Chemotherapy resistance in pancreatic cancer: the DNA repair link

Chemotherapy resistance in pancreatic cancer may depend not only on mutations but also on how efficiently tumour cells repair damaged DNA. In one group of pancreatic tumours, this repair system was unusually active: two proteins in the nucleus switched on genes that helped the cells survive platinum-based drugs.

The study focused on pancreatic ductal adenocarcinoma, the most common form of pancreatic cancer. Even after surgery, the disease frequently returns, and chemotherapy often fails to control it for long. Researchers are therefore looking not only for new drugs, but also for molecular features that can identify especially aggressive tumours before treatment decisions are made.

A subtype linked to early recurrence

The authors first examined data from more than 560 patients in public datasets and then analysed tumour samples from 82 patients whose cancers had been surgically removed. They focused on GSK3β, a kinase involved in many cellular processes. In some tumours, unusually large amounts of GSK3β accumulated inside the nucleus, where the cell stores its DNA.

High nuclear GSK3β alone was associated with poorer outcomes. The clearest difference, however, appeared when the researchers also measured NFATc1. NFATc1 is a transcription factor: it binds regulatory regions of DNA and alters which genes are active.

Tumours with high nuclear levels of both GSK3β and NFATc1 represented about 14% of the resected cancers in the cohort. In this subgroup, median relapse-free survival after surgery was 186 days, compared with 367 days in the remaining patients. Median overall survival was 267 days versus 715 days.

These numbers show an association, not proof that the two proteins alone determine a patient’s outcome. The clinical analysis involved a relatively small cohort of people with resectable disease, and the proposed subtype included only a minority of those cases.

How tumour cells repair chemotherapy damage

Platinum drugs damage DNA, including by generating severe double-strand breaks. A cell that cannot repair such damage may stop dividing or die. Cancer cells, however, can sometimes use their own DNA-repair machinery as a defence against treatment.

In the experimental models, GSK3β helped maintain NFATc1 in an active state. NFATc1 then increased the expression of BRCA1, BRCA2, RAD51 and several FANC-family genes. Their products participate in homologous recombination, a high-fidelity repair process that restores broken DNA by copying information from an intact template.

This creates a biological paradox. A pathway that normally protects genome integrity can become a mechanism of drug resistance in cancer. After exposure to cisplatin or oxaliplatin, cells with active GSK3β–NFATc1 signalling repaired damage more effectively and continued growing.

The researchers tested the mechanism in established cell lines, patient-derived tumour cells and organoids, slices of human pancreatic tumours, and mouse models. When GSK3β or NFATc1 was disrupted genetically or with experimental inhibitors, homologous-recombination activity declined, DNA damage accumulated, and the tumour cells became more sensitive to platinum treatment.

The effect was most evident in tumours with an intact homologous-recombination system. In BRCA2-mutant models, where this repair pathway was already defective, adding a GSK3β inhibitor did not produce the same increase in chemotherapy sensitivity. This distinction matters: any future combination would probably need molecular selection rather than being given to all patients with pancreatic cancer.

Not yet a new treatment

The findings suggest two possible applications. First, high nuclear levels of both GSK3β and NFATc1 could become a marker of tumours at increased risk of early recurrence. Second, blocking this signalling axis might weaken DNA repair and restore sensitivity to platinum drugs.

Neither possibility is ready for clinical use. The study did not test the drug combination in patients and therefore cannot show that GSK3β inhibition improves survival or is safe alongside chemotherapy. The treatment evidence comes from cells, organoids, tumour explants and mice. The prognostic association also needs confirmation in larger, independent cohorts, including patients with unresectable or metastatic disease.

The work nevertheless offers a plausible explanation for why the same chemotherapy can produce very different results in tumours arising from the same organ. In some cases, the decisive difference may not be how much DNA damage the drug causes, but how quickly the tumour can repair it.