Abstract
Thyroid cancer is one of the most common cancers diagnosed in young women (1). On top of the well-known adverse consequences of the disease and its treatment, thyroid cancer diagnosis commonly complicates pregnancy planning. This could partly explain why specifically young female thyroid cancer survivors have a lower quality of life (2). Among young women diagnosed with thyroid cancer, pregnancy planning is affected not only by medical aspects but also by increased psychosocial distress related to the diagnosis (3,4). The latter is further amplified by uncertainties relating to pregnancy-specific disease relapse, thyroid hormone therapy, and a risk for adverse pregnancy or child outcomes. The treating physician has an important role in these situations by providing information and support.
Clinical data have refuted the theoretical concept that various pregnancy-specific physiological changes could promote thyroid cancer (remnant) growth (e.g., increase in estrogen, placental growth hormone, and human chorionic gonadotropin). For example, it has been shown that pregnancy is not associated with thyroid cancer recurrence and that any progression of residual disease only rarely requires additional treatment (5,6). As a consequence of practically all thyroid cancer treatment modalities, the majority of women will require levothyroxine treatment, and all young women will require preconception counselling and gestational monitoring (7). It is reassuring that women with pre-existing hypothyroidism that is well-controlled on levothyroxine have a similar risk of adverse pregnancy outcomes as euthyroid women (8 –10). Nonetheless, pregnant women with a history of thyroid cancer can be difficult to control biochemically using levothyroxine therapy aiming for example for thyrotropin suppression or due to a history of hemithyroidectomy. Furthermore, out of all thyroid cancer patients, young women in particular exhibit more distress and anxiety related to the diagnosis (3,4). Both non-optimal biochemical control and increased stress and anxiety are risk factors for adverse pregnancy outcomes, which could potentially mediate a higher risk in this specific subgroup (10 –12).
In the current issue of Thyroid, Cho et al. studied the association of a history of thyroid cancer with obstetric outcomes by leveraging the data of close to 2.3 million Koreans stored in the Korea National Health Insurance claims database (13). Women who gave birth between 2007 and 2015 were included, and ICD-10 codes were used to identify those with a history of thyroid cancer and/or adverse obstetric outcomes. Crude analyses showed that women with a history of thyroid cancer had a slightly higher risk of preterm birth (absolute risk difference +0.5%), high birth weight (absolute risk difference +0.5%), preeclampsia (absolute risk difference +0.6%), postpartum hemorrhage (absolute risk difference +2.1%), and placenta previa (absolute risk difference +0.5%). From a clinical point of view, these absolute risks seem negligible and should be communicated to the patient as such.
From a research point of view, pregnant women with a history of thyroid cancer were older (mean age ± SD: 30.8 ± 3.9 vs. 32.5 ± 3.5 years, proportion >35 years: 15.6% vs. 26.5%), more likely to be diagnosed with pre-pregnancy hypertension (0.5% vs. 1.3%) or pre-pregnancy diabetes mellitus (0.6% vs. 2.1%), and more likely to have a multiple gestation pregnancy (absolute risk difference +0.6%) than those without a history of thyroid cancer. After additional adjustment for maternal age and pre-pregnancy hypertension or diabetes mellitus among other confounders, only the association for postpartum hemorrhage persisted. This seems to indicate that it is not the history of thyroid cancer per se that is associated with a slightly higher risk of adverse pregnancy outcomes. However, it could be possible that this slightly higher risk of adverse pregnancy outcomes is caused by: (i) older age at conception due to a delay in pregnancy planning due to the diagnosis or increased stress and anxiety; (ii) older age before planning pregnancy, increasing the risk of infertility and the need for assisted reproduction techniques, which is a risk factor for adverse pregnancy outcomes (14); or (iii) a higher risk of hypertension or diabetes mellitus after thyroid cancer treatment, which has been reported before and are both a risk factor for adverse pregnancy outcomes (15). Alternatively, these data could reflect surveillance bias by which women with a history of thyroid cancer have a higher probability of having the study outcome detected because of medical workups related to the disease. Translating these concepts back to a clinical point of view, if anything, it seems that an effort to decrease unnecessary delay in pregnancy by reassuring the patient that there is no medical indication to postpone pregnancy after thyroid cancer treatment (excluding radioactive iodine) (7), as well as early identification of infertility or symptoms related to hypertension or diabetes mellitus, could be of added value in young female thyroid cancer patients.
The strengths of this study include the large sample size and the use of a population-based cohort. A large data set such as this is necessary to perform a reliable study, given the relatively low event rates of certain pregnancy complications, such as placental abruption and placenta previa, and the fact that the number of women in which pregnancy is complicated by a history of thyroid cancer is small (in the current study 7734/2,888,745 [0.26%]). It would be difficult to complete a study such as this one using single institution studies or even pooled data from several institutions, as sample sizes would be small and selection bias would occur. However, the necessity of using claims data does impact the depth of information available. There were no data available on the type of thyroid cancer treatment, infertility treatment, thyroid function during pregnancy, or maternal smoking. Whether certain subgroups of women with a history of thyroid cancer, for example those who underwent high-dose radioactive iodine or those that are poorly controlled biochemically, are at increased risk of adverse pregnancy outcomes remains to be elucidated. Furthermore, replication of these data is needed from other countries because South Korea has seen a surge in thyroid cancer diagnosis due to over-detection by the widespread use of sensitive imaging tools during the study period (16). The over-detection of low-risk thyroid cancers and the risk for errors secondary to use of claims data could cause a decrease in the detected effect estimates.
Despite limitations, these valuable data should reassure both patients and physicians that a history of thyroid cancer does not meaningfully impact the risk of adverse pregnancy outcomes. We look forward to future studies that can replicate these results in other populations and ascertain data on different thyroid cancer treatment modalities and the effects of thyroid hormone therapy on adverse pregnancy outcomes.