Pembrolizumab in patients with CLL and Richter transformation or with relapsed CLL

Abstract

Chronic lymphocytic leukemia (CLL) patients progressed early on ibrutinib often develop Richter transformation (RT) with a short survival of about 4 months. Preclinical studies suggest that programmed death 1 (PD-1) pathway is critical to inhibit immune surveillance in CLL. This phase 2 study was designed to test the efficacy and safety of pembrolizumab, a humanized PD-1-blocking antibody, at a dose of 200 mg every 3 weeks in relapsed and transformed CLL. Twenty-five patients including 16 relapsed CLL and 9 RT (all proven diffuse large cell lymphoma) patients were enrolled, and 60% received prior ibrutinib. Objective responses were observed in 4 out of 9 RT patients (44%) and in 0 out of 16 CLL patients (0%). All responses were observed in RT patients who had progression after prior therapy with ibrutinib. After a median follow-up time of 11 months, the median overall survival in the RT cohort was 10.7 months, but was not reached in RT patients who progressed after prior ibrutinib. Treatment-related grade 3 or above adverse events were reported in 15 (60%) patients and were manageable. Analyses of pretreatment tumor specimens from available patients revealed increased expression of PD-ligand 1 (PD-L1) and a trend of increased expression in PD-1 in the tumor microenvironment in patients who had confirmed responses. Overall, pembrolizumab exhibited selective efficacy in CLL patients with RT. The results of this study are the first to demonstrate the benefit of PD-1 blockade in CLL patients with RT, and could change the landscape of therapy for RT patients if further validated. This trial was registered at www.clinicaltrials.gov as #NCT02332980.

Figure 1.

Changes in tumor burden, duration of therapy, and representative PET response in patients with RT receiving pembrolizumab. (A) The maximal percentage alteration in tumor burden from baseline in 9 RT patients. One patient (RS9), met the criteria for a PMR without having a 50% decrease in tumor burden but with a significant reduction of PET avidity. The color of each bar indicates whether a patient had prior ibrutinib therapy. (B) The response onset, and duration of therapy and DOR in RT patients. The length of the bar shows the time until the patient had a CR or a PR, along with the duration of the response and duration of therapy. Two patients (RS3 and RS7) who had SD, elected to discontinue the study in order to undergo alternative therapy. Three patients (RS1, RS2, and RS9) continued to receive pembrolizumab at the time of this report. Two patients (RS1 and RS2) had been added a signal inhibitor in addition to pembrolizumab (indicated by a blue arrow for the continuation phase) due to CLL progression in marrow (RS2) or a single locus RT progression detected on PET (RS1) (indicated by a solid red circle). RS1 had a DOR of 11 months for single-agent pembrolizumab and then received local radiation directed to one site of progression. He subsequently had a second CR and has maintained a CR with resumed pembrolizumab for 16 months of total therapy by the time of this report. RS2 had a DOR of 5 months for single-agent pembrolizumab and was then added idelalisib, and had pembrolizumab therapy for 12 months by the time of this report. RS6 had a DOR of ∼2 months before CLL marrow progression, leading to G3 thrombocytopenia. He came off therapy and received palliative care before the trial was amended to add a signal inhibitor. RS9 was in a sustained response of 3 months before the analysis cutoff. (C) The representative whole body PET images in RS2 at baseline prior to trial therapy and at the time point after 2 cycles of pembrolizumab treatment. The PET avid diseases were circled in these two PETs to compare the changes of PET-avid tumor. (D) The alteration of tumor burden and platelet count with the duration of therapy in RS2 during the single-agent therapy of pembrolizumab and the double-therapy of pembrolizumab with idelalisib. Arrow indicates the ongoing therapy. Trapezoid sign indicates a brief interruption of pembrolizumab due to thrombocytopenia.

Figure 2.

OS of patients in separate cohorts. (A-B) The OS of 25 patients in the entire cohort (A) and in a separate cohort of CLL (“no” to RT) vs RT (”yes” to RT) patients (B). (C) The OS of patients in 2 cohorts separated by having prior ibrutinib or no prior ibrutinib treatment. OS of patients who had prior ibrutinib treatment (n = 15) separated by CLL vs RT is shown in panel D. NE, not ended.

Figure 3.

Biomarker assessment of PD-1/PD-L1 expression and tumor-infiltrating T cells in treated patients. (A) The expression of PD-1, PD-L1, CD3, and CD20 detected by standard IHC in one representative RT- and one CLL-involved lymph node are shown in images taken using 40× visual field with magnification ×400. Scale bars = 50 μm. (B) The percentage of expression of PD-1, PD-L1, CD3, and CD8 in the baseline lymph nodes/tumors of 10 patients (6 RT and 4 CLL). PD-L1 expression is significantly increased (P = .03 by 2-tailed Student t test using 2-sample equal variance). PD-1 expression shows a trend for an increased level of expression (P = .10) in RT patients who have confirmed response in comparison with the CLL/RT patients who do not have confirmed responses. CD3 and CD8 expression are not different in these above 2 cohorts. Line represents median expression levels of individual antigens. Each open/filled square or triangle represents the % expression of individual antigen in 1 patient. (C) The representative images of PD-1 staining (green) in conjunction with CD3 (red) and PD-L1 expression (red) in conjunction with Pax 5 (green), tested in dual-color immunofluorescent staining using confocal microscopy. PD-L1 has minimal colocalization with Pax5-positive B cells (supplemental Table 2) and PD-L1–positive cells have the morphology of monocytic/histiocytic lineages. PD-1 partially colocalized with CD3, whereas the majority of PD-1 staining appears to be positive on tumor B cells (supplemental Table 2). Arrows point to the colocalization. Images were taken using 40× visual field with magnification ×400. H&E, hematoxylin and eosin.