The Treatment of Primary Mediastinal Large B-Cell Lymphoma: A Two Decades Monocentric Experience on 98 Patients

The first line treatment for primary mediastinal large B-cell lymphoma (PMLBCL) still remains a matter of debate even if the literature confirms that an anthracycline-containing regimen should be the main choice. The European experience shows the superiority of “third-generation” dose-dense regimen like MACOP-B (methotrexate, doxurubicin, cyclophosphamide, vincristine, bleomycin, prednisone) or VACOP-B (same as MACOP-B, with etoposide instead of methotrexate) over CHOP (cyclophosphamide, adriamycin, vincristine and prednisone) or CHOP-like regimen. The addition of rituximab to third-generation regimen does not seem to cause any advantages in terms of overall survival (OS) and progression free-survival (PFS), while external radiotherapy (RT) has shown a good efficacy as a consolidation strategy at the end of induction chemotherapy, especially in converting partial responses (PR) into complete responses (CR).

Between October 1989 and April 2010, 98 (58 females and 40 males) previously untreated PMLBCL patients were diagnosed and subsequently treated at our Institution. All patients were treated with MACOP-B regimen, concurrent rituximab was administered in 57 patients (58.2%) and 67 patients (68.4%) received mediastinal RT. Among 57 patients who received rituximab, 37 (64.9%) underwent RT whereas, among 41 who did not receive rituximab, RT was delivered in 30 (group 4, 73.2%) patients. 11 patients (group 1, 11.2%) received chemotherapy alone and 37 (group 3, 37.8%) received besides immunotherapy and RT (Table 1). All patients were assessed at the diagnosis and after the treatment with computed tomography and positron emission tomography (after 2001) scan.

Main aims of our study were the effectiveness of the regimen measured as overall response rate (ORR) and patients survival. Sixty-one (62.2%) out of 98 patients achieved a CR and 27 (27.6%) were in PR after 12 cycles of MACOP-B regimen (with or without rituximab). Twenty-one patients in PR after (immuno)chemotherapy converted the response into CR with mediastinal RT. At the end of the scheduled treatment, 82 patients (83.7%) achieved a CR and 6 a PR (6.1%), yielding an ORR of 89.8%. At a median follow-up of 5.6 years, 9 patients relapsed within the first 2 years of treatment. During the follow-up 15 patients died, of whom 13 as a consequence of disease relapse or progression. The projected OS at 17 years is 72% with a PFS and a disease free survival (DSF) of 86.8% and 88.4% respectively (Figure 1 A-C). The subgroup analysis shows a statistically significant difference in term of OS (p=0.0003) but not in term of PSF and DFS among the four groups of treatment (Figure 1 D-F). All the patients receiving consolidation RT obtained a CR without differences between subgroup 3 and 4. RT seems to have a small consolidative potential in patients who obtained CR after chemo-immunotherapy alone: there are no differences in term of DFS between subgroup 2 and 3. No statistically significant differences in terms of OS, PFS and DSF occurred among patients received rituximab or not, regardless of a subsequent RT. Our monocentric experience spans a period of 20 years and indicates that a third-generation regimen like MACOP-B is feasible and could be a standard of first line treatment for PMLBCL. In agreement with the literature, adding rituximab doesn’t improve the outcome. Mediastinal RT, delivered as a consolidative strategy, impacts on global survival and on CR rates. In particular, RT after third-generation regimen remains a good strategy to convert PR into CR, but it may be avoided in patients obtaining CR after (immuno)chemotherapy.

Figure 1

View largeDownload slide

Figure 1

View largeDownload slide

Close modal