Inflammatory CADASIL: the cross-link between CADASIL and multiple sclerosis: a report of two cases and systematic review of the literature Authors: Ali Mahmoud Ahmed, Niraj Mistry, Ahmed Samir Gehad, Abdelmonem M. Hassan Published: 2026 ABSTRACT BackgroundMultiple Sclerosis (MS) is sometimes misdiagnosed in people with cerebral autosomal dominant arteriopathy with subcortical infarcts and leukoencephalopathy (CADASIL). We aimed to discuss two CADASIL patients who were misdiagnosed with MS and review the literature for similar cases.MethodsIn addition to reporting our two cases, we searched the literature and all available databases for CADASIL cases with clinical or radiological features suggestive of MS.ResultsWe reported two CADASIL cases misdiagnosed as MS. The review identified 12 reports describing 19 cases with CADASIL and inflammatory or MS-like features. The most prevalent features were sensory manifestations (76.2%), followed by motor manifestations (71.4%), gait imbalance (66.7%) and migraine (61.9%). Majority of the patients (90.5%) had either small vessel disease, demyelinating lesions, or both. Our cases shared these predominant features.Conclusion CADASIL may occasionally present with MS like inflammatory features, . The term “inflammatory CADASIL” can be considered when such red flags are present in atypical MS patients. Genetic testing is imperative for suspected CADASIL.Supplementary InformationThe online version contains supplementary material available at 10.1186/s12883-026-04772-1. Introduction Multiple sclerosis (MS) is a chronic disease that affects the central nervous system (CNS) [1]. The underlying mechanism is autoimmune, compounded by other mechanisms, including failure of remyelination [2]. Proposed causes for this include immune dysregulation, genetics, and environmental factors, such as viral infections [3]. The symptoms of MS widely vary from very mild to significant debilitating and can involve any region of the CNS, including the optic nerve, brainstem, spinal cord, and cerebral hemispheres [3]. Recent advances in MS management significantly improve outcomes, especially with early diagnosis and management [4]. However, owing to the wide variety of MS initial symptoms, overlapping with those of MS mimics [5], some conditions can initially present with clinical features, biomarkers and MRI findings similar to MS, leading to erroneous diagnosis of MS [6]. One such mimic is cerebral autosomal dominant arteriopathy with subcortical infarcts and leukoencephalopathy (CADASIL) [7]. CADASIL is a genetic disorder caused by mutation in the NOTCH3 gene on chromosome 19p13 gene, which leads to the production of an abnormal NOTCH3 receptor protein that impairs the function and survival of vascular endothelial and smooth muscle cells [8]. This disruption can lead to apoptosis of these cells in the brain, leading to blood vessel damage that can cause a variety of symptoms including strokes or transient ischemic attacks (TIAs), dementia, migraine with aura, and mood disturbances [9]. The radiological findings that suggest CADASIL include diffuse white matter changes affecting mainly the external capsule and temporal poles, in addition to lacunar infarcts and microbleeds in pons, thalamus, centrum semiovale, or basal ganglia [10]. The white matter changes are mainly extensive and diffuse, however initially they can appear similar to MS. Furthermore, MS-related biomarkers like cerebrospinal fluid (CSF) oligoclonal bands (OCBs) and neurofilament light chains are not specific, and they can be present in CADASIL [11, 12]. The literature reported some cases who were initially misdiagnosed as MS, and later they develop symptoms and radiological findings that were in concordance with CADASIL. Some of these cases had atypical clinical and radiological features from the start like those who were reported by Khan et al. and Carone [7, 13]. Alternatively, these cases may present with atypical features for both MS and CADASIL as those reported by Emanuele et al. [14]. However, the question yet to be answered is: can CADASIL predispose to MS, or can it present with MS-like manifestations? What features can be considered as red flags indicating genetic testing in a patient fulfilling the McDonald diagnostic criteria [15], the international standards for MS diagnosis that incorporate clinical attacks, MRI dissemination in space and time, optic nerve lesions, CSF OCBs or kappa free light chains, as well as imaging markers such as central vein sign, and paramagnetic rim lesions? If demyelination was present in CADASIL patients, can we consider disease modifying treatments (DMTs)? All these questions have been raised and are still debatable. In this report, we present two cases who were initially diagnosed with MS with radiology reports of demyelinating brain lesions and positive OCBs in the CSF, they received DMTs, however their genetic tests were positive for CADASIL. Furthermore, we reviewed the literature to elaborate all the previously reported cases with the similar features, in order to help answer questions above. Methods This study, comprising a report of two cases and a systematic review of the literature to, was conducted in accordance with the Declaration of Helsinki. It was designed to collect the evidence from all previously published cases similar to our cases. Written informed consent for publication, including images and clinical details, was obtained from both patients. Eligibility criteria of the included reports We included any report of a case or cases with a diagnosis of CADASIL, confirmed either by NOTCH3 mutation or by typical clinical or radiological features, and evidence of neuroinflammation. Eligible cases had one or more of the following: (i) clinical features such as subacute onset, relapsing-remitting course, internuclear ophthalmoplegia, optic neuritis, or sensory or motor symptoms atypical for pure vascular disease, or improvement on corticosteroids or disease-modifying treatments; (ii) radiological features such as demyelinating-appearing lesions, including Dawson’s fingers, gadolinium-enhancing lesions, or spinal cord lesions on MRI; or (iii) molecular or CSF markers of inflammation, including unmatched oligoclonal bands, elevated CSF white cell count or protein, or neuroinflammatory biomarkers. Some of these cases were initially misdiagnosed as MS and some showed similar features as MS. We excluded cases of conventional CADASIL (strokes, TIAs, migraine, cognitive decline without inflammatory features above). Cases who had other explanations of the inflammatory findings rather than CADASIL (like infections or other demyelinating diseases).were also omitted. Database search and screening We searched PubMed, Google scholar, Scopus, and web of science for any publication highlighting the relationship between CADASIL and MS through clinical cases. We used the followed keywords as a core for the search with making the appropriate amendments based on each database; CADASIL, “Autosomal Dominant Arteriopathy with Subcortical Infarcts and Leukoencephalopathy”, “NOTCH Mutation”, “Multiple Sclerosis”, MS, Neuroinflammation, “Inflammatory CADASIL”, and Neuroinflammation. The search terms were amended accordingly to suit each database (supplementary file S1). The results from that search were combined and duplication was removed used EndNote Software ( version 20 for Mac). The resulted publications were screened based on the eligibility criteria. This screening was performed by two reviewers independently to avoid selection bias. Finally, 12 reports were included into the review (Fig. 1). Fig. 1PRISMA flow diagram explaining the cascade of searching several databases, removal of duplicates, screening steps, and reviewing processes PRISMA flow diagram explaining the cascade of searching several databases, removal of duplicates, screening steps, and reviewing processes Data extraction The data of the included cases were collected by two independent reviewers, then the data were combined tabulated. The extracted data included demographic clinical (symptoms and signs, course of the disease, type of MS initially proposed, and medical treatment including DMTs), radiological (MRI findings of the brain and spinal cord), CSF analysis results, and genetic test results. Results Case presentations Case 1 A 37-year-old man who had a medical history of psoriasis and took vitamin D presented in January, 2024 with episodes of severe bifrontal pulsatile headache, vomiting, nausea, photophobia, phonophobia, vertigo, unsteady gait, weakness in both lower limbs, and pain in bilateral upper and lower limbs (with pain scale 9/10). The MRI showed white matter lesions, some of which were identical to MS with evidence of activity (Fig. 2a). Furthermore, OCBs were present in the CSF only without serum matching. The initial impression was MS, and he received three days of pulsed steroids therapy (1 g of methylprednisolone) with marked improvement after two weeks. This improvement reinforced the diagnosis of relapsing remitting multiple sclerosis (RRMS). Subsequently, he commenced Ocrelizumab as a DMT in June 2024 after specialist Neuroradiology review of his MRI scans at the MS multidisciplinary team meeting, which specifically determined that imaging appearances were likely MS and not CADASIL. Fig. 2MRI Brain and cervical spine of case1. A: shows T2 fluid attenuated inversion recovery (FLAIR) in chronal, sagittal, and axial brain images that were performed in April 2025. B: shows T2 FLAIR in chronal, sagittal, and axial brain images, in addition to sagittal T2 image of the cervical spines and cord that were performed in August 2024. These MRI images revealed extensive scattered periventricular and subcortical hyperintense lesions some of them showing possibility of demyelination, as shown in two periventricular lesions perpendicular to the corpus callosum “Dawson fingers”, with some are suggestive of CADASIL, as shown in the temporal white matter lesions. There is no sinister change when comparing 2024 with 2025 imaging. MRI cervical spines that were performed in 2024 did not reveal spinal cord lesions MRI Brain and cervical spine of case1. A: shows T2 fluid attenuated inversion recovery (FLAIR) in chronal, sagittal, and axial brain images that were performed in April 2025. B: shows T2 FLAIR in chronal, sagittal, and axial brain images, in addition to sagittal T2 image of the cervical spines and cord that were performed in August 2024. These MRI images revealed extensive scattered periventricular and subcortical hyperintense lesions some of them showing possibility of demyelination, as shown in two periventricular lesions perpendicular to the corpus callosum “Dawson fingers”, with some are suggestive of CADASIL, as shown in the temporal white matter lesions. There is no sinister change when comparing 2024 with 2025 imaging. MRI cervical spines that were performed in 2024 did not reveal spinal cord lesions In August 2024, another severe episode of bifrontal pulsatile headache with VAS 8/10 with the similar characteristics as the previous episode: associated ataxia, vertigo, and visual impairment. Examination revealed left lower homonymous quadrantanopia. The patient received 1 g of methylprednisolone daily for three days and made a full recovery after one week. The follow up MRI found no change from the first one (Fig. 2b). Furthermore, repeated CSF examination again showed presence of unmatched OCBs. Further investigations revealed normal autoimmune vasculitis screening and metabolic profile. The molecular analysis revealed heterozygous, pathogenic NOTCH3 mutation (NM_000435.2) NM_000435.2:c.544 C > T p.(Arg182Cys). Therefore, CADASIL diagnosis was confirmed. Further clinical assessment did not reveal cognitive impairment or focal neurological symptoms. The patient then discovered that his father had previously been diagnosed with CADASIL; he is a 62-year-old, otherwise healthy man without significant cognitive or other neurological problems. After counselling with the patient and discussing his case in the MS and neuroinflammatory multiple disciplinary team meeting, his Ocrelizumab was stopped. The patient has been relatively stable with episodes of migraine with aura which were not fully controlled on propranolol; therefore, amitriptyline was added. Case 2 A 47-year-old woman who had episodes of diplopia and migraine since 2008 presented in 2012 for the assessment of her migraine and visual disturbance. Her MRI brain showed periventricular white matter lesions suggestive of MS (Fig. 3). She was diagnosed with RRMS and commenced interferon beta 1 A in 2013 until 2015. During this period, she continued to have recurrent migraine episodes with photophobia and visual disturbance. Therefore, her DMT was switched to Fingolimod which was commenced in 2015 and continued until 2025. Over the lest 10 years, she has had multiple episodes of migraine with visual aura associated with nausea, photophobia and phonophobia in addition to neuropathic pain mainly at night (unresponsive to gabapentin), numbness, weakness, tingling, double vision and dysphasia. On assessment, there were no cognitive nor focal neurologic signs. Her main complaint was migraine, ; molecular analysis was requested which revealed heterozygous, likely pathogenic NOTCH3 mutation (NM_000435.2) NM_000435.2:c.2791 A > T p.(Ser931Cys). Therefore, a diagnosis of CADASIL was confirmed. No family history of CADASIL was found. Fig. 3MRI Brain and cervical spine of case 2 that were performed in two different periods. A: Fluid attenuated inversion recovery (FLAIR) sagittal and axial brain images acquired in March 2025 demonstrate scattered periventricular and subcortical hyperintense lesions with appearances suggestive of demyelination. B: FLAIR sagittal and axial brain images from April 2024 show a similar distribution of lesions without significant interval change. Cervical spine MRI performed in 2025 (T2 weighted sagittal sequence, not shown due to unavailability in that sequence) did not reveal spinal cord lesions MRI Brain and cervical spine of case 2 that were performed in two different periods. A: Fluid attenuated inversion recovery (FLAIR) sagittal and axial brain images acquired in March 2025 demonstrate scattered periventricular and subcortical hyperintense lesions with appearances suggestive of demyelination. B: FLAIR sagittal and axial brain images from April 2024 show a similar distribution of lesions without significant interval change. Cervical spine MRI performed in 2025 (T2 weighted sagittal sequence, not shown due to unavailability in that sequence) did not reveal spinal cord lesions Review of the included cases We reviewed 19 cases reported in the included 12 reports. The clinical and radiological features of these 21 cases in addition to our two cases are summarized in Table 1. Table 1Characteristics of our two cases and the previously reported casesPatientSexAgeMutationCSF OCBsMRI brainMRI of spineCognitive impairmentMS subtypeKey red flags for MSTreatment responseDMTsCase 1M37c.544 C > T p.(Arg182Cys)Unmatched OCBsPV and SC lesions with some demyelination (Temporal lobe involved)NormalNormalRRMSMigraine, family history, no effect on 1 DMTIV MP with good responseOcrelizumabCase 2F47c.2791 A > T p.(Ser931Cys)N/RPV and SC lesions with some demyelination (Temporal lobe involved)NormalNormalRRMSMigraine, no effect on 2 DMTsIV MP with good responseInterferon β -1a FingolimodCollongues et al. [16]F53c.346 C > T(p.Cys90Arg)NegativePV and SC lesions with some demyelination, +Gad. (Temporal lobe involved)1 lesion in T spineDysexecutive syndromeRRMSDysexecutive syndrome, thoracic lesion, -ve OCBsIV MP with good responseN/RCollongues et al. [16]F46c.1750 C > T(p.Cys558Arg)NegativePV, SC and IT lesions with demyelination. (Temporal lobe not involved)NormalMemory impairment, dysexecutive syndrome, depressionRRMSEarly mental health problems, negative OCBsIV MP with good responseGlatiramer acetateSchiess et al. [12]F25c.3691 C > T (p.Arg1231Cys)Unmatched OCBsPV, SC and IT lesions with some demyelination. +Gad. (Temporal lobe involved)2 lesions in C spineSevere cognitive impairmentPPMSEarly cognitive impairment, No effect on 1 DMT.IV MP with good responseGlatiramer acetateMotolesea et al. [17]M48(c.994 C > T exon 6)NegativeExtensive PV and SC lesions with demyelination, (Temporal lobe involved)1 lesion in T spineNormalSPMSThoracic lesion, modest response to steroids, -ve OCBsIV MP with modest response.N/RAljaberi et al. [18]M43+ve NOTCH3 mutationNegativePV and SC lesions with some demyelination, (Temporal lobe involved)N/RDifficulty in calculation and processing speedMSExecutive dysfunction, -ve OCBsN/RN/RKhan et al. [7] F61c.544 > T(pArg182Cys)NegativePV and SC lesions with demyelination, lacunar infarction and atrophy(Temporal lobe involved)1 lesion in C spineMemory impairmentSPMSLacunar infarction, memory affection, -ve OCBs, age of the onsetOral prednisoloneGlatiramer aetate,Interferon β -1a Natalizumab, TeriflunomideKhan et al. [7] F55+ve NOTCH3 mutationNegativePV and SC lesions with some demyelination, (Temporal lobe involved)N/RMild memory impairmentMSTemporal lobe lesions, memory affection, -ve OCBsTopiramateN/RKhan et al. [7] F 45c.3691 C > T (p.Arg1231Cys)Unmatched OCBsPV and SC lesions with no prominent demyelination, (Temporal lobe not involved)N/RMemory impairmentRRMSSVD, memory affection, -ve OCBs, modest response to steroidsIV MP with modest response.Dimethyl fumarateMadjidov et al. [19] F 50c.6102dup, (p.Gly2035Argfs60)Unmatched OCBsExtensive PV and SC lesions with no prominent demyelination, (Temporal lobe not involved)N/RNormalNoSVD, good response to antiplateletsDual antiplatelet CCBN/RMadjidov et al. [19] F 51c.6102dup, (p.Gly2035Argfs60)Unmatched OCBsExtensive PV lesions with some demyelination, (Temporal lobe not involved)N/RProgressive cognitive declineMSEarly cognitive declineN/RN/RMorena et al. [14] F 38c.421 C > T, p. (Arg141Cys)Unmatched OCBsPV and SC lesions with no prominent demyelination, (Temporal lobe not involved)1 lesion in C spineN/RRRMSNo demyelination. No effect on 4 DMTsIV MP ClopidogrelGlatiramer acetate Natalizumab Dimethyl fumarateβ-interferon-1aMorena et al. [14] M 49c.3226 C > T (p.Arg1076Cys)Unmatched OCBsPV and BG lesions with some demyelination, (Temporal lobe not involved)> 2 lesions in C and T spinesNormalRRMST spine lesion, no response on 3 DMTsAspirinTeriflunomide Fingolimod OcrelizumabJaffer et al. [20] F 48+ve NOTCH3 mutationNegativePV, SC and IT lesions with some demyelination. (Temporal lobe not involved)NormalMemory impairmentRRMSMemory impairment, -ve OCBsAspirinGlatiramer acetateInterferon β -1aBouley et al. [21] M 31+ve NOTCH3 mutationNegativeExtensive PV and WM lesions mainly vascular. + Gad (Temporal lobe not involved)N/RNormalNOVascular lesions, -ve OCBsIV MPAspirin AtorvastatinN/RJoshi et al. [22] F 23+ve NOTCH3 mutationN/RPV and SC lesions with some demyelination, (Temporal lobe involved)N/RNormalNOTemporal lobe involvementN/RN/RJoshi et al. [22] M 56+ve NOTCH3 mutationNegativePV and SC lesions with some demyelination, (Temporal lobe involved)NormalMemory impairmentMSEarly memory impairment, -ve OCBs, temporal lobe involvement.N/RN/RJoshi et al. [22] F N/R+ve NOTCH3 mutationN/RPV and SC lesions with some demyelination, (Temporal lobe not involved)N/RN/RMSN/RN/RN/RCarone. [13] F 61+ve NOTCH3 mutationN/RPV and SC lesions with some demyelination, (Temporal lobe not involved)Normalcognitive impairmentRRMSEarly cognitive impairment, No response on 2 DMTs, age of the onsetN/RInterferon β -1a FingolimodChitnis [23]M26c.1750 C > T30 WBCs, No OCBBalo’s concentric sclerosis (incomplete ring enhancement)Not doneexpressive aphasia, word-finding difficultyNoAphasia, -ve OCBsIV MPPlasmapheresis, β-interferon-1a with improvementAbbreviation M male F female, OCB oligoclonal band, N/R not reported, PV periventricular, SC subcortical, GAD gadolinium, IT infratentorial, BG basal ganglia, RRMS relapsing remitting multiple sclerosis, PPMS primary progressive multiple sclerosis, SPMS secondary progressive multiple sclerosis, MS multiple sclerosis, MP methylprednisolone, SVD small vessel disease, -ve negative ‘+ve NOTCH3 mutation’ refers to a reported pathogenic or likely pathogenic NOTCH3 variant without specification of the exact cDNA/protein change in the original paper Table 1Characteristics of our two cases and the previously reported casesPatientSexAgeMutationCSF OCBsMRI brainMRI of spineCognitive impairmentMS subtypeKey red flags for MSTreatment responseDMTsCase 1M37c.544 C > T p.(Arg182Cys)Unmatched OCBsPV and SC lesions with some demyelination (Temporal lobe involved)NormalNormalRRMSMigraine, family history, no effect on 1 DMTIV MP with good responseOcrelizumabCase 2F47c.2791 A > T p.(Ser931Cys)N/RPV and SC lesions with some demyelination (Temporal lobe involved)NormalNormalRRMSMigraine, no effect on 2 DMTsIV MP with good responseInterferon β -1a FingolimodCollongues et al. [16]F53c.346 C > T(p.Cys90Arg)NegativePV and SC lesions with some demyelination, +Gad. (Temporal lobe involved)1 lesion in T spineDysexecutive syndromeRRMSDysexecutive syndrome, thoracic lesion, -ve OCBsIV MP with good responseN/RCollongues et al. [16]F46c.1750 C > T(p.Cys558Arg)NegativePV, SC and IT lesions with demyelination. (Temporal lobe not involved)NormalMemory impairment, dysexecutive syndrome, depressionRRMSEarly mental health problems, negative OCBsIV MP with good responseGlatiramer acetateSchiess et al. [12]F25c.3691 C > T (p.Arg1231Cys)Unmatched OCBsPV, SC and IT lesions with some demyelination. +Gad. (Temporal lobe involved)2 lesions in C spineSevere cognitive impairmentPPMSEarly cognitive impairment, No effect on 1 DMT.IV MP with good responseGlatiramer acetateMotolesea et al. [17]M48(c.994 C > T exon 6)NegativeExtensive PV and SC lesions with demyelination, (Temporal lobe involved)1 lesion in T spineNormalSPMSThoracic lesion, modest response to steroids, -ve OCBsIV MP with modest response.N/RAljaberi et al. [18]M43+ve NOTCH3 mutationNegativePV and SC lesions with some demyelination, (Temporal lobe involved)N/RDifficulty in calculation and processing speedMSExecutive dysfunction, -ve OCBsN/RN/RKhan et al. [7] F61c.544 > T(pArg182Cys)NegativePV and SC lesions with demyelination, lacunar infarction and atrophy(Temporal lobe involved)1 lesion in C spineMemory impairmentSPMSLacunar infarction, memory affection, -ve OCBs, age of the onsetOral prednisoloneGlatiramer aetate,Interferon β -1a Natalizumab, TeriflunomideKhan et al. [7] F55+ve NOTCH3 mutationNegativePV and SC lesions with some demyelination, (Temporal lobe involved)N/RMild memory impairmentMSTemporal lobe lesions, memory affection, -ve OCBsTopiramateN/RKhan et al. [7] F 45c.3691 C > T (p.Arg1231Cys)Unmatched OCBsPV and SC lesions with no prominent demyelination, (Temporal lobe not involved)N/RMemory impairmentRRMSSVD, memory affection, -ve OCBs, modest response to steroidsIV MP with modest response.Dimethyl fumarateMadjidov et al. [19] F 50c.6102dup, (p.Gly2035Argfs60)Unmatched OCBsExtensive PV and SC lesions with no prominent demyelination, (Temporal lobe not involved)N/RNormalNoSVD, good response to antiplateletsDual antiplatelet CCBN/RMadjidov et al. [19] F 51c.6102dup, (p.Gly2035Argfs60)Unmatched OCBsExtensive PV lesions with some demyelination, (Temporal lobe not involved)N/RProgressive cognitive declineMSEarly cognitive declineN/RN/RMorena et al. [14] F 38c.421 C > T, p. (Arg141Cys)Unmatched OCBsPV and SC lesions with no prominent demyelination, (Temporal lobe not involved)1 lesion in C spineN/RRRMSNo demyelination. No effect on 4 DMTsIV MP ClopidogrelGlatiramer acetate Natalizumab Dimethyl fumarateβ-interferon-1aMorena et al. [14] M 49c.3226 C > T (p.Arg1076Cys)Unmatched OCBsPV and BG lesions with some demyelination, (Temporal lobe not involved)> 2 lesions in C and T spinesNormalRRMST spine lesion, no response on 3 DMTsAspirinTeriflunomide Fingolimod OcrelizumabJaffer et al. [20] F 48+ve NOTCH3 mutationNegativePV, SC and IT lesions with some demyelination. (Temporal lobe not involved)NormalMemory impairmentRRMSMemory impairment, -ve OCBsAspirinGlatiramer acetateInterferon β -1aBouley et al. [21] M 31+ve NOTCH3 mutationNegativeExtensive PV and WM lesions mainly vascular. + Gad (Temporal lobe not involved)N/RNormalNOVascular lesions, -ve OCBsIV MPAspirin AtorvastatinN/RJoshi et al. [22] F 23+ve NOTCH3 mutationN/RPV and SC lesions with some demyelination, (Temporal lobe involved)N/RNormalNOTemporal lobe involvementN/RN/RJoshi et al. [22] M 56+ve NOTCH3 mutationNegativePV and SC lesions with some demyelination, (Temporal lobe involved)NormalMemory impairmentMSEarly memory impairment, -ve OCBs, temporal lobe involvement.N/RN/RJoshi et al. [22] F N/R+ve NOTCH3 mutationN/RPV and SC lesions with some demyelination, (Temporal lobe not involved)N/RN/RMSN/RN/RN/RCarone. [13] F 61+ve NOTCH3 mutationN/RPV and SC lesions with some demyelination, (Temporal lobe not involved)Normalcognitive impairmentRRMSEarly cognitive impairment, No response on 2 DMTs, age of the onsetN/RInterferon β -1a FingolimodChitnis [23]M26c.1750 C > T30 WBCs, No OCBBalo’s concentric sclerosis (incomplete ring enhancement)Not doneexpressive aphasia, word-finding difficultyNoAphasia, -ve OCBsIV MPPlasmapheresis, β-interferon-1a with improvementAbbreviation M male F female, OCB oligoclonal band, N/R not reported, PV periventricular, SC subcortical, GAD gadolinium, IT infratentorial, BG basal ganglia, RRMS relapsing remitting multiple sclerosis, PPMS primary progressive multiple sclerosis, SPMS secondary progressive multiple sclerosis, MS multiple sclerosis, MP methylprednisolone, SVD small vessel disease, -ve negative ‘+ve NOTCH3 mutation’ refers to a reported pathogenic or likely pathogenic NOTCH3 variant without specification of the exact cDNA/protein change in the original paper Characteristics of our two cases and the previously reported cases Glatiramer aetate, Interferon β -1a Natalizumab, Teriflunomide c.6102dup, ( p.Gly2035Argfs60) c.6102dup, ( p.Gly2035Argfs60) Glatiramer acetate Natalizumab Dimethyl fumarate β-interferon-1a Glatiramer acetate Interferon β -1a IV MP Aspirin Atorvastatin Abbreviation M male F female, OCB oligoclonal band, N/R not reported, PV periventricular, SC subcortical, GAD gadolinium, IT infratentorial, BG basal ganglia, RRMS relapsing remitting multiple sclerosis, PPMS primary progressive multiple sclerosis, SPMS secondary progressive multiple sclerosis, MS multiple sclerosis, MP methylprednisolone, SVD small vessel disease, -ve negative ‘+ve NOTCH3 mutation’ refers to a reported pathogenic or likely pathogenic NOTCH3 variant without specification of the exact cDNA/protein change in the original paper Having the symptom onset, the mean age at which the symptoms started was 38.4 ± 10.1. The most prominent clinical features in these cases were sensory manifestations, mostly paresthesia, (76.2%) followed by motor manifestations, weakness and spasticity, (71.4%), gait imbalance (66.7%) and headache, mostly migraine with or without aura, (61.9%). Unlike conventional CADASIL, the percentage of strokes or TIAs was only 9.5%, while 47.6% had cognitive impairment in the form of mild to moderate executive and memory function impairment (Fig. 4). Fig. 4The clinical, MRI and cerebrospinal fluid finding in the including 21 cases. MRI; magnetic resonance images, CSF; cerebrospinal fluid, SVD; small vessel disease; LVO; large vessel occlusion, BG; basal ganglia, IT; infratentorial, OCBs; oligoclonal bands The clinical, MRI and cerebrospinal fluid finding in the including 21 cases. MRI; magnetic resonance images, CSF; cerebrospinal fluid, SVD; small vessel disease; LVO; large vessel occlusion, BG; basal ganglia, IT; infratentorial, OCBs; oligoclonal bands Considering radiological features, most of the patients (90.5%) had either evidence of white matter small vessel disease, demyelinating lesions (either nonspecific or MS-like), or both. The temporal lobe was involved in 65% of patients. The cerebral cortex, basal ganglia, brain stem or cerebellum were affected in 85.7% of the patients. About the third of the included patients had gadolinium enhancing lesions and 40% had spinal cord lesions. Unmatched oligoclonal bands were present in the CSF in 29.4% of patients and other CSF inflammatory markers were present in 35.3% of the patients (Fig. 3). Discussion The variability of MS presenting symptoms, the presence of different phenotypes, relatively non-specific conventional biomarkers, and prioritization of early diagnosis and DMT can result in misdiagnosis of some cases who have conditions other than MS [24]. CADASIL is one of these conditions. While typical cases can be easily differentiated, atypical presentations might resemble the dissemination in time and in space of demyelinating MRI lesions which is the cardinal principle in the MS diagnosis. In this study, we presented two cases who were initially misdiagnosed with MS and received DMT, then due to clinical suspicion subsequent molecular analysis confirmed pathogenic NOTCH3 mutations. Furthermore, we have collated previously reported CADASIL cases that were initially misdiagnosed as MS or showed overlapping features. CADASIL-associated vascular lesions, especially those associated with large vessel occlusion, are associated with loss of the functions of the affected area, but the MS associated lesion usually show selectivity with some fibers affected and other not in the same area [25]. NOTCH3 receptor dysregulation can cause brain lesions resembling MS lesion through activation of NF-κB [26], a pro-inflammatory transcription factor which upregulates inflammatory genes in smooth muscle cells and macrophages in the vessel wall causing MS-like demyelination [27]. Furthermore, Mutant NOTCH3 extracellular domains form aggregates with extracellular matrix proteins [28]. These complexes may present novel epitopes, stimulating autoimmune responses [29]. Therefore, it would be better for CADASIL which shows inflammatory or demyelinating features to be called “inflammatory CADASIL” to distinguish it from conventional CADASIL. The literature contains 19 published CADASIL cases who were initially misdiagnosed as MS or showed overlapped inflammatory or demyelinating features [7, 12, 14, 16-19, 21-23]. By reviewing these cases and including our two cases we can compile criteria or “red flags”, that if present in cases of MS or related disorders, would indicate further molecular analysis to either confirm or rule-out inflammatory CADASIL. Across our two cases and the previously reported series, most NOTCH3 variants involve cysteine-altering mutations within the epidermal growth factor-like repeats (EGFr) of the extracellular domain, consistent with classical CADASIL pathogenesis [7, 12, 14, 17–20, 30]. These mutations promote abnormal NOTCH3 multimerization and aggregation in the vessel wall, leading to degeneration of vascular smooth muscle cells, thickening of small arteries, and disruption of the blood-brain barrier [8, 9, 28]. Experimental work suggests that mutant NOTCH3 can enhance activation of NF-κB-dependent inflammatory pathways in vascular smooth muscle cells and macrophages, upregulating proinflammatory mediators and adhesion molecules [26, 27, 29]. In addition, a recently described inflammatory CADASIL phenotype associated with a cysteine-sparing NOTCH3 frameshift variant showed complement deposition and markers of vascular inflammation, supporting a direct link between NOTCH3 dysfunction, vascular injury, and secondary immune activation [19]. Together, these observations provide a plausible biological framework in which certain NOTCH3 mutations, or cumulative vascular damage from NOTCH3-mediated small vessel disease, can trigger or amplify local CNS inflammation, intrathecal IgG synthesis, and demyelinating-appearing lesions, and hence generating an ‘inflammatory CADASIL’ phenotype that overlaps with MS. By considering the reported inflammatory features and efficacy of anti-inflammatory and disease modifying treatments [12, 16, 23], we might consider that inflammation and immune system dysfunction plays an important role in the pathogenesis of this subtype. The immune cells infiltrating from diseased vessels and accompanying cytokine release are, as described before, the most accepted explanation. Therefore, continued immune therapy, either acutely as steroids or plasmapheresis or disease modifying treatments could be reasonable in some cases. However, further large-sample size studies including clinical trials are needed to confirm that assumption. The data in this report is not enough to consider inflammatory CADASIL; however, we proposed this term as a direction for future consideration. Another thing that some of included cases were reported as CADASIL cases who were initially misdiagnosed with MS, not as they showed inflammatory features. In fact, the cause of this misdiagnosis is the presence of inflammatory features. The possibility of dual pathology some of the cases may genuinely have MS as well and incidentally having CADASIL. Therefore, considering DMT, is a clue in the management of such patients. In conclusion, CADASIL may occasionally mimic MS, not only clinically but also with presence of unmatched oligoclonal bands in the CSF and MRI features that defy even specialist Neuroradiology interpretation. This MS-like CADSAIL should be considered and genetic testing should be performed when red flags including stroke-like presentation, migraine predominance, disproportionate cognitive impairment, vascular lesions, temporal lobe involvement, or negative oligoclonal bands. The concept of an “inflammatory CADASIL” phenotype remains hypothesis generating and warrants further research. Supplementary Information Supplementary Material 1. Supplementary Material 1.