The global burden of diabetic retinopathy (DR) continues to worsen and DR remains a leading cause of vision loss worldwide. Here, we describe an algorithm to predict DR progression by means of deep learning (DL), using as input color fundus photographs (CFPs) acquired at a single visit from a patient with DR. The proposed DL models were designed to predict future DR progression, defined as 2-step worsening on the Early Treatment Diabetic Retinopathy Diabetic Retinopathy Severity Scale, and were trained against DR severity scores assessed after 6, 12, and 24 months from the baseline visit by masked, well-trained, human reading center graders. The performance of one of these models (prediction at month 12) resulted in an area under the curve equal to 0.79. Interestingly, our results highlight the importance of the predictive signal located in the peripheral retinal fields, not routinely collected for DR assessments, and the importance of microvascular abnormalities. Our findings show the feasibility of predicting future DR progression by leveraging CFPs of a patient acquired at a single visit. Upon further development on larger and more diverse datasets, such an algorithm could enable early diagnosis and referral to a retina specialist for more frequent monitoring and even consideration of early intervention. Moreover, it could also improve patient recruitment for clinical trials targeting DR.

The importance of DDR1 in renal fibrosis has been shown via gene knockout and use of antisense oligonucleotides; however, these techniques act via a reduction of DDR1 protein while we prove the therapeutic potential of inhibiting DDR1 phosphorylation with a small molecule. To date, efforts to generate a selective small-molecule to specifically modulate the activity of DDR1 in an in vivo model have been unsuccessful. We performed parallel DNA encoded library screens against DDR1 and DDR2, and discovered a chemical series that is highly selective for DDR1 over DDR2. Structure-guided optimization efforts yielded the potent DDR1 inhibitor 2.45, which possesses excellent kinome selectivity (including 64-fold selectivity over DDR2 in a biochemical assay), a clean in vitro safety profile, and favorable pharmacokinetic and physicochemical properties. As desired, compound 2.45 modulates DDR1 phosphorylation in vitro as well as prevents collagen-induced activation of renal epithelial cells expressing DDR1. Compound 2.45 preserves renal function and reduces tissue damage in Col4a3−/− mice (the preclinical mouse model of Alport syndrome) when employing a therapeutic dosing regime, indicating the real therapeutic value of selectively inhibiting DDR1 phosphorylation in vivo. Our results may have wider significance as Col4a3−/− mice also represent a model for chronic kidney disease, a disease which affects 10% of the global population.