HER2 targeting redirects a molecular glue degrader toward cyclin K loss

A new preclinical strategy aims to solve a familiar problem in targeted cancer therapy: how to keep a potent drug active where it is wanted and restrained where it is not. In this preprint, investigators built a HER2-directed antibody conjugate around a cyclin K molecular glue degrader, then asked whether receptor targeting could shift the drug’s biology toward antigen-positive breast cancer cells. The answer, at least in cell-based systems, was yes. The conjugate internalized preferentially in HER2-positive cells and drove stronger cyclin K depletion than in HER2-negative cells or in a non-targeting control conjugate.

The rationale is anchored in tumor biology. CDK12 is a transcriptional kinase with a central role in DNA damage response, and it is frequently co-amplified with ERBB2 in breast cancer. CDK12 depends on its obligate partner cyclin K to form an active complex, so degrading cyclin K offers a way to disrupt a transcriptional program that supports malignant growth and genomic stress tolerance. The challenge is that the CDK12/cyclin K axis is not uniquely cancer-specific, which makes broad systemic suppression a poor fit for a highly potent degrader. Antibody-mediated delivery is the obvious countermeasure: confine the payload to cells that express the chosen surface marker, then exploit receptor internalization to bring the degrader inside.

The chemistry matters here because molecular glues do not have the modular linker architecture of PROTACs. A linker can easily ruin the scaffold if it is attached in the wrong place. To avoid that problem, the team designed SAS1001, a cyclin K molecular glue degrader with a solvent-exposed piperazine positioned for conjugation. Free SAS1001 produced potent, ubiquitin-proteasome system-dependent cyclin K degradation and slowed proliferation across several breast cancer cell lines. The attached piperazine was not a cosmetic detail. It was the structural opening that allowed the glue to be turned into a conjugation-ready payload without sacrificing its core degradative activity.

One sentence captures the central advance: linker placement preserved function.

Trastuzumab conjugation sharpened selectivity without abandoning activity

The antibody component was trastuzumab, a clinically established HER2-directed monoclonal antibody. Coupling SAS1001 to trastuzumab generated Tz-SS2009, a HER2-targeted degrader-antibody conjugate that outperformed a matched non-targeting conjugate in HER2-positive settings. Internalization tracked with receptor expression, which is exactly the mechanism the field hopes to harness when it moves beyond simple cytotoxic payloads. Once inside the cell, the conjugate enhanced cyclin K degradation and produced stronger anti-proliferative effects relative to HER2-negative cells and to the isotype control conjugate.

The biological logic is straightforward but important. HER2 is not merely a biomarker for selection; it is also a delivery handle. By using the receptor as an entry point, the conjugate turns an otherwise broadly distributed degrader into a more localized intracellular intervention. In practical terms, this approach aims to widen the therapeutic window by exploiting differential antigen expression rather than relying only on intrinsic selectivity of the small molecule. For a target like cyclin K, where basal activity is tied to essential transcriptional functions, delivery control can matter as much as target affinity.

Whole proteome analysis reinforced the mechanism. Cyclin K depletion was observed, and the downstream protein changes were consistent with the known role of the CDK12/cyclin K complex in transcriptional regulation. The proteomic signature did not read like an unrelated stress response; it aligned with on-target pathway suppression. The data also support the idea that antibody-directed delivery can alter the intracellular geography of degrader activity, concentrating it in cells that present the relevant antigen on their surface.

Why this matters for the degrader field

This work pushes antibody-directed degradation into territory that has been dominated by CRBN-based glues. Most of the field’s practical experience has centered on cereblon-recruiting compounds, where the effector biology is already well mapped. Here, the degrader is different. The scaffold sits in the CR8/DS18 purine series, and the study shows that substitution at a solvent-exposed position can preserve cyclin K degradation even when the payload is carried on a full-length IgG. That is a useful design lesson because it suggests the chemistry of the glue itself can be engineered to tolerate conjugation without collapsing the pharmacology.

The broader implication is not simply that antibodies can carry molecular glues. It is that structural characterization of a glue scaffold can reveal a viable attachment point, making antibody-directed delivery feasible for drug classes that were not built with conjugation in mind. The field has long understood how to attach cytotoxic payloads to antibodies. Extending that logic to a mechanistically distinct degrader requires more than a linker and a target antigen; it requires a scaffold that stays active after modification. SAS1001 appears to meet that requirement in this preclinical system.

The findings stop at the bench, and that boundary matters. No in vivo efficacy, no clinical dosing, and no patient-level safety profile are available here. The preprint also leaves open how much selectivity will survive in complex tissues where HER2 expression is heterogeneous and where internalization kinetics vary by tumor context. Even so, the study establishes a credible route toward antigen-directed delivery of a molecular glue degrader, a strategy that could prove especially useful for targets whose potency is otherwise limited by broad essential expression. The unanswered translational question is not whether cyclin K can be degraded in HER2-positive cells; it is how much additional control antibody delivery can supply once this chemistry leaves cell culture and enters a living organism.


Reporting by Dr. Revathy Mohan V.R., MBBS for Medicine World Foundation, with additional data provided by bioRxiv (Biomedical Preprints).