Patterns of failure of diffuse large B‑cell lymphoma patients after involved-site radiotherapy

Radiotherapy (RT) in combination with chemoimmunotherapy is highly efficient in the treatment of diffuse large B‑cell lymphoma (DLBCL). This retrospective analysis evaluated the efficacy of the treatment volume and the dose concept of involved-site RT (ISRT).

We identified 60 histologically confirmed stage I-IV DLBCL patients treated with multimodal cytotoxic chemoimmunotherapy and followed by consolidative ISRT from 2005-2015. Progression-free survival (PFS) and overall survival (OS) were estimated by Kaplan-Meier method. Univariate analyses were performed by log-rank test and Mann-Whitney U‑test.

After initial chemoimmunotherapy (mostly R‑CHOP; rituximab, cyclophosphamide, doxorubicin, vincristine, and prednisolone), 19 (36%) patients achieved complete response (CR), 34 (64%) partial response (PR) or less. Excluded were 7 (12%) patients with progressive disease after chemoimmunotherapy. All patients underwent ISRT with a dose of 40 Gy. After a median follow-up of 44 months, 79% of the patients remained disease free, while 21% presented with failure, progressive systemic disease, or death. All patients who achieved CR after chemoimmunotherapy remained in CR. Of the patients achieving PR after chemotherapy only 2 failed at the initial site within the ISRT volume. No marginal relapse was observed. Ann Arbor clinical stage I/II showed significantly improved PFS compared to stage III/IV (93% vs 65%; p ≤ 0.021). International Prognostic Index (IPI) score of 0 or 1 compared to 2-5 has been associated with significantly increased PFS (100% vs 70%; p ≤ 0.031). Postchemoimmunotherapy status of CR compared to PR was associated with significantly increased PFS (100% vs 68%; p ≤ 0.004) and OS (100% vs 82%; p ≤ 0.026). Only 3 of 53 patients developed grade II late side effects, whereas grade III or IV side effects have not been observed.

These data suggest that a reduction of the RT treatment volume from involved-field (IF) to involved-site (IS) is sufficient because no marginal failures occurred. The concept of IS will likely reduce the risk for late sequelae of RT.