Acute lymphoblastic leukemia during the third trimester of pregnancy

The diagnosis of cancer during pregnancy is relatively rare with an estimated incidence of 1 in 1000 pregnancies [1]. Before 1960, termination of pregnancy was the standard of care [2], but nowadays, pregnant women with a malignancy are being treated with chemotherapy when possible. The expert opinion is that administration of chemotherapy is contraindicated during the first trimester of pregnancy, but from 14 weeks onwards until 3–4 weeks before delivery, chemotherapy is considered relatively safe in pregnant patients with acute leukemia [3–5], also in the long term [6,7]. Exposure to chemotherapy in the last weeks of pregnancy can induce hematological toxicity with risk of infectious and bleeding complications for mother and child during and after delivery and is therefore advised against. Here, we report on a 37-year-old woman in her 36th week of pregnancy, who presented with acute lymphoblastic leukemia (ALL). We describe the dilemma’s that we faced to time a safe delivery of the baby without significantly delaying curative treatment of the mother and we review the literature on this subject, including the chances of placental transfer of ALL.

This 37-year-old, 36 + 3-week pregnant woman (gravida 2, para 1), presented with extreme fatigue and night sweats. The pregnancy until then had been uneventful and apart from a thrombophlebitis 8 years earlier she had no medical history. Physical examination revealed lymph nodes up to 2 cm in diameter in the neck and axillary regions and an enlarged uterus in accordance with the duration of gestation. Laboratory results demonstrated hyperleukocytosis (376 × 10⁹/L) with 99% leukemic blasts, severe anemia and spontaneous tumor lysis syndrome (Table 1). Bone marrow aspiration showed >95% leukemic blasts, immunophenotypically matching T-acute lymphoblastic leukemia (ALL) (CD45dim CD38 + CD5 + CD7 + CD3 + CD4- CD8- CD1 + CD10 + TdT+; no expression of myeloid markers). Cytogenetic testing revealed a t(10;14)(q24;q11) TLX1 translocation with a complex karyotype; no molecular abnormalities were found. Fundoscopic examination excluded hyperviscosity syndrome. Fetal examination by cardiotocography and ultrasound showed no signs of fetal distress. Rapid blast reduction with leukapheresis was considered but decided against in the absence of hyperviscosity syndrome, and treatment was started within hours after presentation, with a 7-day prephase of prednisolone (30 mg/m² BID) according to protocol (HOVON 100, Dutch Trial Registry #NTR2004). Tumor lysis syndrome was treated with hyperhydration and rasburicase.

Table 1. Laboratory results at T-ALL diagnosis.

Test	Result	Normal range
Hemoglobin (mmol/L)	4.7	7.5–10.0
Platelets (×10^9/L)	81	150–400
Leukocytes (×10^9/L)	376	4.0–10.5
Leukemic blasts (%)	99	0
Creatinine (µmol/L)	102	65–95
Potassium (mmol/L)	4.9	3.5–4.5
Lactate dehydrogenase (U/L)	2361	0–247
Uric acid (mmol/L)	0.60	0.12–0.42
Calcium (mmol/L)	2.44	2.20–2.60
Phosphorus (mmol/L)	0.51	0.70–1.45

Early delivery of the child was discussed in a multidisciplinary deliberation between the hematologist, obstetrician, neonatologist, midwife and anesthesiologist. Vaginal delivery of the patient’s firstborn 1.5 years earlier had been complicated by severe postpartum hemorrhage with two liters of blood loss. The patient had a normal partial thromboplastin time and activated partial thromboplastin time and a moderate thrombocytopenia (70 × 10⁹/L, normal range 150–400). A cesarean section was assumed to delay chemotherapy with at least a week because of the risk of impaired wound healing and infections as a result of chemotherapy and continued treatment with high-dose prednisolone. Temporal withdrawal of ALL treatment was not considered an option given the hyperleukocytosis. It was decided to induce labor at day 6 of prednisolone pre-phase therapy, at a time the number of circulating leukemic blast was expected to be significantly reduced, with daily monitoring of fetal condition by cardiotocography until that day. A detailed back up plan was prepared in advance in case of fetal distress, spontaneous labor, failure to progress during labor or severe postpartum hemorrhage. At the 6th day of prednisolone prephase therapy the leukocyte count was down to 34 × 10⁹/L, and labor was induced as planned by amniotomy followed by intravenous administration of oxytocin. A healthy baby boy was delivered at a gestational age of 37 + 2 weeks with a normal birth weight (3335 grams) and normal Apgar scores (9 and 10 at 1 and 5 min, respectively). There was no significant blood loss.

Forty-eight hours after giving vaginal birth, remission-induction chemotherapy was started per protocol with the start of vincristine (weekly), daunorubicine, PEG-L-asparaginase and continuation of prednisolone. Because of the increased risk of venous thrombosis during pregnancy and puerperium combined with the thrombosis risk associated with newly diagnosed ALL treated with steroids and particularly asparginase, the patient received low-molecular-weight heparin (LMWH) in therapeutic dose. Prophylactic intrathecal methotrexate at days 8 and 15 was postponed given its potential risk on inducing cerebral venous thrombosis [8]. Since treatment with PEG-asparaginase has also been associated with increased risk of thrombosis [9], the first dose was postponed for one week. LMWH therapy was continued for six weeks postpartum. These modest modifications of the treatment scheme allowed full dose antileukemia treatment of this patient, which is important for prognosis [10]. Apart from an uncomplicated urinary tract infection, transient PEG-asparaginase-induced hypertriglyceridemia and mild peripheral neuropathy due to vincistrine, no complications occurred. The patient was advised to refrain from breastfeeding given the potential excretion in breast milk of methotrexate and etoposide, which have demonstrated toxicity to the baby. The patient obtained complete morphologic and molecular remission after remission-induction chemotherapy and has at the time of writing of this paper completed maintenance therapy.

At the moment of delivery, the patient had a peripheral blood leukemic blast count of 34 × 10⁹/L. Microscopic examination of the placenta demonstrated the presence of leukemic blasts in the maternal placental circulation, without invasion of chorionic villi (Figure 1(A)). Accordingly, cord blood samples were analyzed morphologically and immunophenotypically but revealed no signs of T-ALL translocation (Figure 1(B,C)). A patient-specific primer targeting the T-cell receptor β and δ loci of the leukemia was designed for minimal residual disease assessment and was found negative on the cord blood sample. The child is at the time of writing of this paper over 2.5 years of age and thriving.

Figure 1. No placental transfer of ALL. (A) Tissue section of the placenta showing several TdT + lymphoblasts in the intervillous space (arrows) and the absence of leukemic blasts in the chorionic villi. (B) H&E staining of cord blood smear showing no leukemic blasts. Insert: ALL blasts in a peripheral blood smear of the patient taken at the day of delivery. (C) Immunophenotyping of cord blood demonstrating the absence of CD2 + CD1 + CD10 + ALL blasts.

In this case, the placenta served as a barrier against maternal leukemia transfer, but in the very rare event that the placental barrier fails [1,11], the immune system of the infant may clear the leukemia. Clearance of maternal ALL blasts that were detectable by RQ-PCR in a cord blood sample, but not at later time points in the infants circulation, has been described [12]. In another case report, AML transmission from mother to child did lead to the development of AML in the child, 22 months later. The leukemic clone had a 46, XX karyotype while the infant was a boy and matched the mother’s leukemia immunophenotype and HLA type [13]. Apparently the leukemic clone had escaped the infants immune surveillance, as was objectified in vitro by the absence of reactivity of the child’s lymphocytes to his mother’s or his own leukemic blasts in a mixed lymphocyte reaction. Probably both the placental barrier and effective immune surveillance against maternal leukemic cells are important in protecting an infant from maternal leukemia.

A diagnosis of acute leukemia in pregnancy is a rarity, illustrated by a recent case series describing 23 patients between 1962 and 2016 in the Mayo Clinic [14]. Only six of these patients were in their third trimester. ALL in the 36th week of pregnancy poses the health care team for specific challenges, in particular when immediate action, for example because of hyperleukocytosis and severe anemia, is warranted. Rapid tumor reduction was achieved with prednisolone monotherapy, and vaginal birth was initiated in a timed and controlled way, after which the patient started chemotherapy without delay. Despite the presence of ALL blasts in the placental circulation, ALL transfer did not occur. With this report, we intend to give insight in the clinical reasoning as well as risk versus benefit assessments made by all health care providers involved in treating this patient, in order to contribute to the decision making in similar future cases.

Potential conflict of interest

Disclosure forms provided by the authors are available with the full text of this article online at https://doi.org/10.1080/10428194.2017.1375104.

Acknowledgements

We thank the patient for giving permission to publish her data. AVK and MDH wrote the first draft of the paper. All coauthors contributed significantly to the clinical care of the patient and contributed to the writing and revising of the paper.