If you’re interested in live presentations about the latest cancer research, all roads lead to the annual meeting of the AACR, the American Association for Cancer Research. Upon your arrival, you enter a teeming mass. You try to get your bearings as attendees rush in every direction—plenary sessions, major symposia, “advances” sessions in various specialties, and much else. You, too, enter the fray, but you worry that taking in everything on offer is as hopeless as drinking all the water flowing through one of Rome’s aqueducts.

The most recent annual meeting, which took place April 8–13 in New Orleans, was no exception. So, in this brief review of the meeting’s proceedings, we will not attempt anything like exhaustive coverage. Instead, we will focus on cancer immunotherapy. Moreover, we will give ourselves time to explore overlooked byways, rather like travelers who deviate from the usual tourist circuit. The idea is to dwell on unique experiences.

In our excursion, we will encounter researchers who hope to fight cancer by harnessing pro-inflammatory cytokines. If these substances could be tamed, that is, made less toxic, they could direct the immune responses to cancer. At the AACR event, one company reported that it is engineering interleukin-2 (IL-2) prodrugs that may be conditionally activated to minimize off-target effects. Another company reported that it is testing a chimeric PD-1 antibody/IL-2 fusion protein that recruits killer immune cells.

Other researchers at the AACR event indicated that they are exploring alternative approaches to cancer immunotherapy. Some are focusing on small-molecule inhibitors for identifying and targeting key kinases that hamper T-cell responses to tumor cells. Some are developing bispecific antibodies that can influence multiple and complementary immune pathways for greater therapeutic effect. Finally, some are engineering chimeric antigen receptor (CAR) T-cell systems that attack solid tumors and overcome the harsh barriers of the tumor microenvironment (TME).

Conditionally activated prodrugs

An attractive cancer immunotherapy approach is the prodrug. For example, a systemically administered prodrug could remain inactive until it reaches the TME, where it becomes activated and launches a powerful antitumor response. This approach sounds straightforward, but it has proven hard to implement. A common problem is that tumor heterogeneity prevents prodrugs from being adequately activated.

To overcome this problem, Werewolf Therapeutics is utilizing Predator, a protein engineering technology, to create conditionally activated molecules called Indukines. The company asserts that its Indukines can enhance tumor-specific killing while minimizing undesirable off-target effects.

tumor-activated immunotherapeutic prodrugs illustration
Werewolf Therapeutics develops tumor-activated immunotherapeutic prodrugs. The company’s prodrugs are called Indukine molecules, and they are designed to unleash pro-inflammatory mechanisms on tumors while preventing unwanted side effects on nontarget tissue. A typical Indukine consists of a wild-type cytokine, an inactivation domain to block activity in the periphery, a tumor-protease-sensitive linker that activates in the tumor microenvironment, and a half-life extension domain.

“We have spent considerable time and effort designing substrates that can be efficiently cleaved by tumor proteases yet remain stable in normal tissues and plasma,” said Cynthia Seidel-Dugan, PhD, chief scientific officer, Werewolf Therapeutics. “We are focusing on pro-inflammatory cytokines, such as IL-2 and IL-12, which can stimulate potent antitumor immunity. They have great potential for treating people afflicted with cancer. However, their clinical application has been limited because of their systemic toxicities and poor pharmaceutical properties.”

According to Seidel-Dugan, Indukine molecules consist of four modules: the wild-type cytokine (for example, IL-2 or IL-12); an inactivation domain to block cytokine activity in the periphery; a tumor-protease-sensitive linker that activates in the TME; and a half-life extension domain. One of the company’s IL-2 Indukine molecules, the WTX-124 prodrug, has shown promise in preclinical studies. It exhibits potent antitumor activity by activating tumor-infiltrating CD8+ and CD4+ T cells and has a better therapeutic window relative to recombinant human IL-2.

Werewolf Therapeutics also sees potential in one of its IL-12 Indukine molecules, the WTX-330 prodrug. Because of its pleiotrophic effects on immune cells, WTX-330 could be a gamechanger in treatment-resistant types of cancer.

The company’s scientists recently demonstrated that WTX-330 can reprogram tumor-infiltrating CD8+ T cells to drive tumor regression in multiple syngeneic tumor models, and that it can expand the therapeutic window (in comparison with recombinant IL-12). WTX-330 has also been shown to have activity in less immunogenic tumor models.

“We are very excited about this unique approach,” Seidel-Dugan declared. “We expect that it could also be applied to additional pro-inflammatory mechanisms.” Both WTX-124 and WTX-330 are expected to enter Phase I trials later this year.

Targeting specific immune cells

Aldesleukin, a form of IL-2 that is made in the laboratory, is used to treat adults with melanoma or renal cell carcinoma. Indeed, systemic therapy with first-generation, high-dose aldesleukin can help the immune system kill cancer cells by increasing the activity and growth of T cells and B cells. However, the drug has several limitations: a low response rate (just 15%); significant toxicity (including capillary leak syndrome); a narrow therapeutic index; a short half-life; and a preferential expansion of regulatory T cells (Tregs).

Some of these limitations are being addressed by researchers at Roche’s Pharma Research and Early Development (pRED) group. For example, they have developed a chimeric antibody-cytokine protein by fusing a high-affinity anti-PD-1 antibody to a modified IL-2 variant (IL-2v). Unlike wild-type IL-2, IL-2v does not bind to IL-2 receptor subunit alpha (IL-2Ra), which is also known as CD25. Consequently, the preferential targeting of Tregs and endothelial cells through CD25 is abolished.

“The nonsignaling component of IL-2R, that is, CD25 or IL-2Ra, is constitutively and preferentially expressed on Tregs over conventional T cells and also on some endothelial cells,” said Pablo Umaña, PhD, head of cancer immunotherapy discovery, pRED. “IL-2 binding to CD25 can lead to preferential Treg targeting and also compound the side effects of systemic IL-2 therapy. To avoid these problem, IL-2v is devoid of CD25 binding while maintaining binding to the signaling chains of the receptor, CD122 (b chain) and CD132 (g-c chain).

“We and others have brought to clinical trials initial IL-2 variants with reduced or no CD25 binding. In our case, a first version of IL-2v was fused to a tumor-stroma-targeting antibody. While we confirmed this resulted in a lack of preferential Treg activation and also better tolerability as compared to aldesleukin, we were not satisfied by the level of clinical efficacy achieved by those earlier IL-2 variants.

“We have now learned that removing CD25 binding also impairs effective expansion of tumor-specific effector T cells, which also transiently express CD25 upon specific T-cell receptor (TCR) activation. To overcome this challenge, we have now made a next-generation IL-2v, ‘PD1-IL-2v.’”

PD1-IL-2v acts via a cis mechanism. That is, it engages PD-1 on the same cell that expresses IL-2R. Notably, PD-1 is also expressed by tumor-specific T cells upon TCR activation. By exploiting cis delivery of IL-2v to PD-1+ T cells, it is possible to redirect IL-2R agonism away from CD25-expressing cells and toward PD-1-expressing cells.

“This is the first PD-1-targeted IL-2Rbg-biased agonist to be developed,” Umaña asserted. “We are excited by its potential to improve over immune checkpoint inhibition alone and over previous IL-2 generations. We are now assessing PD1-IL-2v in a Phase I trial (NCT04303858).”

Small molecules, big effects

Aside from cancer therapeutic strategies that focus primarily on immune checkpoint axes, there are strategies that involve the targeting of kinases that reduce T-cell responses to tumor cells. One such kinase is hematopoietic progenitor kinase 1 (HPK1), a negative regulator of TCR signaling. Active HPK1 phosphorylates a protein in the TCR that flags it for destruction. It also reduces the TCR signaling that is so critical for mounting an immune response to attack tumor cells.

Scientists at RAPT Therapeutics are developing small-molecule inhibitors of HPK1. “We believe that by inhibiting HPK1 with small-molecule compounds, we can enhance T-cell function, thus restoring the ability of the immune system to mount an effective antitumor response,” said George Katibah, PhD, principal investigator, Discovery Biology, RAPT Therapeutics.

RAPT Therapeutics develops small-molecule therapeutics that can modulate the immune responses underlying oncological and inflammatory diseases. For example, RAPT is exploring ways of inhibiting hematopoietic progenitor kinase 1 (HPK1), an enzyme that weakens the immune response against tumor cells by blocking T-cell receptor signaling. The crystal structure of the dimeric HPK1 kinase domain is shown here. (Individual kinase domains of the HPK1 dimer are colored separately.) Familiarity with HPK1 allows RAPT to pursue the structure-based design of HPK1 inhibitors.

The company utilizes a number of approaches to identify drug candidates. Katibah reported, “Key to our rapid discovery of small molecules is our use of structure- and pharmacophore-based drug design strategies, as well as machine learning–assisted [investigation of] structure-activity relationships to improve potency, selectivity, and pharmacokinetic properties. We couple these approaches with early testing in physiologically relevant immune assays to rapidly identify highly selective orally administered small molecules with the desired profiles. We also identify biomarkers that can guide our clinical development and patient selection strategies to increase the probability of clinical success.”

In preclinical studies, the company demonstrated that one of its molecules, HPK1-054, enhanced cytokine production by human and mouse primary T cells beyond that observed with TCR activation alone. Further, in mice, the treatment enhanced the activation of antigen-specific CD8+ T cells and decreased tumor growth not only as a single agent, but also in combination with clinically relevant checkpoint inhibitor antibodies.

“The combination of small-molecule drugs like HPK1 inhibitors with established checkpoint inhibitor antibodies or cellular therapies is a particularly attractive proposition to expand scope and efficacy,” Katibah summarized. “HPK1 inhibitors will be part of an integrated approach to cancer immunotherapy treatment that could include standalone treatment but will most likely be part of combination therapies with existing biologics like checkpoint inhibitors.”

Bispecific antibodies against multiple pathways

For decades, researchers have been working on bispecific antibodies, molecules that show promise as cancer therapeutics. Thanks to recent advances, bispecific antibodies are nearing the bedside. “Extensive preclinical and now early clinical data indicate that targeting multiple and complementary immune pathways may be necessary to elicit robust antitumor immunity and improve patient outcomes,” stated Maria N. Jure-Kunkel, PhD, vice president, head of late-stage oncology translational research, Genmab.

Maria N. Jure-Kunkel
Maria N. Jure-Kunkel, PhD
VP, Head of Late-Stage Oncology Translational Research, Genmab

“Together with our partner, BioNTech, we are developing two new investigational therapeutics, GEN1046/BNT311 and GEN1042/BNT312 with co-stimulatory receptor agonist activity,” she continued. “These investigational bispecific antibodies are designed to address deficient or suboptimal antitumor immunity by the co-targeting of key immune pathways.” The idea is to deploy a single molecule that can act on two different pathways that regulate antitumor immune responses.

“Conditional activation of co-stimulatory receptors means that activation of these receptors is strictly elicited when both targets for the bispecific molecule are engaged, because signaling occurs only when the receptors are cross-linked,” she explained. “The conditionality due to dual targeting will potentially localize the activity where it matters, in the tumor microenvironment and in associated lymph nodes, as observed in preclinical studies, therefore addressing some of the limitations observed with agonistic antibodies in the clinic.”

GEN1046/BNT311 elicits an antitumor immune response by simultaneous and complementary PD-L1 blockade alongside conditional 4-1BB stimulation. The latter belongs to the tumor necrosis family of receptors. 4-1BB is an inducible T-cell co-stimulatory receptor expressed on activated CD4+ and CD8+ T cells as well as activated natural killer cells. Jure-Kunkel elaborated, “Concurrent PD-L1/PD1 blockade and 4-1BB activation results in improved T-cell function and promotes the killing of tumor cells.”

GEN1046 is currently in a Phase I/II clinical study of solid tumors and a Phase II clinical study as a single agent or in combination with other standard-of-care therapies in relapsed/refractory metastatic non-small cell lung cancer.

GEN1042 targets CD40 as well as 4-1BB. Targeting of CD40 enhances both dendritic cells and antigen-dependent T-cell activation. Jure-Kunkel noted, “As per the preliminary data from the ongoing, first-in-human study, GEN1042 showed a favorable safety profile and encouraging early clinical activity in a heavily pretreated, heterogeneous patient population.” The investigational drug is undergoing a Phase I/II clinical study in solid tumors including combination therapy with standard-of-care therapies.

Modular CAR T cells against solid tumors

Chimeric antigen receptor (CAR) T-cell therapies have led to durable remissions in chemotherapy-refractory B-cell cancers. However, their use against solid tumors has been limited due to many challenges including the immunosuppresive influence of the TME. These challenges are being addressed by Autolus Therapeutics, a company that is developing CAR T-cell therapeutics to target solid tumor cancers, particularly neuroblastoma, the most common “extracranial” solid cancer of childhood.

One of the company’s therapeutic candidates, AUTO6NG, is a CAR T-cell system that targets GD2, a disialoganglioside that is abundantly expressed on most neuroblastomas yet is largely absent on normal tissue. AUTO6NG consists of two engineered modular vectors that are co-transduced together.

The first vector is designed to improve CAR T-cell persistence, while the second vector supports CAR T-cell functioning in the TME. Common to both vectors are GD2-targeting CAR T-cell modules. According to the company, this approach has shown high antitumor activity in patients without concomitant neurotoxicity.