Based on this phase I clinical trial, we arrived to a conclusion that the infusion of NKT cells could be safely consisted of two infusions at Day 1, Day 3 and 28 days designated as a treatment cycle at dose level of 5×109 cells. As no DLT was observed in this study, no MTD was achieved. This dosage could be referred to in the phase II study.
Initial cancer clinical trials using NKT cells as a systemic treatment had documented some adverse events including anemia, fever, lymphopenia, pneumothorax, and elevated glutamic-oxalacetic transaminase, glutamic-pyruvic transaminase and total bilirubin level, which were no serious toxicity (> grade 2)[21, 26]. In this study, the adverse events which were documented were similar to those reported in other clinical trials. The most common adverse events were hematuria, leucocyturia, leukocytopenia, fatigue, fever, hyperbilirubinemia, positive antistreptolysin O and elevated thyroid stimulating hormone. Besides, the less common events were proteinuria, thrombocytopenia, muscular ache, elevated transaminase and creatine kinase. All the adverse events were tolerated and manageable. With the use of nonsteroidal anti-inflammatory drugs, fever was controlled. With the regard of laboratory data, leukocytopenia and thrombocytopenia were too mild to need intervention while abnormal liver function including elevated transaminase and hyperbilirubinemia which was grade 1 or 2 went normal with the use of hepatinica treatment. In this study, some patients underwnet autoimmune response but no severe autoimmune disease which contained positive antistreptolysin O and elevated thyroid stimulating hormone with a normal baseline level, which meant that NKT cells might induce autoimmune response. However, the response was uncorrelated with anti-tumor effect. Meanwhile, elevated creatinine in Patient 3 was attributed to tumor was relived quickly after the surgical operation. No unexpected and severe toxicity was observed.
The efficacy evaluation was limited in phase I study. In previous studies, immunotherapy was demonstrated that could prolonged progression-free survival and overall survival in patients with advanced melanoma, but had worse efficacy in acral and mucosal melanoma patients[27–30]. In this study, most of them had progressed from prior treatment including immunotherapy, chemotherapy or targeted therapy with heavy tumor burden. However, there were three patients with tumor response that achieved partial response, while there were five achieved stable disease. One of them had acral melanoma. Of note, the patients with no liver metastasis and normal LDH might benefit from the infusion of NKT cells even with acral melanoma. 33.3% patients with primary site of cutaneous melanoma experienced clinical benefit and the median progressive-free survival time was 2.2 months even after failure of treatment with immunotherapy consisted of ipilimumab, interferon and interleukin-2. According to previous study, PD-1 expressed not only by tumor-associated T cells but also by other immune cells and its blocking might release NK cells playing a crucial role against tumors[30]. And Natural killer T-cell agonist α-Galactosylceramide combined with PD-1 blockade could reduce tumor development in a preclinical model of colon cancer in prior study[31]. Immune cells associated with anti-PD-1 might synergize to reduce tumor development. However, there was no similar study in melanoma. In this study, the infusion of NKT cells showed potential antitumor response which may be the basis of further phase II clinical trial and combined treatment.
The trend of immune cells was not consistent with the antitumor response. The change in response to NKT cell infusion returned to normal after each 2 cycles of infusion according to previous studies[7, 10] and the results in our study were similar to that. The small sample limited the reliability of analysis, but in Patient 22 who had a PR, positive NKT cell elevation might suggest tumor lysis. The number of NKT cells perhaps could be a predictive factor of antitumor response.
Cytokines have variable regulatory functions in immune responses, including in anti-tumor immunity. Several cytokines were already shown to limit tumor growth and some of them had been applied to solid tumors[32, 33]. So, cytokines are good potential candidates as the biomarkers for predicting response to adoptive cell transfer as cancer treatment. It has been reported that high serum levels of the immunosuppressive cytokines IL-10 in blood are found to be correlated with poor prognosis, cancer stage and disease progression[34] .TNF-α also works as the key player in melanoma aggressiveness[35].
Although, the trends of cytokine profiles in most cases were not consistent with the antitumor response, the dramatically elevated levels of IL-10 and TNF-α were found in the blood of the Patient 19 who shortly progressed after the treatment. Unfortunately, the small sample limited the reliability of analysis and the moderate production of cytokines via immune reconstitution need further exploration.
In summary, having determined the feasibility and safety of infusion of NKT cells, this study showed some signs of antitumor activities and provided a possibility for combined treatment strategy with anti-PD-1 antibody in the further study.