Phenotypic polarization and functional polarization of cells cultured in IL-15
We evaluated the quality of the cell phenotype and compared cell composition at the end of cultivation. No significant differences were found in the composition of the cells between culturing with IL-2 or IL-15 (Figure 1A). Expression of the activation antigen CD25 on CD3+CD56-or CD3+CD56+ T lymphocytes cultured with IL-15 was significantly higher than with IL-2(Figure 1B).
ACTIL-15 cells maintained positive staining for the TCM marker. While the TCM markers CD62L and chemokine receptor (CCR7) were virtually absent on ACTIL-2 cells (0.3%), ACTIL-15 cells (62.1%) retained the expression of CD62L and CCR7 (Figure 2A). Lymphocytes were also counted at indicated time points. Figure 2B illustrates that lymphocytes similarly expanded in media containing IL-2 and IL-15.
We also assessed the functional activity of the generated lymphocytes. ACTIL-15 and ACTIL-2 cells were added to target cells AGS, N87, and MKN-45 in an effector-to-target ratio of 50:1 and tested in europium release assays. ACTIL-15 cells induced significantly more cytotoxicity against AGS and N87 than did ACTIL-2 cells, whereas the in vitro cytotoxicity of ACTIL-15 cells against MKN-45 cells was not higher than that of ACTIL-2 cells (Figure 3A). We also evaluated cytokine secretion and found that ACTIL-15 cells significantly increased IFN-g secretion from AGS and N87 cells but not from MKN-45 cells (Figure 3B). In sum, ACTIL-15 cells exhibited cytotoxicity against gastric cancer cells and their cytotoxicity against AGS and N87 cells was stronger than that of ACTIL-2 cells.
We considered that ACTIL-15 cells might have differential effects in vivo and in vitro and, therefore, may be effective against MKN-45 cells in vivo. After culture in medium containing IL-2 or IL-15, lymphocytes proliferated extensively and similarly. In three independent experiments, ACT cells were transferred into MKN-45 gastric carcinoma-bearing mice, and lymphocyte tumor infiltration and persistence were compared. We measured tumor infiltration of adoptively transferred human cells and the secretion of cytokines by ACT cells in tumor sites. ACTIL-15 cellspromoted tumor infiltration and increased IFN-g secretion potential of adoptively transferred lymphocytes (Figure 4).Moreover,the proliferation of MKN-45 cells was inhibited by ACTIL-15 cells, as measured by TUNEL staining and PCNA immunohistochemistry (Supplementary Figure S1).
We next measured tumor volume and survival in mice receiving ACTIL-2 or ACTIL-15 cells compared with these measures in untreated controls. Treatment with ACTIL-15 inhibited tumor growth upon adoptive transfer (Figure 5A). Moreover, mice that received ACTIL-15 cells had a significantly improved survival (P = 0.049, ACTIL-15 vs. ACTIL-2) (Figure 5B).
These data indicated that the adoptive transfer of ACTIL-15 cells is preferable to ACTIL-2 cells in tumor immunotherapy.
The role of ACTIL-15 in human gastric cancer immunotherapy
The combination with immunotherapy and chemotherapy has been proposed as a therapeutic strategy with the potential for improved survival rate and prognosis for gastric patients with cancer[23]. Thus, we assessed the clinical efficacy and safety of ACTIL-15 administered with S-1 plus oxaliplatin for treating patients with advanced gastric cancer.
Clinical evaluation of patients with gastric cancer
Seventy-three patients were enrolled at the National Cancer Center/National Clinical Research Center for Cancer/Cancer Hospital, Chinese Academy of Medical Sciences, and Peking Union Medical College from November 2014 to February 2019; three were excluded before randomization (Figure 6). The remaining 70 patients were randomized into the combination therapy (ACTIL-15 plus S-1/oxaliplatin) group or the chemotherapy (S-1/oxaliplatin) group (35 patients per group). One patient from the combination therapy group and six patients from the chemotherapy group were eliminated because of poor adherence, leaving 34 patients in the combination therapy group and 29 patients in the chemotherapy group for final analysis (Figure 6). The characteristics of all patients are detailed in Table 1.
Effector cell treatment and assessment of safety
ACTIL-15-related self-limiting adverse drug reactions, including pyrexia, chill, myalgia, and fatigue, were reported in 11% (7/63) of patients; however, the reactions did not delay or stop the treatment. No patients exhibited pulmonary or renal symptoms, signs of infection, hepatic deterioration, or autoimmune disorders.
The effect of ACTIL-15 on OS and PFS
As of March 1, 2019, the median survival of patients in the combination therapy group was 472 days (95% confidence interval (CI), 276–668 days), whereas that of patients in the chemotherapy group was 266 days (95% CI, 200–332 days; P < 0.05, Figure 7A). The 1-year OS rate of the combination therapy group (66.7%) also was significantly higher than that of the chemotherapy group (40.5%; P < 0.05). Multivariate Cox regression model of OS with treatment and other prognostic variables as covariates showed that combined ACT IL-15 and S-1/oxaliplatin treatment(vs chemotherapy alone, HR 0.42; 95%CI, 0.21–0.86; P=0.02), female sex (HR 2.53; 95%CI, 1.19–5.40; P=0.02), and fewer organs with metastases (≤ 2 vs ≥ 3)(HR 2.24; 95%CI, 1.12–4.51; P=0.02) were associated with longer survival time (Supplementary Table S1).
The median PFS of the combination therapy group was 153 days (95% CI, 109–197 days), whereas the median PFS of the chemotherapy group was 136 days (95% CI, 103–169 days; P > 0.05, Figure 7B). The 1-year PFS rate of the combination therapy group (18.2%) was significantly higher than that of the chemotherapy group (14.7%; P < 0.01). These results are evidence that the addition of ACTIL-15 to a standard chemotherapy regimenimproves patient survival.