Sacituzumab govitecan is a targeted therapeutic that couples an antibody directed against Trop-2, a cell-surface protein abundantly expressed in many epithelial cancers, to SN-38, the active metabolite of the widely used chemotherapy drug irinotecan. Once the antibody binds Trop-2 on the tumor cell surface and the complex is internalized, the cleavable linker releases SN-38, which poisons topoisomerase I and stalls DNA replication. In the pivotal ASCENT trial, the drug nearly doubled median progression-free survival compared with single-agent chemotherapy in metastatic triple-negative breast cancer, and it subsequently earned regulatory approval as a standard of care in the later-line setting. Those landmark results, however, were generated in patient populations in which metaplastic cancers were either absent or represented only a vanishingly small subset, leaving clinicians with little more than scattered case reports to inform decisions for this especially aggressive group.

The Massachusetts General team, led by Shivahamy Maheswaran and Arielle J. Medford with senior author Aditya Bardia, addressed that gap by interrogating their institutional database of patients with metastatic triple-negative breast cancer treated between 2000 and 2025. They identified seventeen patients with metaplastic disease who had received sacituzumab govitecan and compared their outcomes with 133 patients who had non-metaplastic triple-negative tumors treated at the same center over the same quarter-century. Metaplastic breast cancer is defined histologically by the presence of squamous, spindle-cell, or other mesenchymal-like differentiation within an otherwise adenocarcinoma framework, a feature thought to reflect profound lineage plasticity and to contribute to chemotherapy resistance. Prior studies have repeatedly documented poorer responses to neoadjuvant chemotherapy and shorter survival in metaplastic cases, making the question of whether antibody-drug conjugates can overcome this resistance clinically urgent.

The headline result was a numerically shorter median progression-free survival on sacituzumab govitecan for patients with metaplastic disease: 2.6 months versus 6.8 months for non-metaplastic triple-negative breast cancer. Because of the small number of metaplastic cases, this difference did not reach conventional statistical significance, with a p-value of 0.16, and the authors are careful to frame it as a signal rather than a definitive conclusion. Multivariable Cox regression was used to adjust for potential confounders when assessing the association between metaplastic histology and both progression-free and overall survival. Most patients with metaplastic tumors showed limited benefit from the drug as monotherapy, a pattern consistent with the subtype’s documented resistance to cytotoxic and targeted agents alike. For a cancer in which every additional month of disease control matters, the roughly four-month gap in median progression-free survival, even if imprecisely estimated, represents a clinically meaningful shortfall that clinicians must weigh when sequencing therapy.

Yet the study was not uniformly pessimistic. Two patients with metaplastic disease who received sacituzumab govitecan in combination with talazoparib, an oral PARP inhibitor, experienced unusually prolonged progression-free survival. This observation carries mechanistic logic. PARP inhibitors exploit a synthetic-lethal vulnerability in tumors with defective homologous recombination DNA repair, typically driven by germline or somatic BRCA1 or BRCA2 mutations, while topoisomerase I poisoning by SN-38 generates replication-associated DNA lesions that PARP-mediated repair would normally resolve. Preclinical work has shown that combining an anti-Trop-2 antibody-drug conjugate with PARP inhibition can be effective even in BRCA-wild-type triple-negative models, and early-phase studies such as SEASTAR, which paired the PARP inhibitor rucaparib with sacituzumab govitecan, have explored this strategy across tumor types. A sequential topoisomerase I and PARP inhibitor approach has also been proposed as a rational therapeutic rhythm in breast cancer. The two long-responding patients in this cohort, though anecdotal, provide a concrete clinical anchor for that hypothesis in metaplastic disease.

Beyond treatment response, the investigators undertook a systematic annotation of genomic and pathology data from all fifty-eight patients with metaplastic breast cancer in their database, regardless of treatment, to characterize how many harbored actionable biomarkers. The results were striking. Thirty-four of fifty-six patients with evaluable HER2 immunohistochemistry qualified as HER2-low, defined as staining of 1+ or 2+ without amplification, a status that renders tumors eligible for trastuzumab deruxtecan, a different antibody-drug conjugate that demonstrated survival benefit in HER2-low metastatic breast cancer. Eight of forty-five patients tested carried germline BRCA1 or BRCA2 mutations, opening the door to PARP inhibitor monotherapy or platinum-based regimens. Nine of thirty patients with sequencing data had somatic PIK3CA mutations, potentially qualifying them for PI3K-alpha inhibitor therapy. In aggregate, a substantial majority of the metaplastic cohort carried at least one biomarker linked to an approved or investigational targeted therapy, a finding that argues strongly for comprehensive molecular profiling at the time of metastatic diagnosis.

The authors also examined how patients with metaplastic tumors fared on matched targeted therapies when such treatments were actually delivered, and the answer was heterogeneity: progression-free survival on matched therapy varied considerably from patient to patient. This variability echoes a broader theme in precision oncology, namely that the presence of an actionable alteration does not guarantee sensitivity to the corresponding drug, particularly in tumors as genomically chaotic as metaplastic carcinomas, which often combine epithelial and mesenchymal features, high mutational burden, and complex copy-number landscapes. Prior genomic characterizations of paired metaplastic samples have documented marked tumor evolution under therapeutic pressure, suggesting that clonal selection may rapidly erode the relevance of a biomarker identified in an earlier biopsy. Real-time, repeated profiling of plasma and tissue may therefore be necessary to keep treatment aligned with the evolving genome.

The study’s limitations are inherent to its design and to the rarity of the disease it examines. It is a single-institution retrospective cohort, spanning a period during which sequencing practices, biomarker definitions, and treatment standards all changed substantially. Seventeen treated metaplastic patients is a small number by any statistical standard, and the authors themselves note that the absence of a significant p-value should not be mistaken for evidence of equivalence. Not every patient had every biomarker assessed, which explains the differing denominators across the HER2-low, germline BRCA, and PIK3CA analyses. Selection bias in who received sacituzumab govitecan, and in who underwent genomic testing, cannot be excluded. These constraints do not undermine the central observations, but they do mean that the estimates of survival and biomarker frequency carry wide uncertainty, and they underscore why no single center can realistically resolve the clinical questions surrounding a disease this uncommon.

That last point is the study’s clearest message to the field. The authors argue that multi-institutional collaboration and deliberately inclusive trial design are essential to optimize treatment for metaplastic breast cancer, and the structure of their own evidence base illustrates why. Case reports and small case series, which until now constituted essentially the entire literature on antibody-drug conjugates in this subtype, cannot distinguish signal from noise. Registry-based efforts, international consortia, and trial protocols that either stratify for or explicitly enroll metaplastic histology are the only realistic path to generating adequately powered evidence. In the meantime, the practical takeaways for clinicians are concrete: comprehensive biomarker testing, including HER2-low assessment by immunohistochemistry, germline BRCA testing, and broad next-generation sequencing of tissue or plasma, should be standard for every patient with metastatic metaplastic triple-negative breast cancer, and combination strategies pairing sacituzumab govitecan with DNA-damage-response agents such as PARP inhibitors deserve prospective evaluation rather than remaining an anecdotal rescue option.

For patients, the study tempers expectations without closing doors. Sacituzumab govitecan remains a rational choice in later-line metastatic triple-negative breast cancer, but the data suggest that metaplastic histology should prompt heightened vigilance for early progression and a low threshold for pursuing molecular profiling and clinical trial enrollment. The identification of targetable alterations in the majority of characterized tumors reframes this rare cancer not as an untreatable outlier but as a genomically diverse disease in which the right drug for the right alteration, delivered at the right time, may still shift the trajectory. Turning that possibility into reliable outcomes will require exactly what the authors call for: shared data, collaborative trials, and a willingness to design studies around the rarest and most resistant corners of breast cancer biology.

Subject of Research: Response of metastatic metaplastic triple-negative breast cancer to the antibody-drug conjugate sacituzumab govitecan and the frequency of actionable genomic biomarkers

Article Title: Metastatic metaplastic triple negative breast cancer response to sacituzumab govitecan and biomarker targetability

Article References: Maheswaran, S., Dedeoglu, A. S., Niemierko, A., Abelman, R. O., Jimenez, R., Rieur, O., Hutchinson, J., Vidula, N., Spring, L. M., Ellisen, L. W., Isakoff, S. J., Wander, S. A., Bardia, A., & Medford, A. J. (2026). Metastatic metaplastic triple negative breast cancer response to sacituzumab govitecan and biomarker targetability. Breast Cancer Research and Treatment, 219(3), Article 26. https://doi.org/10.1007/s10549-026-08079-0

Image Credits: AI Generated

DOI: 10.1007/s10549-026-08079-0

Keywords: metaplastic breast cancer, triple-negative breast cancer, sacituzumab govitecan, antibody-drug conjugate, HER2-low, BRCA mutations, PIK3CA, PARP inhibitor, talazoparib, biomarker testing, progression-free survival, precision oncology

The research, led by Prerna Bansal and supervised by Kamna Srivastava, focused on a signaling protein called VAV2. VAV2 is a guanine nucleotide exchange factor, a molecular switch that activates Rho-family GTPases such as Cdc42, Rac1 and RhoA by exchanging GDP for GTP. These small GTPases regulate the actin cytoskeleton, cell spreading, migration and signaling, and previous work has implicated the Vav family in the uptake of oxidized lipoproteins and in the development of atherosclerotic lesions in mice. Genetic studies have linked VAV2 and its close relative VAV3 to cardiovascular risk factors in humans, and mice lacking Vav2 develop cardiovascular abnormalities, making the protein a biologically plausible, if challenging, drug target.

Targeting VAV2 with conventional drug discovery would be slow and expensive, so the team turned to structure-based drug repurposing, a strategy that asks whether approved or clinically tested medicines already possess the right molecular shape and chemistry to engage a new target. Repurposing offers a well-known advantage: because repurposed candidates have established safety, pharmacokinetic and manufacturing profiles, they can move into new indications far faster than de novo chemical entities. Building on the group’s earlier work that identified repurposed drugs as disruptors of the PCSK9-LDLR axis for lipid lowering, the researchers set out to find compounds capable of interfering with the VAV2 signaling that supports foam-cell formation.

Because a full experimental structure of human VAV2 spanning all of its domains, including the calponin homology, Dbl homology, pleckstrin homology and regulatory regions, was not available, the team constructed a full-length three-dimensional structural model of the protein. Predicted protein structures of this kind, an approach validated by the success of modern deep-learning structure prediction, provide a scaffold for computational ligand screening even when crystallography has not delivered a target structure. The researchers then assembled libraries of clinically relevant compounds, drawing on FDA-approved drug collections and curated chemical databases, and screened them computationally against the VAV2 model to find molecules with predicted binding affinity.

The computational funnel was deliberately stringent. Virtual screening narrowed thousands of candidates, and molecular docking with AutoDock Vina estimated how each ligand might sit within pockets on the VAV2 model. Surviving compounds were then filtered for drug-likeness and predicted pharmacokinetics using established rules for solubility, permeability and molecular descriptors, together with graph-based predictions of absorption, distribution, metabolism, excretion and toxicity. The top-ranked candidates, which included the PARP inhibitor talazoparib, the VEGF receptor inhibitor tivozanib and the calcium channel blocker isradipine, were then subjected to an unusually thorough dynamic assessment: three independent 200-nanosecond molecular dynamics simulations for each protein-ligand complex, run with the GROMACS engine under constant pressure and temperature conditions.

Those simulations allowed the team to ask whether the predicted binding poses were physically stable or merely computational artifacts. Root mean square deviation and radius of gyration measurements tracked the overall convergence and compactness of each complex, while root mean square fluctuation analyses revealed how flexibly individual residues behaved in the bound state. Principal component analysis distilled the dominant collective motions of the protein-ligand systems, and binding free energies were estimated with the molecular mechanics Poisson-Boltzmann surface area method, which decomposes affinity into van der Waals, electrostatic, polar and non-polar solvation contributions. Across these measures, talazoparib, tivozanib and isradipine consistently emerged as the most stable and energetically favorable VAV2 binders among the screened set.

Computation alone, however, cannot establish biological activity, so the investigators moved to cell-based experiments using THP-1 human monocytic cells differentiated into macrophages with phorbol 12-myristate 13-acetate. When these macrophages were exposed to oxidized LDL, they accumulated intracellular neutral lipids, mimicking the foam-cell transformation that occurs in the arterial wall. Treating the cells with talazoparib significantly reduced that oxLDL-induced neutral-lipid accumulation at a concentration that was sub-cytotoxic, meaning the effect could not be explained simply by the drug killing the cells. Cell viability was independently monitored using an MTT assay, an established colorimetric measure of metabolic activity.

A complementary experiment strengthened the case. In a DiI-oxLDL uptake assay, in which oxidized lipoprotein particles are fluorescently labeled so that their binding and internalization can be quantified, all three compounds, talazoparib, tivozanib and isradipine, reduced cell-associated fluorescence relative to vehicle-treated controls. This suggests that each candidate interferes with some step in the handling of oxidized lipoproteins by macrophages, whether at the level of receptor engagement, actin-dependent internalization or downstream signaling through the VAV2 pathway. The consistency between the structural predictions and the cellular readouts is what lends the study its persuasive weight, since purely computational hits frequently fail at this transition.

The authors are careful to frame these results as preliminary. Direct physical binding of any of the three compounds to VAV2 has not been demonstrated biochemically, and no direct inhibition of VAV2’s exchange-factor activity has been measured. The observed reductions in lipid accumulation could, in principle, arise from mechanisms unrelated to VAV2, a possibility that is particularly relevant for isradipine, a calcium channel blocker with well-documented vascular effects, and for talazoparib, whose primary target is the DNA repair enzyme PARP. Confirming the mechanism will require biochemical binding assays, genetic knockdown or knockout experiments, pharmacological controls and, ultimately, in vivo studies in animal models of atherosclerosis.

Even with those caveats, the work illustrates how modern computational pipelines can compress the earliest phase of drug discovery for a notoriously difficult target class. Signaling proteins like VAV2, with their large multidomain architectures and dynamic regulatory conformations, are rarely the first choice for small-molecule drug development, yet the combination of full-length structural modeling, rigorous multi-replica simulation and rapid cellular validation produced three clinically familiar candidates worthy of deeper study. If subsequent biochemical and animal work confirms that talazoparib, tivozanib or isradipine can suppress foam-cell formation through VAV2 or an allied pathway, the field would gain a shortcut to anti-atherosclerotic therapy that bypasses a decade of de novo chemistry. For now, the study stands as a proof of concept that drug repurposing, guided by structure and validated in human macrophages, can surface plausible new weapons against the cellular origins of the world’s deadliest disease.

Subject of Research: Drug repurposing against VAV2 to block macrophage foam-cell formation in atherosclerosis

Article Title: Structure-based drug repurposing and in vitro evaluation of VAV2-associated candidates for suppression of macrophage foam-cell formation in atherosclerosis

Article References: Structure-based drug repurposing and in vitro evaluation of VAV2-associated candidates for suppression of macrophage foam-cell formation in atherosclerosis. (n.d.). https://doi.org/10.1007/s11030-026-11713-0

Image Credits: AI Generated

DOI: 10.1007/s11030-026-11713-0

Keywords: VAV2, atherosclerosis, foam cells, drug repurposing, molecular docking, molecular dynamics, oxLDL, talazoparib, tivozanib, isradipine, macrophages, cardiovascular disease