Photodynamic therapy for recurrent spinal diffuse midline glioma with a modest extension of overall survival: illustrative case

Abstract

BACKGROUND

Diffuse midline glioma (DMG), a highly malignant CNS tumor, is classified as a World Health Organization grade 4 tumor. Its overall survival prognosis, even after standard treatment, is typically extremely poor. Photodynamic therapy (PDT) represents an innovative treatment modality with promising results for intracranial malignant gliomas. However, its use for spinal malignant gliomas has not yet been explored, despite their pathological similarity to intracranial malignant gliomas.

OBSERVATIONS

The present case of a 28-year-old female patient with recurrent spinal DMG represents the first documented report concerning the effects of PDT on spinal malignant gliomas, with the pathological postmortem findings. Despite the successful application of PDT to the resection cavity of the spinal DMG, cervical spinal cord edema developed soon after surgery. The blood flow to the spinal cord parenchyma was thought to be altered but reversible. The patient’s overall survival was 21 months after the initial surgery.

LESSONS

This case highlights the potential of PDT as a feasible and promising adjunctive treatment for spinal malignant gliomas. The observed clinical course and pathological findings may provide valuable insights for future research and therapeutic development.

https://thejns.org/doi/10.3171/CASE25751

Diffuse midline glioma (DMG), classified as a World Health Organization (WHO) grade 4 tumor since the discovery of the associated H3 K27M mutation in 2016, is one of the most malignant CNS tumors.¹ It predominantly develops in midline structures such as the thalamus, brainstem, and spinal cord, with spinal DMG being particularly rare.^2,3 Case reports concerning this specific subtype are therefore limited in the literature. The imaging characteristics and treatment options for spinal DMG, as well as spinal glioblastoma (GBM), which is similarly classified as a WHO grade 4 tumor, remain poorly understood. The standard strategy for spinal intramedullary malignant tumors involves biopsy or partial resection followed by chemoradiation therapy, as maximal resection has not been shown to improve survival compared with less extensive procedures.⁴ The prognosis of spinal GBM is typically quite poor, with a median survival of approximately 12 months, while that of spinal DMG is approximately 17.0 ± 3.7 months.^3,5 Similarly to the case for intracranial malignant gliomas, improving treatment outcomes for spinal GBM and DMG remains a significant challenge.

Photodynamic therapy (PDT) is an innovative and recently developed treatment modality that has shown promising results in managing intracranial malignant gliomas, potentially outperforming standard therapies for this tumor type.^6,7 PDT has already been approved for intracranial malignant gliomas under Japan’s public health insurance system. However, its application for spinal malignant gliomas has not yet been approved, despite their pathological similarity to intracranial tumors, suggesting the potential for comparable efficacy. To assess the safety and efficacy of PDT in treating malignant spinal gliomas, we participated in an investigator-initiated phase II clinical trial focused on this indication.⁸ Herein, we present the case of a patient with recurrent spinal DMG who underwent PDT following partial tumor resection. Postmortem pathological analysis was conducted to evaluate the histological impact of PDT on the spinal cord tissue.

Illustrative Case

PDT in the Spinal Cord

The PDT was conducted in an investigator-initiated phase II clinical trial to evaluate its efficacy and safety in a patient with spinal intramedullary malignant glioma (jRCT2021220040). The photosensitizer utilized in this study was Laserphyrin (talaporfin sodium), manufactured by Meiji Seika Pharma Co. Treatment was administered using a PDT semiconductor laser device (Meiji Seika Pharma Co.). The protocol adhered to the established methods and conditions previously validated for PDT in primary malignant brain tumors.

Ethics Statement

The Ethics Committee of Osaka Medical and Pharmaceutical University (Osaka, Japan) approved this clinical trial. All procedures adhered to the ethical guidelines of the Institutional Research Committee, following the Declaration of Helsinki and its subsequent revisions or equivalent ethical standards. Given the prospective design and use of anonymized clinical data, written informed consent was obtained from the patient before the trial commenced.

Treatment Protocol

The treatment protocol was identical to that used for primary malignant brain tumors. The patient received 40 mg/m² talaporfin sodium the day before surgery (i.e., 22–24 hours prior to PDT). A partial or extensive resection of the spinal tumor was performed, and the resected specimen was submitted for pathological examination. PDT was administered to the resection cavity (one target site per area), with irradiation for 3 minutes at 150 mW/cm² and an energy dose of 27 J/cm² (i.e., the same conditions used for intracranial malignant gliomas).

Patient History and Examination

A 28-year-old woman previously presented to a hospital with complaints of low back pain, urinary dysfunction, and progressive muscle weakness in both lower limbs. MRI performed at that time revealed a tumorous lesion in the thoracic spine (Fig. 1A and B). Two months after the initial evaluation, she underwent an initial surgery at that institution to resect the intramedullary thoracic spine tumor (T8–10). The pathological diagnosis was confirmed as H3 K27M–altered DMG. After surgery, the patient received adjuvant chemotherapy and radiation therapy (45 Gy delivered in 25 fractions) targeting the affected spinal region (T8–12). However, local recurrence was suspected in the thoracic spinal cord 7 months after this initial surgery (Fig. 1C), leading to the initiation of bevacizumab therapy.

FIG. 1.

Serial sagittal MR images. A: T2-weighted image obtained before the initial surgery, suggesting an intramedullary tumor at the T8–10 level. B: T1-weighted contrast image obtained before the initial surgery, suggesting abnormal spinal cord enhancement at the T10 level. C: T2-weighted image obtained 7 months after the initial surgery, suggesting local recurrence at the T6–8 level. D: T2-weighted image obtained 9 months after the initial surgery, suggesting extensive growth of the recurrent tumor over the T3–10 region.

The patient was therefore referred to our institution for further evaluation and treatment. Her motor function had already deteriorated to near-complete paraplegia in both lower limbs, and her neurological function was assessed as grade 5 according to the modified McCormick functional schema. MRI suggested extensive growth of the recurrent tumor, distributed along T3–10 in the thoracic spine (Fig. 1D). Ten months after the initial surgery, the patient underwent a second surgery to remove the recurrent tumor, followed by PDT. A thoracic spinal lesion (T3–7) was exposed, and partial removal was performed. PDT was then successfully administered to the resection cavity (T3–5) (Fig. 2A–D). Pathological examination confirmed the diagnosis of DMG, H3 K27M–altered.

FIG. 2.

Intraoperative photographs taken during the second surgery. A: The recurrent tumor at the T3–5 level was exposed via left posterolateral sulcus myelotomy. B:Successful removal of the recurrent tumor at the T3–5 level. C and D: PDT was administered to the resection cavity (one target site per area) with irradiation for 3 minutes at 150 mW/cm² and an energy dose of 27 J/cm. Note the normal spinal cord surface was covered, such that the PDT laser was solely delivered in the resection cavity of the spinal cord.

The patient experienced no new neurological symptoms immediately after the second surgery with PDT; however, she unexpectedly developed motor weakness in both upper extremities 3 days later. MRI suggested an extension of intramedullary high signals on T2-weighted images of the cervical spine versus images obtained preoperatively (Fig. 3A–D). The intramedullary signal changes gradually disappeared by approximately 10 weeks after the second surgery, and cystic changes were noted in the thoracic spinal cord (Fig. 3E and F). Her upper extremity weakness began to improve approximately 3 weeks after surgery and had recovered to the preoperative level by around 10 weeks postoperatively, coinciding with the resolution of the intramedullary signal changes on MRI. The intramedullary high signal intensity on T2-weighted MRI of the cervical spine was found to be reversible, suggesting edema formation possibly caused by impaired blood flow to the spinal cord parenchyma. Follow-up brain MRI suggested a distant metastasis to the cerebellum. A stereotactic tumor biopsy confirmed that the cerebellar lesion was compatible with DMG, indicating leptomeningeal dissemination. Radiotherapy (60 Gy delivered in 30 fractions) was administered to treat the cerebellar lesions. MRI performed 5 months after the second surgery showed further growth of the tumor at the T3–5 level, but a significant atrophic change in the spinal cord at the T7–10 was also noted (Fig. 3G and H). Radiotherapy targeting the cervical and thoracic spine (C3–T5) was planned at a dose of 36 Gy over 12 fractions. However, this treatment was discontinued after 24 Gy was delivered in 8 fractions owing to significant deterioration of the patient’s respiratory function that eventually necessitated tracheal intubation. She also developed severe headaches and decreased consciousness as a result of hydrocephalus. A ventriculoperitoneal shunt was placed 7 months after the second surgery to relieve symptoms. Despite these interventions, her condition continued to decline over time, and she eventually died 11 months after the second surgery (21 months after the initial surgery).

FIG. 3.

Serial sagittal MR images. A and B:T2-weighted images of the cervical and thoracic regions obtained before the second surgery (also shown in Fig. 1D). C and D:T2-weighted images obtained 3 days after the second surgery, suggesting an extension of high signal intensity in the intramedullary regions of the cervical and thoracic spine. E and F: T2-weighted images obtained 10 weeks after the second surgery, suggesting complete resolution of the high intramedullary cervical spine signal, as well as cystic changes in the thoracic spinal cord. G and H: T2-weighted images obtained 5 months after the second surgery, suggesting further growth of the tumor at the T3–5 T5 levels, as well as significant atrophic spinal cord changes at the T7–10 level.

Postmortem Pathological Findings

A postmortem autopsy was conducted with the consent of the patient’s family. Separate pathological examinations of local spinal cord regions in the thoracic spine were done for the regions that were treated with PDT followed by radiotherapy (24 Gy) at the T3–5 level and those only treated with radiotherapy (45 Gy) at the T8 level.

Tumor progression and prominent tumor invasion of the venous vascular structure were observed in the T3–5 spinal cord treated with PDT and radiation therapy. Furthermore, inflammatory cell infiltration was also noted, accompanied by coagulative necrosis. No degenerative changes were observed in the mixed vascular endothelium (Fig. 4A–C). In contrast, extensive coagulative necrosis was observed in the T8 spinal cord treated with radiation therapy alone, with evidence of tumor proliferation within the necrotic area. No degenerative changes were observed in the mixed vascular endothelium (Fig. 4D and E).

FIG. 4.

Postmortem pathological H&E findings. Pathological examination of the thoracic spinal cord was performed separately for the regions that were treated with PDT followed by radiotherapy (i.e., T3–5; A–C) and those that were treated with radiotherapy alone (T8; D and E). A and B:Tumor progression (A) and prominent tumor invasion (B) of the venous vascular structure were observed in the T3–5 spinal cord region that was treated with PDT followed by radiotherapy. C: Inflammatory cell infiltration was also observed, accompanied by coagulative necrosis. No degenerative changes were observed in the mixed vascular endothelium after PDT (A and B). D: Extensive coagulative necrosis was observed in the T8 spinal cord region (treated with radiotherapy only), with evidence of tumor proliferation within the necrotic area. E: No degenerative changes were observed in the mixed vascular endothelium. Original magnification ×200 (A and E), ×100 (B and C), and ×40 (D).

Informed Consent

The necessary informed consent was obtained in this study.

Discussion

Observations

The main observation in this case was the histopathological findings following PDT for recurrent spinal DMG. Postmortem analysis revealed tumor progression with venous invasion and coagulative necrosis, but no endothelial degeneration or thrombus formation in the PDT-treated region. In contrast, radiation therapy–only areas showed more extensive coagulative necrosis. These results suggest that PDT did not cause direct vascular damage in spinal cord tissue. Additionally, transient cervical cord edema was observed postoperatively, likely representing a reversible vascular or reperfusion-related event. To our knowledge, this is the first report to document both the clinical course and pathological effects of PDT in spinal malignant glioma, highlighting its potential feasibility while emphasizing the need for further validation.

PDT involves three key components: a photosensitizer, light, and oxygen. These components trigger a photochemical reaction that generates singlet oxygen, a highly reactive molecule that causes cell death through either apoptosis or necrosis. The antitumor effects of PDT are mediated by three primary mechanisms: direct cytotoxicity to tumor cells, vascular damage to the tumor, and an inflammatory response that may enhance the systemic immunological tumor response.⁹ PDT has been approved for treating intracranial malignant gliomas by Japan’s public health insurance, where it has demonstrated a significant survival benefit versus the standard treatments.^6,7 Given that spinal malignant gliomas are pathologically identical to their intracranial counterparts, a similar efficacy level is expected. Although the patient survived for 21 months after the initial surgery, this observation cannot confirm any survival benefit due to the absence of a control group. Nonetheless, the case provides valuable insights into the possible clinical feasibility of PDT for spinal malignant gliomas.

In this case, PDT was successfully applied to the resection cavity after the partial resection of a previously treated spinal DMG. However, edema of the cervical spinal cord was observed during the first 3 days postoperatively. The blood flow to the spinal cord parenchyma was thought to be severely but reversibly altered by 10 weeks postoperatively. This event occurring soon after resection followed by PDT was considered a potentially serious adverse PDT-related event by the efficacy and safety evaluation committee for this investigator-initiated phase II clinical trial, but the cause of this edema was never clearly determined. The possibility of venous congestion of the entire spinal cord after partial resection of the radically recurrent tumors was discussed. Reperfusion injury, also known as white cord syndrome, may have been caused by spinal cord decompression following tumor removal.¹⁰ However, the edema was more widespread throughout our patient’s spinal cord parenchyma, possibly indicating that reperfusion injury alone was not responsible. Recently, Karita et al. reported postoperative acute swelling of the spinal cord in a case of DMG.¹¹ As described by the authors, the possibility that surgical intervention itself might induce rapid spinal cord swelling must be considered.

Postmortem pathological examinations of our patient’s thoracic spinal cord were performed separately for the region that was treated with PDT followed by radiotherapy (T3–5) and the one that only received radiotherapy (T8). No endothelial degeneration or thrombus formation was observed in either region. The radiation therapy–only T8 region exhibited more extensive coagulative necrosis than the T3–5 region, which received radiation and PDT. Previous studies have indicated that PDT does selectively target tumors through the vascular shutdown theory, including endothelial degeneration followed by thrombus formation in the tumor vessels as one of the mechanisms.^10,12 Suzuki et al. used human umbilical vein endothelial cells and found that PDT induced destruction of the tumor vessels via the RhoA/Rho-assisted protein kinase pathway by activating the Rho-GTP pathway.¹² In the current case, endothelial degeneration and thrombus formation were not observed in the vessels of the spinal cord after PDT, and there was no clear evidence of vascular damage caused by PDT. The most significant finding of this case lies in the pathological assessment performed after PDT, which revealed detailed histological changes in the treated spinal cord. These observations provide important preliminary data for understanding the tissue-level effects of PDT in spinal malignant gliomas.

Previous studies on intracranial GBM have reported a therapeutic effect of PDT at a penetration depth of 9–18 mm.¹³ Considering the much thinner diameter of the spinal cord than that of the brain, the safety and efficacy of PDT in the spinal cord malignant glioma must be reassured through an investigator-initiated clinical trial of PDT for spinal cord malignant glioma.⁸

This case report highlights the potential of PDT as a novel therapeutic option for spinal malignant gliomas. However, the findings should be interpreted with caution since this is a single case report of an extremely rare case of recurrent malignant spinal cord glioma. Further research should lay the foundation for establishing PDT as a promising treatment modality for this application, potentially improving outcomes for patients with this particularly aggressive malignancy.

Lessons

This is the first report of PDT for a case of recurrent spinal DMG with postmortem pathological evaluation. Pathological analysis showed tumor progression with venous invasion and necrosis but no endothelial degeneration or thrombus formation, providing important insights into the tissue-level effects of PDT. The transient cervical cord edema observed after treatment was regarded as a postoperative complication, suggesting that vascular or reperfusion-related mechanisms may have been involved. While the patient’s survival was relatively long compared with that in previous reports, a direct survival benefit from PDT cannot be confirmed from a single case. Nevertheless, this report demonstrates the technical feasibility of PDT in the spinal cord and highlights its potential as a therapeutic option for spinal gliomas. The ongoing phase II clinical trial is expected to further clarify its safety and efficacy, helping to establish optimal treatment parameters for future clinical application.

Acknowledgments

We thank all the members of this investigator-initiated phase II clinical trial (jRCT2021220040) and the Clinical Trial Administration Office for their generous support.

Disclosures

The authors report no conflict of interest concerning the materials or methods used in this study or the findings specified in this paper.

Author Contributions

Conception and design: Takami, Fukumura, Kazuki, Kashiwagi, Endo, Wanibuchi. Acquisition of data: Takami, Fukumura, Tani, Kazuki, Kashiwagi, Kosaka, Kawabata. Analysis and interpretation of data: Takami, Fukumura, Kazuki, Kashiwagi, Nonoguchi, Endo. Drafting the article: Takami, Fukumura, Kazuki, Kawabata. Critically revising the article: Takami, Fukumura, Kazuki, Kawabata, Endo. Reviewed submitted version of manuscript: Takami, Fukumura, Tani, Kazuki, Kosaka, Kawabata, Nonoguchi. Approved the final version of the manuscript on behalf of all authors: Takami. Administrative/technical/material support: Takami, Kawabata, Nonoguchi, Endo. Study supervision: Takami, Endo.

Correspondence

Toshihiro Takami: Osaka Medical and Pharmaceutical University, Takatsuki, Osaka, Japan. toshihiro.takami@ompu.ac.jp.