Primary mediastinal large B-cell lymphoma (PMBCL) is a distinct clinico-pathological subtype of diffuse large B-cell lymphoma (DLBCL) with unclear prognostic factors, and limited clinical data especially in the Asian population. Optimal treatment and role for radiotherapy is not fully undefined for this entity.

Patients and methods:

We retrospectively studied progression free and overall survival (PFS, OS) of 129 consecutive patients with newly diagnosed PMBCL who received treatment between 2001 to 2016 at 3 cancer centers , 2 in Singapore (n= 113) and 1 in United Kingdom (n= 16). Chemotherapy regimens were R-CHOP (n=73), dose adjusted (DA) R-EPOCH (n= 48). 30% patients also received RT.


Median age of the patients was 28 years (range 11-72), 75 were female and the majority had stage I-II disease (73%) and bulky disease (58%). Median follow up was 41 months (range2.4-188 months). Projected 10-year OS and PFS survival for the entire cohort was 91% and 74% respectively. The type of chemo-radiotherapy regimen showed a significant association with OS (p=0.037) on univariate but not multivariate analysis (Figure 1) No other baseline clinical characteristic (including age, presence of extranodal disease, gender, presence of bulky disease, disease stage, Japanese prognostic score1, IPI) was significantly association with OS. . The Japanese prognostic score and the chemo-radiotherapy regimen was significantly associated with PFS in univariate analysis (Figure 2), but only chemo-radiotherapy regimen remained significant (p=0.02) after multivariate analysis. Patients who received R-CHOP +RT or DA R-EPOCH had better PFS, than those receiving R-CHOP alone. (R-CHOP+RT vs R-CHOP: HR=0.19; 95% C.I. 0.54-0.68; p=0.001; DA R-EPOCH vs R-CHOP HR=0.26; 95% C.I. 0.09-0.73; p=0.011).

In subgroup analysis, of patients with bulky disease (n=68,), those receiving R-CHOP alone (n=20) had the inferior PFS compared to those receiving RCHOP + RT (n=19) or R-EPOCH (n=29), with 5 year PFS of 51 %, 91 % and 93%, respectively (p=0.02), but there was no significant difference in OS (80% vs 93% vs 91%; p=0.314). In contrast, in patients without bulky disease (n=39), the regimen did not have any statistically significant impact on OS (0.05) or PFS (p=0.164).


Both R-CHOP + RT and R-EPOCH are associated with excellent outcomes in patients with PMBCL in the rituximab era. In patients with bulky disease, the use of DA R-EPOCH may be preferable as it allows omission of RT without reduction in efficacy.


No relevant conflicts of interest to declare.

Author notes


Asterisk with author names denotes non-ASH members.

This icon denotes a clinically relevant abstract