Nature sub-issue: Breakthrough! Sui Jianhua's team at Beijing Institute of Biological Sciences has made important progress in broad-spectrum anticancer drugs targeting CD98!

Since rituximab was approved by the US Food and Drug Administration (FDA) in 1997, monoclonal antibodies against tumor-specific antigens or tumor-associated antigens have developed into one of the most important and successful treatments for multiple cancer types[1].

CD98 is a tumor-associated antigen that is highly expressed in a variety of blood and solid tumors and is considered a potential target for the development of new cancer drugs due to its important role in tumorigenesis, development, and metastasis[2].

However, since CD98 is widely expressed in a variety of tissues, including the brain, spleen, small intestine and blood system, and plays an important role in the proliferation and regeneration of hematopoietic stem cells and progenitor cells [3], how to avoid targeted side effects against normal tissues is an important problem
in the current development of CD98 antibody therapy.

Recently, Sui Jianhua's team from the Beijing Institute of Biological Sciences and the Interdisciplinary Institute of Biomedical Sciences of Tsinghua University published important results on CD98 antibody therapy in Nature Biomedical Engineering [4].

They used the human phage antibody library to screen a highly efficient and broad-spectrum anti-tumor activity anti-CD98 antibody (S1-F4), and developed a pH-dependent variant (H15L54) through molecular structure-based antibody engineering, which effectively improved the tumor specificity, pharmacokinetic profile and safety
of anti-CD98 antibody.

This study is of great significance for advancing the application of anti-CD98 antibody therapy in cancer treatment, and also provides important ideas
for the development of antibodies against similar widely expressed antigens.

Screenshot of the first page of the paper

Next, let's take a look at how Sui Jianhua's team carried out this research
.

The researchers first screened a number of antibodies against the extracellular domain of human CD98 through the human phage antibody library, and analyzed the antitumor activity of HN2-G9, which had the best antigen binding effect
.
The results showed that HN2-G9 had similar antitumor activity compared with the existing anti-CD98 antibody IGN523[4], but its antigen-binding effect was worse (1.
33 μM).

Therefore, the researchers optimized the antigen-binding affinity of HN2-G9 by altering its heavy chain and developed S1-F4 (58.
2 nM).

S1-F4 has shown potent anti-tumor activity
in a variety of xenograft mouse tumor models, including B-cell lymphoma, liver cancer, pancreatic cancer, etc.

Since the expression of CD98 in xenografted mouse tumor models is limited to transplanted tumor cells and the immune system is incomplete, the actual therapeutic effect and potential safety problems of S1-F4 in cancer patients may not be accurately reflected
.

Therefore, the researchers constructed a humanized CD98 mouse model by replacing the extracellular domain of mouse CD98 in the context of C57BL/6 mice, and tested the therapeutic effect
of S1-F4 in T-cell lymphoma, colon cancer, fibrosarcoma and melanoma models.
The results showed that S1-F4 had significant anti-tumor effects
in three tumor models except melanoma.

Anti-CD98 antibody S1-F4 has broad-spectrum antitumor activity

To understand the antitumor mechanism of S1-F4, the researchers analyzed the effects of S1-F4 on CD98-mediated amino acid transport in tumor cells, tumor cell proliferative capacity, as well as complement-dependent cytotoxicity (CDC), antibody-dependent cell-mediated cytotoxicity (ADCC), and antibody-dependent cell-mediated phagocytosis (ADCP)
of S1-F4.

The results showed that S1-F4 did not change CD98-mediated amino acid transport, nor did it affect the cell proliferation of HepG2 and HCT-B cell lines, which indicated that the tumor cell killing ability of S1-F4 was not achieved by affecting the physiological function of
CD98.

In vitro CDC, ADCC and ADCP experiments showed that S1-F4 could cause ADCC and ADCP, and this phenomenon disappeared in the S1-F4 variant (S1-F4DANA) where the FcR binding domain was disrupted, indicating that the antitumor effect of S1-F4 depends on Fc-FcR-mediated ADCC and ADCP
.
This conclusion has also been validated
in animal models.

The antitumor effect of S1-F4 relies on Fc-FcR-mediated ADCC and ADCP

Common immune cells expressing FcR are macrophages, dendritic cells (DCs), natural killer cells (NK), and neutrophils
.
To determine which cells are involved in the anti-tumor function of S1-F4, the researchers first eliminated these immune cells separately with specific depleting antibodies, and then evaluated
the anti-tumor function of S1-F4.

The results showed that after clearing macrophages or DCs, the antitumor function of S1-F4 was completely or partially lost
.
The results illustrate that the anti-tumor function of S1-F4 is dependent on macrophages and DCs
.
Further analysis revealed that CD8+ T cells also play an important role in the antitumor effect of S1-F4, and that S1-F4 therapy can induce the generation of antigen-specific long-term immune memory
.

The antitumor effect of S1-F4 relies on macrophages, dendritic cells, and CD8+ T cells

As mentioned above, S1-F4 is not effective in humanized CD98 mouse melanoma models, which is different from its results
in C57BL/6 mouse melanoma models (only tumors expressing human CD98).
Since the difference between the two mice lies mainly in the expression range of human CD98, the researchers speculate that CD98 in normal human tissues may affect the pharmacokinetics (PK) of S1-F4, affecting the anti-tumor effect
of S1-F4.
The results of PK experiments showed that S1-F4 did decay faster (half-life 29.
5 hours) in humanized CD98 mice than C57BL/6 mice (half-life 341.
5 hours).

In order to overcome the potential problems caused by the binding of S1-F4 to CD98 in normal tissues, the researchers plan to modify S1-F4 into a pH-dependent antibody
with low binding force with CD98 under normal tissue pH conditions (pH 7.
2-7.
5) and high binding capacity under tumor weak acidic pH conditions (pH 6.
5-6.
9) based on the unique acidic microenvironment of tumors.

To this end, the researchers analyzed the crystal structure of S1-F4 in the state of binding to the CD98 extracellular domain, and found the key amino acid Y97
, which may increase the pH-dependent binding capacity of antibodies.
By mutating this tyrosine to glutamic acid (Y97E) on S1-F4 and further mutating the HCDR3, LCDR1, and HCDR2 domains of S1-F4, the researchers obtained the pH-dependent anti-CD98 antibody H15L54
.
In vitro attenuation experiments showed that H15L54 had a longer
half-life than S1-F4.

H15L54 has better pH-dependent and pharmacokinetic properties than S1-F4

Compared with S1-F4, the pharmacokinetic properties of H15L54 in serum were significantly improved compared with S1-F4, and the distribution in mice was more concentrated in tumor tissues, and the potential targeted side effects were significantly improved
.
At low doses (5 mg/kg), the antitumor effect of S1-F4 is weakened, while H15L54 still has a strong antitumor effect
.

H15L54 has better tumor selectivity and anti-tumor effect

Overall, Sui Jianhua's team developed an anti-CD98 antibody S1-F4 with broad-spectrum antitumor activity and analyzed its mechanism
of action.
Through the rational design based on crystal structure, the team further developed pH-dependent H15L54, which effectively increased the tumor specificity of anti-CD98 antibody, improved the antibody attenuation problem and potential toxicity, and laid the foundation
for the clinical application of anti-CD98 antibody in the future.

References:

[1] Scott, Andrew M.
, Jedd D.
Wolchok, and Lloyd J.
Old.
"Antibody therapy of cancer.
" Nature reviews cancer 12.
4 (2012): 278-287.

[2] Cantor, Joseph M.
, and Mark H.
Ginsberg.
"CD98 at the crossroads of adaptive immunity and cancer.
" Journal of cell science 125.
6 (2012): 1373-1382.

[3] Hayes, Gregory M.
, et al.
"Antitumor activity of an anti‐CD 98 antibody.
" International journal of cancer 137.
3 (2015): 710-720.

[4] Tian, Xinxin, et al.
"An anti-CD98 antibody displaying pH-dependent Fc-mediated tumour-specific activity against multiple cancers in CD98-humanized mice.
" Nature Biomedical Engineering (2022): 1-16.