Marios Giannakis, MD, PhD, discusses how targeting the WNT signaling pathway could harness a previously underutilized pathway involved in GI tumor development, spotlights the activity and safety of CGX1321 with or without pembrolizumab in phase 1 studies, and emphasizes the need for continued investigation of CGX1321 to further validate this precision medicine approach.
Preliminary efficacy and safety seen with the novel highly potent and selective O- acyltransferase porcupine inhibitor CGX1321 in patients with advanced gastrointestinal (GI) tumors harboring RSPO fusionsdemonstrated the feasibility of targeting the WNT signaling pathway in these specific tumor types, according to Marios Giannakis, MD, PhD. Moreover, administering the agent in combination with a checkpoint inhibitor could potentially curb the development of immunotherapy resistance in this population, Giannakis said, which can be associated with WNT pathway activation.
Phase 1 studies conducted in the United States (NCT02675946) and China (NCT03507998) evaluating CGX1321 alone and in combination with pembrolizumab (Keytruda) showed that in the monotherapy arm, patients whose tumors displayed an RSPO fusion experienced a disease control rate (DCR) of 77%, durable response, and significant circulating tumor DNA reduction. Similarly, the DCR in the combination arm was 83% for patients with microsatellite stable (MSS) tumors and 33% for patients with RSPO fusions. Findings were presented at the 2023 ASCO Annual Meeting.
“The WNT signaling pathway is in all cancers, but especially GI cancers can be targeted. However, we have to be rational and think about the subset of [patients who] are most likely to respond to [WNT inhibition],” said Giannakis, a medical oncologist and clinical investigator at the Gastrointestinal Cancer Center at Dana-Farber Cancer Institute and an assistant professor of medicine at Harvard Medical School in Boston, Massachusetts.
In an interview with OncLive®, Giannakis discussed how targeting the WNT signaling pathway could harness a previously underutilized pathway involved in GI tumor development, spotlighted the activity and safety of CGX1321 with or without pembrolizumab in phase 1 studies, and emphasized the need for continued investigation of CGX1321 to further validate this precision medicine approach.
Giannakis:The WNT signaling pathway is an important pathway for tumor growth among all cancers, especially GI cancers. However, attempts to target and inhibit this pathway have not been very successful so far. With this trial, we are testing a WNT signaling inhibitor against GI cancers, both in a dose-escalation [phase] and in patients whose tumors have specific mutations that are predicted to be sensitive to this approach. WNT signaling also creates an immunosuppressive tumor microenvironment, so another component of the study was to combine WNT signaling inhibition with immune checkpoint inhibition.
[CGX1321] is part of a class of drugs called porcupine inhibitors. It interferes with the secretion of the WNT ligand. The WNT ligand is the molecule that initiates and activates WNT signaling outside the cell. Based on a lot of preclinical models that we and others have generated, this specific inhibitor is predicted to be particularly effective against tumors that have mutations upstream of the WNT signaling pathway, such as our RSPO fusions or RNF43 inactivating mutations.
This was an early-phase study. It included both a phase 1 and 1b component to escalate the porcupine inhibitor alone and in combination with pembrolizumab, which was the PD-1 inhibitor used here. [For the dose escalation of single-agent CGX1321], we selected [patients with] all solid tumors, but the majority of them had GI cancers. The combination [cohort] with immunotherapy also [predominantly consisted of patients with] GI tumors, the majority of which had colorectal cancer [CRC] and small bowel adenocarcinoma. In the dose-expansion phase of the study, we specifically selected [patients with] GI cancers who had RSPO fusions or RNF43 inactivating mutations. We tested the single agent, as well as the combination with pembrolizumab, in these specific patients.
The purpose of a phase 1 study is to determine a dose that is safe and tolerable, and that's what we achieved in the dose-escalation components with a single agent and [combination]. Once we picked the [recommended] doses, we focused on patients with specific mutations in the expansion cohorts. [When assessing the] single-agent porcupine inhibitor [in patients with] GI tumors who had RSPO fusions, [we found that] the drug was able to control the cancer in 77% of cases. However, in [patients with] tumors who did not have RSPO fusions, the DCR was 0%, highlighting the importance of targeting the right tumors with specific WNT signaling inhibitors.
In the combination arm, we chose patients whose tumors had RSPO infusions, we saw a DCR of 83%. We also observed a response in 33% of patients, which is interesting because all of them had MSS CRC or small bowel adenocarcinoma. These efficacy results need to be further tested in larger samples of patients and in a randomized fashion.
Overall, both the single-agent porcupine inhibitor and the combination with immune checkpoint inhibition were well tolerated among patients. For the single agent, we observed that the most frequent adverse effect [AE] was altered taste, or dysgeusia. However, this was not of severe grade. With the combination, the most common AEs were liver enzyme abnormalities and fatigue. Again, these were not severe.
Since this was a phase 1 study, a dose-limiting toxicity was hypercalcemia with the single agent. This [is because] the WNT pathway can be involved in bone resorption. In a small percentage of patients, there were fractures observed. [This] is something that needs to be prophylactically addressed. We did that in the study by adding, for example, denosumab or [zoledronic acid] to prevent bone-related AEs.
What we would like to test next would be whether some of these efficacy results for the single agent and combination hold up in more patients with GI cancers who have these specific RSPO fusions, and [we[ also want to retest our hypothesis for patients with RNF43 mutations, for which we had very few patients [enrolled in the phase 1/1B] trial to make any significant conclusions. We would like to test these findings in a larger, randomized setting. These plans are under discussion right now.
This [study] is an example where a lot of preclinical data and the scientific rationale behind this approach turned out to be at least preliminarily validating, meaning we must match the right patients with RSPO mutations to this specific porcupine inhibitor, based on how we predicted that drug would work and what the preclinical data showed. We'll have to follow precision medicine, but when we do that, perhaps targeting WNT signaling [with CGX1321] alone or [in combination] with immunotherapy is reasonable. Also, the fact that WNT signaling inhibition in the right context could enhance immunotherapy response in patients with MSS tumors is very interesting and exciting. We are all very interested to see this validated and tested further.
Editor’s note: Dr Giannakis reports receiving research funding from Janssen and Servier; he also reports a pending patent on biomarkers of immune response.
Giannakis M, Le DT, Pishvaian M, et al. Phase 1 study of WNT pathway Porcupine inhibitor CGX1321 and phase 1b study of CGX1321 + pembrolizumab (pembro) in patients (pts) with advanced gastrointestinal (GI) tumors. J Clin Oncol. 2023;41(suppl 16):3514. doi:10.1200/JCO.2023.41.16_suppl.3514