A study comparing the efficacy of R-CHOP and R-DA-EPOCH regimens in young, high-risk treatment-naïve diffuse large B-cell lymphoma patients: a multicenter retrospective analysis.

BACKGROUND

Patients with high-risk diffuse large B-cell lymphoma (DLBCL) have poor prognoses, and the comparative efficacy between R-CHOP and R-DA-EPOCH regimens remains controversial.

OBJECTIVE

To compare the efficacy and safety of R-CHOP and R-DA-EPOCH regimens in young, treatment-naïve patients with high-risk DLBCL.

METHODS

This multicenter retrospective cohort study included 148 young, treatment-naïve, high-risk DLBCL patients treated at five tertiary hospitals between March 2023 and March 2025. Patients were divided into the R-CHOP group (n=98) and the R-DA-EPOCH group (n=50). Baseline characteristics, treatment responses, treatment completion, dose intensity, adverse events, and supportive care information were collected. Intergroup comparisons were performed using the Mann-Whitney U test, chi-square test, or Fisher's exact test.

RESULTS

Baseline characteristics were generally balanced between the two groups, except for a significantly higher proportion of ECOG score 1 in the R-DA-EPOCH group (76.0% vs. 50.0%, P=0.007); subsequent multivariate analysis was used to adjust for this difference. The complete response (CR) rate was 46.0% in the R-DA-EPOCH group versus 36.7% in the R-CHOP group, but the difference was not statistically significant (P=0.313); the objective response rates were 58.0% and 61.2%, respectively (P=0.313). Subgroup analysis showed higher CR rates with R-DA-EPOCH in most subgroups, though none reached statistical significance. The relative dose intensity was significantly lower in the R-DA-EPOCH group compared to the R-CHOP group ($0.84\pm 0.06 vs. $0.92\pm 0.04, P<0.001), with significantly higher rates of dose reduction (52.0% vs. 18.4%, P<0.001) and treatment delay (46.0% vs. 18.4%, P=0.001). The R-DA-EPOCH group had significantly higher usage rates of G-CSF (50.0% vs. 31.6%, P=0.033), platelet transfusions (28.0% vs. 8.2%, P=0.003), and a longer median length of hospital stay (7.1 days vs. 4.9 days, P<0.001). Baseline LDH levels showed no significant correlation with treatment response (P=0.561).

CONCLUSION

Compared with R-CHOP, the R-DA-EPOCH regimen showed a trend toward higher CR rates in young, treatment-naïve, high-risk DLBCL patients, but this did not reach statistical significance. Moreover, R-DA-EPOCH was associated with significantly increased hematological toxicity and supportive care burden. Clinicians should carefully weigh the potential benefits against the risks when selecting this regimen.