GTB-3550 TriKE Stimulates Natural Killer Cell Function in MDS and AML

GTB-3550 TriKE was found to safely drive natural killer cell proliferation in patients with high-risk myelodysplastic syndromes (MDS) and acute myeloid leukemia (AML), according to data from initial dose cohorts of a phase 1/2 trial (NCT03214666) presented during the 2020 ASH Annual Meeting & Exposition.

Results showed that the GTB-3550 TriKE, when administered at a daily dose of 5 mcg/kg in a patient with therapy-related AML who had previously been treated with 3 lines of induction chemotherapy, led to stable blasts and improved platelet transfusion needs; this qualified them for retreatment. However, a second patient with relapsed/refractory FLT3-ITD–mutated AML who received the TriKE at the same dose experienced disease progression. This patient had previously received 2 cycles of induction chemotherapy, relapsed after first complete remission (CR), and went on to receive 2 lines of treatment with relapsed disease. Notably, neither of these patients experienced clinically significant toxicities that were associated with treatment.

Two patients who received the agent at a daily dose of 10 mcg/kg were able to achieve stable disease. One patient who had relapsed/refractory AML had received previous treatment with azacitidine for 1 year with disease control before progression and experienced a brief remission with venetoclax (Venclexta) and cytarabine; this patient did not respond to an IDH1 inhibitor. Another patient had therapy-related MDS with residual disease following treatment with a hypomethylating agent (HMA) followed by HMA plus venetoclax. No significant toxicities were experienced by these patients, either.

Two patients received the TriKE at a higher daily dose of 25 mcg/kg. One of these patients had secondary AML following an 18-month CR with the combination of venetoclax and azacitidine; they relapsed with a unique FLT3 mutation and did not respond to gilteritinib (Xospata). The first patient experienced blast reduction with the GTB-3550 TriKE, going from 18% blasts pretreatment to 12% blasts following treatment. This patient also experienced improved platelet transfusion needs. However, the patient also had significant toxicity of neutropenic fever. The other patient had relapsed disease, having progressed following re-induction with venetoclax plus an HMA. This patient achieved stable disease with no clinically significant toxicities.

“Our immune data show that GTB-3550 TriKE stimulates robust natural killer cell activity, with best responses to date including mild blast reduction or stable disease with improvement to platelet transfusion needs in 2 patients,” said Erica Warlick, MD, an associate professor of medicine in the Division of Hematology, Oncology, and Transplantation at the University of Minnesota Medical School during a presentation of the data.

Although several novel targeted agents have emerged and received regulatory approval in the treatment paradigms of relapsed/refractory AML and MDS, many patients’ tumors do not harbor targetable mutations and many have exhausted all chemotherapy options available to them. Natural killer cell infusions, when done after lymphodepleting chemotherapy, has been shown to result in remissions for patients with relapsed/refractory disease, but this approach does not have antigen specificity, said Warlick. To provide these patients with a potential option, investigators developed GTB-3550 TriKE, which is a tri-specific killer engager that targets CD16, IL-15, and CD33.

“[The agent] is composed of 2 single-chain variable fragments [scFvs]: 1 that binds CD16 on natural killer cells and 1 that binds CD33 on myeloid malignancies,” explained Warlick. “Sandwiched between the 2 is an IL-15 linker, bridging the CD16 and CD33 scFvs, leading to sustained cell activation.”

Preclinical data demonstrated that the GTB-3550 TriKE induced natural killer cell activation and had targeted cytolytic activity in xenogenic AML mouse models. Investigators launched the phase 1/2 trial to see whether the agent would have the same success in patients.

To be eligible for enrollment on the single-center phase 1/2 trial, patients needed to have a CD33-positive malignancy. Patients with AML needed to have experienced primary induction failure or have relapsed disease with one failed re-induction attempt. Patients with MDS had to have high-risk disease and have progressed on 2 lines of treatment. All patients needed to be aged 18 years of age or older and have adequate renal, hepatic, cardiac, and lung function. Lastly, patient had to have an absolute lymphocyte count of 200 cells/μl or higher, or absolute circulating CD56-positive/CD3-negative natural killer cell count of over 25 cells/μl 14 days prior to treatment initiation.

The primary end point of the trial was to identify the maximum-tolerated dose of the GTB-3550 TriKE, and correlative objectives included identifying the describe number, phenotype, activation status, and function of both natural killer cells and T cells.

“Two patients per dose cohort were enrolled and went on to the next dose cohort if there were no dose-limiting toxicities,” said Warlick. “To date, we have completed the first 3 dose levels and the first patient at dose cohort 4, which is 50 mcg/kg per day.”

In each block, patients received 4 consecutive 24-hour continuous infusions (CI) of the agent at their assigned dose with a 72-hour rest following blocks 1 and 2. In the dose-finding phase 1 portion of the trial, each participant was given the agent at the assigned dose for 3 sets of 4 consecutive 24-hour infusions; these were separated by a 72-hour rest.

“The most exciting findings that we have seen were with our correlative studies,” Warlick added. “[We observed] increased CD69 expression during TriKE infusion at all cohort levels. This increased expression correlated with egress or exit of natural killer cells from the blood on days 3, 10, and 17 with rebound at days 8, 15, and 22.”

Additionally, sustained natural killer cell activation or proliferation was demonstrated at all dose cohort levels and this was observed across all treatment days.

“We looked at activation of T cells and found no significant activation of CD4-positive T cells and only brief T-cell proliferation at day 8 for CD8-positive T cells,” Warlick noted.

Example data from 1 patient within the 25 mcg/kg per day cohort showed that the experimental agent induced increased target cell killing compared with the pre-treatment sample on days 1, 8, 15, and 22 against K562 and HL60 targets.

With regard to safety, no infusion-related reactions or dose-limiting have been observed. Moreover, there were no signs of clinical immune activation, as defined by treatment-related tachycardia of over 30 beats above baseline or a temperature of 102.4 Fahrenheit or greater persisting for 4 hours or more.

To date, 1 patient has been enrolled onto the 50 mcg/kg per day cohort; this patient has MDS/myeloproliferative neoplasm with red blood cell transfusion dependence following previous treatment with an HMA and luspatercept (Reblozyl). No significant toxicities have been observed in this patient, but data pertaining to disease assessment following treatment with the GTB-3550 TriKE are still pending.

“Dose-escalation is currently ongoing and will show if natural killer cell proliferation translates into clinical efficacy,” Warlick concluded.

Reference

Warlick ED, Weisdorf DJ, Vallera DA, et al. GTB-3550 TriKE™ for the treatment of high-risk myelodysplastic syndromes (MDS) and refractory/relapsed acute myeloid leukemia (AML) safely drives natural killer (NK) cell proliferation at initial dose cohorts. Presented at: 2020 ASH Annual Meeting and Exposition; December 5-8, 2020; Virtual. Abstract 65. https://bit.ly/3qzFUhq.