Patient-derived organoid xenografts reveal the multifaceted role of the lncRNA MALAT1 in breast cancer progression

This study demonstrates that targeting the lncRNA MALAT1 with antisense oligonucleotides in patient-derived organoid xenograft models effectively inhibits triple-negative breast cancer progression by altering alternative splicing and reducing metastatic burden through tumor-stroma crosstalk, thereby validating MALAT1 as a promising therapeutic target.

The Gist

Imagine your body is a bustling city, and the cells within it are the citizens. Sometimes, a few citizens go rogue and start building illegal structures, causing chaos. This is cancer. In the case of Triple-Negative Breast Cancer (TNBC), the "rogue" cells are particularly aggressive because they lack the usual "off switches" (receptors) that doctors use to stop them with standard drugs.

This paper is like a detective story where scientists are hunting for a new way to stop these rogue cells. They found a suspect named MALAT1.

The Suspect: MALAT1

Think of MALAT1 not as a protein (the usual building block of cells), but as a mischievous foreman living inside the cell's control center (the nucleus).

• What does it do? It doesn't build things itself; instead, it directs the construction crew. It tells the crew how to assemble the blueprints (RNA) for the cell's machinery.
• The Problem: In cancer cells, this foreman is overworked and shouting orders that make the cells grow faster, spread to other parts of the city (metastasis), and hide from the police (the immune system).

The Detective's Toolkit: Patient-Derived Organoids

In the past, scientists tested drugs on flat, 2D cell cultures in a petri dish. This is like testing a new traffic law on a toy car in a living room; it doesn't tell you how the law works on a real highway with real traffic.

To get a real answer, this team built Patient-Derived Organoids (PDOs).

• The Analogy: Imagine taking a tiny piece of a patient's actual tumor and growing it into a miniature, 3D city in a lab. These mini-cities look, act, and react just like the real tumor inside the patient.
• They even grew these mini-cities inside mice (called PDO-Xenografts) to see how they behave in a living body with a blood supply and immune system. This is the "real highway" test.

The Weapon: The "Silencer" (ASO)

The scientists created a special tool called an Antisense Oligonucleotide (ASO).

• The Analogy: Think of the ASO as a smart noise-canceling headphone or a glitch in the foreman's walkie-talkie. It is designed to find the specific radio frequency the MALAT1 foreman uses to shout orders and jam the signal.
• When the signal is jammed, the foreman goes silent, and the construction crew stops following the bad orders.

What Happened When They Jammed the Signal?

The team tested this "noise-canceling headphone" on three different types of mini-cancer cities. Here is what they discovered:

1. The Blueprint Chaos (Splicing)

• The Metaphor: Cells build proteins by reading blueprints. Sometimes, they need to cut out a page or add a new one to get the right instructions. This is called "splicing."
• The Finding: When MALAT1 was silenced, the blueprints got messy. The crew started keeping pages they should have thrown away (called Intron Retention).
• Why it matters: This created brand new, weird instructions that the cancer cells had never seen before. It's like the construction crew suddenly building a house with a door in the ceiling. These weird structures might confuse the cancer cells or, interestingly, make them easier for the body's immune system to spot and destroy.

2. The Police Force (Immune System)

• The Metaphor: Cancer cells often hire "mercenaries" (immune cells called macrophages) to protect them and help them grow.
• The Finding: When MALAT1 was silenced, the number of these protective mercenaries dropped significantly. The tumor became less defended.
• The Result: The cancer cells were less able to hide, and the body's natural police (cytotoxic T-cells, though not fully present in these specific mice models) would have a better chance of attacking them.

3. The Spread (Metastasis)

• The Metaphor: Metastasis is when the rogue citizens pack their bags and move to new neighborhoods (like the lungs) to start new illegal colonies.
• The Finding: The mice treated with the "noise-canceling headphone" had significantly fewer cancer cells in their lungs. The "silencer" stopped the cancer from spreading as aggressively.

The Big Takeaway

This study is a breakthrough because it didn't just test the drug on a flat cell; it tested it on real human tumor mini-cities inside a living body.

• It works: Silencing MALAT1 stops the cancer from spreading and makes it less protected.
• It's specific: It works best on the most aggressive types of breast cancer (TNBC).
• It opens a new door: By messing with the "blueprints" (splicing), the drug might be creating new targets for the immune system to attack.

In short: The scientists found a way to mute the "foreman" (MALAT1) that is driving the cancer. When he goes silent, the cancer stops spreading, loses its bodyguards, and starts building weird, broken structures that might be easier for the body to destroy. This gives hope for a new type of treatment for the hardest-to-treat breast cancers.

Technical Summary

1. Problem Statement

• Clinical Gap: While Long Non-coding RNAs (lncRNAs) are known to regulate tumor biology, no lncRNA-targeted therapies have successfully translated to clinical oncology.
• Contextual Ambiguity: The lncRNA MALAT1 is overexpressed in over 20 cancers, including breast cancer. However, its functional role is highly context-dependent. Previous studies using 2D cell lines and mouse models have yielded conflicting results, with some suggesting an oncogenic role and others a tumor-suppressive role.
• Model Limitations: Existing preclinical models (2D cell lines, genetically engineered mouse models) often fail to recapitulate the genetic heterogeneity, stromal interactions, and clinical relevance of human Triple-Negative Breast Cancer (TNBC). There is a critical need to evaluate MALAT1 function and therapeutic potential in clinically relevant, patient-derived models.

2. Methodology

The study employed a multi-tiered approach utilizing patient-derived organoids (PDOs) and patient-derived organoid xenografts (PDO-Xs):

• Model System:
  • Utilized a biobank of 28 human breast tumor PDOs derived from Northwell Health.
  • Established and optimized PDO-X models by injecting 2–4 million cells into the mammary fat pads of NOD scid gamma (NSG) immunocompromised mice.
  • Selected three independent TNBC PDO-X models (DS115T, NH048T, NH85TSc) for in vivo interrogation based on growth rates and engraftment success.
• Therapeutic Intervention:
  • Designed and tested two independent 16-mer gapmer antisense oligonucleotides (ASOs) targeting human MALAT1 (cEt-modified, phosphorothioate backbone).
  • Administered ASOs subcutaneously at 50 mg/kg twice weekly to established tumors in PDO-X models.
• Analytical Techniques:
  • Knockdown Efficiency: Validated via qRT-PCR and RNA-seq.
  • Transcriptomics: Bulk RNA-seq performed on primary tumors and lungs. Data was aligned to a combined human-mouse reference genome to separate tumor-intrinsic (human) and stromal (mouse) responses.
  • Splicing Analysis: Utilized the SpliceCore® platform and SpliceDuo™ algorithm to quantify alternative splicing events (Alternative Acceptor, Donor, Cassette Exon, Intron Retention) and calculate $\Delta$PSI values.
  • Genomic Characterization: Used Affinity-based Cas9-Mediated Enrichment (ACME) followed by Oxford Nanopore long-read sequencing to assess MALAT1 mutations and methylation patterns.
  • Metastasis & Microenvironment: Quantified lung metastases using human-mitochondria IHC and assessed macrophage infiltration via F4/80 IHC and mouse gene expression profiling.

3. Key Contributions

• First Comprehensive In Vivo Interrogation: This is the first study to systematically evaluate MALAT1 perturbation using human PDO-X models, bridging the gap between cell line data and clinical reality.
• Discovery of Splicing Mechanisms: Identified that MALAT1 depletion specifically drives widespread alternative splicing changes, with a unique and significant enrichment of Intron Retention (IR) events, many of which generate previously unannotated isoforms.
• Tumor-Stroma Crosstalk: Demonstrated that tumor-specific MALAT1 depletion induces transcriptional changes in the mouse stromal compartment, specifically reducing macrophage-associated gene signatures, without directly targeting the stromal cells.
• Therapeutic Validation: Provided strong preclinical evidence that ASO-mediated MALAT1 knockdown reduces metastatic burden and alters the tumor microenvironment in aggressive TNBC.

4. Key Results

• Knockdown Efficiency & Model Selection:
  • Screening 28 PDO models revealed that MALAT1 ASOs achieved >70% knockdown efficiency in ~54% of models.
  • High-efficiency models correlated with fast-growing, high-grade, poorly differentiated tumors (mostly TNBC).
  • Efficiency was not limited by endogenous MALAT1 levels, RNase H1 expression, or ASO cellular uptake, suggesting intrinsic tumor biology dictates sensitivity.
• Transcriptional Impact:
  • In vivo knockdown resulted in modest changes in overall gene expression (e.g., 319 genes in DS115T, 28 in NH048T, 952 in NH85TSc), suggesting MALAT1 acts as a fine-tuner of transcription rather than a binary switch.
  • Changes were highly patient-specific, reflecting the heterogeneity of TNBC.
• Splicing Alterations:
  • MALAT1 depletion caused widespread alternative splicing changes across all event types.
  • Intron Retention (IR) was the dominant event type (52–72% of all changes), significantly deviating from the background distribution of known splicing events.
  • A majority of these IR events occurred in unannotated transcripts (Cluster 20), generating novel isoforms.
  • Differentially spliced genes were enriched for targets of 42 shared cancer-associated transcription factors (including MYC, E2F1, YY1, and immune regulators).
• Metastasis and Microenvironment:
  • Metastasis: In PDO-X models, MALAT1 knockdown led to a significant reduction in lung metastatic burden (21–48% reduction in metastatic area), despite the models generally having low metastatic potential.
  • Immune Modulation: Analysis of the mouse stromal transcriptome revealed 78 common differentially expressed genes across all models, 74 of which were downregulated. These genes were strongly associated with macrophages and monocytes.
  • IHC Validation: F4/80 staining confirmed a 7–9% reduction in tumor-infiltrating macrophages upon MALAT1 knockdown.
• Genomic Integrity:
  • Long-read sequencing confirmed no tumor-specific MALAT1 mutations or amplifications. Hypomethylation of the MALAT1 promoter was observed in tumors compared to patient-matched blood, potentially explaining high expression levels.

5. Significance

• Therapeutic Potential: The study provides robust preclinical validation for MALAT1-targeting ASOs as a therapeutic strategy, particularly for aggressive TNBC. The reduction in metastasis and modulation of the immune microenvironment (macrophage reduction) supports its clinical development.
• Mechanistic Insight: The findings redefine MALAT1 not just as a transcriptional regulator but as a critical orchestrator of alternative splicing, specifically influencing intron retention to generate novel, potentially immunogenic neoantigens.
• Model Advancement: The successful establishment and characterization of PDO-X models for functional RNA perturbation studies demonstrate their superiority over traditional PDX or cell line models for studying tumor-stroma crosstalk and metastasis in a human-relevant context.
• Future Directions: The generation of unannotated intron-retaining isoforms suggests a potential mechanism for immune evasion (via neoantigen presentation), warranting future studies in humanized immunocompetent mice to explore combination therapies with immune checkpoint inhibitors.