Acute myeloid leukemia (AML) is a common hematologic malignancy with a high relapse rate, creating an urgent need for new therapeutic agents. T-cell immunoglobulin mucin-3 (TIM-3) is expressed on the surface of leukemia stem cells (LSCs) and blasts in the majority of AML patients but is not expressed on normal hematopoietic stem cells (HSCs). Zou et al. developed an antibody-drug conjugate (ADC), designated KK2845, composed of a fully human anti-TIM-3 IgG1 antibody, a valine-alanine linker, and a potent DNA-crosslinking pyrrolobenzodiazepine (PBD) dimer payload, SG3199. KK2845 demonstrated potent cytotoxicity against AML cells both in vitro and in vivo. The cytotoxicity of KK2845 against AML cells was nearly comparable to that of CD33-ADC, an anti-CD33 antibody conjugated to a PBD dimer which has shown high response rates in clinical studies. Importantly, compared to CD33-ADC, KK2845 exhibited less cytotoxicity against normal human bone marrow cells. Intravenous infusion of KK2845 in cynomolgus monkeys showed a favorable pharmacokinetic profile. These findings suggest that KK2845 may represent a novel ADC for the treatment of AML.
Main Research Content
KK2845 comprises a fully human IgG1 antibody, TIM-3Ab, which targets human TIM-3. SG3249 is conjugated randomly to the cysteine residues of TIM-3Ab via thiol-maleimide coupling. SG3249 consists of a PEG-8 spacer, a valine-alanine linker, and the PBD dimer warhead SG3199 (Figure 1). The valine-alanine linker is cleaved by cathepsins in lysosomes to release the payload.

Figure 1: Structure of KK2845 (KK2845 is an ADC composed of an anti-TIM-3 fully human IgG1 antibody, a valine-alanine linker, and the potent DNA-crosslinking PBD dimer SG3199)
The study utilized Kasumi-3 cells, an AML cell line co-expressing TIM-3 and CD33. When Kasumi-3 cells were incubated with KK2845-Alexa488 on ice for 30 minutes, localization of KK2845-Alexa488 on the cell surface was restricted. However, when Kasumi-3 cells were incubated with KK2845-Alexa488 at 37°C for 2 or 24 hours, KK2845-Alexa488 was internalized into the cells, and a portion of the intracellular signal colocalized with LysoTracker signal (Figure 2). After 24 hours of incubation in Kasumi-3 cells, the internalization rate of KK2845 was nearly comparable to that of CD33-ADC.

Figure 2: Localization and intracellular trafficking of KK2845 (KK2845-Alexa488 was incubated with Kasumi-3 cells on ice (0 hour), or at 37°C for 2 or 24 hours. AlexaFluor488 signal is shown in green; LysoTracker signal is shown in red. DAPI signal is shown in blue; yellow signal indicated by arrows represents colocalization of KK2845-Alexa488 and Lysotracker).
Primary AML cells were incubated with KM8047-SG3249 (an isotype control ADC), KK2845, or CD33-ADC at 37°C for 4 days. The number of viable CD34+, CD34+CD38-, and CD34- cells was counted by flow cytometry. The percentage of viable cells in each treatment group was calculated relative to the untreated control group (set as 100%). The study tested the cytotoxicity of KK2845 and CD33-ADC against three different batches of primary AML cells: AML818BM (Figures 3A and 3B), AML817BM (Figure 3C), and AML942IIBM (Figure 3D). In AML818BM, CD34+ cells, which contain LSCs and leukemic progenitors, were the predominant population. In AML817BM, CD34- blasts were the major component. In AML942IIBM, CD34+ leukemic progenitor cells constituted the main proportion. Viable CD34+ cells, CD34+CD38- cells (enriched for LSCs), and CD34- cells were enumerated after 4 days of treatment with KM8047-SG3249, KK2845, or CD33-ADC.


Figure 3: Cytotoxicity of KK2845 against primary AML cells
The results showed that in AML818BM, compared to KM8047-SG3249, both KK2845 and CD33-ADC killed CD34+ and CD34+CD38- cells in a dose-dependent manner starting at 1 or 10 ng/mL (Figures 3A, 3B). At a concentration of 1 ng/mL, the percentage of viable CD34+ and CD34+CD38- cells after KK2845 treatment was lower than that after CD33-ADC treatment. At concentrations of 10 and 30 ng/mL, the percentage of viable CD34+ and CD34+CD38- cells after KK2845 treatment was nearly comparable to that after CD33-ADC treatment. In AML817BM, compared to KM8047-SG3249, both KK2845 and CD33-ADC killed viable CD34- cells in a dose-dependent manner starting at 1 ng/mL (Figure 3C). At all concentrations tested, the percentage of viable CD34- cells after KK2845 treatment was nearly comparable to that after CD33-ADC treatment. In AML942IIBM, compared to KM8047-SG3249, both KK2845 and CD33-ADC killed CD34+ cells in a dose-dependent manner starting at 3 ng/mL (Figure 3D). At all concentrations tested, the percentage of viable CD34+ cells after KK2845 treatment was nearly comparable to that after CD33-ADC treatment. KK2845 showed varying sensitivity across different batches of primary AML cells, with IC50 values ranging from approximately 1-20 ng/mL. The cytotoxicity of KK2845 and CD33-ADC was nearly comparable.
TIM-3 is expressed only in a subset of granulocyte/macrophage progenitors but not in normal HSCs, whereas CD33 is broadly expressed in both normal HSCs and myeloid progenitors. To compare the hematotoxicity between KK2845 and CD33-ADC in vitro, the study cultured human bone marrow-derived CD34+ cells in the presence of KM8047-SG3249, KK2845, or CD33-ADC at concentrations ranging from 0.1 to 1000 ng/mL (Figure 4). CD34+ cells were induced in vitro to generate myeloid progenitors, and hematotoxicity was monitored.

Figure 4: Cytotoxicity of KK2845 against normal human bone marrow cells
The results showed that treatment with KK2845 at concentrations up to 1000 ng/mL did not affect the viability of CD34+ cells, whereas treatment with CD33-ADC at 100 and 1000 ng/mL inhibited the viability of CD34+ cells (Figure 4A). The viability of myeloid progenitors was unaffected by KK2845 at concentrations up to 1000 ng/mL, but their activity was inhibited by CD33-ADC at concentrations ≥ 10 ng/mL (Figure 4B).
This study developed and characterized KK2845, a novel antibody-drug conjugate (ADC) targeting TIM-3. Nonclinical studies demonstrated that KK2845 possesses potent anti-leukemia activity through an ADC-based mechanism, without affecting normal hematopoiesis. These data support the initiation of a Phase I clinical trial (jRCT2011240008) in patients with relapsed or refractory AML and also provide insights and impetus for the future development of ADC drugs for refractory diseases.
Tek Biotech (Tianjin) Co., Ltd. has established a comprehensive targeted antibody discovery platform based on phage display technology and yeast display technology. We are committed to providing high-quality technical services to scientists worldwide, including targeted antibody discovery, antibody humanization, ADC design and conjugation, cell killing assays, and animal model experiments. Our services aim to provide robust support for clients' research projects and the development of monoclonal antibody therapeutics for refractory diseases.
References
[1] Zou, J., Kinosada, H., Takayanagi, Si. et al. KK2845, a PBD dimer-containing antibody-drug conjugate targeting TIM-3-expressing AML. Leukemia (2025).
![]() | A PBD Dimer-Containing Antibody-Drug Conjugate (KK2845) Targeting TIM-3-Expressing AML |
![]() | A Novel Approach for Cancer Therapy - Modulating TCR Complex Recruitment to the T Cell Antigen Coupler (TAC) Enhances TAC-T Cell Function |
![]() | Development of Therapeutic Monoclonal Antibodies for Disease Treatment |
![]() | Common Problems and Solutions of Prokaryotic Expression of Recombinant Protein |
![]() | Neutralizing Monoclonal Antibody Development Service FAQ |
![]() | A Brief Discussion on Prokaryotic Expression of Recombinant Antibodies |
![]() | Common Problems and Solutions for Recombinant Antibody Yeast Expression Experiments |
![]() | Common Problems and Solutions for Recombinant Antibody Expression Experiments in lactation |
To experience the reliable service of Tekbiotech please subscribe:
Antibody Discovery
Antibody Production
Antibody Modification
Contact
Follow us!
Technical Support
©2026Tekbiotech (Tianjin) Co., Ltd