Key points

Introduction

Primary mediastinal B-cell lymphoma (PMBCL) is a rather infrequent aggressive lymphoma, putatively arising from a transformed thymic B cell. It accounts for less than 5% of all non-Hodgkin lymphomas, and typically affects adolescents and young women in their third or fourth decade. It clinically presents with bulky mediastinal masses, usually exerting compressive effects on nearby vessels and airways ( Fig. 1 ), giving rise to the possible abrupt onset of dyspnea; dysphagia; thoracic pain; and facial, neck, breast, and arm edema ( Fig. 2 ), or infiltrating the adjacent lung parenchyma. It should be often regarded as a hematological emergency and promptly treated; the initial treatment decision is crucial for the management of this disease.

Conventional radiological appearance of a massive mediastinal mass that roughly alters the mediastinal RX profile ( A ). The same adenopathic bulk as it appears on contrast-enhanced CT scan of the chest ( B ).

The superior vena cava syndrome: superficial collateral veins at both the upper limbs and at the upper thoracic quadrants become evident as a consequence of the compression or obstruction of the superior vena cava.

PMBCL was originally recognized as a subtype of diffuse large B-cell lymphoma (DLBCL) since the 1994 Revised European American Lymphoma (REAL) classification, and it has been regarded as a unique clinical and biological entity since the 2001 World Health Organization classification. Apart from its peculiar clinical presentation and pathologic features, this disease also displays a unique molecular fingerprint, which clearly distinguishes it from the rest of DLBCL. That same fingerprint, however, partly overlaps with the molecular profile of nodular sclerosis Hodgkin disease, with which it shares at least a third of its genes, abnormalities on chromosome 9p (in nearly three-quarters of cases), and CD30 expression (although weaker).

In the last 20 years, several studies—mostly retrospective in their fashion due to the rarity of the disease and to the difficulty of designing randomized clinical trials specifically addressed to these patients—have tried to provide answers to several issues regarding the optimal treatment and management of patients. Above all, although the necessity of an anthracycline-based induction is understood, the choice of the initial chemotherapy approach is still debated, including the value of the addition of rituximab. Although most of institutions have agreed to adopt an R-CHOP (cyclophosphamide, doxorubicin, vincristine, and prednisone) induction (as they would do for any other subtypes of DLBCL) others prefer a more intense approach, either based on weekly administered third-generation regimens or on more aggressive strategies.

Apart from the debate on which first-line approach seems more suitable, there are other aspects that require clarification:

• 
  

  The actual role of external beam radiotherapy (RT), recognized as an adjuvant strategy consolidates the a response to chemotherapy, and the possibility of its omission in certain categories of patients to reduce the likelihood of radiation-induced long-term toxicity

  
  
• 
  

  The value of fluorodeoxyglucose (FDG) positron emission tomography (PET) scan and the way in which its results are interpreted in this peculiar context, either when performed at disease staging or after induction treatment

  
  
• 
  

  If PET itself can be regarded as a possible guide to the administration of a subsequent consolidation treatment, whether based on external RT or on a high-dose approach and autotransplant

  
  
• 
  

  The treatment of refractory and relapsing patients, for whom standard approaches have shown rather unsatisfactory results

Some new drugs are now emerging as potentially active agents in the setting of recurrent disease, as their mechanisms of action are based on specific biologic features of PMBCL.

Purpose of this article is to review the currently adopted strategies in the first-line treatment of PMBCL, to analyze the risks and benefits of postinduction RT consolidation, and to discuss the potential role of PET in helping physicians optimize the approach to their patients. A preview to some new agents on the horizon is also provided. A patient evaluation overview is provided in Box 1 .

Introduction

Primary mediastinal B-cell lymphoma (PMBCL) is a rather infrequent aggressive lymphoma, putatively arising from a transformed thymic B cell. It accounts for less than 5% of all non-Hodgkin lymphomas, and typically affects adolescents and young women in their third or fourth decade. It clinically presents with bulky mediastinal masses, usually exerting compressive effects on nearby vessels and airways ( Fig. 1 ), giving rise to the possible abrupt onset of dyspnea; dysphagia; thoracic pain; and facial, neck, breast, and arm edema ( Fig. 2 ), or infiltrating the adjacent lung parenchyma. It should be often regarded as a hematological emergency and promptly treated; the initial treatment decision is crucial for the management of this disease.

Conventional radiological appearance of a massive mediastinal mass that roughly alters the mediastinal RX profile ( A ). The same adenopathic bulk as it appears on contrast-enhanced CT scan of the chest ( B ).

The superior vena cava syndrome: superficial collateral veins at both the upper limbs and at the upper thoracic quadrants become evident as a consequence of the compression or obstruction of the superior vena cava.

PMBCL was originally recognized as a subtype of diffuse large B-cell lymphoma (DLBCL) since the 1994 Revised European American Lymphoma (REAL) classification, and it has been regarded as a unique clinical and biological entity since the 2001 World Health Organization classification. Apart from its peculiar clinical presentation and pathologic features, this disease also displays a unique molecular fingerprint, which clearly distinguishes it from the rest of DLBCL. That same fingerprint, however, partly overlaps with the molecular profile of nodular sclerosis Hodgkin disease, with which it shares at least a third of its genes, abnormalities on chromosome 9p (in nearly three-quarters of cases), and CD30 expression (although weaker).

In the last 20 years, several studies—mostly retrospective in their fashion due to the rarity of the disease and to the difficulty of designing randomized clinical trials specifically addressed to these patients—have tried to provide answers to several issues regarding the optimal treatment and management of patients. Above all, although the necessity of an anthracycline-based induction is understood, the choice of the initial chemotherapy approach is still debated, including the value of the addition of rituximab. Although most of institutions have agreed to adopt an R-CHOP (cyclophosphamide, doxorubicin, vincristine, and prednisone) induction (as they would do for any other subtypes of DLBCL) others prefer a more intense approach, either based on weekly administered third-generation regimens or on more aggressive strategies.

Apart from the debate on which first-line approach seems more suitable, there are other aspects that require clarification:

• 
  

  The actual role of external beam radiotherapy (RT), recognized as an adjuvant strategy consolidates the a response to chemotherapy, and the possibility of its omission in certain categories of patients to reduce the likelihood of radiation-induced long-term toxicity

  
  
• 
  

  The value of fluorodeoxyglucose (FDG) positron emission tomography (PET) scan and the way in which its results are interpreted in this peculiar context, either when performed at disease staging or after induction treatment

  
  
• 
  

  If PET itself can be regarded as a possible guide to the administration of a subsequent consolidation treatment, whether based on external RT or on a high-dose approach and autotransplant

  
  
• 
  

  The treatment of refractory and relapsing patients, for whom standard approaches have shown rather unsatisfactory results

Some new drugs are now emerging as potentially active agents in the setting of recurrent disease, as their mechanisms of action are based on specific biologic features of PMBCL.

Purpose of this article is to review the currently adopted strategies in the first-line treatment of PMBCL, to analyze the risks and benefits of postinduction RT consolidation, and to discuss the potential role of PET in helping physicians optimize the approach to their patients. A preview to some new agents on the horizon is also provided. A patient evaluation overview is provided in Box 1 .

First-line chemotherapy combinations in prerituximab era: different perspectives

Since prerituximab era, the first-line chemotherapy schedule has represented a matter of controversy. Although the CHOP regimen was mainly adopted by US centers, the European experience has carried out evidence that MACOP-B (methotrexate, doxorubicin, cyclophosphamide, vincristine, bleomycin, prednisone) or VACOP-B (etoposide, doxorubicin, cyclophosphamide, vincristine, bleomycin, prednisone), both weekly dosed third-generation dose-dense regimens, could be superior to CHOP. As a consequence of the application of dose-dense regimens, in fact, remission rates and survival functions have appeared to be at least as good as—or probably even better than—those observed for DLBCL patients, thus retracting the initial impression that PMBCL was per se a prognostically unfavorable subset of DLBCL.

However, this conclusion was only drawn from existing reports, since no randomized clinical trial has been carried on so far, and prospective studies are also lacking ( Table 1 ). Overall it is clear that an anthracycline-containing regimen should be regarded as the first approach to PMBCL.

Lazzarino and colleagues first showed the MACOP-B/VACOP-B superiority on CHOP both in terms of complete response (CR) rates and relapse-free survival (RFS), with a 73% CR rate for the former treatment versus 36% for the latter treatment, and a 3-year RFS of 58% versus 38%. The same issue was confirmed by a subsequent retrospective multicenter Italian experience of 138 patients. Patients on MACOP-B/VACOP-B achieved better results than those on CHOP, both with CR and event-free survival (EFS) rates, with a statistically significant difference in low and low–intermediate International Prognostic Index (IPI) risk groups (into which most patients with PMBCL generally fall), although lacking significance in high–intermediate-risk and high-risk disease.

The results of a multinational retrospective analysis published in 2002 by the International Extranodal Lymphoma Study Group (IELSG) clearly showed that patients treated with third-generation regimens or high-dose therapy performed better than those who received a first-generation approach (ie, CHOP, Fig. 3 ), and that RT (delivered to 80% of the enrolled patients) played a pivotal role in the consolidation of response, as it was able to convert partial responses into CR, as documented with a gallium ( ⁶⁷ Ga) scan.

Overall survival curves of the main chemotherapy subgroups applied as first-line treatment in the prerituximab era.

( From Zinzani PL, Martelli M, Bertini M, et al. Induction chemotherapy strategies for primary mediastinal large B-cell lymphoma with sclerosis: a retrospective multinational study on 426 previously untreated patients. Haematologica 2002;87:1261; with permission.)

On the basis of these results, MACOP-B or VACOP-B induction, followed by mediastinal RT, was thus suggested to be a suitable induction strategy for PMBCL patients. Nevertheless, the effective superiority of third-generation treatments over CHOP and CHOP-like regimens has never been validated, either in larger phase 3 studies or in randomized trials. Alternative strategies, such as the application of high-dose methotrexate, have yielded similar response rates, although with a significantly higher incidence of toxic events and hospitalization rates.