Primary Mediastinal Large B-Cell Lymphoma: the role of PET scan response assessment in guiding choice of treatment

On 14^th January 2017, Ferreri and Martelli published a review article on Primary Mediastinal Large B-Cell Lymphoma (PMBCL) in Critical Reviews in Oncology and Hematology.¹

So far, current studies have been unable to fully define the role of PET-CT scan in terms of Complete Metabolic Response (CMR) in patients with PMBCL, and especially in whether consolidation RT may not be required based on a negative scan.

Key Highlights:

• In a MSKCC retrospective study of 54 patients with PMBCL who received R-CHOP/ICE dose-dense therapy without mediastinal RT, in PET-negative patients 3-yr OS and PFS were 88% and 78%, respectively²
• In a BCCA study of 96 patients with PMBCL treated with R-CHOP, out of 59 PET scans after treatment 35 (59%) were PET-negative (two received RT) and 24 (41%) were PET-positive (23 received RT). No survival difference between PET-positive and PET-negative patients were found after median follow-up of >5yrs³
• An Italian retrospective analysis of 74 patients with PMBCL administered R-MACOP-B reported similar results; after treatment 51 (61%) were PET-positive and received mediastinal RT and the other 23 (39%) PET-negative patients did not receive RT. 5-yr EFS was not significantly different between PET-negative (91%) versus PET-positive (90%) patients⁴
• In the phase 2 prospective IELSG 26 study, 115 patients were eligible for central review of PET/CT scan after chemo-immunotherapy treatment using the five-point Deauville score. Eighty-nine percent of these also received RT. CMR, defined as a completely negative scan or residual 18-FDG activity below the mediastinal blood bool (MBP) uptake (score 1–2) was achieved by 47%. In the other 61 patients, residual uptake was >MBP but <liver uptake (score 3) in 27 (23%) patients, and was also slightly and markedly >liver uptake (score 4–5) in 34 (30%) patients. Patients with a score of 3 had equally good outcomes and no recurrence. Using liver uptake as a cut-off (score 4–5) for PET positivity effectively distinguished high and low risk of failure; 5-yr PFS of 99% and 68% (P < 0.0001) and 5-yr OS of 100% and 83% (P = 0.0003)⁵
• In an Italian study of 80 patients with PMBCL administered rituximab based chemotherapy, patients with PET Deauville score 1–3 had a better 3-yr OS versus patients with a score of 4–5 (100% vs 77%; P < 0.05)⁶
• In PMBCL patients administered DA-EPOCH-R, 50% were PET-positive (defined as FDG uptake >MBP after chemo-immunotherapy) and 3/18 had PD without consolidation RT
• The IELSG 26 study also measured SUV, Metabolic Tumor Volume (MTV), and Total Lesion Glycolysis (TLG) at baseline PET/CT according to standard protocol in 103 patients. 5-yr OS in patients with low versus high TLG was 100% vs 80%, respectively (P = 0.0001); PFS was 97% versus 64%, respectively (P < 0.0001)⁷

Based on the results of the BCCA study, Ferreri and Martelli state that the no difference in survival between PET-positive and PET-negative patients indicates that a PET-guided RT approach in patients treated with R-CHOP can potentially reduce the use of unnecessary RT.

The data in DA-EPOCH-R treated patients show that PET-CT has a good negative predictive value, however positive predictive value is poor due to a high false positive rate, which Ferreri and Martelli recommend needs to be further defined before changing therapy for patients based on FDG-PET alone. However, de-escalating therapy based on PET-negativity is becoming accepted in clinical practice. Ferreri and Martelli state that rituximab-induced inflammatory response or thymic rebound may increase uptake of FDG, reducing specificity. To investigate this, the international phase 3 IELSG 37 study is ongoing, aiming to assess consolidation RT in patients with PET-negative mediastinal masses after standard chemo-immunotherapy.⁸ Ferreri and Martelli state this study should show a non-inferior outcome in patients who are not administered RT. They also hypothesize that the trial may ultimately lead to individualized treatment for each individual patient by tailoring treatment to PET response; this will limit additional RT only to PET-positive patients who do not respond to chemo-immunotherapy.

Lastly, the authors of the review recommend that in future trials, TLG level at baseline PET scan should be considered, as it has been demonstrated to be a powerful predictor of outcome in patients with PMBCL. 

Abstract:

Primary mediastinal large B-cell lymphoma (PMLBCL) is a distinct clinical and biological disease from other types of DLBCL. It is more frequent in young female and constitutes 6%-10% of all DLBCL. PMLBCL is characterized by a diffuse proliferation of medium to large B-cells associated with sclerosis. Molecular analysis shows it to be a distinct entity from other DLBCL. Rituximab CHOP/MACOP-B-like regimens followed by with mediastinal radiotherapy (RT) were associated with a 5-years PFS of 75%-85%. More intensive regimens, as DA-EPOCH-R without mediastinal RT, have shown very promising results, but this therapeutic advance needs to be confirmed in further prospective trials. The role of consolidative mediastinal RT should be still better assess in prospective comparative studies. PET-CT scan is a powerful tool to define the real quality of response and it is hoped that future prospective trials may allow its role in the de-escalation of mediastinal RT.