Abstract Background Selective depletion of myeloid lineage leucocytes by adsorptive granulocyte and monocyte apheresis (GMA) with Adacolumn ® was introduced as a nonpharmacologic treatment for ulcerative colitis (UC) in 2000. It has been reported that GMA may be effective in combination with immunosuppressive treatment in a subset of patients. Τhe purpose of our study is the evaluation of the effectiveness and safety of GMA as a complementary treatment in patients with refractory ulcerative colitis. Methods Prospective data collection of a patient cohort with refractory UC receiving Adacolumn ® as an adjunct to their medical treatment. The therapeutic protocol has 2 phases: The induction phase entails two sessions per week for at least 3 weeks. The maintenance includes one weekly session for one month, one session every 15 days for one the next month, and monthly sessions thereafter. The patients’ medical treatment was maintained during the sessions. As a failure to GMA was considered the need for colectomy, the switch to a different treatment and inability to discontinue steroids. Response was evaluated after completion of at least 6 sessions of GMA. Results Ten patients with refractory UC were offered Adacolumn® between February 2021 and September 2023. Mean age was 39.6 years (22-61years). All patients had failed at least one biologic treatment and two-thirds two biologics. Their treatment which was maintained during GMA was: tofacitinib 10mg bid for 4 patients, ustekinumab 90mg sc every 8 weeks for 3, vedolizumab 300 mg every 8 weeks for 2, and corticosteroids 16mg for one. The median duration of treatment was 5.1 months (2-13 months), while the median number of sessions was 16 (6-45). Clinical and endoscopic remission was achieved in two cases (20%), after 13 sessions for both (in brackets in the Table) whereas two patients (20%) responded clinically according to partial Mayo score. Treatment failure was documented for 6 patients (60%) after 6-45 sessions. Patient number 4 performed 45 sessions in different hospitals because he was reluctant to be operated. Three underwent colectomy and three discontinued due to non-response. No adverse effects were observed. The initial median of partial Mayo score was 6 (5-9) while at the end of the evaluation was 4.5 (0-9). Table 1 summarizes the results of our study. Conclusion 1) GMA may be beneficial as an adjunct to biologics in refractory to medical treatment UC patients. 2) Interestingly, all patients on tofacitinib showed a favorable response after the addition of GMA. This observation may help define a subset of UC patients who may benefit the most.