A and B, Progression-free survival and overall survival for all 39 patients according to disease histology. C, Progression-free survival and overall survival for all 36 patients that did not terminate pregnancy at the time of diagnosis according to receipt of antenatal therapy. HL indicates Hodgkin lymphoma; NHL, non-Hodgkin lymphoma.
eTable 1. Antenatal Therapy
Customize your JAMA Network experience by selecting one or more topics from the list below.
Pinnix CC, Osborne EM, Chihara D, et al. Maternal and Fetal Outcomes After Therapy for Hodgkin or Non-Hodgkin Lymphoma Diagnosed During Pregnancy. JAMA Oncol. 2016;2(8):1065–1069. doi:10.1001/jamaoncol.2016.1396
Copyright 2016 American Medical Association. All Rights Reserved. Applicable FARS/DFARS Restrictions Apply to Government Use.
The management of lymphoma diagnosed during pregnancy is controversial and has been guided largely by findings from case reports and small series.
To determine maternal and fetal outcomes of women diagnosed with Hodgkin lymphoma (HL) or non-Hodgkin lymphoma (NHL) during pregnancy.
Design, Setting, and Participants
This retrospective analysis studied a cohort of 39 pregnant women diagnosed with HL and NHL (31 HL and 8 NHL) at a single specialized cancer institution between January 1991 and December 2014.
Main Outcomes and Measures
We examined data on disease and treatment characteristics, as well as maternal and fetal complications and outcomes. The Kaplan-Meier method was used to compare progression free survival (PFS) and overall survival (OS) according to receipt of antenatal therapy and other clinical factors. Univariate and multivariate analyses were performed by using Cox proportional hazard regression models to identify potential associations between clinical and treatment factors and survival.
The median (range) age of the 39 women in the patient cohort was 28 (19-38) years; 32 women (82%) had stage I or II disease at diagnosis, and 13 had bulky disease. Three women electively terminated the pregnancy to allow immediate systemic therapy; of the remaining 36 women, 24 received antenatal therapy (doxorubicin based combination chemotherapy in 20 of 24 patients), and 12 deferred therapy until after delivery. Four women experienced miscarriage, all of whom had received antenatal systemic therapy and 2 during the first trimester. Delivery occurred at a median (range) of 37 (32-42) weeks and was no different based on receipt of antenatal (median [range], 37 [33-42] weeks) vs postnatal (median [range], 37 [32-42] weeks) therapy (P = .21). No gross fetal malformations or anomalies were detected. At a median (range) follow-up time of 67.9 (8.8-277.5) months since the diagnosis of lymphoma, 5-year rates of PFS and OS were 74.7% and 82.4%, respectively; these rates did not differ according to timing of therapy. On univariate analysis, bulky disease (>10 cm), extranodal nonbone marrow involvement, and poor performance status (Eastern Cooperative Oncology Group score, ≥2) predicted increased risk of disease progression. On multivariate analysis, extranodal nonbone marrow disease and performance status remained significant for both PFS and OS.
Conclusions and Relevance
Systemic therapy given for lymphoma after the first trimester of pregnancy is likely safe and results in acceptable maternal and fetal outcomes.
Diagnoses of cancer occur in roughly 1 of 1000 pregnancies.1 Emerging reports suggest that standard chemotherapy administered during the second and third trimesters results in acceptable maternal and fetal outcomes.2,3 We report maternal and fetal outcomes from a single-institution experience of women diagnosed with lymphoma during pregnancy.
Question What are the maternal and fetal outcomes of women diagnosed with lymphoma during pregnancy?
Findings In an analysis of 39 women diagnosed with lymphoma during pregnancy, delivery occurred at a median of 37 weeks and was no different based on administration of antenatal therapy. Five-year progression-free survival and overall survival rates were 74.7% and 82.4% and did not differ according to the timing of therapy.
Meaning Systemic therapy after the first trimester is likely safe and results in acceptable maternal and fetal outcomes.
We identified 47 women diagnosed with HL or NHL while pregnant between 1991 and 2014 after institutional review board approval. A waiver of informed consent was granted due to the retrospective nature of the study. Eight women were excluded for lack of follow-up data or pregnancy during lymphoma treatment; the final cohort included 31 women with HL and 8 women with NHL.
All cases were comanaged with a maternal fetal medicine (MFM) obstetrician. Staging studies typically included magnetic resonance imaging (MRI) and ultrasonography. Computed tomography (CT) and 18-fluorodeoxyglucose positron emission tomography (PET)–CT was generally avoided until the postpartum period. Response assessment was based on the Lugano Classification.4 Progression-free survival (PFS) and overall survival (OS) were defined from the time of lymphoma diagnosis and estimated using the Kaplan-Meier method. Clinical factors were compared using 2-sample t tests, Fisher exact tests, and Cochran-Mantel-Haenszel tests. Multivariate Cox proportional hazard regression and stepwise variable selection was applied to assess interactions between variables and survival. A significance level of 0.05 was used in statistical tests.
Patient and gestational characteristics for the 39 patients are in Table 1. Three women underwent elective termination at diagnosis. Of 36 remaining patients, 24 (61%) began antenatal therapy (AT) and 12 (31%) postponed therapy until delivery (Table 2) (eTable 1 in the Supplement). Four patients received supradiaphragmatic radiation therapy (RT), (median [range] dose, 40.4 [36-50] Gy [to convert milligray to rad, multiply by 0.1]).
Delivery did not differ based on receipt of AT (Table 2). Among the 36 women who did not undergo termination at diagnosis, receipt of AT was not associated with increased rates of preterm delivery. Among the 24 women who received treatment during pregnancy, 7 (29%) experienced preterm delivery compared with 5 of 12 women (42%) who postponed treatment until after delivery (P = .73).
Miscarriage occurred in 4 patients that required immediate therapy during the first (n = 2) and second (n = 2) trimesters. Neonatal outcomes were available for 31 of 32 patients who did not undergo immediate elective termination or experience miscarriage. The infants of these 31 women showed no known anomalies at birth.
Among the 24 patients who initiated treatment during pregnancy, the overall response rate (ORR) was 91.7%, and the complete response (CR) rate was 75.0%. Of the 12 patients who deferred therapy until delivery, both the ORR and the CR rate were 91.7%.
At a median (range) follow-up of 67.9 (8.8-277.5) months, the 5-year PFS and OS rates were 74.7% and 82.4% for the entire cohort (Figure, A and B). For the 31 women with HL, the 5-year PFS and OS rates were 69.9% and 80%, respectively; for the 8 women with NHL, the 5-year PFS and OS rates were 85.7% and 83.3%, respectively. Among the 36 women who did not terminate their pregnancy at diagnosis, no difference was found in PFS or OS based on receipt of AT (P = .84) (Figure, C and D).
For the 36 patients who did not electively terminate the pregnancy at the time of diagnosis, on univariate analysis bulky disease (PFS hazard ratio [HR], 3.6; 95% [CI], .96-13.54; P = .06), extranodal nonbone marrow involvement (HR, 4.2; 95% CI, 1.04-16.99; P = .04), and poor Eastern Cooperative Oncology Group performance status (PS) (HR, 3.9; 95% CI, 1.5-10.14); P = .005) were associated with increased risk of progression; poor PS was associated with OS (HR, 8.88; 95% CI, 2.04-38.7; P = .004). On multivariate analysis, the final Cox proportional hazard model revealed significant associations for extranodal nonbone marrow involvement (OS HR, 73.5; 95% CI, 2.32-2332.86; P = .02; and PFS HR, 8.26; 95% CI, 1.67-40.78; P = .01) and PS (OS HR, 26.7; 95% CI, 3.10-229.53; P = .003; and PFS HR, 4.89; 95% CI, 1.80-13.23; P = .002).
The management of lymphoma diagnosed during pregnancy aims to treat the malignancy adequately without harming the fetus. In the current study, 24 women (roughly two-thirds) received therapy during pregnancy. Maternal outcomes were not affected by the delivery of AT, although the 4 women who had miscarriages received chemotherapy during pregnancy (4 of 24 patients who received AT therapy vs 0 of 12 patients who received postnatal therapy; P = 0.30).
The miscarriage rate in our series was roughly 10% (n = 4), higher than previous studies reporting no spontaneous abortions.5-7 Multi-agent chemotherapy has been given rarely during the first trimester because of the vulnerability of the fetus during organogenesis. This concern is justified because both patients who received first-trimester therapy in this study experienced fetal death. Antenatal therapy is thought to be safer for the fetus during the second trimester.2,3,5-7 Two of the 4 miscarriages occurred in women who received systemic therapy during the second trimester; however, 1 of those patients was critically ill, so other factors likely contributed to the fetal death. The other patient experienced the loss of 23-week twin fetuses conceived via in vitro fertilization, after ABVD (doxorubicin, bleomycin, vinblastine, dacarbazine) therapy was initiated during gestational week 15. We suspect that maternal factors contributed to the adverse fetal outcomes in at least 1 of these 2 cases. In keeping with previous reports, our practice is to consider AT during the second and third trimesters if deferring that therapy until after delivery would be expected to compromise maternal outcome.3,8,9
Radiation therapy was given during pregnancy to four patients. In contrast to the multicenter study by Evens et al,2 in which use of RT during pregnancy was associated with inferior maternal outcomes, we did not detect any adverse effects of RTon PFS (P = .23) or OS (P = .24). However, we do not advocate the routine use of RT during pregnancy. In cases where RT is deemed to be critical, appropriate precautions for the fetus are essential.
Maternal outcomes were acceptable, with a PFS rate of 75% and an OS rate of 82% at 5 years. Known adverse risk factors among nonpregnant patients with lymphoma, such as poor PS and bulky disease, were also associated with inferior outcomes among the 39 patients in this series. Among patients with HL, the 5-year PFS and OS rates were 70% and 80%, respectively. This finding contrasts with the 3-year OS rate of 97% for 40 patients with HL in the report from Evens and colleagues.2 This difference may reflect the higher proportion of patients in our series with adverse prognostic factors. In the study by Evens et al, only 5% of patients with HL patients had bulky disease compared with 35% in our series. Also, roughly one-fourth of patients in our series had a PS score of 2, whereas no patients had a score greater than 1 in the multicenter study.2 Because we found no differences in PFS or OS according to receipt of AT, we contend that disease factors, rather than treatment-related factors, influenced the worse maternal outcomes among patients with HL.
Although the follow-up time is relatively short, no malformations have been observed. Rates of preterm labor were similar among patients who received AT and those who deferred treatment. Preterm delivery, and not fetal chemotherapy exposure, has been shown to be the most strongly predictive of neurocognitive impairment.3,10,11 Thus, an important goal should be prolonging the pregnancy so as to avoid preterm birth.
The diagnosis of lymphoma during pregnancy poses management challenges requiring consideration of the fetus without compromising potentially curative therapy for the mother. In this series, the decision to defer therapy until after delivery did not affect maternal outcomes. Our findings, coupled with those from previous studies, lead us to recommend delaying therapy to the second trimester if this can be accomplished without perceived detriment to the patient.
Corresponding author: Michelle A. Fanale, MD, Department of Lymphoma/Myeloma, Unit 429, The University of Texas MD Anderson Cancer Center, 1515 Holcombe Blvd, Houston, TX 77030 (firstname.lastname@example.org).
Accepted for Publication: April 1, 2016.
Published Online: May 26, 2016. doi:10.1001/jamaoncol.2016.1396.
Author Contributions: Dr Pinnix had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Drs Pinnix and Osborne contributed equally to this article.
Study concept and design: Pinnix, Osborne, Chihara, Dabaja, Wogan, Ballas, Fanale.
Acquisition, analysis, or interpretation of data: Pinnix, Osborne, Lai, Zhou, Ramirez, Oki, Hagemeister, Rodriguez, Samaniego, Fowler, Romaguera, Turturro, Fayad, Westin, Nastoupil, Neelapu, Cheah, Dabaja, Milgrom, Smith, Horace, Milbourne, Fanale.
Drafting of the manuscript: Pinnix, Cheah, Dabaja, Fanale.
Critical revision of the manuscript for important intellectual content: Pinnix, Osborne, Chihara, Lai, Zhou, Ramirez, Oki, Hagemeister, Rodriguez, Samaniego, Fowler, Romaguera, Turturro, Fayad, Westin, Nastoupil, Neelapu, Cheah, Dabaja, Milgrom, Smith, Horace, Milbourne, Wogan, Ballas, Fanale.
Statistical analysis: Pinnix, Osborne, Zhou, Cheah.
Administrative, technical, or material support: Pinnix, Chihara, Oki, Nastoupil, Neelapu, Dabaja, Horace, Fanale.
Study supervision: Pinnix, Samaniego, Fowler, Fayad, Dabaja, Milbourne, Fanale.
Providing care and outcome data: Ramirez.
Conflict of Interest Disclosures: None reported.