Abstract Background Baló’s Concentric Sclerosis (BCS) is a rare heterogeneous demyelinating disease with a variety of phenotypes on Magnetic Resonance Imaging (MRI). Existing literature lacks data especially on the therapeutic approach of the disease which we intended to elucidate by means of suggesting a new possible BCS classification and introducing different therapeutic concepts based on each BCS-subgroup characteristics. Methods We present a retrospective study of eight treated patients with BCS-type lesions, emphasizing on MRI characteristics and differences on therapeutic maneuvers. Results Data analysis showed: at disease onset the BCS-type lesion was tumefactive (size ≥2 cm) in 6 patients, with a mean size of 2.7 cm (± 0.80 SD); a coexistence of MS-like plaques on brain MRI was identified in 7 patients of our cohort. The mean age was 26.3 years (±7.3 SD) at disease onset and the mean follow-up period was 56.8 months (range 9–132 months). According to radiological characteristics and response to therapies, we further categorized them into 3 subgroups: a) Group-1; BCS with or without coexisting nonspecific white matter lesions; poor response to intravenous methylprednisolone (IVMP); treated with high doses of immunosuppressive agents (4 patients), b) Group-2; BCS with typical MS lesions; good response to IVMP; treated with MS-disease modifying therapies (2 patients), c) Group-3; BCS with typical MS lesions; poor response to IVMP; treated with rituximab (2 patients). Conclusions Our study introduces a new insight regarding the categorization of BCS into three subgroups depending on radiological features at onset and during the course of the disease, in combination with the response to different immunotherapies. Immunosuppressive agents such as cyclophosphamide are usually effective in BCS. However, therapeutic alternatives like anti-CD20 monoclonal antibodies or more classical disease-modifying MS therapies can be considered when BCS has also mixed lesions similar to MS. Future studies with a larger sample size are necessary to further establish these findings, thus leading to better treatment algorithms and improved clinical outcomes.