Journal: Cancer Discovery

Article Title: HER2 Heterogeneous Breast Cancer Models Reveal Novel Therapeutic Targets and Subclonal Dynamics during Evolution to Resistance to HER2-Targeted Therapies

doi: 10.1158/2159-8290.CD-25-1459

Figure Lengend Snippet: Effect of HER2-targeting agents on HER2 hi and HER2 lo populations. A, Dose response of the indicated compounds tested at various concentrations in parental 21NT and HER2 hi and HER2 lo derivatives. Comparison with HER2 hi cells, P values in blue (HER2 lo ) or purple (parental). B, Immunoblot analysis of EGFR, HER2, and downstream signaling pathway components in HER2 hi and HER2 lo cells treated with vehicle or neratinib. C, Flow cytometry analysis of the HER2 lo subpopulation of PDOs from HER2 HET tumor T537 treated with T-DXd or neratinib. D, FISH analysis of ERBB2 and CEP17 copy numbers in control 21NT cells and derivatives resistant to T-DXd (TDR), T-DM1 (TMR), neratinib (NR), or tucatinib (TR). Bar plot depicts quantification of FISH signal in 675–1,421 cells per condition. χ 2 without Yates correction. Scale bars, 25 µm. E, Principal component analysis plot of RNA-seq profiles of HCC1954 control cells, HER2 hi and HER2 lo subpopulations, and resistant derivatives. F, Bar graph showing Euclidean distances between PC1 and PC2 of the indicated RNA-seq samples from E . G, Dose response of the indicated compounds tested at various concentrations in control 21NT cells and resistant derivatives. Comparison with vehicle control cells, P value colored for the different cells as indicated. Data are presented as mean ± SEM, n = 3, two-way ANOVA ( A and G ).

Article Snippet: T-DXd (MedChemExpress HY-138298) and T-DM1 (MedChemExpress HY-P9921) were dissolved in PBS.

Techniques: Comparison, Western Blot, Flow Cytometry, Control, RNA Sequencing

Journal: Cancer Discovery

Article Title: HER2 Heterogeneous Breast Cancer Models Reveal Novel Therapeutic Targets and Subclonal Dynamics during Evolution to Resistance to HER2-Targeted Therapies

doi: 10.1158/2159-8290.CD-25-1459

Figure Lengend Snippet: Subclonal dynamics of HER2 hi and HER2 lo cells during evolution to resistance in mono- or coculture. A, Plots depicting viable cell numbers of HCC1954 HER2 hi and HER2 lo cells grown separately in mono- or together in coculture in the presence of DMSO, T-DXd, T-DM1, neratinib, or tucatinib. B, Heatmap depicting the relative bystander effect of T-DXd in a panel of HER2 lo or TNBC cells in coculture with different HER2 hi cells (mean of n = 3). C, Dose response of DXd tested at various concentrations in different HER2 lo and TNBC cells. D, Schematic outline of the experiment in which HCC1954 HER2 hi and HER2 lo cells are infected with the ClonMapper library, whereas barcoded cells are grown in mono- or coculture in the presence of DMSO, T-DXd, or neratinib. E, Bar plot depicting the number of days per passage in the indicated conditions, including the ratio of HER2 hi and HER2 lo cells in initially 1:1 coculture. F, Shannon entropy measuring barcode diversity over different passages of HER2 hi and HER2 lo cells growing in mono- or coculture in the presence of DMSO, neratinib, or T-DXd. G, Bar plot illustrating the number of barcodes remaining in the top 90% of HER2 hi and HER2 lo cells after 10 passages (DMSO and neratinib) or 4 passages (T-DXd). One-way ANOVA comparing all conditions to DMSO neratinib (HER2 hi , P <0.0001; HER2 lo , P <0.0001) and T-DXd (HER2 hi , P <0.0001; HER2 lo monoculture, P = 0.0070, and coculture P <0.0001). H, Stacked bar plot depicting the relative proportion of barcodes within HER2 hi and HER2 lo cells at final passage. Each barcode in the top 30 is represented by one color across each panel, and gray represents all other barcodes in the population. Data are presented as mean, n = 3, two-way ANOVA ( A , C and F ).

Article Snippet: T-DXd (MedChemExpress HY-138298) and T-DM1 (MedChemExpress HY-P9921) were dissolved in PBS.

Techniques: Infection

Journal: Cancer Discovery

Article Title: HER2 Heterogeneous Breast Cancer Models Reveal Novel Therapeutic Targets and Subclonal Dynamics during Evolution to Resistance to HER2-Targeted Therapies

doi: 10.1158/2159-8290.CD-25-1459

Figure Lengend Snippet: Impact of HER2 heterogeneity on tumorigenesis and treatment responses in vivo . A, Tumor growth of mammary fat pad injection of 21NT homogeneous HER2 hi or HER2 lo and HET 1:1 mix ( n = 6 tumors). B, Immunofluorescence for mCherry (HER2 hi ), GFP (HER2 lo ), and HER2 reveals that HER2 hi and HER2 lo retain their different HER2 levels in HET tumors in vivo . Scale bars, 50 µm. C and D, Stacked bar plot representing the neighborhood composition of HER2 hi and HER2 lo cells in 21NT HER2 HET tumors ( C , n = 5 tumors) or human tumors from the CycIF of ( D , n = 20 patients; ref. ), t test ( P value in red for HER2 hi and blue for HER2 lo ). E, PCNA, mCherry (HER2 hi ), and GFP (HER2 lo ) staining on 21NT HER2 HET tumors. Scale bars, 50 µm. F and G, Quantification of staining for PCNA, a proliferation marker, in HER2 hi and HER2 lo cells in HER2 HET 21NT ( F , n = 5) or human tumors ( G , n = 20 patients; ref. ), t test. H and I, Uniform Manifold Approximation and Projection for Dimension Reduction (UMAP) of scRNA-seq of all cells from human HER2 HET breast tumors T537 and T565. Clusters are colored by samples ( H ) and by gene module scores ( I ). J, CellChat analysis of scRNA-seq data depicting the amount of incoming and outgoing interactions of every cell type. K, Plot presenting the pathways of the overall signals comparing HER2 lo and HER2 hi cell populations. L, Presence of PDPN + stromal cells in the neighborhood of HER2 hi and HER2 lo cells ( n = 20 patients, t test). M, Schematic depicting the in vivo experiment setup in which 21NT homogeneous and HET tumors are treated with neratinib or T-DXd. N, Curve of changes in tumor size under treatment with neratinib or T-DXd. n = 9–10, two-way ANOVA. O, Fluorescence pictures of tumors after 2 weeks of treatment showing HER2 hi cells in red and HER2 lo cells in blue. Scale bars, 1 cm. P, Quantification by flow cytometry of the composition of HET tumors in HER2 hi and HER2 lo cells after 2 weeks of treatment. n = 5; one-way ANOVA. Q, Schematic depicting the survival assay in which tumors are treated for 2 weeks with T-DXd or neratinib and mice are followed until tumor recurrence. R, Tumor growth and recurrence of homogeneous and HET tumors upon T-DXd or neratinib treatment. n = 10–12 tumors, two-way ANOVA. S, Survival curves showing time to tumor volume endpoint in mice treated with indicated treatments, comparing homogeneous and HET tumor-bearing mice. n = 5–6 mice, log-rank (Mantel–Cox) test. T, Quantification by flow cytometry of the composition of HET tumors in HER2 hi and HER2 lo cells after recurrence. n = 5, one-way ANOVA. Data are presented as mean ± SEM ( C , D , F , G , L , N , P , and T ).

Article Snippet: T-DXd (MedChemExpress HY-138298) and T-DM1 (MedChemExpress HY-P9921) were dissolved in PBS.

Techniques: In Vivo, Injection, Immunofluorescence, Staining, Marker, Fluorescence, Flow Cytometry, Clonogenic Cell Survival Assay

Journal: Cancer Discovery

Article Title: HER2 Heterogeneous Breast Cancer Models Reveal Novel Therapeutic Targets and Subclonal Dynamics during Evolution to Resistance to HER2-Targeted Therapies

doi: 10.1158/2159-8290.CD-25-1459

Figure Lengend Snippet: CRISPR KO cellular viability screen with T-DXd in cocultures. A, Schematic showing 21NT HER2 lo cells infected with Cas9 and a whole-genome CRISPR KO library and put in coculture with HER2 hi cells in the presence or absence of T-DXd for 1 month. B, Bubble plot illustrating the enrichment in KEGG pathways of the CRISPR hits linked to resistance (red) and synthetic lethality (blue). C, Scatter plot from MLE analysis depicting hits from the CRISPR screen that changed only in the T-DXd condition linked to resistance or synthetic lethality. D, Histogram showing cell count normalized to the untreated control with or without T-DXd for 10 days upon genetic (sg ABCC1 and sg USP9X ) or pharmacologic (reversan or biricodar, G9 or FT709) inhibition of ABCC1 or USP9X in 21NT HER2 lo cells. E, Treatment with 1 μmol/L of G9, FT709, reversan, or biricodar increased the T-DXd effect in 21NT TDR after 10 days. F, Bar graph showing cell counts normalized to the untreated control with or without DXd upon genetic (sg ABCC1 ) or pharmacologic (1 μmol/L of reversan or biricodar) inhibition of ABCC1 in 21NT HER2 lo cells. G, ABCC1 KO in 21NT HER2 lo cells increases the bystander effect of T-DXd in coculture with HCC1954 HER2 hi cells. H, Schematic outline of the survival assay experiment. I, Graph illustrating the volume of individual tumors during the indicated treatment. n = 8–10 tumors, two-way ANOVA. J, Kaplan–Meier plot showing time to tumor volume endpoint in mice with the indicated treatments [ n = 4–5 mice, log-rank (Mantel–Cox) test]. Data are presented as mean ± SEM, n = 3, one-way ANOVA ( D–G ). NES, normalized enrichment score.

Article Snippet: T-DXd (MedChemExpress HY-138298) and T-DM1 (MedChemExpress HY-P9921) were dissolved in PBS.

Techniques: CRISPR, Infection, Cell Characterization, Control, Inhibition, Clonogenic Cell Survival Assay

Journal: Cancer Discovery

Article Title: HER2 Heterogeneous Breast Cancer Models Reveal Novel Therapeutic Targets and Subclonal Dynamics during Evolution to Resistance to HER2-Targeted Therapies

doi: 10.1158/2159-8290.CD-25-1459

Figure Lengend Snippet: Effect of USP9X inhibition on lysosomal targeting of HER2 and response to T-DXd. A, Immunoblot analysis of HER2 immunoprecipitate (IP) probed for ubiquitin (Ub) shows an increase in HER2 ubiquitination upon USP9X inhibition in the presence of T-DXd in 21NT HER2 lo cells. B, Co-IP experiments show an interaction between USP9X and HER2. C, PLA between HER2 and USP9X in 21NT HER2 lo cells treated with vehicle or T-DXd. Scale bars, 20 μm. Quantification of PLA intensity per cell. Fifteen cells from each replicate. D, Immunoblot analysis of HER2 levels in 21NT HER2 lo cells after treatment with T-DXd upon USP9X inhibition. E and F, Cycloheximide chase for HER2 in 21NT HER2 lo cells upon T-DXd treatment with USP9X inhibition, including the quantification of HER2 levels. G, PLA experiment between HER2 and LAMP1 in 21NT HER2 lo cells treated with T-DXd and FT709. H, Quantification of PLA intensity per cell. Fifteen cells from each replicate. I, Immunoblot of total cell lysate (TCL) and lysosome enriched fraction of 21NT HER2 lo cells treated as in panel G . LAMP1, CTCF, and TUFM were used as lysosomal, nuclear, and mitochondrial marker, respectively. HER2 was run on separate gel, tubulin loading controls are included for both blots. J, USP9X mRNA expression in human tumors before and after T-DM1/pertuzumab neoadjuvant treatment from the NCT02326974 clinical trial ( , ), separated by pCR or no pCR and non-HET or HET. K, Schematic depicting the survival experiment in which 21NT HER2 HET tumors are treated with one dose of T-DXd, G9, or both. L, Tumor growth of individual tumors following indicated treatment. n = 8–10 tumors, two-way ANOVA. M, Survival curves showing time to tumor volume endpoint in mice treated with indicated treatments. n = 4–5 mice, log-rank (Mantel–Cox) test. N, Graphical summary showing the role of USP9X and ABCC1 in sensitivity to T-DXd. Data are presented as mean ± SEM, n = 3, one-way ANOVA ( C , H , and J ) or two-way ANOVA ( E and L ). Figure generated using Biorender. CHX, cycloheximide

Article Snippet: T-DXd (MedChemExpress HY-138298) and T-DM1 (MedChemExpress HY-P9921) were dissolved in PBS.

Techniques: Inhibition, Western Blot, Ubiquitin Proteomics, Co-Immunoprecipitation Assay, Marker, Expressing, Generated

Journal: bioRxiv

Article Title: Fluorescence cross-correlation spectroscopy quantifies affinity, cooperativity, and kinetic stability in ternary protein complexes

doi: 10.64898/2026.07.16.738902

Figure Lengend Snippet: ( A ) left: schematic of HER2 (magenta)/trastuzumab (cyan) (co-)diffusing through the confocal volume and the impact of binding on auto-correlation functions (magenta/cyan) as well as cross-correlation function (purple). (B) Schematic of multiplexed FCCS-measurements in 96-well plate format. ( C) Titration of HER2 to trastuzumab. Error bars depict the standard error of the mean. The data is fit with a quadratic binding model (see methods). ( D ) Titration of HER2 to pertuzumab. Error bars depict the standard error of the mean. The data is fit with a Hill equation (see methods)

Article Snippet: Human HER2 extracellular domain, trastuzumab, and pertuzumab (MedChemExpress, HY-P73094, HY-P9907, HY-P9912) were prepared for fluorescent labeling by exchanging the buffer into phosphate-buffered saline (PBS) at pH 7.4.

Techniques: Binding Assay, Titration

Journal: bioRxiv

Article Title: Fluorescence cross-correlation spectroscopy quantifies affinity, cooperativity, and kinetic stability in ternary protein complexes

doi: 10.64898/2026.07.16.738902

Figure Lengend Snippet: ( A, B ) Schematic of epitope competition (top)/orthogonal binder (bottom) experiment: Preformed complexes of labelled antibodies and targets are subjected to large excesses of unlabeled competitors. Replacement of the labeled antibodies by unlabeled competitors results in loss of signal, orthogonal binders leave the signal unchanged. ( C) Dissociation of trastuzumab (purple) and pertuzumab (cyan). The controls show trastuzumab competing with unlabeled pertuzumab and vice versa. ( D ) Normalized cross-correlation fraction retained after 48 h in presence of capture antibodies trastuzumab (trast), pertuzumab (pert), None (NA) or non-specific control IgG (NIST).

Article Snippet: Human HER2 extracellular domain, trastuzumab, and pertuzumab (MedChemExpress, HY-P73094, HY-P9907, HY-P9912) were prepared for fluorescent labeling by exchanging the buffer into phosphate-buffered saline (PBS) at pH 7.4.

Techniques: Labeling, Control

Journal: bioRxiv

Article Title: Fluorescence cross-correlation spectroscopy quantifies affinity, cooperativity, and kinetic stability in ternary protein complexes

doi: 10.64898/2026.07.16.738902

Figure Lengend Snippet: (A) Schematic of ternary complex formation: As HER2 concentration increases more ternary complex is formed. In large excess of HER2 the antibodies are sequestered and rarely bound to the same molecule of HER2, leading to the hook-effect. (B) Ternary complex formation of Trastuzumab-HER2-Pertuzumab as a function of unlabeled Her2 concentration. The error bars depict the standard error of the mean of three independent measurements. The peak is reached at 10 nM HER2, trastuzumab and pertuzumab respectively. The dashed line is a numeric fit with the cooperativity parameter α held constant at 1, the cyan line is a numeric fit with α freely fit. (C) Cryo-EM structure of pertuzumab-HER2-trastuzumab (pdb-ID 6OGE). pertuzumab (cyan) and trastuzumab (purple) bind separate epitopes on HER2 (magenta). Note that only the Fab-fragments of trastuzumab/pertuzumab were used for this structure. (D) Dissociation rates of trastuzumab (purple)/ pertuzumab (cyan) from the binary/ternary complexes. Complexes were preformed at saturating/maximal conditions (10 nM of each component). Data was fit with single-exponential dissociation models (see methods).

Article Snippet: Human HER2 extracellular domain, trastuzumab, and pertuzumab (MedChemExpress, HY-P73094, HY-P9907, HY-P9912) were prepared for fluorescent labeling by exchanging the buffer into phosphate-buffered saline (PBS) at pH 7.4.

Techniques: Concentration Assay, Cryo-EM Sample Prep