Back

A DDR2-targeted PET tracer images activated fibroblasts in early pulmonary fibrosis

Chen, X.; Lai, Y.; Yang, P.; Li, G.; Hu, G.; Rao, J.; Shi, M.; Luo, Q.; Chen, Z.; Jiang, J.; Yang, L.; Gao, C.; Zhang, P.; Wang, H.; Qiu, S.; Wang, K.; Ben, X.; Lv, J.; Cheng, Y.; Liu, J.; Li, J.; Ran, P.; Su, J.

2025-11-28 respiratory medicine
10.1101/2025.11.26.25341068 medRxiv
Show abstract

Molecular imaging of activated fibroblasts holds promise for advancing the early detection and monitoring of idiopathic pulmonary fibrosis (IPF), a fatal lung disease with limited therapeutic options. Current imaging strategies primarily rely on fibroblast activation protein (FAP), but its functional ambiguity complicates the interpretation of signal intensity as a direct indicator of fibrotic activity. Here, we identify discoidin domain receptor 2 (DDR2), a collagen-binding receptor tyrosine kinase, as a superior and coherent target for imaging active fibroblasts in IPF. DDR2 was found to be significantly upregulated in IPF lung tissues, particularly in the early fibrotic phase, compared to FAP. Using a nanobody (1A12) with high affinity, specificity and stability, we developed a DDR2-targeted radiotracer, 68Ga-NOTA-1A12. In both mouse and non-human primate models of pulmonary fibrosis, we demonstrated that 68Ga-NOTA-1A12 PET/CT detected pathological changes earlier and with higher signal uptake than the established 18F-FAPI tracer, underscoring its potential for early diagnosis; moreover, total standardized uptake value of 68Ga-NOTA-1A12 correlated with fibrosis progression, providing a quantitative metric for disease evaluation. Notably, 68Ga-NOTA-1A12 PET/CT safely identified pulmonary fibrotic foci in a patient with interstitial lung disease. Our findings position DDR2-targeted PET/CT as a transformative tool for stratifying IPF patients, guiding anti-fibrotic therapies.

Matching journals

The top 7 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.