This article is more than 5 years old. Information may no longer be current.
Pregnant patients with cancer face delayed protocols, difficult choices
Click Here to Manage Email Alerts
A diagnosis of cancer in a pregnant woman is uncommon, as few as one in 3,000 pregnancies according to a 2006 estimate published in the journal Cancer. A 2010 study by Van Calsteren and colleagues in the Journal of Clinical Oncology estimated that number at one in 1,000.
However, Douglas Yee, MD, HemOnc Today breast cancer section editor and director of the Masonic Cancer Center, University of Minnesota, said, at some point, most oncologists who treat women will have to treat a patient with cancer who is pregnant. Experts interviewed said the experience of treating these types of patients is similar in many ways based on the common characteristics of pregnancy, but is also quite individual depending on the unique characteristics of the various cancers with which the patients present.
In a 2009 study published by the Journal of Clinical Oncology, Stensheim and colleagues found that breast cancer was the most common diagnosis among all women (n=13,106), but that malignant melanoma was the most common malignancy both among pregnant women (n=160) and women who were lactating (n=126). Cervical cancer was the second most common malignancy in both groups. The highest incidences of cause-specific death were observed in lactating women diagnosed with breast (67%) or ovarian (47%) cancers, and among women diagnosed with lymphoma or leukemia during pregnancy (48%).
|
Photo courtesy of Jonathan Kolbe Photography |
But treating these patients can be complicated, despite scientific consensus that, in most cases, pregnant women can tolerate aggressive chemotherapy after the first trimester. A 2001 report published in Clinical Lymphoma by Agustin Avilés, MD, and Natividad Neri, MD, was among one of the first studies to establish that pregnant women and their unborn children could safely tolerate aggressive chemotherapy without adverse effects to offspring. That study presented results on 84 children born to mothers treated with chemotherapy for hematological malignancies. At median follow-up of 18.7 years, none of the children displayed any signs of cancer, or acute or late abnormalities.
Results from a 2005 18-year observation of women in London published in the Journal of Clinical Oncology also showed that chemotherapy could be safely administered in the second and third trimesters according to an evaluation of peripartum complications and immediate fetal outcomes.
Still, experts interviewed by HemOnc Today reported a hesitancy among patients to seek treatment out of fear of harming their unborn children as well as the mistaken belief that if they were found to have cancer, that no doctor would treat them.
“I tell my patients that the most important question they need to ask their oncologists is, ‘How would you treat me if I wasn’t pregnant?’” said Elyce H. Cardonick, MD, associate professor of obstetrics and gynecology with a special interest in cancer and pregnancy at Cooper University Hospital in Camden, N.J. “The saddest cases are the ones where women are afraid to tell their physician about a lump because they are afraid they’ll have to terminate the pregnancy. Then they tell the physician about that lump after giving birth, but the disease has already spread and they die before the baby is even a year old. We want to get these women treated so they’re there for their families.”
Timing of treatment
Yee said for most women of child-bearing age, even pregnant patients with very low-risk tumors, most oncologists would advocate some kind of systemic therapy. “In breast cancer, the question is, what kind of systemic therapy? Certainly, hormonal therapy is contraindicated for any stage of pregnancy. Many drugs, like doxorubicin [Adriamycin, Pharmacia and Upjohn], cytoxan and paclitaxel, drugs we commonly use have been reported to be safe, but [with pregnant patients] there is always a question of the timing of surgery.”
All cancer treatment starts with a conversation between patient and physician about treatment goals and potential outcomes. Yee concurs with Cardonick that the discussion is not markedly different from that of a young woman who is not pregnant.
“The difference is: How do we manage this around the delivery? That part gets trickier; depending on when they present with pregnancy, the options are slightly different,” he said.
Yee cited an example of a recent patient who was diagnosed with breast cancer at 30 weeks of pregnancy. She delayed treatment until her obstetrician said the baby’s lungs had matured; labor was induced and she began treatment. However, that was not a viable strategy in another patient he recently treated, who was diagnosed at week 12. Similarly, waiting to treat is not always an option in fast-growing cancers, and not all patients are willing to terminate a pregnancy to begin treatment. Furthermore, in both cases the patients had HER2 amplified tumors. Trastuzumab has a black box FDA warning during pregnancy, yet is essential to improving outcomes for these patients.
Christine M. Lee, MD, a gynecologic oncologist at Texas Oncology–The Woodlands, said she tries to avoid termination if at all possible. Cancer surgery during pregnancy is safest between 16 weeks and 20 weeks of gestation, but a patient who needs surgery for a gynecologic malignancy during pregnancy is at high risk for pre-term labor and premature delivery, she said. A patient requiring radical surgery should be referred to a specialist because few OB/GYNs are trained for extensive lymph node dissections or radical surgery.
“Any time the internal gynecologic organs are manipulated in pregnancy, one risks premature delivery and rarely, fetal demise,” she said. “That makes operating in this setting very complex.”
Choices of treatment
Pregnancy can also affect the manner in which certain cancers are typically treated, leading to alternative courses of care. Jennifer Rubatt, MD, a gynecologic oncologist with Cancer Centers of North Carolina, treated a patient with a large tumor on her cervix who declined to terminate her pregnancy. Instead, physicians administered chemotherapy to shrink the tumor until the pregnancy was far enough along that labor could be induced.
“If she hadn’t been pregnant, we would have treated her with radiation,” Rubatt said. Instead, “we did something of a combined procedure where we delivered the baby, then conducted the appropriate surgery for surgical cancer.”
|
|
Even more so than most patients, pregnant women with cancer need multidisciplinary treatment, Rubatt said. The treatment team ideally includes a high-risk obstetrical specialist, a surgeon and an oncologist. The team has to amalgamate the patient’s wishes, discern what physicians know about the disease and determine the steps that need to be taken to ensure a safe pregnancy. Once that team is in place, “then we can have a roundtable discussion about how we can best keep all those things in mind while coming up with an effective treatment plan,” she said.
Promising prognoses
When establishing treatment protocols, medical science ideally evaluates promising new therapies through randomized controlled trials. But the experts who spoke to HemOnc Today said it would be extremely expensive to conduct a prospective trial among pregnant women and, because the population is so small, it would be difficult to build a cohort large enough for the findings to be significant.
Lee said that a prospective case-controlled study might be possible theoretically, but such a study was likely impossible to conduct in practice.
“It would depend on the endpoint. We could ask whether pregnant patients have poorer outcomes because their drug is metabolized through the liver because of pregnancy or does pregnancy effect their overall survival or do they have more side effects?” she said. “To conduct a study like that, you’d have to draw blood from a large number of pregnant women and determine the levels of chemicals in their bloodstreams to see how the drug is being metabolized. Then you’d have to do the same thing with a cohort of patients with the same disease who aren’t pregnant. That’s not feasible.”
Cardonick said it might be possible to conduct a prospective study by comparing two drugs known to be safe to determine which is more effective, but physicians usually only know about outcomes in non-pregnant patients. She pointed to idarubicin, noting that it is in the same class as doxorubicin and epirubicin, but there are reports suggesting a link between idarubicin and fetal cardiac effects and cardiomyopathy.
“Just because drugs are in the same class, we can’t assume it will be safe in pregnancy,” she said. “We do the opposite when considering treatment in pregnant patients; instead of using the newest drug, we look at the drugs that have the longest track record for safety.
“You might be able to randomize patients to lumpectomy vs. mastectomy. If we had two equally efficacious drugs with the same safety protocol, you could probably conduct a randomized study, but not with a new drug,” Cardonick said.
Despite these obstacles, evidence has shown that in many cases survival trends are positive. Results from one population-based cohort study published in the Journal of Clinical Oncology in 2009 demonstrated that a diagnosis of cancer during pregnancy or lactation did not lead to an increase in cause-specific death for most malignancies.
Results published in 2004 by Lens and colleagues showed that survival in pregnant women diagnosed with melanoma was equal to that of nonpregnant women (HR=0.58; 95% CI, 0.32-1.05). Pregnancy was not found to be a significant predictor for survival.
The study involved two analyses of women aged 16 to 49 years who were diagnosed with cancer from 1967 to 2002. The first analysis included 41,464 women who were not pregnant, 516 who were pregnant and 531 who were lactating. The second analysis included a control group of 40,200 women and 2,311 women who had post-pregnancy cancers.
As stated earlier, the highest incidences of cause-specific death was observed in lactating women diagnosed with breast (67%) or ovarian (47%) cancer, and among women diagnosed with lymphoma or leukemia during pregnancy (48%). Researchers noted that 54% of women diagnosed with breast cancer during lactation had advanced disease compared with just 34% of the control group.
|
|
In some women, however, pregnancy appeared to have a protective effect in those patients who had a post-cancer pregnancy. For all disease sites, these women had a lower risk for cause-specific death compared with the control group (HR=0.49; 95% CI, 0.41-0.59). Further, women with a post-cancer pregnancy diagnosed with cervical cancer, lymphoma or leukemia had an 80% lower risk for cause-specific mortality compared with women without a subsequent pregnancy.
The quest for dedicated care
Despite reports of positive outcomes, patients often struggle with finding clinicians who are able to effectively handle the intricacies of their cases. Cardonick said she decided to start the International Cancer and Pregnancy Registry in 1997 when she was a fellow in maternal-fetal medicine and treated a pregnant woman with melanoma. The patient declined to terminate the pregnancy, and Cardonick said she could not find an oncologist willing to perform a sentinel node biopsy. The cancer spread while treatment was delayed, but the patient gave birth to a healthy baby and she and her daughter are both doing well.
A few months later, another pregnant woman presented with Hodgkin’s disease and also declined an abortion. Because delaying treatment for 9 months was not a desirable option, Cardonick, at the patient’s behest, argued successfully to the hospital’s ethics board to be able to treat the patient with chemotherapy.
“I decided to [create a] subspecialty in maternal-fetal medicine because nobody knew what to do with these patients,” she said. “The OB-GYN wanted to deal with the pregnancy and the oncologist wanted to deal with the cancer, but there are some changes in treatment that require someone to coordinate all the moving parts.”
Patients entered into the registry are followed from diagnosis to the birth of their child, and Cardonick follows up each year on the child’s birthday. There are currently 296 women included in the registry, and the registry includes data on all the children born to a registered woman. To date, Cardonick has not recorded any negative health effects for any of the children, the oldest of whom are now about 14 years old.
“They’re developing fine,” she said. “The other goal of the registry is to stay in contact with these pediatricians and families to find out how the babies do when they go to school or go through puberty. We want to make sure that even though there were no negative effects at birth, we don’t see any effects from chemotherapy or cancer years later.”
This trend supports findings found in literature that treatment, when delivered at the right time, appears to have little appreciable effect on children born to mothers with cancer.
Although cohorts are not particularly large, in 2009, the Journal of Clinical Oncology published results from a European study of 215 pregnant women with invasive cancer. Patients suffered five spontaneous miscarriages, 30 women terminated their pregnancies, 122 patients (56.7%) underwent treatment and 58 patients delayed treatment until after delivery. Treated patients underwent chemotherapy alone (n=33); chemotherapy plus surgery (n=25); surgery alone (n=49); and one patient underwent radiation and surgery. Three patients had radiotherapy, three had radiation plus surgery and three had chemoradiation plus surgery.
Breast cancer was the most common malignancy (46%), followed by hematologic malignancies (18%) and dermatologic malignancies (10%).
Mean gestational age at delivery was 36.3 ± 2.9 weeks, and delivery was induced in 71.7% of pregnancies. More than half the infants (54.2%) were premature. Incidence of preterm labor was up 11.8% among children exposed to cytotoxic treatment in the womb. Additionally, that group had a higher proportion of newborns with birth weight below the 10th percentile (24.2%).
Of all newborns, 51.2% were admitted to a neonatal ICU, most (85.2%) because of prematurity. There was no increased incidence of congenital malformations. Researchers said that there was no increased risk for major or minor physical malformation among the children, even for children exposed to radiotherapy in utero.
Van Calsteren and colleagues concluded that most of the neonatal problems observed were the result of preventable prematurity. “The findings of this study show an overall good outcome of pregnancies complicated with cancer,” they wrote.
All of the experts interviewed for this article agree that more studies are needed to study the unique characteristics of these types of patients. “If it is true that each oncologist is only going to see one or two [pregnant] patients in his or her career, then the best thing we can do is collect information from all these oncologists,” Cardonick said. – by Jason Harris
For more information:
- Andrieu N. J Natl Cancer Inst. 2006;98:535-544.
- Avilés A. Clin Lymphoma. 2001;2:173-177.
- Donegan WL. CA Cancer J Clin. 1983;33:194-214.
- Hahn KME. Cancer. 2006;107:1219-1226.
- Lau WY. Cancer. 1995;75:2669-2676.
- Lens MB. J Clin Oncol. 2004;22:4369-4375.
- Morice P. J Clin Oncol. 2009;27:2298.
- Mulvihill JL. Cancer. 1987;60:1143-1150.
- Ring AE. J Clin Oncol. 2005;23:4192-4197.
- Stensheim H. J Clin Oncol. 2008;27:45-51.
- Tabata T. Int J Gynecol Cancer. 2008;18:181-184.
- Turchi JJ. Cancer. 1988;61:435-440.
- Van Calsteren K. J Clin Oncol. 2009;28:683-689.
- Zauzu J. Cancer. 1991;67:703-709.
Disclosures: Drs. Cardonick, Lee, Rubatt and Yee reported no relevant financial disclosures.
Is attempting to preserve fertility safe for pregnant women with cancer?
It largely depends on the type of cancer.
For invasive cervical cancer, the definitive treatment is, in some way, going to render the patient physically unable to have more children because we’re performing a hysterectomy. That doesn’t mean the patient can’t have their ovaries preserved for egg harvesting, so they can be a biological mother with a surrogate.
|
|
In that case, you have to weigh the type of cancer you’re dealing with. Is it a cancer that is highly likely to spread to the ovaries, in which case the ovaries should be removed, or is the risk for metastases in the ovaries so low that you can leave the ovaries in place?
That’s a tougher question for patients with ovarian cancer. Some of the ovarian cancer malignancies, like germ cell malignancies, can afflict very young people, age 16 to 20 years. You would want to do everything possible to preserve fertility in a pregnant patient that age who developed cancer because these tumors are very sensitive to chemotherapy, so a complete hysterectomy is unnecessary. The epithelial ovarian cancers are more delicate because those cancers can spread to the other ovary.
You have to know what type of cancer you’re dealing with, understand where the cancer is likely to spread to and what are the chances for recurrence. That might help you direct the patient in recommendations for more conservative or more aggressive treatment.
S. Diane Yamada, MD, is Professor of Obstetrics/Gynecology and Chief of Gynecologic Oncology at the University of Chicago Medical Center. She reports no relevant financial disclosures.
The patient, with guidance and education from the physician, must balance the desire for more children against receiving less than standard treatment, particularly in the case of breast cancer.
|
|
A 34-year-old pregnant woman is diagnosed in the third trimester with ER-positive stage II breast cancer. She delivers the baby and receives adjuvant chemotherapy and is placed on tamoxifen. She has retained menstrual function and wants to have another child. She doesn’t want to go through 5 years of tamoxifen because she knows that , in general, fertility declines with age and the impact of prior adjuvant chemotherapy on fertility is presently unclear. Her question is: Can we shorten the course of tamoxifen?
The standard duration of tamoxifen is 5 years. We know from randomized trials that 1 year of tamoxifen is inferior to 2 years, and 2 years is inferior to 5 years. In counseling this woman, we have to balance the desire for another child with the underlying risk for distant metastases. For example, take two women with breast cancer: one has a very favorable prognosis with an absolute risk of distant metastases of 10%; the other one has a 50% risk of developing distant metastases. In the former, perhaps truncating the tamoxifen course to less than 5 years, say 2 or 3 years, would be more acceptable, whereas the latter parent may be more inclined to take the standard 5 years of treatment.
A major limitation is that we can’t tell an individual woman whether she’s going to develop distant metastases and die of breast cancer, or die of a non breast cancer-related cause and experience a “personal cure.” We can only estimate the risks of developing distant metastases and the treatment benefits. The Oncotype DX assay gets to a finer detail in terms of predicting risk for distant metastases on the basis of genes expressed in the tumor and who benefits from adjuvant chemotherapy, but when dealing with each woman who has early-stage breast cancer, it is either 100% or 0% they will or will not have distant metastases. Given this uncertainty, a woman who wants to have more children may choose to come off tamoxifen early no matter what the risks of developing a distant metastases.
Charles L. Shapiro, MD, is Director of Breast Medical Oncology at James Cancer Hospital and The Ohio State University Comprehensive Cancer Center. He reports no relevant financial disclosures.