Central nervous system (CNS) germ cell tumors (GCTs) are a family of predominantly malignant neoplasms impacting pediatric, adolescent and young adult populations, with a median age at diagnosis from 12 to 16 years old [1]. Overall, GCT patients are overwhelmingly male, with >90% of pineal lesions [2], and approximately 50% of neurohypophyseal lesions arising in men [3]. GCTs arise primarily in midline structures including the pineal gland and neurohypophysis most frequently, followed by basal ganglia, thalamus, cerebrum, cerebellum, and spinal cord [4, 5].

Histopathologically, germinoma accounts for 50–60% of GCTs, while non-germinomatous GCTs (NGGCTs) are a minority [1]. According to WHO classification, NGGCTs include mature and immature teratoma (MT and ImT), teratoma with somatic-type malignancy, yolk sac tumor (YST), choriocarcinoma (CC), embryonal carcinoma (EC), and mixed GCT, which may demonstrate any combination of the aforementioned histological components [6]. Germinoma responds well to platinum-based chemotherapy and whole ventricular irradiation (WVI), with > 90% survival in long-term follow-up studies [1, 7, 8]. By contrast, NGGCTs are prognostically unfavorable in comparison and demonstrate only 60–70% long-term survival, in particular among tumors with a malignant component (YST, CC, and EC) [1]. Correspondingly, treatment protocols are more aggressive and frequently incorporate intensified chemotherapy regiments with both platinum and alkylating agent chemotherapy as well as well as irradiation, although radiation coverage protocols are heterogeneous and vary by treatment location and active clinical trial participation, among other factors [9,10,11,12].

Teratoma biology is heterogeneous, and although the MT phenotype is less aggressive and treatment is oriented around surgical resection alone, ImT treatment remains an area of active study. At present, COG clinical trials include ImT with other NGGCTs if tumor markers is elevated or the diagnosis has been histopathologically proven; SIOP protocols allow for ImT treatment planning on a case-by-case basis [8, 13, 14]. In Japan, ImT is classified within the “intermediate risk group,” prompting treatment with chemotherapy and irradiation that is more intensive than germinoma protocols but less aggressive than NGGCTs with a dominant malignant component [5]. Correspondingly, diagnostic differentiation between germinoma, MT/ImT, and phenotypically malignant NGGCTs has significant implications with regard to treatment pathways, patient education, and prognostic counseling.

GCT diagnosis is established through various methods, which may incorporate imaging findings, histopathological specimens, and tumor markers from serum or cerebrospinal fluid (CSF). In COG and SIOP protocols, tumor marker thresholds are defined for the diagnosis of NGGCTs (with a malignant component), which in turn prompts protocol-based therapeutic assignment [11]. However, a preceding report argued that elevation of tumor markers was not limited to NGGCTs with a malignant component [15], while another study demonstrated that HCG RNA was expressed across all histological subtypes [16]. Correspondingly, establishing a GCT diagnosis and proceeding with intensive therapies on the basis of tumor markers alone remains controversial and potentially risky. In tandem, histopathological diagnosis based on a small biopsy specimen predisposes to sampling error and associated under-estimation in the prevalence of malignant components. Collectively, these complexities have yielded a highly heterogenous set of protocols with regard to how tumor markers are incorporated into algorithms for GCT diagnosis and treatment across the world [17].

The goal of the current study was to assess the distribution of elevated tumor markers in resection-based histopathologically confirmed GCT cases, as well as the relationship between tumor markers and histopathologic diagnosis, with particular attention to the range of marker abnormalities observed in GCTs without a YST or CC component. Further emphasis was placed on examining the relative yield of tumor markers measured from serum, CSF, or both, as well as the relationship between marker levels, clinical phenotype, and treatment response in ImT.