GTN is a malignant tumor that shows variable epidemiologic features and geographical patterns from published studies across the regions [3]. Data of choriocarcinoma from Asian countries were heterogeneous, ranging between 63 to 202 per 100,000 pregnancies in surveys from China, India, and Indonesia and below 23 per 100,000 pregnancies in studies from Japan and Singapore [3, 19–23]. In contrast with relatively high ratios of GTN found in these regions, a remarkably low ratio was found in Europe, North America, Australia, and some area of Latin America [3, 19, 20, 24, 25]. These regional differences of GTN data is unavoidable, which may, in part, be due to differences in calculating methods of disease ratio and histological classification. Although a considerable number of retrospective studies and surveys is published to describe the clinical features for GTN patients in China [9, 10], most reports are based on major hospitals that locate in the southeast, especially in some developed regions, and thus lack precise researches needed for a reliable assessment of GTN in under-developed regions.
GTN patients in the northwest of China were treated at different hospitals rather than in designated disease centers. The First Affiliated Hospital of Xi'an Jiaotong University is one of the largest disease centers in the northwest of China, which plays an important role for GTN patients. This study totally included 387 GTN patients from the First Affiliated Hospital of Xi'an Jiaotong University between 2009 and 2020. By analyzing differences of clinical features between 2009–2014 and 2015–2020 groups, we found that parity and metastatic conditions have changed over time. In this study, as much as 33 (18.7%) of 177 patients before 2014 had at least two babies, 61 (29.0%) of 210 patients had at least two babies after 2015, indicating an increased fertility desire among the patients in 2015–2020 group. This finding is consistent with a published Chinese study involving 291 patients, with a mean parity of 1.29 ± 0.81 and 0.90 ± 1.00 for choriocarcinoma patients in the 2005–2009 group and 2000–2004 group, respectively [26]. However, analyses were not conducted to know that children are born before or after they get GTN. Before patients getting GTN, the two-child policy in China that introduced in recent years may be a reason for increased parity among patients in 2015–2020 group [27]. After disease remission, scientific guidance regarding effects of chemotherapy regimen on pregnancy and timing of another pregnancy may increase opportunities for patients to successfully become pregnant. This study may not completely explain the difference in parity and delivery between two groups, but it is tempting to speculate that main reasons affecting the desire for pregnancy of patients in 2009–2014 group are not GTN itself.
Approximately 30% GTN cases have distant metastases at the time of initial assessment, most commonly to lungs (80% of patients), followed by vagina (30%), liver (10%), and brain (10%). Therefore, chest radiography is recommended for all patients with GTN at the time of diagnosis. Metastases in lungs can be detected on chest X-ray or CT scan and the latter one is much more sensitive to detect smaller lesions with greater accuracy [28]. However, a debate regarding whether CT scan is superior to chest X-ray in detecting lung metastases is ongoing and some previous studies found conflicting results [11, 12]. For this reason, chest CT is not commonly used in patients with GTN. In this study, patients in 2015–2020 group have more distant invasions affecting organs, especially lungs, which may attribute to heavy use of CT scan that confuses pulmonary nodule with metastatic lesions. In lung cancer screenings with CT scan, the average nodule detection rate per round in randomized controlled trials and cohort studies ranged from 3–50% [29, 30]. Therefore, lung nodules might be misdiagnosed as metastatic lesions with evident invasive mole and choriocarcinoma [31, 32]. We, thus, cannot simply conclude that the time has changed features of GTN. However, this study could remain gynecologic oncologist and radiologist to distinguish metastatic lesions that originate from GTN between lung nodules.
GTN is a group of pregnancy-related malignancies with few data given their rarity that occurs in women of reproductive age. Some previous studies have confirmed that maternal age was a significant factor in the risk for molar pregnancies [4, 5, 33]. According to a Canadian study involving 428 women that were identified with gestational trophoblastic disease (GTD), the rates of hydatidiform mole were highest in both younger (less than 20 years old) and older age groups (30 or more years old) [33]. It was also noted that GTN was disproportionately higher in those at extremes of age including 10 to 19 years and 40 to 54 years, based on a recent study [4, 5]. In our study, the mean age of GTN patients is 32.4 years, which is similar to the median age (32 years) at the start of treatment for patients in a UK study [34]. The proportion of patients aged over 40 years in 2015–2020 group was higher than that reported in GTN patients with low-risk scores (25.2% vs 10.3%) [35], indicating a failure to apply contraception. Dantas et al. found that hormonal contraception, the most common contraceptive methods, does not increase the occurrence of postmolar GTN and the clinical progression of GTN or retard β-hCG normalization [36]. Therefore, when considering maternal age at the time of GTN diagnosis, the effects of age for clinical progression and another pregnancy should be concerned.
Hydatidiform mole was the most common type of antecedent pregnancy among GTN patients in this study, followed by abortion and term or preterm gestation. The composition of antecedent pregnancy of GTN patients has no significant differences between 2009–2014 and 2015–2020 groups. Although abortion has been found to be one of the most frequent antecedent pregnancy among patients with ultra high-risk GTN [9], this was not observed in our study because we did not divide patients into groups using a FIGO score. Invasive mole usually developed from hydatidiform mole, whereas choriocarcinoma could derive from other gestational types including miscarriage and term pregnancy, which may lead to differences in primary symptoms [1, 2]. Vaginal bleeding was the most typical and common symptoms, followed by enlarged uterus for gestational age, theca lutein cyst. Elevated β-hCG level and local invasions were main causes for vaginal bleeding and enlarged uterus for gestational age. Symptoms or signs vary with disease location. Patients with lung metastasis might have a combination of haemoptysis and shortness of breath, whereas those with brain metastasis might present with seizure or headache. Therefore, when considering clinical manifestations for reproductive women showing respiratory and nervous symptoms, we should consider the possibility of GTN.
The FIGO 2019 guideline recommended that GTN patients with FIGO score less than 7 would receive monochemotherapy [37]. In this study, 309 (79.8%) of 387 patients scored less than 7, but only 157 (40.6%) of 387 patients received single-agent chemotherapy, indicating that some cases with a FIGO score less than 7 were given a multi-drug chemotherapy. Although FIGO scoring system would be helpful for identifying low-risk patients who need single-agent chemotherapy, the following conditions warrant a switch to multi-drug chemotherapy, including poor response to initial therapy, significant elevation in β-hCG level, development of metastasis, or resistance to sequential single-agent chemotherapy regimens [37]. In this study, 230 (59.4%) of 398 patients received multi-drug chemotherapy. In our center, FAM is the most used multi-drug regimen that given patients with a score ranging 5 to 6, which was not displayed in this study. In our center, patients scoring 5 or over would be given multi-agent chemotherapy including FAV, FAM, FEM, AEM, FAEV, and EMA-CO. However, some regimens were not widely used and we will conduct a cohort study to compare these regimens. For patients with drug-resistant disease, surgical resection can aid cure and β-hCG level will drop rapidly if lesion has been completely removed.
Some limitations of our study should be mentioned. First, this is a retrospective study that has incomplete information and recall bias; for example, history of spontaneous abortions may have been ignored. Second, we cannot exclude an inherent selection bias linked to the retrospective nature of the analysis and multicenter origin of patients. Third, we did not include patients diagnosed before 2009 because the information of these patients was recorded using paper-based medical documents that is not comprehensive. Finally, there are also some confounding factors, such as socioeconomic policies.
Overall, the results of this study showed that demographics and clinical features of patients with GTN did not change over time, with the exception of parity and metastatic conditions which have increased in the period of 2015 to 2020. As such, parity condition and functions of imaging tests in diagnosing GTN patients should be re-analyzed in the future.