Hematological malignancies and pregnancy: Treat or no treat during first trimester
Treatment of hematological malignancies (HM) during pregnancy remain unsolved, although the use of chemotherapy during second and third trimester has been accepted because of the low rate of toxicities, the use of cytotoxic drugs during first trimester is generally forbidden. Most of the concerns are related to congenital abnormalities and development, but long-term follow-up of these children are not available. From 1975 to 2008, we diagnosed and treated 15,750 cases of HM, and 143 female patients were pregnant during this time that were treated with combined chemotherapy. In our study, we present the long-term follow-up (the median follow-up was 22.4 years with a range of 3.8–32.0 years) of 54 newborns, whose mothers received chemotherapy during the first trimester of pregnancy with an intent-to-cure HM. Physical and neurological development were carefully assessed, and cardiac and chromosomal studies were performed until the age of 20 years to evaluate late toxicities. The obstetrical development of pregnancy was normal, chemotherapy was used at doses and schedules used in normal patients. Low-weight birth was the most frequent finding. No congenital abnormalities were detected. Physical, psychological and neurological developments were normal. Education and academic degree were according to the economical and social factors. Cardiac function and chromosomal examination were normal. No neoplasm or acute leukemia has been observed in these children. Forty-three mothers are alive and disease-free and can be considered cured. The use of cytotoxic drugs during the first trimester to treat HM seems to be beneficial to both the mother and fetus, and chemotherapy during the first trimester can be considered if the cure of the patient is the goal.
Cancer is the leading cause of death in women of childbearing age. However, its simultaneous occurrence during pregnancy is uncommon, with a reported incidence of 0.007%–0.2%. The coincident occurrence of pregnancy and cancer present complex therapeutics problems for the patient, oncologist, hematologist, obstetricians and pediatricians, almost legal, ethical and religious problems.
The most common hematological malignancies (HM) during pregnancy are acute leukemia (AL), Hodgkin's lymphoma (HL) and non-Hodgkin's lymphoma (NHL). Several studies and recent review articles have analyzed the dilemmas of the diseases associated with pregnancy.1–14 Owing to the fact that anticancer therapy to treat the mother could exert serious adverse events to the developing fetus, there has been a call for the development of tailored strategies for the patient.
In fact, it seems that active-specific agents in HM can be used during the second and third trimester of pregnancy, because no adverse events have been documented in the newborn.6, 8, 10, 11, 13, 15–17 However, the use of the aforementioned treatment during the first trimester remains to be defined, because these treatments have been associated with an increase incidence of spontaneous abortion, fetal death and major congenital malformations.4–15
In our institution, we began a prospective study to assess the use and toxicity of modern chemotherapy in an intent-to-cure HM, both in mother and fetus. Considering the status of the mother and the disease, delayed treatment was believed to be very dangerous to the fetus and the mother. In the mother, the goals were whether the chemotherapy affected the course of pregnancy, tolerance to treatment, response and overall survival, while in the fetus, the goals were the presence of congenital abnormalities, late toxicities in specific organs, namely, brain, heart, endocrine and reproductive system, and outcome in physical development, behavior in school and apprenticeships and academic degree.
Patients and Methods
This was an open study performed in a single center, our hospital has population coverage of 31,000,000 and as a national reference center, we received different complicated hematological cases, thus in some cases, we can attend the total population of our Institute: 53,000,000 of people. From 1975 to 2008, we diagnosed and treated 15,750 cases of HM, and 143 cases were complicated with a pregnancy (< 0.1%). During the study period, 58 cases of HM diagnosed pregnant (< 12 weeks), and thus they were included in our study. Eighty mothers were diagnosed during second or third trimester of pregnancy and were treated with curative intent, some of these have been previously reported.16 Three patients (two HL and one malignant lymphoma) prefer abortion and received standard chemotherapy, two of these group of patients die secondary to tumor progression (12 and 14 months after pregnancy, and one patient (HL), relapse at 15 months and was treated with stem cell transplant, she is well at 8 years after pregnancy). Thus, two of the three patients who prefer abort died. Two patients, both with AL showed massive infiltration with multiple organ failure and die (4 and 5 days after diagnosis) and did not receive any treatment, thus, they were excluded in our study, and for these reasons, we diagnosed and treated 143 cases of HM during pregnancy.
The criteria for beginning chemotherapy during the first trimester were clinically advanced disease with organ failure secondary to infiltration for neoplasm and risk of death secondary to neoplasm itself (40 cases), patient and family refused abortion procedure (ten cases) and excessive risk of death or complication if abortion was performed (eight cases). All cases, including family, legal and religious counselors, were carefully informed about the risks to the mother and fetus if chemotherapy was administered and the considerations of delay treatment.
Pregnancy was carefully evaluated by obstetricians in an obstetric clinic for high-risk pregnant. Chemotherapy was administered according to conventional doses and schedules. Additional drugs were antiemetics, antibiotics and in some cases, granulocyte colony factor to treat severe granulocytopenia or infection. If possible, chemotherapy was stopped 3 or 4 weeks before delivery to avoid the presence of severe granulocytopenia in the newborn. Delivery was performed according to obstetrics conditions.
At birth, the newborn was carefully examined by pediatric team to detect congenital abnormalities. Height and weight were considered. Complete blood count and serum chemistry were performed. Subsequently, children were carefully evaluated at the age of 3, 6, 12, 18 and 24 months, followed by annual interval until now (December 2010). At each visit, biometric data were obtained and compared to normal standards of clinical development in Mexican children. Laboratory tests were complete blood count and serum chemistry. Findings of any disease and treatment were carefully recorded. The teachers completed questionnaire about attendance, learning and social development. Intelligence Wechsler test were performed at 5, 10 and 15 years of age. Cardiac evaluation was performed with an echocardiographic examination at 3, 5, 7, 10, 15 and 20 years, as previously reported.18 Chromosome studies were performed at 12 months and were repeated if any abnormality was observed. Some of these children were previously reported.17, 19–21
The mothers were evaluated with respect to response type, duration of progression-free disease and overall survival and were matched with patients of the same age and diagnosis to compare the response type. In some cases, when their children became adult and parenthood was possible, the children, termed second generations were observed and when possible previous authorization of the fathers, several studies were performed.
Three patients received chemotherapy including monoclonal antibody, that were well tolerated and without any toxic effect in newborn, similar to other reports.22, 23
The study was approved by the Ethical and Scientific Committee of our Institution (HG 1975-32).
From 1973 to 2009, all women with AL (14 patients), HL (25 patients) and NHL (19), who received chemotherapy during the first trimester of pregnancy were included in NHL and early stage HL, because their treatment could be delayed without risk to the mother. Table 1 shows the main clinical characteristics of these patients. Most of the patients had HL and NHL. Different schedules of treatment were used, according to the time when the treatment was used. In all cases, the chemotherapy regimens used were in an intent to cure. No methotrexate was used in any patient, and two patients with AL received idaurubucin. Furthermore, no radiotherapy was administered in these patients, and surgery was not necessary.
Table 1. Mothers clinical characteristics
The complete response (CR) was higher in all the types of HM; also, improvement in progression free-survival and overall survival was observed when we compare our results with those observed in normal patients treated in our Institution. Toxicity was common to the administered regimen. Hematological toxicity, namely granulocytopenia and thrombocytopenia Grades 3 and 4, was common with the use of aggressive chemotherapy used in AL and Burkitt's-like lymphoma. Twenty-five episodes of infection related to granulocytopenia (septicemia in 12 cases, neumonia in ten patients and urinary infection in three patients) were observed and were treated with broad-spectrum antibiotics. Delivery was planned according to obstetrical indications. Chemotherapy, when possible was stopped at least 3 weeks before delivery to avoid the possibility of severe hematological toxicity in newborns. Thus, myelosuppression, secondary to chemotherapy in newborns were minimum, only three children developed mild granulocytopenia (Grade 2), infection-related granulocytopenia were not observed and hematological recovery was noted between 3 and 8 days. Pathological studies were performed in 43 placentas, and no evidence of infiltration was observed. Median survival was 20.4 years for AL (range 4.5–30.6 years), 21.4 years for malignant lymphoma (range 3.8–28.9 years) and 23.0 years for HL (range 6.0–32 years). The median follow-up was 22.4 years (range 3.8–32.0 years). They were compared to nonpregnant patients with similar diagnosis, age and that were treated with similar chemotherapy regimens. Median survival for AL was 11.4 years (range 3.8–16.4 years), for malignant lymphoma was 14.5 years (range 3.7–21.0 years) and for HL 17.8 years (median 5.1–21.5 years).
Table 2 shows the newborn characteristics. Four fetuses were lost, including two miscarriage at 19 and 20 weeks (early miscarriages) and two stillborns at 5 and 6 months. Thus, 54 newborns were examined. No differences were observed with respect to sex. Four newborn were premature (< 32 weeks) and needed attention in neonatal intensive unit. Ten children had low weight at birth, but they recovered the normal weight within the next 10 weeks. One mother developed infection during fetal period, and the newborn was low weight. Furthermore, no congenital abnormalities were observed.
Table 2. Newborn clinical characteristics
Physical, hematological, neurological, psychological and cardiac studies were normal. Physical development was according to the standards of the Mexican population. No chromosomal abnormalities were observed. No evidence of AL or other neoplasm has been observed until date. Academic development was according to age, as well as economic and social status. Intelligence test, including verbal and performance IQ, was within normal ranges as compared to the control group of 102 children of the same age and socio-economic status. Twenty-four children achieved academic degree. Echocardiography was within normal ranges, with cardiac dimensions, ejection fraction and fractional shortening. We did not perform fetal echocardiograms. Until now, these children have been parents in 12 occasions, all the children of this second generation were normal, unfortunately most of the parents did not accept to perform any study.
Table 3 shows the analysis of the mothers with HM and pregnancy that were recollected from case reports, some series and reviews.6–13 It is important to mention that most reports only included the moment of pregnant, without follow-up of mother and newborn, also they did not mention laboratory studies, cardiac and neurological evaluation and presence of late complications. Most of the cases were HL. A total of 216 patients received chemotherapy, and CR was achieved in 177 patients (82%). At the time of reporting, 157 (69%) patients were alive and free of disease, but longer follow-up was not recorded.
Table 3. Cases diagnosed and treated during first trimester. Literature review
Table 4 shows the data of the fetus and newborn. A total of 163 fetuses were lost, most of them secondary to induced abortion: 119 cases (73%). One hundred and ninety-six newborn were available for evaluation, 86 (48%) were premature (< 32 weeks), 87 (40%) newborn had low weight and 25 (13%) were reported with different congenital abnormalities. According to the dates reported, nine of the abnormalities were lethal, some corrected by surgery and some were minimal that did require follow-up.
Table 4. Children who received chemotherapy during first trimester. Literature review
Treatment of cancer during pregnancy remain as a challenge, because the common treatment include chemotherapy that is been considered to produce detrimental effect on a fetus during pregnancy, including malformations, mutations, carcinogenesis, intrauterine growth restriction, mental retardation and low birth weight. In general, it is considered that chemotherapy could be safely administered during the second and third trimester in patients suffering from HM; however, it is recommended that certain drugs should always be avoided, such as folate antagonists, idaurubicine and concomitant radiotherapy.1–17, 19–22
The use of chemotherapy during the first trimester is considered dangerous to the fetus, and abortion is recommended; however, the affirmation has been based on animal experiment, and the dose to produce fetal malformations is higher than that used for therapeutic use in humans. On the other hand, the best available data document the immediate postpartum period, and delayed effects are noted only in case reports and limited retrospective studies. Today, it is very difficult to define the real impact of cytotoxic agents in fetus, because chemical interactions in humans are a point of current concern, and in HM, chemotherapy combinations of different drugs targeting different cell sites of action are used.23 Moreover, there exist great variations between the metabolism of drugs in mother and fetus and is very difficult to predict the impact of the same drug in different pregnancies.24 In a mother with AL, who was treated during two different pregnancies using cytosine arabinoside and anthracyclines; Schafer25 demonstrated that one child was normal, and the other showed many congenital abnormalities that were not lethal.
In our study, we have reported the clinical outcome of 58 newborns, who were exposed to chemotherapy during the first trimester of pregnancy. No congenital abnormalities were observed; only four newborns were considered premature, the mean weight at birth was between normal range with ten low-weight newborns. Neonatal period was normal, and all recovered weight after 3 months of birth. Physical and neurological development was normal, including intelligence, academic attendance and behavior. It has been established that anthracyclines can cross the placenta and can be found in fetal tissue,26 but the children examined did not show any alterations in cardiac function measured by echocardiogram.18 On reviewing the literature, we found that when mothers received adequate chemotherapy for HM, the clinical outcome was better, with higher number of CR and > 75% of the mothers being alive and free of disease, in contrast to those who were not treated, among whom higher number of deaths secondary to tumor progression was observed. Furthermore, the outcome among children was better, with clinical and neurodevelopment outcome similar to those who did not receive chemotherapy.27, 28 Some studies have shown that neonatal development was not affected when mothers with cancer and pregnancy were treated with chemotherapy and that congenital abnormalities were minimum.29 The recent data of International Registries for pregnant mothers and breast cancer confirmed these findings and demonstrate that mothers can be treated with aggressive chemotherapy during the second and third trimester, because congenital abnormalities were found to be similar to general population, and children exposed to chemotherapy in utero did not develop acute or late side effects secondary to the use of chemotherapy.12, 30
We agree that our experiment is unique, but we have a high number of patients with this special condition and longer follow-up and with clinical examination to evaluate the outcome until the adulthood of these children is required. We have proved that it is possible to treat mothers with HM during the first trimester, owing to the possibility of cure for mother and children with longer survival and normal clinical, neurological, cardiac and psychological development. However, death of the mother and fetus has been noted to be common when chemotherapy is not administered.28 Thus, we considered that most of the reports have not performed an adequate follow-up of the mother and newborn, and hence the consequence of chemotherapy has not been defined in most of the cases. Taking this fact into consideration as well our previous experience, we continued treating mothers with HM and pregnancy, during the first trimester and believe that the use of chemotherapy must be reconsidered in this very special clinical condition, because although the risk of congenital abnormalities increases when HM is treated during the first trimester, patients diagnose will be treated with conventional chemotherapy because termination of pregnancy could not be unsafe and prognosis is better for both mother and fetus and some instances pregnant milieu seem to improve response rate and outcome.31
The work was performed with the owner resources of the Mexican Institute of Social Security.