Finerenone and Diabetic Kidney Disease: Reading FIDELIO and FIGARO

For adults with type 2 diabetes and chronic kidney disease who are already on the highest tolerated dose of an ACE inhibitor or angiotensin receptor blocker, adding finerenone lowered the combined risk of kidney and heart events. Two trials divided the work: FIDELIO-DKD followed the kidney, FIGARO-DKD followed the heart, and the pooled FIDELITY analysis showed both benefits held across the spectrum of disease. The price of admission was more hyperkalemia, a rise in blood potassium that stayed manageable when it was monitored on schedule. That single trade-off, benefit against a watchable harm, is the whole story.

Key points#

• Finerenone is a nonsteroidal blocker of the mineralocorticoid receptor, added on top of standard renin-angiotensin system (RAS) therapy, not instead of it.
• FIDELIO-DKD (more advanced kidney disease) cut a kidney composite; FIGARO-DKD (earlier disease) cut cardiovascular events, driven mostly by fewer heart failure hospitalizations.
• The pooled FIDELITY analysis of 13,026 patients confirmed both signals, and the kidney benefit survived a stricter endpoint definition.
• Hyperkalemia was roughly twice as common on finerenone, but serious events were rare and no hyperkalemia deaths were reported in the pooled data.
• Whether the trade is worth it for one person depends on baseline potassium, kidney function, and other medicines. That is a clinician's call, not a trial average.

The gap that RAS blockade leaves open#

For decades, the anchor of diabetic kidney care has been blocking the renin-angiotensin system with an ACE inhibitor or an ARB. It works, but it does not finish the job. Even on maximal RAS therapy, many patients keep losing kidney function and keep landing in the hospital for heart failure. That leftover danger has a name in the literature, residual risk, and one of its drivers is the mineralocorticoid receptor.

When this receptor is overactivated, it pushes inflammation and scarring (fibrosis) in kidney and heart tissue through pathways that run separately from blood pressure. So lowering pressure alone does not switch it off. That insight is why an old drug class came back into the conversation.

Why not just use spironolactone#

The receptor already had blockers: spironolactone and eplerenone, both steroidal. In diabetic kidney disease they ran into two walls. They raised potassium enough to be risky in exactly the patients who needed protection, and spironolactone in particular hit other steroid receptors, producing off-target effects. Neither was ever proven in a large kidney-outcome trial in this group.

Finerenone was built differently. It is nonsteroidal, so it grips the receptor in a distinct way, spreads more evenly between heart and kidney tissue rather than concentrating in one, and clears quickly with no lingering active metabolites. The open question was simple to state and hard to answer: do those chemistry differences turn into fewer real clinical events, and at what safety cost?

Two trials, two ends of one spectrum#

Rather than run a single study, the program split diabetic kidney disease into its two natural halves.

FIDELIO-DKD (New England Journal of Medicine, 2020) took the sicker end. It randomized 5,734 patients with type 2 diabetes and more advanced kidney disease, defined by albuminuria plus reduced filtration, to finerenone or placebo on top of maximal RAS therapy. Its primary endpoint was a kidney composite: kidney failure, a sustained drop of at least 40 percent in estimated glomerular filtration rate, or death from kidney causes. Over a median of 2.6 years the composite hit 17.8 percent on finerenone versus 21.1 percent on placebo (hazard ratio 0.82, 95 percent CI 0.73 to 0.93). A secondary heart composite also leaned in finerenone's favor (hazard ratio 0.86, 95 percent CI 0.75 to 0.99). A kidney trial, in other words, carried a heart signal alongside its main result.

FIGARO-DKD (2021) took the earlier end and flipped the question. It enrolled 7,352 patients with better-preserved filtration but albuminuria, and made the cardiovascular composite the primary endpoint. Over a median of 3.4 years that outcome occurred in 12.4 percent on finerenone versus 14.2 percent on placebo (hazard ratio 0.87, 95 percent CI 0.76 to 0.98). What drove it matters to you: the win came mostly from fewer hospitalizations for heart failure (hazard ratio 0.71, 95 percent CI 0.56 to 0.90), not from large separate cuts in heart attack, stroke, or cardiovascular death.

Lay the two side by side and they cover a broad band of the disease, from early kidney damage with heavy protein leak to advanced impairment, pointing the same way in both.

Do the composites actually hold up#

Composite endpoints deserve suspicion, because a headline result can rest entirely on its softest ingredient. FIDELIO's kidney composite counted a 40 percent filtration drop, which is more sensitive and catches change earlier than the older benchmark of a doubling of serum creatinine (roughly a 57 percent decline). If you are a fair skeptic, you could ask whether the benefit lived only inside that gentler threshold.

The pooled FIDELITY analysis (European Heart Journal, 2022) put your worry to the test. It combined both trials, 13,026 patients, and applied the tougher 57 percent threshold to the kidney composite. The cardiovascular composite held (hazard ratio 0.86, 95 percent CI 0.78 to 0.95), and the stricter kidney composite still favored finerenone (hazard ratio 0.77, 95 percent CI 0.67 to 0.88). A result that survives a harder definition is one worth trusting more, and this one did.

The potassium trade-off, weighed plainly#

Nothing here is free. The cost was hyperkalemia. In FIDELIO-DKD, investigator-reported high potassium occurred in 18.3 percent on finerenone versus 9.0 percent on placebo, and led to stopping the drug in 2.3 percent versus 0.9 percent. In the pooled FIDELITY data it appeared in 14.0 percent versus 6.9 percent, with hospitalization for hyperkalemia in 0.9 percent versus 0.2 percent and permanent discontinuation in 1.7 percent versus 0.6 percent.

Two facts shrink that signal without deleting it. First, the trials screened out people whose baseline potassium already ran high and then required scheduled potassium checks with dose adjustment, so these rates come from a monitored setting, not a leave-it-alone one. Second, dangerous events were uncommon, and the pooled analysis reported no deaths tied to hyperkalemia. The honest summary: finerenone swaps a modest, watchable rise in potassium for a measurable reduction in kidney and heart events, and whether that swap favors your patient turns on baseline potassium, kidney function, and the rest of their medication list.

Diabetic kidney disease is a useful case study in evidence appraisal. The lesson it teaches is not that finerenone is good or bad in the abstract, but that a real average benefit and a real average harm can both be true at once, and that the decision only becomes meaningful one patient at a time.