Introduction
Central neurocytomas (CNs) of World Health Organization (WHO) grade II [
1] are uncommon tumors of the central nervous system and represent approximately 0.25-0.5% of all intracranial tumors. CN was described for the first time in 1982 by Hassoun and co-workers, and, by the early 1990s, it had become a well-defined clinical and pathological entity [
2]. CNs are typically located supratentorially in the lateral ventricle(s) and/or the third ventricle. CN used to be considered a benign tumor due to the excellent clinical outcomes after treatment. However, long-term follow-up showed that CNs had a higher recurrence rate than expected, even after complete resection [
3].
Because of the rarity of this tumor, no randomized clinical trials on CNs have been performed. Treatment strategies have been based on case reports and retrospective studies on small populations [
4]-[
6]. Surgery is the mainstay of treatment, and the extent of surgery is the most important prognostic factor affecting clinical outcomes [
7]. In addition, radiotherapy (RT) can successfully control residual tumor after incomplete resection or recurrence [
3],[
8]. It is important to investigate the role of adjuvant RT after resection, given the dismal local control rates in CN patients who undergo incomplete resection (IR). For example, in a study of 45 CN patients [
3], one-third of the patients developed recurrence after gross tumor resection (GTR). The recurrence rates were even worse in patients who underwent IR. Rades et al. [
9] conducted a meta-analysis of all the published data since 1982 from 91 centers and reported that the 10-year local control rate was only 35% in the 109 CN patients who received IR. The role of adjuvant RT after resection remains controversial, and the standard treatment for this tumor has not yet been established [
7],[
9]-[
14]. The main concern about the use of adjuvant RT in CN patients is that it may increase the risk of late neurotoxicity in these patients [
4],[
5]. However, neurotoxicity of RT in patients with CN has not been reported till now, and the concerns about radiotoxicity mainly come from the long-term toxicity observed in cases of other low-grade primary cerebral neoplasms like glioma [
15].
Here, we retrospectively investigated the long-term survival and late toxicities associated with adjuvant RT for neurocytoma; we hope that these findings will help guide oncologists in treatment decisions in such cases in the future.
Methods and materials
Patient population
Sixty-three patients with CN underwent surgical resection between January 2001 and October 2010 at Beijing Tiantan Hospital. All the patients received adjuvant RT after surgical resection at Beijing Tiantan Hospital, Beijing Shijitan Hospital or Beijing Puren Hospital. Institutional review board approval was obtained at these hospitals for retrospective review of the data. In all the patients, the CN diagnosis was confirmed by pathological tests. The dates of patients’ follow-up appointments were obtained through a review of hospital records, telephone interviews, and patient consultations.
Treatment
Of the 63 patients with CN, 24 underwent gross total resection (GTR); 28, subtotal resection (STR); 9, partial resection (PR); and 2, biopsy only (Bx). The extent of resection was assessed by contrast-enhanced MR within 2 weeks after surgery. Based on the surgeon’s intraoperative observation and the MR image, the extent of resection was classified as GTR (absence of a macroscopically visible residue), STR (less than 10% residue remaining) or PR (10-50% residue remaining). Patient selection for postoperative RT was mostly based on the surgeon’s discretion. Because there is no standard management procedure for this rare malignant tumor, a majority of surgeons at Tiantan Hospital preferred adjuvant RT to decrease the chances of recurrence. From the 63 patients, 38 (60.3%) were treated with conventional focal RT, and 25 (39.7%) with three-dimensional conformal RT (3D-CRT, n =18) or intensity-modulated radiation therapy (IMRT, n =7). Simulation was performed generally a month after surgical resection. The median interval between surgery and RT was 47 days (range, 13 to 160 days).
Tumor volume is best defined using pre- and postoperative imaging. The planning target volume (PTV) represented a 3-mm geometric expansion of the clinical target volume, which was an anatomically constrained 10-mm expansion of the gross-residual tumor and/or tumor bed. In the whole series, the median total dose was 54 Gy (range, 46-60 Gy), and it was delivered in 1.8-2 Gy daily fractions. All patients completed RT according to the schedule.
Follow-up
Follow-up ended on September 1, 2012. Fifty-two (82.5%, 52/63) patients underwent contrast-enhanced MRI of the brain every year after radiation, while eleven (17.5%, 11/63) patients discontinued the MRI scans after 1 to 4 years of follow-up. The median follow-up was 69 months, and the mean follow-up was 66 months (range, 15-129 months). Overall survival (OS) was calculated from the time of surgery until death or last follow-up, and progression-free survival (PFS) was calculated from the time of surgery until tumor recurrence or last follow-up. The date of progression was confirmed retrospectively based on the findings from the MRI scans or based on symptomatic deterioration consistent with tumor progression. Toxicity was scored retrospectively using the National Cancer Institute Common Terminology Criteria for Adverse Events version 4.0 [
16].
Statistical analysis
The Kaplan-Meier method was used to estimate the time from surgery to relapse and death, and the log-rank test was used to compare survival between groups. All analyses were carried out using the SPSS 17.0 statistical software, and p values less than 0.05 were considered statistically significant.
Discussion
To the best of our knowledge, this is one of the largest original studies investigating the long-term survival, local control and toxicity outcomes of adjuvant RT for CN. To date, data on the long-term outcomes of RT have been limited (Table
4). One meta-analysis [
9] did show that RT improved the 10-year OS and 10-year LC in patients who underwent IR; moreover, this study also showed that RT significantly improved the OS and LC in cases of both typical CN and atypical CN. However, the results of the published meta-analysis were limited by the fact that it was conducted on preexisting published data from many institutions. Our results suggested that the addition of adjuvant RT led to the comparable survival outcomes to those who underwent GTR.
Table 4
Findings of some large retrospective studies on central neurocytomas
Current study | 2013 | 63 | GTR + RT(24) | 96.6 | N/A | 96.5 | N/A | 69 mo | GTR + RT(1/24) |
STR + RT(28) | (15-129) | STR + RT(1/28) |
PR + RT(9) | | PR + RT(1/9) |
Bx + RT(2) | Bx + RT(0/2) |
| 2012 | 71 | GTR(43) | N/A | N/A | N/A | N/A | 48 mo | GTR(4/43) |
STR(13) | | | | | (6-204) | STR(9/28) |
STR + RT(4) | | | | | | |
STR + S(6) |
STR + CT(2) |
| 2011 | 19 | GTR(10) | N/A | 82 | N/A | 60 | 104.5 mo | GTR(1/10) |
STR(8) | | | | | (0.75-261.7) | STR(4/8) |
STR + RT(1) | | | | | | STR + RT(1/1) |
| 2007 | 45 | GTR(15) | 83 | 83 | 67 | 60 | 10 y | LR(15/45) |
STR(14) | | | | | (1.6-23.4) | GTR(5/15) |
GTR + RT(6) | | | | | | |
STR + RT(8) |
BX + RT(1) |
| 2006 | 20 | GTR(10) | N/A | N/A | N/A | N/A | 7 y (3-20) | GTR(1/10) |
STR + RT(2) | STR + RT(0/2) |
STR(1) | STR(0/1) |
PR(3) | PR(3/3) |
PR + RT(4) | PR + RT(3/4) |
| 2006 | 438 | GTR(152) | 99 | 99 | 84 | 74 | 44 mo | N/A |
GTR + RT(43) | 97 | 97 | 87 | 87 | (12-456) |
IR(109) | 82 | 82 | 41 | 35 |
IR + RT(134) | 89 | 89 | 83 | 76 |
In this study, only three patients had recurrences during follow-up. According to the published data, the extent of resection is the most important prognostic factor, local control is better after GTR than after IR [
14]. In some studies, high MIB-1proliferation index is also a prognostic factor which is associated with poorer local control [
18]-[
20]. Radiation is important factor that can retard residual tumor growth and improve local control [
14], but the effect of radiation dose to local control remained unclear and should be addressed. The local control rate in our study is excellent, even in patients who underwent IR and adjuvant RT. Tumor location seemed to be irrelevant to local recurrence, but involvement of periventricular parenchyma was associated with poor local control in some cases [
21]. In our study, one of three patients with the lateral ventricle and the third ventricle involvement had recurrence. It is hard to draw conclusion from such limited dada, but we guess that the extent of tumor invasion may affect the local recurrence and deserve further investigation. Other factors such as age and gender may not correlate to local recurrence based on the published data.
Neurotoxicity of RT in patients with CN has not been reported till now. Although Hallock et al. [
13] did report the first large study on toxicity outcomes in CN patients, mainly the sequelae of surgery, only one patient received adjuvant RT (1/19) in their study. Our results showed that surgery plus adjuvant RT produced minimal late toxicity: only seven (11.7%) and three patients (5%) had grade 2 and 3 late toxicity, respectively, and no grade 4 and 5 late toxicity was observed. The grade 3 late toxicity was caused by surgery in two patients and by cerebral infarction 5 years after RT in the third patient. Although vasculopathy with stroke are rare sequelae of irradiation [
22], it is still possible that they are related to adjuvant RT. Nonetheless, this is only a speculation, and this possible association should be studied further in future studies.
The most common late neurotoxicity was short-term memory impairment, which had no impact on the daily life and work of the affected patients. It is not clear whether this is related to RT, because the records of acute toxicity after surgery are incomplete. It has been reported that radiation doses to key structures such as the hippocampus may lead to long-term memory effects [
23],[
24]; however, dosimetric comparison of the group with memory impairment and the group without memory impairment showed that the mean and maximum doses applied to the hippocampus did not differ significantly. Therefore, it seems that it is unlikely that RT was the main cause of the memory deficit. Instead, some studies [
25],[
26] have shown that surgical resection has adverse effects on memory in patients in whom the tumor is located in the lateral ventricle and/or the third ventricle. Friedman et al. [
26] retrospectively analyzed the neuropsychological data of 33 patients who received surgical treatment for brain tumors in the third ventricle: 43% of the patients displayed particularly severe memory impairment. Moreover, Shi et al. [
27] showed that CN patients who underwent surgical resection showed impaired memory function in comparison with healthy individuals. In contrast, with regard to the effect of RT, most prospective studies [
28]-[
33] investigating radiation-induced neurotoxicity in low-grade glioma patients did not find a relationship between RT and cognitive impairment. Moreover, most studies that suggested a decline in cognitive function after RT were retrospective and often included a small number of patients [
34]. Therefore, based on the published findings, we believe that the late neurotoxicities observed in our study were predominantly caused by surgery. Taken together, the findings in our study imply that the adverse effects of adjuvant RT for CN patients are limited and acceptable. Thus, in patients for whom it is difficult to perform GTR, IR with adjuvant RT avoids the potential risk of complications associated with a radical surgical approach.
Based on some studies, another reason for defering RT in patients who undergo IR is that they may be successfully salvaged with surgery or radiation in case of disease progression, these authors suggested that careful surveillance alone after IR may be a safe strategy [
5],[
13]. However, other authors [
9] argued that both OS and LC were important endpoints. At the time of recurrence, initially benign neurocytomas may be associated with severe complications such as craniospinal dissemination and intraventricular hemorrhage [
35],[
36]. A second cranial operation appears to entail significant risk. Therefore, in our opinion, RT would be more effective if the tumor burden is minimal, and lower radiation dose and a limited radiation field are feasible at the time. The benefits of adjuvant RT seemed to overweigh its potential risks for CN patients after IR. Although it is hard to draw a conclusion based on the small sample size in this retrospective study, we believe that the role of adjuvant RT should be re-evaluated for these patients. Longer follow-up is needed to further investigate this issue.
For CN patients who undergo GTR, most authors agree that the clinical outcomes are favorable. Rades et al. [
9] reviewed 152 CN cases of GTR and found the 5-year OS and 5-year LC to be 99% and 84%, respectively; these results are in accordance with ours (94.4% and 95%, respectively). All these data suggest that GTR alone is sufficient for the treatment of most CN patients.
In our study, two patients who underwent biopsy and RT (one underwent surgery 6 months after RT) survived for 65 and 91 months without tumor progression. Similar to these cases, Leenstra et al. [
3] summarized a total of 11 cases in which the patients were treated with conventional RT alone after stereotactic biopsy; tumor relapse occurred in only three cases. These limited data suggest that RT may be the treatment of choice for CN patients who undergo biopsy and do not require resection.
The optimal target volume of adjuvant RT for CN remains unclear because the data for this are scarce. Following the generally accepted guidelines for low-grade brain tumors, many institutions adopt a target volume that includes the preoperative tumor volume and a 2-cm margin [
37]. Since CNs are indolent tumors and are mostly confined to the lateral ventricles without invasion to the surrounding brain parenchyma, we think that the target volume maybe somewhat different from that of other low-grade brain tumors. In recent years, we have performed conformal irradiation with a 10-mm clinical target volume margin in order to reduce the adverse effects of RT. Although the follow-up time was relatively short, no recurrence was observed in our patients. In the most recent largest prospective study [
33] addressing the late effects in 78 pediatric patients with low-grade glioma treated with 3D-CRT, a 10-mm clinical target volume margin was routinely used; the results showed that the adverse effects were limited and predictable for most patients. Moreover, the 10-year OS and 10-year event-free survival were 96% and 75%, respectively. According to these limited data, we think that a 10-mm clinical target volume margin may be sufficient for CN patients who undergo 3D-CRT.
The optimal radiation dose for CN patients also remains unclear. The study by Rades [
37] is the only one that investigates the appropriate radiation dose after IR; according to this study, a dose of ≥54 Gy significantly improves local control in patients with subtotally resected neurocytomas. However, in contrast to these findings, in our study, most of the patients received a total dose of 50-54 Gy, and higher doses (≥54 Gy) were not found to have an impact on clinical outcomes. Thus, a total dose of 50-54 Gy may be appropriate for CN patients who undergo IR.
One of the limitations in our study is that the MIB-1 labeling index of CN was not routinely examined at Tiantan Hospital in the past. The MIB-1 labeling index for this tumor is usually low, less than 2%, and tumors with indices greater than 2% or 3% have been referred to as “atypical neurocytomas” and are associated with a significantly shorter recurrence-free interval [
18]-[
20]. However, a recent large multicenter study of 71 cases by Vasiljevic and co-workers [
7] reported that histologic criteria may not be reliable markers for “atypical” CNs, and that the extent of surgery was the main prognostic factor. Further, the long-term clinical results of 58 CN patients also challenged previous reports that the MIB-1 labeling index is related to the risk of recurrence [
38]. In light of these reports, it seems that more reliable criteria are needed to define atypical groups of CNs.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made.
The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.
Competing interests
The authors declare that they have no competing interests.