Combining CD38 antibody with CD47 blockade is a promising strategy for treating hematologic malignancies expressing CD38

Song Li, Dianze Chen, Yannan Yang, Huiqin Guo, Dandan Liu, Nana Sun, Xing Bai, Kaili Wang, Tengfei Li, Guanghui Li, Chunmei Yang, Wei Zhang, Li Zhang, Gui Zhao, Liang Peng, Sijin Liu, Xiaoping Tu, Ruliang Zhang, Wenzhi TianFrontiers in Immunology 2024 · DOI

This research paper, published in Frontiers in Immunology, details the development of a novel bispecific antibody (BsAb) designed to treat blood cancers by simultaneously hitting two targets: CD38 and CD47. By using a clever "affinity-balancing" strategy, the researchers created a drug that is potent against tumors but significantly safer for healthy red blood cells than previous attempts at targeting these pathways.

TL;DR

Researchers developed a "mAb-trap" bispecific antibody that targets CD38 and blocks the CD47 "don't eat me" signal on cancer cells. By engineering the drug to bind strongly to CD38 but weakly to CD47, they achieved complete tumor eradication in mouse models without the toxic side effects on red blood cells typically seen with CD47 inhibitors.

What Problem Does This Paper Address?

Hematologic malignancies—like multiple myeloma (MM) and various lymphomas—often express the protein CD38. While drugs like daratumumab target CD38 successfully, patients frequently develop resistance.

Another promising target is CD47, a protein cancer cells use as a "don't eat me" signal to evade the immune system. However, CD47 is also found on healthy red blood cells (RBCs). Most anti-CD47 drugs are "too good" at their job; they bind to RBCs, causing severe anemia and "antigen sink" effects where the drug is soaked up by healthy cells before it can reach the tumor. The challenge is creating a drug that blocks CD47 only when it is on a cancer cell.

Methods

The authors utilized a proprietary 2 + 2 “mAb-trap” platform. This structure consists of:

  • Two anti-CD38 arms: Derived from high-affinity antibodies screened via hybridoma technology.
  • Two SIRPαD1 domains: These act as the "trap" to block CD47. Crucially, these were engineered to have a lower affinity for CD47.
  • Engineered Fc Region: The researchers introduced specific mutations (E333A, S298A, K334A) into the antibody's "tail" (Fc region) to enhance its ability to recruit immune cells to kill the tumor.

The team screened nine anti-CD38 candidates, narrowing them down to five leads (4A8, 12C10, 26B4, 35G5, and 65A7) for bispecific construction. They then tested these in various in vitro assays (cell killing, enzymatic inhibition) and in vivo using NCI-H929 (multiple myeloma) xenograft mouse models.

Datasets

The study relied on a battery of established cancer cell lines and primary human samples:

  • Tumor Cell Lines: Raji (lymphoma), Daudi (lymphoma), NCI-H929 (multiple myeloma), L-1236 (Hodgkin’s lymphoma), and others.
  • Control Lines: CHO cells engineered to express human or monkey CD38.
  • Human Samples: Peripheral blood from healthy donors was used to test for RBC binding and agglutination (clumping).
  • Animal Models: CB17-SCID mice were used for the xenograft efficacy studies.

Key Insights and Ideas

  • Avidity-Driven Selectivity: This is the "aha moment" of the paper. By making the CD47-binding arm weak, the drug won't stick to a cell that only has CD47 (like an RBC). However, if a cell has both CD38 and CD47 (like a tumor cell), the strong CD38 binding anchors the drug, allowing the weak CD47 arm to engage. This creates a "zip-lock" effect exclusive to the cancer.
  • Multi-Modal Killing: The BsAbs don't just block signals; they kill via four distinct pathways:
  • ADCC: Recruiting NK cells to blast the tumor.
  • ADCP: Inducing macrophages to "eat" the tumor.
  • CDC: Activating the complement system to poke holes in the tumor.
  • Direct Apoptosis: Triggering the cell's internal self-destruct sequence.
  • Enzymatic Inhibition: CD38 is an enzyme that produces immunosuppressive metabolites. One lead, IMM5605-26B4, was particularly effective at shutting down this enzymatic activity, potentially making the tumor's environment less hostile to the immune system.

Key Results

  • Safety Profile: Unlike the benchmark anti-CD47 antibody (Hu5F9), the new BsAbs showed zero binding to RBCs and did not induce RBC clumping, even at high concentrations (100 nM).
  • Superior Potency: The BsAbs were significantly more effective at blocking the CD47/SIRPα interaction on tumor cells than SIRPα-Fc fusion proteins alone.
  • Tumor Eradication: In the NCI-H929 mouse model, the lead candidate IMM5605-12C10 (at a dose of 1 mg/kg) eradicated all established tumors in 100% of the mice. This outperformed the combination of separate anti-CD38 and anti-CD47 therapies.

Limitations and Future Work

  • Preclinical Scope: While the results in SCID mice are impressive, these mice lack a full adaptive immune system. Human clinical trials are the necessary next step to see if the safety and efficacy translate to patients.
  • Monkey Cross-Reactivity: Only some of the leads (like 65A7 and 12C10) showed binding to cynomolgus monkey CD38, which is a requirement for standard toxicology studies.
  • Resistance Mechanisms: The paper does not explore how tumors might eventually adapt to this dual-targeting (e.g., through target shedding or down-regulation).

Overall, this paper presents a compelling case for "low-affinity" CD47 targeting as a way to unlock the potential of immunotherapy while sidestepping the hematologic toxicity that has plagued the field for years.

Summary updated

Abstract

A panel of BsAbs targeting CD38 and CD47 developed based on the "mAb-tarp" platform showed potent tumor-killing ability <i>in vitro</i> and <i>in vivo</i>. As BsAbs had lower affinity for binding to CD47, higher affinity for binding to CD38, no affinity for binding to RBCs, and did not induce RBC agglutination, we concluded that CD38/CD47 BsAbs are safe and have a satisfactory tolerability profile.
BibTeX
@inproceedings{li2024combining,
  author = {Song Li and Dianze Chen and Yannan Yang and Huiqin Guo and Dandan Liu and Nana Sun and Xing Bai and Kaili Wang and Tengfei Li and Guanghui Li and Chunmei Yang and Wei Zhang and Li Zhang and Gui Zhao and Liang Peng and Sijin Liu and Xiaoping Tu and Ruliang Zhang and Wenzhi Tian},
  title = {{Combining CD38 antibody with CD47 blockade is a promising strategy for treating hematologic malignancies expressing CD38}},
  booktitle = {Frontiers in Immunology},
  year = {2024},
  doi = {10.3389/fimmu.2024.1398508},
}
Figure