Introduction

Follicular lymphoma (FL) is the most frequent indolent lymphoma diagnosed. In limited stages, radiation-based therapy is thought to have curative potential. In advanced stages, when patients lack symptoms, watch and wait strategy is applied, whereas chemotherapy is the therapy of choice when treatment is needed (Yuda et al. 2016). Over decades, no progress had been seen in the therapy of advanced FL. The introduction of rituximab has changed the therapy of follicular lymphomas (FL) significantly, in that the combined immuno-chemotherapy not only prolongs progression- and event-free survival, but more importantly, overall survival (Herold et al. 2007; Marcus et al. 2008). Transformation into high-grade diffuse large B-cell lymphoma is observed in approx. 4% after 5 years in rituximab-treated patients, which seems lower than in the pre-rituximab era (Janikova et al. 2018).

Predictive factors that discriminate patients who might have a higher benefit from anti-CD20 therapy are not known. Serum concentration of APRIL or low recovery of IgG has been associated with poor survival (Kusano et al. 2018; Li et al. 2008; Taskinen et al. 2007). In contrast, Blaker et al. found that a higher number of intrafollicular CD68-positive macrophages was associated with a higher likelihood of transformation of follicular lymphoma patients into aggressive B-cell lymphoma (Blaker et al. 2016). In their study, all patients had been treated with rituximab-based protocols. However, our study addressed the difference between a small cohort of follicular lymphoma patients treated with rituximab-based chemotherapy and compared this with the cohort treated with chemotherapy only. Both studies thus focused on different issues.

Our study suffers from small numbers, and we used a criterion of p < 0.05 for nominal statistical significance. Therefore, the data presented in this paper need to be interpreted with caution. However, the patients were balanced between the two arms of the original protocol, and the clinical outcome of the 18 patients analyzed here reflected the original trial. As the numbers are small, the data need to be confirmed in larger trials. For such future trials, we propose that the levels as determined for each antigen in the present study (see suppl. Table 3) may be applied as cut-off values since we here demonstrated that the cohort of 18 patients analyzed was representative for the whole trial (see Fig. 1). Naturally, as the chemotherapy given in this trial is not used any more (MCP), we cannot rule out that the herein observed rituximab-based immune activating effect would also be observed when using other chemotherapy regimens such as CHOP or bendamustine. It would also be interesting to apply—in addition to the markers analyzed here—the mutational load of the respective samples, as given in the m7-FLIPI (Pastore et al. 2015).

Taken together, we show here that rituximab exerts beneficial effects especially in the subgroup of follicular lymphoma patients with low intrafollicular CD3, CD5, ZAP70 and CD8, and high CD56 and CD68 expression. Although the study by Dave et al. did not use immunohistology (Dave et al. 2004), they showed that lower T cell and higher macrophage cell signature were both associated with worse prognosis in advanced follicular lymphoma. Our study, therefore, suggests that rituximab therapy could overcome the worse prognostic feature of lower T-cell and higher macrophage infiltration in follicular lymphoma. As this was a retrospective analysis on a small subgroup of patients, these data need to be corroborated in larger clinical trials. Whether rituximab resistance may be overcome by higher dosages of CD20-specific antibodies must be addressed in prospective trials. A hint could be the so-called GADOLIN trial, in which rituximab-resistant patients with indolent lymphomas were randomly assigned to either bendamustine alone or bendamustine plus obinutuzumab that was given at a significant increased dosage as compared to the previously applied rituximab (Sehn et al. 2016).