Hoskins, K. | Search bioRxiv

bioRxiv2026

Integrative pipeline to profile and target endocrine therapy-insensitive cell populations in ER+ breast cancer

Up to 40% of patients with estrogen receptor positive breast cancer will experience relapse, either while on endocrine therapy (ET) or after ET is completed. A major contributor to ET failure is the presence of ET-insensitive cell populations within tumors. Here, we developed an analytical pipeline to systematically identify and target these populations by integrating single-cell RNA sequencing of ER+ tumors from the FELINE clinical trial with functional validation in a panel of patient-derived xenograft organoid models. We found that ET-insensitive cells are detected in all tumors regardless of clinical response and exhibit higher transcriptional heterogeneity than ET-sensitive populations. Using our pipeline, we identified and validated new therapeutic options that target patient-specific and shared ET-insensitive populations. Our integrated workflow provides a robust platform for identifying and targeting ET-insensitive cells and offers a translational framework to develop precision medicine approaches to improve outcome in breast cancer patients.

cancer biology↗

bioRxiv2026

Container-less Acoustic Levitation Expands Plasma Extracellular Vesicle Proteome Coverage by Mitigating Size-Dependent Peptide Loss

Proteomic profiling of plasma-derived extracellular vesicles (EVs) is limited in part by adsorptive loss of peptides and proteins to container walls during sample preparation. Here we apply an automated, environment-controlled acoustic levitation platform (Levcell) to the tryptic digestion of small EVs (sEVs) isolated from pooled breast cancer patient plasma, and compare it directly with digestion in low-bind microcentrifuge tubes. Across three parallel technical replicates per method, container-less digestion identified 309 {+/-} 22 protein groups versus 261 {+/-} 6 for tubes (+18.4%; Welch t-test p = 0.053), with equivalent or better quantitative reproducibility (median CV 12.0% vs 14.4%). The gain was strongly asymmetric: 66 protein groups were recovered only under levitation while 10 were recovered only in tubes (exact McNemar p = 3 x 10-11). Peptides recovered exclusively by levitation were longer and heavier than those exclusive to tubes (median 14 vs 12 residues, 1611 vs 1358 Da; p < 2 x 10-6; Cliffs {delta} {approx} 0.19-0.20), whereas the total peptide pools were indistinguishable and mean missed-cleavage rates were equivalent (0.276 vs 0.263, p = 0.41), excluding differential digestion efficiency as an explanation. No systematic difference in hydropathy, isoelectric point or hydrophobic residue frequency was detected. The levitation-rescued sub-proteome was enriched for ribosomal, proteasomal, chaperonin and RNA-binding complexes -- canonical sEV luminal cargo (MYC targets 16/19, odds ratio 31.7, q = 5.6 x 10-8) -- and covered 33 of the 100 ExoCarta reference markers versus 22 for tubes, gaining 14 markers while losing three (a single ezrin/moesin/radixin protein group; McNemar p = 9.8 x 10-4). We also report two findings that temper the approach: levitated samples carried an approximately 3.7-fold higher keratin burden, consistent with airborne contamination in an open chamber, and no individual marker showed a significant abundance difference after correction for multiple testing. Container-less processing therefore offers a reproducible gain in sEV proteome coverage attributable to reduced size-dependent peptide loss, provided that contamination control is addressed.

biochemistry↗