Primary intracranial histiocytic sarcomas: a report of six cases and a pooled analysis of individual patient data

Primary intracranial histiocytic sarcomas (PIHSs) are extremely rare with limited reported cases, making their prognostic factors and management uncertain. This study aims to describe the clinical characteristics of PIHSs and propose a treatment protocol for this entity. Clinical data were collected from six patients diagnosed with PIHSs at Beijing Tiantan Hospital between March 2011 and October 2022. Additionally, a comprehensive search of the PubMed database was conducted using the keywords "primary intracranial" or "primary central nervous system" combined with "histiocytic sarcoma" or "histiocytic sarcomas" between 1996 and 2022, identifying 24 cases. A pooled analysis of individual patient data was performed to assess risk factors for overall survival (OS). The six cases included four males and two females, with a mean age of 42.2 ± 13.3 years. In total, 24 cases of PIHSs were identified from previous studies. Multivariate Cox regression analysis revealed that gross total resection (GTR) was the only factor predicting a longer OS (p = 0.027). Kaplan–Meier analysis demonstrated that GTR (p = 0.0013), solitary lesions (p = 0.0048), and radiotherapy (p = 0.0492) were associated with a longer OS. PIHSs are rare brain tumors with poor clinical prognosis. Patients with solitary lesions have a longer OS than those with multifocal lesions. Gross total resection must be the first choice. Radiotherapy may bring benefits for these patients, but chemotherapy may not be useful. Further studies involving larger cohorts are necessary to validate these findings.


Introduction
Histiocytic sarcoma (HS) is an uncommon malignancy arising from the hematopoietic system, primarily affecting lymph nodes, the intestinal tract, and soft tissues (Zanelli et al. 2017). Primary intracranial histiocytic sarcomas (PIHSs) are extremely rare, with only 24 reported cases until recent times. According to the 2021 WHO Classification of Tumors of the Central Nervous System, HS is classified as a subtype of histiocytic tumors (Louis et al. 2021). This rare entity is characterized by distinctive morphological and immunophenotypic features that are characteristic of histiocytes (Hung and Qian 2020). Previous studies showed rapid disease progression and poor clinical prognosis of the tumor (Torres et al. 1996;Cheuk et al. 2001;Sun et al. (2003). However, due to the rarity of PIHSs, their clinical features and management remained unknown. In this study, we reported six patients with PIHSs in our single institute and tried to propose treatment protocols to manage PIHSs (see Table 1).

Methods
All patients diagnosed with PIHSs at Beijing Tiantan Hospital between March 2011 and October 2022 underwent pathological examination to confirm the diagnosis. Secondary HSs were ruled out through general examination. A comprehensive review of relevant clinical data, including age, sex, symptoms, duration of symptoms, extent of resection, radiological features, histopathology, treatment, and outcomes, was conducted. The radiological findings, such as tumor location, peritumoral edema, solid or cystic appearance, signal intensity, and enhancement, were independently reviewed by two senior radiologists. The extent of resection was determined based on pre-and postoperative magnetic resonance imaging (MRI) scans and categorized as gross total resection (GTR) or non-GTR. The diagnosis of PIHSs was confirmed by the Department of Neuropathology at Beijing Neurosurgical Institute, where two independent neuropathologists assessed the tumors using the 2021 WHO Classification of Tumors of the CNS. Follow-up was conducted through telephone interviews. Furthermore, a search of the PubMed database was conducted, covering the period between 1996 and 2022, using the keywords "primary intracranial" or "primary central nervous system" in combination with "histiocytic sarcoma" or "histiocytic sarcomas." A total of 24 cases of PIHSs with confirmed pathology were identified (see Table 2). The potential risk factors for overall survival (OS), including sex, age, extent of resection, multifocality, radiotherapy, and chemotherapy, were evaluated through univariate and multivariate Cox regression analyses using the pooled cases from our institute (5 cases) and the literature (22 cases). Survival curves were generated using the Kaplan-Meier method, and statistical significance was assessed using the log-rank test. Statistical analyses were performed using SPSS Statistical Package software (version 22.0, IBM Corp.), with a significance level set at p < 0.05 (see Table 3).

Results
The present study included four males and two females, with a mean age of 42.2 ± 13.3 years (range: 27-67 years). The duration of symptoms ranged from 0.5 to 6 months, with a mean duration of 2.4 ± 2.1 months. Common symptoms reported were headache (n = 2), dizziness (n = 2), gait difficulty (n = 1), limb weakness (n = 1), nausea (n = 1), vomiting (n = 1), impaired vision (n = 1), and hearing loss (n = 1). Radiologically, 33.3% of the patients had multifocal brain lesions, while other locations included the frontotemporal lobe, frontoparietal lobe, triangle area, and occipital lobe. Preoperative MRI scans were available for five patients, and a preoperative CT scan was available for one patient. The lesions appeared hypointense-isointense in one case and hypointense in four cases on plain T1WI. On plain T2WI, two cases showed isointense-hyperintense and three cases showed hypointensity. CT scans revealed high-density masses. Contrast enhancement was observed in all five available patients, and peritumoral edema was observed in all six patients. Gross total resection (GTR) was performed on three patients, while two patients underwent non-GTR, and one patient underwent biopsy. Unfortunately, case 3 was lost to follow-up. Three patients (60%) received postoperative radiotherapy and chemotherapy. After a mean follow-up of 25.3 ± 27.9 months (range 1-70 months), three patients died, resulting in a mean overall survival (OS) of 18.6 ± 14.1 months (range 10.2-34.8 months) (see Table 4).
The mean follow-up period for the pooled cohort, which included five cases from our institute and 24 cases from the literature, was 13.5 ± 16.6 months. The 0.5-year and 1-year overall survival (OS) rates were 65.6% and 39.1%, respectively. Univariate Cox regression analysis revealed that gross total resection (GTR) (p = 0.005) and solitary lesions (p = 0.010) were predictive of a better OS. However, factors such as age ≤ 40 (p = 0.976), male gender (p = 0.382), radiotherapy (p = 0.066), and chemotherapy (p = 0.638) did not show a significant association with OS. Subsequently, GTR, solitary lesions, radiotherapy, and chemotherapy were included in the multivariate Cox regression analysis, which revealed that GTR was the only factor significantly predicting a better OS (p = 0.027). The Kaplan-Meier analysis demonstrated that GTR (p = 0.0013) and solitary lesions (p = 0.0048) were associated with a better OS. The median survival time for patients with multifocal lesions was 4 months, whereas for patients with solitary lesions, it was 34.8 months (see Fig. 2).

Case 1
A 41-year-old male presented with symptoms of dizziness, gait difficulty, and hearing loss persisting for approximately 3 months. Neurological examination revealed ataxia of the left limb. An MRI scan revealed the presence of multifocal lesions in various areas of the brain, including the right CPA area, interventricular foramen area, right   frontal lobe, and left temporal lobe. The lesion located in the right CPA area appeared as hypointense-isointense on T1WI, isointense-hyperintense on T2WI, and demonstrated homogeneous enhancement, indicating it as the primary mass responsible for the symptoms. Following multidisciplinary discussions, it was decided to perform the removal of the CPA lesion to alleviate the patient's symptoms. A right retrosigmoid approach was chosen for the surgery. During the procedure, the tumor was visually observed to have a gray and white appearance with a well-developed blood supply. The tumor was found to originate from the lower cranial nerves and exhibited strong adherence to the brainstem and cerebellum, rendering gross total resection unattainable. Pathological examination of the tumor specimen revealed the presence of aggregates of histiocytoid cells characterized by abundant eosinophilic cytoplasm and marked nuclear pleomorphism. Immunohistochemistry analysis demonstrated positive staining for CD68, lysozyme, CD34, and vimentin, while negative staining was observed for CD3, CD20, CD1a, S-100, and CD79a. The Department of Neuropathology at Beijing Neurosurgical Institute confirmed the diagnosis of histiocytic sarcoma.
Following the surgery, the patient developed aspiration pneumonia, which was successfully treated with systemic antibiotic therapy. Unfortunately, the patient succumbed to tumor progression 10.7 months after the surgical intervention.

Discussion
Histiocytic sarcoma (HS), initially described by Mathé et al. in 1970, is a rare and malignant hematopoietic neoplasm (Mathé et al. 1970). Via analyzing 330 HS cases, Kommalapati et al. concluded that surgical treatment alone or in combination with radiotherapy could bring benefits in OS for patients with HS in skin and connective tissue and surgery may also improve OS in patients with HS in respiratory system and gastrointestinal (Kommalapati et al. 2019). PIHS is extremely rare and it was first reported in 1996 by Torres et al. (1996). PIHS was classified as a subtype of histiocytic tumors, based on the 2021 WHO Classification of Tumors of the Central Nervous System (Louis et al. 2021). In our single institute, there are approximately 740 cases were pathologically diagnosed of primary intracranial hematolymphoid tumors (including 574 cases with lymphomas and 166 cases with histiocytic tumors) between 2011 and 2022 over a 10-year period. Only six cases were pathologically confirmed PIHSs and the incidence was approximately 0.81% among all types of primary intracranial hematolymphoid tumors at Beijing Tiantan Hospital.
In the pooled analysis, we found that this entity usually occurred in middle-aged or elderly people without gender predominance. Primary central nervous system lymphoma can be found in immunosuppressed patients, but  immunodeficiency related diseases were not found in all 30 cases (including six cases from our hospital and 24 cases from published paper). Symptoms were related to the tumor location. Relatively short duration of preoperative symptoms suggested the malignant biological behavior of the tumor. Radiologically, approximately one third patients showed multifocal lesions in the brain. For the solitary counterpart, supratentorial area (17/20, 85%) is the most frequently involved region. The clinical prognosis of patients with multifocal lesions was found to be exceedingly poor. The median survival for patients with multifocal lesions and patients with solitary lesions were 4 VS 34.8 months. On MRI scans, the most common findings included hypointensity or isointensity on T1-weighted images (T1WI) and hypointensity or isointensity on T2-weighted images (T2WI), accompanied by pronounced enhancement and peritumoral edema. Histopathologically, this condition exhibited characteristic histiocytoid cells displaying abundant eosinophilic cytoplasm and notable nuclear pleomorphism (Bai et al. 2014). However, the definitive diagnosis of histiocytic sarcoma (HS) requires the exclusion of other hematolymphoid tumors and confirmation of histiocytic markers, such as CD68, lysozyme, and CD163. Notably, CD163 has recently been identified as a more specific histiocytic marker compared to CD68, potentially serving as a valuable biomarker in the future (Nguyen et al. 2005). In our study of six cases, all patients exhibited positive staining for histiocytic markers while testing negative for T-cell and B-cell markers. Additionally, comprehensive examination results effectively ruled out the possibility of secondary HSs.
The clinical outcome of the rare entity is very poor. In this study, the median survival for the pooled cohort was 10 months, longer than previous studies reported (4.5 months) (Cao et al. 2007;Laviv et al. 2013). Via statistical analysis, we found that the extent of surgical resection has a significant impact in OS. If feasible, the tumors should be strived for gross total resection. The effect of radiotherapy remains controversial (Wang et al. 2012;Ma et al. 2018). We found that postoperative radiotherapy reveal some statistically insignificant improvement in OS, which may indicate the potential role of radiotherapy. The effect of chemotherapy remained uncertain. Our research could not prove the relationship between chemotherapy and OS. Primary leptomeningeal histiocytic sarcoma is exceedingly rare and only three cases had been reported previously (Torres et al. 1996;Pérez-Ruiz et al. 2013;Zanelli et al. 2017). Pérez-Ruiz reported a 41-year-old female with primary leptomeningeal histiocytic sarcoma had been alive for 42 months after total resection of the tumor (Pérez-Ruiz et al. 2013). However, because of extensive involvement of the leptomeningeal, the other two patients died soon without surgery (Torres et al. 1996;Zanelli et al. 2017). Our Case 2 has been alive for 70 months without tumor recurrence after GTR in combination with radiotherapy and chemotherapy. Two independent Fig. 2 The Magnetic Resonance Imaging (MRI) findings demonstrated a hypointense-isointense signal on T1-weighted images (T1WI), b isointense-hyperintense signal on T2-weighted images (T2WI), and c-e significant enhancement of a solid mass located at the right cerebellopontine angle (CPA) area and the third ventricle.
The postoperative Computed Tomography (CT) scan revealed f subtotal resection of the tumor at the right CPA area. Histopathological examination of the H&E-stained section showed g atypical tumor cells with marked nuclear pleomorphism and abundant eosinophilic cytoplasm. (H&E, × 200 magnification) institute (including our institute) both confirmed the case as a histiocytic sarcoma. Based on our current understanding, this case exhibits the most favorable clinical outcome observed to date.
Small-molecule targeted therapy may bring hope to these patients. Idbaih et al. (2014) reported a 40-year-old male of PIHS with BRAFV600E mutation and he was administered with vemurafenib. Although the tumor relapsed four months after administration with vemurafenib, he was stable for 3 months and significant radiologic tumor response was also observed. May et al. published a report on a 47-year-old female who had a mutation in the platelet-derived growth factor receptor, but the patient did not tolerate Dasatinib (May et al. 2018). While the targeted therapy used in this case did not yield satisfactory results, the expression of PD-L1 and PD-L2 suggests the possibility of utilizing immunotherapy as a potential treatment option.
Overall, in this study, firstly, we also observed that complete surgical resection improved OS for patients. Therefore, whenever feasible, neurosurgeons should strive for complete tumor resection. Secondly, despite undergoing complete surgical resection, the majority of patients still succumb to tumor recurrence. Future research should focus on determining the appropriate postoperative adjuvant therapies to prolong patient survival in this rare disease.
In a previous retrospective study involving 330 cases of histiocytic sarcomas, it was found that surgery combined with radiotherapy could prolong OS (Kommalapati et al. 2019). Our study also revealed that postoperative radiotherapy can improve overall survival (OS) in patients with PIHSs, but chemotherapy did not demonstrate significant benefits. Based on our pooled analysis, we believe that complete surgical resection followed by adjuvant radiotherapy represents the optimal treatment approach for PIHSs.

Conclusion
PIHSs are rare brain tumors with poor clinical prognosis. Patients with solitary lesions have a longer OS than those with multifocal lesions. The maximum safe resection should be first choice at present. Radiotherapy could bring benefits in OS. Based on present data, chemotherapy could not bring survival benefits. Targeted therapy and immunotherapy may bring hope for these patients. Future studies with large cohorts should be performed to verify our findings.

Limitations
The limitations of this study are small sample size, retrospective nature, and short observation periods. Firstly, the retrospective nature of our study introduces certain constraints. We conducted a pooled analysis of a relatively small sample size, including only 27 cases, which is attributed to the rarity of the disease under investigation. Secondly, the observation periods in our study were relatively short, with a mean follow-up time of 13.5 ± 16.1 months. Considering the potential long-term effects and outcomes of the condition, this duration may not fully capture the complete spectrum of disease progression and related factors. Therefore, a prospective multicenter study with a large series of PIHSs is needed. Funding This study was funded by Multicenter clinical big data study and multi-path tumorigenesis mechanisms and precision treatment research on brainstem glioma (JINGYIYAN2018-7).

Availability of data and material
The datasets generated during the current study are available from the corresponding author on reasonable request.

Declarations
Conflict of interest All authors declared that there were no conflicts of interest to be disclosed.
Ethics approval This study was approved by the human research ethics committee of Beijing Tiantan Hospital. The written informed consent was obtained from the subjects.

Consent to participate All patients signed an informed consent form.
Consent for publication Written informed consent for publication was obtained from all participants.