A total of 303 inpatients with malignant tumors who received CIK treatment were recruited from the Department of Biotherapy, the Fourth Affiliated Hospital of China Medical University (Shenyang, China) between January 2008 and December 2011. Of these patients, 133 received treatment with CIKs obtained from an autologous blood source (the autologous group) and 170 underwent treatment with allogeneic CIKs obtained from the blood of human leukocyte antigen (HLA) haploidentical donors (the allogeneic group). The demographic and disease characteristics of the patients are shown in Table I. Symptoms, signs and the outcomes of routine blood tests, and kidney and liver function tests, as well as electrocardiographic examinations were compared prior to and following the CIK infusion. In addition, adverse events were recorded using the National Cancer Institute Common Terminology Criteria for Adverse Events version 3.0 (2003). An evaluation of immune function identified that CD3⁺, CD3⁺CD4⁺, CD3⁺CD8⁺, CD3⁻CD56⁺CD16⁺ and CD19⁺ cells, and IgG, IgM, IgA, C3 and C4 were routinely detected in the peripheral blood. Prior to treatment, written informed consent was obtained from patients and their relatives, and the possible adverse events were explained. The entire protocol was approved by the Ethics committee of the Fourth Affiliated Hospital of China Medical University.

Venous blood was collected prior to chemotherapy or at least one month following chemotherapy. For the offspring (HLA haploidentical donors), the routine blood tests, and liver and kidney function tests were normal and negative for hepatitis A virus-IgM, hepatitis B surface (HBs) antigen (Ag), HBs antibody (Ab), hepatitis B e (HBe)-Ag, HBe-Ab, hepatitis B core-Ab, hepatitis C virus (HCV)-Ab, HCV-IgG, Syphilis-Ab and human immunodeficiency virus-Ab. As reported in our previous study (13), 50 ml of peripheral venous blood was collected and the mononuclear cells were isolated using Lymphocyte Separation Medium (Haoyang Biological Manufacture, Co., Ltd., Tianjin, China). These cells were added to flasks precoated with RetroNectin (Takara, Shiga, Japan) and CD3 monoclonal antibodies (Johnson & Johnson, New Brunswick, NJ, USA) followed by incubation in serum-free KBM551 (Takara) at 37°C in a 5% CO₂ atmosphere. On the day of culture, 1,000 U/ml human recombinant interferon-γ (PeproTech, Rocky Hill, NJ, USA), 500 U/ml recombinant human interleukin (rhIL)-1α (Invitrogen Life Technologies, Carlsbad, CA, USA) and 1,000 U/ml rhIL-2 (Quangang Pharmaceutical Co., Ltd., Jinan, China) were added. Four days later, 1,000 U/ml rhIL-2 was added and the cells were transferred into a GT-T610 culture bag (Takara). Cell growth was observed every other day and cells were stained with 0.4% trypan blue (Yocon Biotechnology Company, Co., Ltd., Beijing, China) and the viable cells were counted. Following 18 days of culture, cells were infused once daily (>1×10⁹ cells; viability rate, >95%). A cycle of treatment comprised of between three and five infusions and >5×10⁹ cells were infused per cycle. Prior to infusion, the surface immunophenotype of CIKs was determined by a standard fluorescence cytometry labeling protocol using fluorochrome-conjugated monoclonal mouse anti-human antibodies against CD3, CD8, CD4, CD45, CD19 and CD16CD56 (BD Biosciences, San Jose, CA, USA). Assays for the detection of bacteria, mildew and endotoxin were performed. Briefly, the presence of bacteria and mildew was investigated by incubating cultured CIKs on agar at 37°C, with subsequent inspection for the growth of microbial contaminants. Endotoxin levels were determined using a chromogenic endotoxin assay kit (Chinese Horseshoe Crab Reagent Manufactory, Co., Ltd., Xiamen, China), according to the manufacturer’s instructions. Only sterile and endotoxin-free preparations were used clinically.

The paired method was performed on the patients with pathologically confirmed stage IIIb-IV non-small cell lung cancer (NSCLC). Clinical data, including name, gender, age, Karnofsky performance status (KPS), pathological type (14), TNM stage (15) and history of treatment, were collected. These patients had a KPS score of ≥60 and a predicted survival of more than three months. Patients were divided into the following two groups: The CIK group and the optimal support group. In the CIK group, patients were treated with allogeneic CIKs for at least two cycles. Observations were performed from the initial therapy until mortality or until the final follow-up, and survival was determined. In the optimal support group, patients receiving optimal support therapy in the interim between receiving CIK infusions were recruited, and gender, age, KPS score and disease stage were matched between the two groups. The follow-up was performed via telephone or hospital visit.

SPSS software package version 13.0 (SPSS, Inc., Chicago, IL, USA) was employed for statistical analysis. Quantitative data were expressed as the mean ± standard deviation and comparisons were performed using Student’s t-test. Qualitative data were compared using the χ² and Fisher’s exact tests. Survival was delineated by the Kaplan-Meier method, followed by the log-rank test and P<0.05 was considered to indicate a statistically significant difference.

The general information of the patients with malignancies who received CIK therapy is shown in Table I. Age, gender and histological type were comparable between the two groups and a total of 647 cycles were applied (n=308 in the autologous group and n=339 in the allogeneic group). The total number of CIKs used for therapy in the autologous group was between 1.76×10¹⁰ and 1.35×10¹¹ (mean, 2.11±0.32×10¹⁰) and between 1.16×10¹⁰ and 6.71×10¹⁰ (mean, 2.29±0.36×10¹⁰) in the allogeneic group, with no marked differences identified between the two groups (t=1.147; P>0.05). The majority of patients received less than four cycles of therapy (Table II) and the therapy was completed within one year. In 13 and four patients, therapy endured for more than one and two years, respectively. In addition, four patients received >10 cycles of therapy (autologous group, n=3; allogeneic group, n=1). In the autologous group, the highest number of therapy cycles was 20 for one patient, with therapy lasting four years with a total of 1.35×10¹¹ CIKs. In addition, one patient underwent 13 cycles of therapy, which was completed in two years with a total of 7.3×10¹⁰ CIKs and one patient received 11 cycles of therapy, which was completed over 2.5 years with a total of 6.5×10¹⁰ CIKs. In the allogeneic group, one patient received 11 cycles of therapy, which was completed over 3.5 years with a total of 6.71×10¹⁰ CIKs.

During and following the infusion of CIKs, among the 133 patients in the autologous group and 170 patients in the allogeneic group, euphoria was observed in 66 (49.6%) and 86 (50.6%) patients, physical improvements in 21 (15.8%) and 28 (16.5%) patients, and pain relief in 15 (11.3%) and 20 (11.8%) patients, respectively. The major adverse events included insomnia, fever, nausea, vomiting and mild abdominal pain, which were identified in nine (6.8%), eight (6.0%), two (1.5%) and one (0.8%) patients, respectively, in the autologous group and 11 (6.5%), 10 (5.9%), one (0.6%) and one (0.6%) patients, respectively, in the allogeneic group. No marked differences in adverse events were identified between the two groups (P>0.05). Adverse events usually occurred on the day of infusion. Insomnia was non-responsive to diazepam and spontaneously recovered between three and five days later. Fever was mild or moderate, body temperature was ≤38.5°C and concomitant flu-like symptoms were present; the fever recovered spontaneously between one and three days later. Nausea, vomiting and mild abdominal pain were rarely observed in these patients and when present, the patients often recovered following symptomatic treatment. Symptoms of allergic reaction or rejection, such as rashes, diarrhea, jaundice and hematuria, were rare, with rashes observed in only one patient. These adverse events were not associated with the number of CIKs used for therapy, the frequency of the therapy, chronic rejection or accelerated rejection due to repeated infusions. Notably, after two months, no evident rejection was observed in the four patients who received CIKs from the blood of two of their offspring. In addition, the outcomes of routine blood tests and liver and kidney function tests, as well as electrocardiogram examinations were compared between the two groups, prior to and following therapy within each group, and no significant differences were identified (P>0.05; Table III). Data for the autologous group are not shown.

To determine whether the patients’ stage was associated with the occurrence of adverse events, 94 patients with lung cancer (Table I) that were treated with CIKs were divided according to their TNM stage, and the adverse events were compared. Among the 40 patients in the autologous group, 10 patients with stage I–IIIa disease developed fever, insomnia and flu-like symptoms, and 30 patients with stage IIIb-IV disease developed nausea, vomiting, fever and flu-like symptoms. Of the 54 patients in the allogeneic group, seven patients with stage I–IIIa disease developed fever, insomnia, rashes and mild abdominal pain, and 47 patients with stage IIIb-IV disease developed nausea, vomiting, fever and flu-like symptoms. No marked differences were observed in adverse events between stage I–IIIa and IIIb-IV patients within the same therapy group, and between patients in the same TNM stage but different therapy group (P>0.05).

With regard to immune function, cellular and humoral immune function were comparable between the patients in the two groups prior to and following treatment (P>0.05; Table IV). To evaluate the impact of high-dose short-term infusion of CIKs on immune function, four patients (one in the autologous group and three in the allogeneic group) with KPS scores of >90 and stable disease were recruited and treated only with intravenous infusion of CIKs for two cycles (>1×10¹⁰ per cycle for 28 days). Immune function was detected prior to and two, seven and 14 days following the infusion (Fig. 1). On day two of the first cycle of CIK infusion, the proportion of CD3⁺, CD3⁺CD4⁺ and CD3⁺CD8⁺ cells, and the CD4⁺/CD8⁺ ratio were increased, which was most evident for the increasing ratio of CD4⁺/CD8⁺ (Fig. 1A). However, on day seven, the above indexes returned to levels, which were comparable with or slightly lower than the levels prior to infusion. This condition continued into day 14 following the second cycle of CIK infusion. IgG levels continued to increase over two consecutive cycles of CIK infusion (Fig. 1B). The proportion of CD19⁺ and CD3⁻CD56⁺CD16⁺ (Fig. 1A), and humoral immunity indexes, IgA, IgM, C3 and C4 (Fig. 1B), were almost invariant over two consecutive cycles of CIK treatment.

Between January 2008 and December 2010, a total of 54 patients were recruited for the present study. In the allogeneic CIK and optimal support groups (n=27 per group), no marked differences were identified among age, gender, KPS score and pathological type or stage (P>0.05). In the allogeneic CIK and optimal support groups, the duration of follow-up was between four and 18 and six and 24 months, the median duration of follow-up was 13.0 and 14.0 months and the median survival was 11.0 (95% confidence interval (CI), 9.2–11.8 months) and 8.0 (95% CI, 9.2–11.8 months) months, respectively (χ²=5.618; P=0.018; Fig. 2).