Background: The tumor microenvironment (TME) plays a role in follicular lymphoma (FL) pathogenesis and survival. PD-1/PDL-1 inhibitors have been effective in non-Hodgkin lymphoma subtypes, such as primary mediastinal B-cell lymphoma, and pembrolizumab + rituximab (P-R) produced a complete response (CR) in 50% of patients with rituximab-sensitive, relapsed/refractory (R/R) FL (Nastoupil ASH 2017, Blood Adv 2022). In 2018, we conducted a phase 2 study to assess the efficacy and safety of P-R in R/R FL and diffuse large B cell lymphoma (DLBCL) including rituximab-resistant patients. Methods: This phase 2, single-center, open-label, non-randomized trial aimed to determine the objective response rate (ORR) of P-R in R/R FL (Arm A) and R/R DLBCL (Arm B). A 3 rd arm with P + Obinutuzumab (O) for R/R FL was added later (Arm C). Eligible patients had R/R disease; for FL, patients must have received prior anti-CD20 therapy and have an indication for treatment; for DLBCL, patients must have relapsed after or be ineligible for autologous stem cell transplantation. Arm A sought to include > 9 patients with rituximab-refractory disease. Weekly R 375 mg/m2 x 4 and 4 doses of P (200 mg in q21 day cycles) were given as induction; responders could receive extended P monotherapy for up to 2 years. In Arm C, patients received weekly O x 4; those with at least SD and clinical benefit could receive P for up to 2 years and 6 more doses of O given every 3 months. CT was used for on-treatment tumor assessments after cycles 4 and 8, then every 12 weeks. A 2-stage design required response at cycle 4 among at least 3 of the first 7 patients in Arm A, and among 3 of the first 9 patients in Arm B, to permit accrual to 20 (Arm A) and 17 patients (Arm B). In Arm C, at least 6 responses in the first 12 patients would be needed to proceed to total accrual of 25 patients. Additional stopping rules were included for excess deaths (>10%) and serious adverse events (SAEs) (>50%). Results: Between 4/2018 and 11/2021, 18 patients were treated on study (Table 1). All were previously treated with rituximab, with 9 refractory to their last rituximab-containing regimen. Median age was 56, 78, and 45 and median prior lines of therapy was 4, 3, and 2 in Arms A, B, and C respectively. In Arm A, 2/7 patients had rituximab-refractory disease, and int- or high-risk FLIPI. All patients in Arm B were refractory to their most recent line of therapy, and 75% had IPI > 3. In Arm A, ORR was 29% (1 CR, 1 PR). In Arm B, no patients responded (6 PD, 1 SD, 1 unevaluable). In Arm C, 3 were accrued and achieved SD as best response. Figure 1 shows outcomes of all accrued patients. Treatment-related adverse events (AE) occurred in 83% of patients, with 9 grade 3/4 AEs occurring in 5 (28%) patients (1 acute kidney injury, 2 diarrhea, 1 hypokalemia, 1 arthralgia, 1 infusion reaction, 1 back pain, 1 generalized muscle weakness, 1 heart failure). There were 5 possible grade 3/4 immune-related AEs, including 2 suspected colitis treated with steroids. With a median follow up of 4.2 years, there were 8 deaths, 6 from progressive lymphoma, 1 from AML likely related to prior therapy, and 1 from a combination of progressive lymphoma and therapy-related AML. Arms A and B closed due to futility, having failed to exceed minimum efficacy for further accrual. Arm C was closed after demonstration of no responses and discussion with the study sponsor following accrual of 3 patients. Conclusion: P-R has modest, expected toxicity but low efficacy in R/R FL and DLBCL patients previously treated with R. Our data align with CHECKMATE-140 in FL (4% ORR, Armand Blood 2021) and data from post-transplant DLBCL (10% ORR post-ASCT, Ansell JCO 2019). Despite similarities in age and FLIPI distribution with data from Nastoupil et al, we failed to confirm those efficacy findings, a finding partly attributable to inclusion of rituximab-resistant patients. Our data suggest limited utility of combined anti-CD20 and anti-PD-1 therapy in R/R FL and DLBCL, and support exploration of alternative strategies to target the TME in these diseases.