Non-Adherence Increases Risk of Preterm Birth in Women With SLE

“Pregnancies in women with SLE are often complicated,” investigators stated. “Accordingly, it is crucial to optimize drug therapy to promote disease control and improve outcomes.”

While hydroxychloroquine (HCQ) pharmacokinetics (PK) shortened the drug’s half-life by 10 days during pregnancy, women with systemic lupus erythematosus (SLE) who did not adhere to their medication regimen were more likely to experience preterm birth, according to a study published in Lupus Science & Medicine.1

Investigators urge pregnant patients to optimize adherence rather than focusing on adjusting HCQ dosage to mitigate physiological changes, such as weight gain. HCQ concentrations of ≤100 ng/mL were indicative of non-adherence, regardless of trimester.

“Pregnancies in women with SLE are often complicated by fetal loss, preterm birth and in severe cases, maternal death. Additionally, many women with SLE who are undertreated experience flares of their disease during pregnancy, which is independently associated with worse pregnancy outcomes,” investigators stated. “Accordingly, it is crucial to optimize drug therapy in lupus pregnancies to promote disease control and improve outcomes.”

PK analysis was performed using 2 observational pregnancy registries at Duke University from November 2013 to January 2020. Eligible patients were pregnant women with SLE who were receiving HCQ at least 3 months prior to and throughout their pregnancy. Serum, concomitant medication, neonatal outcomes, and disease activity data were collected at each clinical visit. Dosing simulations were conducted, using the PK model, and 0%/20%/40%/60% non-adherence was calculated to analyze the impact of bodily changes compared with adherence on concentrations of HCQ. Medication adherence was defined as a Medication Adherence Self-Reported Inventory (MASRI) score of ≥80 on the visual analogue scale. Additionally, investigators created a virtual population of 1000 pregnant women to represent real-world pregnant patients with SLE. Adjusted logistic regression was used to understand the effect of non-adherence and preterm birth.

A total of 56 women, with a combined total of 61 pregnancies and 202 clinical visits, were enrolled in the study. MARSI scores were available for 18 women with 18 pregnancies (37/202 visits; 18.3%). HCQ levels were highest for those deemed adherent when compared with non-adherence (266.6 vs 93.7, respectively). With the serum cut-off of ≤100 ng/mL in mind, 25.2% (n = 51) patients were non-adherent during the PK model development. Of these, 19.7% (n = 12) were chronically non-adherent throughout pregnancy and postpartum.

For women who were fully adherent to their treatment, concentrations of ≤100 ng/mL should only appear about 0.3% of the time. This increases to 24.2% for those who miss 60% of doses.

HCQ clearance (CL), at a dosage of 400 mg/day, steadily increased throughout pregnancy, culminating in a 59.6% increase by the third trimester. The volume of distribution also increased by 31.2% due to weight gain, resulting in the half-life decreasing from 32.4 days postpartum compared with 22.1 days in the third trimester. HCQ concentrations continued to decrease throughout pregnancy when compared with postpartum concentrations. A significant decrease occurred in patients with medication non-adherence.

Neonatal data was available for 56 pregnancies (54 live births). Median gestational age at birth was 37 weeks, with 31.5% (n = 17) patients delivering preterm. Of those with HCQ concentrations of ≤100 ng/mL, 66.7% delivered preterm and 55.6% with concentrations ≥500 ng/mL delivered preterm, compared with only 6.7% who had concentrations between 101-500ng/mL.

Patients with low HCQ concentrations also experienced greater maternal disease activity, as calculated by the Physician Global Assessment (PGA) when compared with the >100ng/mL cohort (0.75 vs 0.24, respectively).

The study was strengthened by the population PK modeling, which differentiated between physical changes and non-adherence to medication. However, data quality issues, such as disease misclassification and missing data, may have occurred as data was collected via a real-world registry. Due to the small sample size, investigators were more flexible with the cut-off value for covariate selection (p=0.05), which may have led to a false positive between trimester and clearance. This is not very likely, as HCQ increases the glomerular filtration rate by up to 40% during pregnancy. Finally, renal clearance may have been underestimated by excluding visits with very low HCQ concentrations during the development of the PK model. However, the PK model was more accurate when omitting these concentrations.

“Despite the importance of maintaining therapeutic drug exposure during pregnancy, few studies have evaluated the PK of HCQ in pregnancy… As a result of these PK changes, women with SLE may be at risk for subtherapeutic HCQ exposure as pregnancy progresses. In addition to changes in PK, subtherapeutic HCQ exposure can also occur in the setting of drug non-adherence,” investigators concluded. “Accordingly, drug exposure in pregnancy must be interpreted in the context of both physiological changes, disease effects and adherence.”


Balevic SJ, Weiner D, Clowse MEB, et al. Hydroxychloroquine PK and exposure-response in pregnancies with lupus: the importance of adherence for neonatal outcomes. Lupus Sci Med. 2022;9(1):e000602. doi:10.1136/lupus-2021-000602