Malignant central nervous system (CNS) germ cell tumors (GCTs) are uncommon, accounting for less than 1% of primary brain tumors, and occur predominantly during childhood. The incidence of CNS GCTs has been reported to exhibit a marked association with race, appearing to be more common in Asian countries than in the West [1-13]. However, the majority of previous reports were based on single institution experiences rather than the general population. Recently, a population-based comparison study between Japan and the United States found no significant difference in the incidence of malignant CNS GCTs between these two countries [14]. In contrast, the Central Brian Tumor Registry of the United States (CBTRUS) report published in 2014 stated that Asians/Pacific Islanders (age-standardized rate [ASR], 0.19) were associated with a higher incidence rate of CNS GCTs than Caucasian (ASR, 0.10) and African-American (ASR, 0.07) individuals in the United States [15]. Thus, the existence of racial differences in the incidence of CNS GCT remains unclear.

The aim of this study was to determine the Korean nationwide incidence and survival rate for primary malignant CNS GCTs. Using the Korea Central Cancer Registry (KCCR) database, we reported the precise CNS GCT incidence by age, sex, histological type, and topography in Korea, as well as the survival rate during 2005-2012.

The incidence of CNS GCTs is known to be higher in Asia, Japan, and Korea than in other areas of the world. In particular, CNS GCTs account for relatively high proportions of pediatric brain tumors in East Asia, 7.8% in Japan [18], 14.0% in Taiwan [9], 7.9% in China [8], and 9.5% in Korea [19], whereas their frequencies in North America [6,15] and Europe [7,20] are less than 3%. According to the CBTRUS report, the overall incidence of malignant CNS GCTs between 2007 and 2011 was 0.07 per 100,000 in the United States, with the Asian/Pacific Islander group associated with a higher CNS GCT incidence rate (0.19 per 100,000) than Caucasian and African-American participants [15].

In children, the incidences of CNS GCT, including benign tumors, were 0.21 per 100,000 in total and 0.21, 0.14, and 0.51 per 100,000 in Caucasian, African-American, and Asian/ Pacific Islander populations, respectively [15]. Goodwin et al. [21] also described differences in the incidence of pediatric GCTs according to race (Caucasian, 1.29 per 100,000; African-American, 0.33 per 100,000; and Asian, 2.6 per 100,000). Moreover, the childhood incidence of primary CNS GCTs in a regional survey in Kumamoto Prefecture in southern Japan was found to be 0.45 per 100,000 [4]. These data suggest that an Asian genetic background is strongly associated with the development of CNS GCT. However, a recent study demonstrated that the incidence of CNS GCT in Japan was not significantly higher than that in the United States. McCarthy et al. [14] estimated the incidence of malignant CNS GCTs diagnosed during 2004-2006 in 14 population-based registries representing 31.8% of the total population of Japan and analyzed the Surveillance Epidemiology and End Results program database for 17 population-based cancer registries covering 26% of the population of the United States. Unexpectedly, they found the incidence of malignant CNS GCTs in Japan (0.096 per 100,000) to be similar to that in the United States (0.075 per 100,000), and showed the same sex-based patterns by location [14]. The incidence of CNS GCTs in Japan was low compared to those reported in previous studies. These results prompted us to investigate the actual nationwide population-based incidence of CNS GCTs in Korea.

Between 2005 and 2012, 654 cases of newly diagnosed primary malignant CNS GCTs were registered in the KCCR. More than 90% of these tumors were diagnosed with histological confirmation through surgical procedures. The incidence of CNS GCTs in Korea was more than two times higher than that in Japan and the United States. Furthermore, the incidence of pediatric CNS GCTs in Korea (0.434 per 100,000) was similar to the incidence of CNS GCTs in Kumamoto prefecture (0.45 per 100,000). These findings suggest possible geographical variations in CNS GCT incidences in Japan. Kumamoto and Korea may also share a similar genetic background because of their geographical proximity. An international collaborative study would be required to investigate these findings further.

The survival rate of pure germinoma is known to be very high, with the 10-year overall survival rate attaining 90%. For nongerminomatous GCT, the 10-year overall survival rates were reported to be 30%-80% [22]; however, CNS GCTs showed better prognosis among pediatric patients. Matsutani et al. [23] reported that the 10-year survival rates for pure germinoma and mature teratoma were both over 90%. Even though we did not report 10-year survival rates because of our short follow-up period, the survival outcomes observed in this study suggest that CNS GCTs are curable. The 5-year observed survival rate for all patients with CNS GCTs was 91.2%, with no differences observed according to sex, age group, or primary tumor site. However, patients with germinoma showed a greater 5-year survival rate (95.3%) than those with nongerminomatous GCTs (77.6%) and malignant teratomas (76.5%).

It should be noted that our study had some limitations. Specifically, the KCCR does not record cancer risk factors or specific treatment-related factors, making it difficult to determine specific prognostic factors for CNS GCT. Nevertheless, the KCCR data encompasses over 97% of the population of Korea [17], and includes up-to-date information with longer follow-up periods than the data sources used in previous studies. This allows incidence rates to be analyzed with excellent internal validity and generalizability.

In conclusion, our nationwide population-based study revealed that the incidence of CNS GCTs in Korea is significantly higher than the reported incidence for the overall population of the United States, but is comparable to the rate reported for Americans of Asian/Pacific Islander backgrounds.