2025 Archives | Alliance for Cancer Gene Therapy

Renier Brentjens, MD, PhD, helped create the first chimeric antigen receptor (CAR) T cells to cure patients with blood cancer. Now, as an Alliance for Cancer Gene Therapy (ACGT) Research Fellow, he’s tackling solid tumors and engineering next-generation CAR T cells that could potentially treat all cancers.

As a young postdoctoral researcher at Memorial Sloan Kettering Cancer Center in the early 2000s, Dr. Brentjens worked on one of the very first ACGT grants in the lab of Michel Sadelain, MD, PhD. Ultimately, the research led to the launch of the first-in-human trials of CD19-targeting CAR T cells for leukemia and the first cell therapy approvals. The incredibly rapid and complete remissions in children and adults were the first clinical sparks that lit the cell therapy revolution in oncology.

And yet, more than two decades later, almost all solid cancers still lack satisfactory cell therapy solutions.

Closing that gap is the challenge that now occupies Brentjens, who currently serves as Deputy Director, Chair of the Department of Medicine, and The Katherine Anne Gioia Endowed Chair in Cancer Medicine at the Roswell Park Comprehensive Cancer Center in Buffalo, NY.

To describe his approach, he referenced a memory from years ago on an airport tarmac in London. Admiring the Rolls-Royce logo on the jet engines, a friend and colleague remarked how unlikely it was that all four would give out in the air. Jet engines are designed to the most exacting standards, so that their function persists even in extreme circumstances, because the entire plane depends on them. Rolls-Royce doesn’t make the wings, or the wheels, or the seats. It only focuses on the engines, but it makes them better than almost anyone.

Brentjens thinks about cancer the same way. For him, everything begins with the engine— the CAR T cell itself—and with building one powerful and reliable enough to work where others have failed. And with his ACGT Investigator Award in Cell and Gene Therapy for Gynecological Cancer Research, he has big ambitions.

“We want to be the Rolls-Royce of CAR T-cell engines,” he declared. “To make an engine that won’t fail.”

Why current CAR T-cell engines still stall

The first barrier to designing effective CAR T cells for solid cancers, Brentjens explains, is the tumor environment itself. Solid tumors contain significant numbers of non-tumor cells; the supporting cast the cancer recruits and manipulates to protect itself. These include immunosuppressive immune cells that shut down an attack and fibroblasts that give the tumor structure and help it spread. Even when the immune system recognizes the cancer, that hostile tumor environment defuses its response. Blood cancer cells, circulating and exposed, have no such fortifications, so the CAR T cells that target them were never designed for this tougher terrain.

The second barrier is that solid tumor cells are not uniform. They mutate unevenly, forming a heterogeneous population in which cells on one side of a tumor may look nothing like those on the other. Engineer a CAR T cell against a single target, and it could clear every cancer cell that carries that target, but the cells that don’t carry that target will simply escape detection and regrow the tumor. This phenomenon, known as antigen escape, is a central obstacle for CAR T cells in solid tumors.

An effective CAR T-cell engine must drive a response that overcomes both problems: the hostile tumor environment and the antigen escape route.

The cytokine that changes the fight

The main enhancement that sets Brentjens’s engine apart is a single, potent addition: his CAR T cells are engineered to secrete a powerful pro-inflammatory cytokine called interleukin-18 (IL-18).

When Brentjens’ CAR T cells reach the cancer, the IL-18 they release reprograms the tumor environment around them, flipping suppressive macrophages into inflammatory ones, keeping the CAR T cells from being shut down, and, as he puts it, “exposing the tumor to the full force of the immune system.” The patient’s own immune cells previously held in check reactivate and join the assault, and the engineered CAR T cell activates a far larger immune response.

That recruitment of the immune system is also the key to the antigen escape problem. Once the person’s native immune system is drawn in, it begins recognizing other targets on the tumor in addition to the original CAR T-cell target. This phenomenon is called antigen spreading, and it allows the immune response to reach the cancer cells that a single-target therapy would have missed.

However, none of this engineering will work unless the CAR T cells reach the tumor and know what to strike. That is the job of the CAR itself, which Brentjens’ team aims at MUC16, a protein overexpressed on most ovarian tumors and rarely found on healthy tissue. MUC16 is also tied to the disease’s ability to spread, making it not just a risk marker but also a potential therapeutic vulnerability.

These two modifications work together to target cancer cells effectively. The anti-MUC16 receptor tells the engine where to drive while the IL-18 gives it the power to win the race.

Brentjens’ CAR T cells are manufactured basically the same way as the CAR T-cell therapies already approved for blood cancers: from a patient’s own T cells that are harvested, enhanced, and then reinfused. But because he will be treating patients with ovarian cancer, his therapeutic plan is to deliver the CAR T cells both intravenously and directly into the peritoneal cavity, the abdominal space where ovarian tumors grow and spread. With ACGT support, Brentjens’s team is advancing the therapy from the lab into its first trial in patients with recurrent, platinum-resistant ovarian cancer.

The support from ACGT, in Brentjens’ view, not only helps pay for the trial, but also provides funds for the costly follow-up analyses that reveal not just whether the therapy works, but why.

“If it works, we need to know why it works,” he said. “And more importantly, if it doesn’t work, we need to know why it doesn’t work.”

A trial measured on survival alone might show a small benefit and leave no clue how to build on it. That deeper analysis, Brentjens argues, is how a field actually makes leaps forward.

Questioning assumptions

The desire to dig deeper for insights and improvements extends to one of the field’s most entrenched assumptions. Conventional CAR T-cell therapy requires lymphodepletion, a harsh round of chemotherapy that clears out a patient’s existing T cells before the engineered CAR T cells are infused. The field treats it as essential. Brentjens is not so sure.

For traditional CAR T cells, especially in blood cancer, he says you should not skip it. But CAR T cells that secrete IL-18 are built to attract and activate the patient’s own immune cells—and lymphodepletion “wipes out exactly the cells that you’re trying to recruit”. Clear the field first, and you may be committing friendly fire against waiting reinforcements, forfeiting the antigen-spreading approach the therapy aims to elicit. Furthermore, if lymphodepletion could be avoided, it might be possible to administer CAR T-cell therapy in an outpatient setting rather than an ICU.

Brentjens holds both ideas at once: that lymphodepletion can be genuinely valuable, but also that it may be counterproductive in the very cases his CAR T cell is designed to treat.

That tension—lymphodepletion as both ally and obstacle—points to something larger about how Brentjens sees the field’s future. The answer, he suspects, won’t be a single therapy that beats cancer, but a decision made, to some extent, tumor by tumor and patient by patient. A biopsy that tells the lab which targets to pursue and how to tune the engine to drive responses in a particular tumor, and whether pretreatment conditioning might be important. Different cancers, and different patients, will demand different CAR T-cell engines and therapeutic strategies.

Every one of them, he insists, must be built for performance first. Only once a CAR T-cell engine wins reliably does it make sense to add noncritical features, like being cheaper and faster to manufacture and simpler to administer. Cell therapy isn’t there yet, Brentjens believes. As long as these treatments still can’t reliably cure patients with solid tumors, the field can’t afford to optimize cost and convenience over therapeutic power.

“We need to focus on CAR T-cell engines that meet F1 performance standards,” he said. “Once we’ve succeeded there, we can make more consumer-friendly versions.”

Ovarian cancer remains the deadliest of the gynecologic cancers, and three obstacles account for much of that lethality. Tumors surround themselves with a dense, fibrous shield that blocks immune cells from entering. Not every tumor cell carries the same markers, so cells that lack the one being targeted survive treatment and regrow the disease. And the environment inside the tumor wears down whatever defenders manage to get in. CAR (chimeric antigen receptor) T-cell therapy, which reprograms a patient’s own T cells to recognize and kill cancer cells, has transformed the treatment of blood cancers, but these barriers have kept it from delivering comparable results here.

Our institution has tested that proposition directly. We have run CAR T cell trials targeting mesothelin and folate receptor-alpha (FRα), two proteins found broadly on ovarian tumors, and while both proved safe, neither produced lasting benefit. Few of the infused cells reached the tumor, and those that did failed to settle in as tissue-resident memory cells, the long-lived form T cells must adopt to persist inside tissue. The problem was not the target. It was the terrain, and how long our cells could hold it.

Our recent work points toward a solution with three coordinated parts. We combined mesothelin CAR T cells with VCN-01, a tumor-killing virus engineered to carry an enzyme that digests the fibrous shield. In the first patients to receive it, the combination produced a partial response, but the CAR T cells still did not persist well, and still could not reach every tumor cell. Separately, we found that briefly conditioning CAR T cells with a signaling protein called TGF-beta pushes them toward that resident state while preserving their killing power. We also developed a bispecific antibody: a two-armed molecule that grabs a cancer cell with one arm and a T cell with the other, forcing them together. Ours targets mesothelin while ignoring the decoy fragments ovarian tumors shed to misdirect an attack, and the CAR T cells can manufacture it themselves.

This proposal tests whether those pieces work better together than apart. Our central idea is that VCN-01 does more than open a door: by clearing the shield and releasing signals that draw immune cells in, it creates local conditions where FRα-targeted CAR T cells can take up residence and stay. Aim 1 will determine whether the virus creates those conditions, and measure how long CAR T cells persist inside tumors after priming with it. Aim 2 will ask whether the bispecific antibody closes the escape route left open by targeting one marker alone, recruiting the patient’s own T cells against tumor cells that lack FRα without exhausting the engineered cells.

The novelty of this work lies in a single framework that prepares the ground, programs T cells to hold it, and amplifies local immunity to close the escape routes that limited earlier CAR T therapies. Because several of these components have already been tested in patients at our center, this study is positioned to define the dose, route, and timing needed for a first-in-human trial, and to reveal more broadly how a tumor’s structure, signals, and shifting markers govern whether immunotherapy succeeds, in ovarian cancer and beyond.

ACGT – Edward Netter Memorial Investigator Award in Cell and Gene Therapy for Pancreatic Cancer Research

Pancreatic cancer is a devastating disease with a dismal prognosis. Our lab develops treatments for difficult-to-treat cancers, such as pancreatic cancer. The main treatment we focus on is CAR (chimeric antigen receptor) T cell therapy. CAR T cells are made by taking a patient’s immune cells, reprogramming them to target the patient’s cancer, and putting them back into the patient. The reprogramming informs the T cells how to distinguish tumor cells from normal cells through markers specifically expressed by the tumor cells.

This therapy has worked well for blood cancers but has been less effective against solid tumors, like pancreatic cancer. Solid tumors are more difficult to treat with CAR T cells because they are less accessible by immune cells, which are usually injected into the blood, and they don’t survive as well once they get into the tumor. The tumor cells can also decrease the expression of the tumor-specific marker, causing them to go unrecognized by the CAR T cells and, therefore, evade killing.

To overcome these challenges, we have used a high-throughput screening approach to identify additional changes we can induce in CAR T cells to make them more effective in solid tumors. We created a mixture of CAR T cells, all targeting the same marker but with one additional, unique change per cell. This mixture was then injected into mice with pancreatic tumors, and after 4 weeks, the CAR T cells making it into the tumor and surviving during that time were identified, representing CAR T cells with changes that enhanced their presence in the tumor over time. Although these changes improved tumor killing, the tumors eventually returned, demonstrating room for further improvement.

In this proposal, we will make CAR T cells with combinations of two of the changes we identified to improve function, to see if combining them will make them more effective and control tumor growth for longer. We will also repeat our screening approach using tumor cells with different levels of the tumor-specific marker to identify changes to the CAR T cells that make them more effective when there are lower maker levels. Overall, this work will provide valuable information on how to adjust CAR T cell therapy for pancreatic tumors to ultimately make them an effective treatment for pancreatic cancer.

ACGT – Edward Netter Memorial Investigator Award in Cell and Gene Therapy Research for Pancreatic Cancer

Cell-based therapies for cancer have been highly successful in blood cancers such as leukemia and lymphoma. These therapies take the form of chimeric antigen receptor T cells (CAR T cells), which recognize and kill cancer cells. However, translating these therapeutic successes to cancers of internal organs – tumors of the lung, breast, colon, etc. – has been challenging. A major reason for this difficulty is that these “solid tumors” are protected by layers of protein and cells that shield the cancer from immune attack. This barrier to anti-tumor immunity is called the tumor microenvironment (TME), and pancreatic cancer represents the most extreme case of this impermeability. Our proposal seeks to disrupt this obstacle so that CAR T cells may gain entry to the tumor and exert their anti-cancer activity.

Our strategy contains two approaches. The first involves KRAS – a cancer-causing mutant protein that is the source of over 90% of pancreatic cancers. Recent advances in drug development have resulted in several chemical inhibitors of KRAS with dramatic anti-cancer effects in animal models and patients (leading already to FDA approvals in certain cancers). KRAS inhibitors exert most of their activity by killing cancer cells. However, a fortuitous side benefit is that these drugs also reshape the TME to allow immune cells (especially T cells) to infiltrate tumors. Our preliminary data in mice demonstrate that treatment with a KRAS inhibitor allows more T cells (including CAR T cells) to enter the tumor, resulting in greater anti-cancer activity. Our second approach involves the use of CAR T cells directed against a component of the tumor barrier – fibroblast activation protein (FAP). We have shown that CAR T targeting FAP can force their way into tumors, bringing other T cells (including other CAR T cells) along with them. As a result, combinations of anti-FAP CAR T cells with either chemotherapy or cancer-targeted CAR T leads to greater anti-tumor effects.

Our proposal includes studies in both animals and patients. Aim 1 extends our work on KRAS inhibitors and FAP CAR T cells in mouse models to understand how these interventions reshape the TME. We will also determine whether combining the two therapies can have additive effects, leading to cures (as suggested by our preliminary studies). Aims 2 and 3 focus on patients. Aim 2 builds on an ongoing clinical trial in pancreatic cancer of CAR T cells against mesothelin. Our goal is to determine whether pretreatment with a KRAS inhibitor before CAR T infusion results in a greater number of CAR T cells inside the tumor. Aim 3 involves a clinical trial of FAP CAR T cells scheduled for 2025 in which we will assess whether these CAR T cells penetrate patient tumors (as they do in our animal models) and reshape the TME in a favorable manner.

Collectively, these studies aim to break down the physical and immunological barriers limiting the efficacy of CAR T cell therapy for pancreatic cancer.

ACGT – Edward Netter Memorial Investigator Award in Cell and Gene Therapy for Pancreatic Cancer Research

Patients with pancreatic cancer (PDA) generally present with advanced disease, and standard treatment regimens have provided limited benefit in this setting. Immunotherapy has proven to be a promising new therapy for many malignancies, but yielded only marginal benefit thus far in PDA. We have pursued a strategy for engineering immune cells to be able to recognize tumors, and then administering large numbers of these cells to treat cancers. We demonstrated in a mouse model that CD8 T cells engineered with a tumor-specific T cell receptor (TCR) targeting the tumor antigen Mesothelin can infiltrate pancreatic tumors and mediate therapeutic anti-tumor activity.

This led to a clinical trial in which patients with metastatic PDA were treated with their own CD8 T cells that we engineered ex vivo with a human MesotheIin-specific TCR. Biopsies were obtained after T cell infusions, infiltrating T cells isolated, and the T cells and tumor cells analyzed in depth to elucidate obstacles to efficacy. In both the mouse model and clinical trial, the infused CD8 T cells by day 21 had acquired in the tumor characteristics of exhaustion, becoming dysfunctional and failing to expand at the tumor site. Therefore, we have been developing synthetic strategies for further engineering of T cells to enhance activity for mediating more sustained responses.

For this trial, we have isolated a human TCR specific for mutated KRAS, which in the vast majority of PDA cases is an obligate driver of the cancer, making it difficult for the tumor to evade responses by losing the antigen. To better sustain T cell responses, we are introducing the TCR plus CD8 genes, which improve TCR binding to its target, into both CD4 and CD8 T cells to create a coordinated T cell response in which both CD4 and CD8 T cells can recognize and respond to the same and adjacent tumor cells. These genes can now create functional CD4 T cells, which has been shown in multiple models to not only promote CD8 T cell function, proliferation, and survival, but also delay or prevent exhaustion. Therefore, in this clinical trial for advanced PDA we will treat patients with their own CD4 and CD8 T cells engineered to express both a TCR specific for mutant KRAS and the CD8 genes.

Patients will be biopsied before and after T cell infusions, as our prior experiences highlighted the value of in-depth analysis of the tumor and infiltrating T cells to elucidate reasons for success or failure for building next generation strategies. The high dimensional data will be used to engineer T cells that can overcome encountered obstacles impeding tumor eradication and will be validated in preclinical PDA models. We have already generated synthetic proteins that can convert inhibitory or death signals into costimulatory and survival signals, or suppressive signals into proliferative signals, and will prioritize advancing to next generation trials the synthetic strategy(s) most effective for overcoming observed obstacles.