Primary Angiitis of Central Nervous System-Tumor-Like Mass Lesions: Report of Six Biopsy Confirmed Cases
Correspondence Address: Source of Support: None, Conflict of Interest: None DOI: 10.4103/0028-3886.314536
Source of Support: None, Conflict of Interest: None
Keywords: Diffusion restriction, granulomatous angiitis, heterointense lesion, lymphocytic angiitis, necrotizing angiitis, primary angiitis of central nervous system, tumor-like-mass lesion
Primary angiitis of the central nervous system (PACNS), angiitis confining to brain and spinal cord, is a very uncommon disorder with a reported annual incidence rate of 2.4 cases per person-years. An extremely rare subtype of PACNS is a tumor-like mass lesion (ML-PACNS). In the Cleveland Clinic/Massachusetts General Hospital study this lesion accounted for 4% of PACNS cohort and it was 6.8% in the Mayo Clinic PACNS cohort. In the French COVAC cohort (Cohort of Patients With Primary Angiitis of Central Nervous System), of the 60 adult patients with PACNS, 7 (11.7%) had ML-PACNS lesion on MRI. This report presents the clinical, pathological, and radiological characteristics of six patients with histologically proven ML-PACNS and also reviews the literature to characterize the clinical features of ML-PACNS.
This study is a report of six patients with histologically verified ML-PACNS and not a registry-based study. Case records of the six patients with ML-PACNS were reviewed and the data collected included demographic data, clinical characteristics, radiological, and histological features, and treatment outcomes. Four patients had partial excision of the lesion biopsy and two had biopsy of the lesion. Postoperatively all the six patients had work-up for systemic vasculitis which was negative. This study has the approval of the Institutional Ethics Committee (Institutional Ethics Committee meeting dated May 8, 2020, Item-17).
Radiological studies included contrast computer tomography (CCT) brain, plain and contrast magnetic resonance imaging (MRI). MRI sequences included diffusion-weighted imaging (DWI), apparent diffusion coefficient (ADC), T1 and T2 weighted, fluid-attenuated inversion recovery (FLAIR) and susceptibility-weighted imaging (SWI). The images were reviewed by a neuroradiologist. MRIs were done on 1.5 T machine. Lesions were classified into Type I infiltrative pattern with no clear demarcation from the brain parenchyma and Type II discrete lesions with or without perilesional edema.
The biopsies were processed for formalin-fixed paraffin-embedded sections and stained with hematoxylin and eosin (H&E), Masson trichrome (MT), Verhoef-Vangieson (VVG), Perl's stain, Zeihl-Nielson (Z-N) stain, Gomori methenamine silver (GMS) stain, and Congo red. Biopsies were reviewed for features of vasculitis: Transmural and perivascular infiltrate, composition of inflammatory infiltrate, disruption or duplication of internal elastic lamina, fibrinoid necrosis, fibrosis, and hyalinization. Other features noted were: Presence of extravasated red blood cells, hemosiderin pigment, gliosis, or infarct in the adjacent brain parenchyma Based on the histological features, vasculitis was classified as granulomatous, necrotizing, or lymphocytic type.,
Clinical characteristics, imaging features, pathological findings, treatment, and outcome are given in [Table 1]. The mean age was 44.5 years (range 25–68) and four were males. The duration of symptoms ranged between one week and 4 months. Presenting symptoms were headache (5), focal neurologic deficits (5), and seizures (4). Postoperative workup for systemic vasculitis was negative.
CCT brain was done in two patients and MRI was done in all the 6 patients. In three patients all the sequences including contrast study were done. One patient had no contrast study and another no DWI study. One patient had only a noncontrast MRI. We could not do a contrast study as he had elevated creatinine.
Of the six patients, 5 had Type II lesions and 1 had Type 1 lesion. The lesion was unifocal in two and multifocal in 4 patients. The lesion location was periventricular or deep white matter in 3, lobar in 2, and subcortical in one. The individual lesions in the cases with multifocal lesions and the cases with single lesion showed heterointense internal morphology with areas of diffusion restriction [Figure 1] and [Figure 2]. Different patterns of enhancement were observed, uniform, peripheral or heterogeneous. Three cases showed perilesional edema [Figure 3].
Vasculitis was classified as: Granulomatous in 1 [Figure 4]a lymphocytic in 3 [Figure 4]b, and necrotizing in 2 [Figure 4]c. Granulomatous vasculitis showed perivascular granulomas with foreign body type giant cells and lymphomononuclear infiltrate. Z-N and GMS stains for tuberculous bacilli and fungi respectively were negative. Amyloid staining was negative. Adjacent brain parenchyma showed infarcts. The necrotizing vasculitis seen in two samples showed fibrinoid necrosis, highlighted on MT with transmural and perivascular lymphocytic infiltrate. Disruption of the internal elastic membrane on VVG was seen in one. Organization and recanalization of the thrombus was seen in one [Figure 4]d, One sample showed hyalinized and thickened vessels in addition. Adjacent brain parenchyma showed infarction, gliosis, hemorrhages. and hemosiderin-laden macrophages [Figure 4]e and [Figure 4]f. Congored stain for amyloid was negative. Lymphocytic vasculitis seen in three patients, microscopic examination showed perivascular lymphocytic infiltration and hyalinized vessels. Infarction, gliosis. and hemosiderin-laden macrophages were seen in the adjacent brain parenchyma [Figure 4]d.
Of the six patients two died, one (lymphocytic vasculitis) due to neutropenic sepsis and another (necrotizing vasculitis) due to refractory status epilepticus. In the remaining four patients, prednisolone +/- cyclophosphamide treatment was stopped with the resolution of symptoms and neuroimaging findings. Three patients were back to their previous occupation. Case 3 had mild cognitive deficits and he could be provided with a semiskilled job. In all the four patients, antiseizure medication (ASM) was withdrawn after a 1 to 2 year of seizure freedom.
In this study, the mean age of the patients was 44.5 years (range 25-68). Similar young age was observed in all the reported series.,, In our series males were predominant similar to the observation in Mayo clinic series. However, equal gender distribution was reported in the literature., Comparison study in Mayo clinic series showed that patients with ML-PACNS were more likely to be men when compared to patients with PACNS. The common symptoms in these patients were headache, focal neurological deficits, diffuse deficits, and seizures. Both the Mayo clinic and French series observed that patients with ML-PACNS were more likely to have seizures at presentation when compared to patients with PACNS.,
The MRI features in patients with ML-PACNS reported in the literature were: Single or multiple; lobar or sub-lobar (white matter) location; almost exclusively supratentorial; and contrast-enhancing lesions with or without perilesional edema. Some lesions showed central necrosis,,,, In none of the patients reviewed had MRI-DWI or susceptibility-weighted (SWI) sequences. Other rarely documented features were small infarcts, single or multiple in the same or opposite hemisphere (4/13, 30%), and meningeal enhancement (4/13, 40%). Intralesional hemorrhage was described in one case reported by Kumar et al. and in three cases in French cohort (Supplemental Figure II: Tumor presentation of PACNS). Of these four patients, the lesions were hemorrhagic (1), partially hemorrhagic (2), and microbleeds (1).
The cases included in this series presented with unifocal or multifocal lesions. The individual lesions in the cases with multifocal lesions and the case with single lesion showed heterointense internal morphology with areas of diffusion restriction. Different patterns of enhancement were observed, uniform, peripheral or heterogeneous. Three cases showed perilesional edema. Although a larger case series would be more conclusive. Heterogeneous internal morphology, areas of diffusion restriction, postcontrast enhancement appeared to be consistent findings. Histopathological features such as acute infarcts, foci of hermitages, hemosiderin-laden macrophages, and gliosis seen in the biopsy material can explain the morphological features of the mass lesions seen on MR-imaging.
Biopsy evidence of angiitis is gold stranded for the diagnosis of PACNS. Histological types of vasculitis include Granulomatous, granulomatous arteritis associated with vascular amyloid deposits; lymphocytic, and necrotizing.,, In this study the distribution type of vasculitis was: Granulomatous vasculitis in 1, and lymphocytic in 3 and necrotizing vasculitis in 2 and none of the biopsies was positive for amyloid staining. In the study of 38 patients with ML-PACNS by Molloy et al. 20 (53%) had granulomatous vasculitis and 18 (47%) lymphocytic vasculitis. The distribution of vasculitis type in 71 patients in the three studies was: Granulomatous vasculitis in 31 (43.6%), lymphocytic vasculitis in 38 (53.5%) and necrotizing vasculitis in 2 (2.8%) patients.,,
With treatment the overall outcomes were better in patients with ML-PACNS. All the 13 patients in the Mayo Clinic series responded to therapy and recovered with no, slight, or moderate disability. Of the 38 patients reviewed by Molly et al. 26 (68%) patients entered remission as defined by regression of the lesion and resolution or stabilization of neurological symptoms. There were 6 (16%) deaths. Poor outcomes were seen in amyloid group, 5 deaths. Of the non-amyloid group, better outcomes were seen in the patients treated with corticosteroids and cyclophosphamide as compared with patients treated with corticosteroids alone., In our series there were two deaths, one related to the treatment-related neutropenia and sepsis and one related to the disease-related complication, refractory status epilepticus.
Patients with ML-PACNS are likely to be younger at onset of symptoms. The presenting clinical symptoms are similar to any patient with intracranial space-occupying mass lesion. These patients are more likely to have seizures at presentation. A consistent image morphology of the lesion were: Heterogeneous internal morphology, internal areas of DWI and post-contrast enhancement. We suggest that in young patients with features of space-occupying and seizures, these MRI features may give a clue for the possible diagnosis of ML-PACNS preoperatively.
Institutional ethics committee approval
This study has the approval of the Institutional Ethics Committee (Institutional Ethics Committee meeting dated May 8, 2020, Item-17).
Data sharing statement
Not applicable, all the data is given in [Table 1].
This study was not funded by any funding agency.
Patient and Public involvement
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
[Figure 1], [Figure 2], [Figure 3], [Figure 4]