Historically, patients (pts) with early stage DLBCL treated without consolidative IFRT more commonly relapsed in sites of initial disease. While event-free survival (EFS) was superior with IFRT after CHOP in the short-term in a large randomized trial (NEJM 339:21), prolonged follow-up demonstrated equivalent overall survival due in part to late relapses (Blood 98:724a). The optimal therapy for pts with early stage DLBCL in the rituximab-CHOP (R-CHOP) era has not been established. To determine the impact of IFRT after R-CHOP treatment on outcome of early stage DLBCL, we reviewed medical records of 67 consecutive pts with stage I and II DLBCL who were primarily treated with R-CHOP with or without IFRT between February 2001 and March 2006. Three to 8 cycles of R-CHOP were administered, and 30–45 Gy IFRT was administered for those irradiated. The Kaplan-Meier method was used for the estimates of outcome. EFS was measured from the date of diagnosis until first documented disease progression or death from any cause. Comparisons were adjusted for IPI by Cox proportional hazards regressions, and all tests were two sided. Median follow-up was 2.6 years (range 0.5–5.5). 45 (67%) of the pts were men. Median age at diagnosis was 60 years (range 20–92); 18 (27%) had ‘B’ symptoms; bulky disease (defined as mass >10cm) was present in 15 (22%); 27 (40%) had stage I and 40 (60%) had stage II disease. Stage-adjusted IPI score distribution was: IPI 0, n=6 (9%); IPI 1, n=17 (25%); IPI 2, n=24 (36%); IPI 3, n=16 (24%); IPI 4, n=4 (6%). 39 (58%) pts had extranodal disease (23% soft tissue, 21% head & neck, 56% others). As expected, since the majority of pts had more than 1 IPI risk factor, the 2-year event-free survival (EFS) was 68% (median EFS not reached). 11 (16%) pts had primary refractory disease including 3 pts who died during treatment, and they were excluded from further analysis. Of the remaining pts, 34 (61%) received IFRT, and 22 were treated with R-CHOP alone. The IPI scores of the 2 groups were balanced, but as expected, stage II disease was more common in pts who received R-CHOP alone. The EFS at 2 years for pts treated with consolidative IFRT was 87%, and with chemotherapy alone was 72%. When controlling for IPI score and the presence of bulky disease, there was a significant benefit in EFS for patients treated with R-CHOP + IFRT compared with R-CHOP alone (p=0.027). Therefore, in pts with early stage DLBCL treated with R-CHOP, we find improved outcomes with a combined modality approach including consolidative IFRT compared with chemotherapy alone as was observed in the CHOP era. Our study has inherent biases due to its retrospective nature, and limited follow-up. However, to our knowledge, this is the first study that compare R-CHOP with and without IFRT in early stage DLBCL, and therefore has important implications on future clinical trial designs.

Author notes

Disclosure: No relevant conflicts of interest to declare.