A “PET” topic in primary mediastinal large B-cell lymphoma: positive or negative, and how to handle it in the end

Primary mediastinal large B-cell lymphoma (PMLBCL) is regarded as a separate entity among the aggressive large B-cell lymphomas, characterized by unique clinical, pathologic and molecular features [1]. Immunochemotherapy with rituximab–cyclophosphamide, doxorubicin, vincristine and prednisone (R-CHOP) is still viewed as the current standard of care, but several promising more aggressive regimens have also been used in this disorder [2–6]. However, until now these alternative combination chemotherapy regimens have not yet been directly compared systematically [2–6]. Radiotherapy is frequently used as an additional treatment modality for this lymphoma, but its precise role still needs to be further defined. Since progression-free survival (PFS) for PMLBCL after therapy is now 75–90% and long-term overall survival 85–90%, earlier identification of the minority of patients who are destined to fail treatment is important [2–6]. In this respect it is possible that utilizing imaging techniques such as positron emission tomography–computed tomography (PET-CT) during the course of the disease and at the end of therapy may provide important data, which could help in recognizing a subgroup of patients who are most likely to fail therapy.

PET-CT is increasingly used in Hodgkin lymphoma (HL) and diffuse large B-cell lymphoma (DLBCL) [7–9]. In the interim setting, PET-CT (iPET) may facilitate early treatment modification, while end-of-treatment PET-CT (ePET) may well guide decisions regarding the addition of complementary radiotherapy or further salvage therapy. Until now, such decisions have essentially been based on empirical data, since the results of large ongoing randomized trials are still awaited. However, in the specific setting of PMLBCL, both iPET and ePET are widely used in daily practice, based merely on projection of data and conclusions derived from prior clinical experience gained with HL and DLBCL.

In this issue of the journal, Cheah et al., from Peter McCallum Cancer Center in Melbourne, Australia, report their single-institution experience on the use of PET-CT for interim and final evaluation of response in 28 patients with PMLBCL [10]. In this series R-CHOP-14 or R-CHOP-21 were the most frequently used regimens (n = 21; no R in one case), while only four and three patients received R-da-EPOCH (R-dose adjusted etoposide, prednisolone, vincristine, cyclophosphamide, doxorubicin) and Hyper-CVAD ± R (hyperfractionated cyclophosphamide, vincristine, doxorubicin, dexamethasone, with or without R), respectively.

Regarding iPET (mostly performed after two or three cycles), 35% of 23 patients with available data were reported as positive according to the standard Deauville criteria (score 4 or 5) [11], while 30% had positive findings when evaluated according to a modified version of the International Harmonization Project (IHP) criteria [12,13]. Irrespective of the method used, iPET appeared to have little impact on the eventual patient outcome. Nevertheless, in this study chemotherapy was escalated in some patients in response to receiving a report of a positive iPET (iPET-pos), with a potential further reduction of the positive predictive value (PPV), which was only 12–14% in this series. This observation is generally in keeping with other studies: Moskowitz et al. reported the results of iPET after four cycles of chemotherapy in 51 patients treated with augmented R-CHOP-14 × 4 followed by additional consolidation therapy with three cycles of ICE (ifosfamide, carboplatin, etoposide) chemotherapy, in cases who were iPET-negative (iPET-neg) or iPET-pos but with a negative biopsy [14,15]. The minority of iPET-pos patients, who had a positive biopsy, received a similar consolidation regimen plus R, followed by autologous stem cell transplant (ASCT). As iPET positivity was more loosely defined here as uptake > background, 47% of patients were regarded as iPET-pos. After the administration of a non-cross-resistant subsequent treatment, iPET had no impact on the outcome at all, and using a change in maximum standardized uptake value (ΔSUV_max)-based criterion was also shown to be insufficient in this respect [15]. In another recently published study, Avigdor et al. analyzed iPET results in 30 patients, mostly treated with R-VACOP-B (R–etoposide, doxorubicin, cyclophosphamide, vincristine, prednisone, bleomycin) (R-CHOP-21 was also used but to a lesser extent) without radiotherapy [16]. In this series, 47% of the patients were regarded as iPET-pos using the IHP criteria, and 5-year PFS for iPET-neg vs. -pos patients was 94% vs. 57%, respectively. Although this difference was statistically significant, its magnitude was much less impressive than that observed in HL [8]. However, in this respect, it should be stressed that very heterogeneous definitions for iPET positivity were used in the different studies [10,14–16]. Based on the above data, current evidence for the routine use of iPET in PMLBCL outside clinical trials still appears to be too inconclusive to recommend it definitively as a routine procedure.

The most interesting and instructive part of the study reported here by Cheah and colleagues relates to the implications of ePET results on the outcome and further management of patients with PMLBCL [10]. Using the same criteria for positivity as for iPET, the authors set a much more stringent than the usual definition for end-of-treatment PET positivity [10–13], and as a result, ePET was considered positive only if the Deauville score was 4 or 5 (but not 3) or if a modified IHP criterion was met. Under this strict definition, only seven of 28 (25%) patients remained as ePET-pos and, while the negative predictive value (NPV) for ePET was 95%, the positive predictive value (PPV) was only 43%. These seven patients and the two patients with Deauville score 3 were not irradiated but received salvage therapy with R-ICE or Hyper-CVAD and ASCT (in 8/9), although 5/9 had achieved a conventional radiographic response and no biopsy was performed prior to the initiation of salvage therapy. After ASCT, six of these nine patients remained in remission, and this is the reason for the low PPV of ePET. It is also of interest that among four mediastinal biopsies performed after ASCT because of persistent PET positivity, three were negative, revealing foamy macrophages and/or necrotic changes.

Recent moderately sized or large studies have attempted to address the issue of ePET in PMLBCL [5,17–21]. Using less strict criteria (either the IHP or Deauville score ≥ 3), 41–69% of patients had a positive ePET. When radiotherapy was applied uniformly to ePET-pos patients, a positive ePET was associated with inferior PFS, but the magnitude of the difference between ePET-neg and -pos patients was < 25–30% [18,20], and statistical significance was borderline or even absent in some studies [17,19,21]. Generally, ePET-pos patients enjoyed long-term PFS of > 65% if given radiotherapy, illustrating that this was adequate treatment for most patients, and leaving salvage chemotherapy and ASCT as a questionable option. However, in patients with more intense fluorodeoxyglucose (FDG) uptake, the efficacy of radiotherapy appeared to be more limited, as most relapses or disease progressions occurred in responding patients with Deauville scores of 5 (without primary progression) or 4 or in those with SUV_max ≥ 5 (roughly corresponding to Deauville score 5 or the highest level of uptakes for Deauville 4) [17,18,20]. Furthermore, according to personal experience (unpublished data) and some National Cancer Institute (NCI) data as well [5], several ePET-pos patients still remain PET-pos after radiotherapy without experiencing relapse in the long-term. This observation may also support the possibility of false positive ePET interpretations in PMLBCL, especially if FDG uptake is relatively low. Finally, it is noteworthy that the above observations appear valid irrespective of the chemotherapy regimen used, i.e. R-CHOP [17–19], R-MACOP-B (R–methotrexate, doxorubicin, cyclophosphamide, vincristine, prednisone, bleomycin)/R-VACOP-B [17,20,21] or R-da-EPOCH [5]. Thus in our view, the conclusion drawn by Cheah et al. [10] appears to be extremely valid, clinically relevant and worth stressing again: “Positive ePET scan by visual analysis or 5PS [5-point scale] predicts inferior PFS, but an inflammatory response resulting in metabolic abnormality is frequently observed and is difficult to differentiate from residual disease. Metabolically active residual masses after the completion of treatment should be biopsied to confirm viable lymphoma if salvage therapy is planned.” [10].

In conclusion, there is currently no convincing evidence for the use of interim PET-CT in PMLBCL. On the other hand, at the end of immunochemotherapy, PET-CT is a valuable tool for the evaluation of response. The ongoing International Extranodal Lymphoma Study Group (IELSG)-37 study will, in the near future, no doubt reveal whether radiotherapy can be omitted in cases with a negative ePET [22]. However, in the mean time, it is important to recognize that PET-CT positivity at the end of therapy is not universally an indication for salvage therapy and ASCT. In this regard, it may be prudent to suggest that only patients with radiographic progression, as documented by conventional imaging, should proceed to ASCT. In contrast, the majority of ePET-pos patients, i.e. those who have achieved at least a partial response, should receive radiotherapy or follow a biopsy-driven approach to achieve satisfactory final outcomes.

Potential conflict of interest

Disclosure forms provided by the authors are available with the full text of this article at www.informahealthcare.com/lal.