Abstract
Animal studies with imatinib have shown it to be teratogenic, embryotoxic and to result in increased rates of post-implantation loss. As a consequence there are justifiable physician/patient concerns regarding the safety of tyrosine kinase inhibitors during pregnancy. We can now report on 180 mothers (the majority with chronic myeloid leukemia but some treated for other indications) exposed to the drug during pregnancy. Timing of exposure to imatinib by trimester is available for 146 (81%). 71% of these were exposed in the 1st trimester (includes 4 exposed in both 1st and 2nd trimesters), 38 (26%) were exposed in all trimesters and 4 only after the 1st trimester. Outcome data are known for 125/180 (69%). 63 pregnancies resulted in the birth of normal live infants (50% of those with known outcome (KO), 35% of all pregnancies). 35 women (28% of KO or 19.5% of all pregnancies) underwent elective terminations, 3 following identification of fetal defects. The remainder either had no defects or were of unknown status. There were 12 pregnancies with fetal abnormalities (9.6% KO), resulting in 8 live and 1 still birth. 18 pregnancies (14.4% KO) ended in spontaneous abortion. Although within expected limits for the general population (spontaneous abortion rates of 10–15%) these data may be skewed by absent data from 55 cases. In 51 patients whose pregnancies were reported after a known outcome (as opposed to reporting at the time of confirmation of pregnancy) 11 (22%) resulted in spontaneous abortion- a higher rate than expected in the general population which may suggest an abortifacient effect. Abnormalities included one infant with premature closure of the skull sutures (craniosynostosis), one with hypoplastic lungs, exomphalos, left duplex kidney, right absent kidney, hemivertebrae and a right shoulder anomaly, another child with exomphalos, right renal agenesis and hemivertebrae and a fourth with a small exomphalos and scoliosis. These 4 cases are pertinent as in animal models imatinib causes similar bony defects including exencephaly, encephaloceles and deformities of the skull bones. Other severe defects included a child with communicative hydrocephalus, cerebellar hypoplasia, atrial septal defect, over-riding aorta, ascites and pericardial effusion - this infant was born live but later died. A baby with a meningocoele was still born. The remaining cases include cleft palate and polydactyly, 2 cases of hypospadias, 1 case of pyloric stenosis, 1 abnormal ultrasound with raised AFP and 1 case of warfarin embryopathy. Despite this information, balancing the risk to the foetus of continuing imatinib vs. the risk to the mother of stopping it remains difficult. In a recent report of 10 women who interrupted treatment with imatinib due to pregnancy, 6 had an increase in Ph positive metaphases. At a median of 18 months since restarting imatinib only 3/10 have achieved a complete cytogenetic response. Given the above findings the current recommendation to avoid imatinib during pregnancy remains unchanged. In cases of accidental or desired pregnancy, risk/benefit evaluations must be carried out on an individual basis with careful counselling of both parents. As experience with imatinib continues to grow, further information will become available which should clarify the situation. To facilitate this a global pregnancy registry will be established in the near future.
Warfarin embryopathy: Balancing maternal and fetal risks with anticoagulation therapy
