Sponsored by the the AACR Chemistry in Cancer Research (CICR) working group, this year’s three-part New Drugs on the Horizon sessions featured the first disclosures of structures of 12 novel agents, both small molecules and biologics, that have recently entered, or imminently will enter, into phase I clinical trials.
Parts 1 and 2 took place Saturday, April 10, and Part 3 took place Monday. Registrants can watch replays of all three sessions through June 21, 2021.
“Inhibition of PD-1/PDL-1 pathway represents a breakthrough in the treatment of many cancer types, and multiple PD-1 or PDL-1 monoclonal antibodies have been approved for various cancer indications,” said Liangxing Wu, PhD, Incyte Corporation. “Small molecule PD-1/PDL-1 inhibitors could represent a novel approach for cancer treatment, so we set out to discover potent, selective, and orally bioavailable small molecules which can block PD-1/PDL-1 interactions.”
The investigators reasoned that small molecule inhibitors may potentially offer several advantages, Wu said, such as convenience of oral dosing, improved tumor penetration, titratable control of systemic drug concentration for rapid response to side effects, and potential convenience for small molecule combinations.
“We rationally designed a series of small molecules that bind PDL-1 and block PD-1/PDL-1 interactions,” Wu said. “Cellular potency was significantly improved after structure-guided drug design, and systematic structure modification led to improvement of ADME and PK properties, and ultimately these efforts led to the discovery of clinical candidate INCB086550 (Compound 550).”
Wu reported that preclinical studies demonstrated that INCB086550 induced PDL-1 internalization in vitro in a dose-dependent manner, noting that it also increased tumor-infiltrating T cells in the in vivo studies. INCB086550 exhibited single-agent activity in humanized mouse models, Wu said, and antitumor efficacy response was only seen in immunocompetent mice. It is being evaluated in patients with advanced solid tumors.
“There are many agents currently in development acting to agonize CD137, but we would argue that none of them yet fully meet the design goals dictated by the underlying biology of this receptor family,” said Nicholas J. Keen, MA, PhD, Bicycle Therapeutics. “Namely, that activity needs to be localized to the tumor microenvironment. We want to potentiate immune activation at the site where it’s needed, not systemically throughout the whole body.”
Rapid onset of action, complete tumor penetration, and controllable duration of activity are important for this receptor family.
“And ideally, we want a molecule that has no Fc receptor interactions,” Keen said. “Those interactions with Fc receptors are thought in part to have led to the hepatic toxicity of urelumab since Fc receptors are very highly expressed within the liver.”
The investigators selected Nectin-4 as their target tumor antigen and sought to effectively turn that tumor antigen into an analog of the physiologic ligand. That led to the development of the tumor-targeted immune-cell agonist BT7480.
“BT7480 is a completely novel, fully chemically synthetic, tumor antigen-dependent agonist of CD137 with very high biological potency, and it is at least 30-fold smaller than any comparator biologic that we know of,” Keen said. “It induces very robust antitumor responses to Nectin-4 expressing tumors in immune competent mouse models and, importantly, it induces immunologic memory to those tumors.”
Keen said that BT7480 demonstrated an extremely benign profile in preclinical safety evaluation with no liver effects observed. BT7480 will enter human trials this year. He said that this molecule may benefit more than 50 percent of breast, head and neck, ovarian, and esophageal cancer patients.
“IL-2 was one of the first immunotherapies to be approved for the treatment of cancer; however, it has had limited utility because of the high systemic toxicity it induces, in particular capillary leak syndrome,” said Pablo Umana, PhD, Roche Glycart AG. “In addition, it has a very short circulatory half-life, and when reducing the dose, it preferentially expands regulatory T cells.”
Over the last few decades, he said that researchers have attempted to improve IL-2 by engineering it in different ways, including mutating it to favor binding, in a preferential way, to the different receptor chains.
Umana and colleagues have generated a new, next-generation PD1-IL2v fusion protein that binds with high affinity to PD-1 and blocks the binding of PD-1 to PDL-1.
“It replaces the binding of CD25 by PD-1 binding for in-cis potentiation of IL-2Rβγ agonism by IL2v on PD-1 positive T cells, while avoiding the preferential expansion of Tregs,” he said. “At well tolerated doses preclinically, it leads to superior expansion of highly proliferative and cytotoxic effector cells and anti-tumor efficacy, as compared to the combination of non-PD-1-targeted IL2v and high dose of anti-PD-1 antibodies. It is the first PD-1-targeted IL-2Rβγ agonist immunocytokine to enter clinical trials.”
Chimeric antigen receptors (CARs) were the first approved gene therapy products in the U.S. and have shown great clinical efficacy, but there have also been significant challenges.
“These include a pretty high frequency of cytokine release syndrome (CRS), on target/off tissue toxicities, and neurological toxicities such as ICANS that are associated with CRS,” said Travis Young, PhD, Scripps Research Institute. “So, the challenge is how to create a CAR-T cell product that has both the remarkable potency of these approved products, but also introduces components that could potentially improve safety and include greater versatility.”
Young presented data on the development of a two-component platform that comprises a switchable CAR T-cell, CLBR001, and an antibody-based switch, SWI019, which represents a novel switchable platform that has recently entered the clinic for patients with relapsed or refractory B cell malignancies.
“This creates a two-component platform—the first is the switchable CAR-T cell and the second is an antibody-based switch—and when they’re combined, the switch forms a bridge, or an immunological synapse, between the CAR-T cell and the target cell,” Young said. “When it engages in this immunological synapse, it activates the CAR-T cell and causes lysis of the target antigen-positive cell.”
Young said that this platform potentially provides completely reversible on/off control, meaning adverse events can be mitigated simply by stopping dosage of the antibody-based switch.
A first-in-human study is ongoing to assess the safety, tolerability, and PK and PD of both CLBR001 CAR-T cells and of SWI019.
Jeffrey Johannes, PhD, and Elisabetta Leo, PhD, AstraZeneca Oncology, discussed the background and latest data on the experimental PARP-1 inhibitor AZD5305.
“When looking across the landscape of existing PARP inhibitors, we saw an opportunity in exploring the potential differentiation of a PARP1 inhibitor that is selective against PARP2 and other PARP family members, and clean with respect to secondary pharmacology,” Johannes said. “Recent literature has suggested that only PARP1 inhibition and trapping is required for synthetic lethal efficacy in homologous recombination repair deficient (HRD) cells.”
A growing body of evidence to suggests that PARP2 is linked to hematological toxicity, one of the main clinical adverse events observed with first generation PARP inhibitors. Combining these observations, the investigators hypothesized that a selective PARP1 inhibitor and DNA trapper may improve the therapeutic index versus first-generation PARP inhibitors.
“Our data show that AZD5305 is a next-generation PARP inhibitor with multiple clinical options as monotherapy and in combination, which are now being tested in Phase 1 clinical trials,” Leo said. “It is a potent, selective inhibitor and PARP1-DNA trapper with excellent secondary pharmacology, physiochemical properties, and oral bioavailability.
In the tumor microenvironment, various cells limit a robust anti-tumor immune response by producing immunosuppressive cytokines. Additionally, antigen-presenting cells in the tumor microenvironment are often tumor-supportive rather than tumor-destructive.
“One hypothesis that we and others have had is if you can wake up these antigen-presenting cells, this could result in a more productive and durable anti-tumor immune response,” said David Dornan, PhD, Bolt Biotherapeutics.
Building on that hypothesis, Dornan cited a paper published by investigators at Stanford University (Carmi, et al. Nature 2015) that showed that tumor-targeting monoclonal antibodies (mAbs) combined with various stimuli can activate antigen-presenting cells and resulted in complete eradication in various tumor models.
“To find what could be combined with tumor-targeting mAbs, we turned to the pattern recognition receptors. It has been shown that these can actually poise the tumor immune microenvironment for an anti-tumor immune response,” Dornan said. “The challenge of these agonists is that they are largely restricted to intratumoral administration due to their systemic toxicity, but in our mind, these provided a really great biological lead to potentially combine with antibody-targeting therapy.”
Dornan described the development of BDC-1001, a HER2 immune-stimulating antibody conjugate (ISAC), comprised of a HER2-targeting biosimilar, trastuzumab, which is conjugated to one of Bolt’s proprietary TLR7/8 agonists and is designed to maximize the potential anti-tumoral response of the immune system. BDC-1001 is currently in clinical development for the treatment of patients with HER2-expressing solid tumors, including HER2-low tumors.
HER2 is known to be overexpressed in about 20 percent of breast cancer patients, and there are many HER2-targeting compounds, antibodies, and antibody-drug conjugates (ADCs) already approved or in clinical development in HER2 positive breast and gastric/gastroesophageal cancer.
“Besides these indications, HER2, or the proto-oncogene ErB2, is also overexpressed in other indications, such as colorectal, lung, bladder, ovarian, and salivary gland cancers,” said Jenny Karlsson, PhD, Bayer US Healthcare. “However, there are still areas of high unmet clinical need for new HER2-targeting drugs with alternative mechanisms of action in tumors with low HER2 expression and other cancers beside breast and gastric cancer.”
HER2-TTC is a targeted thorium-227 conjugate, targeting HER2-expressing cancers. The compound relies on the high linear energy transfer upon alpha decay leading to DNA double-strand breaks and subsequent tumor-cell cytotoxicity.
“Based on this unique mode of action, HER2-TTC differentiates from other HER2-targeting agents, with the potential to overcome resistance to HER2 targeted therapy and target HER2 low-expressing tumors,” Karlsson said. “Interestingly, HER2-TTC demonstrates synergistic anti-tumor activity in combination with olaparib in a BRCA2 knockout model.”
A first-in-human study is ongoing and recruiting in the dose-escalation cohort.
Theresa Kolben, MD, PhD, F. Hoffmann-La Roche Ltd., presented preclinical as well as clinical data of RG6292 and how this antibody modifies the tumor microenvironment in a phase 1 single-agent trial.
“RG6292 is a first-in-class Fc-optimized IL-2 non-blocking CD25 Treg depleter,” Kolben said. “The preliminary single-agent dose-escalation data indicate a good tolerability, so no DLTs, no grade 4 or 5 events, and with manageable skin toxicity being the most frequent adverse event.”
Additionally, she said that RG6292 exhibits a linear PK profile, time-independent PK, and no anti-drug antibodies.
“Nicely, there is also a dose-dependent peripheral Treg depletion, and when we look into the tumor, we see an immune-cell relocation and conversion of tumors to a CD8 inflamed phenotype,” Kolben said. “Our next steps, besides the ongoing single-agent dose escalation, include looking into the combination of RG6292 with atezolizumab.”
Robert L. Hoffman, Pfizer Global Research & Development, discussed PF-07104091, a CDK2-selective inhibitor for the treatment of cyclinE-amplified cancers. CDK2, a member of a large family of cyclin dependent kinases, requires binding to a regulatory cyclin subunit, such as A or E.
“The cyclin E/CDK2 complex plays a critical role in controlling the progression of the cell cycle from G1 to S phase by phosphorylation of the tumor suppressor protein retinoblastoma, or RB,” Hoffman said. “Specifically, the CDK2/cyclineE complex, along with CDK4/6 cyclinD complexes, successfully phosphorylate RB, which releases the transcription factor E2F and promotes S phase entry and progression of the cell cycle.”
CyclinE/2 amplification occurs across a diverse set of tumor types such as lung, breast, and ovarian cancers. Amplification represents intrinsic resistance to CDK4/6 inhibitors.
Hoffman said there has been a focus on developing cyclinE CDK2 as an oncology target for the last 30 years. Hampering successful development of CDK2 inhibitors has been the toxicity associated with a lack of selectivity against CDK2 isoforms.
Hoffman’s presentation described efforts to discover a CDK2-selective inhibitor. These efforts yielded the compound PF-07104091, which has 100-fold selectivity against CDK1, 200 to 400-fold selectivity against CDK4 and CDK6, 170-fold selectivity against CDK9, and greater than 500-fold selectivity against GSK3B.
In a tumor growth inhibition study in an ovarian cancer model, single-agent PF-07104091 demonstrated a dose-dependent tumor growth reduction with 86 percent maximal effect and demonstrated single-agent efficacy across multiple ovarian models. PF-07104091 also re-established cell-cycle control in cancer types where CDK4/6 inhibitors lack efficacy.
PF-07104091 is currently being tested in phase I/IIa dose escalation trial with the first patient dosed in October 2020.
Leslie Burnett, PhD, Revolution Medicines, Inc., presented background on RMC-5552, a selective bi-steric inhibitor of mTORC1 that suppresses 4EBP1 phosphorylation for the treatment of mTORC1-activated tumors, including RAS pathway escape.
“We envision that RMC-5552 can be used as companion to RAS pathway inhibitors as well as single-agent therapy for tumors bearing mutations that results in mTORC1 activation,” Burnett said.
The Pi3K/mTOR signaling pathway is commonly dysregulated in a wide range of human cancers. mTOR kinase sits downstream of receptor tyrosine kinase and PI3 kinase signaling, and forms the core of two distinct complexes, mTORC1 and mTORC2.
mTORC1 phosphorylates and inactivates S6K and modulates activity of tumor suppressor 4EBP1, Burnett said. In cancer, phosphorylation of 4EBP1 releases eIF4E, which drives oncogene translation and promotes tumorigenesis.
Two generations of mTOR inhibitors have been previously developed; however, these were not effective at inhibiting 4EBP1 phosphorylation in the clinic due to mTORC2-mediated dose-limiting toxicities, Burnett said. Revolution wanted to develop a drug that would be a potent inhibitor of 4EBP1 and S6K phosphorylation inhibition and was selective against mTORC2 and other lipid kinases.
In an MCF-7 CDX model, RMC-5552 showed dose-proportional tumor inhibition and was tolerated at all tested dose levels. Researchers observed 9 out of 10 regressions at the highest dose level.
In an upcoming phase I trial, RMC-5552 will be given as a monotherapy for tumor genotypes linked to hyperactivated mTORC1. RMC-5552 will also be tested in combination with RAS inhibitors for patients with tumor harboring both RAS mutations and mutations resulting in the activation of mTOR.
Robert Macleod, PhD, Ionis Pharmaceuticals, Inc., presented data on ION-537, a next-generation antisense oligonucleotide (ASO) inhibitor of YAP1, from preclinical cancer models.
“YAP1 is a transcriptional co-activator that is part of the Hippo pathway,” Macleod said. “The Hippo pathway is an important development pathway that is hijacked in cancer. It is a tumor suppressor, and its activation restricts YAP1/TAZ proteins to the cytoplasm.”
Nuclear YAP levels or high expression is correlated with shorter survival in multiple tumor types, most notably HCC and head and neck squamous cell carcinoam (HNSCC), Macleod said, and YAP1 is difficult to drug target.
Macleod discussed a pilot study that showed that YAP1 ASOs induced regression of HCC tumor in a Sav knockout mouse model. A broader Sav knockout mouse clinical trial comparing YAP1 ASOs to control ASOs confirmed this activity.
Looking at HNSCC tumors, Macleod and colleagues found that tumors harboring LOF alterations in the tumor suppressor gene FAT1, or YAP1 gene amplification were highly sensitive to ASO-mediated YAP1 depletion.
Ionis identified ION-537 as the YAP1 human candidate drug and sought to identify a biomarker that could be used to optimize the use of the drug. They found that nuclear YAP1 levels correlated with YAP1 RNA inhibition induced by YAP1 ASOs as well as antiproliferative activity in cell lines, several xenograft models, and human clinical models.
Preclinical studies have also been done in HNSCC cell lines and in vivo studies.
A phase I dose-escalation study in patients with advanced cancer started in November 2020. Once maximum-tolerated dose and the recommended phase II dose are found, an expansion phase will be initiated in HNSCC patients with FAT1 mutations and/or positive nuclear YAP1 or advanced solid tumor positive for nuclear YAP1 expression.
M4205 is a highly selective inhibitor of disease-relevant cKIT mutations for unresectable metastatic or recurrent gastrointestinal stromal tumors (GIST).
“GIST are the most common tumor of the GI tract. There are about 5,000 to 6000 new cases every year in the U.S.,” said Andreas Blum, PhD, Merck KGaG.
Activating mutations in cKIT act as primary oncogenic drivers for the GISTs. Imatinib is active on the primary driver mutation of cKIT, which occurs in exon 11, but resistant mutations occur upon treatment. Sunitinib is used as a second-line therapy but has a low overall response rate, Blum said.
That means there is a high unmet medical need for patients with GIST that has resistance mutations against imatinib. Blum and colleagues sought a candidate drug with a target profile that comprises mutations against exons 13 and 17, has high kinase selectivity, and a superior safety profile compared with sunitinib.
M4205 met the outlined criteria. It had high activity against exon 11, 13 and 17 mutations of cKIT, with high kinase selectivity, with excellent metabolic stability, permeability, and solubility.
Preclinical studies of M4205 in GIST xenograft models expressing mutations in exon 11 showed that M4205 delivered strong responses compared with imatinib. Compared with second-line standard of care sunitinib, M4205 treatment resulted in deeper anti-tumor response in xenograft models expressing secondary resistance mutations in exon 13. Similar results were seen in a xenograft model expressing secondary resistance mutations in exon 17.
“We see M4205 as a best-in-class option for GIST,” Blum said. There is a phase I clinical study of GIST in preparation.”