Pembrolizumab and Copanlisib for the Treatment of Relapsed or Refractory Mature T-Cell Lymphomas

Background: Outcomes for relapsed or refractory mature T-cell lymphomas (RR TCL) are poor and have remained largely stagnant despite approval of several novel single agents in this setting over the last decade (Chihara D, et al. Br J Haematol. 2017). The observed chemoresistance of many aggressive T-cell lymphomas may be overcome by non-cytotoxic treatment strategies. Rationale for exploring checkpoint blockade in RR TCL is provided by genetic and immunological features of malignant T-cells and their microenvironment (Phillips T, et al. J Immunother Cancer. 2016). The PD-1 inhibitor pembrolizumab has demonstrated moderate single agent activity in RR TCL (Barta SK, et al. Clin Lymphoma Myeloma Leuk. 2019). Nevertheless, there is concern that PD-1 inhibition in RR TCL may accelerate T-cell lymphoma progression in selected subtypes. One postulated mechanism is upregulation of PD-1 expression and PD-1 functions via the PTEN/PI3K/AKT pathway through oncogenic T cell signaling leading to growth inhibition in an auto-inhibitory fashion. Induced PD-1 deficiency results in accelerated tumor growth, which has the potential to be abrogated by PI3K inhibition (Wartewig T, et al. Nature. 2017). Furthermore, PI3K inhibition, specifically inhibition of the PI3Kβ, PI3Kγ, and PI3Kδ isoforms, can enhance antitumor immunity and susceptibility to immune checkpoint inhibition, and may result in synergistic efficacy (De Henau O, et al. Nature. 2016; Liu N, et al. Hematological Oncology. 2017). Lastly, PI3K inhibitors themselves have significant single agent activity in RR TCL (Horwitz SM, et al. Blood. 2018; Dreyling M, et al. Ann Oncol. 2017). This provides the rationale for the combination of the PD-1 inhibitor pembrolizumab with the pan-PI3K inhibitor copanlisib as a novel biological combination.

Study Design and Methods: The combination of pembrolizumab and copanlisib is being explored in a multi-center phase I/II single arm clinical trial, which is registered at www.clinicaltrials.gov as NCT02535247. Patients ≥18 years old with RR systemic mature TCL, who have progressed after a minimum of 1 systemic therapy, have measurable disease, and have adequate organ function, are eligible. Key exclusions criteria include the subtypes Adult T-cell leukemia/lymphoma (ATLL) and T-cell prolymphocytic leukemia (PLL), current immunosuppressive therapy, active autoimmune disease or GVHD, uncontrolled hypertension or diabetes, and active CNS involvement. In the phase I portion of the trial, we will determine the recommended phase II dose (RP2D) of the combination in a traditional "3+3" design using 3 dose levels: pembrolizumab will be given at a flat dose of 200 mg on day 1 of a 3 week cycle in all dose levels. Copanlisib will be given at a dose of 45 mg on day 1 for dose level -1; 45 mg on days 1 and 8 for dose level 0; and 60 mg on days 1 and 8 for dose level +1. Once the RP2D has been determined, a total of 24 patients treated at the RP2D will be enrolled using a Simon's 2-stage design in the phase II portion. The primary endpoint of the phase I portion is determination of the RP2D, and overall response rate for the phase II portion. Secondary endpoints include safety and tolerability, 6- and 12-month progression-free and overall survival, as well as duration of response. If ≤3 respond of the first 8 patients treated at RP2D, the trial will be stopped for futility. The null hypothesis (ORR ≤30%) will be rejected if ≥11 responses are observed in 24 patients treated at the RP2D and the combination deemed worthy of further consideration. This design yields a power of 80% with a type I error of 5% when the true response rate is ≥60% ORR. The probability of early stopping under the null is 80%. Correlative studies include changes in in composition, activation, and function of peripheral blood T-, NK-, and myeloid cells by flow cytometry, RNA sequencing of tumor tissue to identify a gene expression signature that may correlate with response, as well as non-invasive disease monitoring using T-cell receptor high throughput sequencing of circulating tumor DNA.