Intracranial germ cell tumors (GCTs) are rare, heterogeneous, and management challenging, with a geographically variable incidence of 0.6–2.7 per million (1, 2). GCTs usually develop in the region of the third ventricle along the midline axis, including sellar region, pineal region, or both regions involved (3, 4). In rare circumstances, GCTs also originate off the midline in the basal ganglionic region, thalamus, or other ventricular sites. It is reported that GCTs in basal ganglionic region are more likely to be germinomas, which are prone to delayed diagnosis for their insidious onset and atypical clinical presentation (3, 5, 6).

Basal ganglia consist of subcortical nuclei embedded in the deep brain hemispheres responsible for control of movement, behavior, cognition, and emotions (7). Slowly progressive hemiparesis and cognitive decline are frequent in patients with GCTs in the basal ganglia (8, 9). In this study, we define basal ganglia area based on the imaging, including internal capsule and peripheral white matter in addition to anatomical BG (striatum, claustrum, and amygdaloid body).

Due to the rarity of basal ganglia germ cell tumors (BGGCTs), there is a paucity of research on it and a lack of study on large populations. Our study aims to determine the demographic and clinical characteristics, associated factors, and impact of delayed diagnosis and treatment on outcome of BGGCTs with or without sellar involvement. We reviewed 92 BGGCT patients with complete clinical information and follow-up data since 2000, including 16 patients diagnosed with BGGCTs at our institution and 76 cases from other published studies. To the best of our knowledge, this study included the largest population so far.

The interquartile range of age at diagnosis is from 9.75 to 14 years, with a median age of 12.5 years and a peak between 10 and 14 years. As shown in Figure 1, it is noticeable that males are more likely to have BGGCTs with or without sellar involvement in all age groups. Young-age onset and male predominance are two remarkable features of BGGCTs, which is consistent with previous studies (6, 10).

Basal ganglia are a group of evolutionarily conserved deep forebrain nuclei, which form multiple parallel loops and reentering circuits and are involved in motor, cognition, and affective control (7, 11). Abnormalities in the basal ganglia area can lead to movement disorders, alterations of mood, and cognitive disorders. Distinctive clinical features can lead to early diagnosis and optimize the treatment. However, patients with BGGCTs with or without sellar involvement present with a variety of nonspecific clinical manifestations. Slowly progressive hemiparesis has been recognized as the initial and most common symptoms (12, 13), indicating the progressive lesions of the basal ganglia and internal capsule area (8). In this study, hemiparesis (67.0%) is also the most common initial complaint, and many previous studies have also reported the same finding (6, 13–15). Cognitive disturbance (24.2%) follows behind, indicating an abnormality in the basal ganglia area. Signs of increased intracranial pressure, like nausea/vomiting (22.0%) and headache (16.5%), are relatively frequent symptoms. Diabetes insipidus (19.6%) is observed and may indicate tumor involvement of the sellar area.

Magnetic resonance imaging (MRI) is highly sensitive in detecting intracranial GCTs, though it is limited in distinguishing GCTs with different pathological types (3). Brain MRI is routinely performed on all patients at the first time of evaluation. Abnormalities on MRI in patients with BGGCTs involve hemiatrophy, cystic components, mass effect, intratumoral hemorrhage, and peritumoral edema (13–15). Cerebral hemiatrophy or hemorrhagic or cystic formation is highly suggestive of BGGCTs (13). Hyperintensity in T2WI and heterogeneous enhancement seem more likely to be observed. Hemiatrophy, usually presenting as ipsilateral peduncle and hemispheric atrophy, is the most common radiologic feature.

In one study, five patients, among eight cases observed with hemiatrophy on MRI, present with hemiparesis (14), which may imply an association between hemiparesis and hemiatrophy. Atrophy of the basal ganglia, even before the development of hemiparesis, was considered the earliest and most characteristic diagnostic feature in patients with BGGCTs (16).

BGGCTs can be classified as different types by the MRI features, but there is a lack of consensus on the universal standard. In the current study, we adopt a method of categorizing BGGCTs into four distinct patterns: a subtle lesion with faint or no enhancement (type 1), a small lesion <3 cm with enhancement (type 2), type 2 combined with subependymal seeding (type 3), and a large lesion ≥3 cm (type 4), among which type 1 lesions are easily misdiagnosed with non-tumorous conditions and have a longer delay in taking a biopsy (6).

Delayed diagnosis, which is defined as an interval of ≥6 months (from the onset of symptoms to the date of diagnostic MRI) (5), has been a topic of great concern in intracranial GCTs. It was considered that a longer diagnostic delay increased the risk of disseminated disease, negatively influenced treatment outcomes, left the occult lesions with enough time to develop into a full-blown disease, and significantly shorten OS (5, 6, 17). Compared with GCTs in other regions, GCTs in basal ganglia had a longer delay in diagnosis (15). In this study, delayed diagnosis was identified as a significant independent prognostic factor (OR 2.33, p = 0.045) for BGGCTs with or without sellar involvement, consistent with previous studies (6, 17). Besides, we observed that patients with BGGCTs and delayed diagnosis were more likely presenting with cognitive disturbance, mental disturbance, and diabetes insipidus (Table 1). In terms of clinical characteristics, delayed diagnosis was more likely in patients with multifocal lesions or tumor dissemination than that with a solitary lesion (Table 2). Although there is no definite cause and effect between clinical features/characteristics mentioned above and delayed diagnosis, it is recommended to carry out active diagnostic procedures in patients with high suspicion.

Different classification systems of intracranial GCTs were proposed based on histopathology, prognosis, and immunohistochemical markers. Germinoma GCT (GGCT) and NGGCTs are two main tumor patterns divided by pathological features, with the latter one further subdivided into embryonal carcinoma, endodermal sinus tumor/yolk sac tumor, choriocarcinoma, mixed GCT, and teratomas based on histology, tumor markers, and protein markers secreted by the tumor cells (18, 19). Diverse treatment outcomes and prognosis were observed in different histological subtypes, categorizing intracranial GCTs into three therapeutic groups with good, intermediate, and poor prognosis (20). Pure germinoma and mature teratoma were classified into the good prognosis group. In this study, patients with basal ganglia pure germinoma were more likely to achieve complete remission (p = 0.002) and had significantly higher OS rates and PFS rates (p = 0.003 and p = 0.001; Figure 3). Pure germinoma was revealed as a significant independent prognostic factor (OR 4.64, p = 0.002) for basal GCTs. Pure germinoma in basal ganglia showed similar prognostic characteristics to intracranial pure germinoma.

It was considered that distinct therapeutic modalities should be applied to fit the patients in different prognosis groups (20, 21). Germinomas, making up over half of all intracranial GCTs, are extremely sensitive to RT and CT. All patients with intracranial germinoma were recommended to receive RT (2). It was reported that more than 90% of patients with intracranial germinomas were cured with CSI, and the long-term OS was from 90% to 100% after RT alone with a total cranial dose of 40–50 Gy (22). CT was considered an effective way to reduce the dose and volume of RT without weakening the survival rates (2, 19, 21, 23, 24). CT alone was not recommended for its significantly inferior outcomes in patients with intracranial GCTs compared with CT combined with RT (21, 23, 25). Following neoadjuvant CT, WVI was necessary and sufficient for patients with localized germinoma, minimizing adverse effects of RT without affecting outcome (2, 19, 21, 22, 26). For patients with spinal dissemination, craniospinal irradiation was required. In terms of NGGCTs, multidisciplinary therapy, including tumor resection, radiation, and CT, was established (2, 19, 24, 27). Patients with BGGCTs displayed significantly worse intelligent performance and health-related quality of life than those with tumors in sellar and pineal regions (26, 28). WBI rather than WVI was suggested in patients with basal ganglia germinoma, since the WVI was likely not sufficient for the deep brain lesions (14, 28). This study observed a statistically significant difference in PFS rates (p = 0.001; Figure 4) between patients with basal ganglia pure germinoma receiving CSI or WVI or WBI and those receiving only PB or others. For patients with non-pure germinoma, survival analysis revealed a higher PFS rate in CSI or WVI or WBI treatment group and was not significant.

A consensus emerged that surgical biopsy was required for intracranial germ-cell tumor diagnosis without elevated AFP or HCG, regardless of radiological imaging findings (2). Despite many disadvantages, the surgical biopsy was critical to obtain an accurate pathological subtype, which predicted treatment regimens and prognosis (29, 30). The extent of surgical resection for BGGCTs is still unproven. As the evolution of neurosurgical techniques and the reduction of surgery-related morbidities, benefits may be received from more aggressive resection (14, 23). In this study, we investigated the impact of surgical debulking on clinical outcomes compared with surgical biopsy. The survival analysis of the relationship between operation types and survival was shown in Figure 5. Although without statistical significance, a trend was observed that the PFS rate of surgical debulking was higher than surgical biopsy in patients with basal ganglia pure germinoma ≥3 cm (p = 0.436; Figure 5). However, in terms of tumor size <3 cm, the surgical biopsy revealed a trend to have a better prognosis (p = 0.137; Figure 5). The result suggested that surgical debulking for basal ganglia pure germinoma with tumor size more than 30 mm may provide some benefits. Further studies with a large sample size are needed to check the hypothesis on surgical debulking.

There are several limitations in our study. This study is a retrospective study whose data were retrieved from electronic medical records in our institution and collected from previously published studies. Consequently, selection bias, missing data, and inaccurate information are inevitable. Patients included in this study were heterogeneous, leading to selection bias and additional confounders. For example, a higher proportion of patients in our institute presented with a combined sellar region involved, but the stratified analysis was not performed because of the small sample size. Moreover, analysis related to complications was limited in this study as a lack of adequate clinical and prognostic information. Finally, the small sample size of this study was not sufficient to give more reliable and precisive results, suggesting more patients involved in future studies.

In conclusion, we discussed the clinical presentations, MRI features, delayed diagnosis, prognosis, and treatment of BGGCTs with or without sellar involvement. Young-age onset and male predominance are two outstanding features. Atrophy of the basal ganglia and hemiatrophy on MRI act as diagnostic characteristics in patients with BGGCTs, responsible for hemiparesis and cognitive disturbance. Delayed diagnosis, RT field (CSI or WVI or WBI), and pure germinoma were identified as significant independent prognostic factors. Pure germinoma in basal ganglia shows similar prognostic features to pure intracranial germinoma. RT keeps playing a pivotal role. The extent of surgical resection for BGGCTs is not yet clear. This study finds potential benefits from surgical debulking for basal ganglia pure germinoma ≥3 cm. Further studies with a large sample size are suggested to check the hypothesis on surgical debulking.

The raw data supporting the conclusions of this article will be made available by the authors without undue reservation.

The studies involving human participants were reviewed and approved by the Ethics Committee of Peking Union Medical College Hospital. Written informed consent to participate in this study was provided by the participants’ legal guardian/next of kin.

Conceptualization, HZ and YY. Data curation, YZ. Formal analysis, YZ, LW. Funding acquisition, YY and YZ. Writing—original draft, YZ and LW. Writing—review and editing, LW, WM, HP, and RW. All authors contributed to the article and approved the submitted version.

This study was supported by the Chinese Academy of Medical Sciences Innovation Fund for Medical Sciences (No. 2016-I2M-1-002) for YY and Youth Science Foundation of Peking Union Medical College Hospital (No. pumch201911867) for YZ.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.