Efficacy of adjuvant immunotherapy with cytokine-induced killer cells in patients with locally advanced gastric cancer

All patients who had undergone gastrectomy and were histologically confirmed gastric adenocarcinoma were consecutively recruited through the Department of Tumor Biological Treatment Clinic at The Third Affiliated Hospital, University of Soochow between May 2002 and June 2005, as part of a clinical trial study on gastric adenocarcinoma. These patients were diagnosed and histologically confirmed with stage IIIA, IIIB, or IV (M0) according to the International Union Against Cancer (UICC 2002) TNM system; had to have received 6 cycles of adjuvant chemotherapy based on 5-FU; and had to have had an Eastern Cooperative Oncology Group performance status of 2 or less before adjuvant chemotherapy. The exclusion criteria for eligibility were receipt of adjuvant radiotherapy or other immunotherapy, concurrent active malignancy, and recurrence identified within 6 months after operation. All patients underwent R0 curative gastrectomy with D2 lymph node, that is, N1 and N2, dissection. Indications for total stomach resection were diffuse or mixed-type cancer according to Lauren’s classification and intestinal-type tumor of the middle and upper stomach with a proximal margin of no less than 5 cm. This study was conducted according to the principles of the Declaration of Helsinki and was approved by the Ethics Committee of The Third Affiliated Hospital of Soochow University. All the patients provided informed consent prior to enrollment.

Of the 158 enrolled patients, 151 patients were eligible for the study. If patients who received immmunotherapy with CIK cells were treated as an ‘immunotherapy group’ or ‘treatment group’, the rest without immunotherapy were treated as a ‘no immunotherapy group’ or ‘control group’.

The peripheral blood mononuclear cells (PBMCs) were isolated by Ficoll-Conray density gradient centrifugation, as described previously [25], and were collected using blood cell separators (Baxter, Deerfield, IL, USA). The number of seeded PBMCs is about 1 × 10⁷. After the viability of the PBMCs was assessed by trypan blue exclusion, the PBMCs (2.0 × 10⁶/ml) were plated onto 6-well dishes (Nunc, Roskilde, Denmark) and cultured with Medium I containing RPMI 1640 in the presence of human interferon-gamma (1.0 × 10⁶ U/L, Shanghai Fosun Pharma Co., Shanghai, China); recombinant human interleukin 2 (5.0 × 10⁵ U/L, Shandong Quanguang Pharmaceutical Co., Quanguang, China); 10 % inactivated human serum; 25 mM HEPES; and 2 mM l-glutamine. The cells were incubated in a humidified atmosphere with 5 % CO₂ at 37 °C. After 24 h, monoclonal antibody against CD3 (100 μg/L, Antibody Diagnostic Inc., New York, NY, USA) and interleukin-1 alpha (1.0 × 10⁵ U/L, Promega Biological Products, Ltd., Shanghai, China) were added. After another 48 h, the supernatant was removed by aspiration and the cells were cultured in Medium II (Medium I without interferon-gamma). The medium was replaced every 3 days. On days 1, 7, 14, 21, and 28, the cells were identified and sorted by flow cytometry (Beckman-Coulter, Fullerton, CA, USA). Before infusion, the viability of CIK was tested by the dye-exclusion with no less than 95 % viable cells. The cytotoxic activity of the CIK cells was determined by co-incubation with the natural killer cell-sensitive K562 cell line (American Type Culture Collection, Manassas, VA, USA) as described previously [24].

Before enrollment in this study, all patients received 6 cycles of multidrug adjuvant chemotherapy based on 5-FU. The patients in the immunotherapy group received at least 3 cycles of CIK-cell therapy after adjuvant chemotherapy unless recurrence was ascertained. The patients in the control group did not receive immunotherapy. The first cycle was started 6 weeks after the end of adjuvant chemotherapy, and the subsequent cycles were started at the intervals of 8–12 weeks. More than 1 × 10⁹ CIK cells were transfused into patients within 1 h every second day via superficial vein. Five transfusions were defined as 1 cycle. When the patients in either the immunotherapy group or the control group were diagnosed with recurrence, second-line chemotherapy or palliative surgery was performed.

The postoperative baseline and follow-up investigations were documented. The baseline assessments included a complete medical history, physical examination, and diagnostic imaging, including abdominal ultrasonography or computed tomography and chest radiography.

Follow-up was the same for the immunotherapy and control groups, and performed every 3 months for the first 2 years after CIK-cell therapy, every 6 months for the next 3 years, and yearly thereafter. Follow-up consisted of physical examination, blood counting, liver function, CEA level, abdominal ultrasonography or computed tomography, and chest radiography. Gastroscopy was also performed for patients in whom regional recurrence was suspected. In patients with recurrence and patients who died, the site and date of the first recurrence and the date of death were recorded. Disease recurrence was diagnosed by physical and radiological examinations, and routine histological examinations were carried our as needed. Patients were followed up until they were lost to follow-up or died or until October 25, 2010.

In the immunotherapy group, lymphocyte subsets were detected by flow cytometry in peripheral blood 2 weeks after the completion of adjuvant chemotherapy (baseline) and 1 and 8 weeks after the start of every cycle of CIK-cell therapy. In the control group, lymphocyte subsets were detected by flow cytometry 6 months after the completion of adjuvant chemotherapy.

Differences in distribution of selected demographic and clinical characteristics between the immunotherapy and control groups were evaluated using the Student’s t test and χ² test. Lymphocyte subsets before and after CIK therapy were compared by the paired-sample t test, and continuous data at multiple time points in the same individual were analyzed by repeated-measures analysis of variance. The main end point was overall survival (OS). Secondary end points were disease-free survival (DFS) and cellular immune response. OS and DFS were defined as the time from the date of operation to the date of death from any cause or the first occurrence of a neoplastic event (relapse or second malignancy) or the date of the last follow-up. Participants who were alive or recurrence free at the end of the study period or lost to follow-up were censored. OS and DFS curves were estimated using the Kaplan–Meier method, and the Cox’s model for hazard ratio (HR) and 95 % confidence interval (CI) was performed for comparison of the immunotherapy with control groups. Multivariable model was adjusted with possible underlying variables that were statistically significant in the univariate analysis. P < 0.05 was considered statistically significant. Data were analyzed using SPSS software (version 13.0, SPSS Inc., Chicago, IL, USA.).

A total of 158 patients were recruited for the study at The Third Affiliated Hospital of Soochow University from April 2002 to June 2005. Seven patients were excluded as 4 patients in immunotherapy group were refused further CIK-cell therapy after the first cycle; the other 3 were in the control group and received other immunotherapies after enrollment. Therefore, the final analyses included 151 eligible patients, 77 in the control group, and 74 in the immunotherapy group. There were no significant differences in demographic and clinical characteristics between the 2 groups except that the proportion of female patients was higher in the immunotherapy group (P = 0.037) (Table 1).

The proliferation and phenotypes of the PBMCs after CIK induction varied between individuals. The cell number increased more than 100-fold on average after 14-day incubation. The number of CIK (CD3⁺CD56⁺) cells increased greatly, from 400-fold to more than 1,400-fold depending on the individual, with an average of 700-fold. The number of CIK cells peaked at day 14 and then slightly decreased by day 28. All CIK cells administered met the following criteria: The percentages of CD3⁺ and CD8⁺ cells exceeded 70 and 40 %, respectively, and CD3⁺CD56⁺ cells were no less than 30 %. The cytotoxic activity of the CIK cells was highest at day 14 (70.5 ± 5.2 %). These results are similar to those in our previously published articles [24, 26]. The final cell products were assessed for viability by the dye-exclusion test and checked twice for possible contamination by bacteria, fungi, and endotoxins. At least 1 × 10⁹ CIK cells were harvested and transfused into patients within 1 h every second day since day 14 to day 22 for 5 times.

Three patients (1 in the immunotherapy group and 2 in the control group) were not followed up, and they were censored at the time of their last visit. Four patients in the immunotherapy group received only 2 cycles of CIK-cell therapy because the recurrence was diagnosed within 6 months after adjuvant chemotherapy. These patients were not excluded from the study in order to keep balance between the 2 groups.

The median follow-up period was 50.5 months (range, 18–82 months). By the end of follow-up, 137 patients (90.7 %) had died, and 143 patients (94.7 %) had been diagnosed with recurrence. The causes of death and sites of recurrence are shown in Table 2. Of the patients with recurrence, 54.5 % had a hematogenous recurrence which was most frequent, 27.3 % had a peritoneal recurrence, 19.6 % had a lymphatic recurrence, 18.2 % had a locoregional recurrence, and 19.6 % had multiple forms of recurrence. The hematogenous recurrence rate of the immunotherapy group was 50.0 %, which was moderately lower than that of the control group (58.7 %); however, the difference between the groups was not significant. Tumor-related deaths accounted for 78.5 % of the deaths in the immunotherapy group and 88.9 % of the deaths in the control group. Univariate analysis showed that the following factors were associated with improved OS: lower grade (P = 0.000), smaller tumor size (P = 0.007), younger age (P = 0.000), lower tumor location (P = 0.007), earlier tumor UICC stage (P = 0.000), and partial gastrectomy (P = 0.001). In contrast, sex, chemotherapy regimen, and performance status did not significantly influence OS.

OS and DFS curves between the immunotherapy and control groups are presented in Figs. 1 and 2, respectively. The patients in immunotherapy group had a borderline significantly and significantly better OS and DFS than the patients in control group (log-rank, P = 0.071 for OS and P = 0.044 for DFS), respectively. The 3- and 5-year OS rates were 54.5 and 23.4 % in the control group versus 67.6 and 32.4 % in the immunotherapy group. The median OS durations were 42.1 months in the control group and 48.1 months in the immunotherapy group. In contrast, the 3- and 5-year DFS rates were 36.4 and 10.4 % in the control group versus 47.3 and 28.3 % in the immunotherapy group. The median DFS durations were 34.1 months in the control group and 40.4 months in the immunotherapy group. Furthermore, compared with the patients in the control group, the patients in immunotherapy group had borderline and significantly reduced risk of overall death and recurrence (HR, 0.78; 95 % CI, 0.53–1.04 for overall death and HR, 0.72; 95 % CI, 0.52–0.99 for recurrence), respectively.

A retrospective subgroup analysis of patients by Lauren’s histological type showed that patients with diffuse or mixed-type tumors seemed not to benefit from adjuvant immunotherapy. The OS and DFS were not significantly different between the 2 groups (log-rank, P = 0.970 for OS and P = 0.962 for DFS, Fig. 3). Within this subgroup, the 3- and 5-year OS rates were 38.5 and 7.7 % in the control group versus 40.7 and 7.4 % in the immunotherapy group. The 3- and 5-year DFS rates were 11.5 and 0.0 % in the control group and 14.8 and 3.7 % in the immunotherapy group. In addition, there was no significant difference in risk of overall death and recurrence between the control group and the immunotherapy group (HR, 1.01; 95 % CI, 0.57–1.74 for overall death and HR, 0.98; 95 % CI, 0.57–1.69 for recurrence). However, for the patients with intestinal-type tumors, not only OS but also DFS was significantly better in the immunotherapy group than in the control group (log-rank, P = 0.045 for OS and P = 0.023 for DFS, respectively, Fig. 4). The 3- and 5-year OS rates were 62.7 and 31.4 % in the control group versus 87.2 and 46.8 % in the immunotherapy group. The 3- and 5-year DFS rates were 49.0 and 15.7 % in the control group versus 66.0 and 42.4 % in the immunotherapy group. The patients in immunotherapy group had significantly reduced risk in overall death and recurrence compared with the patients in the control group (HR, 0.65; 95 % CI, 0.43–1.00 for overall death and HR, 0.62; 95 % CI, 0.41–0.95 for recurrence).

We evaluated the cellular immune response in the immunotherapy group by analyzing changes in lymphocyte subsets between baseline (2 weeks after adjuvant chemotherapy) and various points during immunotherapy shown in Table 3. At baseline, the percentages of CD3⁺, CD4⁺ cells, and the CD4⁺/CD8⁺ ratio were 50.8 ± 8.5 %, 26.5 ± 6.1 %, and 0.9 ± 0.4, respectively. One week after the end of the first CIK-cell treatment cycle, the percentages of CD3⁺, CD4⁺ cells, and the CD4⁺/CD8⁺ ratio were significantly higher (62.6 ± 11.3 %, 36.0 ± 6.6 %, and 1.4 ± 0.3, respectively). By 2 months after the end of the first treatment cycle, the values declined to near baseline values (51.8 ± 9.2 %, 28.0 ± 7.6 %, and 1.0 ± 0.2 %, respectively). Two months after the end of the third cycle of CIK-cell treatment, values increased to 60.7 ± 8.2 %, 34.2 ± 7.1 %, and 1.3 ± 0.3, respectively, which was significant higher than the baseline. Additionally, the changes of immunologic data in the different subgroups were compared according to Lauren’s histological type. Compared to the patients with intestinal-type cancer, similar cellular immune responses were observed in the patients with diffuse-type and mixed-type cancer.

A total of 351 cycles of CIK-cell therapy were performed. Two patients received 2 cycles, 18 patients received 3 cycles, 32 patients received 4–5 cycles, 22 patients received more than 6 cycles, and the rest of 77 patients were in the control group. During and after each of the 351 cycles of CIK-cell therapy that were administered, side effects, including fever, chills, headache, rash, nausea and vomiting, diarrhea, shock, abnormalities of routine blood test, hepatic dysfunction, and renal dysfunction, were recorded. The most common side effects were fever (20.8 %), chills (15 %), headache (10 %), rash (5 %), and nausea and vomiting (5 %). There were no instances of diarrhea, shock, abnormalities of routine blood test, hepatic dysfunction, or renal dysfunction (Table 4). All of the side effects were resolved without intervention within 24 h or were treated successfully by simple allopathy, such as anti-inflammatory treatment, anti-allergy treatment, and antiemetic treatment.