Abstract

Background: Diffuse bone-marrow metastasis of high-grade glioma associated with hematological abnormalities is extremely rare.

Case Description: A 32-year-old man was referred and admitted to our hospital for treatment of three remote recurrent brain lesions. He had been treated at the referring hospital for a primary brain tumor in the right frontal lobe. One of the recurrent lesions was resected and diagnosed as a grade 4 isocitrate dehydrogenase (IDH)-mutant astrocytoma. Stereotactic radiation therapy (SRT) was performed on all three lesions. During this hospitalization, a lumbar spine magnetic resonance imaging (MRI) showed signal changes in the first and fourth vertebral bodies, suggesting lumbar metastasis. In addition, blood tests showed a gradual increase in the lactate dehydrogenase (LDH) level. Three months later, the patient was referred to our hospital again for palliative SRT of metastatic lumbar vertebral lesions invading the psoas major muscles. Laboratory data showed pancytopenia and a marked increase in the LDH level. A lumbar spine MRI showed signal changes in all lumbar and sacral vertebrae. To rule out hematological malignancy, biopsies of the psoas major and iliac bone marrow were performed. They showed invasion of grade 4 astrocytoma cells in both areas, leading to a diagnosis of diffuse bone-marrow metastasis. The patient died 12 days after the second admission.

Conclusion: We present a rare case of diffuse bone-marrow metastasis of grade 4 IDH-mutant astrocytoma associated with hematological abnormalities. Progressive LDH elevation might predict diffuse bone-marrow metastasis in patients with high-grade glioma.

Keywords: Gliomatosis of the bone marrow, High-grade glioma, Lactate dehydrogenase, Pancytopenia, Vertebral metastasis

INTRODUCTION

High-grade gliomas, such as grade 4 isocitrate dehydrogenase (IDH)-mutant astrocytoma and glioblastoma, are aggressive and prevalent malignant brain tumors. However, they rarely metastasize outside the central nervous system (CNS): the incidence of extraneural metastasis from primary intracranial glioblastoma has been estimated at 0.4–0.5% of cases.[ 44 ] Metastases have been discovered near previous surgical sites in at least 36.9% of cases, and other sites of metastasis have included bone (47.9%), lung (25.6%), lymph nodes (25.1%), scalp (19.2%), and liver (14.2%).[ 19 ] Although the major site of metastasis is bone, widespread bone-marrow metastasis of high-grade glioma associated with hematological abnormalities, which has been described in the field of cancer as disseminated carcinomatosis of the bone marrow (DCBM), has rarely been reported.[ 17 ]

Here, we describe a case of diffuse bone-marrow metastasis of grade 4 IDH-mutant astrocytoma that was associated with hematological abnormalities. We also reviewed other cases of bone metastasis of high-grade glioma with hematological abnormalities, in which we expected that the primary brain tumors had biological characteristics similar to ours.

MATERIALS AND METHODS

A PubMed search was conducted in accordance with the guidelines of Preferred Reporting Items for Systematic Reviews and Meta-Analyses for articles containing the terms “glioblastoma,” “astrocytoma,” “oligodendroglioma,” “oligoastrocytoma,” “glioma,” “bone metastasis,” and “bone metastases.” Studies were limited to those written in English, French, German, or Japanese. Primary studies, such as case reports, were included, but secondary studies, such as systemic reviews and meta-analyses, were excluded unless new patients were described. One author (N.N.) performed the initial screening of all titles and abstracts. Full-text reviews of the studies that were selected for this review were conducted. The last search was performed on 5 March 2025. Cases of bone metastasis of high-grade glioma – CNS World Health Organization (WHO) grade 3 and 4[ 3 ] – with hematological abnormalities were selected and analyzed. From these selected cases, we extracted the following variables: histologic diagnosis; age at the time of diagnosis of primary high-grade glioma; gender; skeletal pain as a representative symptom; hematological abnormalities; level of lactate dehydrogenase (LDH) as a marker of hemolysis and tumor lysis; level of alkaline phosphatase (ALP) as a marker of bone metabolism; prior resection as a cause of extraneural metastasis (if the patient had symptoms of bone-marrow metastasis immediately after the resection, we judged the case as prior-resection negative, as we considered that the bone metastasis had already started before the resection in that case); immunophenotype and genetic alteration observed in the primary brain tumor or extraneural metastatic lesion; treatments for diffuse bone-marrow metastasis with hematological abnormalities; survival time from the time of diagnosis of high-grade glioma; time from the diagnosis of high-grade glioma to the diagnosis of diffuse bone-marrow metastasis with hematological abnormalities; and survival time from the time of diagnosis of bone-marrow metastasis with hematological abnormalities.

CASE DESCRIPTION

A 32-year-old man was referred and admitted to the Department of Neurosurgery at Shiga University of Medical Science Hospital for the treatment of remote recurrent lesions of a brain tumor. The original lesion had been located in the right frontal lobe [ Figure 1a ], removed subtotally 8 months before [ Figure 1b ], diagnosed as anaplastic oligodendroglioma by histopathological analysis (without either chromosome or gene testing), and treated at the referring hospital with radiation therapy and chemotherapy with temozolomide (TMZ). Recurrent lesions were located in the left cerebellar hemisphere [ Figure 1c ], both caudate nuclei [ Figure 1d ]) and the right occipital lobe [ Figure 1e ]. Laboratory data on admission to our hospital showed only mildly elevated liver enzymes. Because magnetic resonance imaging (MRI) showed signs of subependymal dissemination of the tumor, we considered the possibility of malignant transformation of the original tumor. For accurate diagnosis based on molecular classification and for further treatment based on accurate diagnosis and gene analysis, we decided to perform tissue sampling. The right occipital lesion was subtotally resected endoscopically [ Figure 1f ] and was histopathologically diagnosed as grade 4 IDH-mutant astrocytoma [ Figures 2a and b]. Fluorescence in situ hybridization showed only 19q deletion, without 1p deletion, suggesting that the first histopathological diagnosis as anaplastic oligodendroglioma had been incorrect. The specimens contained a highly cellular monomorphic population of tumor cells with microvascular proliferation and some mitoses adjacent to normal brain tissue [ Figure 2a ], with IDH1 mutation on immunohistochemistry [ Figure 2b ]. In that lesion, nuclear α-thalassemia/mental-retardation-syndrome-X-linked gene (ATRX) expression was lost, and p53 was overexpressed [ Figure 2c ], with a high Ki-67 proliferation index. Gene analysis of the specimen showed IDH1 R132H mutation, loss of ATRX, TP53 mutation, and cyclin-dependent kinase inhibitor 2A (CDKN2A) homozygous deletion without telomerase reverse transcriptase gene (TERT) promoter mutation, indicating astrocytoma, IDH-mutant, grade 4. O^6-methylguanine-DNA methyltransferase (MGMT) promoter methylation was observed. Gene analysis also showed other alterations: bromodomain-containing protein 4 (BRD4) rearrangement intron 17, CDKN2B loss, Capicua (CIC) loss, and phosphoinositide-3-kinase regulatory subunit 1 (PIK3R1) R577 deletion.

Figure 1:

(a) Post-contrast T1-weighted imaging in the axial plane on first admission of the patient to the referring hospital, (b) after the first operation, (c–e) at recurrence, and (f) after the second operation. A large cystic lesion with ring enhancement was located in the right frontal lobe (a) and was subtotally removed (b). Remote recurrent lesions were found in the cerebellum (c), the caudate nucleus bilaterally (d), and the right parieto-occipital white matter (e). The lesion in the right parieto-occipital white matter was subtotally removed (f).

Figure 2:

(a) Hematoxylin and eosin staining of the intracerebral metastatic lesion at 400× magnification. (b) Mutant IDH-1 immunohistochemistry at 400× magnification. (c) P53 immunohistochemistry at 400× magnification. Scale bar: 50 mm in (a). (d) T1-weighted, (e) T2-weighted, and (f) post-contrast T1-weighted sagittal imaging of the lumbar spine after removal of the metastatic brain lesion. Arrowheads in (d, e) show hypointense signal change on T1-weighted images and hyperintense signal change on T2-weighted images in the first and fourth lumbar vertebrae; these may correspond to early-stage bone- marrow metastases. Note that the post-contrast T1-weighted imaging obscures these lesions [arrowheads in (f)]. Arrow in the magnified inset in (f) shows a remote small lesion attached to the cauda equina.

As the tumors had been growing rapidly, we had planned the resection for diagnosis as soon as possible. However, before the surgery, there had been no booking availability for MRI of the whole spine. Consequently, we performed the resection before we were able to assess the possibility of leptomeningeal metastases in the spine, and the MRI was performed 4 days after the surgery. It showed hypointense changes on T1-weighted imaging (T1WI) [ Figure 2d ] and hyperintense changes on T2-weighted imaging (T2WI) [ Figure 2e ] in the first and fourth lumbar vertebrae, without deformation, in addition to a small lesion attached to the cauda equina on postcontrast T1WI [ Figure 2f , inset]. The hypointense vertebral lesions on T1WI were enhanced on postcontrast T1WI, giving them almost the same intensity as the normal vertebral area and making them inconspicuous [ Figure 2f ]. Computed tomography (CT) of the lumbar spine showed neither osteolysis nor hyperostosis (Data not shown). The lesions were suspected to be vertebral extraneural metastases, along with a small leptomeningeal metastatic lesion in the cauda equina [ Figure 2f ]. As tissue samples for diagnosis of the recurrent tumor had already been obtained, a vertebral biopsy was not planned. CT of the chest, abdomen, and pelvis showed no second malignancy.

Stereotactic radiation therapy (SRT) of all three brain recurrence lesions was performed. Maintenance chemotherapy with TMZ was resumed, and molecularly targeted therapy with bevacizumab (Bev) was started. After a month of hospitalization, the chemotherapy and molecularly targeted therapy were maintained at the outpatient department of the referring hospital.

Twelve months after the first resection and 4 months after the day of admission to our hospital, the patient was referred to our hospital again for palliative radiotherapy for the metastatic lesion in and around the fourth lumbar vertebra. The patient reported lower back pain, left hip joint pain, and left knee pain. Postcontrast T1WI of the lumbar spine showed the metastatic lesion in the fourth lumber vertebra expanding into the epidural space and the psoas major muscles [ Figures 3a - c ]. Preadmission laboratory data showed hematological abnormalities: pancytopenia (white blood cells: 2.1 × 10³/µL [normal range: 3.3–8.6 × 10³/µL]; hemoglobin: 8.1 g/dL [13.7–16.8 g/dL]; low platelet count: 52 × 10³/µL [158–348 × 10³/µL]) with normal mean corpuscular volume (87.9 fL [83.6–98.2]), polychromasia, anisocytosis, and poikilocytosis. They also showed markedly elevated serum levels of LDH (9836 U/L), fibrin degradation products (FDPs) (67.0 µg/mL [0.0–5.0 µg/mL]), and D-dimer (26.1 µg/mL [0.0–0.9 µg/mL]), with moderate elevation of ALP (206 U/L [38–113 U/L]). The serum aspartate aminotransferase level and total bilirubin level were mildly elevated (66 U/L [13–30 U/L] and 1.71 mg/dL [0.40–1.50 mg/dL], respectively), but the level of potassium was normal (4.4 mmol/L [3.6–4.8 mmol/L]). LDH isozyme analysis showed dramatically increased percentages of LDH-1 (46.9%) and LDH-2 (41.5%), suggesting tumor lysis or hemolysis, or both. These results prompted us to review the laboratory data obtained during the patient’s previous hospitalization when his LDH level gradually increased from 189 U/L on admission to 635 U/L at discharge. MRI of the lumbar spine showed a mild compression fracture of the fourth lumbar vertebra, but it did not show either apparent deformity or apparent destructive changes in other bones [ Figure 3a ]. Three days after the referral visit, the patient was admitted to our hospital, and on the same day, a biopsy of the psoas major was performed. The next day, a biopsy of the bone marrow was also performed from the iliac crest to rule out hematologic malignancies. Both specimens showed invasion by grade 4 IDH-mutant astrocytoma cells [ Figures 3d , e ]. The Ki-67 labeling index was approximately 100% in both lesions. The bone marrow was diffusely invaded and replaced by oligodendrocyte transcription factor 2 (OLIG2)-positive glioma cells [ Figure 3f ], with few megakaryocytes observed [ Figure 3e ]. These neoplastic cells were strongly positive for p53 [ Figure 3g ], as observed in the specimens of the recurrent brain lesions [ Figure 2c ]. IDH1 mutation was confirmed on immunohistochemistry [ Figure 3h ]. Retrospectively, we confirmed signal changes in all lumbar vertebrae and sacral vertebrae on T1WI [ Figure 3i ; compare with Figure 2d ] and T2WI [ Figure 3j ; compare with Figure 2e ] of the lumbar spine. All vertebrae were enhanced on postcontrast T1WI [ Figure 3a ] in comparison with noncontrast T1WI [ Figure 3i ]. These MRI findings and the histological findings from the iliac bone-marrow biopsy suggested the presence of widespread diffuse bone-marrow metastasis in multiple bones, perhaps causing the hematological abnormalities. CT of the chest, abdomen, and pelvis showed no second malignancy, and tumor marker tests were negative. SRT of the metastatic mass lesions in and around the fourth lumbar vertebra and in the psoas major muscles was performed to palliate lumbar pain. The patient was diagnosed with probable disseminated intravascular coagulation (DIC) on the basis of the increased levels of FDPs.[ 35 ] After red blood cell and platelet transfusions, the pancytopenia transiently improved, and the patient was discharged 11 days after the second admission. Unfortunately, he died the following day from sudden cardiac arrest, and we were unable to obtain permission to perform further gene analysis of the bone-marrow metastasis.

Figure 3:

Post-contrast T1-weighted (a) sagittal, (b) axial, and (c) coronal imaging of the lumbar spine after the metastatic lesion in the fourth lumbar vertebra [arrowhead in (a)] had expanded into the epidural space [arrow in (b)] and the psoas major muscles [arrowheads in (b) and (c)]. (d–e) Hematoxylin and eosin staining of the metastatic tumor in the psoas major (d) and in bone marrow from the iliac crest (e) at 400× magnification. (f) OLIG2 immunohistochemistry of the metastatic tumor in the bone marrow at 400× magnification. Almost all tumor cells were OLIG2 positive, suggesting that they were of glial origin. (g) P53 immunohistochemistry at 400× magnification. Scale bar: 50 mm in (d). (h) Mutant IDH-1 immunohistochemistry at 400× magnification. Scale bar: 50 mm in (h). (i) T1-weighted and (j) T2-weighted sagittal imaging of the lumbar spine. Comparison with Figures 2c and 2d reveals that hypointense signal changes on T1-weighted images in (i) and hyperintense signal changes on T2-weighted images in (j) extend from the 12th thoracic vertebra to the second sacral vertebra, suggesting that these vertebrae have been diffusely invaded by grade 4 astrocytoma cells. Comparison with T1-weighted images in (i) reveals that all vertebrae are enhanced on post-contrast T1-weighted images in (a).

SYSTEMATIC LITERATURE REVIEW RESULTS

The PubMed search yielded 773 unique hits. Initial screening of the titles and abstracts excluded 544 studies, primarily because they were not relevant to bone metastasis or were relevant to the diagnosis of pathologies other than high-grade glioma. Screening of the citations, references, and related articles yielded 31 additional hits. A total of 223 studies were assessed through full-text review for eligibility, with 192 studies being excluded because of diagnoses other than glioma, no bone metastasis, no hematological abnormalities, or low grade. Thirty-seven studies could not be retrieved. Finally, a total of 31 articles were included in this systematic review, which included 34 unique patients with bone metastasis of high-grade glioma associated with hematological abnormalities [ Figure 4 ].

Figure 4:

Preferred Reporting Items for Systematic Reviews and Meta-Analyses flow diagram of literature screening and exclusion criteria.

Including ours [ Table 1 ], the average age at the time of diagnosis of high-grade glioma was 44.2 years (range, 12–74 years), with 74% of patients being male. In most of the cases, skeletal pain was observed at the site of bone-marrow metastasis [ Table 1 ]. X-ray (typical findings: lytic bone lesions), bone scan (increased uptake), CT (osteolytic lesions), MRI (signal changes), and [(18)F]-fluorodeoxyglucose positron emission tomography (FDG-PET) (hypermetabolic lesions) were used for the diagnosis of bone-marrow metastasis [ Table 1 ]. Other than cytopenia, DIC was observed in some cases [ Table 1 ]. Biochemical examination revealed elevation of LDH levels, ALP levels, or both in some cases [ Table 1 ]. In 80% of cases, prior resection of the primary brain lesion had been performed. Mutations in, for example, TP53, CDKN2A/B, or TERT were reported [ Table 2 ]. Treatment of bone-marrow metastasis with hematological abnormalities consists of various combinations of chemotherapy, palliative radiotherapy for skeletal pain, bisphosphonate treatment, transfusion, and supportive care [ Table 2 ]. The median overall survival time was 12 months, and the median survival time from the time of diagnosis of bone-marrow metastasis with hematological abnormalities was 2 months [ Table 2 ]. The median time from diagnosis of intracranial high-grade glioma to the diagnosis of bone-marrow metastases with hematological abnormalities was 9 months. In grade 4 glioma, including glioblastoma, grade 4 IDH-mutant astrocytoma, and H3 K27M-mutant diffuse midline glioma, the average age at diagnosis was 44.8 years (range, 12– 74 years), with 81% of patients being male [ Tables 1 and 2 ]. In grade 4 glioma, the median overall survival time was 9 months, and the median survival time from the time of diagnosis of bone-marrow metastasis with hematological abnormalities was 2 months. The median time from diagnosis of intracranial grade 4 glioma to the diagnosis of bone-marrow metastases with hematological abnormalities was 7.25 months. In grade 3 glioma, including anaplastic astrocytoma and anaplastic oligodendroglioma, the average age at diagnosis was 43.4 years (range, 28–71 years), with 64% of patients being male [ Tables 1 and 2 ]. In grade 3 glioma, the median overall survival time was 22 months, and the median survival from the time of diagnosis of bone-marrow metastasis with hematological abnormalities was 3.5 months. The median time from the diagnosis of intracranial grade 3 glioma to the diagnosis of bone-marrow metastases with hematological abnormalities was 21 months.

Table 1:

Clinical data on bone-marrow metastasis of high-grade glioma with hematological abnormalities, as described in the literature.

Table 2:

Additional clinical data on bone-marrow metastasis of high-grade glioma with hematological abnormalities, as described in the literature.

DISCUSSION

Here, we reported a case of diffuse bone-marrow metastasis of grade 4 IDH-mutant astrocytoma associated with hematological abnormalities, namely pancytopenia and probable DIC, with a dramatic increase in the serum level of LDH. The laboratory data during the patient’s first hospitalization at our hospital had shown a gradual increase in the level of LDH coincident with the signal changes in the first and fourth lumbar vertebrae on MRI, suggesting that bone-marrow metastasis of grade 4 IDH-mutant glioma with hematological abnormalities had already started at that time.

Grade 4 IDH-mutant astrocytoma featuring microvascular proliferation or necrosis in the WHO 2021 classification[ 3 ] was assigned to IDH-mutant glioblastoma in the 2016 classification [ 34 ], and it had been assigned to glioblastoma for a long time before the 2016 classification. Hence, we discuss our case by comparing it with metastatic glioblastoma and other cases of grade 4 glioma. However, the first histological diagnosis in our patient was anaplastic oligodendroglioma, so we also reviewed the data on metastatic grade 3 gliomas.

Smith et al.[ 45 ] reported that, among about 8000 neuroectodermal tumors primary to the CNS, there were 35 with extraneural metastasis, consisting of 23 cases of glioblastoma, 8 of medulloblastoma, and 4 of other tumors such as ependymoma and oligodendroglioma. A recent review of 211 unique patients with metastatic glioblastoma reported that the most metastatic site was the bone (47.9%).[ 19 ] Of the cases in another study with bone metastasis from glioblastoma, 63% involved the vertebral column, whereas the remainder involved lesions within the skull, sternum, rib cage, and appendicular skeleton.[ 46 ] Another study that reviewed 61 patients with metastatic oligodendroglioma reported that the most frequent metastatic site was the bone and bone marrow (42.7%), followed by the lymph nodes (20.0%), liver (6.4%), scalp (5.5%), and lung (5.5%).[ 20 ] Therefore, bones are the major metastatic sites of gliomas, but widespread bone metastasis associated with hematological abnormalities, as in our case, has rarely been reported. In the literature, we found only 18 similar cases of glioblastoma and two cases of H3 K27M-mutant diffuse midline glioma [ Table 2 ]. In grade 3 glioma, we found only three cases of anaplastic astrocytoma and 11 cases of anaplastic oligodendroglioma [ Table 1 ].

Although we found only a limited number of cases of bone-marrow metastasis of high-grade glioma associated with hematological abnormalities, the epidemiology derived from our review was as follows. In the grade 4 glioma cases, the mean age at diagnosis (44.8 years; range, 12–74 years) [ Table 1 ] was similar to the epidemiology of the overall cases of metastatic glioblastoma (mean age, 45.6 years) and cases of bone metastasis of glioblastoma (44 years).[ 19 , 46 ] However, bone-marrow metastasis in grade 4 glioma associated with hematological abnormalities showed a higher male predominance, with 81% of patients being male [ Table 1 ] than in the overall cases of metastatic glioblastoma (67%) or bone metastasis of glioblastoma (69.6%).[ 19 , 46 ] In grade 3 glioma, bone-marrow metastasis with hematological abnormalities was reported more often in anaplastic oligodendroglioma (n = 11) than in anaplastic astrocytoma (n = 3), suggesting that oligodendroglioma might have more affinity for the bone marrow and a greater tendency than astrocytoma to result in hematological abnormalities [ Table 1 ].

Analysis of the clinical profiles in bone-marrow metastasis of high-grade glioma associated with hematological abnormalities revealed that many patients had skeletal pain, which was also observed in our patient [ Table 1 ]. As hematological abnormalities, anemia, thrombocytopenia, pancytopenia, and (in a few cases) DIC were observed [ Table 1 ]. In many cases, resection of the primary brain tumor preceded the bone-marrow metastasis, but in some cases, bone-marrow metastasis occurred without resection surgery or seemed to occur before the resection surgery, suggesting that glioma cells can metastasize without surgical invasion [ Table 2 ]. After the diagnosis of bone-marrow metastasis with hematological abnormalities, the patients’ condition rapidly deteriorated, and they died [ Table 2 ]. The median overall survival in grade 4 glioma (9 months) was shorter than the overall survival of patients with bone metastasis from glioblastoma (16 months) [ Table 2 ].[ 46 ] These short survival times in patients with bone-marrow metastasis with hematological abnormalities might have resulted from the highly aggressive nature of the metastasis, which occurred in far more than a few bones – not only all lumbar and sacral vertebrae but also the ilium – in our patient. The hematological abnormalities and short survival times might have been the result of rapid, diffuse, and widespread bone-marrow metastasis. In our case, the patient had multiple metastases, including brain and bone metastases, 8 months after the first resection; he had hematological abnormalities 12 months after the first resection and died 2 weeks after the hematological abnormalities were found; this disease course was similar to those in previous reports [ Table 2 ]. In our case, the survival time was a little longer than the median overall survival of grade 4 glioma cases (9 months), suggesting that the relatively new treatment combination of TMZ, Bev, and SRT, or the patient’s IDH-mutant status, or both, might have been responsible.

In many cases of bone-marrow metastasis of high-grade glioma with hematological abnormalities, skeletal pain led to a detailed examination and contributed to the diagnosis of bone-marrow metastasis. Typical findings were lytic bone lesions on X-ray and CT, signal changes and postcontrast enhancement of bone on MRI, and increased uptake on bone scan and FDG-PET [ Table 1 ]. However, diffuse bone-marrow metastases were sometimes difficult to find on X-ray or CT, and in those cases, MRI and FDG-PET were found useful. In our case, MRI showed signal changes in the vertebrae very clearly, but there was a pitfall: when the diffuse bone-marrow metastasis progressed, all bones had signal changes on MRI [ Figure 3 ], and we could not find the abnormalities without normal past images for comparison. Biochemical examination revealed elevation of LDH levels in some cases [ Table 1 ]. Nagata et al.[ 31 ] reported that severe pancytopenia progressed rapidly, with increased LDH levels, before the diagnosis of bone-marrow metastasis in their patient. Our patient suffered from intractable low back pain with markedly elevated LDH levels and pancytopenia, which were consistent with the descriptions given in previous reports. In our patient, blood tests also showed polychromasia. Because marked increases in LDH-1 and LDH-2 with LDH-1 predominance can be observed in tumor lysis or hemolysis, or both, and polychromasia can be observed in hemolysis, we inferred that not only tumor lysis but also hemolysis occurred in our patient. Thus, skeletal pain with increased LDH-1 and LDH-2 levels with LDH-1 predominance and cytopenia during the treatment of high-grade glioma might indicate widespread and diffuse bone-marrow metastasis. Furthermore, in our patient, serum LDH elevation started 3 months before the diagnosis of bone-marrow metastasis with hematological abnormalities, suggesting that LDH is a sensitive predictor of bone-marrow metastasis with hematological abnormalities.

As treatments for bone-marrow metastasis of high-grade glioma with hematological abnormalities, various combinations of supportive care, transfusions, and palliative radiotherapy for skeletal pain were chosen [ Table 1 ]. When bone-marrow metastasis of high-grade glioma with hematological abnormalities was diagnosed, the patient’s condition was already too poor for him to undergo chemotherapy. In cases where the patient can tolerate chemotherapy, platinum-containing drugs or BCNU (carmustine) were administered in grade 4 glioma before TMZ went mainstream. In anaplastic oligodendroglioma, procarbazine–lomustine–vincristine chemotherapy has also been used [ Table 2 ]. Recently, TMZ Bev, or both, have been used in some cases [ Table 2 ]. As expected, the outcome of bone-marrow metastasis of high-grade glioma with hematological abnormalities has been poor.

The latest systematic review of glioblastoma with extraneural metastasis described an overall increased frequency of these cases.[ 19 ] The cause may be the improvement in, and increased use of, medical diagnostic tools and the increased survival of glioblastoma patients treated with improved treatment modalities. From 1940 to 2009, there was progressive lengthening of the interval from detection of extracranial metastasis to death, at a rate of 0.7 months per decade, and the use of MRI has been associated with an increase in overall survival.[ 23 ] The accumulated treatment effects of surgery, chemotherapy, molecularly targeted therapy, conventional and stereotaxic radiotherapy, tumor-treating fields, and cerebrospinal fluid shunting may be contributing to the longer survival times. In fact, patients treated with surgery, radiation, chemotherapy, and cerebrospinal fluid shunts have had the longest average survival intervals from metastasis to death.[ 23 ] The longer survival times from metastasis to death may enable the detection of the growth of extraneural metastatic tumors using advanced diagnostic modalities such as 3-T MRI or FDG-PET/CT.

The pathogenesis of high-grade glioma metastasis remains largely unknown.[ 3 ] Extension within and along perivascular spaces is typical, but invasion of the vessel lumen is not frequent, and penetration of the dura, venous sinuses, and bone is exceptional.[ 22 ] One hypothesis for how glioblastoma cells spread extra-axially is that tumor cells seeded in the scalp during surgery reprogram to a more mesenchymal-like cell type and spread to distant sites through the parenchyma of the surgical defect, local lymphatic channels, or vascular pathways.[ 19 ] In fact, metastases to local tissues such as the dura, subcutaneous tissue, skin, scalp, or parotid have been observed in 36.9% of metastatic glioblastoma cases.[ 19 ] However, as seen in some cases of bone-marrow metastasis from high-grade glioma, distant extraneural metastases can happen without prior surgery, and metastasis to remote sites such as bone, lung, or liver can occur without the tumor first metastasizing near the previous surgical site.[ 19 ] One of the hypothetical mechanisms of spread in these cases is through spontaneously circulating tumor cells (CTCs), and another is through the glymphatic system.[ 19 ] CTCs have been observed in the peripheral blood of 39% of patients with glioblastoma, and this phenomenon might explain the remote metastasis of glioma to such sites as the bone marrow.[ 48 ] As the transplantation of organs from patients with glioblastoma can cause metastases in the recipient patients, immunosuppressive treatment after organ transplantation might lower the defenses against extraneural adhesion and engraftment of CTCs.[ 4 , 40 ] In the review by Strong et al., the vertebral bodies were the most common sites of metastasis of glioblastoma.[ 46 ] They mentioned that one of the main mechanisms of vertebral metastasis of glioblastoma may involve a hematogenous route through the robust plexus surrounding the vertebral bodies.[ 46 ] The spine lies close to the spinal cord and surrounding dura.[ 46 ] Dural cells release chemokine (C-X-C motif) ligand 2 (CXCL2), which is overexpressed in glioblastoma and is important in glioblastoma progression through angiogenesis.[ 47 , 52 ] CXCL2 released by the dura might contribute to hematogenous vertebral metastasis from spinal cord lesions through angiogenesis. In our patient, the glioma cells metastasized to many bones in the absence of immunosuppressive treatment. We speculate that our patient’s glioma cells might have acquired the ability to escape immune surveillance and develop an affinity for the bone marrow at an early stage. In our patient, a small leptomeningeal metastatic lesion was concurrently present near two metastatic vertebral lesions [ Figure 2f ]. Tumor cells might have metastasized from the leptomeningeal lesion to the vertebrae through a hematogenous route, using the venous plexus or blood vessels newly generated by the action of CXCL2. However, why did the metastasizing glioma cells prefer the bone marrow to the surrounding muscles or connective tissues? Hira et al.[ 15 ] reported similarities between glioblastoma stem-cell niches and hematopoietic stem-cell niches in the bone marrow, showing that glioblastoma stem-cell niches expressed bone-marrow niches proteins such as stromal-cell-derived factor-1α, osteopontin, cathepsin K, CD44, hypoxiainducible factor-1α, and vascular endothelial growth factor. These similarities might explain the predilection of the glioblastoma cells for the bone marrow.

Some studies have focused on potential genomic drivers of extraneural metastasis in glioblastoma. TP53 mutations are highly enriched in glioblastomas with extraneural metastasis – particularly in patients with shorter survival times.[ 36 , 54 ] Other mutations observed in patients with extraneural metastatic glioblastoma include those of RB1, PTEN, TERT, ATRX, and NF1.[ 33 ] Zhang et al.[ 54 ] reported that grade 4 IDH-mutant (p.R132H) astrocytomas are at least as likely to metastasize as IDH wild-type glioblastomas. In our case, the glioma was a grade 4 IDH-mutant astrocytoma with overexpression of p53; it might, therefore, potentially have had genetic characteristics that led to extraneural metastasis. In a case of metastatic glioblastoma reported by Zhang et al.,[ 54 ] CIC mutation and MGMT promoter methylation were also observed with TP53 mutation; these gene mutation conditions were similar to those in our case. Although CIC loss results in the activation of proliferative pathways, inhibition of differentiation, and poorer outcomes in patients with a 1p19q co-deletion glioma,[ 13 ] it was not clear in our case whether CIC loss would have contributed to the extraneural metastasis. PIK3R1 mutations have been identified in patients with grade 4 IDH-mutant astrocytoma and are associated with shorter overall survival times.[ 5 ] These mutations might have contributed to the short survival time of our patient. Genetic alterations between the primary lesion and metastases have been reported,[ 28 , 36 ] but we wereunable to get permission to perform further gene analysis of the bone-marrow metastases.

The bone-marrow metastases with hematological abnormalities observed in our patient had many characteristics in common with DCBM. DCBM is characterized by widespread bone metastasis (i.e., bone-marrow infiltration) from solid tumors – especially gastric cancers – and is associated with hematological abnormalities such as DIC and microangiopathic hemolytic anemia.[ 17 ] The typical patient with DCBM complains of various combinations of low back pain and hemorrhagic or anemic symptoms, with elevated serum ALP or LDH levels or both. Some previous cases with bone-marrow metastasis of glioma have had findings similar to those of DCBM; they have been reported as “bone-marrow metastasis of glioblastoma mimicking acute myeloid leukemia” and “leukemia-like onset of bone-marrow metastasis from anaplastic oligodendroglioma.”[ 31 , 49 ] Because of the similarity of our case to DCBM, we diagnosed our patient with gliomatosis of the bone marrow.

In our case, during the early phase of bone-marrow metastasis, elevation of the serum LDH level began. This elevation may have indicated the start of tumor lysis hemolysis, or both, suggesting aggressive metastasis to the bone marrow. We, therefore, recommend a systemic diagnostic work-up – and especially whole-spine MRI – in patients diagnosed with high-grade gliomas if the patients show gradual but progressive elevation of LDH. On the contrary, if MRI reveals signs of bone metastasis, the level of LDH should be monitored. If bone metastasis on MRI is observed concomitant with an increase in LDH levels in a patient with high-grade glioma, then FDG-PET and, if possible, bone-marrow biopsy may be recommended as a further work-up.

Further studies are needed for us to understand the mechanisms of mutation better, spread, and predilection for the bone marrow. Once bone-marrow metastasis of high-grade glioma with hematological abnormalities has occurred, then the treatment options are very limited in terms of preventing hematological deterioration. To elucidate the etiology of this condition, we think that routine genetic and molecular analyses of not only primary but also metastatic high-grade gliomas should be performed. This would help to identify the potential clinicopathological associations between genetics and the pathophysiology of tumor metastasis selectively to the bone marrow.

CONCLUSION

We have presented a rare case of bone-marrow metastasis of astrocytoma, IDH-mutant, grade 4, with hematological abnormalities. Elevation of serum LDH levels preceded the pancytopenia, suggesting that elevated serum LDH is one of the predictors of bone-marrow metastasis of grade 4 glioma with hematological abnormalities.

Ethical approval:

Institutional Review Board approval is not required.

Declaration of patient consent:

The authors certify that they have obtained all appropriate patient consent.

Financial support and sponsorship:

Nil.

Conflicts of interest:

There are no conflicts of interest.

Use of artificial intelligence (AI)-assisted technology for manuscript preparation:

The authors confirm that there was no use of artificial intelligence (AI)-assisted technology for assisting in the writing or editing of the manuscript and no images were manipulated using AI.

Disclaimer

The views and opinions expressed in this article are those of the authors and do not necessarily reflect the official policy or position of the Journal or its management. The information contained in this article should not be considered to be medical advice; patients should consult their own physicians for advice as to their specific medical needs.

References

1. Aftahy AK, Butenschoen VM, Hoenikl L, Liesche-Starnecker F, Wiestler B, Schmidt-Graf F. A rare case of H3K27-altered diffuse midline glioma with multiple osseous and spinal metastases at the time of diagnosis. BMC Neurol. 2023. 23: 87

2. Al Sharie S, Talafha M, Abu Laban D, Al Awabdeh T, AlMousa A, Al-Masri N. H3 K27M-mutant diffuse midline glioma with osseous metastases: A case report and a literature review. Clin Neuropathol. 2022. 41: 263-70

3. Brat DJ, Cahill DP, Cimino PJ, Huse JT, Kleinschmidt-Demasters BK, Komori T, editors. Astrocytoma, IDH-mutant. WHO classification of tumours editorial board. Central Nervous System Tumours. Lyon, France: IARC; 2021. p. 19-27

4. Chen H, Shah AS, Girgis RE, Grossman SA. Transmission of glioblastoma multiforme after bilateral lung transplantation. J Clin Oncol. 2008. 26: 3284-5

5. Chen W, Guo S, Wang Y, Shi Y, Guo X, Liu D. Novel insight into histological and molecular astrocytoma, IDH-mutant, Grade 4 by the updated WHO classification of central nervous system tumors. Cancer Med. 2023. 12: 18666-78

6. Cordiano V, Miserocchi F, Storti M. Bone marrow metastases from anaplastic oligodendroglioma presenting with pancytopenia and hypogammaglobulinemia: A case report. Tumori. 2011. 97: 808-11

7. Dawson TP. Pancytopaenia from a disseminated anaplastic oligodendroglioma. Neuropathol Appl Neurobiol. 1997. 23: 516-20

8. Demeulenaere M, Duerinck J, Four SD, Fostier K, Michotte A, Neyns B. Bone marrow metastases from a 1p/19q Co-deleted oligodendroglioma-a case report. Anticancer Res. 2016. 36: 4145-9

9. Dewar JM, Dady PJ, Balakrishnan V. Metastatic astrocytoma. Aust N Z J Med. 1985. 15: 745-7

10. Didelot A, Taillandier L, Grignon Y, Vespignani H, Beauchesne P. Concomitant bone marrow metastasis of a glioblastoma multiforme revealed at the diagnosis. Acta Neurochir (Wien). 2006. 148: 997-1000

11. Friedman JH, Liu HM, Spremulli E, Calabresi P. Distant metastases from a malignant glioma: Unusual complications associated with treatment of a glioblastoma: Distant metastases and focal white matter degeneration. J Neurol Neurosurg Psychiatry. 1987. 50: 237-8

12. Gerrard GE, Bond MG, Jack AS. Bone marrow infiltration by a parietal lobe grade III oligodendroglioma. Clin Oncol (R Coll Radiol). 1995. 7: 321-2

13. Gleize V, Alentorn A, Connen De Kérillis L, Labussière M, Nadaradjane AA, Mundwiller E. CIC inactivating mutations identify aggressive subset of 1p19q codeleted gliomas. Ann Neurol. 2015. 78: 355-74

14. Heinig S, Aigner T, Bloß HG, Grabenbauer GG. Spinal and cervical nodal metastases in a patient with glioblastoma. Strahlenther Onkol. 2024. 200: 838-43

15. Hira VV, Wormer JR, Kakar H, Breznik B, Van Der Swaan B, Hulsbos R. Periarteriolar glioblastoma stem cell niches express bone marrow hematopoietic stem cell niche proteins. J Histochem Cytochem. 2018. 66: 155-73

16. Hsu E, Keene D, Ventureyra E, Matzinger MA, Jimenez C, Wang HS. Bone marrow metastasis in astrocytic gliomata. J Neurooncol. 1998. 37: 285-93

17. Iguchi H. Recent aspects for disseminated carcinomatosis of the bone marrow associated with gastric cancer: What has been done for the past, and what will be needed in future?. World J Gastroenterol. 2015. 21: 12249-60

18. Kleinschmidt-Demasters BK. Diffuse bone marrow metastases from glioblastoma multiforme: The role of dural invasion. Hum Pathol. 1996. 27: 197-201

19. Laurin BJ, Treffy R, Connelly JM, Straza M, Mueller WM, Krucoff MO. Mesenchymal-type genetic mutations are likely prerequisite for glioblastoma multiforme to metastasize outside the central nervous system: An original case series and systematic review of the literature. World Neurosurg. 2025. 193: 397-426

20. Li G, Zhang Z, Zhang J, Jin T, Liang H, Gong L. Occipital anaplastic oligodendroglioma with multiple organ metastases after a short clinical course: A case report and literature review. Diagn Pathol. 2014. 9: 17

21. LoRusso PM, Tapazoglou E, Zarbo RJ, Cullis PA, Austin D, Al-Sarraf M. Intracranial astrocytoma with diffuse bone marrow metastasis: A case report and review of the literature. J Neurooncol. 1988. 6: 53-9

22. Louis DN, Brat DJ, Ohgaki H, Stupp R, Suva ML, Biernat W, Louis DN, Ohgaki H, Wiestler OD, Cavenee WK, editors. Glioblastoma, IDH-wildtype. WHO classification of tumours of the central nervous system. Lyon: International Agency for Research on Cancer; 2016. p. 28-45

23. Lun M, Lok E, Gautam S, Wu E, Wong ET. The natural history of extracranial metastasis from glioblastoma multiforme. J Neurooncol. 2011. 105: 261-73

24. Malca SA, Roche PH, Pellet W. Secondary localizations of cerebral glioblastoma. Pathogenic and anatomoclinical focus apropos of a case of multiple bone metastases disclosed by vertebral involvement. Neurochirurgie. 1993. 39: 315-21

25. Martens T, Matschke J, Müller C, Riethdorf S, Balabanov S, Westphal M. Skeletal spread of an anaplastic astrocytoma (WHO grade III) and preservation of histopathological properties within metastases. Clin Neurol Neurosurg. 2013. 115: 323-8

26. Mazza E, Belli C, Terreni M, Doglioni C, Losio C, Cantore M. Breast metastases from oligodendroglioma: An unusual extraneural spread in two young women and a review of the literature. Crit Rev Oncol Hematol. 2013. 88: 564-72

27. Merrell R, Nabors LB, Perry A, Palmer CA. 1p/19q chromosome deletions in metastatic oligodendroglioma. J Neurooncol. 2006. 80: 203-7

28. Mohme M, Maire CL, Schliffke S, Joosse SA, Alawi M, Matschke J. Molecular profiling of an osseous metastasis in glioblastoma during checkpoint inhibition: Potential mechanisms of immune escape. Acta Neuropathol Commun. 2020. 8: 28

29. Montagne K, Marie B, Cahn V, Hennequin V, Didelot A, N’Seir R. Systemic metastasis at the time of diagnosis of a glioblastoma. Ann Pathol. 2004. 24: 268-70

30. Mousavi M. Bone marrow metastasis from glioblastoma multiforme. J Med Soc N J. 1980. 77: 904-5

31. Nagata A, Kanemasa Y, Kikuchi M, Otani R, Yamada R, Motoi T. Bone marrow metastasis of glioblastoma multiforme mimicking acute myeloid leukemia. Oxf Med Case Reports. 2020. 2020: omaa040

32. Newman HF, Howard GC, Reid PM. Metastatic oligodendroglioma presenting as a leukoerythroblastic anaemia. Eur J Surg Oncol. 1985. 11: 287-8

33. Noch EK, Sait SF, Farooq S, Trippett TM, Miller AM. A case series of extraneural metastatic glioblastoma at memorial Sloan Kettering cancer center. Neurooncol Pract. 2021. 8: 325-36

34. Ohgaki H, Kleihues P, Von Deimling A, Louis DN, Reifenberger G, Yan H, Louis DN, Ohgaki H, Wiestler OD, Cavenee WK, editors. Glioblastoma, IDH-mutant. WHO classification of tumours of the central nervous system. Lyon, France: IARC; 2016. p. 52-6

35. Papageorgiou C, Jourdi G, Adjambri E, Walborn A, Patel P, Fareed J. Disseminated intravascular coagulation: An update on pathogenesis, diagnosis, and therapeutic strategies. Clin Appl Thromb Hemost. 2018. 24: 8-28.S

36. Park CC, Hartmann C, Folkerth R, Loeffler JS, Wen PY, Fine HA. Systemic metastasis in glioblastoma may represent the emergence of neoplastic subclones. J Neuropathol Exp Neurol. 2000. 59: 1044-50

37. Rajagopalan V, El Kamar FG, Thayaparan R, Grossbard ML. Bone marrow metastases from glioblastoma multiforme-a case report and review of the literature. J Neurooncol. 2005. 72: 157-61

38. Romo CG, Palsgrove DN, Sivakumar A, Elledge CR, Kleinberg LR, Chaichana KL. Widely metastatic IDH1-mutant glioblastoma with oligodendroglial features and atypical molecular findings: A case report and review of current challenges in molecular diagnostics. Diagn Pathol. 2019. 14: 16

39. Rubinstein LJ. Development of extracranial metastases from a malignant astrocytoma in the absence of previous craniotomy. Case report. J Neurosurg. 1967. 26: 542-7

40. Seoane J, De Mattos-Arruda L. Escaping out of the brain. Cancer Discov. 2014. 4: 1259-61

41. Sharma A, Agarwal A, Sharma MC, Anand M, Agarwal S, Raina V. Bone marrow metastasis in anaplastic oligodendroglioma. Int J Clin Pract. 2003. 57: 351-2

42. Shimizu H, Motomura K, Ohka F, Aoki K, Tanahashi K, Hirano M. Multiple metastases to the bone and bone marrow from a 1p/19q-codeleted and IDH2-mutant anaplastic oligodendroglioma: A case report and literature review. Neurooncol Adv. 2020. 2: vdaa101

43. Shinmura F, Chen M, Itoh T, Ariwa R. An autopsy case of extraneural metastases of giant cell glioblastoma with intracerebral hemorrhage. No Shinkei Geka. 1985. 13: 1245-50

44. Simonetti G, Silvani A, Fariselli L, Hottinger AF, Pesce GA, Prada F. Extra central nervous system metastases from glioblastoma: A new possible trigger event?. Neurol Sci. 2017. 38: 1873-5

45. Smith DR, Hardman JM, Earle KM. Metastasizing neuroectodermal tumors of the central nervous system. J Neurosurg. 1969. 31: 50-8

46. Strong MJ, Koduri S, Allison JA, Pesavento CM, Ogunsola S, Ogunsola O. Bone metastasis from glioblastoma: A systematic review. J Neurooncol. 2022. 158: 379-92

47. Strong MJ, Rocco S, Taichman R, Clines GA, Szerlip NJ. Dura promotes metastatic potential in prostate cancer through the CXCR2 pathway. J Neurooncol. 2021. 153: 33-42

48. Sullivan JP, Nahed BV, Madden MW, Oliveira SM, Springer S, Bhere D. Brain tumor cells in circulation are enriched for mesenchymal gene expression. Cancer Discov. 2014. 4: 1299-309

49. Tanaka Y, Nobusawa S, Ikota H, Yokoo H, Hirato J, Ito H. Leukemia-like onset of bone marrow metastasis from anaplastic oligodendroglioma after 17 years of dormancy: An autopsy case report. Brain Tumor Pathol. 2014. 31: 131-6

50. Taule EM, Brekke J, Miletic H, Sætran H, Maric S, HogenEsch I. Breaking boundaries: A rare case of glioblastoma with uncommon extraneural metastases: A case report and literature review. Brain Spine. 2024. 4: 103927

51. Terheggen HG, Müller W. Extracerebrospinal metastases in glioblastoma. Case report and review of the literature. Eur J Pediatr. 1977. 124: 155-64

52. Urbantat RM, Blank A, Kremenetskaia I, Vajkoczy P, Acker G, Brandenburg S. The CXCL2/IL8/CXCR2 pathway is relevant for brain tumor malignancy and endothelial cell function. Int J Mol Sci. 2021. 22: 2634

53. Yung WK, Tepper SJ, Young DF. Diffuse bone marrow metastasis by glioblastoma: Premortem diagnosis by peroxidase-antiperoxidase staining for glial fibrillary acidic protein. Ann Neurol. 1983. 14: 581-5

54. Zhang X, Katsakhyan L, LiVolsi VA, Roth JJ, Rassekh CH, Bagley SJ. TP53 mutation and extraneural metastasis of glioblastoma: Insights from an institutional experience and comprehensive literature review. Am J Surg Pathol. 2021. 45: 1516-26