FSGS With Chronic Scarring: Immunosuppress, or Trust Sparsentan Instead
A single patient with FSGS and real, already-present chronic scarring. The disagreement is whether immunosuppression still earns its toxicity once meaningful scarring exists, or whether a newer non-immunosuppressive drug with an unproven long-term benefit is the more honest choice.
R.M. found out he had kidney disease the same way he's found out most things about his health over the past decade — during a routine blood draw for his hypertension refill, where a nurse mentioned in passing that his protein looked high and he should probably get that checked. He's sixty-two, has managed hypertension for fifteen years without much drama, and had no symptoms at all when the biopsy came back three weeks later: focal segmental glomerulosclerosis, not-otherwise-specified variant, with 30% of glomeruli already globally sclerotic — old, permanent scar tissue, not active injury the biopsy caught in progress. His eGFR is already down to 45 and his proteinuria is nephrotic-range at 5.8 grams a day — but his albumin is holding at 3.3 and he has no edema, so he has nephrotic-range proteinuria without nephrotic syndrome, a distinction that turns out to decide which of the two drugs on the table he is actually eligible for.
That biopsy detail — nearly a third of glomeruli already scarred — is what actually complicates this decision, more than the proteinuria number does. CureGN's accumulated cohort data speaks to exactly this tension: whether immunosuppression's benefit holds once meaningful chronic scarring is already present, or whether it's mainly protecting glomeruli that are still salvageable, offering little to ones that already aren't. There's no clean cutoff in that data, only a real, unresolved pattern practitioners genuinely read differently. Sparsentan offers a non-immunosuppressive alternative built around a different mechanism entirely — dual endothelin-A and angiotensin II receptor blockade, reducing intraglomerular pressure rather than suppressing an immune process. DUPLEX tested it head-to-head against maximally dosed irbesartan and met its prespecified interim endpoint, partial remission of proteinuria at 36 weeks (42% versus 26%); its primary endpoint, eGFR slope over 108 weeks, was missed at final analysis. The FDA nonetheless approved sparsentan for FSGS in April 2026 — the first drug ever approved for the disease — but wrote the indication narrowly, for patients without nephrotic syndrome, and the trial data explain why: in DUPLEX's nephrotic subgroup proteinuria fell 39% on sparsentan against 37% on irbesartan, essentially no separation, with a numerically greater eGFR decline on the drug. In the non-nephrotic subgroup the same trial found 48% against 27%. His albumin of 3.3 and absent edema put him on the side of that line where the drug both works and is labeled, which is the reason it is a real option for him and would not be for a patient with the same proteinuria and an albumin of 2.6.
In clinic, after the biopsy result
Nephrotic-range proteinuria with primary FSGS is the standard indication for immunosuppression, full stop. I know the scarring is real, but a biopsy slide can't fully separate active injury from old scar in the same tissue — if there's an active component, and there very well may be, withholding treatment because of what's already scarred risks losing what isn't yet.
I hear that, but CureGN's own cohort data speaks directly to this pattern — immunosuppression benefit is attenuated once substantial chronic scarring is already present, and thirty percent globally sclerotic is substantial. He's sixty-two; steroid toxicity is a real, proportionate cost, and I'm not convinced the expected benefit here clears that bar. Sparsentan gives us a real proteinuria-reduction option through a completely different mechanism — and he is inside its label, which matters more than it sounds. The FSGS indication is written for patients without nephrotic syndrome, and his albumin of 3.3 with no edema puts him there. In DUPLEX's nephrotic subgroup the drug did essentially nothing to proteinuria; in the non-nephrotic subgroup it nearly doubled the irbesartan response.
A classic indication written for FSGS in general doesn't automatically apply once we know his specific biopsy already shows this much irreversible scarring — the indication assumes a kidney that still has more to protect.
I don't think either of you is wrong, and I don't think we have to fully resolve the CureGN question to make a good decision today. Biopsy genuinely can't distinguish primary, treatment-responsive FSGS from an adaptive or genetic form that was never going to respond to immunosuppression regardless of scarring — the histology looks similar even though the biology is different.
So: trial immunosuppression, but explicitly time-limited, with a hard checkpoint at ten weeks. A partial response tells us this is active, primary disease worth continuing to treat. No response at all tells us we're not fighting the right process, and we move to sparsentan and RAAS optimization without having exposed him to months more of steroids first. One condition on that contingency worth writing down now: sparsentan's approval covers him only while he stays non-nephrotic. If ten weeks of no response also brings his albumin below 3.0 with edema, he has crossed out of both the label and the subgroup where the drug separated from irbesartan, and the fallback has to be rethought rather than executed. That's a real answer this patient's own disease gives us, not a cohort inference we're applying from outside.
Agreed: a ten-week trial of corticosteroid therapy with proteinuria checked at weeks 4 and 10 against an explicit, pre-agreed partial-response threshold. Sparsentan is the named next step, not a repeat or extended steroid course, if that threshold isn't met. Albumin is checked at both visits, not as a severity marker but because it governs whether he remains inside sparsentan's approved population.
Continue immunosuppression per standard duration; his disease is read as active and treatment-responsive.
Stop immunosuppression and start sparsentan on top of optimized RAAS blockade — provided he is still non-nephrotic at that point. Albumin rechecked at week 10 alongside proteinuria for exactly this reason.