Central neurocytoma (CN) is a rare central nervous system tumor that occurs in adolescents and accounts for approximately 0.1 to 0.5% of intracranial tumors (1–4). Spinal cord dissemination is rare (5). In 1982, Hassoun identified the first two cases of CN, coining the term to describe intra-lateral ventricular tumors with typical features of neuronal differentiation (6). According to the 2006-2014 epidemiological survey of the Central Brain Tumor Registry of the United States (CBTRUS), the annual incidence of CN is 0.022 (0.021–0.024), and is slightly lower in men than in women (7). The majority of CNs (approximately 70%) occur in young people aged 20 to 40 years and are found in the third and lateral ventricles (3, 8). Patients initially clinically present with headache and visual disturbance caused by intracranial hypertension and obstructive hydrocephalus (9). CNs originate from neural stem cells with bidirectional differentiation potential around the ventricles. They are neuronal and mixed neuronal glial tumors which are classified as grade 2 in accordance with the 2021 World Health Organization classification (10). Although most CNs are well differentiated and generally have a good prognosis, some exhibit malignant biological behavior (11, 12).

Where possible, complete surgical resection is the preferred treatment modality for CN; otherwise, incomplete surgical resection followed by adjuvant radiotherapy (RT) or stereotactic RT to control any residual tumor and reduce tumor recurrence is considered (2, 9). With clinical, radiological, and pathological developments in recent years, our understanding of CN has improved; however, due to the rarity of CN, the current selection of treatment options is informed only by case reports, small-sample retrospective analyses, and meta-analyses, and there are no prospective studies with large samples and prolonged follow-up (2, 13, 14). Furthermore, the treatment options for recurrent tumors are still controversial (15, 16). Consequently, there is still much debate about surrounding the best way to treat this disease.

In this study, we collected clinical information from 101 patients with CN from 2010 until 2020 including basic characteristics, treatment patterns, and long-term survival outcomes. We then analyzed this information and reviewed the literature in order to gain insight into the characteristics of CN and identify optimal treatment modalities and prognostic factors.

In our study of 101 patients with CN, survival analysis showed that the 5- and 10-year OS rates were 88.7% and 82.8%, respectively, and the 5- and 10-year PFS rates were 86.5% and 64.9%, respectively. This 5-year PFS rate is excellent, and the 5-year OS rate is similar to results reported in previous studies (20–22). The largest meta-analysis to date was performed by Rades et al.; it included 438 patients and found 5- and 10-year OS rates of 91%, a 5-year PFS rate of 73%, and a 10-year PFS rate of 66% (22). Han’s study included 67 patients with CN, which represented the largest single-institution cohort before ours (20). In that study, the 10-year OS and PFS rates were 84.8 ± 6.1% and 79.8 ± 9.5%, respectively, with the latter value significantly different to that from our study. However, in Hallock et al.’s study (23), the 10-year OS rate was 82% and the 10-year local control (LC) rate was 61%, which are comparable to our values. Some of the major reports on survival rates for CN are summarized in Table 4 . Larger sample sizes and long-term follow-up are needed in the future.

Surgical resection is generally accepted as the first-choice treatment for CN. Not only can it remove the tumor and reduce the symptoms of intracranial hypertension caused by hydrocephalus, but it can also clarify the nature of the tumor pathology (26). For patients with CN, surgical resection can achieve a 5-year LC rate exceeding 70% (4, 27). In our study, all patients underwent surgery as the first treatment, and there was a statistically significant survival benefit in the GTR group as compared to the STR group in terms of 5-year OS and PFS. Many other studies have suggested that GTR plays a crucial role in the treatment of CN and is associated with a lower CN recurrence rate (17, 22). The feasibility of GTR is influenced by the location, depth, and size of the tumor, and it is not always a practical treatment strategy. However, complete surgical resection should always be performed when possible.

Transcortical or interhemispheric transcallosal approaches are usually used for CN (28, 29). Han et al. showed that there is no significant difference between these surgical approaches in terms of the effect on the patient’s memory (20). The transcortical approach avoids the parasagittal sinus vein, but disrupts the projection fibers of the frontal lobe. The transcallosal approach, with its high flexibility of exposure, is popular due to its smaller incision, but it also entails a higher incidence of disconnection syndrome (28). For large tumors, a transcortical approach is usually chosen; however, this carries the risk of seizures, hemiparesis, aphasia, and memory impairment. For small or medium-sized tumors, an interhemispheric transcallosal approach is usually chosen, which risks the occurrence of bridging vein injury or disconnection syndrome (30, 31). Many investigators prefer the transcallosal approach because of the smaller incision (20). Therefore, the tumor location and characteristics, minimization of blood loss, maximization of the extent of tumor removal, and minimization of complications should all be taken into account when determining the surgical approach to take.

Postoperative RT plays an important role in LC. Since complete resection is not always feasible and CNs tend to occur in young adults, adjuvant RT is now widely accepted as a treatment (2, 11). A meta-analysis by Rades et al. including 438 patients showed that ITR had significantly lower LC rates than CTR alone or in combination with other treatments (22). 5 and 10-year local control rates for ITR alone were 46% and 35%, respectively; however, ITR plus adjuvant RT increased the 5-year local control rate to 83% and the 10-year local control rate to 76%, which are comparable to the rates for CTR with or without RT (22). Several authors (27) have reported that postoperative RT has no effect on OS or PFS in patients who have undergone GTR, whereas it can significantly improve survival for patients who have undergone STR. In Han et al.’s study, there was no significant difference in OS between patients treated with GTR+RT and those treated with STR+RT (P=0.263) (20). A study from Mayo Clinic showed postoperative RT improved LC in patients with atypical CN with a high risk of recurrence (2). Our study produced similar results. In our study, there was no significant OS benefit in the GTR+RT group over the GTR alone group (P=0.602), but there was a significant OS benefit in the STR+RT group over the STR alone group (P<0.001). However, the OS (P=0.842) and PFS (P=0.915) of the STR+RT and the GTR without RT groups were comparable.

Regarding the effect of different RT technologies in the treatment of CNs, an international multicentric study did not observe any differences in survival for CNs associated with RT devices (21). Paek et al. suggest that more sophisticated radiation techniques should be used to reduce the subsequent sequelae in normal brain tissue (32). Modern RT techniques such as IMRT, volumetric arc radiation therapy (VMAT), and HT protect the surrounding normal tissue well, but HT increases low dose volume. The impact on the late toxicities in young people treated by HT needs to be evaluated in the long term. In our study, no grade 3 or 4 radiotoxicity was observed in patients who had been treated using modern RT techniques.

Adjuvant RT can lead to side effects such as memory impairment and secondary tumors (33, 34). Chen et al. reported that a common toxicity is temporary memory impairment (34). In our study, multivariate analysis showed that RT increased the risk of memory impairment (OR=3.871, 95% CI 1.028-14.577, P=0.045). However, many studies have shown that surgery is associated with a risk of memory impairment in patients with lateral ventricular tumors (35, 36). Schild et al. reported that the wait-and-see approach protects 50% of patients from the risk of RT toxicity while also leading to a risk of local treatment failure and progression, with a 5-year LC rate of 100% for patients treated with RT vs. 50% for patients not treated with RT (P<0.02) (4). Radiation as a carcinogenic factor has been clearly demonstrated (37, 38). The risk of radiation-induced tumors in central nervous system has been documented to be about 1-3% (37, 39). Basic criteria for radiation-induced tumor were established by Cahan et al. (40). A meta-analysis showed that the mean latency from RT to secondary tumors was 9 years (41). And radiation-induced tumors after SRS are rare. In our study, 1 patient was treated twice with SRS and the second pathological result was WHO grade IV tumor. It does not exclude radiation-induced secondary tumor after SRS.

In 1997, Schild et al. reported the first case of a patient treated with Gamma Knife SRS after STR, but no efficacy was reported (4). Compared to EBRT, SRS precisely concentrates single high doses of radiation to target and have the rapid radiation fall-off outside the target, leaving surrounding tissue exposed to low dose. SRS is usually finished in fewer fractions while EBRT treatment usually takes several weeks. SRS is suitable because CNs are inert and have clear boundaries with brain parenchyma (9, 42). A multicenter retrospective study showed that LC rates for SRS group, EBRT group were 93% and 88%, respectively (P=0.40) (9). Many studies have shown little neurotoxicity or complications after SRS treatment (43, 44). However, the smaller target area and the lower dose of SRS results in a higher risk of recurrence (9). In our study, 1 patient was treated by Gamma Knife SRS for the second time as salvage therapy. The survival period of 40 months suggests that SRS may be used as an adjuvant treatment after incomplete resection and as salvage therapy.

To date, there have been few reports on the optimal RT dose and delivery device for CNs (45). Rades et al. concluded that for patients treated with STR+RT, a radiation dose of 54 Gy appeared to be sufficient to improve LC, but not statistically significantly (15). In another report, in patients with typical CN, there was no significant difference in OS or LC between those who received a radiation dose of >54 Gy and <54 Gy, whereas for patients with atypical CN who underwent partial resection, the LC rate was better with a radiation dose >54 Gy (P=0.05). Therefore, patients with atypical CN undergoing STR were recommended to receive a radiation dose of 56–60 Gy (22). On the other hand, an international multicentric study showed no significant difference in OS and PFS for radiotherapy dose of 54Gy and greater than 54Gy (P>0.05) (21). In our study, the median EBRT dose was 54.54Gy (46-61.60Gy). Patients who received EBRT had no recurrence. Therefore, we suggest that a dose of 54 Gy may be appropriate for patients with CN.

The morphological appearance of CNs tumor cells under light microscopy is uniformly small, round cells of uniform size and shape, with visible calcification and a perinuclear halo around the nucleus due to the lack of cytoplasm. Immunohistochemistry is positive for synaptophysin (Syn), Neuronal Nuclei (NeuN), neuron-specific enolase (NSE), but generally does not express vimentin (Vim), oligodendrocyte transcription factor-2 (Olig-2), and a few Glial fibrillary acidic protein (GFAP). The pathological features of CN are shown in Figure 4 .

Although CNs are benign tumors, atypical CNs are more aggressive than typical CNs and tend to show a high propensity for recurrence due to their relatively malignant biological behavior. In our long-term follow-up, 1 patient relapsed and developed into a WHO grade IV malignant embryonic tumor, demonstrating its potential malignant biology. Atypical CNs have varying degrees of histological mitosis, vascular proliferation and local necrosis (17). Bertalanffy et al. reported a recurrence rate of 21% (8-43%) for CNs (46). Some studies have suggested that patients with atypical CNs have a worse prognosis and a higher recurrence rate (2, 47). Rades et al. consider proliferative potential as a prognostic factor (15). The most common thresholds for the clinical use of MIB-1 LI in classifying atypical and typical CNs remain controversial, including 2% (48), 3% (49), 4% (50). However, the survival outcomes of these studies (48–50) did not differ significantly between typical and atypical CNs. Consistent with these studies, our study showed that there was no statistically significant difference in OS (P=0.338) and PFS (P=0.277) between typical and atypical CNs.

Whether atypical CNs should receive more treatment remains a topic of debate. A meta-analysis of 310 patients with CN by Rades et al. showed that in patients with atypical CN who underwent STR with RT increased the 5-year LC rate from 5% to 65% and the 5-year OS rate from 46% to 69% (27). Several studies have demonstrated that RT exerts good control over the residual tumor (45, 51). The results of our study are in line with these findings (22, 24). We observed a significant improvement in the OS and PFS of patients with atypical CN who received RT after STR compared to those who received STR alone, and the 5-year PFS improved from 25% to 75% (P=0.004). Therefore, we believe that patients with atypical CN who can only undergo STR will benefit from postoperative RT.

Our multivariate Cox models and competing risk model showed that STR is an independent prognostic factor for poor prognosis in patients with CNs. Consistent with the data published previously, the extent of surgical resection is the most important prognostic factor (15). In our study, the extent of surgical resection was a confounding factor when the effects of RT were examined, and Model 3 showed that postoperative RT significantly improved OS (HR=0.145, 95% CI 0.028–0.746, P=0.021). Furthermore, postoperative RT was also an independent prognostic factor in our competing risk model (P=0.014). Many studies concluded that STR combined with postoperative RT is superior to STR alone, and that it shows no significant difference in OS compared to GTR (21, 22). By combining the two models, we concluded that postoperative RT was an independent prognostic factor affecting for survival.

This study is subject to some limitations. It is a retrospective study with inherent bias; consequently, there may be undetected confounding factors in the subgroup and prognostic analyses, which may affect the statistical models and the results of our analyses. However, there are also some clear advantages to this study. For instance, it is, to the best of our knowledge, the largest retrospective single-center study to date on the long-term survival of patients with CN in China. Also, sensitivity analysis was performed to verify the robustness of the results and evaluate the effects of RT. Therefore, our data will provide a valuable contribution to the existing literature and will support future meta-analyses of this rare tumor.

In summary, the first-choice treatment for CNs is complete surgical resection. The survival rate of incomplete resection followed by adjuvant RT is comparable to that of complete resection. The prognostic value of atypical compared with typical CNs require further investigation, and RT may have a survival benefit on prognosis.

CNs are a rare type of tumor. This study has analyzed the efficacy of treatment modalities and prognostic factors for CNs. Total surgical resection is recommended whenever possible, provided that side effects can be minimized. Adjuvant RT improves the prognosis of patients who have undergone subtotal resection, especially for atypical CNs, which have a high tendency for recurrence. Modern RT techniques have proven to be effective, and have acceptable levels of associated toxicity. Therefore, it is essential to weigh the pros and cons of available treatment modalities to develop optimal, individualized treatment plans for patients with CN. The results of this study will help to guide the treatment of CNs.

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

QX and RW are responsible for the conception and design; RW is responsible for the administrative support; BX, LO, MW, and ZT are responsible for the provision of study materials or patients; YH, XY, JH, and ZL are responsible for the collection and assembly of data; QXis responsible for the data analysis and interpretation; all authors are responsible for the manuscript writing; and all authors are responsible for the final approval of the manuscript. All authors contributed to the article and approved the submitted version

This study was supported by the grants from the Hunan Provincial Natural Science Foundation of China (NO. 2021JJ31104, NO. 2019JJ40497).

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.