ADC payloads are the ultra-potent cytotoxic (or increasingly, non-cytotoxic) small molecules that do the actual cell-killing once delivered by the antibody, and they fall into three dominant mechanistic classes plus a growing set of emerging categories.
Microtubule Inhibitors (Tubulin-Targeting Agents)
This is historically the most widely used payload class, present in the majority of approved ADCs, working by binding tubulin and blocking microtubule polymerization or dynamics, causing mitotic arrest at G2/M phase and apoptosis. Key subtypes include:
-
Auristatins (MMAE, MMAF): synthetic derivatives of the natural product dolastatin 10; MMAE is used in Adcetris, Padcev, and Polivy, and is typically paired with a cleavable Val-Cit linker enabling bystander killing.
-
Maytansinoids (DM1, DM4): natural-product-derived agents used in Kadcyla (DM1, non-cleavable linker) and Elahere (DM4, cleavable linker).
-
Tubulysins and eribulin/halichondrin derivatives: newer, less clinically established microtubule disruptors still being explored in ADC pipelines.
Potency is in the sub-nanomolar range (IC50), and the main clinical limitations are peripheral neuropathy and susceptibility to multidrug-resistance efflux pumps.
DNA-Damaging Agents
These payloads directly damage genomic DNA rather than disrupting cell division machinery, and reach picomolar potency — roughly 1,000x more potent than microtubule inhibitors — making them useful for targets with low antigen density. Subtypes include calicheamicins (double-strand DNA breaks; used in Mylotarg and Besponsa), pyrrolobenzodiazepine (PBD) dimers (DNA crosslinking; used in Zynlonta), and duocarmycins (DNA alkylation, non-cleavable linker delivery). The trade-off is a narrow therapeutic window and off-target genotoxicity risk, and this class has historically struggled with poor plasma stability and high aggregation tendency due to hydrophobic, planar chemical structures.
Topoisomerase I Inhibitors
The newest and fastest-growing major class, based on camptothecin-derived scaffolds, works by stabilizing the topoisomerase I-DNA cleavage complex and inducing double-strand DNA breaks. Key examples are SN-38 (active metabolite of irinotecan, used in Trodelvy) and DXd/exatecan derivatives (used in Enhertu and Datroway). This class is prized for its strong bystander effect and effectiveness in tumors with heterogeneous antigen expression, though it carries a rare but serious interstitial lung disease (ILD) risk.
Comparison Table
Emerging, Non-Traditional Payload Classes
The field is actively diversifying beyond these three legacy classes to overcome resistance and expand into non-oncology-style mechanisms:
-
Immunostimulatory agents: STING and TLR7/8/9 agonists that trigger anti-tumor immune responses rather than direct cytotoxicity.
-
RNA polymerase II inhibitors: alpha-amanitin-class payloads, effective even against non-dividing cells.
-
Anti-apoptotic protein inhibitors, splicing modulators, proteasome inhibitors, and NAMPT inhibitors: newer mechanism-based payloads in earlier development stages.
-
Anthracyclines (e.g., PNU): re-emerging DNA-intercalating payloads being re-optimized for better physicochemical properties.
Why Payload Choice Matters So Much
Payload selection is inseparable from linker design — a hydrophobic, highly potent payload like PBD or duocarmycin often forces a lower DAR and more careful linker engineering to avoid aggregation, whereas a moderately potent but bystander-capable payload like DXd allows a much higher DAR (8) when paired with a hydrophilic linker. This interdependency between payload class, potency, hydrophobicity, and linker chemistry is exactly why Enhertu's topoisomerase I/hydrophilic-linker combination has outperformed older microtubule-inhibitor-based designs, and why current R&D increasingly treats payload and linker as a single co-optimized system rather than independent components.
Payload Potency: The Extreme End of Pharmacology
ADC payloads occupy a unique potency niche — they must be roughly 100–1,000 times more potent than conventional chemotherapy because only a tiny amount actually reaches the tumor after systemic dilution and antibody-mediated delivery. Research shows sub-nanomolar potency (IC50) is essentially a prerequisite: payloads with activity weaker than 1 nM are far more likely to produce inactive ADCs, while free-payload IC50s below 1 nM reliably translate into effective conjugates. This is a hard design threshold, not a preference.
Potency Ranked by Payload Class
Actual measured potency varies dramatically across payload families, spanning three orders of magnitude:
Interestingly, ADC potency doesn't map linearly to free-payload potency — Enhertu's DXd payload is only moderately potent as a free drug (1.7–9.0 nM) but the conjugated ADC achieves a striking 43-fold potency boost (down to 0.04–0.16 nM) once delivered via targeted internalization, illustrating how linker/DAR engineering can compensate for using a "gentler" payload. This matters clinically: gentler payloads that still hit the sub-nanomolar bar (like DXd) tend to have better tolerability profiles than picomolar DNA-damaging agents, which is part of why topoisomerase inhibitors have become the preferred modern payload class.
Payload Resistance: Efflux Pumps Are the Dominant Mechanism
The best-characterized payload-level resistance mechanism is drug efflux via ATP-binding cassette (ABC) transporters, detected in an estimated 30–40% of ADC-resistant tumors. The two major players are:
-
P-glycoprotein (P-gp/MDR1/ABCB1): actively exports hydrophobic payloads including MMAE, DM1, and PBD dimers out of the cell before they can act.
-
BCRP (ABCG2): specifically implicated in resistance to SN-38 and to DXd, the payload used in Enhertu.
A particularly important nuance for bystander-capable payloads: the same membrane permeability that lets DXd diffuse into neighboring antigen-negative cells also makes it susceptible to being pumped back out by ABCG2 in those same neighboring cells, meaning efflux resistance can develop at a population level even when the originally-targeted antigen-positive cells remain fully drug-sensitive. Strategies being explored to counter this include engineering payloads with reduced P-gp/BCRP affinity, co-administering efflux pump inhibitors, and modulating payload structure to evade transporter recognition altogether.
Physicochemical Engineering of Payloads
Beyond potency and resistance, payload chemists actively tune physicochemical properties (hydrophobicity, charge, alkyl chain length) to balance efficacy against aggregation risk. A notable exatecan-derivative study found ADC aggregation rate correlated directly with alkyl chain length on the payload linker, with 2–4 methylene units representing the optimal window for balancing cytotoxicity against aggregation propensity. Similarly, systematic modification of PBD-dimer payloads showed that introducing basic chemical moieties to improve solubility could backfire — one variant dramatically worsened animal tolerability despite improving physical properties, showing that payload optimization involves real trade-offs rather than a single "better" direction.
Newest Frontier: Expanding the Chemical Space of "Conjugatable" Payloads
A 2026 Nature Communications paper addressed a longstanding limitation: many potent small molecules (drugs with hydroxyl groups) simply couldn't be stably attached using existing linker chemistries. Researchers developed self-immolative phosphoramidate linker units capable of stable serum attachment and traceless release of aliphatic and aromatic alcohol-containing drugs, testing ten structurally diverse cytotoxins and confirming that essentially any payload with sub-nanomolar potency could be converted into an effective ADC using this platform. This kind of platform innovation is significant for the field because it decouples payload selection from the historical constraint of "does this molecule have a linker-compatible reactive group," opening the door to repurposing a much larger universe of existing potent compounds as ADC payloads.
Reviews:
(1) Dumontet, C., Reichert, J. M., Senter, P. D., Lambert, J. M., & Beck, A. (2023). Antibody–drug conjugates come of age in oncology. In Nature Reviews Drug Discovery (Vol. 22, Issue 8, pp. 641–661). Nature Research. https://doi.org/10.1038/s41573-023-00709-2