The cannabinoid-1 receptor (CB1R) contributes to albuminuria and renal fibrosis. Although animal studies support the use of CB1R antagonism in diabetic nephropathy, a recent clinical trial failed to confirm this approach in humans. Pointeau and colleagues have now re-investigated the CB1R as a potential therapeutic target in diabetic nephropathy, speculating that its effect might become particularly apparent during combination treatment. Indeed, the CB1R inverse agonist monlunabant, administered either alone or in combination with the sodium-glucose co-transporter 2 (SGLT2) inhibitor empagliflozin, ameliorated renal injury in a murine model of diabetic nephropathy. This study is noteworthy in two respects: first, for confirming CB1R as a potential therapeutic target in chronic kidney disease and, second, for evaluating its effects in combination with SGLT2 inhibition (Pointeau et al., 2025). Not all basic science discoveries reach clinical application. This translational gap has multiple causes, one of which is the potential oversimplification by basic scientists who often focus on single-agent interventions—such as knocking out a single gene or administering a single inhibitor. Of course, preclinical studies often simplify in order to investigate mechanisms; however, once a novel mechanism is identified, it is essential to take the next step and conduct experiments that are more clinically relevant. In addition, such interventions are often initiated before disease onset, which does not reflect the clinical reality. In 2025, patients with diabetic nephropathy are commonly treated with combination therapies that are initiated after disease manifestation. As physicians, we face yet another challenge: keeping pace with evolving guidelines and ensuring that our treatment strategies for patients with diabetic nephropathy align with current recommendations. So, should we now treat all our patients with diabetic nephropathy using the currently advocated four-drug regimen? A number of respected experts say “yes”—and these voices carry significant weight. They publish extensively in high-impact journals and advocate their positions at major congresses (Neuen et al., 2024). As clinicians, we may be inclined to prescribe all four therapies. Yet, guided by evidence rather than eminence, we must ask: in which trial has this combination actually been tested? The answer is sobering. Despite broad enthusiasm, such combination therapy remains largely untested in randomised controlled trials. In the pivotal FIDELIO-DKD and FIGARO-DKD trials of finerenone in diabetic nephropathy, 99.9% of patients received RAS blockade, but only 5–8% an SGLT2 inhibitor and 7–8% a GLP-1 receptor agonist (Bakris et al., 2020; Pitt et al., 2021). The recent FLOW trial showed that the GLP-1 receptor agonist semaglutide reduces kidney outcomes and cardiovascular death in patients with diabetic nephropathy. Whereas 95% received RAS inhibition, only 16% were treated with an SGLT2 inhibitor, and the use of finerenone was not even mentioned (Perkovic et al., 2024). Of course, planning and conducting megatrials requires considerable effort, with additional time for data analysis and publication. During this period, new therapies often emerge—meaning that enrolled patients may not receive what is later seen as optimal background therapy. This problem partly explains why optimal background therapy is not always used in trials. Moreover, demonstrating treatment effects is more challenging when background therapy is near-optimal, as clinical event rates decline (Zhu et al., 2025). This challenge necessitates longer trials or larger sample sizes, substantially increasing costs. Megatrials with hard cardiovascular endpoints in chronic kidney disease or heart failure are already highly complex and expensive, making further expansion not always feasible. Endpoints: Drug approval for highly prevalent chronic conditions such as diabetes, heart failure or arterial hypertension should rely on reductions in hard endpoints. At the same time, improvements in surrogate parameters—such as blood glucose levels, serum brain natriuretic peptide levels or blood pressure—are insufficient to justify regulatory approval. But what exactly qualifies as a hard endpoint? In renal disease, doubling of serum creatinine or progression to end-stage renal disease (ESRD) are gold-standard trial endpoints. The slope of estimated glomerular filtration rate (eGFR) decline is increasingly discussed as an alternative. However, most nephroprotective therapies cause an initial eGFR drop before slowing long-term decline, reducing a trial's sensitivity to detect long-term benefits. To address this issue, trials often use two-slope models or a pre-randomisation run-in phase. Yet, is it appropriate to disregard early, temporary kidney function worsening when evaluating long-term outcomes? Albuminuria is another common endpoint and a major risk factor for end-stage renal disease (ESRD). This raises a key regulatory question: Should every drug that meaningfully lowers albuminuria be eligible for approval? And what is a meaningful decrease of albuminuria? 10, 30 or 50%? The debate continues, balancing the need for robust trials with hard endpoints—which may no longer be feasible—and the adoption of softer endpoints that may not always reflect true patient benefit. Combination therapy: Megatrials must ensure that all participants receive optimal background medication. Investigators and authorities must remain flexible to adapt protocols if new treatment strategies emerge shortly before trial initiation. Fixed-combination trials, such as the recently launched EASi-KIDNEY study, testing empagliflozin with the aldosterone synthase (CYP11B2) inhibitor vicadrostat in 11,000 patients, may serve as a model for future research (Judge et al., 2025). Another approach is to test therapies that have reduced hard endpoints in earlier trials under suboptimal background therapy in follow-up studies using the state-of-the-art background and combination therapy but with surrogate endpoints. The CONFIDENCE trial (COmbinatioN effect of FInerenone anD EmpaglifloziN) is evaluating finerenone and empagliflozin, alone or in combination, using albuminuria reduction as the primary endpoint (Agarwal et al., 2025). An added advantage of fixed combinations might be a reduced occurrence of side effects. For instance, SGLT2 inhibition on top of aldosterone blockade lowers the risk of hyperkalaemia that is associated with the latter (Tuttle et al., 2024). Preclinical studies also must incorporate optimal background therapy, to be recognised and published in high-impact journals. Moreover, treatment should not be administered solely before disease onset, but also after the disease has been established—reflecting real-world clinical conditions. In this regard, the study of Pointeau and colleagues is exemplary: they applied combination therapy and studied mice with established renal injury. Particularly mentioned should be the recent work by Zhu et al., which serves as a model of best practice. In this study, mice with established Alport disease were randomised to receive ramipril monotherapy, ramipril plus empagliflozin, or a triple combination of ramipril, empagliflozin and finerenone. The primary endpoint was survival. Notably, experiments were conducted in both male and female mice—an approach that should become standard in preclinical research (Zhu et al., 2023). In conclusion, there is room for improvement in both clinical and preclinical research. Scientists, funding agencies, authorities and journals must ensure that experimental and clinical approaches address real-world needs. Many journals, such as the British Journal of Pharmacology, already require a statement on translational relevance. The study by Pointeau et al. is commendable because it exactly addresses a real-world need (pairing a validated drug with a novel target) using a clinically relevant model and starting treatment after the onset of disease. Patient needs also must remain central in clinical trials. We must balance scientific rigour—including optimal background therapy and meaningful endpoints—with the feasibility of real-world studies. Translational success depends on uniting rigorous preclinical design with clinically relevant research. The ideas exist; their execution must follow. Shanshan Li, Leonie Dreher, A. H. Jan Danser and Ulrich O. Wenzel wrote the manuscript, and Ulrich O. Wenzel prepared the figure. The study was supported by RECORD (Research Center on Rare Kidney Diseases), iDfellows (German Research Foundation) to Leonie Dreher and German Research Foundation grant We 1688/19-3 to Ulrich O. Wenzel. Open Access funding enabled and organized by Projekt DEAL. The authors declare no conflicts of interest.
No takes yet. Share an insight, caveat, or question.
Li et al. (2025) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: