In Alien, the xenomorph doesn't just survive - it adapts to every corridor, every airlock, every desperate human countermeasure. Pancreatic cancer cells, it turns out, pull the same trick. A new study in Cancer Research reveals that when pancreatic tumor cells spread to different organs, they don't rely on genetic mutations to fit in. They just... redecorate. Same DNA, completely different furniture.

The Deadliest Shapeshifter You've Never Heard Of

Pancreatic ductal adenocarcinoma (PDAC) is the honey badger of cancers. It doesn't care about your treatment plan. With a five-year survival rate hovering around 12%, and roughly 80% of patients already at advanced stages by diagnosis, PDAC is responsible for a wildly disproportionate share of cancer deaths. The reason it's so lethal? Metastasis - cancer cells packing their bags and colonizing distant organs like unwanted houseguests who immediately rearrange your living room to suit their taste.

But here's the question that's been bugging researchers: when cancer cells move from the pancreas to, say, the liver or the stomach wall, do they succeed because they genetically evolved to thrive there? Or are they just really, really good at reading the room?

One Patient, Eleven Tumors, Forty-Five Thousand Cells

A team led by Alejandro Jiménez-Sánchez and colleagues at Memorial Sloan Kettering took a remarkably detailed look at this question using tissue from a rapid autopsy - a procedure where samples are collected within hours of death, preserving the full landscape of metastatic disease. Think of it as getting a snapshot of every franchise location a cancer has opened, all on the same day.

They performed single-nucleus RNA sequencing on the primary pancreatic tumor and nine metastatic sites (liver, omentum, peritoneum, stomach wall, lymph node, and diaphragm), capturing over 45,000 individual cancer cell transcriptomes. That's like reading the diary of every single employee at every branch office to figure out who's following corporate policy and who's gone rogue (Jiménez-Sánchez et al., 2026).

The Cancer Cells Are Method Actors

Using unsupervised archetype analysis - a technique that identifies the "extreme roles" cells can play - the team found that cancer cells at different metastatic sites were running wildly different gene programs. Peritoneal metastases had cranked up lipid metabolism genes, like they'd moved to a new city and immediately adopted the local cuisine. Stomach wall lesions activated gastrointestinal programs, essentially cosplaying as stomach cells. Each site had its own vibe, and the cancer cells matched it.

A companion spatial transcriptomics study in Nature confirmed these patterns across 13 additional autopsy cases, showing that this isn't a one-patient fluke - it's a recurring survival strategy (Pei et al., 2025).

Enter PICASSO (No, Not That One)

To untangle whether these transcriptomic costume changes were driven by genetic differences between cell clones, the team built a new computational tool called PICASSO (Phylogenetic Inference from Copy-number Alterations in Single-cell Sequencing Observations). It's a probabilistic framework that reconstructs the family tree of cancer clones using the noisy copy-number signals buried in single-cell RNA data, cross-referenced with whole-exome sequencing.

The punchline? Cells from the same genetic clone behaved completely differently depending on where they landed. A clone in the liver looked nothing like its sibling in the peritoneum, transcriptionally speaking. The clonal genotype contributed minimally to the observed phenotypic differences. It's like identical twins raised in different countries - same genome, completely different personalities.

If you're the kind of person who likes visualizing complex relationships between data points, tools like mapb2.io can help you sketch out these kinds of branching evolutionary trees and gene program networks in a way that actually makes sense to a human brain.

Why This Matters (Beyond Being Cool Science)

This finding has uncomfortable implications for how we think about treating metastatic PDAC. If cancer cells are chameleons that adapt their gene expression to whatever organ they've invaded, then a drug targeting a specific transcriptomic subtype might only work on some metastatic sites while the others shrug it off. It's the therapeutic equivalent of playing whack-a-mole against an opponent who can turn into the mole and the hammer.

The study also identified that basal-like cancer cells tend to cozy up with specific cancer-associated fibroblasts (myCAFs) and actively exclude immune cells through CXCR4-CXCL12 signaling. Translation: the most aggressive cancer cells are literally building themselves a bodyguard squad that keeps your immune system out.

The Bottom Line

Cancer metastasis isn't just about which cells escape the primary tumor. It's about what those cells become after they arrive. This study - and the growing body of archetype-based single-cell work, including foundational studies in small cell lung cancer (Barkley et al., 2022) - suggests that transcriptomic plasticity, not genetic evolution, is the main engine of metastatic adaptation. The tumor genome sets the stage, but the microenvironment writes the script.

Future therapies may need to target the plasticity itself - the ability to switch states - rather than any single state. Which is a harder problem, sure. But at least now we know what we're actually fighting.

References

1. Jiménez-Sánchez, A., Persad, S., Hayashi, A., et al. (2026). Transcriptomic Plasticity Is a Hallmark of Metastatic Pancreatic Cancer. Cancer Research, 86(7), 1769-1796. DOI: 10.1158/0008-5472.CAN-25-1117. PMID: 41379552.

2. Pei, M., et al. (2025). Spatial mapping of transcriptomic plasticity in metastatic pancreatic cancer. Nature, 642, 212-221. DOI: 10.1038/s41586-025-08927-x. PMID: 40269162.

3. Barkley, D., et al. (2022). Archetype tasks link intratumoral heterogeneity to plasticity and cancer hallmarks in small cell lung cancer. Cell Systems, 13(9), 690-710.e17. DOI: 10.1016/j.cels.2022.07.006.

4. Ho, W.J., et al. (2023). Single cell transcriptomic analyses implicate an immunosuppressive tumor microenvironment in pancreatic cancer liver metastasis. Nature Communications, 14, 5123. DOI: 10.1038/s41467-023-40727-7.

5. bioRxiv preprint: 10.1101/2025.02.28.640922.

Disclaimer: This blog post is a simplified summary of published research for educational purposes. The accompanying illustration is artistic and does not depict actual model architectures, data, or experimental results. Always refer to the original paper for technical details.