Abstract
Background and purpose
Data on pregnancy outcomes following fetal exposure to disease‐modifying drugs (DMDs) in women with multiple sclerosis (MS) are sparse although growing.
Methods
Data from the Danish Multiple Sclerosis Registry were linked with nationwide registries enabling an investigation of adverse pregnancy outcomes in newborns of women with MS following fetal exposure to injectable first‐line treatments, dimethyl fumarate, glatiramer acetate, or natalizumab. Logistic regression models accounting for clustered data were used to estimate odds ratios (ORs) with 95% confidence intervals (CIs) for individual and composite adverse outcomes after adjusting for relevant covariates.
Results
A total of 1009 DMD‐exposed pregnancies were compared with 1073 DMD‐unexposed pregnancies as well as 91,112 pregnancies from the general population. No association of an increased risk of any perinatal outcome was found when comparing newborns with fetal exposure with the general population, including preterm birth (OR = 1.19, 95% CI = 0.86–1.64), small for gestational age (OR = 1.38, 95% CI = 0.92–2.07), spontaneous abortion (OR = 1.04, 95% CI = 0.84–1.27), congenital malformation (OR = 0.99, 95% CI = 0.68–1.45), low Apgar score (OR = 0.62, 95% CI = 0.23–1.65), stillbirth (OR = 1.05, 95% CI = 0.33–3.31), placenta complication (OR = 0.53, 95% CI = 0.22–1.27), and any adverse event (OR = 1.10, 95% CI = 0.93–1.30). Similar results were found when comparing DMD‐exposed pregnancies with DMD‐unexposed pregnancies.
Conclusions
We found no increased association of adverse pregnancy outcomes in newborns with fetal exposure to DMDs when compared with either DMD‐unexposed pregnancies or the general population.
Keywords: adverse pregnancy outcomes, fetal exposure, multiple sclerosis
INTRODUCTION
Teratogenicity is a genuine concern if one becomes pregnant while treated with disease‐modifying drugs (DMDs) for multiple sclerosis (MS). Drugs with a low molecular weight such as teriflunomide and cladribine are able to cross the placental barrier and have demonstrated teratogenicity in animal studies; hence, they are contraindicated in women planning pregnancy [1, 2]. Natalizumab is a recombinant humanized monoclonal antibody very similar to immunoglobulin G and also able to cross the placental barrier. The placental transfer of monoclonal antibodies depends on the gestational age and happens in a linear fashion, starting around Week 13 and increasing hereafter [3, 4]. Natalizumab was approved in Europe in 2006, and growing data from the natalizumab pregnancy registry, case reports, and observational studies indicate that natalizumab is not fetotoxic. However, hematological abnormalities have been reported in several cases, with increasing association the later in the gestational period the exposure occurred [5, 6, 7].
On the contrary, drugs with a high molecular weight such as interferon‐β (including pegylated forms; interferons) or glatiramer acetate are not considered to be able to permeate the placental barrier. Dimethyl fumarate has not demonstrated teratogenicity in rat studies, but data on outcomes in pregnancies exposed to the drug are very sparse; therefore, cessation is advised if planning pregnancy.
This study aimed to investigate individual and composite adverse perinatal outcomes in pregnancies exposed to injectable first‐line treatments (interferons or glatiramer acetate), dimethyl fumarate, or natalizumab compared with DMD‐unexposed pregnancies in women with MS and pregnancies in the general population.
METHODS
We conducted a cross‐sectional study using population‐based data on all pregnancies resulting in either abortion, live birth, or stillbirth from 1 January 1997 to 31 December 2018. The nationwide Danish Multiple Sclerosis Registry was used to identify pregnancies in women with MS [8]. Data were cross‐linked with the Danish Medical Birth Register (data on perinatal outcomes, delivery mode, smoking during pregnancy) [9], the National Patient Registry (public hospital registrations related to abortions) [10], and the Register for Legally Induced Abortions (private practitioner registrations related to abortions) [11] to obtain and ensure complete information. Only pregnancies occurring after the MS diagnosis were included to ensure information about exposure to DMDs.
Injectable first‐line treatments, dimethyl fumarate, or natalizumab were included as possible exposures. According to clinical practice in the 13 Danish MS clinics, injectables have been used up until confirmed pregnancy for several years. Cessation of dimethyl fumarate is recommended if becoming pregnant, although teratogenicity has not been demonstrated. The clinical practice of stopping dimethyl fumarate before conception varies; therefore, the drug was included as a possible exposure. Natalizumab is often continued until conception in women who wish to becoming pregnant, but only continued during the first and second trimester of pregnancy in case of highly active disease. Treatment adherence of at least 30 consecutive days prior to the last menstrual period was required for all exposures. Eligible exposed pregnancies were pooled into one combined group to increase statistical power.
Study populations
Exposed group
The first day of the last menstrual period (LMP) was used to define exposure. LMP for births was calculated by subtracting the gestational age in days from the birthdate. For abortions LMP was calculated by subtracting the gestational week (GW)*7 from the date of the procedure. Data on gestational age in days for stillbirths and live births were extracted from the Danish Medical Birth Register and for abortions from the National Patient Registry and Registry for Legally Induced Abortions.
Exposure to injectable first‐line treatments and dimethyl fumarate was defined as either injection or administering treatment at the day of or after LMP. For the high‐efficacy treatment (natalizumab), exposure was defined as infusion within 8 weeks prior to the LMP or onward.
Control groups
Two control groups were used for comparison: DMD‐unexposed pregnancies in women with MS (unexposed group) and pregnancies in the general population (control group).
DMD‐unexposed pregnancies
A pregnancy was considered unexposed to DMD if (i) the last injection/intake of glatiramer acetate, interferons, or dimethyl fumarate was administered any time before LMP; (ii) the last infusion of natalizumab was given >3 months prior to LMP; (iii) the last intake of fingolimod happened >2 months prior to LMP; (iv) the last infusion of rituximab or alemtuzumab was administered >4 months prior to LMP; (v) the last infusion of mitoxantrone was given >6 months prior to LMP; (vi) the last intake of teriflunomide was at least 2 years prior to last LMP; or (vii) the subject was never treated. All definitions of exposure were based on half‐life times of the drug in question.
General population
A random sample of 5% of the general population was established through the Danish Civil Registration [12] and comprised the group from which pregnancies in the control group were identified.
Outcome definitions
Perinatal outcomes
Preterm birth was defined as delivery before GW 37 and stillbirth as birth of a dead fetus after completion of GW 22, in accordance with the literature [13].
Small for gestational age (SGA) [14] was calculated for each individual gestational week separated by biological sex and defined as a birth weight below the mean two times minus the SD as described by Sankilampi et al. [14]. Additionally, the calculations were performed separately for singleton and twin births.
The Apgar score is used as a quick assessment of the postnatal condition of the newborn and is defined as “reassuring of normality” if the score is >6/10 points based on muscular tonus, vitals on heart and lung, and coloration of the newborn [15].
Congenital malformations (CMs) are generally recorded at birth, but as some anomalies are not detected at birth, delayed identification up to 1 year after birth is possible [9]. Only major malformations based on the EUROCAT definition were included [16], and in case of more than one registered malformation, these attributed to one event.
Any adverse event (AE) consisted of at least one of the following: congenital malformation, spontaneous abortion, preterm birth, stillbirth, SGA, or low Apgar score. Any AE was recorded as a binary (yes/no) regardless of the number of individual AEs.
Classifications
Delivery mode was classified into spontaneous delivery, emergency cesarean section (C‐section), elective C‐section, induced delivery, or instrumental delivery.
Abortions were classified as spontaneous or elective. Additionally, ectopic pregnancies were identified and included descriptively. Abortions or ectopic pregnancies closer than 8 weeks apart only counted as one [17].
Statistical analyses
Variables were summarized using either mean and SD or median and interquartile range (Q1–Q3) depending on the distributional shape. Logistic regression models using generalized estimating equations to account for the clustered nature of data were used to estimate odds ratios (ORs) with 95% confidence intervals (CIs). All analyses were performed crude and subsequently adjusted for clinically relevant covariates related to the outcome. SAS Enterprise Guide version 7.15 was used for all statistical analyses, and statistical significance was defined as p < 0.05.
Sensitivity analyses
A less conservative approach for exposed pregnancies and a more conservative approach for unexposed pregnancies were applied to perform sensitivity analyses. The definition window to be included in the exposed group was extended to treatment 3 months prior to LMP for injectable first‐line treatments and dimethyl fumarate and 4 months prior to LMP for natalizumab.
Unexposed pregnancies were defined as (i) treatment discontinuation of interferons/glatiramer acetate, natalizumab, rituximab, alemtuzumab, dimethyl fumarate, or fingolimod at least 6 months prior to LMP; (ii) last infusion of mitoxantrone 12 months prior to LMP; (iii) last intake of teriflunomide at least 3 years prior to last LMP; or (iv) subject never treated.
Furthermore, complete‐case analyses with smoking included as a binary variable were performed, as smoking is associated with increased risk of preterm birth, CM, and SGA in newborns [18]. Information on smoking habits was only available in the Danish Medical Birth Register; hence, only pregnancies resulting in birth were eligible.
Study approval
This study was approved by the Danish Data Protection Agency (j.nr.: 2012‐58‐0004). Noninterventional register‐based studies do not require ethical approval in Denmark.
RESULTS
A total of 1009 pregnancies to 704 individual women were included in the exposed group (Figure 1). The women were on average 26.8 years old at diagnosis and 31.6 years old at LMP, with a disease duration of 6.7 years at LMP (Table 1). The unexposed group consisted of 1073 pregnancies in 709 individuals with almost identical age at diagnosis, age at LMP, and disease duration at LMP as the exposed group (Table 1). The control group comprised 91,112 pregnancies to 40,845 women who were slightly younger at LMP (29.7 years; Table 1).
FIGURE 1.
Inclusion flowchart of exposed pregnancies
TABLE 1.
Maternal characteristics in the exposed MS group of pregnancies (n = 1009 pregnancies, 704 individuals), unexposed MS pregnancies (n = 1073 pregnancies, 709 individuals), and the control group (n = 91,112 pregnancies, 40,845 individuals)
| Characteristic | Exposed | Unexposed | Controls |
|---|---|---|---|
| Age at diagnosis, years, mean (SD) | 26.8 (5.1) | 27 (4.9) | n/a |
| Age at onset, years, mean (SD) | 24.8 (5.4) | 24.9 (5.1) | n/a |
| Disease duration at diagnosis, years, mean (SD) | 1.9 (2.7) | 2.1 (2.8) | n/a |
| Disease duration at LMP, years, mean (SD) a | 6.7 (4.1) | 6.5 (4.2) | n/a |
| Maternal age at LMP, years, mean (SD) | 31.6 (4.8) | 31.4 (4.9) | 29.7 (5.6) |
| Educational level, n (%) b | |||
| Primary school | 134 (13.3) | 136 (12.7) | 14,612 (16) |
| High school or equivalent | 327 (32.4) | 405 (37.7) | 34,510 (37.9) |
| Bachelor level or equivalent | 343 (34) | 358 (33.4) | 29,331 (32.2) |
| Above bachelor level | 205 (20.3) | 174 (16.2) | 12,659 (13.9) |
| Parity, n (%) | |||
| 1 | 382 (54) | 399 (48.2) | 29,362 (45.6) |
| 2 | 244 (34.5) | 341 (41.2) | 23,584 (36.6) |
| >2 | 82 (11.5) | 88 (10.6) | 11,512 (17.9) |
| Exposure, n, (%) | |||
| Injectable first‐line treatments | 774 (76.7) | ||
| Interferon‐β | 633 (62.7) | n/a | n/a |
| Glatiramer acetate | 141 (13.9) | n/a | n/a |
| Dimethyl fumarate | 58 (5.7) | n/a | n/a |
| Natalizumab | 177 (17.5) | n/a | n/a |
Abbreviations: LMP, last menstrual period; MS, multiple sclerosis; n/a, not applicable.
First day of the LMP.
Highest attained educational level.
Live births represented 70.5%, 77.2%, and 70.2% of the total number of pregnancies in the exposed, unexposed, and control groups, respectively, with very similar rates of abortions, stillbirths, and ectopic pregnancies among the three groups (Figure 2; Table 2). On the other hand, elective C‐section was the delivery mode with the greatest contrast, namely 17.2%, 16.6%, and 8.6% (Table 2).
FIGURE 2.
Pregnancy outcomes. Any AE, any adverse event, placenta; CM, congenital malformation; CS, cesarean section; Ectopic; ectopic pregnancy; El. abortion, elective abortion; Preterm, preterm birth; SGA, small for gestational age; Sp. abortion, spontaneous abortion
TABLE 2.
Delivery mode and characteristics for the included pregnancies
| Mode/characteristic | Exposed, n = 1009 | Unexposed, n = 1073 | Control group, n = 91,112 |
|---|---|---|---|
| Live births, n (%) | 708 (70.2) | 828 (77.2) | 64,458 (70.2) |
| Stillbirths, n (%) | 3 (0.3) | 3 (0.3) | 229 (0.2) |
| Ectopic pregnancies, n (%) | 6 (0.6) | 14 (1.3) | 1044 (1.1) |
| Abortions, n (%) | |||
| Spontaneous | 109 (10.8) | 113 (10.5) | 9259 (10) |
| Elective | 183 (18.1) | 115 (10.7) | 16,122 (18.2) |
| Delivery mode, n (%) | |||
| Spontaneous delivery | 338 (47.5) | 405 (48.7) | 39,119 (60.5) |
| Acute cesarean section | 94 (13.2) | 93 (11.2) | 7559 (11.7) |
| Elective cesarean section | 122 (17.2) | 138 (16.6) | 5539 (8.6) |
| Induced delivery | 103 (14.5) | 123 (14.8) | 7908 (12.2) |
| Instrumental delivery | 54 (7.6) | 72 (8.7) | 4562 (7.1) |
| Live birth outcomes, n (%) | |||
| Placenta complication a | 5 (0.7) | 13 (1.6) | 767 (1.2) |
| Apgar score < 7 | 4 (0.6) | 2 (0.2) | 502 (0.8) |
| Small for gestational age | 31 (4.4) | 28 (3.4) | 1965 (3) |
| Preterm birth | 57 (8) | 60 (7.2) | 4205 (6.5) |
| Congenital malformations | 30 (2.9) | 23 (2.2) | 2450 (2.7) |
| Any adverse event b | 223 (22.1) | 229 (21.3) | 17,842 (19.6) |
Placenta accreta or abruptio.
Any of the following: spontaneous abortion, low Apgar score, congenital malformation, small for gestational age, stillbirth, or preterm birth.
Among the exposed pregnancies resulting in live birth, the majority (64.7%) were exposed to interferons, with a treatment duration of 1.6 years (0.9–3.2) prior to LMP and a fetal exposure of 6.8 weeks (4.4–12; Table 3). Glatiramer acetate‐exposed pregnancies comprised 12.8%, with a treatment duration of 1.2 years (0.4–2.5) prior to LMP and a median fetal exposure of 7.7 weeks (4.5–12). A minority were exposed to dimethyl fumarate (34/711 pregnancies), with a treatment duration of 1.2 years (0.9–1.8) prior to LMP and fetal exposure of 5 weeks (1.9–6.3). Natalizumab exposure comprised 17%, with a median fetal exposure of 1.9 weeks (0.6–4.4) and a treatment duration of 1.6 years (1.1–2.5) prior to LMP (Table 3).
TABLE 3.
In utero exposure and treatment duration prior to the first day of the last menstrual period in pregnancies resulting in live birth from the exposed group (n = 708) a
| n (%) | Live births, n = 708 | |
|---|---|---|
| Fetal exposure in utero, weeks, median (Q1–Q3) | ||
| Moderate‐efficacy treatment | ||
| Interferon‐β | 460 (64.7) | 6.9 (4.4–12) |
| Dimethyl fumarate | 34 (4.8) | 5 (1.9–6.3) |
| Glatiramer acetate | 91 (12.8) | 7.7 (4.5–12) |
| High‐efficacy treatment | ||
| Natalizumab | 123 (17.3) | 1.9 (0.6–4.4) |
| Treatment duration prior to last menstrual period, years, median (Q1–Q3) | ||
| Moderate‐efficacy treatment | ||
| Interferon‐β | 460 (64.7) | 1.6 (0.9–3.2) |
| Dimethyl fumarate | 34 (4.8) | 1.2 (0.9–1.8) |
| Glatiramer acetate | 91 (12.8) | 1.2 (0.4–2.5) |
| High‐efficacy treatment | ||
| Natalizumab | 123 (17.3) | 1.6 (1.1–2.5) |
Data on stillbirths (n = 3) on the individual level were excluded, as small cell information is prohibited to present according to Danish law on data protection.
No association of an increased risk of any perinatal outcomes was found when comparing exposed newborns with controls, including preterm birth (OR = 1.19, 95% CI = 0.86–1.64), SGA (OR = 1.38 95% CI = 0.92–2.07), spontaneous abortion (OR = 1.04, 95% CI = 0.84–1.27), CM (OR = 0.99, 95% CI = 0.68–1.45), low Apgar score (OR = 0.62, 95% CI = 0.23–1.65), stillbirth (OR = 1.05, 95% CI = 0.33–3.31), placenta complication (OR = 0.53, 95% CI = 0.22–1.27), and any AE (OR = 1.10, 95% CI = 0.93–1.30; Table 4).
TABLE 4.
Association of delivery mode and perinatal outcomes among exposed pregnancies (n = 1009) compared with unexposed pregnancies (n = 1073) and the control group (n = 91,112)
| Unexposed | Control group | |||||
|---|---|---|---|---|---|---|
| Unadjusted, OR (95% CI) | Adjusted, OR (95% CI) a | p | Unadjusted, OR (95% CI) | Adjusted, OR (95% CI) a | p | |
| Delivery mode | ||||||
| Spontaneous delivery | Reference | Reference | Reference | Reference | ||
| Acute cesarean section | 1.22 (0.86–1.74) | 1.13 (0.79–1.63) | 0.50 | 1.45 (1.12–1.87) | 1.18 (0.91–1.52) | 0.22 |
| Elective cesarean section | 1.06 (0.78–1.44) | 1.15 (1.81–1.63) | 0.43 | 2.55 (2.02–3.23) | 2.30 (1.82–2.93) | <0.0001 |
| Induced delivery | 0.99 (0.74–1.34) | 0.92 (0.68–1.26) | 0.61 | 1.50 (1.18–1.90) | 1.17 (0.91–1.49) | 0.20 |
| Instrumental delivery | 0.90 (0.60–1.36) | 0.90 (0.59–1.37) | 0.62 | 1.37 (1.0–1.88) | 1.15 (0.83–1.58) | 0.41 |
| Perinatal outcomes | ||||||
| Apgar score < 7 | N/A | N/A | N/A | 0.72 (0.27–1.94) | 0.62 (0.23–1.65) | 0.33 |
| Any adverse event b | 1.05 (0.84–1.30) | 1.01 (0.81–1.27) | 0.92 | 1.16 (0.99–1.37) | 1.11 (0.95–1.30) | 0.20 |
| Stillbirth | N/A | N/A | N/A | 1.19 (0.38–3.73) | 1.05 (0.33–3.31) | 0.93 |
| Congenital malformation | 1.55 (0.88–2.71) | 1.59 (0.88–2.86) | 0.12 | 1.12 (0.77–1.63) | 0.99 (0.68–1.45) | 0.96 |
| Small for gestational age | 1.31 (0.75–2.28) | 1.32 (0.75–2.33) | 0.34 | 1.45 (0.97–2.18) | 1.38 (0.92–2.07) | 0.12 |
| Preterm birth | 1.12 (0.73–1.73) | 1.10 (0.72–1.72) | 0.64 | 1.25 (0.91–1.72) | 1.19 (0.86–1.64) | 0.30 |
| Spontaneous abortion | 0.97 (0.74–1.26) | 0.99 (0.76–1.31) | 0.98 | 1.07 (0.87–1.31) | 1.05 (0.86–1.29) | 0.63 |
| Placenta complication c | 0.45 (0.15–1.29) | 0.43 (0.15–1.22) | 0.15 | 0.59 (0.25–1.43) | 0.53 (0.22–1.27) | 0.15 |
Abbreviations: CI; confidence interval; N/A; not available; OR; odds ratio.
Adjusted for prior abortion, maternal age at last menstrual period, educational level, prior cesarean section, and calendar year of birth.
Including one of the following: spontaneous abortion, congenital malformation, small for gestational age, preterm birth, stillbirth, or low Apgar score.
Placenta accreta or abruptio.
Similar results were found when comparing exposed with unexposed pregnancies (Table 4). Data are not shown for low Apgar score and stillbirth, as too few events occurred to enable statistical analyses.
Descriptive, crude, and adjusted analysis for exposures separately are displayed in Tables S1.
Sensitivity analyses
The exposed sensitivity group was defined by broader criteria related to the timing of exposure and included 1172 pregnancies, whereas a more rigorous approach with firmer restrictions on treatment discontinuation related to LMP resulted in 833 pregnancies in the unexposed sensitivity group (Table S7).
The unexposed sensitivity group was first compared with the exposed group (n = 1009) and thereafter with the exposed sensitivity group (n = 1172; Tables S8 and S9). Likewise, the control group was compared with all groups. No association of increased risk of any perinatal outcomes were found in either comparison (Tables S8 and S9).
Lastly, pregnancies with data on smoking habits for the complete‐case analysis were identified. The exposed group consisted of 698 births with registration on smoking habits related to the index pregnancy, at which 88.8% were nonsmokers. The corresponding numbers were 814 pregnancies with 87.6% nonsmokers and 59,441 pregnancies with 84.7% nonsmokers among the unexposed and control groups, respectively. Neither the exposed versus unexposed nor versus control group showed any association of increased risk in any perinatal outcomes when adding smoking habits to the analyses (Table S10).
DISCUSSION
In this cross‐sectional study, we compared a range of individual adverse perinatal outcomes as well as one composite outcome in newborns with fetal exposure to injectable first‐line treatments, dimethyl fumarate, or natalizumab with DMD‐unexposed pregnancies and the general population. The prevalence of individual outcomes as well as the composite AE outcome was comparable among the three groups. After adjusting, first for confounding and second for exposure definition, the OR and related 95% CI remained statistically insignificant in all analyses. However, we did find an increased occurrence of elective C‐sections in the exposed group when compared with the general population, although comparable to the occurrence found among DMD‐unexposed pregnancies.
The contraindication label for usage during pregnancy and breastfeeding was removed from glatiramer acetate and interferons by the European Medicine Agency in April 2017 and September 2019, respectively [19, 20, 21]. The largest study to date investigating pregnancy outcomes following glatiramer acetate exposure included 5042 pregnancies. Although not finding an increased prevalence of CM in exposed pregnancies, a major limitation to this study is that the fetal exposure to glatiramer acetate was unknown [21]. Later studies with early exposure to glatiramer acetate [22] and exposure during all three trimesters [23] have not demonstrated an association between exposure to glatiramer acetate and increased risk of CM, spontaneous abortion, preterm birth, birthweight, birth length, or C‐sections. In our study, 91 newborns were exposed to glatiramer acetate, with a median fetal exposure of 7.5 weeks (Q1–Q3 = 4.5–12). Our results were in concordance with previous studies in terms of CM, preterm birth, spontaneous abortion, and prevalence of C‐sections. Extensive data on pregnancy outcomes following early exposure to interferons have not demonstrated increased risk of adverse outcomes such as CM, preterm birth, stillbirth, or risk of spontaneous abortion [19, 20, 24]. The prevalence of CM was slightly higher in our study (5%, 23/460), demonstrated in the drug‐stratified analysis (Table S1), when compared to the three abovementioned studies, although comparable to our control group (3.9%) and that of the general population of Denmark (5%) [25].
Only few pregnancies (58/1009) in our cohort were exposed to dimethyl fumarate, of which 24 were aborted (nine spontaneously and 15 electively). The remaining 34 pregnancies all resulted in full‐term newborns with no registrations of stillbirth, placenta complications, or low Apgar score. Fewer than four newborns presented with CM or being SGA. Data on dimethyl fumarate‐exposed pregnancies are very sparse, and only few studies have reported outcomes. Clinical trial data on 39 exposed pregnancies with known outcome reported the same rate of births and spontaneous abortion as in the general population, with no information on CM in live births [26]. A recently published interim analysis from the international TecGistry collecting data on pregnancy outcomes following exposure to dimethyl fumarate included 351 pregnancies with known outcomes [27]. The median exposure time was 5 weeks, with only one pregnancy exposed later than the first trimester. The prevalence of spontaneous abortions and malformation was 8% and 2.9%, respectively, both within the range reported in the general population.
Diverging results following early exposure to natalizumab have been reported. Portaccio et al. found that exposure to natalizumab was a significant predictor of spontaneous abortion, although the reported prevalence of spontaneous abortion among the exposed pregnancies (17.4%) was similar to that of the general population [28]. Similarly, Ebrahimi et al. demonstrated an increased prevalence of spontaneous abortion in pregnancies with early natalizumab exposure (17.3%) when compared to healthy controls (4.1%) [29]. However, after excluding women older than 35 years from both groups, the prevalence in the two groups was very alike, 11.5% versus 12% respectively. Friend et al. reported an increased rate of CM in natalizumab exposed newborns (5.05%); however, no specific pattern was observed [30]. The prevalence of both spontaneous abortion (10.2%) and CM (2.4%) in our study is below that reported in the aforementioned studies and in concordance with those of the general population. Furthermore, none of the natalizumab‐exposed newborns was SGA or presented with low Apgar score. Overall, the proportion of elective C‐sections was increased among women with MS when compared with the general population, but an additional increment was observed among those treated with natalizumab (Table S1). This might be due to a higher disease burden in these women hence a more precautious approach regarding their birth plan. A notably larger proportion of women treated with natalizumab experienced relapse activity severe enough to require corticosteroid treatment (6.7%) compared with 3.5% and 1.3% among those treated with glatiramer acetate and interferons, respectively (data not shown). This corresponds with existing literature demonstrating an increased risk of relapse following natalizumab discontinuation [31, 32]. The occurrence of preterm birth was elevated among women treated with natalizumab (13%) when compared with the general population (6.5%). None of the preterm natalizumab‐exposed newborns had been exposed to corticosteroid treatment during pregnancy, and the majority of the preterm babies (12/16) were born in GW 36 and hence considered late preterm.
Hematological abnormalities have been reported in studies investigating outcomes in pregnancies with second or third trimester exposure to natalizumab, with increasing tendency the later in the pregnancy the exposure occurred [5, 6, 7]. Placental transfer is possible at the earliest at GW 13 and most likely during the third trimester based on the pharmacokinetics of the drug [4]. Hematological abnormalities were not represented among the registered CMs in natalizumab‐exposed newborns, which was as expected, as none of them had been exposed after GW 13.
Comparability of measurements encompassing birthweight and gestational age is difficult, as different definitions are used across studies. A large prospective cohort included 75,296 pregnancies from 12 European countries in which the prevalence of SGA varied from 4.6% to 15.3% [33]; hence, the prevalence of 4.4% among exposed newborns in our study was in the lower range. Furthermore, we did not find an increased risk of SGA newborns in the exposed group when compared with the unexposed newborn or the control group.
Use of corticosteroids during pregnancy has demonstrated an association with low birth weight [34], and if administered during the first trimester, it is also considered to be associated with an increased risk of orofacial abnormalities [35, 36]. According to the national Danish clinical practice, use of corticosteroids during the first trimester is solely restricted to patients experiencing a very severe relapse, and during second and third trimester only if deemed clinically necessary and after consultation with an obstetrician [37]. Therefore, treating physicians strive to minimize the use of corticosteroids during pregnancy; thus, treatment of relapses during pregnancy with corticosteroids was not included in the statistical analysis. Descriptively, a total of 25 women were reported with corticosteroid treatment due to relapses during pregnancy. Of these, 48% (12/25) of patients were exposed to natalizumab, eight of 25 to interferons, and the remaining five to glatiramer acetate. Among all newborns exposed to corticosteroids during pregnancy, there were three or fewer registrations of SGA and preterm birth and no registrations of malformations (data not shown).
The nationwide prevalence of elective C‐sections in Denmark from 1997 to 2018 was 8.2% [9, 38], which corresponds with the control group in our study. Elective C‐section was the only parameter in which the exposed and unexposed groups both differed significantly from the control group, although not from each other (17.2% and 16.6% vs. 8.6%, respectively). Most studies report C‐sections as one delivery mode without distinguishing between elective and acute C‐section, which complicates direct comparison with our results. A previous Danish cohort study compared perinatal outcomes and reported an increased prevalence in elective C‐sections of 13.8% in the MS population versus 7.8% in the general population [39]. Similarly, two Portuguese studies reported increased rates of C‐sections in women with MS (14.3% and 17.5%) [40, 41]. Possible explanations for the increased rate of elective C‐sections among women with MS might include MS‐related symptoms such as fatigue and/or neuromuscular perineal weakness, which could prompt both the clinician and women to be precautious.
We used nationwide registry data to include women with MS linked with population‐based registries on reproductive data that have virtually complete national coverage, which should eliminate recall and indication bias. Furthermore, access to health care in Denmark is government‐funded, providing equal access for all citizens. As all Danish registries have national coverage, our findings represent national generalizability. We found very similar results in all sensitivity analyses, regardless of the approach, which confirmed robust models and results. However, we were not able to stratify the analyses by trimester due to very few exposures in the second or third trimester, which calls for further investigation in future research. Data on pregnancies with early exposure to dimethyl fumarate should be considered preliminary, as the sample was quite small. Although the sensitivity analysis was adjusted for smoking, this variable has limitations of its own. First, it was a self‐reported outcome that is generally known to be exposed to recall bias. Second, it was adjusted for as a binary variable, as very few women reported smoking. Lastly, it was only available in relation to pregnancies resulting in birth; hence, the adjusted results on any AE outcome might be flawed, as spontaneous abortions were part of this composite outcome. Data on alcohol consumption or use of substances were not available and therefore not included. A final limitation of this study is related to abortions. The reported prevalence of spontaneous abortions is possibly underestimated, as only hospitalized registrations were available. Nonetheless, the underestimation is expected to be equally distributed among the groups and therefore unlikely to influence the results. Elective abortion is optional until the end of GW 12 in Denmark; thus, no record of the reason for the procedure is registered. The routine ultrasound performed to detect abnormalities is usually performed around GW 18–21; thus, it is unlikely to be the cause of the termination.
CONCLUSIONS
Our study confirms that newborns exposed to injectable first‐line treatments, dimethyl fumarate, or natalizumab demonstrate a similar prevalence of stillbirths, spontaneous abortions, placenta complications, low Apgar score, SGA, preterm birth, and CM as the general population. However, we did find an increased occurrence of elective C‐sections among both groups of women with MS when compared with the general population.
FUNDING INFORMATION
This study was funded by The Danish Multiple Sclerosis Society (grant number A‐34757).
CONFLICT OF INTEREST
J.B.A. has received travel and congress participation funds from Merck. F.S. has served on scientific advisory boards, been on the steering committees of clinical trials, served as a consultant, received support for congress participation, received speaker honoraria, or received research support for his laboratory from Biogen, Merck, Novartis, Roche, Sanofi and Genzyme, and Teva. M.M. has served on scientific advisory board for Biogen, Sanofi, Teva, Roche, Novartis, and Merck; has received honoraria for lecturing from Biogen, Merck, Novartis, Sanofi and Genzyme; and has received research support and support for congress participation from Biogen, Genzyme, Teva, Roche, Merck, and Novartis.
Supporting information
Table S1
Table S2
Table S3
Table S4
Table S5
Table S6
Table S7
Table S8
Table S9
Table S10
ACKNOWLEDGMENTS
We would like to thank the Danish Multiple Sclerosis Group for ongoing data collection around the country, which enables us to conduct studies on a continuously growing and updated nationwide cohort.
Andersen JB, Sellebjerg F, Magyari M. Pregnancy outcomes after early fetal exposure to injectable first‐line treatments, dimethyl fumarate, or natalizumab in Danish women with multiple sclerosis. Eur J Neurol. 2023;30:162‐171. doi: 10.1111/ene.15559
DATA AVAILABILITY STATEMENT
Anonymized data will be shared on request from any qualified researcher under approval from the Danish Data Protection Agency.
REFERENCES
- 1. EMA . European Medicines Agency ‐ Find medicine ‐ Aubagio, http://www.ema.europa.eu/ema/index.jsp?curl=pages/medicines/human/medicines/002514/human_med_001645.jsp&mid=WC0b01ac058001d124 (accessed 10 April 2018).
- 2. EMA . EMA Mavenclad. Accessed February 9, 2020. https://www.ema.europa.eu/en/documents/product‐information/mavenclad‐epar‐product‐information_en.pdf
- 3. Saji F, Samejima Y, Kamiura S, Koyama M. Dynamics of immunoglobulins at the feto‐maternal interface. Rev Reprod. 1999;4:81‐89. [DOI] [PubMed] [Google Scholar]
- 4. Palmeira P, Quinello C, Silveira‐Lessa AL, Zago CA, Carneiro‐Sampaio M. IgG placental transfer in healthy and pathological pregnancies. Clin Dev Immunol. 2012;2012:1‐13. doi: 10.1155/2012/985646 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Haghikia A, Langer‐Gould A, Rellensmann G, et al. Natalizumab use during the third trimester of pregnancy. JAMA Neurol. 2014;71:891‐895. [DOI] [PubMed] [Google Scholar]
- 6. Hellwig K, Thiel S, Meinl I, et al. Long‐term exposure to natalizumab during pregnancy ‐ a prospective case series from the German multiple sclerosis and pregnancy registry; 2018. Abstract 204.
- 7. Triplett JD, Vijayan S, Rajanayagam S, et al. Pregnancy Outcomes amongst Multiple Sclerosis Females with Third Trimester Natalizumab Use. Mult Scler Relat Disord. 2020;40:1‐4. doi: 10.1016/j.msard.2020.101961 [DOI] [PubMed] [Google Scholar]
- 8. Magyari M, Joensen H, Laursen B, Koch‐Henriksen N. The Danish multiple sclerosis registry. Brain Behav. 2020;11:1‐10. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Bliddal M, Broe A, Pottegård A, Olsen J, Langhoff‐Roos J. The Danish medical birth register. Eur J Epidemiol. 2018;33:27‐36. [DOI] [PubMed] [Google Scholar]
- 10. Schmidt M, Schmidt SAJ, Sandegaard JL, Ehrenstein V, Pedersen L, Sørensen HT. The Danish National Patient Registry: a review of content, data quality, and research potential. Clin Epidemiol. 2015;7:449. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Tølbøll Blenstrup L, Knudsen LB. Danish registers on aspects of reproduction. Scand J Public Health. 2011;39:79‐82. [DOI] [PubMed] [Google Scholar]
- 12. Schmidt M, Pedersen L, Sørensen HT. The Danish civil registration system as a tool in epidemiology. Eur J Epidemiol. 2014;29:541‐549. doi: 10.1007/s10654-014-9930-3 [DOI] [PubMed] [Google Scholar]
- 13. Sundhedsdatastyrelsen. www.esundhed.dk. 2019. http://end2019.esundhed.dk/sundhedsregistre/MFR/Sider/MFR06A.aspx
- 14. Sankilampi U, Hannila ML, Saari A, Gissler M, Dunkel L. New population‐based references for birth weight, length, and head circumference in singletons and twins from 23 to 43 gestation weeks. Ann Med. 2013;45:446‐454. [DOI] [PubMed] [Google Scholar]
- 15. American Academy of Pediatrics Committee on fetus and newborn. Use and abuse of the Apgar score. Pediatrics. 1996;98:141‐142. [PubMed] [Google Scholar]
- 16. EUROCAT Guide 1.4 and Reference Documents. 2018. https://eu‐rd‐platform.jrc.ec.europa.eu/sites/default/files/Full_Guide_1_4_version_28_DEC2018.pdf
- 17. Kolte AM, Westergaard D, Lidegaard Ø, Brunak S, Nielsen HS. Chance of live birth: a nationwide, registry‐based cohort study. Hum Reprod. 2021;36:1065‐1073. [DOI] [PubMed] [Google Scholar]
- 18. Li R, Lodge J, Flatley C, Kumar S. The burden of adverse obstetric and perinatal outcomes from maternal smoking in an Australian cohort. Aust New Zeal J Obstet Gynaecol. 2019;59:356‐361. [DOI] [PubMed] [Google Scholar]
- 19. Hellwig K, Geissbuehler Y, Sabidó M, et al. Pregnancy outcomes in interferon‐beta‐exposed patients with multiple sclerosis: results from the European interferon‐beta pregnancy registry. J Neurol. 2020;267:1715‐1723. doi: 10.1007/s00415-020-09762-y [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Hakkarainen KM, Juuti R, Burkill S, et al. Pregnancy outcomes after exposure to interferon beta: a register‐based cohort study among women with MS in Finland and Sweden. Ther Adv Neurol Disord. 2020;13:175628642095107. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Sandberg‐Wollheim M, Neudorfer O, Grinspan A, et al. Pregnancy outcomes from the branded glatiramer acetate pregnancy database. Int J MS Care. 2018;20:9‐14. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Herbstritt S, Langer‐Gould A, Rockhoff M, et al. Glatiramer acetate during early pregnancy: a prospective cohort study. Mult Scler J. 2016;22:810‐816. [DOI] [PubMed] [Google Scholar]
- 23. Hellwig K, Neudorfer O, Melamed‐Gal S, Baruch P, Qassem S. Pregnancy outcomes in patients with multiple sclerosis and exposure to branded glatiramer acetate during all three trimesters (P4.362). Neurology. 2018;90(15 Suppl). https://n.neurology.org/content/90/15_Supplement/P4.362 [Google Scholar]
- 24. Thiel S, Langer‐Gould A, Rockhoff M, et al. Interferon‐beta exposure during first trimester is safe in women with multiple sclerosis‐a prospective cohort study from the German multiple sclerosis and pregnancy registry. Mult Scler J. 2016;22:801‐809. [DOI] [PubMed] [Google Scholar]
- 25. Broe A, Damkier P, Pottegård A, Hallas J, Bliddal M. Congenital malformations in Denmark: considerations for the use of Danish health care registries. Clin Epidemiol. 2020;12:1371‐1380. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Gold R, Phillips JT, Havrdova E, et al. Delayed‐release dimethyl fumarate and pregnancy: preclinical studies and pregnancy outcomes from clinical trials and postmarketing experience. Neurol Ther. 2015;4:93‐104. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27. Hellwig K, Rog D, McGuigan C, et al. Interim analysis of pregnancy outcomes after exposure to dimethyl fumarate in a prospective international registry. Neurol ‐ Neuroimmunol Neuroinflammation. 2022;9:e1114. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28. Portaccio E, Moiola L, Martinelli V, et al. Pregnancy decision‐making in women with multiple sclerosis treated with natalizumab: I: fetal risks. Neurology. 2018;6:e823‐e831. [DOI] [PubMed] [Google Scholar]
- 29. Ebrahimi N, Herbstritt S, Gold R, Amezcua L, Koren G, Hellwig K. Pregnancy and fetal outcomes following natalizumab exposure in pregnancy. A prospective, controlled observational study. Mult Scler J. 2015;21:198‐205. [DOI] [PubMed] [Google Scholar]
- 30. Friend S, Richman S, Bloomgren G, et al. Evaluation of pregnancy outcomes from the Tysabri® (natalizumab) pregnancy exposure registry: a global, observational, follow‐up study. BMC Neurol. 2016;16(1):150. doi: 10.1186/s12883-016-0674-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31. Portaccio E, Annovazzi P, Ghezzi A, et al. Pregnancy decision‐making in women with multiple sclerosis treated with natalizumab II: maternal risks. Neurology. 2018;90:e823‐e831. [DOI] [PubMed] [Google Scholar]
- 32. Sorensen PS, Koch‐Henriksen N, Petersen T, Ravnborg M, Oturai A, Sellebjerg F. Recurrence or rebound of clinical relapses after discontinuation of natalizumab therapy in highly active MS patients. J Neurol. 2014;261:1170‐1177. [DOI] [PubMed] [Google Scholar]
- 33. Ruiz M, Goldblatt P, Morrison J, et al. Mother's education and the risk of preterm and small for gestational age birth: a DRIVERS meta‐analysis of 12 European cohorts. J Epidemiol Community Health. 2015;69:826‐833. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34. Rodriguez A, Wang Y, Ali Khan A, Cartwright R, Gissler M, Järvelin MR. Antenatal corticosteroid therapy (ACT) and size at birth: a population‐based analysis using the Finnish medical birth register. PLoS Med. 2019;16:e1002746. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35. Pradat P, Robert‐Gnansia E, Di Tanna GL, et al. First trimester exposure to corticosteroids and oral clefts. Birth Defects Res Part A Clin Mol Teratol. 2003;67:968‐970. [DOI] [PubMed] [Google Scholar]
- 36. Kemp MW, Newnham JP, Challis JG, Jobe AH, Stock SJ. The clinical use of corticosteroids in pregnancy. Hum Reprod Update. 2015;22:240‐259. [DOI] [PubMed] [Google Scholar]
- 37. Neurologisk National Behandlingsvejledning. Dansk Neurologisk Selskab. Accessed December, 2021. https://neuro.dk/wordpress/nnbv/behandling‐af‐ms‐attak/
- 38. Fødte og fødsler (1997). Accessed October 3, 2021. https://www.esundhed.dk/home/emner/graviditet foedsler og boern/foedte og foedsler 1997.
- 39. Andersen JB, Kopp TI, Sellebjerg F, Magyari M. Pregnancy‐related and perinatal outcomes in women with multiple sclerosis. Neurol Clin Pract. 2021;11:280‐290. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40. Jesus‐Ribeiro J, Correia I, Martins AI, et al. Pregnancy in multiple sclerosis: a Portuguese cohort study. Mult Scler Relat Disord. 2017;17:63‐68. [DOI] [PubMed] [Google Scholar]
- 41. Novo A, Castelo J, de Sousa A, et al. Pregnancy outcomes in Portuguese women with multiple sclerosis: the PREGNIMS study. Mult Scler Relat Disord. 2019;28:172‐176. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Table S1
Table S2
Table S3
Table S4
Table S5
Table S6
Table S7
Table S8
Table S9
Table S10
Data Availability Statement
Anonymized data will be shared on request from any qualified researcher under approval from the Danish Data Protection Agency.