Invikafusp Alfa Demonstrates Single-Agent Activity in PD-(L)1–Resistant Solid Tumors

James L. Gulley, MD, PhD

Treatment with the first-in-class, selective, dual T-cell agonist invikafusp alfa led to a disease control rate (DCR) of 50% in patients with heavily pretreated anti–PD-(L)1–resistant solid tumors, according to findings from the phase 1/2 STARt-001 trial (NCT05592626) presented at the 2024 SITC Annual Meeting.¹

Data compared favorably to the 32% DCR seen with prior anti–PD-(L)1 therapy across all dose cohorts. Additionally, more patients experienced tumor shrinkage within the optimal biologic dose range. In this population (n = 14), 7 patients experienced tumor shrinkage, including 2 confirmed partial responses.

“[This] first-in-human phase 1 data validates a new class of selective T-cell receptor [TCR] Vβ dual T-cell agonists for precision cancer immunotherapy,” lead study author James L. Gulley, MD, PhD, said in a presentation. “The strongest efficacy signal emerged in the tumor mutational burden–high [TMB-H] population, particularly in patients with microsatellite stable [MSS] colorectal cancer [CRC].”

Gulley is the acting codirector of the Center for Cancer Research at the National Cancer Institute in Bethesda, Maryland.

“Invikafusp alfa introduces into the clinic a novel way to activate T cells. You have the anti-Vβ TCR portion that the antibody binds to. It binds to a selective set of T cells that express Vβ6 or Vβ10, and it also brings [interleukin 2 (IL-2)] on the other arm of the antibody,” Gully added. “That IL-2 will only bind to the same T cells binding in CIS and therefore activate, very specifically, those T cells. These Vβ6 or Vβ10 T cells [where these] binds [occur] happen to be among the highest-expressed T cells in tumor-infiltrating lymphocytes in patients across a wide range of different tumors.”

To be eligible for enrollment in the trial, patients needed to have unresectable, locally advanced, or metastatic solid tumors that were TMB-H, microsatellite instability–high (MSI-H)/mismatch repair-deficient (dMMR), or were virally associated cancers. Also required was an ECOG performance status of 0 or 1; prior treatment with anti–PD-(L)1 therapy was allowed.

Phase 1 consisted of a standard, 3+3 dose-escalation schema in which patients received invikafusp alfa intravenously at a dose of 0.01 mg/kg, 0.02 mg/kg, 0.04 mg/kg, 0.08 mg/kg, 0.12 mg/kg, and 0.16 mg/kg every 2 weeks.

The primary objectives of phase 1 were to establish the recommended phase 2 dose (RP2D) and evaluate safety and tolerability. Secondary objectives were to evaluate preliminary antitumor efficacy and pharmacokinetics. Phase 2 dose expansion employed an optimal Simon’s 2-stage design, enrolling patients into 1 of 3 cohorts: TMB-H tissue agnostic, MSI-H tissue agnostic, and TMB-H and/or MSI-H/dMMR CRC.

The primary cohort, which enrolled 28 patients, was well balanced in terms of baseline characteristics with that of the overall population (n = 35). A total of 16 tumor types were represented overall, including adrenocortical (2.9%), ampullary (2.9%), anal (14.3%), breast (2.9%), cervical (17.1%), colon (11.4%), esophageal (2.9%), melanoma (2.9%), Merkel cell (5.7%), nasopharyngeal (5.7%), non–small cell lung cancer (5.7%), pancreatic (2.9%), rectal (2.9%), skin (2.9%), and vulvar (2.9%).

Within the primary cohort, the median patient age was 57.9 years (range, 34-78), and most patients were female (53.6%), White (78.6%), and had an ECOG performance status of 1 (64.3%). Most patients had received at least 4 prior lines of therapy (57.1%), including prior anti–PD-(L)1 therapy (89.3%). Best response to prior immunotherapy was progressive disease (42.9%) or stable disease (17.9%). Half of all cancers were virally associated compared with being TMB-H (42.9%) or MSI-H (7.1%).

In terms of the pharmacokinetic and pharmacodynamic activity of the agent, Gulley explained, “We saw very nice linear increases in the peak concentrations as we increased the dose. At the higher dose levels, you can see a remarkable and very selective expansion of peripheral Vβ6 and Vβ10 cells. One patient had a 500% increase in the Vβ6 cells. If you look at flow cytometry, you can see that the peak effect is [reached at] about the 0.08-mg/kg dose level.”

Gulley noted that the safety profile of the treatment was consistent with its mechanism of action, which reflects T-cell activation and expansion in vivo.

Temporary, low-grade treatment-related adverse effects (TRAEs) occurred in the first and second cycles of treatment without immune effector cell–associated neurotoxicity syndrome or grade 4 cytokine release syndrome (CRS). The maximum tolerated dose was not reached, and 0.08 mg/kg was selected as the RP2D.

“[We saw a] manageable, on-target safety profile without prophylactic use of corticosteroids, tocilizumab [Actemra], or step-up dosing. Common anti–PD-(L)1–related, immune-related AEs [such as] colitis and pneumonitis were not seen,” Gulley said.

AEs that were seen in dose escalation (n = 28) included chills, nausea, vomiting, pruritus, pyrexia, arthralgia, infusion-related reactions (IRRs), CRS, hypoxia, hypotension, fatigue, and rash. TRAEs that occurred when optimal biologic doses (0.08 and 0.12 mg/kg) were administered (n = 14) consisted of chills (grade 1/2, 35.7%; grade 3, 0%), nausea (grade 1/2, 50.0%; grade 3, 7.1%), vomiting (grade 1/2, 57.1%; grade 3, 0%), pruritus (grade 1/2, 42.9%; grade 3, 21.4%), pyrexia (grade 1/2, 64.3%; grade 3, 0%), arthralgia (grade 1/2, 21.4%; grade 3, 7.1%), IRRs (grade 1/2, 42.9%; grade 3, 0%), CRS (grade 1/2, 71.4%; grade 3, 14.3%), hypoxia (grade 1/2, 7.1%; grade 3, 7.1%), hypotension (grade 1/2, 42.9%; grade 3, 21.4%), fatigue (grade 1/2, 21.4%; grade 3, 0%), and rash (grade 1/2, 42.9%; grade 3, 7.1%).

Gulley also shared a case report highlighting the potential for rapid, deep, and durable single-agent response with invikafusp alfa. This patient was a 46-year-old Caucasian female with metastatic CRC (mCRC). She had been heavily pretreated with 6 prior lines of therapy and was resistant to PD-1 therapy. Her CRC was deemed TMB-H with 10 mut/mb, MSS, and RAS wild-type. She received 0.08 mg/kg of invikafusp alfa in the seventh-line setting. Nine months after receiving therapy, she had a confirmed partial response with a 58% reduction in her target lesion. Moreover, tumor burden reduction was seen across all lesions and elevated baseline CEA levels returned to normal range.

“Three cohorts [for] patients with TMB-H disease, MSI-H/dMMR tumors, and mCRC in the phase 2a expansion portion have opened and are enrolling,” Gulley concluded.

Disclosures: Dr Gulley disclosed that he is employed by the National Cancer Institute (NCI) and that the NCI has a cooperative research and development agreement with Marengo Therapeutics, the manufacturer of STAR0602.

Reference

Gulley JL, Sullivan RJ, Friedman CF, et al. STARt-001: a phase 1/2 study of invikafusp alfa, a first-in-class TCR β chain-targeted bispecific antibody, as monotherapy in patients with antigen-rich solid tumors resistant to anti-PD(L)1. Presented at: 2024 SITC Annual Meeting; November 6-10, 2024; Houston, TX. Abstract 1470.