Treatment of atypical neurocytomas

Abstract

BACKGROUND

The majority of central neurocytomas are benign. Approximately 25% of these rare central nervous system tumors are more aggressive, with an MIB-1 labeling index > 2% or atypical histologic features, and are classified as atypical neurocytomas. The objective of this analysis was to define the optimal treatment for patients with these atypical tumors.

METHODS

The first atypical neurocytoma was described in 1989. The patients reported since then were reviewed for age, gender, extent of resection, MIB-1 index, histology, irradiation, local control, and survival. In addition to the data available in the literature, more relevant data were obtained from the authors of that literature. Treatment approaches included complete resection alone (CR), complete resection followed by radiotherapy (CR-RT), incomplete resection alone (IR), and incomplete resection followed by radiotherapy (IR-RT). These four therapies were compared for local control and survival using Kaplan–Meier analysis and the log-rank test.

RESULTS

Complete data were available on 85 patients (15 patients who underwent CR, 13 patients who underwent CR-RT, 17 patients who underwent IR, and 40 patients who received IR-RT). Local control rates at 3 years and 5 years were 73% and 57% after CR, 81% and 53% after CR-RT, 21% and 7% after IR, and 85% and 70% after IR-RT, respectively (P < 0.0001). Survival rates at 3 years and 5 years and were 93% and 93% after CR, 90% and 90% after CR-RT, 65% and 43% after IR, and 87% and 78% after IR-RT, respectively (P = 0.0076).

CONCLUSIONS

Patients who underwent CR achieved better local control and survival rates compared with patients who underwent IR. After IR, patients appeared to benefit from RT. A beneficial effect of RT after CR was not observed. Cancer 2004;100:814–7. © 2004 American Cancer Society.

Neurocytomas are rare tumors of the central nervous system and usually are considered benign. In 1989, a more aggressive variant of this tumor called atypical neurocytoma was reported.1 Neurocytomas were classified as atypical if they exhibited an MIB-1 labeling index > 2%2 or atypical histologic features, such as focal necrosis, vascular proliferation, and increased mitotic activity.3

For patients with typical neurocytomas, complete resection is considered the best treatment; and, after incomplete resection, patients benefit from radiotherapy.4–6 To our knowledge, little is known regarding the treatment of atypical neurocytomas. The current study was performed to determine the optimal therapy for patients with these more aggressive tumors.

MATERIALS AND METHODS

A Medline analysis was performed for the term neurocytoma. All patients with neurocytoma reported since 1989, when the first atypical neurocytoma was described, were reviewed for age, gender, histology, MIB-1 labeling index, extent of resection, radiotherapy, local control, and overall survival. If the published data were incomplete regarding these parameters, then the authors were contacted for additional data.

This analysis included patients with tumors that exhibited an MIB-1 labeling index > 2% or atypical histologic features, such as focal necrosis, vascular proliferation, and increased mitotic activity, and, thus, were classified with atypical neurocytomas. The patients were divided into four groups according to therapy: complete tumor resection alone (CR), complete resection followed by radiotherapy (CR-RT), incomplete resection alone (IR), and incomplete resection followed by radiotherapy (IR-RT). The minimum follow-up was 12 months.

The four treatment approaches were compared for local control and survival. Statistical analysis was performed using Kaplan–Meier analysis7 and the log-rank test.

RESULTS

Eighty-five patients, 75 patients from the literature1–3, 8–37 and 10 patients from our institutions, met the criteria for inclusion in this analysis: 15 patients who underwent CR, 13 patients who underwent CR-RT, 17 patients who underwent IR, and 40 who underwent IR-RT. Patient characteristics are summarized in Table 1.

Table 1. Patient Characteristics

Characteristic	CR	CR-RT	IR	IR-RT
CR: complete tumor resection; RT: radiotherapy; IR: incomplete tumor resection.
Age (yrs)
Median	22	27	29	30
Range	18–66	8–57	16–47	5–76
Gender: No. (%)
Female	10/15 (66)	4/13 (31)	5/17 (29)	16/40 (40)
Male	5/15 (33)	9/13 (69)	12/17 (71)	24/40 (60)

During the period of follow-up, local failure occurred in 34 of 85 patients (40%), and tumor-related deaths were observed in 17of 85 patients (20%). The Kaplan–Meier curves7 demonstrated that CR, CR-RT, and IR-RT were associated with better local control and survival rates compared with IR (Figs. 1, 2). The 3-year and 5-year local control rates were 73% and 57% after CR, 81% and 53% after CR-RT, 85% and 70% after IR-RT, and 21% and 7% after IR, respectively (Fig. 1) (P < 0.0001). The survival rates at 3 years and at 5 years were 93% and 93% after CR, 90% and 90% after CR-RT, 87% and 78% after IR-RT, and 65% and 43% after IR, respectively (Fig. 2) (P = 0.0076).

Figure 1. Local control according to Kaplan–Meier analysis7 after complete tumor resection alone (CR), complete resection followed by radiotherapy (CR-RT), incomplete resection alone (IR), and incomplete resection followed by radiotherapy (IR-RT) (P < 0.0001 for comparison of the 4 curves; log-rank test).

Figure 2. Overall survival according to Kaplan–Meier analysis7 after complete tumor resection alone (CR), complete resection followed by radiotherapy (CR-RT), incomplete resection alone (IR), and incomplete resection followed by radiotherapy (IR-RT) (P = 0.0076 for comparison of the 4 curves; log-rank test).

Patients who underwent IR were evaluated separately from patients who underwent CR to evaluate better the effects of RT on each group. In patients who underwent IR, IR-RT was associated with a better outcome compared with IR alone in terms of both local control and survival. In patients who underwent CR, the comparison between CR and CR-RT did not reveal a significant difference for either local control (P = 0.46) or survival (P = 0.96).

Of the patients included in the analysis, 25% (21 of 85 patients) had extraventricular cerebral tumors. In those 21 patients, the local control rates at 3 years and at 5 years were 53% (vs. 67% for intraventricular lesions; P = 0.647; chi-square test) and 18% (vs. 39%; P = 0.454), respectively. The survival rates at 3 years and at 5 years were 57% (vs. 85%; P = 0.390) and 22% (vs. 67%; P = 0.188), respectively.

Among the patients who underwent IR, 4 patients could be identified in whom the extent of resection was < 25% of the tumor volume. In those patients, the 3-year and 5-year rates of local control were 25% and 0%, respectively, compared with 60% and 31% for patients who had greater degrees of IR (P = 0.578 and P = 0.538, respectively). The 3-year and 5-year survival rates were 25% and 0% for patients who had an extent of resection < 25% of the tumor volume, compared with 78% and 57% for patients who underwent more complete (although incomplete) resection (P = 0.386 and P = 0.498, respectively).

At the time of local failure, the type of salvage treatment was known in 12 patients. Five patients received RT without surgery, and six patients underwent surgery alone. The crude 3-year survival rates for these patients were 75% (3 of 4 patients) after RT and 80% (4 of 5 patients) after surgery.

DISCUSSION

The current analysis was performed to define the optimum treatment for the aggressive variant of central neurocytoma, termed atypical, by comparing four different therapeutic approaches for local control and survival. Data were obtained from the literature and from contact with most of the authors. This approach provided a more detailed analysis and a longer follow-up than obtaining the data from the literature alone.

Biases may have been introduced because of the retrospective nature of this analysis. In addition, the wide time frame of therapy (ranging from 1989 to 2002) and the multitude of institutions that contributed data also may have introduced biases. Furthermore, the cases included in this analysis did not have a central pathologic review. Thus, the possibility of misclassification cannot be excluded completely. Therefore, caution should be used in interpreting these results. However, this methodology appeared to be the only possible way to accumulate such a large series of patients with this exceedingly rare tumor.

According to the results of the current study, CR achieved better local control and survival compared with IR. After undergoing IR, both local control and survival appeared to be improved with postoperative RT. In patients who underwent CR, no local control or survival advantage could be demonstrated for postoperative RT. Therefore, we would not recommend RT after CR for atypical neurocytoma.

For the location of the tumor (extraventricular vs. intraventricular) and for the degree of resection in patients who underwent IR (< 25% resection vs. other degrees of IR), no significant impact on local control or survival was observed. Regarding the degree of IR, this may have been due to the small numbers of patients who underwent < 25% resection of their tumor volume.

Screening for atypical histologic features and indications of the MIB-1 labeling index are essential to identify the atypical central neurocytomas. This is significant because the prognosis is worse for patients who have these atypical tumors compared with patients who have typical central neurocytomas. The beneficial effect of postoperative RT after IR appears to be more striking for patients who have atypical central neuromas compared with patients who have typical central neurocytomas. This analysis demonstrates that, after undergoing IR of atypical lesions, RT significantly improves both local control (70% vs. 7% at 5 years) and survival (78% vs. 43% at 5 years). After undergoing IR of central neurocytomas, in general, with almost 75% typical lesions, RT affected local control (83% vs. 46% at 5 years), but not survival (90% vs. 86% at 5 years).4, 5

The type of treatment at the time of local failure was known in 12 patients. Thus, it is very difficult to make specific recommendations. However, reresection can be considered if it can be performed safely, and RT can be considered especially if it was not administered at the time of primary therapy.

To our knowledge, only very little experience has been published to date concerning chemotherapy in the management of patients with atypical central neurocytomas.9, 24, 33 Thus, the value of chemotherapy remains uncertain.

In the current study, CR was found to be the best treatment for patients with atypical central neurocytomas and achieved better local control and survival rates compared with IR. However, in those patients who could not undergo CR safely, postoperative RT appeared to improve both local control and survival.

Acknowledgements

The authors thank the contacted authors, who provided additional data, for their kind cooperation and support.