Background:Diffuse large B‐cell lymphoma (DLBCL) is the most spread subtype of NHL. Outcomes of standard treatment in high‐risk patients are poor, especially in LMIC with limited access to expensive drugs and 2nd line treatment.Aims:We have started multicenter prospective cohort study with the aim to compare efficacy, toxicity and cost‐effectiveness of treatment of high‐risk DLBCL patients.Methods:Newly diagnosed patients with DLBCL non‐specified, with 3 to 5 points according to IPI, age 18 years and older, without known CNS involvement were included into the study. Chemotherapy regimen (CHOP‐like, R‐CHOP or R‐DA‐EPOCH) was chosen by the investigators opinion. The patients received radiation therapy 30–36 Gy by standard indications and some CNS prophylaxis by investigator choice. Primary end‐point was 2‐year event‐free survival (EFS). Secondary end‐points were overall response (OR) rate, 2‐year overall survival (OS), toxicity rates, quality of life (QOL), index of cost‐effectiveness (ICE) for each couple of regimens (for the pharmaco‐economy analysis). The analysis of direct costs of treatment and Quality‐adjusted life years (QALY) was used for calculation of ICE. As an acceptable level of ICE was chosen amount of 20000 US dollars.Results:Four Ukrainian centers recruited 140 patients during the period since January 2014 till Decmber 2016. Final analysis was performed in July 2016. Median age was 53 years. CHOP‐like regimens received 53 patients (I group), R‐CHOP – 60 patients (II group), R‐DA‐EPOCH – 27 patients (III group). Average number of treatment cycles was (5.8 ± 1.7). Radiation therapy received 20 patients (27.0 %), CNS prophylaxis – 15 patients (20.3 %). Median FU period was 24.7 months. Groups were comparable by demographic and clinical data, except level of CNS prophylaxis, that was higher in the III group (P = 0,04). 2‐year EFS was higher in the III group 86,3 ± 7,5 % vs 64,3 ± 10,4 % in the II group and 42,7 ± 11,4 in the I group (P = 0,045). OR rate didn’t differ statistically in three groups, but was higher in the III group. 2‐year OS was higher in the III group as well, but not statistically significant. Number of patients, who needed further treatment was 7.4% in the III group, 35.0% ‐ in the II group, 49.1% in the I group (p = 0.001). Anemia was less frequent in the group II (P = 0.04) as well as neutropenia (P = 0.04). Among non‐hematological complication significant difference was revealed in the rate of neurotoxicity that was higher in the group III (P = 0,023) and hyperglycemia, that was higher in the group I (P = 0,05). Treatment toxicity did not influence on QOL. The cost of 6 treatment cycles of CHOEP regimen (as the most frequent among CHOP‐like regimens) was 2363 US dollars, with R‐CHOP ‐ 350412 US dollars and with R‐DA‐EPOCH – 467178 US dollars. Average QALY was (0,8 ± 1,6) in the I group, (0,86 ± 1,55) – in the II group and (1,63 ± 1,34) – in the III group (P = 0,14). ICER‐CHOP
vs
CHOP‐like was 74850 US dollars, ICER‐DA‐EPOCH
vs
CHOP‐like was 18801 US dollars, and ICER‐DA‐EPOCH
vs
R‐CHOP was 5833 US.Summary/Conclusion:R‐DA‐EPOCH was the most effective regimen for treatmet of high‐risk patients with DLBCL with acceptable toxicity level and similar QoL compare to R‐CHOP and CHOP‐like regimens. The pharmaco‐economy analysis showed acceptable level of ICE of R‐DA‐EPOCH compare to R‐CHOP regimen. Thus, we consider R‐DA‐EPOCH regimen as a satisfying option for treatment of high‐risk DLBCL even for LMIC.