Pathologic Features of Brain Hemorrhage After Radiation Treatment: Case Series with Somatic Mutation Analysis

Abstract

Background:

Radiation treatment for diseases of the brain can result in hemorrhagic adverse radiation effects. The underlying pathologic substrate of brain bleeding after irradiation has not been elucidated, nor potential associations with induced somatic mutations.

Methods:

We retrospectively reviewed our department’s pathology database over 5 years and identified 5 biopsy specimens (4 patients) for hemorrhagic lesions after brain irradiation. Tissues with active malignancy were excluded. Samples were characterized using H&E, Perl’s Prussian Blue, and Masson’s Trichrome; immunostaining for B-cells (anti-CD20), T-cells (anti-CD3), endothelium (anti-CD31), macrophages (anti-CD163), α-smooth muscle actin, and TUNEL. DNA analysis was done by two panels of next-generation sequencing for somatic mutations associated with known cerebrovascular anomalies.

Results:

One lesion involved hemorrhagic expansion among multifocal microbleeds that had developed after craniospinal irradiation for distant medulloblastoma treatment. Three bleeds arose in the bed of focally irradiated arteriovenous malformations (AVM) after confirmed obliteration. A fifth specimen involved the radiation field distinct from an irradiated AVM bed. From these, 2 patterns of hemorrhagic vascular pathology were identified: encapsulated hematomas and cavernous-like malformations. All lesions included telangiectasias with dysmorphic endothelium, consistent with primordial cavernous malformations with an associated inflammatory response. DNA analysis demonstrated genetic variants in PIK3CA and/or PTEN genes but excluded mutations in CCM genes.

Conclusions:

Despite pathologic heterogeneity, brain bleeding after irradiation is uniformly associated with primordial cavernous-like telangiectasias and disruption of genes implicated in dysangiogenesis but not genes implicated as causative of cerebral cavernous malformations. This may implicate a novel signaling axis as an area for future study.

INTRODUCTION

Benefits of radiation treatment are severely limited by adverse radiation effects (ARE), also referred to as radionecrosis, with a reported incidence of 5–68% after brain irradiation(1). Clinical and imaging definitions have not been consistent(2), and the specific prevalence of brain bleeding within the context of ARE is not known. Pathologic substrates and molecular features of hemorrhagic ARE have not been systematically studied(3).

Herein we report a series of 5 ARE lesions with confirmed brain hemorrhage in 4 patients. We characterize the pathologic features of hemorrhagic AREs and any associated somatic mutations related to dysangiogenesis.

METHODS

Our department’s neurovascular pathology database was queried from 2018–2022 and identified 4 patients who underwent 5 biopsies for ARE, and had manifested evidence of hemorrhage on brain MRI (Figure 1). Tissues with active malignancy were excluded. Search criteria and methods for histology, immune-multiplexing, immunohistochemistry, DNA sequencing and mutation analyses are detailed in the Supplementary Methods. All patients had provided written consent for tissue collection in accordance with the Declaration of Helsinki and approved by our Institutional Review Board (IRB10–295-A). This case series is reported according to the CARE (Case Report) guidelines(4).

Figure 1. Axial FLAIR Magnetic Resonance Images of Hemorrhagic AREs.

(A) Case 1- ARE after radiation treatment for AVM. (B) Case 2- ARE after radiation treatment for AVM. (C) Case 3- ARE after craniospinal radiation for medulloblastoma. (D) Case 4A- ARE outside of primary treatment field for AVM. (E) Case 4B- ARE in the AVM bed after radiation treatment. Additional clinical characteristics in Table S1.

RESULTS

Histopathologic Characterization

One lesion developed among multifocal microbleeds outside the tumor bed after craniospinal irradiation for medulloblastoma treatment. Three cases arose in the bed of focally irradiated arteriovenous malformations (AVM) after confirmed obliteration. A fifth specimen was isolated from a site distinct from an irradiated AVM bed. The clinical presentations and histological findings are detailed in the Supplementary Results (Tables S1 and S2). We identified encapsulated hematomas and cerebral cavernous malformation (CCM)-like lesions, with common features of telangiectasias and inflammatory cell infiltrates (Figure 2).

Figure 2. Representative Histological Analysis of Hemorrhagic AREs with H&E and Immunofluorescent Staining for Inflammatory Cells.

(A) Cavernous vessels in a bed of organizing hemorrhage in Case 3 (H&E, bar=100μm). (B) Cerebellar encapsulated hematoma far from the irradiated AVM bed in Case 4A (H&E, bar=1mm). (C) Telangiectasias with fibrin-like substance extravasation in Case 1 (H&E, bar=50μm). (D) Macrophage (arrowheads) infiltration observed around caverns in Case 4B (immunofluorescence, CD31-red, nuclei-blue, CD163-green, bar=50μm). (E) B and T cell infiltration of telangiectasias in Case 2 (immunofluorescence, CD31-red, nuclei-blue, CD20-green, CD3-yellow, bar=50μm). For additional characteristics and feature comparison between cases please refer to Table S2 and Fig. S1.

Genetic Variant Analysis

We identified genetic variants [single nucleotide variant (SNV), insertion or deletion] as detected by next generation sequencing (NGS), along with their functional consequences using UCSC Genome Browser (http://genome.ucsc.edu), in 3 specimens with DNA of sufficient quality/quantity for analysis (>5 ng/ul). Pathogenetic variants, per ClinVar database (https://www.ncbi.nlm.nih.gov/clinvar/), in PTEN or PIK3CA genes were identified in 2 specimens(5). Variants in PTEN, PTENP1, BRAF and CREBBP were also identified in all three samples (Figure 3). A full list of detected variants is detailed in Supplemental Results (Table S3).

Figure 3. Variants Identified by NGS in 3 Specimens and Related Signaling Mechanisms.

Genetic variant analysis suggests that AREs may occur similar to the known pathophysiology of CCM endothelial dysmorphism. Here we show candidate genes across three patients to motivate a novel framework for hemorrhagic ARE with possible shared pathways seen in CCM disease. (A) Venn diagram showing overlap between cases. (Cases 1 & 2: List of genes identified from AREs after radiation treatment of AVMs; Case 3: List of genes identified from AREs [far from tumor bed] after radiation treatment for anaplastic medulloblastoma). Genetic variants identified in all sequenced samples, confirmed in two panels, color-coded by most severe variant found in each subgroup. (B) Signaling mechanisms linked to genetic variants. ¹Variants present in different chromosomal positions. ²Previously reported pathogenetic variant per ClinVar database.

DISCUSSION

It has been shown that radiation triggers endothelial injury with fibrin deposition in blood vessels, leading to progressive luminal occlusion and wall thickening in association with immune cell infiltration(6). This likely results in ischemia and infarction(3), and a hypoxia response mechanism that initiates neovascularization, with VEGF implicated as a substrate(1, 3).

Previous reports have described post-radiation microbleeds, CCMs, pseudocavernomas, cystic hemorrhagic radiation necrosis, and encapsulated hematomas(3, 7) in hemorrhagic ARE cases, with an unknown respective prevalence. These features were also identified in our cases, with CCM-like malformations and encapsulated hematoma in the different cases. Despite this pathologic heterogeneity, all 5 biopsies involved common features of dilated capillaries (telangiectasias) and robust immune cell infiltration. Interestingly, telangiectasias are the earliest lesions that develop in mouse CCM models(8). Yet, immune cell infiltrates in our cases did not involve the dominant B-cell response reported in CCM(9), and we excluded mutations of CCM1, 2, or 3 genes in any of our specimens.

Interestingly, all specimens regardless of dominant histopathology or underlying disease motivating radiation, manifested genetic variants in PIK3CA and/or PTEN, reflecting pathways involved in endothelial dysfunction and vascular malformation development(10) (Figure 3). Common BRAF SNVs were also identified in all specimens, and these have been implicated in endothelial dysfunction(11). A CREB-Binding Protein (CREBBP) SNV was also identified in all 3 of our samples with potential implications to CCL8 expression and macrophage infiltration(12), as part of the known irradiation response. Clonal expansion and infiltration of T cells have also been shown to be mediated by microglia-derived CCL2/CCL8 chemokines in animal models of ARE(13).

Our results are hypothesis-generating regarding potential signaling aberrations contributing to brain bleeding after irradiation. This study was limited by the small sample size, the lack of control cases of ARE without hemorrhage, and non-lesional tissue outside of the pathology. Future studies will be necessary to confirm these findings with proper controls and in animal models to identify the role of these genes and signaling pathways in radiation damage and brain bleeding. Transcriptomic studies of different cell types in the biopsied tissue will also contribute mechanistic insights, as has been done in cavernous malformations(14) and cancer(15).

In conclusion, we describe five specimens of brain hemorrhage associated with ARE. These reveal common pathologic features of telangiectasias and immune cell infiltrates, and somatic mutations implicating vascular dysmorphism and an immune response. We propose a signaling axis that involves regulators of endothelial function and immune response as an area of future study, to be confirmed in controlled analyses and in cases with adverse radiation effects with and without bleeding.

Highlights.

• Hemorrhagic AREs display primordial cavernous-like changes and telangiectasias

• Hemorrhagic AREs have been called cavernous-like malformations or pseudocavernomas

• Gene sequencing shows variants in PIK3CA and PTEN but not in CCM genes

• Mutations converge on pathways implicated in endothelial dysfunction