Ruxolitinib, is a JAK1/2 inhibitor approved for the treatment of intermediate and high risk myelofibrosis (MF). The recommended starting doses for ruxolitinib, are 20 mg PO BID or 15 mg PO BID if platelet (PLT) counts are ≥200 x109/L or 100 to 200 x109/L, respectively. However, drug related side effects which includes grade 3/4 anemia (45.2%), thrombocytopenia (12.9%), fatigue (25.2%) and diarrhea (23.2%) are common using standard dosing. Moreover 70% of patients (pts) started on the standard regimen require dose reduction because of drug related toxicities. Although hematologic side effects are generally expected during early phases of ruxolitinib therapy and reversible in most cases, a therapeutic approach that can avoid myelosuppression while still providing adequate clinical responses is warranted. Therefore, to obviate toxicities and achieve symptom improvement in pts with MF, we employed a dose escalation (DE, starting at 5 mg QOD) approach for ruxolitinib and compared outcomes with another group that previously received standard regimens (SR, 15 or 20 mg BID). In addition to baseline characteristics, clinical data including hematologic parameters, bone marrow (BM) results, Dynamic International Prognostic Scoring System-Plus (DIPSS-plus) risk, and transfusion (TF) requirement were collected. Hematologic and non-hematologic side effects based on CTCAEv.4 and evidence of clinical response (reduction of splenomegaly by palpation) were assessed at 8, 12 and 16 weeks after starting treatment (post-Tx). Descriptive data were summarized as frequency of counts and percentages and continuous data were presented as mean and ranges and the comparison between two groups was performed. Survival outcomes were analyzed using the Kaplan-Meier method. Plasma concentrations for interleukins (IL) including IL1A, IL1B, IL2, IL4, IL6, IL8, IL10, IL12, IL17A, INFg, TNFa and GM-CSF were also measured pre and post- ruxolitinib treatment using a Multi-Analyte ELISArray Kits (SABiosciences)A total of 42 pts seen in our MPN clinic were studied, including 23 females (54%) and 19 males (46%). The median age of the cohort was 69 (43-83 years). Disease subtypes include 23 PMF, 10 post-PV MF, 6 post-ET MF and, 1 CMML-1. Based on the DIPSS-plus risk score, there were 17 high, 16 intermediate-2, and 8 intermediate-1 risk pts. Thirty-two pts (76%) were JAK2V617F mutant and 10 were wild type. The mean duration of treatment was 8 mos (range, 3-17) and follow up was 11 mos (4-36). Twenty-six pts started on DE and 16 received SR. The mean ruxolitinib dose in the whole cohort was 10 mg BID (range; 5mg QOD-25mg BID). The mean maintenance dose in the DE cohort was 5mg BID while in the SR cohort was 15mg BID. In terms of clinical response, the mean pre and post-Tx spleen size was 13.69 cm and 6.8 cm in the DE cohort compared to 17.68 cm and 13.07 cm in SR cohort (p = .03). In terms of hematological side effects, the DE approach resulted in less frequent grade 3-4 or any grade anemia compared to SR (G3-4: 3% vs. 18%, p =.02; overall anemia: 14% vs. 50%, p =.04). In pts treated with DE and SR, 57% and 75% were TF dependent pre-Tx. Post-Tx, 35% of pts treated with DE became transfusion independent vs 0% in the SR group. The DE approach also resulted in less frequent grade 3-4 or any grade thrombocytopenia compared to SR (G3-4: 11% vs. 31%, p= .04; overall thrombocytopenia: 23% vs. 43%, p=.06). There was no difference in the frequency or degree of neutropenia between the DE and SR treated pts. Non-hematologic side effects were also less frequent in the DE group compared to the SR group (bruising: 36 vs 75%, diarrhea: 28 vs 56%, lower extremity edema: 21% vs 50%, and dizziness: 15 vs 31%). In 3 pts treated with DE, the degree of reticulin fibrosis reduced from MF3 to MF2 confirmed with bone marrow biopsy performed 6-12 months post-Tx. We also compared the differences in cytokine profile between the patients treated with DE (n=2) and SR (N=2). No difference in the degree of reduction of MF relevant cytokines including IL-1A, IL-1B, IL-2, IL-4, IL-6, IL-12, TNF-α and GM-CSF were observed between both groups (p=0.2). At the time of analysis, 38 pts are still alive and no difference in survival were observed between DE and SR treated patients (log rank p=0.69). In conclusion, a dose escalation approach for ruxolitinib therapy is better tolerated and with preserved clinical responses compared to the standard dosing regimen in MF patients. Disclosures:No relevant conflicts of interest to declare.