Currently, there are 15 antibody-drug conjugates (ADCs) approved worldwide, among which the term 'bystander effect' often appears in the data of the magic drug DS-8201, making DS-8201 unique from other ADCs. Then, what is the bystander effect of ADCs?
Figure 1. Schematic illustration of the key processes that determine the local distribution and effects of an ADC. Source: https://doi.org/10.3390/molecules25122861
Bystander Effect of ADCs
Antibody-drug conjugates (ADCs) are designed to target antigen-expressing (Ag+) cells in tumors. Once treated by Ag+ cells, ADCs can release cytotoxic drug molecules that diffuse from Ag+ cells into neighboring antigen-negative (Ag-) cells, thereby inducing their cytotoxicity. This additional efficacy of ADCs on Ag- cells in the presence of Ag+ cells is known as the 'bystander effect'. However, not all ADCs have bystander effects, which are often related to the biochemical properties of linkers and cytotoxins.
Figure 2. Schematic narrating the mechanism of bystander effect induced by ADCs in a heterogeneous population of antigen-positive (Ag+) and antigen-negative (Ag−) cells. Source: reference 
Pharmacological Determinants Of The Bystander Effect
Several studies have tried to identify the major pharmacological determinant of the bystander killing efect of ADCs, let's analyze from the following three aspects.
The chemical stability of the antibody-linker is critical, and a balance between ADC specificity and damage to normal tissues is required if a cytotoxic payload is to be released prior to ADC internalization, thereby producing a killing effect on cells surrounding the target antigen-positive cells.
The stability of the linker is also a factor contributing to the bystander effect. Linkers are usually classified into cleavable and non-cleavable linkers.
Cleavable linkers can be further divided into chemically unstable (e.g., PH-sensitive hydrazone bonds, disulfide bonds, etc.) and enzymatically unstable (e.g., peptides cleavable by histone B, β-glucuronidase-sensitive, β-galactosidase-sensitive, etc.).
Non-cleavable linkers are represented by covalent linkers (e.g., thioether bonds, etc.), which usually require intracellular cleavage (e.g., by lysosomal enzymes) to release the payload. Therefore, ADCs with non-cleavable linkers are more stable in plasma and internalization is essential for the release of payload.
Indeed, preclinical evidence highlights the importance of cleavable linkers that can release payload from the Ab fraction and hydrophobic warheads that can diffuse through the cell membrane to neighboring cells.
To achieve the bystander effect, the intracellularly released payload must embody specific biochemical properties in order to exhibit membrane permeability, diffuse through the cell wall, and reach neighboring cells: it should be lipophilic, hydrophobic, and uncharged.
It should be noted that ADCs with a non-cleavable linker, such as T-DM1, have no bystander effect due to its incomplete cleavage, resulting in a positively charged lysine remaining on the payload, which prevents it from crossing the cell membrane.
Figure 3. Schematic representation of ADCs with and without bystander killing effect. Source: reference 
Examples of Bystander Effect: T-DM1 VS. T-DXd
The bystander effect has been more studied in HER2-ADC, and it may partially explain the difference in efficacy and safety of the two trastuzumab-based ADCs - T-DM1 and T-DXd.
T-DM1 uses a non-cleavable linker whose payload DM1 is polar and cannot penetrate the cell membrane. Its effectiveness in the treatment of breast cancer has been demonstrated. However, in HER2+ gastric and colorectal cancers, T-DM1 has failed to replicate the survival benefit in breast cancer. Various hypotheses have been proposed regarding the reasons for the differences in efficacy of T-DM1 in different tumor types, such as the different driving roles played by HER2 in different tumor histological types, the intracellular metabolism of T-DM1 in gastric cancer, and the heterogeneous and dynamic HER2 expression profile.
T-DXd (DS-8201) employs a cleavable linker and the payload DXd can penetrate cell membranes. T-DXd is approved for the treatment of HER2+ breast and gastric cancers and has shown some antitumor activity in colorectal cancer. In addition, in breast cancer patients with low HER2 expression, T-DM1 has shown poor efficacy, while T-DXd has shown benefit.
Figure 4. T-DXd and T-DM1 and their cleaved toxin structures (Anti-HER2-DXd (2) differs from T-DXd in that the cleaved toxin DXd (2) has a positively charged amino group), source: doi: 10.1111/cas.12966
HER2 expression is heterogeneous, especially in gastric cancer. Whether the difference in efficacy between T-DM1 and T-DXd is attributable to a bystander effect needs to be clarified by additional studies. However, this may be one of the clinical evidence supporting the existence of a bystander effect.
It should also be noted that payloads spreading near the target tumor cells or entering the circulation may increase toxicity. The highly potent chemical drug payloads used in currently marketed ADCs, which show cytotoxicity at sub-nanomolar concentrations, are not suitable for systemic delivery as free drugs. Available clinical data suggest that most of the new ADCs have higher adverse effects compared to T-DM1.
Figure 5. Incidence of HER2-ADC-associated interstitial lung disease (ILD). Source: JAMA Oncol. 2021; 7(12): 1873-1881.
A Cautious and Objective View of the Bystander Effect
It was found that a fraction of ADCs had more potent killing activity despite the absence of bystander effects, such as ARX788, an anti-Her2-p-acetylphenylalanine- MMAF conjugate with a DAR of 1.9. ARX788 exhibits high serum stability and a relatively long half-life of 12.5 days in mice. This stability is thought to reduce the shedding of payload in circulation, thereby improving safety and cytotoxicity delivery to cancer cells. At the same time, the design of this conjugate prevents the induction of bystander effects. Nevertheless, the compound showed an initial but impressive response rate of 74% in T-DM1-pretreated patients with HER2-positive advanced breast cancer, and preclinical evidence suggests that it may also be active against HER2 low-expressing breast cancers.
Figure 6. ARX788, T-DM1 and isotype control ADC were used as tumor suppressor control in HER2 high expression PDX model, ARX788 showed stronger anti-tumor activity than T-DM1.
Figure 7. 3 mg/kg ARX788 significantly inhibited tumor growth, while treatment with the same dose of T-DM1 was ineffective. Therefore, ARX788 had significant antitumor activity in the PDX model of T-DM1-insensitive breast cancer.
This suggests that some caution is needed in attributing the sole cause of the current development of novel ADCs to bystander effects. In fact, these are complex drugs whose activity is determined by multiple factors, not just the linkage technology or the membrane permeability of the payload.
ADCs represent a class of drugs with high antitumor activity, and the existence of a "bystander effect" has received widespread attention. However, there are also preclinical or clinical data showing that more stable ADCs without bystander effects can also show efficacy in tumors with low expression of target antigens. Therefore, more research is needed to answer the question of how to balance the benefit of bystander effect with off-target toxicity, and whether it is a necessary feature to achieve an ideal ADC, or whether it is simply a factor to be considered in the structural design of an ADC.
Huateng Pharma is dedicated to being your most reliable partner to provide chemical synthesis and high-quality PEG linkers for ADC drugs. We are committed to promoting the progress of your ADC discovery and development projects.
 Giugliano F, Corti C, Tarantino P, Michelini F, Curigliano G. Bystander effect of antibody-drug conjugates: fact or fiction? Curr Oncol Rep. 2022 Jul;24(7):809-817.
 Singh AP, Sharma S, Shah DK. Quantitative characterization of in vitro bystander effect of antibody-drug conjugates. J Pharmacokinet Pharmacodyn. 2016 Dec;43(6):567-582. doi: 10.1007/s10928-016-9495-8. Epub 2016 Sep 26. PMID: 27670282; PMCID: PMC5112145.
 Skidmore L, Sakamuri S, Knudsen NA,et al. ARX788, a Site-specific Anti-HER2 Antibody-Drug Conjugate, Demonstrates Potent and Selective Activity in HER2-low and T-DM1-resistant Breast and Gastric Cancers. Mol Cancer Ther. 2020 Sep;19(9):1833-1843.
Approved Antibody–Drug Conjugates (ADCs) and In Clinical Trials
Directions for Next Generation Antibody-Drug Conjugates
Clinical Toxicity Of Antibody Drug Conjugates
ADC Linker - Development and Challenges
What Are PEG Linkers and Their Applications?
Trodelvy: Any Chance Against Dato-DXd and Enhertu?