FDG-PET as a Guide for Consolidation Radiotherapy in Primary Mediastinal Lymphoma

 IELSG37: A Randomized, Open Label, Multicenter, Two-Arm Phase III Comparative Study Assessing the Role of Mediastinal Radiotherapy after Rituximab-Containing Chemotherapy Regimens to Patients with Newly Diagnosed Primary Mediastinal Large B-Cell Lymphoma

 NCT01599559

 International Extranodal Lymphoma Study Group (IELSG)

IOSI in Bellinzona, Switzerland

752 patients

 Adult patients with previously untreated primary mediastinal lymphoma (PMBL) confirmed by immunohistochemistry with a dominant mass in the anterior mediastinum are registered at the time of initial diagnosis and receive a course of standard chemotherapy such as CHOP-14 or CHOP-21, high-dose CHOP, dose-adjusted (DA-) EPOCH, ACVBP, VACOP-B, or MACOP-B together with at least six doses of rituximab (R-chemotherapy). Within five to six weeks of completing the chemotherapy, subjects undergo a restaging fluoro-deoxyglucose-positron emission tomography (FDG-PET) scan, the results of which are centrally reviewed by a panel of experts via a dedicated website. Those with uptake of score three or less on the five-point Deauville scale (FDG uptake at or below the level of the liver) will be randomized 1:1 either to undergo consolidation mediastinal field radiotherapy or to receive no further treatment. The randomization is stratified by country, gender, type of chemotherapy given, and FDG-PET score. The radiotherapy dose is 30Gy given as 15 to 20 daily fractions, commenced within eight weeks after the last administration of R-chemotherapy. Intensity-modulated radiotherapy is allowed to decrease the dose to avoid injuring normal structures including the heart, lungs, and breasts. Patients with a partial response and a positive FDG-PET result (FDG uptake in a residual mass greater than in the liver) will go on to receive either consolidation radiotherapy or other salvage regimen, according to local practice. Follow-up is performed by scheduled CT scans at six, 12, and 24 months and clinical review up to five years.

The primary endpoint is the progression-free survival (PFS) at two years in all the randomized patients. The trial is powered to determine a noninferior outcome in patients not irradiated, based on a hazard ratio of 1.77 (one-sided α error 0.05), which requires a sample size of 376 randomized patients. Overall survival, safety, and long-term outcomes are secondary endpoints. Two interim analyses are planned. When 10 patients with an end of treatment PET score of three or less have been randomized to observation and followed for six months, the Independent Data Monitoring Committee will review the event rate of all randomized patients to determine if there is evidence of early progression in the cohort that did not undergo radiation treatment. A further interim analysis will be performed when 25 percent of the total expected events have occurred, with a critical p value ≤0.0005 based on a log-rank test comparing PFS times.

Given the young age of many patients with PMBL and the need to avoid long-term toxicity from mediastinal radiotherapy wherever possible, it is important to establish definitively which patients do not require irradiation. Incorporating FDG-PET assessment at the end of chemotherapy as the basis for randomization is one approach to addressing this issue.   A previous cohort study by the IELSG of 115 patients with PMBL demonstrated excellent results using conventional R-chemotherapy and mediastinal radiotherapy in most cases, with five-year overall survival and PFS rates of 92 percent and 86 percent respectively and highly significant differences between those having a PET score of one to three versus four to five at the completion of chemotherapy (5-year PFS 99% vs. 68%, p<0.0001) (Martelli M et al. J Clin Oncol, 2014;32:1769-1775). There are data from a cohort treated with R-DA-EPOCH to suggest that, in many patients, consolidation radiotherapy may not be necessary (Dunleavy K et al. N Engl J Med. 2013;368:1408-1416), a finding supported by a retrospective study from the British Columbia Cancer Agency where R-CHOP was used as the R-chemotherapy regimen (Savage KJ et al.Blood. ASH Annual Meeting Abstracts. 2012;120:303.). In light of these findings, it is logical to undertake a prospective study of the safety of omitting mediastinal irradiation in those patients with the most favorable response to chemotherapy.

 The PFS figures for patients treated with R-CHOP and mediastinal radiotherapy are excellent, but there is a feeling that we may be overusing radiation and that limiting its use to those with residual active disease would be preferable. The complicated R-DA-EPOCH regimen seems to result in cures without radiotherapy, but there is reluctance to adopt this regimen widely in the absence of confirmatory results from prospective randomized trials. The difficulty of salvaging PMBL that has relapsed is another cautionary factor. Second-line chemotherapy has a poor record in this illness, so compelling data are needed if de-escalation of initial therapy is to be recommended for some patients.