Novel CAR NK Cell Therapies Show Early Signs of Safety, Efficacy in AML and NHL

Two chimeric antigen receptor (CAR)-engineered natural killer (NK) cell therapies, NKX101 and NKX019, showed promising signs of safety and efficacy in treatment of heavily pretreated patients with acute myeloid leukemia (AML) and non-Hodgkin lymphoma (NHL), according to a press release from Nkarta, Inc.¹

These data come from 2 separate dose finding-studies, 1 examining the NKX101 in relapsed/refractory AML and myelodysplastic syndrome (MDS), and the other being a phase 1 study evaluating the safety and anti-tumor activity of NKX019 as a multi-dose, multi-cycle monotherapy in patients with relapsed/refractory B-cell malignancies (NCT04623944, NCT05020678).

“We’re excited to see our CAR NK co-lead candidates, NKX101 and NKX019, show such striking early single-agent activity in heavily pretreated patient populations, with an exceptional safety profile without the side effects associated with CAR T cell therapies,” said Paul J. Hastings, president and CEO of Nkarta, in a press release. “These encouraging data across multiple indications further validate Nkarta’s best-in-class NK cell platform, as we seek to transform cancer treatment by bringing together the safety advantages of NK cells with an off-the-shelf modality designed to make the benefits of cell therapy accessible in a community setting.”

In the study of NKX101, 5 patients with relapsed/refractory AML were treated with 1 billion or 1.5 billion CAR NK cells per dose. After the infusion, 60% of patients experienced a complete response (CR). Of those patients, 2 of the 3 also tested negative for minimal residual disease (MRD).

The second study consisted of evaluating the CD19-targeted therapy, NKX019. When administered in a 3-dose regimen with 1 billion CAR-positive NK cells per dose, induced responses were found in 5 of 6 patients with relapsed or refractory NHL.

There were no dose-limiting toxicities observed in either study as well as no commonly observed adverse effects (AEs) related to CAR T-cell therapies. Additionally, no cases of cytokine release syndrome, neurotoxicity, or graft-vs-host disease were reported in either study after CAR NK cells were administered.

By April 21, 2022, a total of 21 patients, 17 with AML and 4 with high-risk myelodysplastic syndrome (MDS), were enrolled within the phase 1 study exploring NKX101. Eligibility for enrollment was open to patients with an ECOG performance status of less than 2, adequate organ function, a platelet count ≥30,000/uL, and a suitable haplo-matched related donor who is able and willing to undergo leukapheresis.

Primary end points of the trial include safety and tolerability, response rate to NKX101, and proportion of subjects experiencing dose-limiting toxicities. Secondary end points include assessment of NKX101 half-life, duration of persistence, evaluation of host immune response against NKX101, and response rate.

Patients were heavily pretreated and received a median of 3 prior lines of therapy (range, 1-12), including prior venetoclax (Venclexta) for those with AML. The median percentage of blast cells were present in 27% of bone marrow samples (range, 3% to 85%) at baseline.

In part 1, patients received fludarabine/cyclophosphamide lymphodepletion followed by 3 or 2 weekly doses of NKX101. Subjects in part 2 will receive either fludarabine/cyclophosphamide lymphodepletion or fludarabine/cytarabine (ara-C), followed by 3 or 2 weekly doses of NKX101 depending on specific expansion cohort. Within regimen A, patients will be administered 3 doses of NKX101 on days 0, 7, and 14 of a 28-day cycle while those given regimen B will have 2 doses of NKX101 on days 0 and 7 of a 28-day cycle.

In the 2-dose group, the highest dose consisted of 1.5 billion cells per infusion. This group enrolled 3 patients with AML, and there were no responses observed at this dose level. The second group in the 2-dose cohort received 150 to 450 million cells per infusion. There were 3 patients with AML and 2 with MDS in this group. In those with AML, the overall response rate (ORR) was 33% and there was no response observed in patients with MDS.

Within the 3-dose arm, the highest dose patients received was 1.0 billion to 1.5 billion CAR+ NK cells per infusion. This group contained 5 patients with AML and 2 with MDS. There were no responses observed in those with MDS and the ORR was 60% for those with AML, all of which were CRs. A lower dose of 100 million to 300 million was also given in the 3-dose cohort. Six patients with AML received this dose, with a response seen in 4 (ORR, 67%). There were no CRs. No patients with MDS were treated with this dose. Across all dose sizes in the 3-dose group, the ORR was 47% for those with AML (8 of 17) with a CR rate of 18% (3 of 17).

The most common adverse events (AE) were those associated with lymphodepletion, the most common grade ≥3 AEs were thrombocytopenia (48%), febrile neutropenia (38%), neutropenia (33%), and anemia (29%). Two patients experienced a grade 2 infusion reaction.

“Relapsed/refractory acute myeloid leukemia is a historically hard-to-treat disease, and given the lack of effective treatments, people with cancer and those who treat them are faced with few options,” investigator in the NKX101 clinical trial Marcello Rotta, MD, Colorado Blood Cancer Institute (CBCI), a part of the Sarah Cannon Cancer Institute at Presbyterian/St. Luke’s Medical Center, said in a statement. “Complete responses with corresponding MRD negativity in R/R AML using engineered NK cells, as seen in these preliminary findings, is encouraging. We look forward to leading further investigation to better understand the full potential of a CAR NK approach.”

At the time of the data analyses of April 21, 2022 of the study evaluating NKX019,13 patients were enrolled with various histologies. These included study 5 patients with aggressive large B-cell lymphoma (LBCL), 3 with B-cell acute lymphoblastic leukemia (B-ALL), 3 with follicular lymphoma, 1 with mantle cell lymphoma (MCL), and 1 with marginal zone lymphoma (MZL).

Inclusion in the study is open to patients with a histologically or cytologically confirmed diagnosis of relapsed/ refractory B-cell NHL or chronic lymphocytic leukemia or B-ALL, an ECOG performance status 0 or 1, measurable disease, have received ≥2 lines of therapy, have not responded or relapsed within 12 months of completion of their prior line of therapy, adequate organ function, and more.

The coprimary end points of the study are safety/tolerability, and the proportion of subjects experiencing dose-limiting toxicities of NKX019. Secondary end points consist of assessment of NKX019 half-life, duration of persistence, evaluation of host immune response against NKX019, and ORR.

Within the study, patients received a median of 4 prior lines of therapy (range, 2-7) and enrolled patients with extensive disease in both the United States (n = 3) and Australia (n = 10). Additionally, 3 doses of NKX019 were administered at either 300 million cells per dose or 1 billion cells per dose on days 0, 7, and 14. Prior to the first dose of NKX101, all patients received fludarabine and cyclophosphamide lymphodepletion.

The ORR across all NHL subtypes was 50% at the 300 million dose and 83% at the 1 billion cell dose. These groups showed a CR rate of 25% and 50%. No responses were seen in those with B-ALL. Within the 300 million dose group, no patients were treated with MCL or MZL, and of those with LBCL treated at this dose (n = 3), 1 partial response (33%) was shown. Further, 1 patient with follicular lymphoma treated at this dose had a CR.

A total of 2 patients with LBCL received 3 doses of NKX019, each at 1 billion cells. A CR experienced by 1 patient (50%). There was also 1 patient with MCL and 1 with MZL treated at this dose level who each experienced a CR. There were 2 patients with follicular lymphoma treated at this dose level, with an ORR of 100% that consisted entirely of partial responses.

In regard to safety, NKX019 demonstrated similarity to what was seen in the study of NKX101. The most common grade or higher 3 AEs were neutropenia (69%), thrombocytopenia (38%), febrile neutropenia (23%), and anemia (15%). Additionally, 1 patient experienced a grade 1 infusion site reaction.

“The curative potential of CAR T-cell therapy is truly remarkable, but many eligible patients are still not cured, and the safety and logistical challenges of approved autologous CAR T therapy are barriers,” investigator in the NKX019 trial Michael Dickinson, MD, lead, Aggressive Lymphoma disease group, Clinical Hematology, Peter MacCallum Cancer Center and Royal Melbourne Hospital, said in a statement. “The NKX019 trial exemplifies the continued progress of our field. NKX019 showed clear activity, in patients with a range of NHL histologies, without the sort of toxicities expected of other cellular therapies, supporting continued exploration of this CAR NK candidate.”

Both studies are currently enrolling patients at t3-dose regimens of 1.5 billion NK cells per dose in the dose finding portions, according to Nkarta Inc. Follow-up and updates on the data from the higher dose cohorts are also to be submitted for presentation at a future medical meeting.

References

Nkarta announces positive preliminary dose finding data for two lead engineered natural killer cell programs. News release. Nkarta, Inc. April 25, 2022. Accessed April 26, 2022. https://bit.ly/3LevaPy

NKX101, Intravenous Allogeneic CAR NK Cells, in Adults With AML or MDS. Clinicaltrials.gov. Accessed April 26. 2022. https://clinicaltrials.gov/ct2/show/NCT04623944

NKX019, Intravenous Allogeneic Chimeric Antigen Receptor Natural Killer Cells (CAR NK), in Adults With B-cell Cancers. Clinicaltrials.gov. Accessed April 26, 2022. https://clinicaltrials.gov/ct2/show/NCT05020678