Relation Between Hysterectomy, Oophorectomy and the Risk of Incident Differentiated Thyroid Cancer

The E3N Cohort

Agathe Guenego; Sylvie Mesrine; Laureen Dartois; Laurence Leenhardt; Françoise Clavel-Chapelon; Marina Kvaskoff; Marie-Christine Boutron-Ruault; Fabrice Bonnet


Clin Endocrinol. 2019;90(2):360-368. 

In This Article


In this large prospective study with more than 400 incident thyroid cancer cases occurring in women aged 40–65 at baseline, we found a positive relationship between the risk of differentiated thyroid cancer and a history of hysterectomy, but not with a history of oophorectomy. Results were similar whatever the age at hysterectomy or time since hysterectomy. We observed also an association between a history of fibroids and the risk of thyroid cancer, which appears to be independent of a history of hysterectomy.

Hysterectomy/Oophorectomy Status and Risk of Thyroid Cancer

A recent meta-analysis reported an association between a history of hysterectomy and the risk of thyroid cancer with a summary relative risk (SRR) of 1.43.[5] However, the results were contrasting as only five out of 24 studies found that hysterectomy was associated with an increased risk of thyroid cancer. Other studies focusing specifically on the link between differentiated thyroid cancer risk and a history of hysterectomy were also conflicting.[7,11,14,15,17,19,20] Methodological flaws may partly explain these discrepancies, due to case-control design and retrospective self-reported surgery history. Besides, studies concluding that there was a positive relationship between a history of hysterectomy and thyroid cancer risk did not perform adjustment on history of benign thyroid disease or family history of thyroid cancer, and thus could not discard the possibility of confounding.[14,15,17,19]

In contrast to previous studies,[14,15,17] we did not observe any change in the association between hysterectomy and the risk of thyroid cancer in relation to the time elapsed since hysterectomy or according to the age at hysterectomy. This finding is an argument against a potential detection bias associated with gynaecological surgery. However, we cannot exclude such a detection bias. It could be speculated that thyroid cancer may be more likely to be found among women with hysterectomy or fibroids, because of medical attention received, as compared to those without the conditions. However, the slightly stronger relation observed between hysterectomy or fibroids and macro-cancers compared with micro-cancers is not in favour of a potential screening bias, which involved the detection of a majority of micro-carcinoma.

Surgically induced menopause, often defined in studies as hysterectomy and/or bilateral oophorectomy has previously been associated with an increased risk of thyroid cancer.[8,16,21] When we considered simultaneously hysterectomy and oophorectomy status, we found that a history of oophorectomy was not related to the risk of thyroid cancer and did not substantially affect the association between hysterectomy and thyroid cancer. This is consistent with the WHI cohort for which Kabat et al[7] did not find any association between thyroid cancer risk and bilateral oophorectomy. Some authors found either a higher,[22] a lower[15,17] or a similar[11,14,16] risk of thyroid cancer in women with hysterectomy and bilateral oophorectomy compared to women with hysterectomy alone or hysterectomy with partial oophorectomy.

Benign Gynaecological Diseases History and Risk of Thyroid Cancer

Our study revealed a positive association between a history of fibroids and the risk of differentiated thyroid cancer, which is consistent with the relationship between hysterectomy and thyroid cancer as uterine fibroids are the most common cause of hysterectomy.[12] The twofold increased risk of thyroid cancer linked to surgically treated fibroids was similar to results reported in two American studies.[11,15]

Studies on the relationship between endometriosis history and thyroid cancer risk are more conflicting, with HR/SIR varying between 0.85 and 3.09. Only two studies found an increased risk of thyroid cancer in women with an endometriosis evolving from more than 5 years[23] and in parous women.[24] It may reflect discrepancies in the definition of endometriosis cases, differences in the populations and the design of the studies: some used retrospective cohort of infertile women,[25] other used data extracted from inpatient and cancer registers[23,24,26] or self-reported endometriosis history.[12]

As previous studies,[11] we found no relationship between an ovarian cyst history and thyroid cancer risk, which is in line with the lack of link between oophorectomy and thyroid cancer in our study.

Our results seem not to be related to detection bias. Uterine fibroids are the most common pelvic tumours, occurring in nearly 70% of all reproductive-aged women.[12] Hyper- or hypothyroidism both enhance the likelihood of their diagnosis or treatment[27] by causing abnormal uterine bleeding and have been linked with an increased risk of differentiated thyroid cancer.[28,29] However, the fact that we found a similar thyroid cancer risk whatever the time elapsed since surgery argues against this. Moreover, the slightly stronger relation observed between hysterectomy or fibroids and macro-cancers compared with micro-cancers is not in favour of a potential screening bias.

Potential Mechanisms Linking Uterine Fibroids and Differentiated Cancer Thyroid

Uterine fibroids have been associated with thyroid nodules [30,31] and hypothyroidism [32] in transversal studies, suggesting shared mechanisms between fibroids and thyroid diseases. Both fibroids and thyroid cancers are thought to be sex steroid dependent.

If the link between hormonal factors and the risk of fibroids is well established,[11] the relationship between sex hormones and thyroid cancer is much less characterized. Both oestrogen receptors α (ER α) with a proliferative and anti-apoptotic function, and oestrogen receptors β (ERβ) with a pro-apoptotic function, are expressed in normal and thyroid tumour cells. Thyroid cell proliferation and neoplastic development might depend on the imbalance between ER α and ERβ.[6,33] Moreover, progesterone receptors have already been described in thyroid follicular cells, and progesterone has been shown to upregulate genes involved in thyroid function and growth on normal human thyroid cells in vitro.[34] Besides, thyroid cancer cells and fibroids are both able to biosynthesize estradiol in situ through the action of aromatase.[35,36]

However, the link between hormonal factors and thyroid cancer remain inconclusive in several studies[5,7–9] suggesting that not only direct oestrogen action but also other pathways may be shared between fibroids and thyroid cancer, such as growth factors pathways and TSH/thyroid hormones induced pathways. Growth factors and non-genomic estradiol pathways converge towards the aberrant activation of Ras/Raf/MEK/ERK and PI3K/Akt/mTOR signalling in both thyroid and fibroids cells. In addition, due to common nucleotide sequence in ER and thyroid hormone receptor, these receptors can interact and regulate several transcriptional responses to environmental stimuli.[37] Moreover, thyrotropin-releasing hormone (TRH), thyroid-stimulating hormone (TSH) and thyroid hormone receptors were shown to be present in monkey uterus.[38] Another study described a smooth muscle cells proliferation after stimulation by TSH.[39]

Somatic genetic mutations play a part in the molecular pathogenesis of both fibroids (MED12 mutations)[12] and differentiated thyroid cancer (eg, BRAF mutations).[1] Thus, a chronic reduction in DNA repair capacity might explain partly a common susceptibility to differentiated thyroid cancer and fibroids. The increased risk of both thyroid cancer and fibroids in atomic bomb survivors illustrates this hypothesis.[40] Increased insulin resistance or related hyperinsulinemia might be a common underlying factor shared by women with uterine fibroids which may predispose to the development of thyroid cancer as recently suggested.[41]

Strengths and Limits

Strengths of our study include its prospective design, large sample size, histological confirmation of all thyroid cancer cases and availability of data for most differentiated risk factors including a history of benign thyroid conditions. Exposure data were collected before diagnosis of thyroid cancer, avoiding potential recall biases. To our knowledge, it is the first study that investigated an association between hysterectomy, uterine fibroids and micro- and macro-thyroid cancer risk.

This cohort is not representative of the French population and our results cannot be extrapolated to the entire French population. As in all observational studies, residual confounding may subsist. Misclassification of exposure status is a potential limitation because of the self-reported assessment of surgery, benign gynaecological and thyroid diseases. Moreover, when we restricted the analysis to the treated cases, the associations remain consistent. Women included in our study, although younger than in the WHI and PLCO studies, were mostly postmenopausal, and our results may not be generalizable to young premenopausal women.