All the patients were treated at the Department of Oncology, Tianjin Union Medicine Center (Tianjin, China). HPC patients who received immunotherapy, including an autologous dendritic cell (DC) vaccine and cytokine-induced killer (CIK) cell therapy comprised the immunotherapy group (I group). All the patients provided informed consent prior to DC-CIK treatment and were required to meet the following inclusion criteria: i) A cytological or histological diagnosis of HPC; ii) hospitalization between January 4, 2012 and December 31, 2012; iii) adequate kidney and liver function and normal coagulation; iv) a Karnofsky performance status score of >60; and v) expected survival of >3 months.

The control group included patients who satisfied the abovementioned inclusion criteria, but did not receive immunotherapy [non-immunotherapy (NI) group]. The follow-up for all the patients was completed on November 14, 2013. The median survival time (MST) was compared between the two groups and the benefit of the DC vaccine and CIK cell-based immunotherapy regimens was evaluated in terms of DTH and improvement in QOL.

This retrospective study was an open-label, parallel-group, single-institution, non-randomized study and complied with the class III medical techniques policy of the Ministry of Health of China, as described in the guidelines for treatment with autologous immune cells (T cells, natural killer cells) (12,13). The study protocol was approved by the Ethics Committee of the Tianjin Union Medicine Center. Prior to treatment, all the patients were asked to provide written informed consent.

The DC vaccine and CIK cell therapy schedule was as follows: The day on which peripheral blood mononuclear cells (PBMCs) were collected was defined as day 0. On day 8, 1×10⁷ DCs were infused intravenously once/week for 6 weeks, and on day 11, 1×10⁹ CIK cells were infused intravenously once/day for 4 days (Table I).

Using the Fresenius Kabi system (Fresenius Kabi AG, Bad Homburg, Germany), PBMCs were collected from each patient under electrocardiographic monitoring. The cells were cultured overnight, then adherent cells (including monocytes) and suspension cells (including lymphocytes) were collected separately (14–16).

The DC vaccine was prepared according to the following protocol: Isolated PBMCs were resuspended in a serum-free culture medium (Hyclone, Logan, UT, USA). The PBMCs were then seeded into a 175-cm³ flask to a concentration of 2–4×10⁶/ml, divided among three bottles and then placed in a 37°C, 5% CO₂ incubator (Thermo Fisher Scientific, Waltham, MA, USA). The cell concentration and the number of bottles used may be adjusted according to the number of cells. After 2 h of incubation, the 175-cm³ flasks were gently agitated; the suspension cells were discarded (or collected and used for culturing CIK cells), then washed 1–2 times with serum-free medium and, following addition of the DC medium, placed again in the 37°C, 5% CO₂ incubator. Autologous serum, granulocyte-macrophage colony-stimulating factor, interleukin (IL)-4 and tumor necrosis factors (all from PeproTech EC, Ltd., London, UK) were added on the 3rd day. The tumor antigen (Peprotech, Inc., Rocky Hill, NJ, USA) was added on the 5th day and the ripening factor was added following addition of the tumor antigen. Flat blood tests were performed and the flasks were again placed in the incubator for 2 h. After the flasks were washed twice with serum-free 1640 medium, the DC culture solution was added and the flasks were placed back in the incubator. DC cells were collected between the 7th and 8th days and then washed three times with saline. The quality of the preparation was determined at the same time.

The CIK cells were prepared according to the following protocol: The culture bottles were pre-coated with a 5 µg/ml final concentration of CIK reagents before preparing the CIK cells, then placed in a 4°C environment overnight and washed with phosphate-buffered saline (PeproTech EC, Ltd.) 2–3 times on day 0. The CIK cells were resuspended in the initial CIK cell medium at a concentration of 2–3×10⁶/ml and spread into a culture bag according to the number of cells. Autologous plasma was then added and cultured in a 37°C, 5% CO₂ incubator (PeproTech EC, Ltd.). IL-2, interferon 2γ (both from PeproTech EC, Ltd.) and CD3 monoclonal antibody (100 µg/ul; cat. no 20150605; Beckman Coulter, Inc., Brea, CA, USA) were added using 5-ml disposable syringes with needles. To ensure that all factors were added in the culture bag, the aspiration was repeated five times, after which the culture bags were placed in the 37°C, 5% CO₂ incubator. Samples were prepared and counted every 2–3 days using 2-ml disposable syringes with needles. Appropriate additional CIK expansion bottle medium was added according to the results of the count in order to maintain a CIK cell concentration of 1–3×10⁶/ml and the total number of cells was recorded at the same time.

The immune phenotype markers of HLA2DR, CD86 and CD83 were used for DCs; CD3, CD8 and CD56 were used for CIK cells. The bacteria, endotoxin and fungus levels of the cultured samples satisfied the release criteria for infusion (Table II).

To determine DTH, 4 µg of tumor lysate was injected intradermally 1 week after the last infusion and the results were read 48 h later. The diameter of the induration corresponded to the result, i.e., >10 mm was considered as strongly positive, 5–10 mm as positive, 2–5 mm as weakly positive and <2 mm as negative. Fever, insomnia, anorexia, arthralgia and skin rash were evaluated as adverse effects during therapy. Certain patients developed ≥2 events. The general status in terms of physical strength, appetite, sleep and body weight were evaluated as QOL indices. QOL scores were defined as follows: 0, none of the 4 indices; 1, any one of the 4 indices; 2, any 2 of the 4 indices; 3, any 3 of the 4 indices; and 4, all indices. The degrees of 4, 3, 2 and 1 were considered as an improvement in the general status (Table III) (15,16).

Overall survival was measured from the time of enrollment to the date of death and compared between groups using the Kaplan-Meier method (Table IV, Fig. 1).

Clinical data were collected from the Inpatients Electronic Medical Records of Tianjin Union Medicine Center, re-analyzed and documented for use with the EpiData software, version 3.02 (Epidata Association, Odense, Denmark). Statistical analyses were performed using the SAS statistical software package (SAS Institute Inc., Cary, NC, USA). The MST curves were calculated using the Kaplan-Meier method. The associations between variables were compared using the Pearson's Chi-square test. P<0.05 was considered to indicate a statistically significant difference.

A total of 407 patients (270 men and 137 women) were enrolled in this study, with a median age of 66.4±11.5 years (range, 37–93 years). Among these, 157 patients had hepatocellular cancer, 46 had gallbladder cancer, 54 had cholangiocarcinoma and 150 had pancreatic cancer. Of the 407 patients, 124 underwent primary tumor resection, 33 received radiotherapy, 44 received chemotherapy and 36 received Chinese medicine treatment. All these treatments were administered within the 3 months preceding the initiation of immunotherapy (Table V). The characteristics of the patients in the two groups were well-balanced with respect to the evaluated factors, including age, gender, diagnostic methods and tumor type. The treatment strategies also did not differ significantly between the two groups.

A DTH skin test used to assess the immune response to cell therapy for all patients and the results were recorded for 72 of the 77 patients in the I group. Twelve patients (16.5%) had a strongly positive response; 12 (16.5%) had a weakly positive response; 18 (25%) had a positive result; and 30 (42%) patients failed to show an immune response. Thus, 42 (58%) patients in total exhibited a positive effect. QOL was recorded for 72 of the 77 patients in the I group. Of these, 9 (13%) patients exhibited improvements in all QOL events; 13 (18%) exhibited improvements in 3 events; another 13 (18%) exhibited improvements in 2 events; 12 (16%) exhibited improvement in 1 event; and 25 (35%) exhibited no improvements of the QOL factors. Adverse effects were assessed in 72 of the 77 patients in the I group. Of these, 37 (51%) did not have fever or cold symptoms and 5 (7%) developed cold symptoms. A total of 30 patients developed fever: Low fever occurred in 10 (14%), moderate fever in 11 (15%) and high fever in 9 patients (13%). A total of 12 patients (17%) developed anorexia; 18 (25%) developed insomnia; 22 (31%) developed arthralgia; and 8 (11%) developed skin rash (located at the intradermal injection site in 7 cases and on the upper as well as the lower body in 1 case; Table III).

The MST differed between the I and NI groups (134 vs. 115 days, respectively; Table IV) and it was significantly longer (19 days) in the I group compared with that in the NI group (P=0.014). The overall survival curves for the I and NI groups are shown in Fig. 1.