer and oral contraceptives: collaborative reanalysis of data from 45 epidemiological studies including 23,257 women with ovarian cancer and 87,303 controls. Lancet. 2008;371:303-314. [DOI] [PubMed] [Google Scholar]
9. Collaborative Group on Epidemiological Studies on Endometrial Cancer. Endometrial cancer and oral contraceptives: an individual participant meta-analysis of 27 276 women with endometrial cancer from 36 epidemiological studies. Lancet Oncol. 2015;16:1061-1070. [DOI] [PubMed] [Google Scholar]
10. Karlsson T, Johansson T, Hoglund J, Ek WE, Johansson A.. Time-dependent effects of oral contraceptive use on breast, ovarian, and endometrial cancers. Cancer Res. 2021;81:1153-1162. [DOI] [PubMed] [Google Scholar]
11. Breast cancer and hormonal contraceptives: further results. Collaborative Group on Hormonal Factors in Breast Cancer. Contraception. 1996;54(suppl 3):1S-106S. [DOI] [PubMed] [Google Scholar]
12. Appleby P, Beral V, Berrington de Gonzalez A, et al. ; International Collaboration of Epidemiological Studies of Cervical Cancer. Cervical cancer and hormonal contraceptives: collaborative reanalysis of individual data for 16,573 women with cervical cancer and 35,509 women without cervical cancer from 24 epidemiological studies. Lancet. 2007;370:1609-1621. [DOI] [PubMed] [Google Scholar]
13. Fitzpatrick D, Pirie K, Reeves G, Green J, Beral V.. Combined and progestogen-only hormonal contraceptives and breast cancer risk: a UK nested case-control study and meta-analysis. PLoS Med. 2023;20:e1004188. [DOI] [PMC free article] [PubMed] [Google Scholar]
14. Jareid M, Thalabard JC, Aarflot M, Bøvelstad HM, Lund E, Braaten T.. Levonorgestrel-releasing intrauterine system use is associated with a decreased risk of ovarian and endometrial cancer, without increased risk of breast cancer. Results from the NOWAC Study. Gynecol Oncol. 2018;149:127-132. [DOI] [PubMed] [Google Scholar]
15. Soini T, Hurskainen R, Grénman S, Mäenpää J, Paavonen J, Pukkala E.. Impact of levonorgestrel-releasing intrauterine system use on the cancer risk of the ovary and fallopian tube. Acta Oncol. 2016;55:1281-1284. [DOI] [PubMed] [Google Scholar]
16. Phung MT, Lee AW, Wu AH, et al. ; Ovarian Cancer Association Consortium. Depot-medroxyprogesterone acetate use is associated with decreased risk of ovarian cancer: the mounting evidence of a protective role of progestins. Cancer Epidemiol Biomarkers Prev. 2021;30:927-935. [DOI] [PMC free article] [PubMed] [Google Scholar]
17. Soini T, Hurskainen R, Grénman S, Mäenpää J, Paavonen J, Pukkala E.. Cancer risk in women using the levonorgestrel-releasing intrauterine system in Finland. Obstet Gynecol. 2014;124:292-299. [DOI] [PubMed] [Google Scholar]
18. Iversen L, Fielding S, Lidegaard Ø, Hannaford PC.. Contemporary hormonal contraception and cervical cancer in women of reproductive age. Intl Journal of Cancer. 2021;149:769-777. [DOI] [PubMed] [Google Scholar]
19. Tuesley KM, Webb PM, Protani MM, et al. Nitrogen-based bisphosphonate use and ovarian cancer risk in women aged 50 years and older. J Natl Cancer Inst. 2022;114:878-884. [DOI] [PMC free article] [PubMed] [Google Scholar]
20. Australian Bureau of Statistics. Socio-Economic Indexes for Areas (SEIFA) Australia, 2011. Australian Bureau of Statistics; 2011. Accessed September 19, 2019. http://www.abs.gov.au/websitedbs/censushome.nsf/home/seifa2011?opendocument&navpos=260
21. Australian Bureau of Statistics. Geography Publications. 2017. Accessed September 19, 2019. http://www.abs.gov.au/websitedbs/D3310114.nsf/home/ABS+Geography+Publications
22. The Royal Australian College of General Practitioners (RACGP). Intrauterine devices/systems—resources for general practice. 2024. https://www.racgp.org.au/clinical-resources/clinical-guidelines/key-racgp-guidelines/view-all-racgp-guidelines/iuds-systems-resources-for-general-practice
23. Pearson S, Stewart M, Bateson D. Implanon NXT: expert tips for best-practice insertion and removal. Aust J Gen Pract. 2017;46:104-108. [PubMed] [Google Scholar]
24. Law A, Liao L, Lin J, Yaldo A, Lynen R.. Twelve-month discontinuation rates of levonorgestrel intrauterine system 13.5 mg and subdermal etonogestrel implant in women aged 18–44: A retrospective claims database analysis. Contraception. 2018;98:120-124. [DOI] [PubMed] [Google Scholar]
25. Saloranta TH, Gyllenberg FK, But A, Gissler M, Laine MK, Heikinheimo O.. Free-of-charge long-acting reversible contraception: two-year discontinuation, its risk factors, and reasons. Am J Obstet Gynecol. 2020;223:886.e1-886.e17. [DOI] [PubMed] [Google Scholar]
26. Lu CY, Barratt J, Vitry A, Roughead E.. Charlson and Rx-Risk comorbidity indices were predictive of mortality in the Australian health care setting. J Clin Epidemiol. 2011;64:223-228. [DOI] [PubMed] [Google Scholar]
27. Dhamangaonkar PC, Anuradha K, Saxena A.. Levonorgestrel intrauterine system (Mirena): an emerging tool for conservative treatment of abnormal uterine bleeding. J Midlife Health. 2015;6:26-30. [DOI] [PMC free article] [PubMed] [Google Scholar]
28. Ali S, Na R, Waterhouse M, et al. Predicting obesity and smoking using medication data: a machine-learning approach. Pharmacoepidemiol Drug Saf. 2022;31:91-99. [DOI] [PubMed] [Google Scholar]
29. Lash TL, Fox MP, MacLehose RF, Maldonado G, McCandless LC, Greenland S.. Good practices for quantitative bias analysis. Int J Epidemiol. 2014;43:1969-1985. [DOI] [PubMed] [Google Scholar]
30. Australian Institute of Health and Welfare (AIHW). National Drug Strategy Household Survey 2022–2023. 2024. Accessed October 18, 2024. https://www.aihw.gov.au/reports/illicit-use-of-drugs/national-drug-strategy-household-survey/data
31. Dore M, Filoche S, Danielson K, Henry C.. Efficacy of the LNG-IUS for treatment of endometrial hyperplasia and early stage endometrial cancer: can biomarkers predict response? Gynecol Oncol Rep. 2021;36:100732. [DOI] [PMC free article] [PubMed] [Google Scholar]
32. NPS Medicinewise. Mirena Levonorgestrel consumer medicine information leaflet: MIMS. November 2021. https://www.nps.org.au/medicine-finder/mirena-intrauterine-device
33. Glasziou PP, Jones MA, Pathirana T, Barratt AL, Bell KJ.. Estimating the magnitude of cancer overdiagnosis in Australia. Med J Aust. 2020;212:163-168. [DOI] [PMC free article] [PubMed] [Google Scholar]
34. Kailasam C, Cahill D.. Review of the safety, efficacy and patient acceptability of the levonorgestrel-releasing intrauterine system. Patient Prefer Adherence. 2008;2:293-302. [DOI] [PMC free article] [PubMed] [Google Scholar]
35. Family Planning NSW. Contraception in Australia 2005-2018. Ashfield; 2020. [Google Scholar]
36. Kaunitz AM. Long-acting injectable contraception with depot medroxyprogesterone acetate. Am J Obstet Gynecol. 1994;170:1543-1549. [PubMed] [Google Scholar]
37. Trabert B, Sherman ME, Kannan N, Stanczyk FZ.. Progesterone and breast cancer. Endocr Rev. 2020;41:320-344. [DOI] [PMC free article] [PubMed] [Google Scholar]
38. Conz L, Mota BS, Bahamondes L, et al. Levonorgestrel-releasing intrauterine system and breast cancer risk: a systematic review and meta-analysis. Acta Obstet Gynecol Scand. 2020;99:970-982. [DOI] [PubMed] [Google Scholar]
39. Cervenka I, Mahamat-Saleh Y, Savoye I, et al. Oral contraceptive use and cutaneous melanoma risk: a French prospective cohort study. Int J Cancer. 2018;143:2390-2399. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

djae282_Supplementary_Data

djae282_supplementary_data.pdf^{(488.6KB, pdf)}

Data Availability Statement

[djae282-B1] 1. Bahamondes L, Fernandes A, Monteiro I, Bahamondes MV.. Long-acting reversible contraceptive (LARCs) methods. Best Pract Res Clin Obstet Gynaecol. 2020;66:28-40. [DOI] [PubMed] [Google Scholar]

[djae282-B2] 2. Grzeskowiak LE, Calabretto H, Amos N, Mazza D, Ilomaki J.. Changes in use of hormonal long-acting reversible contraceptive methods in Australia between 2006 and 2018: a population-based study. Aust N Z J Obstet Gynaecol. 2021;61:128-134. [DOI] [PubMed] [Google Scholar]

[djae282-B3] 3. Joshi R, Khadilkar S, Patel M.. Global trends in use of long-acting reversible and permanent methods of contraception: seeking a balance. Int J Gynaecol Obstet. 2015;131(suppl 1):S60-S63. [DOI] [PubMed] [Google Scholar]

[djae282-B4] 4. Eeckhaut MCW, Rendall MS, Zvavitch P.. Women's use of long-acting reversible contraception for birth timing and birth stopping. Demography. 2021;58:1327-1346. [DOI] [PMC free article] [PubMed] [Google Scholar]

[djae282-B5] 5. Hellstrom A, Gemzell Danielsson K, Kopp Kallner H.. Trends in use and attitudes towards contraception in Sweden: results of a nationwide survey. Eur J Contracept Reprod Health Care. 2019;24:154-160. [DOI] [PubMed] [Google Scholar]

[djae282-B6] 6. Iversen L, Sivasubramaniam S, Lee AJ, Fielding S, Hannaford PC.. Lifetime cancer risk and combined oral contraceptives: the Royal College of General Practitioners' Oral Contraception Study. Am J Obstet Gynecol. 2017;216:580.e1-e9. [DOI] [PubMed] [Google Scholar]

[djae282-B7] 7. Wilson LF, Antonsson A, Green AC, et al. How many cancer cases and deaths are potentially preventable? Estimates for Australia in 2013. Int J Cancer. 2018;142:691-701. [DOI] [PubMed] [Google Scholar]

[djae282-B8] 8. Beral V, Doll R, Hermon C, Peto R, Reeves G; Collaborative Group on Epidemiological Studies of Ovarian Cancer. Ovarian cancer and oral contraceptives: collaborative reanalysis of data from 45 epidemiological studies including 23,257 women with ovarian cancer and 87,303 controls. Lancet. 2008;371:303-314. [DOI] [PubMed] [Google Scholar]

[djae282-B9] 9. Collaborative Group on Epidemiological Studies on Endometrial Cancer. Endometrial cancer and oral contraceptives: an individual participant meta-analysis of 27 276 women with endometrial cancer from 36 epidemiological studies. Lancet Oncol. 2015;16:1061-1070. [DOI] [PubMed] [Google Scholar]

[djae282-B10] 10. Karlsson T, Johansson T, Hoglund J, Ek WE, Johansson A.. Time-dependent effects of oral contraceptive use on breast, ovarian, and endometrial cancers. Cancer Res. 2021;81:1153-1162. [DOI] [PubMed] [Google Scholar]

[djae282-B11] 11. Breast cancer and hormonal contraceptives: further results. Collaborative Group on Hormonal Factors in Breast Cancer. Contraception. 1996;54(suppl 3):1S-106S. [DOI] [PubMed] [Google Scholar]

[djae282-B12] 12. Appleby P, Beral V, Berrington de Gonzalez A, et al. ; International Collaboration of Epidemiological Studies of Cervical Cancer. Cervical cancer and hormonal contraceptives: collaborative reanalysis of individual data for 16,573 women with cervical cancer and 35,509 women without cervical cancer from 24 epidemiological studies. Lancet. 2007;370:1609-1621. [DOI] [PubMed] [Google Scholar]

[djae282-B13] 13. Fitzpatrick D, Pirie K, Reeves G, Green J, Beral V.. Combined and progestogen-only hormonal contraceptives and breast cancer risk: a UK nested case-control study and meta-analysis. PLoS Med. 2023;20:e1004188. [DOI] [PMC free article] [PubMed] [Google Scholar]

[djae282-B14] 14. Jareid M, Thalabard JC, Aarflot M, Bøvelstad HM, Lund E, Braaten T.. Levonorgestrel-releasing intrauterine system use is associated with a decreased risk of ovarian and endometrial cancer, without increased risk of breast cancer. Results from the NOWAC Study. Gynecol Oncol. 2018;149:127-132. [DOI] [PubMed] [Google Scholar]

[djae282-B15] 15. Soini T, Hurskainen R, Grénman S, Mäenpää J, Paavonen J, Pukkala E.. Impact of levonorgestrel-releasing intrauterine system use on the cancer risk of the ovary and fallopian tube. Acta Oncol. 2016;55:1281-1284. [DOI] [PubMed] [Google Scholar]

[djae282-B16] 16. Phung MT, Lee AW, Wu AH, et al. ; Ovarian Cancer Association Consortium. Depot-medroxyprogesterone acetate use is associated with decreased risk of ovarian cancer: the mounting evidence of a protective role of progestins. Cancer Epidemiol Biomarkers Prev. 2021;30:927-935. [DOI] [PMC free article] [PubMed] [Google Scholar]

[djae282-B17] 17. Soini T, Hurskainen R, Grénman S, Mäenpää J, Paavonen J, Pukkala E.. Cancer risk in women using the levonorgestrel-releasing intrauterine system in Finland. Obstet Gynecol. 2014;124:292-299. [DOI] [PubMed] [Google Scholar]

[djae282-B18] 18. Iversen L, Fielding S, Lidegaard Ø, Hannaford PC.. Contemporary hormonal contraception and cervical cancer in women of reproductive age. Intl Journal of Cancer. 2021;149:769-777. [DOI] [PubMed] [Google Scholar]

[djae282-B19] 19. Tuesley KM, Webb PM, Protani MM, et al. Nitrogen-based bisphosphonate use and ovarian cancer risk in women aged 50 years and older. J Natl Cancer Inst. 2022;114:878-884. [DOI] [PMC free article] [PubMed] [Google Scholar]

[djae282-B20] 20. Australian Bureau of Statistics. Socio-Economic Indexes for Areas (SEIFA) Australia, 2011. Australian Bureau of Statistics; 2011. Accessed September 19, 2019. http://www.abs.gov.au/websitedbs/censushome.nsf/home/seifa2011?opendocument&navpos=260

[djae282-B21] 21. Australian Bureau of Statistics. Geography Publications. 2017. Accessed September 19, 2019. http://www.abs.gov.au/websitedbs/D3310114.nsf/home/ABS+Geography+Publications

[djae282-B22] 22. The Royal Australian College of General Practitioners (RACGP). Intrauterine devices/systems—resources for general practice. 2024. https://www.racgp.org.au/clinical-resources/clinical-guidelines/key-racgp-guidelines/view-all-racgp-guidelines/iuds-systems-resources-for-general-practice

[djae282-B23] 23. Pearson S, Stewart M, Bateson D. Implanon NXT: expert tips for best-practice insertion and removal. Aust J Gen Pract. 2017;46:104-108. [PubMed] [Google Scholar]

[djae282-B24] 24. Law A, Liao L, Lin J, Yaldo A, Lynen R.. Twelve-month discontinuation rates of levonorgestrel intrauterine system 13.5 mg and subdermal etonogestrel implant in women aged 18–44: A retrospective claims database analysis. Contraception. 2018;98:120-124. [DOI] [PubMed] [Google Scholar]

[djae282-B25] 25. Saloranta TH, Gyllenberg FK, But A, Gissler M, Laine MK, Heikinheimo O.. Free-of-charge long-acting reversible contraception: two-year discontinuation, its risk factors, and reasons. Am J Obstet Gynecol. 2020;223:886.e1-886.e17. [DOI] [PubMed] [Google Scholar]

[djae282-B26] 26. Lu CY, Barratt J, Vitry A, Roughead E.. Charlson and Rx-Risk comorbidity indices were predictive of mortality in the Australian health care setting. J Clin Epidemiol. 2011;64:223-228. [DOI] [PubMed] [Google Scholar]

[djae282-B27] 27. Dhamangaonkar PC, Anuradha K, Saxena A.. Levonorgestrel intrauterine system (Mirena): an emerging tool for conservative treatment of abnormal uterine bleeding. J Midlife Health. 2015;6:26-30. [DOI] [PMC free article] [PubMed] [Google Scholar]

[djae282-B28] 28. Ali S, Na R, Waterhouse M, et al. Predicting obesity and smoking using medication data: a machine-learning approach. Pharmacoepidemiol Drug Saf. 2022;31:91-99. [DOI] [PubMed] [Google Scholar]

[djae282-B29] 29. Lash TL, Fox MP, MacLehose RF, Maldonado G, McCandless LC, Greenland S.. Good practices for quantitative bias analysis. Int J Epidemiol. 2014;43:1969-1985. [DOI] [PubMed] [Google Scholar]

[djae282-B30] 30. Australian Institute of Health and Welfare (AIHW). National Drug Strategy Household Survey 2022–2023. 2024. Accessed October 18, 2024. https://www.aihw.gov.au/reports/illicit-use-of-drugs/national-drug-strategy-household-survey/data

[djae282-B31] 31. Dore M, Filoche S, Danielson K, Henry C.. Efficacy of the LNG-IUS for treatment of endometrial hyperplasia and early stage endometrial cancer: can biomarkers predict response? Gynecol Oncol Rep. 2021;36:100732. [DOI] [PMC free article] [PubMed] [Google Scholar]

[djae282-B32] 32. NPS Medicinewise. Mirena Levonorgestrel consumer medicine information leaflet: MIMS. November 2021. https://www.nps.org.au/medicine-finder/mirena-intrauterine-device

[djae282-B33] 33. Glasziou PP, Jones MA, Pathirana T, Barratt AL, Bell KJ.. Estimating the magnitude of cancer overdiagnosis in Australia. Med J Aust. 2020;212:163-168. [DOI] [PMC free article] [PubMed] [Google Scholar]

[djae282-B34] 34. Kailasam C, Cahill D.. Review of the safety, efficacy and patient acceptability of the levonorgestrel-releasing intrauterine system. Patient Prefer Adherence. 2008;2:293-302. [DOI] [PMC free article] [PubMed] [Google Scholar]

[djae282-B35] 35. Family Planning NSW. Contraception in Australia 2005-2018. Ashfield; 2020. [Google Scholar]

[djae282-B36] 36. Kaunitz AM. Long-acting injectable contraception with depot medroxyprogesterone acetate. Am J Obstet Gynecol. 1994;170:1543-1549. [PubMed] [Google Scholar]

[djae282-B37] 37. Trabert B, Sherman ME, Kannan N, Stanczyk FZ.. Progesterone and breast cancer. Endocr Rev. 2020;41:320-344. [DOI] [PMC free article] [PubMed] [Google Scholar]

[djae282-B38] 38. Conz L, Mota BS, Bahamondes L, et al. Levonorgestrel-releasing intrauterine system and breast cancer risk: a systematic review and meta-analysis. Acta Obstet Gynecol Scand. 2020;99:970-982. [DOI] [PubMed] [Google Scholar]

[djae282-B39] 39. Cervenka I, Mahamat-Saleh Y, Savoye I, et al. Oral contraceptive use and cutaneous melanoma risk: a French prospective cohort study. Int J Cancer. 2018;143:2390-2399. [DOI] [PubMed] [Google Scholar]

PERMALINK

Long-acting, progestin-based contraceptives and risk of breast, gynecological, and other cancers

Karen M Tuesley, PhD

Katrina Spilsbury, PhD

Sallie-Anne Pearson, PhD

Peter Donovan, MSc

Andreas Obermair, MD

Michael D Coory, PhD

Sitwat Ali, MPH

Nirmala Pandeya, PhD

Louise Stewart, PhD

Susan J Jordan, PhD

Penelope M Webb, DPhil

Roles

Abstract

Background

Methods

Results

Conclusion

Introduction

Methods

Study design and participants

Case and control definition

Figure 1.

Concessional subcohort

Exposure variables

Covariates

Statistical analyses

Sensitivity analyses

Results

All cancer

Table 1.

Figure 2.

Breast cancer

Endometrial cancer

Epithelial ovarian cancer

Table 2.

Cervical cancer

Other cancers

Other sensitivity analyses

Discussion

Supplementary Material

Acknowledgments

Contributor Information

Author contributions

Supplementary material

Funding

Conflicts of interest

Data availability

Ethical approval

References

Associated Data

Supplementary Materials

Data Availability Statement

ACTIONS

PERMALINK

RESOURCES

Similar articles

Cited by other articles

Links to NCBI Databases