Treatment following neurosurgical resection or biopsy of a central neurocytoma depends on the pathologic analysis and proliferative activity demonstrated, the extent of surgical resection, the neuroanatomic location and accessibility of residual tumor, and the patient’s postoperative condition. The risks and benefits of additional surgery, radiation therapy, and/or chemotherapy should be weighed and individualized based on these factors and guided by the existing literature.
Key points
- •
Increased proliferative activity (as indicated by the MIB-1 index) as seen on histopathologic specimens may indicate a higher risk of recurrence.
- •
Pharmacologic treatment measures are not well described and would not be considered a standard of care based on reported outcomes.
- •
Surgical treatment is a mainstay for well-differentiated central neurocytomas; with subtotal resection or recurrence, operative resection and radiation-based therapies should be considered.
- •
Stereotactic radiosurgery may be equivalent and may be better tolerated for recurrent/residual neurocytoma in carefully selected patients.
Introduction
Central neurocytomas (CNs) are World Health Organization grade II tumors restricted to regions within and adjacent to the cerebral ventricular system. Well-differentiated CNs are generally associated with a favorable prognosis following gross total resection (GTR), with survival nearing 100% at 10 years from the time of resection. Adjuvant therapies continue to be studied and have contributed to high reported rates of local control. The rate of recurrence of all CNs, however, is estimated at between 20% and 25% with long-term follow-up (>12 years) and may reflect the highly variable proliferative rates observed in patients with these tumors. Over time central neurocytomas may transform, developing aggressive phenotypic behaviors; these have been described as atypical CNs.
Treatment following neurosurgical resection or biopsy of a central neurocytoma depends on several factors. The pathologic analysis of the tissue resected, the amount of tumor resected, the neuroanatomic location and accessibility of residual tumor, the availability of and skill in utilizing adjunctive therapies, and the patient’s condition all direct the course of therapy.
Introduction
Central neurocytomas (CNs) are World Health Organization grade II tumors restricted to regions within and adjacent to the cerebral ventricular system. Well-differentiated CNs are generally associated with a favorable prognosis following gross total resection (GTR), with survival nearing 100% at 10 years from the time of resection. Adjuvant therapies continue to be studied and have contributed to high reported rates of local control. The rate of recurrence of all CNs, however, is estimated at between 20% and 25% with long-term follow-up (>12 years) and may reflect the highly variable proliferative rates observed in patients with these tumors. Over time central neurocytomas may transform, developing aggressive phenotypic behaviors; these have been described as atypical CNs.
Treatment following neurosurgical resection or biopsy of a central neurocytoma depends on several factors. The pathologic analysis of the tissue resected, the amount of tumor resected, the neuroanatomic location and accessibility of residual tumor, the availability of and skill in utilizing adjunctive therapies, and the patient’s condition all direct the course of therapy.
Pathologic considerations
Well-Differentiated Central Neurocytomas: Histopathology and Histogenesis
CNs exist as neuronal and glial entities. Histopathologic specimens show a uniform tumor cell population with clear cells and round nuclei within a delicate microvascular stroma. Large fibrillary areas may mimic rosettes. Calcifications are often noted. Given the morphologic similarities existing between oligodendrogliomas and neurocytomas on light microscopy, neuronal lineage-specific features may be used to assist in diagnosis. Immunohistochemistry and electron microscopy may identify neuronal cytoskeletal proteins (neurofilament), neuronal cell adhesion molecule (NCAM), microtubule-associated protein (MAP2 and tau), synapsin, and synaptophysin, of which synaptophysin is the most reliable diagnostic marker. These markers are identifiable in both primitive and differentiating neuronal cell lineages. The presence of NeuN, a nuclear protein, can be identified by immunohistochemistry in almost all cases. CNs have been found to have high levels of γ-aminobutyric acid (GABA), a neuronal marker. On the other hand, astrocytic differentiation has been observed and glial fibrillary acidic protein immunoreactivity has been described in 10% to 12% of patients, based on immunohistochemical and culture data. Evidence of neuronal (primitive and differentiating) and glial lineages, in addition to the ventricular/periventricular location of these tumors, have led investigators to suggest the subventricular zone or circumventricular organs as sites of histogenesis.
Atypical Histologic Features and Proliferative Index
Highly proliferative CNs are generally considered to be those that have an MIB-1 labeling index (MIB-1 LI) greater than 3%, and histologic features such as mitotic figures, focal necrosis, and vascular proliferative changes distinguish atypical CNs (WHO 1214, 2335, 2336). Atypical histologic features (necrosis, vascular proliferation, increased mitotic activity) have been associated with an increased rate of local failure, although evidence contrary to this has been reported. The association between proliferative activity and local recurrence is similarly mixed in the current literature. However, meta-analysis data from 129 patients performed in 2004 examining the prognostic utility of the MIB-1 LI in patients with neurocytomas demonstrated that an MIB-1 LI greater than 3% was significantly associated with lower rates of local control and survival. Table 1 outlines studies examining MIB-1 LI and recurrence with respect to CNs. Together these data seem to indicate an association between proliferative activity and clinical response to therapy. In terms of surgical treatment, there is no clear association between proliferative activity of a tumor and ease of complete surgical resection.
| Authors, Ref. Year | N | MIB-1 LI (%) | Intervention | Outcomes | Notes |
|---|---|---|---|---|---|
| Kuchiki et al, 2002 | 2 | 7–7.8 | Surgery + RT | No recurrence noted after 16 and 15 mo follow-up (despite elevated MIB-1 LI) | Histologic features may not predict recurrence. Patients treated with 50 Gy RT given MIB-1 LI |
| Kaur et al, 2013 | 10 | Mean 5.4 recurrent, 1.5 nonrecurrent | Surgery, Surgery + RT/GK, Surgery + chemotherapy | MIB-1 LI >4% provided statistically significant cutoff above which recurrence was 100% | 2-y and 4-y recurrence rates in patients with MIB-1 LI >4% were 50% and 75%, respectively. In patients with STR and MIB-1 LI <4%, no recurrence |
| MacKenzie, 1999 a | 15 | 0.1–6 | Surgery, Surgery + RT | All patients with poor outcome (4) had MIB-1 LI >2%, only 1 had histologic atypia | MIB-1 useful as a predictor of clinical outcome. Advocates classification based on proliferative potential vs histologic, qualitative findings |
| Genc et al, 2011 | 18 | 0–5.7 | Surgery + GK | Reduction in tumor volume in 15 patients following GK | High MIB-1 did not affect efficacy of GK based on volumetric reduction on serial imaging after GK |
| Chen et al, 2008 | 9 | <0.1–6.8 | Surgery | Higher MIB-1 LI may predict recurrence | |
| Lenzi et al, 2006 | 20 | Surgery + RT | MIB-1 LI >4% correlated with unfavorable outcome (death or worsening neurologic status) | Histologic atypia significant negative risk factor for survival ( P = .02) | |
| Bertalanffy et al, 2005 | 14 | 0.8–11 | Surgery + GK | GK effective for patients with recurrence after surgery | MIB-1 LI ≥2% in all but 1 disease recurrence. Also, MIB-1 LI in nonrecurrent tumors ≥2% in all but 1 tumor. Tumors have a higher tendency to recur during long-term follow-up even after GTR. MIB-1 LI may not be a reliable in this subset; pathologic methods differ making institutional comparisons difficult |
| Rades et al, 2004 b | 129 | MIB-1 LI >3% associated with statistically significant worse prognosis for local control and survival | Meta-analysis | ||
| Fujimaki et al, 1997 | 10 | <0.1–5.6 | Surgery + RT (50–60 Gy) | Central neurocytomas have relatively high proliferative potential. Radiation therapy provided good control of residual tumors | RT resulted in control (5 patients), lowered proliferative activity in 1 patient (5.6%–2%) |
| Söylemezoglu et al, 1997 a | 36 | 0.1–8.6 | Surgery | 22% recurrence in patients with MIB-1 LI <2%; 63% recurrence in patients with MIB-1 LI >2%. Highly significant difference in disease-free survival between groups (lower in proliferative group) | MIB-1 LI >2% correlated with presence of vascular proliferation |
| Christov et al, 1999 | 1 | Surgery | MIB-1 LI 0.7% following first operation (GTR); MIB-1 LI 3.9% found at time of recurrence | No atypical histologic features at time of first operation; recurrence noted after 9-y disease-free interval | |
| Favereaux et al, 2000 | 10 | Surgery | MIB-1 LI <2.3% in 8 patients; MIB-1 LI >5.2% in 2 patients; these 2 patients had atypical features on histology and recurred | ||
| Li et al, 2012 | 11 | 0.8–12.5 | Surgery | MIB-1, GFAP positivity, vascular proliferation may suggest malignant phenotype. 6/11 patients with no recurrence | |
| Sharma et al, 2006 | 20 | 0.1–3 | Surgery + RT | Patients with higher MIB-1 tended to have mitoses and foci of necrosis noted on histopathologic specimens | |
| De Tommasi et al, 2006 | 2 | <1%–1.7% | Surgery | No recurrence at 32 mo |
a Data included in meta-analysis.
Differential Diagnosis
It is important to rule out other entities before establishing a diagnosis of recurrent or residual CN. Oligodendrogliomas bear similarities to central and extraventricular neurocytomas on light microscopy. With the latter entity, it is especially important to differentiate the 2 types of tumors. Immunostaining for synaptophysin will often identify CNs. Oligodendrogliomas can harbor 1p/19q codeletions and mutations of the TP53 oncogene but, importantly, the absence of these genetic alterations does not exclude the diagnosis. Although oligodendrogliomas have been shown to harbor 1p/19q codeletion along with neurocytic changes, this seems to be rare. Ultrastructurally, clear cell ependymomas and dysembryoplastic neuroepithelial tumors (DNETs) can mimic CNs. The ependymal tumors have perivascular pseudorosettes seen on microscopy, and show immunoreactivity to epithelial membrane antigen; both DNETs and ependymomas lack 1p/19q chromosomal losses.
Patient evaluation
Clinical Presentation
Patients most often present with symptoms from mass effect and/or hydrocephalus. Intratumoral hemorrhage has been reported ; bleeding diatheses, a need to take antiplatelet or anticoagulant medications, and/or the presence of residual tumor may affect postoperative management. Persistent symptoms of headache, visual changes, endocrinopathies (from hypothalamic impingement), or personality changes may affect the urgency and purpose of postoperative management strategies.
Radiologic Studies
The postoperative appearance of recurrent or residual tumor depends on the timing and quality of imaging obtained. On magnetic resonance (MR) imaging, central neurocytomas appear as lobulated, well-circumscribed tumors with heterogeneous to marked enhancement on the administration of gadolinium. The presence of cysts and calcifications (which are present in as many as 50% of tumors) can decrease the avidity of enhancement seen. Of interest, the level of enhancement has been shown to be associated (without statistical significance) with elevated MIB-1 LI. CNs have a hypointense appearance on T1-weighted sequences and are variable on T2-weighted sequences. Recurrent or residual tumor may not appear well circumscribed. The presence of blood products or surgical hemostatic material may complicate the interpretation of postoperative imaging studies. Table 2 summarizes radiologic adjunctive studies that may be useful in diagnosing recurrent or residual tumor.
| Modality | Characteristic Finding |
|---|---|
| H-MRS | High choline peaks compared with NAA or creatine phosphate |
| Tl-SPECT | Significant thallium uptake not relating to MIB-1 index |
| FDG-PET | Typically low, but may show increased uptake in tumors with elevated MIB-1 index |
Location of Recurrent or Residual Tumor
The location of recurrent or residual tumor affects the chosen treatment modality and its efficacy. CNs occur most commonly in regions near the lateral ventricles, but cases have been described in the third and fourth ventricles. Dissemination throughout the cerebrospinal fluid is rare but has been described. Periventricular parenchymal involvement may complicate decision-making regarding adjunctive therapy; localization here has been associated with poor outcome (1103, 1896 WHO).
Pharmacologic treatment options
Dodds and colleagues published the first report describing the use of chemotherapy in treating central neurocytoma in 1997. A 15-year-old boy with a large CN was treated via a transcallosal approach and debulking. Increased mitotic activity was noted, and the patient was started on carboplatin, ifosfamide, and etoposide given unfavorable proposed radiation planning. MIB-1 LI was not available. The patient again underwent surgical debulking followed by radiation therapy (RT). A retrospective analysis of 95 patients diagnosed with oligodendrogliomas revealed 3 patients with CNs based on radiologic and pathologic review. Two of these patients (both of whom did not receive radiation postoperatively) suffered early recurrence. Neither patient had atypical histologic features. One received carboplatin and had an observed decrease in the tumor size. A report described the use of procarbazine, lomustine, and vincristine in a 20-year-old female patient who had previously underwent 4 craniotomies within 3 years to resect a central neurocytoma. Pathologic assessment demonstrated an MIB-1 LI less than 1.0%, and there were no atypical features noted on histopathologic specimens. She had a small focus of residual tumor, which remained stable 16 months following treatment. A series of 3 heterogeneous patients treated in Italy demonstrated responses to a regimen of cisplatin, etoposide, and cyclophosphamide. Although stabilization of disease was noted with remission in 1 patient, 2 of the patients received postoperative radiation (1 received stereotactic radiosurgery [SRS] and the other received fractionated RT), thus confounding the results. Detailed pathologic analysis was not performed, but 1 patient had cerebrospinal fluid (CSF) dissemination. The investigators noted significant toxicity from the chemotherapeutic regimen. In 2008, complete response to a 3-agent regimen (topotecan, carboplatin, ifosfamide) was reported in a 5 year-old with recurrent central neurocytoma and drop metastasis. A recent case report demonstrated treatment with vincristine and carboplatin with a recurrence-free interval of 18 months in a 3-year-old with craniospinal dissemination of a central neurocytoma.
At the time of writing, chemotherapy does not show a clear benefit in treating recurrent or residual CN. Given that the isolated cases described are disparate and pathologic analysis is lacking, it is difficult to ascertain exactly which patients may benefit from chemotherapy. Based on anecdotal evidence, children or patients with widespread disease may benefit, especially if they have contraindications to RT.
Radiation-based treatment options
Radiation Therapy
RT has been described as an adjunctive treatment following resection of de novo CNs; suggesting radiosensitivity. Most of the studies reported herein describe outcomes based on conformal and intensity-modulated RT, but fractionated RT has also been used as an adjunctive treatment. Postoperative RT following incomplete resection has been supported, although long-term concerns about RT including neurocognitive effects must be considered, particularly in a developing patient. In such cases, these risks may not justify prophylactic RT following first-time resection of a well-differentiated central neurocytoma resected entirely.
Although disputed, the efficacy (in terms of local control and survival) of RT seems to rely on the tumor’s pathologic status as demonstrated following surgical excision or biopsy. Atypical CNs seem to be a different entity in regard to response to radiation treatment. Based on meta-analysis data, tumors with atypical features and elevated MIB-1 LIs were shown to have significantly improved rates of local control and survival with postoperative RT in comparison without RT. With CNs lacking atypical features (well-differentiated tumors), the same pattern has been observed in terms of local control but not with survival. In comparing 5-year local control rates between patients with subtotal resections with radiation and those with subtotal resections and no radiation, there was a nonsignificant but improved local control rate in those patients receiving RT.
With the exception of meta-analysis data looking at RT as an adjuvant therapy, there is limited statistical power with respect to the use of RT in treating recurrent CN. The best available data suggest that RT may have a role in the treatment of residual and recurrent tumors, but not necessarily for well-differentiated tumors resected completely. The presence of atypical or malignant pathologic features supports a decision to treat residual tumor with RT. A radiation dose of 54 Gy in the postoperative setting has been recommended for well-differentiated tumors, whereas atypical neurocytomas may benefit from a higher therapeutic dose in the range of 56 to 60 Gy. Table 3 lists studies describing nonadjuvant measures to treat recurrent tumors.
| Authors, Ref. Year | 1° | EOR | 2° | Radiosurgery/Radiation Dose | N | MIB-1 LI % of Recurrent or Residual Tumor | Outcome Assessment | Outcome | Follow-up (mo) | Adverse Outcomes | Comments |
|---|---|---|---|---|---|---|---|---|---|---|---|
| Park et al, 2012 | Surgery | N/A first resection; GTR in 3/3 second resection | Surgery | N/A | 3 | N/A | N/A | N/A | N/A | N/A | Characteristics of 3 patients undergoing secondary operations not discussed specifically. Outcomes analysis includes those having primary resections. Two recurrent tumors had MLB-1 LI >2% |
| Genc et al, 2011 | Surgery, biopsy | N/A | GK | 16.4 Gy (12–22) Margin | 18 | <1–5.7 | Rad | 6 tumors stable, 15 decreased in size, 1 increased in size (included other pathology) | 36.7 | None | For STR, recommended treatment with GK, especially if MIB-1 LI >2%. For GTR, close follow-up advised. Five patients with 2 prior operations |
| Yen et al, 2007 | Surgery, biopsy | GTR; STR | GK | 9–20 Gy Margin | 6 | N/A | Rad | 4/9 tumors treated disappeared, 4 shrank significantly | 60 | None | Continued surveillance for recurrent/residual tumors |
| Javedan et al, 2003 | Biopsy (endoscopic) | STR | GK | 18 Gy Margin | 1 | N/A | Rad | Minimal decrease in size of tumor | 25 | None | |
| Tyler-Kabara et al, 2001 | Surgery, biopsy | STR | GK | 14–20 Gy Margin | 4 | N/A | Rad | Decreases in size of tumor | >36 | None | |
| Anderson et al, 2001 | Surgery | GTR | GK | 16 Gy (16–20) Margin | 4 | N/A | Rad | Decreases in size of tumor | 14–28 | None | GK performed median 17.5 mo after surgery; stated no pathologic difference between recurrent tumors |
| Pollock & Stafford, 2001 | Surgery | GTR | GK | 18 Gy Margin | 1 | N/A | Rad | Decreases in size of tumor | 34 | None | |
| Cobery et al, 2001 | Surgery | STR | GK | 9–10 Gy Margin | 4 | N/A | Rad | Decreases in size of tumor | 12–99 | None | |
| Bertalanffy et al, 2001 | Surgery | N/A | GK | 9.6–13 Gy Margin | 3 | 2.4–8.7 | Rad | Decreases in size of tumor | 12–60 | None | |
| Kim et al, 2013 | Surgery and GK (primary) | N/A | GK (secondary) | 15.4 Gy (9–20) Margin | 10 | N/A | Rad | Local control | Mean 100 | Peritumoral edema noted 6 mo following retreatment. Resolved | Ten patients received GK as a secondary treatment (recurrent cases). Four patients with continued disease with “out of treatment field” recurrences |
| Karlsson et al, 2012 | Surgery and GK (primary) | N/A | GK (secondary) | N/A | 1 | N/A | Rad | Local control | N/A | N/A | Two patients with local recurrences out of 42 patients following GK. 1 retreated with GK, other patient refused treatment |
| Matsunaga et al, 2010 | Surgery | N/A | GK | 13.9 Gy (12–20) Margin | 1 | N/A | Rad | Decreases in size of tumor | N/A | Progression with ICH: 1 patient | Unclear dates of treatment (treated 1–50 mo following surgery); difficult to differentiate adjunctive SRS vs SRS for delayed recurrence. Investigators recommend GK with early detection of recurrence |
| Rades et al, 2005 | Surgery | STR | Surgery + fractionated RT | 50.4 Gy Total | 1 | 5.40 | Recurrence-free survival | 8 years without recurrence | None | Meta-analysis (2005) with case report. Analysis revealed 60 patients with recurrences. Treated with surgery (21), surgery + RT (7), surgery + chemotherapy (4), SRS (9), RT alone (7), chemotherapy (2), no treatment (5). Patients with no treatment died within 3 mo. Deaths in surgery alone (1 patient, 11 mo) and RT alone (1 patient, 14 mo) groups | |
| Kim et al, 1997 | Surgery | GTR; STR | Surgery + fractionated RT | 50.4 Gy Total | 2 | N/A | N/A | N/A; death after unspecified period due to tumor progression | 23–26 | Death | Lacking in pathologic data, no MIB-1 LI. Unclear follow-up in recurrent cases |
Related posts:
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
