CRMD — CorMedix Inc.

Deep-value analysis — CorMedix Therapeutics (NASDAQ: CRMD)

Analysis date: September 8, 2026 · Reference price: … (09/04/2026) · Market cap: … · Shares outstanding: 77.667 mil. (06/30/2026) · Diluted shares (Q2 2026): 92.479 mil.

Primary sources: 10-K 2026-03-05 (fiscal 2025), 10-Q 2026-08-13 (Q2 2026), 10-Q 2026-05-14 (Q1 2026), 10-Q 2025-11-12 (Q3 2025), 8-K/ER 2026-08-13, DEF 14A 2026-04-29, SEC XBRL companyfacts, data pack data-pack-CRMD-20260908.md, Graham/Buffett/Lynch tracker. External checks: analyst consensus (stockanalysis.com, 5 analysts), CMS ESRD PPS.

Executive summary

CorMedix is no longer the company the mechanical tracker sees. The screen shows a P/E of 3.87, a free-cash-flow yield of …%, and a net margin of 40.5% — figures that, taken literally, would make CRMD the cheapest profitable company in the entire covered universe. All three are artifacts of a reimbursement window that closed on June 30, 2026. This is the whole thesis, in two sentences: the market isn’t wrong about the apparent cheapness, it has already priced the cliff; the real question isn’t “why is it at 3.9x earnings,” it’s “what is the business worth AFTER TDAPA disappears.”

The mechanics. DefenCath (taurolidine + heparin, a catheter-lock solution for hemodialysis patients) entered TDAPA — Transitional Drug Add-on Payment Adjustment, a transitional Medicare supplement paid at 100% of average sales price, on top of the bundled ESRD payment — on July 1, 2024. Two years of full-price reimbursement drove revenue from … (2024) to … (2025) and operating profit from −… to +… On July 1, 2026, TDAPA converted into the post-TDAPA Add-On Adjustment, fixed by CMS at … per dialysis session for Q3 and Q4 2026 (10-K 2026-03-05, line 459). The company states it plainly: “the reimbursement level … has decreased significantly, and as a result CorMedix expects a corresponding decrease in net price” (10-Q Q2 2026, line 693).

Management’s own guidance quantifies the cliff more brutally than any model of mine. FY2026 guidance maintained: consolidated revenue …-345 mil., adjusted EBITDA …-140 mil. Realized in H1 2026: revenue …, adjusted EBITDA … By subtraction, H2 2026 implies revenue of …-116 mil. and adjusted EBITDA between −… and +… The company will go, in a single quarter, from a 42% operating margin to roughly zero. It isn’t a slowdown, it’s a cliff — scheduled, announced, but absent from any trailing indicator.

What’s left on the other side. Three things, all real: (1) the Melinta portfolio, acquired in August 2025 for … net of cash — six hospital anti-infectives plus TOPROL-XL, bringing ~… per quarter of revenue not exposed to dialysis CMS; (2) the add-on’s reset on January 1, 2027, which the company estimates at 3-5x the … level, i.e. …-11.85 per session, valid through June 30, 2029; (3) a contract signed in Q2 2026 with the last major dialysis operator, completing top-5 US coverage, with the pilot starting in Q3 — new volume right when the price is at its low.

My free cash flow bridge. Simple FCF over the trailing 12 months is … USD, and I reject it as a valuation base, for three reasons verifiable in the filings: (a) gross-to-net accruals (rebates, chargebacks, returns owed to clients) grew from … (Dec. 2024) to … (Dec. 2025) and … (June 2026) — a substantial share of CFO is customers’ money in transit, not profit; (b) the deferred tax asset was fully consumed in H1 2026 (… → 0), so tax becomes a cash outflow from now on; (c) TDAPA-era prices don’t repeat. Normalized 2027 owner earnings: …, before contingent royalties (treated separately, as debt). This is the figure I fed to the simulation.

Verdict. At … CRMD trades at ~9.3x normalized owner earnings, with ~… of cash net of convertibles and a … buyback ceiling active through the end of 2027. Monte Carlo (20,000 scenarios, OE 71, g1 6%, r 12%, gt 0.5%) gives a median intrinsic value of …, a median safety margin of …, and 75% probability of undervaluation — but with a P10-P90 range from … to …, i.e. a distribution where the bottom quarter of scenarios earns nothing. Triangulation of the five models runs from … (bear: add-on at 3x, the June 2029 cliff, oritavancin impaired) to … (bull: add-on at 5x, REZZAYO in prophylaxis, DefenCath in TPN).

Verdict: SPECULATIVE BUY, small position (1/3 of the standard slice), with a waiting trigger. The cheapness is real only to the extent the 2027 add-on approaches the top end of the company’s estimate. The verification moment isn’t far off: the CMS final rule for CY2027 comes out in November 2026, and the Q3 reporting on November 16, 2026 will show the first full post-TDAPA quarter. Whoever buys now pays for the right not to wait for those two dates; whoever waits probably pays more, but with one of the two major uncertainties resolved. This isn’t a full-size position: earnings quality is medium, the accounting has three soft spots documented below, and the securities litigation is entering the expert-report phase.