PD-1 Blockade Combined with R-CHOP in Patients with Newly Diagnosed EBV-Positive Diffuse Large B-Cell Lymphoma: A Retrospective Analysis

Background

Epstein-Barr virus (EBV) -positive diffuse large B cell lymphoma (EBV+ DLBCL) is a relatively rare aggressive B-cell lymphoma. Most patients present with advanced disease, high incidence of extranodal involvement of 30% to 80% (Keane, Eur J Haematol, 2019), as well as poor prognosis when treated with immunochemotherapy. However, no consensus on standard treatment has been established by now. It was found that the expression of programmed death ligand-1 (PD-L1) was increased in EBV+ DLBCL (Kim, Cancer Res Treat, 2019). In addition, chronic EBV infection results in T-cell exhaustion that can be reversed by PD-1 blockade. Compared with PD-L1-positive EBV-negative DLBCL cell lines, the application of PD-1 blockade in PD-L1-positive EBV+ DLBCL cell lines can significantly restore T cell proliferation and activity(Quan, PLoS One, 2015), suggesting that targeting the PD-L1/PD-1 pathway represents a potential therapy for EBV+ DLBCL. Therefore, we retrospectively analyzed the data of nineteen patients with EBV+ DLBCL who were treated with PD-1 blockade combined with RCHOP in our center, hoping to provide reference for subsequent research.

Methods

Eligible pts had treatment-naive histologically confirmed diagnosis of EBV+ DLBCL. Additional eligibility criteria included at least one measurable lesion, acceptable haptic and hematological function. Pts with primary CNS lymphoma, primary mediastinal large B-cell lymphoma, secondary CNS involvement were excluded. Additional exclusion criteria included history of immune system disease, uncontrolled infection. Once deemed eligible, pts will receive six cycles of RCHOP combined with PD-1 blockade(21 days per cycle) as induction therapy and subsequently maintenance with PD-1 blockade for 1 year(21 days per cycle). End of treatment(EOT) responses were assessed by PET-CT according to the Lugano Classification 2014. The primary objective was to assess the overall response rate(ORR) at EOT.

Results

Between May 1, 2019, and Sep 1, 2022, 19 pts were enrolled. Pts' baseline demographics, clinical and pathological characteristics are shown in Table 1. The median age was 64 years old without significant gender difference. The majority of pts were intermediate- to high-risk by IPI. EBV DNA can be detected in blood serum at diagnosis in more than half of all patients. No patient was diagnosed with double expressor lymphoma among 18 specimens that were detected c-Myc and BCL-2 with immumohistochemical staining. EBV-encoded RNA (EBER) was positive(>80% of the tumor cells) in all specimens. PD-L1 positivity(>10% of the tumor cells) or CD30 positivity(>5% of the tumor cells) was also quite common. Besides, no double hit lymphoma was confirmed in patient with Fluorescence situ hybridization test(n=13).

Treatment course and efficacy evaluation of all pts are summarized in Figure 1. All patients enrolled in this study had completed 6 cycles of induction therapy. Totally, the ORR at EOT was 94.7%, and all achieved CMR. The ORR of patients with PD-L1 positivity(n=10) was numerically higher than that of patients with negative PD-L1 expression (n=5) (100% vs 80%). EBV DNA detected in blood serum of thirteen patients have decreased to undetectable level after induction therapy.

The immune-related adverse events (irAEs) were emphatically analyzed. One patient had recurrent fever after every dose of PD-1 blockade without obvious evidence of infection. Another patient developed grade 3 diarrhea since the second cycle of PD-1 blockade maintenance. Colonoscopy showed mucosal swelling and multiple superficial ulcers in the whole colon, which indicated immune colonitis. Maintenance therapy was therefore discontinued. No irAE was observed in the rest of patients.

With a median follow-up time of 13.1m (7.2-38.5m), the median PFS and OS were not reached. Both of 1 year OS and PFS was 89.5%. Overall, 2 pts experienced disease progression (1 at EOT, 1 at 2 months after ASCT), 2 pts died (1 due to progressive disease, 1 due to COVID-19 infection ).

Conclusions

PD-1 blockade combined with R-CHOP as induction therapy in patients with newly diagnosed EBV+ DLBCL has demonstrated satisfactory response and limited irAEs. Longer follow-up is still warranted to confirm the long-term benefit, especially for patients with maintenance of PD-1 blockade. Still, we believe this therapeutic strategy can be regarded as a successful attempt at improving prognosis for pts with EBV+ DLBCL.