Diabetic kidney disease (DKD) represents an enormous burden in patients with type 2 diabetes mellitus (T2DM). Preclinical studies using most glucose-lowering agents have suggested renal-protective effects, but the proposed mechanisms of renoprotection have yet to be defined, and the promising results from experimental studies remain to be translated into human clinical findings to improve the prognosis of patients at risk of DKD. Also, it is important to distinguish effects on surrogate endpoints, such as decreases in albuminuria and estimated glomerular filtration rate (eGFR), and hard clinical endpoints, such as progression to end-stage renal disease (ESRD) and death from renal causes. Data regarding insulin therapy are surprisingly scarce, and it is nearly impossible to separate the effects of better glucose control from those of insulin per se, whereas favourable preclinical data with metformin, thiazolidinediones and dipeptidyl peptidase (DPP)-4 inhibitors are plentiful, and positive effects have been observed in clinical studies, at least for surrogate endpoints. The most favourable renal results have been reported with glucagon-like peptide-1 receptor agonists (GLP-1RAs) and sodium–glucose cotransporter type-2 inhibitors (SGLT2is). Significant reductions in both albuminuria and eGFR decline have been reported with these classes of glucose-lowering medications compared with placebo and other glucose-lowering agents. Moreover, in large prospective cardiovascular outcome trials using composite renal outcomes as secondary endpoints, both GLP-1RAs and SGLT2is added to standard care reduced renal outcomes combining persistent macro-albuminuria, doubling of serum creatinine, progression to ESRD and kidney-related death; however, to date, only SGLT2is have been clearly shown to reduce such hard clinical outcomes. Yet, as the renoprotective effects of SGLT2is and GLP-1RAs appear to be independent of glucose-lowering activity, the underlying mechanisms are still a matter of debate. For this reason, further studies with renal outcomes as primary endpoints are now awaited in T2DM patients at high risk of DKD, including trials evaluating the potential add-on benefits of combined GLP-1RA–SGLT2i therapies.