Research Into 'Cold' Tumors Heating Up in Pancreatic Cancer

Get Permission

Immunotherapy has changed the treatment paradigm for cancer, inducing durable responses in a subset of patients with previously refractory disease. However, current approaches are successful in only approximately 20% of cancers (so-called hot tumors). For the nearly 80% of cancers that are “cold” with weakly active immunity and few intratumoral T cells, immunotherapy has been largely ineffective.

Vinod P. Balachandran, MD

Vinod P. Balachandran, MD

According to Vinod P. Balachandran, MD, of Memorial Sloan Kettering Cancer Center, New York, pancreatic cancer is a model “cold” tumor that could inform immunotherapeutic principles needed for other treatment-resistant cancers. At the 2020 American Association for Cancer Research (AACR) Virtual Special Conference: Pancreatic Cancer, Dr. Balachandran discussed two potential novel immunotherapeutic approaches to activate immunity in pancreatic cancer.1

“Cold tumors pose a critical challenge to immuno-oncology to expand its scope beyond hot tumors,” said Dr. Balachandran. “Neoantigen delivery and harnessing innate lymphoid cells could potentially inform new approaches, not only for pancreatic cancer but also for cold tumors at large.”

Effective Therapy Sorely Needed

As Dr. Balachandran explained, “hot” tumors that are responsive to immunotherapy have strongly preactivated antitumor immunity reflected by many intratumoral T cells. Current immunotherapies leverage this immunity by directly targeting antitumor T cells using a variety of different strategies, including cytokines, modified T-cell receptors, or blocking inhibitory immune checkpoints.

In humans as well as in mice, however, pancreatic ductal adenocarcinoma is “very cold, very resistant, and very deadly,” said Dr. Balachandran, with one of the lowest response rates to current immunotherapies. Current chemotherapy, radiation therapy, and targeted therapies are also largely unsuccessful, leading to a dismal 5-year survival rate of 9%.

“Pancreatic ductal adenocarcinoma is an ideal cold tumor because it is really desperate for an effective therapy,” said Dr. Balachandran. “Perhaps, immunotherapy is its most promising option.”

Neoantigens: T-Cell Targets in Cold Tumors

In the rare patients with pancreatic ductal adenocarcinoma who are long-term survivors, Dr. Balachandran and colleagues have found that the density of CD8-positive T cells in tumors is among the strongest factors to correlate with long-term survival. Based on data in humans as well as mice, the researchers have focused on neoantigens as a possible target of CD8-positive T cells to devise rational strategies to harness them therapeutically.

Using a cohort of long-term survivors and a stage- and treatment-matched cohort of short-term survivors, Dr. Balachandran and colleagues performed a variety of immunoprofiling, computational, and functional assays to enable antigen discovery. They found that the tumors of long-term survivors were “hotter,” ie, enriched in intratumoral T cells, including CD8-positive T cells.2 What’s more, among patients whose tumors had the highest density of neoantigens and CD8-positive T cells, the researchers identified a subgroup of patients with extreme survival, suggesting that some neoantigens were more immunogenic than others.

The researchers then developed a computational model to select these immunogenic neoantigens and potentially identify an immunogenicity hierarchy. This model, which described higher immunogenicity to tumors with high-quality neoantigens, was prognostic of survival in two pancreatic cancer cohorts and by other investigators in glioblastoma (another cold cancer) as well as in immunotherapy-treated hot tumors (melanoma and lung cancer). In addition, the researchers detected that neoantigen-reactive T cells persisted in the blood of pancreatic cancer survivors, suggesting they may induce potent immunologic memory.

“We’ve demonstrated that neoantigens can be T-cell targets in cold tumors and that neoantigen quality is a biomarker of tumor immunogenicity; it identifies non–self-neoantigens and may be used to allow rational neoantigen prioritization,” said Dr. Balachandran. He noted that one implication of the model is that the immunogenicity of clones in primary tumors may impact the evolutionary trajectory of metastases.

Dr. Balachandran and colleagues are now attempting to harness neoantigens as therapies for pancreatic cancer. They are currently enrolling patients with chemotherapy-naive, resectable pancreatic cancer in a phase I study of adjuvant personalized vaccines. Patients will be treated initially with surgery, followed by sequential PD-L1 blockade, personalized neoantigen vaccination, and then modified FOLFIRINOX (leucovorin, fluorouracil, irinotecan, oxaliplatin) chemotherapy.

Harnessing Innate Lymphoid Cells

As Dr. Balachandran explained, in long-term survivors of pancreatic cancer, there are other signals in tumors that activate CD8-positive T cells. Priming CD8-positive T cells to tumor antigens is, in fact, a multistep process. Although cold tumors appear to have global defects in these pathways, he said, tumors in long-term pancreatic cancer survivors do not.

In an exploratory cohort of long- and short-term survivors,
Dr. Balachandran and colleagues found that the tumors of long-term survivors were enriched in both cytotoxic cells and innate lymphoid cells: antigen-independent lymphocytes that are developmentally linked to T and B cells. However, unlike T cells, which recognize antigens and mostly circulate, innate lymphoid cells primarily reside in tissues.

By examining tissues from patients with pancreatic cancer with the presence of innate lymphoid cells, John Alec Moral, MD, now a surgery resident at UC San Diego, identified cells that lacked expression of conventional lineage markers but expressed markers consistent with innate lymphoid cells.3 These innate lymphoid cells were expanded in tumors compared with adjacent organs and were phenotypically group 2 innate lymphoid cells, as they expressed interleukin-33 (IL-33) receptor and GATA3. What’s more, said Dr. Balachandran, the higher density of these group 2 innate lymphoid cells correlated with longer pancreatic cancer survival.

After research in pancreatic cancer mouse models showed that the cytokine IL-33 expands tumor group 2 innate lymphoid cells, Dr. Balachandran and colleagues were able to demonstrate that recombinant IL-33 could be used to activate these cells in tumors to indirectly stimulate immunity.

To further boost this response, the researchers tested PD-1 blockade in pancreatic cancer mouse models. They found that recombinant IL33-activated group 2 innate lymphoid cells may amplify responses to anti–PD-1 in both tumors that are partially sensitive to PD-1 as well as those that are resistant to PD-1.

“We think these group 2 innate lymphoid cells activate CD8-positive T cells in pancreatic cancer, and this effect is indirect and tissue-specific,” said Dr. Balachandran. “We also think these antitumor group 2 innate lymphoid cells can be harnessed through two cooperative strategies: recombinant IL-33 and PD-1 blockade.”

“More broadly, we think this identifies that immune checkpoints not only regulate T lymphocytes in tumors, but also other unique lymphocytes as well,” he added.

According to Dr. Balachandran, co-targeting group 2 innate lymphoid cells and T cells could potentially be a novel immunotherapeutic approach for cold cancers. The researchers are currently working on clinical recombinant IL-33 drug development to take these concepts to patients. 

DISCLOSURE: Dr. Balachandran has received research support from Bristol Myers Squibb and Genentech.


1. Balachandran VP: Dissecting the immune complexity of pancreatic cancer survivors through team science. AACR Virtual Special Conference: Pancreatic Cancer. Presented September 30, 2020.

2. Balachandran VP, Łuksza M, Zhao JN, et al: Identification of unique neoantigen qualities in long-term survivors of pancreatic cancer. Nature 551:512-516, 2017.

3. Moral JA, Leung J, Rojas LA, et al: ILC2s amplify PD-1 blockade by activating tissue-specific cancer immunity. Nature 579:130-135, 2020.