Rituximab maintenance therapy in diffuse large B-cell lymphoma: is XY the most important variable?

In B-cell non-Hodgkin lymphomas (NHL) rituximab has extended the disease-free intervals of hundreds of thousands of patients. At the inception of rituximab a considerable amount of academic vigor was invested in finding the appropriate dose and frequency during induction therapy. This was followed by consideration of rituximab maintenance or extended dosing strategies. However, if maintenance rituximab does not significantly improve treatment outcomes it only represents expensive plasma. An integral step in harnessing the excitement for maintenance rituximab is to look for patients’ characteristics that can help to tailor or risk-adapt rituximab dose and/or duration of use with the goal of providing benefit to all. The primary endpoint of interest, improvement in overall survival, has only been seen in meta-analyses, leaving surrogate markers of benefit, such as event-free survival and progression-free survival in trials, to be debated at podiums and in patients’ examination rooms without a clear consensus being reached.1

The original report that triggered the spark of enthusiasm for maintenance rituximab was published by Dr. Ghielmini and colleagues and concerned patients with follicular lymphoma (FL) in whom prolonged rituximab treatment extended the duration of remission.2 The use of rituximab in FL subsequently expanded as results of randomized trials emerged showing remission prolongation with maintenance rituximab after single agent rituximab and combined rituximab-chemotherapy and then similar results in mantle cell lymphoma.73 A theme began to develop: rituximab maintenance was most useful in B-cell NHL subtypes in which the majority of patients do not have durable remissions. However, in diffuse large B-cell lymphoma (DLBCL), the most common NHL, in which the majority of patients who achieve a complete remission after rituximab-chemotherapy are cured, maintenance rituximab therapy has not been felt to be efficacious.

Nevertheless, Huang and colleagues reported a randomized trial of maintenance rituximab in patients with an objective response after six cycles of R-CHOP-14 (rituximab, cyclophosphamide, doxorubicin, vincristine, and prednisone). The maintenance rituximab was administered monthly during the first 12 months and once every 3 months during the second year.8 Patients who received maintenance rituximab had a progression-free survival rate at 5 years of 45% compared to 34% in the patients who were observed (P=0.006). The overall survival rate at 5 years was 62% with maintenance rituximab and 49% with observation (P=0.03). Maintenance rituximab improved the progression-free and overall survival of patients in all International Prognostic Index groupings. The lower progression-free and overall survival rates might be expected and were probably related to the fact that all patients who had an objective response (i.e., not just complete remissions) were included in the analysis. In this study, the results were not reported by gender, so it is not possible to determine whether the observed benefit was greater in males than in females.

In a subsequent, larger randomized trial carried out by the Eastern Cooperative Oncology Group (EGOG; ECOG 4494) in the USA, patients over 60 years of age with DLBCL were randomly assigned to receive R-CHOP or CHOP; there was then a second randomization to maintenance rituximab or no maintenance rituximab.9 Thus, this was a four-arm study including patients who received R-CHOP and no maintenance rituximab, R-CHOP with maintenance rituximab, CHOP with maintenance rituximab, and CHOP without maintenance rituximab. The results were comparable within the three groups who received rituximab during induction, as maintenance, or both, but distinctly inferior in the patients who never received rituximab. The fact that maintenance rituximab did not add to the outcome when administered after rituximab during induction has been taken as strong evidence that maintenance rituximab was superfluous in patients who achieved a remission with R-CHOP. Interestingly, a subsequent analysis of this study demonstrated that men did less well than woman in those arms given any rituximab (3 arms), but there was no sex difference in the patients who received only CHOP (1 arm).10

In 2012, the German High Grade non-Hodgkin Lymphoma Study Group (GHLSG) reported that men >60 years old had more rapid rituximab clearance than women.11 They proposed that the resultant higher plasma rituximab levels in women might be the explanation for the observed superior treatment outcome with R-CHOP in women than in men. The same group tested this hypothesis in a pharmacokinetic study in which eight doses of rituximab were administered on specific days in combination with CHOP-14 in the SMARTE–R-CHOP-14 study. They found that this seemed to eliminate the poor outcome in men that had been seen previously without the intensified rituximab therapy.12

More recently the GHLSG tested a higher dose of rituximab in men in the SEXIE-R-CHOP-14 study.13 This study investigated six cycles of R-CHOP-14 dosed at 500 mg/m in men, but the dose of rituximab in women was the standard 375 mg/m. Pharmacokinetic assays were included within the study and showed that men achieved higher peak levels of rituximab, but their total exposure to the drug was approximately the same as that of women because of more rapid drug clearance. The 3-year progression-free survival in men was 74% versus 68% in women (P=NS); the overall survival was also not significantly different.

So how do the above data integrate with those of the NHL13 trial reported by Jaeger et al. in this issue of Haematologica¿14 First of all, it is important to consider the specifics of the NHL13 trial. This was an industry-supported trial that included not only patients with DLBCL, but also those with follicular FL grade IIIb. Only a small fraction of patients had central review of the initial biopsy, but the International Prognostic Index score was available for everyone. Furthermore, it appears that some patients with FL grade III, especially FL grade IIIb have aggressive lymphoma that responds to therapy similarly to DLBCL. Nevertheless, the pathological grading of FL is not precise.15 Some patients with FL grade IIIa will have a course similar to that seen in patients with grade 1–2 FL. In those patients, maintenance rituximab would be expected to prolong remission duration on average but most would eventually relapse. However, since only about 3% of patients in the study by Jaeger et al. had FL grade IIIb, it is unlikely that this greatly affected the results.

What the authors did discover in a prospective fashion is that maintenance rituximab appeared to improve the outcome in men but not women, with no obvious impact of age on the outcome - something that was not tested in prior trials. The authors concluded that maintenance rituximab eliminated the previously found poorer outcome in men with DLBCL. This study also found that rituximab improved the treatment outcome for patients who had bone marrow involvement by NHL. We cannot tell from the paper how many of these patients had large cell lymphoma in the marrow rather than only small cell involvement. Whether this is a statistical “quirk” because of the large number of analyses done, or a real finding is not clear. Approximately 10% of the patients in the study had bone marrow involvement.

How do we interpret and apply all this information? There is increasing evidence that sex may play a part in DLBCL outcomes: in particular, men >60 years of age are relatively under-dosed with the “standard” dose of rituximab compared to women. They might, therefore, benefit from a higher dose or longer duration of treatment with rituximab. Does this mean that we should consider maintenance rituximab in men with DLBCL or only in those >60 years old? From a practical point of view, giving a higher dose of rituximab, such as the 500 mg/m administered by the GHLSG as part of R-CHOP, would be easier and more cost-effective (i.e., less total rituximab and fewer treatment visits) if the results were comparable, but this has been tested only in R-CHOP-14 and not in R-CHOP-21, which is more commonly used. This begs the question, based on the available data, should we change practice immediately? Although there are controversial aspects of this information, we believe both men and women should be made aware of the data and that older men should be offered a higher dose of rituximab or maintenance. Women on the other hand should not be offered a higher dose or rituximab maintenance after treatment for DLBCL. Whether payers will buy into this approach is unknown, but as the data now stand we believe it is a reasonable approach supported by facts not anecdotes.

Footnotes

• Financial and other disclosures provided by the author using the ICMJE (www.icmje.org) Uniform Format for Disclosure of Competing Interests are available with the full text of this paper at www.haematologica.org