Trastuzumab administration during pregnancy: an update

Background Over than one third (28–58%) of pregnancy-associated breast cancer (PABC) cases are characterized by positive epidermal growth factor receptor 2-positive (HER2) expression. Trastuzumab anti-HER2 monoclonal antibody is still the benchmark treatment of HER2-positive breast tumors. However, FDA has categorized Trastuzumab as a category D drug for pregnant patients with breast cancer. This systemic review aims to synthesize all currently available data of trastuzumab administration during pregnancy and provide an updated view of the effect of trastuzumab on fetal and maternal outcome. Methods Eligible articles were identified by a search of MEDLINE bibliographic database and ClinicalTrials.gov for the period up to 01/09/2020; The algorithm consisted of a predefined combination of the words “breast”, “cancer”, “trastuzumab” and “pregnancy”. This study was performed in accordance with the PRISMA guidelines. Results A total of 28 eligible studies were identified (30 patients, 32 fetuses). In more than half of cases, trastuzumab was administered in the metastatic setting. The mean duration of trastuzumab administration during gestation was 15.7 weeks (SD: 10.8; median: 17.5; range: 1–32). Oligohydramnios or anhydramnios was the most common (58.1%) adverse event reported in all cases. There was a statistically significant decrease in oligohydramnios/anhydramnios incidence in patients receiving trastuzumab only during the first trimester (P = 0.026, Fisher’s exact test). In 43.3% of cases a completely healthy neonate was born. 41.7% of fetuses exposed to trastuzumab during the second and/or third trimester were born completely healthy versus 75.0% of fetuses exposed exclusively in the first trimester. All mothers were alive at a median follow-up of 47.0 months (ranging between 9 and 100 months). Of note, there were three cases (10%) of cardiotoxicity and decreased ejection fraction during pregnancy. Conclusions Overall, treatment with trastuzumab should be postponed until after delivery, otherwise pregnancy should be closely monitored.


Background
Pregnancy-associated breast cancer (PABC) is defined as any breast carcinoma diagnosed during pregnancy or during the first postpartum year [1]. It occurs in 1 to 3000 pregnancies while it has been estimated that up to 3% of breast cancers may be diagnosed in pregnant women [1,2]. The incidence of PABC is also increasing due to advanced maternal age in today's society. Median age of disease is 33 years (23-47 years), while there is a 2-to 3-fold decreased risk of PABC in women younger than 30 [1]. Interestingly enough, there is an increased incidence (54-80%) of estrogen receptor (ER)negative tumors in pregnancy-related tumors [1]. This could be explained by downregulation of the receptors as a negative feedback effect of estrogen and progesterone upon hormonal receptor expression. The greater incidence of ER-negative breast cancer in pregnant women mainly stems from the young age of onset. However, some studies demonstrated that the percentage of ER-positive pregnancy-associated breast cancers was not significantly different from that of non-pregnant age-matched patients [3,4]. On the other hand, epidermal growth factor receptor 2 -positive (HER2) tumors compose the 28-58% of PABC [3][4][5]. Although Elledge et al. found 7 out of 12 pregnant patients (58%) to be positive for HER2, Middleton et al. found no difference in the HER2 expression rate (28%) between pregnant and young nonpregnant women [3,4]. Amant et al. reported an 31.8% incidence of HER2-positive tumors in pregnant women which is consistent with the results provided by Cardonick et al (27%) [6,7]. Overall, the incidence of HER2positive tumors was approximately equal to this of patients with breast cancer younger than 35 years old (39%), although it still remains a significant proportion [8].
Treatment of pregnant women with breast cancer represents a clinically challenging case in terms of maternal and fetal safety. Treatment of HER-2 positive PABC relies on the administration of trastuzumab anti-HER2 monoclonal antibody which remains the standard-ofcare for all HER2-positive breast tumors. Trastuzumab binds HER2 on the C-terminal portion of domain IV and inhibits HER2 proteolytic cleavage and release of the extracellular domain in breast cancer cells [9]. Cells treated with trastuzumab undergo arrest during the G1 phase of the cell cycle leading to reduced proliferation. Trastuzumab exerts its antitumor activity through antibody-dependent cell-mediated cytotoxicity. However, our knowledge remains limited on the use and safety of trastuzumab during pregnancy because of its cytotoxic nature. Adverse effects of trastuzumab treatment include hematological and gastrointestinal disorders as well as cardiovascular effects that could potentially threaten pregnancy outcome.
In vivo studies conducted in cynomolgus monkeys at doses up to 25 times that of the weekly human maintenance dose of 2 mg/kg Herceptin revealed no evidence of harm to the fetus. However, trastuzumab transfer through the placenta has been observed during the early (days 20-50 of gestation) and late (days 120-150 of gestation) pregnancy period [10]. A warning about trastuzumab administration during pregnancy states that administration should be avoided during gestation unless it is mandatory for mother's health. As for patients with breast cancer that become pregnant while receiving Trastuzumab or within 7 months after the last dose, close monitoring is indispensable.
The aim of this systematic review is to provide un updated consensus regarding trastuzumab administration during pregnancy after synthesizing all existing data emerging from case reports and individual cases. We previously conducted a relevant systematic review assessing exposure to trastuzumab during pregnancy that was published in 2012 [11]. Since there is new emerging evidence from additional cases during all these years, an updated review of literature would contribute to revision of existing data and reconsideration of current practice.

Methods
This systematic review was performed in accordance with PRISMA guidelines [12]. Eligible articles were identified by a search of MEDLINE bibliographic database and ClinicalTrials.gov for the period up to September 2020. The search algorithm consisted of the following keywords: (breast AND (carcinoma OR carcinomas OR cancer OR cancers OR neoplasm OR neoplasms)) AND (pregnancy OR pregnant OR gestation) AND (trastuzumab OR herceptin). In order to maximize the amount of synthesized information, we meticulously examined the reference lists of the relevant reviews and articles retrieved for potentially eligible papers. Language restrictions were not applied. All studies that examined the efficacy and safety of trastuzumab during pregnancy were eligible for this systematic review, no matter of sample size. All cases where therapeutic or spontaneous abortion occurred were excluded. In addition, articles assessing trastuzumab administration before or after the gestation period were considered ineligible. Eligible studies required the administration of trastuzumab at some point during pregnancy even if treatment commenced prior to pregnancy initiation. Moreover, reviews were ineligible, while all prospective and retrospective studies, as well as case reports, were eligible for this systematic review. In cases where overlapping publications emerging from the same study were identified, the larger size study was included. Two independently working reviewers (FZ and AA) performed the selection of studies and any disagreements were resolved by team consensus.
Oligohydramnios or anhydramnios was the most common (58.1%) adverse event reported in all cases [22, 24-27, 29-33, 35, 36, 38, 39, 43, 44, 47, 49]. Only one of the six cases (16.7%) of trastuzumab exposure exclusively during the first trimester of gestation was complicated with oligohydramnios or anhydramnios. In contrast, seventeen out of 24 pregnancies (70.8%) where trastuzumab was administered during the second or/and third trimester were complicated with oligohydramnios or anhydramnios. The difference was statistically significant (P = 0.026, Fisher's exact test). The trend pointing to a positive association between the duration of trastuzumab treatment and the development of oligohydramnios or anhydramnios did not reach statistical significance (OR = 1.05, 95% CI: 0.96-1.14, increment: 1 week, P = 0.316).
As far as maternal outcome is concerned, all patients with breast cancer were alive at a median follow-up of 47.0 months (ranging between 9 and 100 months), while only two patients relapsed during follow-up according to existing data. It should be noted that there were three cases (10%) of cardiotoxicity and decreased ejection fraction during pregnancy [21,28,48].
Detailed information of all eligible studies is provided in Table 1.

Discussion
PABC is a rare but complex entity which demands multidisciplinary management. Amant et al. reported similar survival rates between pregnant patients with breast cancer and the matched non-pregnant population, despite the preceding belief that PABC is associated with a poor outcome [6]. The finding that survival rates of pregnant BC patients are comparable to those of nonpregnant is essential for mother counselling and optimization of PABC management. Breast cancer treatment during pregnancy does not jeopardize maternal prognosis.
Specific guidelines have been developed for PABC treatment. Surgery is not contraindicated during pregnancy and the type of surgery chosen should be based on usual criteria (mastectomy versus breast conserving surgery) [50]. The lack of wound complications in pregnant BC patients supports surgical management of PABC. Despite the concern of milk fistulae development and that of postoperative hematoma due to the hypervascularization of the breasts, there was no apparent increase in surgical complications between pregnant and nonpregnant patients [50].
Chemotherapy should be started after the first trimester and should be stopped 2-3 weeks prior to delivery for a chemotherapy-free interval. The teratogenicity of chemotherapy depends on time of exposure, dose administered and placental transfer. During the first 2 weeks of pregnancy, spontaneous abortion is more common after chemotherapy treatment, rather than teratogenic effects on the fetus. However, as organogenesis happens from 2nd to 8th gestational week, the embryo is more vulnerable to congenital malformations during this period. The risk of chemotherapy-induced congenital malformations during the 1st trimester is 20%, whereas it declines to 1-2% during the 2nd and 3d trimester [51][52][53].
Regarding trastuzumab administration during pregnancy, we report that in approximately two third of cases oligohydramnios or anhydramnios were developed. The risk for intrauterine complications and oligohydramnios/anhydramnios development was minimal in pregnancies were fetal exposure to trastuzumab occurred exclusively during the first trimester (16.7% vs 70.8%; P = 0.026). In addition, a completely healthy neonate was born in 75% of cases that trastuzumab treatment affected only the first trimester in contrast with 41.7% of cases where the fetus was exposed during second and/or third trimester as well, although this association failed to reach statistical significance (P = 0.311; Fisher's exact test). Our results are consistent with existing knowledge. Fetal exposure to trastuzumab is considered to be low during the first trimester while it gradually increases during the second half of gestation to reach mother levels at delivery [54]. Trastuzumab in an IgG1 monoclonal antibody with a molecular mass that does not permit transport across the placenta via simple diffusion. Active transport of these antibodies require binding to the Fc receptor of the syncytiotrophoblast, however Fc receptor is hardly detectable before the 14th week of gestation. Therefore, placental transfer of trastuzumab during the first trimester is minimal [54].
Antibody transfer to fetal endocrine organs has been reported from 4th to 6th week of development, although concentration was rather low [55]. Indeed, fetuses exposed to trastuzumab exclusively in the 1st trimester tend to be healthy at delivery. This observation is important for the correct consultation of women diagnosed with pregnancy while being on trastuzumab treatment.
US FDA has categorized Trastuzumab as a category D drug due to fetal complications [56]. No congenital malformations or teratogenic effects were reported in our review, apart from some prematurity-related problems [44,47]. Indeed, animal studies did not reveal any teratogenic effects even at doses up to 25 times the recommended weekly human dose [56].
The most common fetal complication observed in our study was oligohydramnios or anhydramnios during second or third trimester. This is a result of EGFR receptor blockade in fetal renal epithelium by trastuzumab, where EGFR receptors are highly expressed [57]. Indeed, EGFR binding affinity in human fetal kidney between 6 and 11th GA week is 4-5 times greater than in normal renal tissue. This abundance of EGFR binding sites in fetal kidney falls rapidly in the postnatal period to low levels observed in adult renal epithelium [57]. EGF increases DNA synthesis in human fetal kidney cells, while anti-EGFR antibody leads to the opposite result. Therefore, trastuzumab attenuates the important role of EGFR receptors in fetal renal cell proliferation and nephrogenesis resulting in the aforementioned cases of oligo−/anhydramnios, fetal renal failure [29,47,49] and renal agenesis [24,47] reported in our study. In addition, even in the presence of normal kidneys, this receptor blockade leads to decreased urinary output and empty fetal bladder visualization. Another explanation of the trastuzumab-induced oligohydramnios is the decreased expression of vascular endothelial growth factor (VEGF). VEGF regulates amniotic fluid production and absorption via modulation of the rate of intramembranous absorption of amniotic fluid by both passive and nonpassive mechanisms [58]. Trastuzumab downregulates VEGF expression and may affect the amniotic fluid level.
Another possible mechanism of trastuzumab-mediated reduction of amniotic fluid may be through altering the function of aquaporins, a family of cell membrane water channels responsible for intramembranous fluid exchange in various tissues as proposed by Sekar and Stone [33]. More specifically, Aquaporin-3 is expressed in placenta, chorion, and amnion and regulation of its expression may contribute to amniotic fluid homeostasis [59]. Moreover, it was shown that Aquaporin-8 and Aquaporin-9 levels were significantly decreased in amnion and increased in placenta in fetuses suffering from oligohydramnios, further supporting the effect of aquaporins in amniotic fluid levels [60].
Whatever the mechanism, there is increased evidence that oligohydramnios induced by trastuzumab is reversible upon discontinuation of treatment [25,27,29]. The mean half-life elimination time for trastuzumab weekly schedule is 6 days (range: 1-32 days) and for the 3-week schedule 16 days (range: 11-23 days) [53]. This may indicate the time required for recovery of oligo−/anhydramnios after trastuzumab treatment. Moreover, it has been shown that the risk of oligo/anhydramnios is analogous to the duration of exposure to trastuzumab during pregnancy. Trastuzumab exposure for a relatively short period does not seem to substantially affect the pregnancy outcome. In contrast, a more prolonged period of exposure is associated with increased risk of fetal harm. In our study, there was a trend to a positive association between the duration of trastuzumab treatment and the development of oligohydramnios or anhydramnios, although not statistically significant (OR = 1.05, 95% CI: 0.96-1.14, increment: 1 week, P = 0.316).
Berveiller et al. reported one case of ectopic cervicoisthmic pregnancy while on trastuzumab treatment [19]. ErbB2 is required embryo implantation process [61]. On days 1-4 ErbB2 mRNA is expressed in uterine epithelial cells, while on days 6-8 the mRNA was accumulated in both implantation and interimplantation sites [61]. Given the crucial role of HER2 in embryo implantation process, it could be postulated that trastuzumab was responsible for the incidence of this ectopic pregnancy.
Azim et al explored the effect of previous or concurrent trastuzumab administration on pregnancy outcome based on data emerging from HERA trial, one of the largest Phase III trials evaluating trastuzumab treatment in the adjuvant setting [20]. Azim et al. reported that 25% of pregnancies that occurred while on trastuzumab treatment resulted in spontaneous abortions. In consistence with our results, no congenital anomalies were reported in the study. However, there were also no cases of oligohydramnios or anhydramnios recorded in the study. The study demonstrated that women that conceived after a period of 3 months after trastuzumab cessation had an uneventful pregnancy, a finding that contributes significantly to the existing knowledge [20]. Of note, results from the very interesting MOTHER trial are anticipated [23].

Conclusions
Overall, medical oncologists encounter the dilemma of choosing between the optimal therapy for the mother and survival of the fetus. Considering that trastuzumab is equally effective when administered within 6 months from breast cancer diagnosis, it may be delayed until after delivery [62]. However, if trastuzumab administration is inevitable as in the case of metastatic disease, close monitoring of both mother and the fetus is required. It should be noted that inadvertent conception while taking Herceptin during the first trimester only is not an indication for termination.
NR: Not reported.