Harnessing TCR-Based Immunotherapy Approaches for Pancreatic NETs

In a discussion with CancerNetwork®, Mauro Cives, MD, outlined his Investigator Award-winning research project funded by the Neuroendocrine Tumor Research Foundation (NETRF). Collaborating with Anguraj Sadanandam, PhD, FRSB, at the Institute of Cancer Research (ICR), Cives is spearheading an effort to transition pancreatic neuroendocrine tumors (NETs) from "immunologically cold" targets to viable candidates for precision cellular immunotherapy.

The cornerstone of this research lies in the discovery that liver metastases in patients with pancreatic NETs harbor specific tumor-infiltrating lymphocytes (TILs). These cells exhibit a memory progenitor stem-like phenotype—which are notably double negative for CD39 and CD69—suggesting potent, long-term antitumor activity. Moreover, Cives found that these reactive T cells are not scattered randomly but concentrated within specialized "immune niches," indicating a highly organized immune response within the tumor microenvironment.

Cives outlined the project’s next phase, which will involve leveraging these biological insights to design TCR-based cellular immunotherapies. Unlike traditional CAR T-cell therapies that are limited to surface proteins, TCR-based approaches can recognize intracellular peptides presented by HLA molecules, significantly expanding the library of targetable antigens to include oncogenic drivers and transcription factors.

Looking ahead, Cives speculated a shift in the pancreatic NET treatment algorithm. By positioning TCR therapies after targeted agents and before chemotherapy, clinicians may avoid the immune dysfunction often associated with alkylating agents like temozolomide (Temodar). While the spatial logic of immune niches remains a mystery, he expressed a hope that future research with the NIH will help “shed light” on this phenomenon.

Cives is an associate professor of Medical Oncology in the Department of Interdisciplinary Medicine at University of Bari Aldo Moro.

CancerNetwork: Could you outline the research you conducted that resulted in your receipt of the NETRF grant?

Cives: First, let me acknowledge NETRF because [they are] doing an amazing job in funding NET research. All the [work] that we did in the past was funded by NETRF, so they were key in this success. The current research project was submitted by me and professor Anguraj Sadanandam, PhD, FSRB, and [his team] at the ICR in London. It lies on some fundamental answers that we did to our former research questions.

Our former research question was, “Can we isolate tumor infiltrating lymphocytes from pancreatic NET liver metastases?” The answer to this research question was yes. The following question was, “Are these numbers clinically meaningful?” The answer was, again, yes. Then, we went on characterizing the tumor-infiltrating lymphocytes coming out from the pancreatic NET liver metastasis.

Quite surprisingly, we found that most of them have a memory progenitor stem-like phenotypes––they are double negative for CD39 and CD69. This is something that is well known to predict antitumor activity of these T cells when injected back into patients. Moreover, we demonstrated that these tumor-infiltrating lymphocytes were able to recognize and attack autologous tumors and that antitumor reactivities were especially restricted to immune niches rather than dispersed throughout the tumors. What I’m saying here is that not all tumor-infiltrating lymphocytes within pancreatic liver metastases are created equally, but a fraction of them is able to exploit antitumor activity against the tumor cells.

How do you plan to apply the grant to the next steps of your research?

Now that we have demonstrated that tumor-infiltrating lymphocytes are able to kill intervention tumor cells, we plan to elucidate the TCR-specificity of tumor-infiltrating lymphocytes and harness this information to design novel TCR-based cellular immunotherapies. We aim to generate an atlas of TCR antigen pairs, functionally validate the predicted TCR antigen interaction, and assess its special relevance based on what I [previously] mentioned. Then, we would like to clone the TCRs with the antitumor activities that were capable of NET antigen recognition.

What criteria are most critical for identifying shared intracellular antigens that offer high tumor specificity while minimizing the risk of adverse effects in healthy endocrine tissues?

These are key questions in the immune oncology field, and multiple adoptive cellular therapies failed because of this problem. We are trying to narrow, as much as we can, the TCR-specificity of the T cells so that we can avoid off-target effects against normal endocrine and non-endocrine tissue. To do so, we are focusing on both canonical and non-canonical tumor neoantigens. We are investigating neoantigens derived from RNA––editing and alternative splicing––but we are also looking at cancer-testis antigens and some lineage-specific antigens where we know that hormone replacement therapy might be possible.

The bottom line is that we would first like to have absolutely specific, not restricted, antigens—like neoantigens, cancer-testis antigens, or lineage-specific antigens—provided that hormone replacement is possible. To do so, we are trying to be stringent in defining the antigen landscape of these malignancies. We are employing a combination of methods, including immunopeptidomes and TCR-sequencing, to have an in-silico prediction of the neoantigen landscape…By applying this concept, we are well-positioned to depict the antigen landscape of pancreatic NETs.

Could you elaborate on why the TCR-based approach, specifically its ability to recognize MHC-presented intracellular peptides, is uniquely positioned to succeed in pancreatic NETs?

As a matter of a background, pancreatic NETs were once considered immunologically cold tumors. Now, we know that this is not completely true, and that there is a lot of heterogeneity, particularly in a subset of pancreatic NETs that exhibit T-cell infiltration. Quite importantly, T cells that infiltrate these tumors are also exhausted. This means that they were primed by antigens. Moreover, we now know that HLA1 is widely expressed by pancreatic NETs, and that there is also expression of the antigen-processing machinery.

What I’m saying is that these neoplasms have all the features and characteristics needed to effectively present antigens to the T-cell counterparts. Now, many of the most biologically relevant tumor drivers and most lineage-defining proteins in pancreatic NETs are intracellular. This makes them inaccessible to surface-targeting modalities; for example, CAR T-cell therapies, monoclonal antibodies, bispecific antibodies, and so on.

A TCR-based approach is uniquely positioned to succeed because T-cell receptors may recognize peptides that are derived from intracellular proteins that are processed and presented on HLA molecules. This has the potential to dramatically expand the universe of targetable antigens beyond surface proteins to include oncogenic drivers, transcription factors, and tumor-specific neoantigens.

Could there be a synergistic rationale for using low-dose radiation to induce immunogenic cell death prior to TCR T-cell infusion?

We should take into consideration 2 main aspects. Aspect no 1, palliative radiotherapy [PRT] can lead to immunogenic cell death, leading to antigen release, and this can stimulate the immune system. If you think about a sequence of PRT and then TCR-based therapies, there is a strong rationale from this standpoint.

At the same time, we should always remember that the target of peptide receptor radionuclide therapy is somatostatin receptors, and that T cells express somatostatin receptors. Indeed, lymphopenia is one of the most common adverse [effects] of PRT. From one side, there is the potential for PRT to induce immunogenic cell death, but on the other side, there is the potential of PRT in decreasing the number of T cells, perhaps decreasing their antitumor activity.

The question is intriguing. Data are needed to test this option. We have a few preliminary data testing a similar concept, like having chemotherapy first and then TCR-based therapies. What we have seen in this context, at least in vitro, is that chemotherapy with the alkylating agents [like] temozolomide can reduce the antitumor activity of T cells.

What liquid biopsy markers are you prioritizing to monitor the expansion and contraction of the engineered T-cell population in real time?

We are using a combination of established and innovative markers. Of course, we are using classical cytokines—including IFN-γ, IL-2, and TNF-α—to see if there is an expansion of the T-cell compartment upon recognition of the tumor cells. At the same time, there is emerging evidence that neoantigen-reactive T cells may be found in the bloodstream, and we are trying to monitor [that]. We don’t have mature data in this regard, but it will be intriguing to see if there is an expansion of the bloodstream compartment of neoantigen-reactive T cells in patients who respond to these kinds of treatments.

If the primary hurdle of T-cell exhaustion is solved, where do you see TCR-based therapies sitting in the pancreatic NET treatment algorithm?

It’s [too] early to respond to this question, personally. I’m just speculating on this point. It might be plausible to see TCR-based therapies [given] after approved therapies, possibly before chemotherapy, to avoid the chemotherapy-induced immune dysfunction. In the treatment of pancreatic NETs, and NETs in general, you may have somatostatin analogs; [radioligands]; and targeted agents including everolimus [Afinitor], sunitinib [Sutent], and cabozantinib [Cabometyx]. Most of these therapies do not induce immune dysfunction. TCR-based therapies might be used after these therapies, but chemotherapy may at least theoretically induce immune dysfunction. We have some very preliminary data at this [time]. If possible, I would use TCR-based therapies before chemotherapy.

Is there anything else that you would like to highlight?

There is an aspect [of this field] that doesn’t allow me to sleep at night; this regards the special distribution of tumor-reactive T cells. In essence, what we have demonstrated so far is that tumor-reactive T cells are not dispersed throughout the tumor. They are, for some reason, compartmentalized to what some call “immune niches.” But what we do not understand as of today is, “Why? Why [are] these T cells located in these immune niches, and what are the molecular clues? What are the molecular determinants that make T cells stay there and not go somewhere else?” We’re trying to address these questions through spatial profiling, and we are collaborating with Jaydira del Rivero, MD, at the NIH in Bethesda, MD. Experiments are ongoing, and we hope to shed some light on this topic.