European Journal of Obstetrics & Gynecology and Reproductive Biology 181 (2014) 145–156

Contents lists available at ScienceDirect

European Journal of Obstetrics & Gynecology and Reproductive Biology journal homepage: www.elsevier.com/locate/ejogrb

Review

Feto-maternal outcomes of pregnancy complicated by ovarian malignant germ cell tumor: a systematic review of literature Michiko Kodama a, Brendan H. Grubbs b, Erin A. Blake c, Sigita S. Cahoon d, Ryusuke Murakami e, Tadashi Kimura a, Koji Matsuo d,f,* a

Department of Obstetrics and Gynecology, Osaka University Graduate School of Medicine, Suita, Osaka, Japan Division of Maternal-Fetal Medicine, Department of Obstetrics and Gynecology, Los Angeles County Medical Center, University of Southern California, Los Angeles, CA, USA c Department of Obstetrics and Gynecology, University of Colorado, Denver, CO, USA d Gynecologic Oncology, Department of Obstetrics and Gynecology, Los Angeles County Medical Center, University of Southern California, Los Angeles, CA, USA e Department of Gynecology and Obstetrics, Kyoto University Graduate School of Medicine, Kyoto, Japan f Norris Comprehensive Cancer Center, University of Southern California, Los Angeles, CA, USA b

A R T I C L E I N F O

A B S T R A C T

Article history: Received 13 February 2014 Received in revised form 19 July 2014 Accepted 29 July 2014

Malignant germ cell tumors (MGCT) are a rare type of ovarian cancer with poorly understood behavior during pregnancy. This systematic review evaluated feto-maternal outcomes and management patterns of 102 ovarian MGCT-complicated pregnancies identified in PubMed/MEDLINE. Mean age was 25.8. The most common histology type was dysgerminoma (38.2%) followed by yolk sac tumor (30.4%). Abdomino-pelvic pain (35.3%) was the most common symptom. The majority were stage I disease (76.4%) with a mean tumor size of 17.9 cm. Most cases had live births (77.5%) at term (56.6%). Tumor surgery without fetal conservation took place in 22 (21.6%) cases (Group 1). This group was characterized by the first trimester tumor detection and intervention, non-viable pregnancy, and frequent concurrent hysterectomy. There were 59 (57.8%) cases which underwent expectant management of pregnancy: mean delay 16.4 weeks for 46 (45.1%) cases with tumor surgery and fetal conservation (Group 2); and 7.8 weeks for 13 (12.7%) cases with tumor surgery after delivery (Group 3). The live birth rate in Groups 2 and 3 was 98.3%. There were 21 (20.6%) cases in which the tumor was incidentally found intra/ postpartum (Group 4). Group 2 showed the highest 5-year overall survival rate (92.8%) followed by Group 4 (79.5%), Group 3 (71.4%), and Group 1 (56.2%, p = 0.028). Group 1 had more advanced-stage disease when compared to Group 2 (proportion of stages II–IV disease, 36.4% versus 11.4%, p = 0.023). In multivariate analysis, age 20 (p = 0.032) and stages II–IV (p = 0.02) remained independent prognosticators for decreased overall survival in all cases. Expectant management of pregnancy was not associated with poor survival outcome in multivariate analysis (p = 0.43). In conclusion, our analysis demonstrated that timing of tumor intervention and delivery significantly impacted feto-maternal outcome of ovarian MGCT-complicated pregnancies. It is suggested that early detection and tumor intervention with expectant management of pregnancy is an acceptable option in early-stage ovarian MGCT-complicated pregnancies. ß 2014 Elsevier Ireland Ltd. All rights reserved.

Keywords: Malignant germ cell tumor Ovarian cancer Pregnancy Review

Contents Introduction . . . . . . . . . Materials and methods Source . . . . . . . . Eligibility . . . . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

146 146 146 146

* Corresponding author at: Division of Gynecologic Oncology, Department of Obstetrics and Gynecology, Los Angeles County Medical Center, University of Southern California, 2020 Zonal Avenue, IRD520, Los Angeles, CA 90033, USA. Tel.: +1 323 226 3416; fax: +1 323 226 2743. E-mail address: [email protected] (K. Matsuo). http://dx.doi.org/10.1016/j.ejogrb.2014.07.047 0301-2115/ß 2014 Elsevier Ireland Ltd. All rights reserved.

M. Kodama et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 181 (2014) 145–156

146

Clinical information . . . . . . . . . . . . . . . . Definition . . . . . . . . . . . . . . . . . . . . . . . . Statistical analysis. . . . . . . . . . . . . . . . . . Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Clinical characteristics . . . . . . . . . . . . . . Pregnancy outcomes . . . . . . . . . . . . . . . . Oncologic outcomes . . . . . . . . . . . . . . . . Management pattern and feto-maternal Expectant management of pregnancy . . Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Funding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Disclosure. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

......... ......... ......... ......... ......... ......... ......... outcomes. ......... ......... ......... ......... .........

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

Introduction Malignant germ cell tumor (MGCT) is a relatively uncommon subtype of ovarian cancer accounting for less than 5% of all ovarian cancers [1,2]. When compared to epithelial ovarian cancer, MGCT is characterized by younger age and earlier stage at diagnosis. It also demonstrates an increased risk of tumor rupture due to rapid tumor growth, tumor necrosis, and intratumoral hemorrhage [3– 5]. The overall prognosis of ovarian MGCT has dramatically improved following the introduction of platinum-based combination chemotherapy in its management [3,4,6]. Malignancy has been reported in approximately one in 1000 pregnancies, and the incidence of any type of ovarian cancer comprises 2.8–11 per 100,000 pregnancies [7–10]. Due to its preponderance for reproductive aged women, ovarian MGCT accounts for 18–26% of all ovarian cancers complicating pregnancy [8,9,11]. Pregnancy is characterized by physiologic changes to multiple organ systems, and it is not known whether these changes, will impact the behavior and prognosis for MGCT. It is also not known how the tumor impacts the mother and the fetus. Collectively, pregnancy complicated by ovarian MGCT is an extremely rare clinical condition. Indeed, available previous literature only shows sporadic case reports for this entity, which limits determination of the optimal treatment strategy for the pregnancies complicated by ovarian MGCT. The management of cancer in pregnancy is complicated, as there are three separate but interactive entities, i.e. mother, fetus, and malignancy, which must be managed simultaneously. Optimal management must be determined by a multidisciplinary approach. The aim of the current study was to conduct a systematic review of literature describing feto-maternal outcomes and management patterns of pregnancies complicated by ovarian MGCT.

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

146 147 147 148 148 148 148 150 151 151 154 154 154

description of the cases. Finally, 77 articles including 102 cases were eligible for the study analysis [8,9,15–89]. Eligibility Inclusion criteria were case reports or case series that had an adequate description for (i) clinical demographics, (ii) pregnancy outcomes, (iii) tumor characteristic, (iv) type of intervention and treatment, and (v) maternal survival outcome (any length of follow-up time). References for each evaluated article were also reviewed, and the corresponding articles that met the inclusion criteria were included in the study. Excluded articles included reviews, and case reports in which there was inadequate clinical data or the diagnoses of MGCT was made prior to the pregnancy or during the postpartum period. Clinical information Among eligible articles, the following variables were abstracted from the case description: (i) clinical demographics included year

Materials and methods Source This review was conducted based on MOOSE guideline for systematic reviews by using PubMed and MEDLINE [12]. The entry keywords used for the search were ‘‘pregnancy’’ and each type of ovarian MGCT, including ‘‘dysgerminoma’’, ‘‘endodermal sinus tumor’’, ‘‘yolk sac tumor’’, ‘‘immature teratoma’’, ‘‘malignant transformation, mature cystic teratoma’’, ‘‘mixed germ cell tumor’’, ‘‘non-gestational choriocarcinoma’’, ‘‘polyembryoma’’, ‘‘embryonal carcinoma’’, and ‘‘gonadoblastoma’’. The search was limited to articles published in English between 1955 and June 1, 2013 as in our previous works [13,14]. The screening process was shown in Fig. 1. A total of 2817 articles were searched from the databases, and 88 articles met our criteria. These articles were further assessed and 11 articles were removed because of inadequate

Fig. 1. Selection criteria of literature search. PubMed/MEDLINE were used to identify the articles with entry keywords ‘‘pregnancy’’ and each type of ovarian malignant germ cell tumor.

M. Kodama et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 181 (2014) 145–156

and country of publication, age at diagnosis, parity, and presenting symptoms; (ii) pregnancy outcomes included gestational age at delivery, route of delivery, birth weight, Apgar scores, outcome of pregnancy, and neonatal complications including malformation, abnormal laboratory data, and death; (iii) tumor characteristics included histology subtype, gestational age at tumor diagnosis, tested tumor markers and its results, tumor laterality and size/ weight, and stage; (iv) type of intervention and treatment included gestational age for tumor surgery, type and timing of chemotherapy, and type and timing of radiotherapy; and (v) maternal survival outcome included presence of absence of recurrence and its location, type of treatment for recurrence, and overall survival (OS). Definition If the study did not specify the exact gestational age, ‘‘early’’ and ‘‘late’’ pregnancy was defined as first and third trimester, respectively. Intrauterine growth restriction (IUGR) was defined as less than the 10th percentile in birth weight [90]. The expectant management period was determined by the duration between the initial diagnosis of ovarian MGCT during pregnancy and the date of delivery. In cases in which gestational age at delivery was described as term, the expectant delay was calculated on the assumption of delivery at 37-week gestation. If the survival period had been determined from completion of initial cancer treatment, this was recalculated to survival from the initial diagnosis of ovarian MGCT. Serious adverse events were defined as: tumor rupture or torsion, obstructed labor, placental abruption, fetal malformation among cases with in-utero chemotherapy exposure,

147

maternal death, and neonatal death. Endodermal sinus tumor cases were categorized as yolk sac tumor. The International Federation of Gynecology and Obstetrics (FIGO) staging system was used for staging [91], and for immature teratoma grading [92]. Cisplatin/bleomycin-based regimen included bleomycin–etoposide–cisplatin or carboplatin (BEP or CEB), etoposide–cisplatin (EP), and cisplatin–vinblastine–bleomycin (PVB) regimen. Abnormal laboratory data was defined as CA-125 over 100 U/ml, alphafetoprotein (AFP) value greater than 2.5 multiples of the median of normal pregnant population, or lactate dehydrogenase (LDH) increased over the normal laboratory values in uncomplicated pregnancies [93]. To examine the significance of management patterns, evaluated cases were grouped in the following manner based on the sequence of management patterns (timing of tumor diagnosis and tumor surgery) as shown in Fig. 2; ante-partum tumor diagnosis and tumor surgery without fetal conservation (Group 1), ante-partum tumor diagnosis and tumor surgery with fetal conservation (Group 2), ante-partum tumor diagnosis and tumor surgery postponed until intra-/post-partum period (Group 3), intra-partum tumor diagnosis and intra-/post-partum tumor surgery (Group 4). The systematic review is exempt from the Institutional Review Board (IRB) approval in our institutions. Statistical analysis The normality of continuous variables were tested by Kolmogorov–Smirnov test, and described as mean (SD) or median (range). The frequencies and means among the four groups of management pattern were evaluated by the chi-square test and oneway ANOVA test, respectively. The Wilcoxon signed rank test was

Fig. 2. Management patterns of malignant germ cell tumors of the ovary. Cases were categorized by the pattern of management during pregnancy based on timing of diagnosis and fetal conservation at tumor surgery.

148

M. Kodama et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 181 (2014) 145–156

used to test Apgar scores among the four groups. Fisher’s exact test was used for the comparison of two categorical variables described as odds ratio (OR) and 95% confidence interval (CI). Survival curves were constructed by the Kaplan–Meier method, and the Log-rank test was used to determine statistical significance. Cox proportional hazard regression test was used to determine the significance of survival outcome in multivariate analysis expressed as hazard ratio (HR) with 95% CI. p-values of less than 0.05 were considered statistically significant (two-tailed). All statistical analyses were performed using Statistical Package for Social Scientists software (SPSS, Inc., version 21.0, Chicago, IL, USA).

Results Clinical characteristics Patient characteristics are shown in Table 1. The mean age of the patients was 25.8. The majority of cases were reported from Asia (52.0%), and dysgerminoma and yolk sac tumor were the two most common histologic subtypes (38.2% and 30.4%, respectively). The most common symptom in all cases was abdominal or pelvic pain (35.3%) followed by abdominal distention or a growing mass (each, 19.6%). Incidental tumor diagnosis occurred in 21.6% of all cases. Multiple symptoms were reported in 18.6% of cases. Abnormal AFP and LDH were frequently reported in ovarian MGCT during pregnancy (80.8% and 85.7%, respectively, Table S1). Ultrasonography was commonly used as the diagnostic test during pregnancy (48.0%).

Table 1 Demographics of pregnancy complicated by ovarian malignant germ cell tumor. Number

n = 102

Year of publication 1955–1979 1980–1999 2000–2013

19 (18.6%) 30 (29.4%) 53 (52.0%)

Area of publication Asia North America Europe Others Age 20 21–34 35 Not recorded

53 (52.0%) 23 (22.5%) 22 (21.6%) 4 (3.9%) 25.8 (5.5) 19 (18.6%) 73 (73.7%) 7 (7.1%) 3 (2.9%)

Histology Dysgerminoma Yolk sac tumor Immature teratoma Gonadoblastoma Malignant mixed germ cell tumor Malignant transformationa Embryonal carcinoma

39 31 16 5 5 4 2

Presenting symptomb Abdominal/pelvic pain Abdominal distention Growing mass

36 (35.3%) 20 (19.6%) 20 (19.6%)

Extent of symptoms Multiple Single None (incidental)

19 (18.6%) 61 (59.8%) 22 (21.6%)

(38.2%) (30.4%) (15.7%) (4.9%) (4.9%) (3.9%) (2.0%)

Number (%) or mean (SD) are shown. a Malignant transformation of mature cystic teratoma. b Three most common symptoms are shown (included multiple symptoms per case).

Pregnancy outcomes Gestational age at tumor diagnosis was evenly distributed (1st, 2nd, and 3rd trimester: 30.7%, 38.6%, and 30.7%, respectively). Gestational age at delivery was mainly in the third trimester (1st, 2nd, and 3rd trimester: 11.1%, 10.1%, and 78.8%, respectively). Pregnancy outcomes based on the correlation of gestational age at tumor diagnosis and delivery are shown in Table 2. A distinct shift in outcome was observed based on the correlation of two time points. Details of pregnancy outcomes are further shown in Table 3. The majority of cases resulted in live birth (77.5%) at term (56.6%) via Cesarean section (65.8%). The mean birth weight was 2553 g and IUGR was common (22.8%). Other pregnancy outcomes included elective termination (11.8%), in-utero fetal demise (IUFD, 3.9%), ectopic pregnancy (2.9%), and spontaneous abortion (2.9%). Serious maternal and fetal adverse events were relatively common, with at least one event occurring in 20.6% of all cases (Table 3). Among 80 cases of intentional pregnancy, there were 22 (27.8%) cases with in-utero chemotherapy exposure. There was no statistical difference in the risk of IUGR between in-utero chemotherapy exposure as compared to no exposure (31.9% versus 19.3%, OR 1.95 95% CI 0.64–5.94, p = 0.26). The type of chemotherapy did not impact the risk of IUGR (p = 0.12), and no loss of hearing acuity was reported in chemotherapy exposed cases. Only one fetal malformation (1%) was identified, a case of ventriculomegaly that developed at 27 weeks following a course of BEP at 25 weeks [32]. Three neonates (2.9%) died of prematurity, one affected by Potter syndrome [27,71,89]. During the pregnancy course, obstructed labor, tumor rupture and torsion occurred in 2.9%, 8.8%, and 1.0%, respectively [20,77,85]. Placental abruption occurred in one case at 34 weeks of pregnancy [51]. Maternal deaths occurred in three cases (2.9%), due to (i) pulmonary embolism [52], (ii) postoperative hemorrhage with disseminated intravascular coagulopathy [40], and (iii) tumor rupture with IUFD followed by systematic coagulopathy [78]. Oncologic outcomes The results are shown in Table 4. Most of ovarian MGCTs were unilateral (84.3%) with stage I (76.4%) disease, and approximately two thirds of cases underwent unilateral adnexectomy. The mean tumor size in all cases was 17.9 cm, with 30% measuring >20 cm. Lymphadenectomy was infrequently performed during pregnancy (pelvic, 1 case; and para-aortic, none). Among 22 cases of stages II– IV cases, only 8 (36.4%) cases reported the extent of residual disease (no residual disease n = 3, residual tumor 20, 46.4% versus 90%, p < 0.01, Fig. 3A). Histologic subtype did not impact on survival (p = 0.50, Fig. 3B). Larger tumor size and advanced-stage disease were associated with worse OS (tumor size >20 versus 20 cm, 62.1% versus 95.2%, p < 0.01, Fig. 3C; and stages II–IV versus stage I, 28.4% versus 96.8%, p < 0.01, Fig. 3D). Advanced-stage disease was statistically associated with younger age (proportion of age 20, stages II–IV versus I, 36.4% versus 14.3%, OR 3.43 95% CI 1.17–10.1, p = 0.031) and tumor size (proportion of tumor 20 cm, 69.2% versus 21.1%, OR 8.43, 95% CI 2.21–32.2, p = 0.001) but not with histologic type (p = 0.14). When controlling for age and tumor size, advanced-stage disease remained an independent prognosticator of poor survival outcome (HR 21.6, 95% CI 2.06–226, p = 0.01). Chemotherapy type was different across years of publication (proportion of BEP combination therapy, before 1980 versus after 1980, 0% versus 60%, p = 0.031). Amongst cases that received chemotherapy, cisplatin/bleomycin-based chemotherapy was Table 3 Outcomes of pregnancy complicated by ovarian malignant germ cell tumor. Variables

N (%)

Timing of birth Full term Preterm

43 (56.6%) 33 (43.4%)

Delivery route Abdominal Vaginal

50 (65.8%) 26 (32.9%)

Birth weight (g) IUGR Apgar score 1 min 5 min

2553 (787) 18 (22.8%)

8 (4–10) 9 (6–10)

Expectant management Duration (week)

59 (57.8%) 14.7 (7.7)

Serious adverse event Rupture of tumor Maternal death Neonatal death Obstructed labor Torsion of tumor Placental abruption Fetal major anomaly Any events

9 3 3 3 1 1 1 21

(8.8%) (2.9%) (2.9%) (2.9%) (1.0%) (1.0%) (1.0%) (20.6%)

Number (%), median (range), or mean (SD) are shown. Abbreviations: GA, gestational age; IUFD, in-utero fetal demise; IUGR, in-utero growth restriction.

Table 4 Oncologic outcomes of ovarian malignant germ cell tumor. Number

n = 102

GA at tumor surgery 1st trimester 2nd trimester 3rd trimester Postpartum

9 49 33 11

Tumor laterality Unilateral Bilateral Not recorded

86 (84.3%) 3 (2.9%) 13 (12.7%)

Type of surgery Adnexectomy Unilateral Bilateral Other

68 (66.7%) 31 (30.4%) 3 (2.9%)

Additional proceduresa Hysterectomy Pelvic lymphadectomy Aortic lymphadectomy Omentectomy Appendectomy

29 14 8 39 6

Tumor size (cm) >20 cm 20 cm Missing

17.9 (9.3) 21 (20.6%) 49 (48.0%) 32 (31.4%)

Tumor weight (gram)

(8.8%) (48.0%) (32.4%) (10.8%)

(28.4%) (13.7%) (7.8%) (38.2%) (5.9%)

2105 (1587)

Stage I II–IV Missing

78 (76.4%) 22 (21.6%) 2 (2.0%)

Chemotherapy

53 (52.0%)

Timing of chemotherapy Fetus in-uterob Postpartumc

22 (41.5%) 45 (84.9%)

Type of chemotherapy CDDP/bleomycin-based Others

40 (76.9%) 12 (23.1%)

Radiotherapy (postpartum)

10 (9.8%)

Recurrence during pregnancy

7 (6.9%)

Number (%) or mean (SD) are shown. Abbreviations: GA, gestational age; CDDP, cisplatin; AFP, alpha-fetoprotein; BEP, bleomycin-etoposide-cisplatin. a Either at the time of initial tumor surgery or subsequent separate surgery. b 14 cases received both in-utero and postpartum chemotherapy. c 31 cases received postpartum chemotherapy alone.

150

M. Kodama et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 181 (2014) 145–156

Fig. 3. Survival outcomes of malignant germ cell tumors of the ovary. Overall survival are shown based on (A) age at diagnosis, (B) histology type, (C) tumor size, (D) stage, (E) chemotherapy type, and (F) year at publication. Log-rank test for p-values. Abbreviations: CDDP, cisplatin.

associated with improved outcome (cisplatin/bleymyocin-based versus others, 90.7% versus 0%, p < 0.01, Fig. 3E). Year of publication did not impact survival outcome (5-year OS rate, before 1980 versus after 1980, 69.2% versus 81.4%, p = 0.18). Fertility sparing surgery for stage I disease was not associated with changes in survival outcomes (data not shown). Management pattern and feto-maternal outcomes Cases were grouped based on the management pattern as shown in Fig. 2. Group 2 was the most common (45.1%), followed by Group 1 (21.6%), Group 4 (20.6%), and Group 3 (12.7%). Demographics of the four groups are shown in Table 5. The age at diagnosis was similar across the groups (p = 0.56). There was a

trend toward an increased proportion of mixed MGCT/pure nondysgerminoma in Group 1 than in the other groups (p = 0.056) but histology type was not associated with gestational age at tumor diagnosis (p = 0.16). Incidental tumor diagnosis was common in Group 4 (47.6%, p = 0.008). In 13.0% of Group 2 cases the ovarian tumor was identified during the evaluation of elevated maternal serum AFP. Pregnancy outcomes based on management patterns are shown in Table 6. When management pattern was evaluated, Group 1 had more cases with the first trimester diagnosis (59.1%) than other groups (p < 0.01). There were significant differences in pregnancy outcome across the four groups, and none in Group 1 resulted in viable pregnancy (p < 0.01). While non-viable pregnancies were seen in nearly half of Group 1 (spontaneous abortion 13.6%, tubal

M. Kodama et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 181 (2014) 145–156

151

Table 5 Demographics based on management patterns. Group 1

Group 2

Group 3

Group 4

Number

n = 22

n = 46

n = 13

n = 21

Year of publication 1955–1979 1980–1999 2000–2013

5 (22.7%) 7 (31.8%) 10 (45.6%)

8 (17.4%) 14 (30.4%) 24 (52.2%)

2 (15.4%) 3 (23.1%) 8 (61.5%)

4 (19.0%) 6 (28.6%) 11 (52.4%)

Area of publication Asia North America Europe Others

11 (50.0%) 7 (31.8%) 3 (13.6%) 1 (4.5%)

22 (47.8%) 13 (28.3%) 10 (21.7%) 1 (2.2%)

6 1 4 2

(46.2%) (7.7%) (30.8%) (15.4%)

14 (66.7%) 2 (9.5%) 5 (23.8%) 0

Age 20 yr

24.6 (5.7) 6 (27.3%)

26.6 (5.9) 8 (18.6%)

25.7 (4.5) 2 (15.4%)

25.5 (5.2) 3 (14.3%)

Histology Dysgerminoma Yolk sac tumor Immature teratoma Gonadoblastoma Malignant mixed germ cell tumor Malignant transformationa Embryonal carcinoma

6 8 2 4 1 0 1

(27.3%) (36.4%) (9.1%) (18.2%) (4.5%)

14 (30.4%) 17 (37.0%) 10 (21.7%) 0 2 (4.3%) 3 (6.5%) 0

7 3 1 0 1 0 1

(53.8%) (23.1%) (7.7%)

12 (57.1%) 3 (14.3%) 3 (14.3%) 1 (4.8%) 1 (4.8%) 1 (4.8%) 0

Presenting symptom Abdominal/pelvic pain Abdominal distention Growing mass Abnormal MSAFP screening Weight loss Nausea/vomiting Shock vital Virilization Incidental

8 6 3 0 2 0 2 0 5

(36.4%) (27.3%) (13.6%)

19 (41.3%) 6 (13.0%) 12 (26.1%) 6 (13.0%) 0 0 0 0 6 (13.0%)

5 5 2 0 0 1 0 1 1

(38.5%) (38.5%) (15.4%)

p-value

0.99

0.20

0.56 0.71 0.056b

(4.5%)

(9.1%) (9.1%) (22.7%)

(7.7%) (7.7%)

(7.7%) (7.7%) (7.7%)

4 (19.0%) 3 (14.3%) 3 (14.3%) 0 1 (4.8%) 1 (4.8%) 0 0 10 (47.6%)

0.36 0.15 0.52 0.051 0.18 0.22 0.06 0.08 20 cm, 9.1%, p < 0.01). Stage distribution was equal across the four groups (p = 0.10). However, in post hoc analysis, Group 1 had more advanced-stage disease when compared to Group 2 (36.4% versus 11.4%, p = 0.023). In addition, there were more gonadoblastoma and embryonal carcinoma and fewer immature teratoma and malignant transformation cases in Group 1 when compared to Group 2 (p = 0.039). The majority (77.8%) of Group 3 cases received chemotherapy with the fetus inutero (p < 0.01). In Group 2, disease recurrence following surgery occurred in 7 (15.2%) cases during the course of the pregnancy. Of the management patterns, Group 2 showed the highest 5-year OS rate (92.8%) followed by Group 4 (79.5%), Group 3 (71.4%), and Group 1 (56.2%, p = 0.028, Fig. 3F).

Expectant management of pregnancy Group 2 and 3 both underwent expectant management of pregnancy in our study (Fig. 2). The mean duration of expectant management for Groups 2 and 3 were 16.4 and 7.8 weeks, respectively (p < 0.01), and 98.3% resulted in a live birth. In order to assess if expectant management of pregnancy impacts survival outcome of women with ovarian MGCT, multivariate analysis was performed (Table 8). Expectant management of pregnancy did not remain an independent prognosticator for OS (p = 0.43) after controlled for age, histology, tumor size, stage, gestational age for tumor surgery, and chemotherapy. Age 20 and stages II–IV disease remained independent prognostic factors associated with decreased OS (p = 0.032 and 0.02, respectively, Table 8). Discussion Ovarian MGCT is an extremely rare complication of pregnancy, and our search yielded only 77 articles over a half century. Given its rarity, clinical characteristics and ultimate management recommendations remain understudied for pregnancies complicated by ovarian MGCT. Salient factors which should be taken into consideration for its management should include optimizing outcomes for both mother and fetus. Several key areas in the study deserve special mention. The timing of delivery for pregnancies complicated by malignancies should take into account the balance between prolonging pregnancy for fetal benefit and delivery of the fetus for maternal benefit. Indeed, preterm delivery is one of the most common complications in pregnancy affected by malignancies.

152

M. Kodama et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 181 (2014) 145–156

Table 6 Outcomes of pregnancy based on management patterns. Group 1

Group 2

Group 3

Group 4

p-value

Number

n = 22

n = 46

n = 13

n = 21

GA at diagnosis 1st trimester 2nd trimester 3rd trimester

13 (59.1%) 6 (27.3%) 3 (13.6%)

16 (34.8%) 29 (63.0%) 1 (2.2%)

2 (16.7%) 4 (33.3%) 6 (50.0%)

0 0 21 (100%)

Outcome of pregnancy Spontaneous abortion Ectopic pregnancy IUFD Fetal lost after surgery Elective termination Live birth

3 (13.6%) 3 (13.6%) 4 (18.2%) 0 12 (54.5%) 0

0 0 0 1 (2.2%) 0 45 (97.8%)

0 0 0 0 0 13 (100%)

0 0 0 0 0 21 (100%)

GA at delivery 1st trimester 2nd trimester 3rd trimester

10 (47.6%) 7 (33.3%) 4 (19.0%)

0 3 (6.8%) 41 (93.2%)

0 0 13 (100%)

0 0 21 (100%)

Timing of birth Full term Preterm

– –

25 (58.1%) 18 (41.9%)

4 (33.3%) 8 (66.7%)

14 (66.7%) 7 (33.3%)

Delivery route Abdominal Vaginal

– –

21 (50.0%) 21 (50.0%)

9 (69.2%) 4 (30.8%)

20 (95.2%) 1 (4.8%)

Birth weight (g) IUGR

– –

2549 (775) 11 (24.4%)

2435 (629) 3 (23.1%)

2697 (1014) 4 (19.0%)

Apgar score 1 min 5 min

– –

8 (4–10) 9 (7–10)

9 (9–9) 10 (9–10)

7.5 (7–8) 9 (6–10)

Feto-maternal outcomes of pregnancy complicated by ovarian malignant germ cell tumor: a systematic review of literature.

Malignant germ cell tumors (MGCT) are a rare type of ovarian cancer with poorly understood behavior during pregnancy. This systematic review evaluated...
971KB Sizes 0 Downloads 3 Views