Menu Close
  • Home
  • The BCRF Legacy Project
Lee/Oesterreich Laboratory · University of Pittsburgh · UPMC Hillman Cancer Center Request organoids
Open resource · Patient-derived organoid biobank

The BCRF ILC Legacy Project

The Leigh Pate Living Biorepository of Invasive Lobular Breast Cancer

Nine comprehensively characterized patient-derived organoid models of invasive lobular breast cancer, along with the molecular data, built to provide researchers the faithful lobular models the field has long lacked, and shared openly with the community.

Request the models Explore the science
BCRF ILC Legacy Project overview: nine validated ILC PDOs characterized across six molecular modalities, in honor of Leigh Pate.
9
Organoid lines
8
Unique patients
8 + 1
ILC + 1 mixed (mDLC)
3 / 6
Primary / metastatic
6
Molecular layers per line
✓
Matched to patient tumor

The Legacy collection brings together nine patient-derived organoid lines of invasive lobular breast cancer, each validated against the defining biology of the disease and profiled across up to six molecular layers, alongside data from the very tumor each line came from.

Black-and-white portrait of Leigh Pate
In memory of

Leigh Pate

This resource exists because of Leigh Pate. Originally from rural North Carolina and later based in Seattle, Leigh was a writer and communications specialist who, after her 2011 diagnosis with lobular breast cancer, became one of the most influential advocates the ILC community has known.

She co-founded the Lobular Breast Cancer Alliance in 2017, shaped its mission, and helped bring lobular breast cancer to the attention of researchers and clinicians worldwide. She led the first white paper defining what was known and unknown about the disease, and was honored as a Susan G. Komen Advocate in Science and Scholar and with the Beth Caldwell Advocacy and Leadership Award.

Though Leigh’s life was cut short in 2022, her gift to the Breast Cancer Research Foundation established this living biorepository and carries her purpose forward. In her honor, the organoid lines are called LIOs, short for Leigh’s Legacy ILC Organoids.

In her own words and images: a life in advocacy. Click any photo to enlarge.
Leigh Pate
Leigh Pate
Leigh Pate
Leigh Pate
Leigh Pate
Why this resource exists

Faithful models for a disease that lacked them

Invasive lobular carcinoma is the most common special histologic subtype of invasive breast cancer, accounting for roughly 10–15% of diagnoses and tens of thousands of new cases each year in the US. Lobular cells lose E-cadherin, the protein that holds cells together, so they grow in single-file strands rather than a lump. That makes invasive lobular carcinoma harder to see on imaging and prone to spreading to unusual places such as the abdomen, ovaries, and gastrointestinal tract.

For decades, invasive lobular carcinoma research has been held back by one gap: too few laboratory models that actually behave like lobular cancer. Leigh Pate understood this, and her gift set out to close it.

Only models that genuinely reproduce lobular biology were kept.

Invasive lobular tumors are slow-growing and notoriously hard to establish in the lab. Building this collection took many attempts, patient growth monitoring, and rigorous validation: every candidate was tested for the molecular hallmarks of lobular cancer, loss of E-cadherin and cytoplasmic p120, and only those that genuinely reproduced that biology were kept. The result is nine validated organoid lines that faithfully carry the biology of the patients they came from.

In total, more than 30 attempts were made to establish ILC organoids from patient tissue. After establishment and growth-phenotype screening, 9 lines were validated by immunofluorescence for the lobular hallmarks, loss of E-cadherin and cytoplasmic p120, to form the final Legacy cohort.

Journey of a Legacy ILC Organoid: ten-step path from patient tissue sample to a shared biorepository resource.
The journey of a Legacy ILC Organoid, from a patient tissue sample to a shared research resource.
What we found · 1 of 5

CDH1 loss is the lobular signature, and the organoids keep it

Loss of the CDH1 gene, which makes the cell-adhesion protein E-cadherin, is the molecular hallmark of invasive lobular breast cancer. Seven of the nine Legacy lines carry a CDH1-truncating mutation, faithfully mirroring the disease they model. The two that keep CDH1 intact include the cohort’s one mixed ductal/lobular line, exactly as expected.

TP53 and PIK3CA mutations each occur in just over half the cohort (56%), frequently co-occurring with CDH1 loss, consistent with the most common co-mutation patterns reported in invasive lobular carcinoma.
CDH1 loss defines the Legacy organoid cohort: oncoprint of CDH1, TP53, and PIK3CA mutation status across nine organoid lines.
What we found · 2 of 5

Losing CDH1 switches off E-cadherin, exactly as invasive lobular carcinoma does

The CDH1 mutation doesn’t just sit there. It has the expected biological effect. The seven organoid lines with a CDH1 mutation show sharply lower E-cadherin activity than the two lines that keep CDH1 intact, the molecular switch that defines lobular breast cancer.

E-cadherin transcript levels (bulk RNA-seq, z-scored across the cohort) separate cleanly by CDH1 genotype. The wild-type group includes the cohort’s mixed lobular/ductal line, which also retains membranous E-cadherin protein by immunofluorescence.
Losing CDH1 switches off E-cadherin: E-cadherin gene activity is much lower in CDH1-mutant organoid lines than in CDH1 wild-type lines.
What we found · 3 of 5

Immunofluorescence confirms the lobular phenotype, line by line

Beyond the genetics, every organoid line was directly imaged for the proteins that define lobular cancer. Across the classical and pleomorphic invasive lobular carcinoma lines, E-cadherin is lost from the cell membrane and p120 shifts into the cytoplasm, the same hallmark pattern seen in patient tumors. The cohort’s one mixed ductal/lobular line, LIO-083L, stands apart, retaining clear membranous E-cadherin signal.

IF for E-cadherin and p120 was performed across all nine organoid lines; loss of membranous E-cadherin with a shift to cytoplasmic p120 was the validation criterion used to confirm the lobular phenotype for inclusion in the final Legacy cohort.
Immunofluorescence for DAPI, E-cadherin, and p120 across all nine Legacy organoid lines, confirming loss of membranous E-cadherin and cytoplasmic p120 redistribution in classical and pleomorphic ILC lines.
What we found · 4 of 5

The organoids stay true to the patient’s tumor

When each organoid is compared with the tumor it came from, the same cancer-driving mutations show up in both. Across five matched patient–organoid pairs, 91% of the tumor’s driver genes are retained in the organoid, including CDH1, in every single case. These models aren’t just similar to the original cancer; they’re faithful working copies of it.

A driver was called retained if mutated in the matched organoid by targeted (MSK-IMPACT), WGS, or WES. One additional patient was excluded from this analysis because its tumor carried only variants of uncertain significance.
Organoids stay true to the patient's tumor: 91 percent of driver genes retained, shown as matched tumor and organoid mutation panels for five patient pairs.
What we found · 5 of 5

A shared landscape of cell states across patients

Looking at the organoids one cell at a time, using combined gene-expression and chromatin-accessibility profiling across more than 56,000 cells, cells settle into nine recurring molecular states that reappear across different patients, suggesting they reflect core biology of invasive lobular cancer.

Joint RNA + ATAC embedding (multi-omic states) resolves nine discrete states spanning 931 to 14,213 cells each, annotated by their dominant transcriptional and chromatin-accessibility signatures.
Multi-omic states across the ILC organoid cohort: joint RNA and ATAC embedding, cells colored by their multi-omic state.

Explore the state-level data

The full multi-omic picture behind the summary above, for researchers who want to go deeper into how these states were defined, where they sit across the cohort, and what drives them. Click any figure to enlarge.

Individual state distributions across the multi-omic embedding, one panel per state.
Where each state sits in the landscapeOne panel per state, from Hypoxia (14,213 cells) to TNFa/NFkB (931 cells)
Shared states show up in every organoid line: composition of cell states per line, with tumor context below.
Shared states show up in every lineEach line’s mix of states, with source-tumor context below
Dot plot showing where each cell state sits across the nine organoid lines.
Where each state sits across the nine linesDot size = share of a line’s cells; color = over/under-representation
Organoids share a core of states and keep private ones: UMAP colored by shared versus line-specific state membership.
A shared core, plus private states21,511 cells in shared states; 35,078 in line-specific ones
What each state is made of: metaprogram activity heatmap across the nine cell states.
What each state is made ofContributory transcriptional programs and their top marker genes
Transcription-factor activity that defines each cell state, from chromVAR motif accessibility.
The transcription factors behind each stateChromatin-accessibility motif activity, state by state
The collection

The Legacy organoid lines

All nine lines at a glance. Each line is accompanied by its full molecular dataset; matched patient-tumor data (whole-genome sequencing, digital pathology) is available for a subset.

LineHistologyOriginER / PR / HER2E-cad / p120Data available (organoid)
LIO-030*ILCPrimary+ / + / −− / cytoWES · bulk RNA-seq · snRNA-seq · snATAC-seq · MSK-IMPACT · IF
LIO-046ILC (pILC)Primary+ / + / −− / cytoWES · bulk RNA-seq · snRNA-seq · snATAC-seq · MSK-IMPACT · IF
LIO-053LILCMetastasis, bone+ / − / −− / cytoWES · bulk RNA-seq · snRNA-seq · snATAC-seq · MSK-IMPACT · IF
LIO-053RILCMetastasis, bone+ / − / −− / cytoWES · bulk RNA-seq · snRNA-seq · snATAC-seq · MSK-IMPACT · IF
LIO-086ILCMetastasis, bone+ / − / −− / cytoWES · bulk RNA-seq · snRNA-seq · snATAC-seq · MSK-IMPACT · IF
LIO-153ILCPrimary+ / + / −− / cytoWES · bulk RNA-seq · snRNA-seq · snATAC-seq · MSK-IMPACT · IF
LIO-223ILCMetastasis, lymph node+ / + / −− / cytoWES · bulk RNA-seq · snRNA-seq · snATAC-seq · MSK-IMPACT · IF
LIO-268ILC (pILC)Metastasis, lymph node+ / + / +− / cytoWES · bulk RNA-seq · snRNA-seq · snATAC-seq · MSK-IMPACT · IF
LIO-083L*mixed (mDLC)Metastasis, ovary+ / + / ++ / memWES · bulk RNA-seq · snRNA-seq · snATAC-seq · MSK-IMPACT · IF
Key: E-cad / p120: −/cyto = E-cadherin negative, p120 cytoplasmic (lobular hallmark); +/mem = E-cadherin positive, membranous. ER / PR / HER2 shown as +/−. LIO-053L and LIO-053R are from the same patient.
* Slow-growing and difficult to maintain in culture.
Use the resource

Access the models and data

Available now

Request the organoids

All nine Legacy organoid lines are available to the research community through the Organoid Research Core (ORC) at the University of Pittsburgh.

Request a line →
Coming soon

Explore the data

A dedicated data portal with full access and visualization tools is in development. This section will be updated with details on access and visualization as they become available.

Team

The people and partners behind the Legacy Project

The Legacy project is led by the Lee/Oesterreich Laboratory at the University of Pittsburgh / UPMC Hillman Cancer Center and Magee-Womens Research Institute, in partnership with Memorial Sloan Kettering Cancer Center and the Institute for Precision Medicine, and supported by the Breast Cancer Research Foundation in honor of Leigh Pate.

University of Pittsburgh 8 contributors
  • Adrian LeePhD
    Women’s Cancer Research Center, UPMC Hillman Cancer Center
    Institute for Precision Medicine, University of Pittsburgh
    Department of Pharmacology & Chemical Biology, University of Pittsburgh, Pittsburgh, PA, USA
  • Brent T. SchlegelBS
    Lee/Oesterreich Laboratory, University of Pittsburgh, Pittsburgh, PA, USA (computational analysis)
    Department of Pharmacology & Chemical Biology, University of Pittsburgh, Pittsburgh, PA, USA
  • Daniel D. BrownPhD
    Institute for Precision Medicine, University of Pittsburgh, Pittsburgh, PA, USA
  • Jagmohan HoodaPhD, MBA
    Women’s Cancer Research Center, University of Pittsburgh Medical Center (UPMC), Hillman Cancer Center
    Magee-Womens Research Institute, Pittsburgh, PA, USA
  • Jian ChenMS
    Women’s Cancer Research Center, University of Pittsburgh Medical Center (UPMC), Hillman Cancer Center
    Magee-Womens Research Institute, Pittsburgh, PA, USA
  • Priscilla F. McAuliffeMD
    Department of Breast Surgical Oncology, University of Pittsburgh, Pittsburgh, PA, USA
  • Rohit BhargavaMD
    Department of Pathology, University of Pittsburgh, Pittsburgh, PA, USA
  • Steffi OesterreichPhD
    Women’s Cancer Research Center, UPMC Hillman Cancer Center
    Department of Pharmacology & Chemical Biology, University of Pittsburgh, Pittsburgh, PA, USA
Memorial Sloan Kettering Cancer Center 6 contributors
  • Britta WeigeltPhD
    Department of Pathology and Laboratory Medicine, Memorial Sloan Kettering Cancer Center, New York, NY, USA
  • Fresia ParejaMD, PhD
    Department of Pathology and Laboratory Medicine, Memorial Sloan Kettering Cancer Center, New York, NY, USA
  • Pier SelenicaMS
    Department of Pathology and Laboratory Medicine, Memorial Sloan Kettering Cancer Center, New York, NY, USA
  • Hunter GreenBS
    Department of Pathology and Laboratory Medicine, Memorial Sloan Kettering Cancer Center, New York, NY, USA
  • Kaitlyn GillBS
    Department of Pathology and Laboratory Medicine, Memorial Sloan Kettering Cancer Center, New York, NY, USA
  • Jorge Reis-FilhoMD, PhD
Presentations

Conference Presentations

Acknowledgements

With Thanks To

In honor of Leigh Pate. Her legacy, our mission, their future.
© Lee/Oesterreich Laboratory · University of Pittsburgh · UPMC Hillman Cancer Center