Pleomorphic xanthoastrocytomas (PXAs) are rare brain tumors which often occur in children and young adults (1–3). In 1999, Giannini et al. defined ‘PXA with anaplastic features’ as PXA exhibiting increased mitotic activity, >5/10 HPF mitotic figures with or without accompanying necrosis (4). ‘PXA with anaplastic features’ are labelled as anaplastic pleomorphic xanthoastrocytoma, according to the 2016 World Health Organization Classification of Tumors of the Central Nervous System (3). In the fifth edition of the WHO Classification of Tumors of the Central Nervous System, the term “anaplastic pleomorphic xanthoastrocytoma” is no longer listed, however, according to histopathological features of PXA, this entity can still be classified into WHO grade 2 or WHO grade 3 tumors (5). The clinical prognosis of patients with PXAs is usually satisfactory, with a 5-year survival rate of 80% (6). However, patients with HGPXAs have a worse clinical outcome than patients with PXAs (7). Due to the rarity of the tumor, their management and prognostic outcomes remain unclear. In this study, we reported 26 cases with HGPXAs in our institute and performed a pooled analysis of individual data (including cases from our institute and 56 cases from the literature) to propose a treatment protocol.

We performed a retrospective analysis of 26 cases of high grade pleomorphic xanthoastrocytomas (HGPXAs). All the patients accepted surgery in Beijing Tiantan Hospital between August 2014 and September 2021. The following clinical data were included: age, sex, imaging characteristics, extent of tumor resection, histopathological results, treatment protocol and outcomes. Pre- and postoperative MRI images were performed to evaluate the extent of tumor excision, which was defined as gross total resection (GTR) and non-GTR. The follow-up was performed by telephone interview every six months. This research was approved by the Beijing Tiantan Hospital Research Ethics Committee. The pathological diagnosis of HGPXAs was confirmed by the Department of Neuropathology at Beijing Neurosurgical Institute according to the 2021 World Health Organization Classification of Tumors of the Central Nervous System. Histopathological examination showed malignant glial component with numerous mitoses. Eosinophilic granules and ribosome-lamellar complexes can be observed under the light microscope, which are characteristic features of PXA. All cases exhibiting increased mitotic activity, >5/10 HPF mitotic figures with or without accompanying necrosis. The mutation status of BRAF, IDH, and TERT promoter, the methylation status of the MGMT promoter, and the co-deletion status of 1p/19q were also assessed in some of our cases.For the pooled analysis of HGPXAs, we performed a search in the PubMed database between January 1997 and October 2022. The keyword used was “anaplastic” combined with “pleomorphic xanthoastrocytoma(s)” and a total of 56 cases were included. All cases were pathologically diagnosed as “anaplastic pleomorphic xanthoastrocytoma”, which were classified into WHO grade 3 PXAs.

Cox regression models were used to evaluate variables and their association with PFS and OS. The Kaplan-Meier method was used to determine the OS and PFS differences with p-values calculated from the log-rank test. Analyses were performed using SPSS Statistical Package software with the significant set at p < 0.05.

Pleomorphic xanthoastrocytoma (PXA) is a rare brain tumor, which was first reported in 1979 (48, 49). ‘PXA with anaplastic features’ are labelled as anaplastic pleomorphic xanthoastrocytoma (APXA), WHO grade III, according to the 2016 World Health Organization Classification of Tumors of the Central Nervous System (3). According recent World Health Organization Classification of Tumors of the Central Nervous System, PXA was classified into WHO grade 2 or WHO grade 3 tumors (5). Most of the information available about APXA comes from isolated case reports. In our institute, 200 cases with PXAs were pathologically confirmed from August 2014 and September 2021. Among them 26 cases were grade 3 PXAs and 174 cases were grade 2 PXAs, HGPXAs are estimated to comprise 13% of all PXAs in our single center.

Mallick et al. reported the median age of grade 2 PXA patients was 20 years via analyzing 167 cases and Rodrigues et al. reported the median age of grade 2 PXA patients was 21 years via analyzing 346 cases (49, 50). In our pooled cohort, HGPXAs tended to affect middle-aged patients with a median age of 30.5 years, compared to their grade 2 PXA patients. We also found that age>33 years was a risk factor for PFS (p=0.018) and OS (p=0.023) via multivariate cox regression analysis. There is no gender predominance in HGPXAs in this study, which is consistent with previous studies (50). Most intracranial HGPXAs were located in supratentorial area and temporal lobe was the first commonest region, only 5 cases located in infratentorial area. Preoperative seizure is common with the occurrence of 35.9%. 8 cases from our institute presented with preoperative seizure were free of seizure after GTR. HGPXAs can be divided into primary tumors or secondary tumors. Of the pooled analysis, 67 cases were primary tumors and 13 cases were secondary tumors, and secondary tumors had a poorer OS than primary tumors (p=0.001). Radiologically, HGPXAs often appear T1 isointense and T2 hyperintense and the cystic component is often hypodense On Magnetic Resonance Imaging (MRI). PXA usually contains solid and cystic components and the solid component often enhances (8, 51). She et al. reviewed MR imaging features of 9 patients with APXA and 10 patients with PXA. The presence of heterogeneous enhancement of solid mass was observed more frequently in patients with APXA than in those with PXA (52). In our study, enhancement was observed in almost all HGPXAs patients (96.6%) and solid tumors are more common than cystic-solid tumors (35 VS 26 cases). 73.3% patients with HGPXAs had obvious peritumoral edema. However, through cox regression analysis, no peritumoral edema could not predict a better PFS or OS.

Due to the rarity of HGPXAs, there is limited research to identify risk factors for OS or PFS. Patibandla et al. pointed that a complete surgical excision is required for a prolonged disease-free interval (28). Rodrigues et al. analyzed 62 cases of APXA from SEER database and concluded that complete surgical resection could not bring improved outcomes. In our study, we found non-GTR group was associated with a poor PFS (p=0.015) and OS (p=0.001) than GTR group. Maximum safe resection, if feasible, should be the first goal of the neurosurgeon. The role of postoperative radiotherapy and chemotherapy for APXAs is still uncertain (1). Marton et al. reviewed 9 cases with APXA underwent conventional fractionated radiotherapy, but the effect of treatment is not significant (53). Koga et al. reported a case of APXA treated with postoperative stereotactic irradiation (STI) that resulted in long-term control of the tumor (1). Postoperative chemotherapy has been commonly considered ineffective for the treatment of PXAs (54). In our pooled analysis, we found that no postoperative radiotherapy group was associated with a poor PFS (p=0.007) and OS (p=0.008) than postoperative radiotherapy group and no postoperative chemotherapy can predict a better PFS (p=0.042). Based on these, we recommended that postoperative radiotherapy and chemotherapy should be added to conventional treatment protocols. Molecular markers are increasingly used to help doctors to diagnose or subclassify gliomas. BRAF mutations were found in 70% PXAs, but involved less common in HGPXAs (55). Phillips et al (56) performed comprehensive genomic profiling on 15 cases with APXA and found 5 cases with TERT promoter hotspot mutation. In our study, BRAF p.V600E mutations were detected in 48.8% of patients (20/41), and TERT promoter mutations were found in 23.1% of patients (3/13). It is reported that compared to the BRAF wild-type PXA, BRAF-mutated PXA revealed prolonged survival (6). However, we did not find any relationship between BRAF mutation and PFS (p=0.9275) or OS (p=0.4755) for HGPXAs. In previous reports, small molecule drugs, such as BRAF, MEK inhibitors were used for some patients. Brown et al. reported a 21-year-old female suffered occurrence of APXA with a BRAFV600 mutation following 2 operations and radiotherapy. Then she was orally administered with BRAF inhibitors. She was well up to the last follow-up (32). Hussain et al. reported a 43-year-old male underwent GTR plus radiotherapy of an APXA, but, unfortunately, the tumor progression happened 2 years later. Because the tumor harbored a BRAFV600 mutation, combination of vemurafenib (BRAF inhibitor) and trametinib (MEK inhibitor) was added to treatment strategy. Clinical stability and radiographic improvement were achieved for 6 months after the targeted therapy (36). Targeted therapy may bring hope to HGPXAs patients, but more data is needed to prove its effectiveness.

According to previous studies, OS of PXA is favorable with 5-year survival rates of >75% and PFS rates at 5 years are >60% (4, 57). Patibandla et al (28) reviewed 17 cases of APXA reported in the literature till 2010 and found that among 13 patients whose follow-up information was available, 8 patients died with the mean overall survival was 26.1 months (ranges from 1 to 66 months). Compared to PXA, we found that HGPXAs have a relatively poor clinical outcomes with 5-year OS of 51.5% and 5-year PFS of 26.1% and 19 patients died with a mean overall survival was 19.8 months (ranges from 0.75 to 66 months). Although no tumor metastasis was found in our institute, 13.3% (10/75) patients experienced spinal metastasis were founded in reported cases. Spinal MRI should be taken into account during postoperative follow-up for early treatment of the metastatic tumor.

High grade pleomorphic xanthoastrocytomas are very rare brain tumors. Children and younger adults have better clinical outcome than elderly patients. Secondary HGPXAs had worse OS than primary HGPXAs. Complete surgical excision plus RT and CT is recommended for this entity. The frequency of BRAF mutations in HGPXAs is 47.5% (19/40) in this study, however, we do not find the connections between BRAF mutations and clinical outcomes. Future studies with larger cohorts are necessary to verify our findings.

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

The studies involving human participants were reviewed and approved by human research ethics committee of Beijing Tiantan Hospital. The patients/participants provided their written informed consent to participate in this study.

Writing, original draft, Conceptualization: PZ; Statistical analysis: TS, WW; Literature review: TL, YW; Follow up: MZ; Methodology, Resources: ZW, JZ; Review, editing, Supervision: LZ. All authors contributed to the article and approved the submitted version.