High Complete Response Rates with Pembrolizumab in Combination with Rituximab in Patients with Relapsed Follicular Lymphoma: Results of an Open-Label, Phase II Study

Background : Follicular lymphoma (FL) tumors are infiltrated with antitumor T cells, however, their function is impaired by immune checkpoints such as PD-1/PD-ligand pathway. Blocking PD-1 enhances the function of antitumor T cells in FL [Nattamai D & Neelapu SS, Blood, 2007]. In addition, blocking PD-1 on NK cells has been shown to enhance the ADCC effect of NK cells [Gorgun et al, Clin Cancer Res, 2015]. We reasoned that the combination of pembrolizumab, an anti-PD-1 antibody, and rituximab (R), an anti-CD20 antibody that induces ADCC, is likely to be synergistic through activation of both the innate and adaptive immune systems and result in enhanced clinical activity in FL.

Methods: We evaluated pembrolizumab and R in an open-label, non-randomized, single institution, phase II trial (N=30). Key inclusion criteria included adult patients (age ≥ 18 years), with FL grade 1-3a, ECOG performance status (PS) 0-1, in relapse after ≥1 prior therapy and R sensitive disease, defined as a complete (CR) or partial response (PR) lasting at least 6 months after the most recent R-containing therapy. Patients received R (375 mg/m² IV) on days 1, 8, 15, and 22 of cycle 1 and pembrolizumab (200mg IV) q 3 weeks for up to 16 cycles starting on day 2 of cycle 1. Primary endpoint was overall response rate (ORR). Response assessment was performed as per the Lugano Classification (Cheson et al, J Clin Oncol, 2014).

Results: As of February 2, 2017, 30 patients have initiated therapy and 10 are still in therapy. All 30 patients are evaluable for safety. Twenty-five patients have had at least 8 infusions of pembrolizumab or stopped therapy due to progression or toxicity and are evaluable for efficacy. Median age was 64 years (range 43-84), 57% were male, 73% had an ECOG PS of 0, 73% had intermediate or high risk FLIPI, and 50% had high tumor burden as defined by GELF criteria. Median prior therapy = 1 (range 1-4). Adverse events (AE) regardless of causality were mild, most were grade 1-2. Grade 3 AE's included nausea (N=2), infusion reaction (N=2), hypertension (N=1), aseptic meningitis (N=1), and pneumonia (N=1). Immune-related (IR) AEs included diarrhea (grade 1 N=6, grade 2 N=4), rash (grade 1 N=2, grade 2 N=2), transaminitis (grade 1 N=4), pneumonitis (grade 2 N=1), pancreatitis (grade 2 N=1), and hypothyroidism (grade 1 N=1). Five discontinued therapy as a result of recurrent grade 2 IR AEs (diarrhea N=3, rash N=1, pneumonitis N=1). Among evaluable patients, ORR is 64% (CR N=12; PR N=4), CR rate is 48%. With a median follow up of 11 months (range 3.7-21), 15 (60%) patients are still in ongoing remission. No deaths have been observed. PD-L1 expression was tested in 8 baseline tumor samples using PD-L1 22C3 IHC pharmDx and was detected in histiocytes in all 8 tumors, but was present in only 1-8% of tumor cells in 5 tumors and was not associated with response. Immune cell gene signature analysis in baseline tumors by Nansotring in 12 patients showed an association between presence of high levels of CD8+ T-effector score and induction of a CR. Additional biomarker analyses from baseline tumors and serial peripheral blood mononuclear cell samples are ongoing and will be presented along with updated clinical data.

Conclusions: Promising efficacy was observed with pembrolizumab and R in relapsed FL with meaningful overall and CR rates. The combination also appears to be well tolerated. Analysis of immune cell gene signatures in baseline tumors may identify a potential predictive biomarker. Further investigation is warranted.