Our group investigated prevention methods for pancreatic cancer, using transgenic mice models.
Relevant to the line of work, there was a new paper on precancerous lesions in pancreas. The paper appeared in Nature, a prestigious scientific journal.
So I checked out the paper.
The authors' group was from Johns Hopkins. They recently developed a new methodology that they call CODA.
Basically, the method can generate 3-dimensional imaging model from a stack of histopathology slides, visualizing tissue structures (fat, duct, smooth muscle, epithelia, etc) as well as precancer lesions and cancers, in a high resolution.
Use of AI was involved in figuring out histopathology from images (e.g., assigning which part is precancer, cancer, or normal part).
By combining microdissection and DNA/RNA extraction, they can analyze genetic/genomic alterations occurring in cancer and precancer lesions.
Nature journal editorial office seems to like papers using new technology to answer old questions.
In this case, they revealed precancer lesions in normal-looking pancreas from pancreatic (and other) cancer patients with 3D microanatomy and information on genomic mutations.
Precancer lesions in humans are difficult to study.
(a) The size is small (<5mm for pre-pancreatic cancer lesions), thus hard to find.
(b) Normal(-looking) human tissues are hard to come by for research purposes in general.
(c) In many cases, precancer lesions do not grow to cancer. As such, treatment may not necessarily be prioritized in clinic.
These were hurdles to investigate precancer lesions.
We did a journal club at our worksite, and the paper's contents were presented.
A few surprises in the report include;
(1) Pancreatic precancer lesions (called PanINs. Another type of lesion called IPMNs was not included in this study) are many, estimated about 1000 (!) in pancreas.
(2) PanINs carry different sets of mutations, suggesting their independent origins. Implicating that they grow like mushrooms in a field.
Below are my impressions. The authors did not spend much of the writing on speculations and translational/therapeutic implications.
(3) As another implication, the various mutations can help PanINs to evade specific mutation-targeting drugs.
Pancreatic cancer (Pancreatic Ductal Adeno Carcinoma PDAC) has been a difficult cancer with 5-year survival rate of 5-8%.
It was known that over 90% of PDAC have a mutation in an oncogene called KRAS, and that the mutation is a driving force for PDAC development.
Naturally, researchers made efforts to develop drugs inhibiting mutated KRAS. KRAS G12D and KRAS G12C are among most commonly occurring oncogenic mutations, and these mutations were targets to develop specific inhibitors. After years of struggle, they succeeded in this past few years.
Specific inhibitors for KRAS G12D and G12C are available, and showing success at laboratory level. There is a hope of using the new drugs for KRAS-mutated cancers (not only for PDAC) in clinic.
Yet, the new results implies that, at least for pancreatic cancer, efficacy of a mutation-specific drug can drop, once cancer or precancer lesions adopt to the specific mutation-targeting drug.
Or, in another scenario, non-responsive cancer/lesions to the drug may start to grow. For example, G12D-specific drug can be overridden by other mutations, rendering the drug ineffective.
The new results indicate that normal-looking parts of pancreas from PDAC patients can carry 1000 PanIN precancer lesions. After treatments, the PanINs, with a mix of different mutations, may develop to another PDAC.
Perhaps, PanINs work like Hydra's heads, growing back to another PDAC, contributing to the difficulty to beat PDAC in clinic.
Sounds like trouble? Indeed. Well, knowing novel aspects of human precancer lesions will help us to beat them, eventually.
Whether normal people, not cancer patients, have many precancer lesions/PanINs or not is an open question.
Personally, I don't want to believe I have hundreds of PanINs.
Also, the "field effect", whole pancreas developing precancer lesions, was peculiar. It looks as if entire pancreas is a field to grow precancer lesions and/or cancers.
In other organs like in colon, numbers of precancer lesions are usually much lower, unless there are genetic predispositions.
We saw that hepatocellular carcinomas (liver cancer) develop from heavily fibrotic (cirrhosis) liver. But such "field effects" come into play after healthy liver was affected in some ways, like hepatitis virus infection, aflatoxin (fungus toxin) exposure, alcohol abuse, or fatty liver with inflammation, over many years.
I wonder if environmental carcinogen or virus are in effect for pancreatic cancers as well.
The paper was, by nature, descriptive. But was also rather thought provoking.
Braxton et al., "3D genomic mapping reveals multifocality of human pancreatic precancers".
Nature 2024, May 1. online ahead of print.
https://pubmed.ncbi.nlm.nih.gov/38693266/
