As part of its coverage of the ASH meeting this year, cancernetwork.com spoke with Andrew Evens. Dr. Evens presented an oral abstract at ASH entitled “Lymphoma in Pregnancy: Excellent Fetal Outcomes and Maternal Survival in a Large Multicenter Analysis.” We discussed his research with him, and the issue of fertility preservation across the cancers.
Andrew Evens, DO, MSc, is the deputy director for clinical and translational research, and medical director of the Clinical Research Office at the University of Massachusetts Memorial Health Care Cancer Center of Excellence. He is also associate professor of medicine and leads the lymphoma program at the University of Massachusetts Medical School.
—Interviewed by Rachel Warren
CancerNetwork: Can you start by telling our audience a little about the study?
Dr. Evens: Yes. It was a retrospective study, as this is somewhat of a rare entity—not that lymphoma is rare, it's approximately the sixth most common cancer, but its actual occurrance during a pregnancy is relatively uncommon. Cancer of any type occurs in approximately 1 in every 1,000 gestations, or pregnancies, but of course there are many thousands of pregnancies that occur every year in the US. This means that, each year in the US, approximately 3,000 to 3,500 pregnant women will have some form of a cancer diagnosed. The most commonly diagnosed cancer is breast, but the second most common are hematologic malignancies, of which lymphoma is the most common.
When you distill it down to just lymphoma, there are probably a couple hundred cases diagnosed during pregnancy in a year. I'm a lymphoma specialist at U Mass; we're a relatively tight-knit community across the US and we all deal with this. We might have a handful of these in total—a couple of cases a year. However, when it's a rare entity, there's not as much published data in the literature to guide even us specialists, much less for physicians out in the community when they see these cases. For these reasons, there were 10 different centers that came together at the same time and said, “Can we put together our collective experiences and try to help guide, not only ourselves, but treating oncologists across the world?”
CancerNetwork: Can you tell us a little about the methods and the results of the study?
Dr. Evens: Again, it was a retrospective analysis but we had a definite design to the study: we collected different information that we wanted to identify and we all worked from the same template of questions to be asked. The study only included cases where a lymphoma occurred during the pregnancy—but at any time during the pregnancy. Then we wanted to determine, what type of lymphoma was diagnosed, how the patient did, and were they treated or were they not. The study included multiple centers: Stanford University, Miami, Fred Hutchinson, University of Nebraska, Roswell Park, Northwestern University, Mayo Clinic, and Washington University—and lymphoma leaders from each. Across all of these institutions, we were able to come up with 90 total cases, which is pretty remarkable because if you look at the data that're published right now—cases of actual lymphoma during a pregnancy—the largest was Hodgkin's disease (HL), with 10 cases, and the number is about the same for non-Hodgkin's lymphoma (NHL), so we're really pleased in a way that we were able to come up with that many cases.
It was mainly a descriptive analysis, describing the different types of lymphoma, but by extension we also looked at outcomes. Everyone was IRB [Institutional Review Board]–approved, throughout the institutions. Initially we put all our data together and we had 90 cases—but we had to exclude 8 because either there was insufficient information for them or the actual diagnosis did not fit our pre-defined definition for them. Obviously the diagnosis had to occur during a pregnancy, and a couple of these had occurred after or before. The final data set was 82 cases. The median age of patients was 31, but the range of patients was 18 through 40. We then looked at the median time of diagnosis, in terms of weeks of gestation—gestation is usually 40 weeks—and diagnosis occurred at a median of 24 weeks, and the range was anywhere from 5 weeks, up until the 40th week of pregnancy.
The majority, 46%, were in the second trimester of pregnancy, with the next most common time of diagnosis in the third trimester. In this pool of patients, there were 43 cases of NHL and 39 cases of HL. Then we distilled it down even further, because within NHL there are many different subtypes, so of those 43 cases of NHL the majority were the more aggressive lymphomas—most notably, diffuse large B-cell lymphoma—but there were some T-cell lymphomas, which are also aggressive. For HL, there aren't as many subtypes, but the most common subtype was nodular sclerosis.
Obviously this was a retrospective report, so therapy had happened in the past—everything was at the discretion of the treating physicians at the time. There were questions like, should everyone have a bone marrow biopsy, what kind of scanning should we do—because certainly you have to be more sensitive to what kinds of radiation you use, and even the CT scanning is controversial. The most common imaging modality in these patients was an MRI, because that doesn't have classic radiation used with it; however, there are still some concerns, in particular with the contrast. We found that two-thirds of the NHL patients had advanced-stage, and half of the HL patients had advanced-stage—the rest had early-stage disease.
One other interesting thing that we found: obviously when women are pregnant they should gain weight, but in the study we found that the median weight gain was only 3%. The common presentation of lymphoma is weight loss (weight loss is classified as a B symptom), so obviously the weight gain in this group was relatively low compared to that of the healthy population. They didn't lose weight, but they didn't gain much, and that was an interesting finding.
Now I'll move on to the treatment of the patients. There were a few cases that occurred very early in the pregnancy, during embyrogenesis, when all the organs of the human body are being formed. I think most people would agree that it's pretty tricky to give chemotherapy during the first trimester, and there would be a risk of malformations and other problems. In total, there were five patients in the first trimester, and one patient very early on, who had termination of the pregnancy because they had aggressive lymphomas and needed to start immediate chemotherapy. That left us with 76 patients, and of course treatment was dependent on what was actually going on at the time, but therapy was actually deferred in 28 of the patients who were diagnosed at a median of 34 weeks. For these patients (who were asymptomatic) the therapy was deferred, then they delivered and then started therapy. But 63% of patients (48 patients) did start chemotherapy during pregnancy. When we look at these 48 patients, they were diagnosed at a median of 25 weeks, so understandably the treating physicians said “we cannot wait 2 months, the lymphoma could progress, it could be fatal, so we have to start treatment.”
There are data in the literature (albeit, as I mentioned, really only in case reports) that it is relatively safe for certain chemotherapies to be given—ones that are not antimetabolites, like methotrexate. Therapy was given in these 48 patients and it was really the classic, standard therapy. In Hodgkin's disease, it was often a common regimen known as ABVD [doxorubicin, bleomycin, vinblastine, dacarbazine], and for diffuse large B-cell, it was rituximab [Rituxan] with CHOP [cyclophosphamide, doxorubicin, vincristine, prednisone].
Our hypothesis was that chemotherapy would be tolerable for the patient and that fetal outcomes would be good—and it was tolerated really well: we saw no differences in toxicity or side effects in the women themselves. I should say that every single case was co-managed with high-risk maternal fetal medicine, and the goal in working together, even in cases where chemotherapy was started early, was to carry the baby to term, to try to get it beyond 36 weeks because you want lung maturation to be able to take place. In 85% of cases, the pregnancy did last beyond 36 weeks. We also analyzed any perinatal complications, and we saw a relatively low risk of pre-eclampsia and other complications, relative to pregnancies that were not associated with lymphoma.
When we looked at birth weights of the infants, from the cases that received chemotherapy vs cases that did not receive chemotherapy, they were the same, and in all the cases, there were no malformations that were detected.
Finally, what we saw was that the outcomes were pretty good. The outcomes for diffuse large B-cell and Hodgkin's disease were similar to those that might occur if the patients were not pregnant. We divided it between B-cell lymphoma, in which the 3-year overall survival was over 80%, while in HL the 3-year overall survival was over 90%. For T-cell lymphoma, which is a more aggressive entity, it was a little bit lower, but the main point is that these outcomes appeared pretty consistent with non–pregnancy-associated lymphoma outcomes.
CancerNetwork: Thanks. You actually answered many of my questions within your description of the results.
Dr. Evens: That was my goal—to give a long-winded answer so there would be very few follow-up questions!