Pancreatic cancer is unique from an immunological perspective. First, intratumoral effector T-cells are rare, in contrast to many other solid tumors for which infiltration of effector T-cells is often prominent. Second, the RAS oncogene drives an inflammatory program that establishes immune privilege in the pancreatic tumor microenvironment (PTME). Third, pancreatic cancer is associated with a massive infiltration of immunosuppressive leukocytes into the tumor microenvironment. Fourth, the development of pancreatic cancer is associated with a strong desmoplastic reaction that consists of multiple cell types, molecular factors, and extracellular matrix.
Immunotherapy is one of these novel strategies that has been under investigation in pancreatic cancer:
Synthetic vaccines: Cancer vaccines aim to stimulate the immune system against tumor cells by generating humoral and/or cellular immune responses. Many forms of cancer vaccines exist but generally they can be divided into synthetic and cellular-based vaccines.
Synthetic vaccines: Synthetic vaccines are typically made from whole protein or peptides that match a pre-determined antigen to induce a T- cell response. Despite multiple large trials targeting MUC1, VEGF, telomerase and gastrin-17, none have shown a meaningful survival benefit. These trials, however, were able to show significant immune responses to the targeted antigens. Some trials using mutated RAS peptide vaccine alone or in combination with gemcitabine have shown clinical benefit. Importantly, this benefit was mainly seen in patients who demonstrated positive immune responses.
Cellular-based vaccines： Cellular-based vaccines use cancer cells (either whole cells or cell lysates) as the source of the antigens, allowing the immune system to utilize multiple antigens rather than a single epitope. Overall the results of these trials are encouraging; however, each included a small number of patients making it difficult for meaningful interpretation.
The efficacy of adoptive T- cell transfer remains to be determined in solid tumors including pancreatic cancer. Kawaoka et al., investigated MUC1-specific cytotoxic T lymphocytes (CTLs) in 28 patients and showed 19 % 3-year survival in patients with resectable disease along with increased effector lymphocytes and decreased T-regs. More recently, the University of Pennsylvania group presented their experience using autologous T- cells modified with a chimeric antigen receptor (CAR) that recognizes mesothelin in pancreatic cancer patients with refractory metastatic disease. Of the 6 patients treated 2 had stable disease with one patient had a decreased PET avidity of hepatic metastases. Overall, the treatment was well tolerated. The carcinoembryonic antigen (CEA) is another attractive target in pancreatic cancer that is currently been explored in clinical trials (NCT01723306, NCT00004178 and NCT01212887). New generations of CAR therapies are under investigation with a focus on increasing their activities and specificities and decreasing their toxicities. The CAR T-cells efficacy could be enhanced by engineering the intracellular domain to contain co-stimulatory molecules such as 41BB and OX40 or combining CAR T-cells therapy with immune modulators such as cyclophosphamide in order to deplete T-regulatory cells (NCT02465983) or immune checkpoint inhibitors such as CTLA-4 and anti-PD1 antibodies.
The first study to investigate the co-stimulatory agonists in pancreatic cancer used CD40 agonist in combination with gemcitabine in 21 patients with locally advanced or metastatic disease. This study demonstrated promising outcomes with an improvement of median overall survival compared to gemcitabine alone and one patient having a complete resolution of his liver metastases. In the area of immune checkpoint inhibitors, both CTLA-4 and PD-L1 inhibitors were investigated in patients with locally advanced or metastatic pancreatic cancer in 2 clinical trials. The clinical outcomes were disappointing, although, only small number of patients were treated on both trials. To date, the only immune checkpoint inhibitor to show activity in pancreatic cancer is MEDI4736 (anti-PD-L1), which showed a PR rate of 8 % in a preliminary analysis of this going trial. While this suggests a response can be achieved with single immunotherapy, it remains dismal and maybe improved by combination therapy.
Despite the ongoing efforts outlined in this review, the prognosis of pancreatic cancer remains dismal. With the recent progress in cancer immunotherapy, there are glimmers of hope in new immune targets with more being identified each year. These advancements are moving from the bench to the bedside at a rapid pace, with the hope of translating into improvements in clinical outcomes.
Kunk P R et al. From bench to bedside a comprehensive review of pancreatic cancer immunotherapy[J]. Journal for ImmunoTherapy of Cancer, 2016, 4(1): 1.
Yadav D et al. The epidemiology of pancreatitis and pancreatic cancer. Gastroenterology. 2013;144(6):1252–1261.