Disc Medicine (IRON): FDA Said No — the Stock Went Back Anyway. Is IRON a Buy?

Disc Medicine, Inc. (IRON) Q2 2026 — Reported before the open on July 30 (calendar Q2 ended June 30, 2026). There is no revenue line: R&D $46.9M (+1.3%), SG&A $18.1M (+20.2%), total operating expenses $65.1M, net loss $59.5M vs $55.2M, loss per share $(1.54) vs $(1.58) on 38.69M weighted shares. Cash, equivalents and marketable securities $717.7M, runway guided into 2029. The stock closed $73.67 on July 29, rose 7.5% to $79.21 on the print, and finished July 31 at $77.00. The line nobody printed: first-half cash used in operations was $106.4M, up 19.2% from $89.3M — the burn GREW after February's restructuring. Cash fell only $73.4M because $30.0M of the $34.0M of financing was a drawdown on the Hercules term loan. Part of the comforting cash pile is borrowed at a floor of 8.25%, secured on every asset except the intellectual property. THE CALL: AVOID (3/5, PRICED FOR A TRIAL THAT HAS NOT READ OUT) — base-case value ~$45.0 vs ~$77.0 today. KEY METRICS: - CALL: AVOID 3/5 — fair value ~$45 vs $77.00 (-42%). No revenue means no DCF, so we use a scenario tree on 38.84M economic shares (38.39M common at July 23 plus 445,422 pre-funded warrants). BEAR 40%: APOLLO misses the sunlight co-primary, leaving $654M net cash plus two mid-stage antibodies = $25. BASE 45%: APOLLO hits, bitopertin approved mid-2027, ~$700M peak sales = $50. BULL 15%: APOLLO hits clean AND selcodebart and DISC-3405 deliver = $82. Weighted at an 11% discount rate = $45 ($52 at 9%, $39 at 13%). STREET: Buy — 11 buy / 0 hold / 0 sell, 11 analysts; average target ~$101.60, range $85-$125 (Morgan Stanley $85 Overweight, Stifel $94, H.C. Wainwright $118). We DIFFER and are far more CAUTIOUS. - THE PRINT — THERE IS NO REVENUE LINE: Disc has never recorded a dollar of revenue. Q2 2026 R&D $46.933M vs $46.319M (+1.3%, flat only because more selcodebart and DISC-3405 spend offset less bitopertin manufacturing and a $10.0M milestone in the comparable quarter); SG&A $18.142M vs $15.091M (+20.2%); total operating expenses $65.075M; other income $5.612M (interest on the cash, net of loan interest); net loss $59.532M vs $55.247M; loss per share $(1.54). First half: R&D $92.837M, SG&A $41.762M, net loss $123.036M, $(3.19) per share. Accumulated deficit $633.2M against $1,279.0M of paid-in capital. - WHAT THE FDA ACTUALLY SAID (the whole thesis): the February 13, 2026 Complete Response Letter did NOT dispute the chemistry. The FDA agreed AURORA and BEACON showed bitopertin significantly lowers whole-blood metal-free PPIX (-40.7% at 60mg vs +8.0% for placebo) and conceded strong mechanistic and biological plausibility for the biomarker. It rejected the accelerated-approval filing because the trials 'did not show evidence of association' between percent change in PPIX and the sunlight-exposure endpoints. In AURORA, time in sunlight was a PRE-SPECIFIED KEY SECONDARY endpoint and it MISSED statistical significance on a strong placebo response; only a post hoc MMRM re-analysis was nominally significant (+7.1h p=0.026 at 20mg, +8.0h p=0.013 at 60mg). The Phase 3 APOLLO trial reading out in Q4 2026 carries that same endpoint as a CO-PRIMARY alongside PPIX. The endpoint that already missed once is now one bitopertin must beat outright. - THE BURN GREW AFTER THE RESTRUCTURING: H1 cash used in operations $106.437M vs $89.294M (+19.2%). Cash and marketable securities fell only $73.4M ($791.152M to $717.746M) because financing contributed $34.032M — of which $29.979M was debt, not equity. Q2 opex of $65.075M annualises near $260M. Thirteen days after the CRL, on February 26, the board cut roughly 20% of the workforce, a reduction the 8-K says 'primarily reflects focused workforce reductions among the Company's commercial functions'. H1 SG&A had risen $14.5M YoY, $12.6M of it personnel, explicitly for 'developing our commercialization capabilities'. SG&A then fell from $23.620M in Q1 to $18.142M in Q2, down 23% sequentially. The 'runway into 2029' is therefore computed on a company that is NOT launching anything — if APOLLO succeeds they must rebuild the commercial organisation they just dismantled. - THE COVENANT TIED TO THE SHARE PRICE: the Hercules Capital facility allows up to $200.0M; $60.0M is drawn ($30M in November 2024, another $30M on the June 25, 2026 First Amendment). Interest is the greater of 8.25% or Prime +1.75%, interest-only through November 2028, maturing December 1, 2029, with a 6.75% end-of-term charge ($4.1M) and a first-priority lien on all assets except intellectual property. Principal due: $4.422M in 2028, $55.578M in 2029. The minimum-cash covenant is 'waived during all times that the Company's market capitalisation is greater than or equal to $1.0 billion' — switched off today at ~$3.0B, but it switches ON precisely if APOLLO fails and the equity falls roughly two-thirds. The same amendment pushed the first test date to July 1, 2028. A subjective acceleration clause carries no date at all. - THE PIPELINE IS BETTER THAN THE ONE-ASSET FRAMING: selcodebart (DISC-0974, anti-hemojuvelin) showed a 56% major anaemia response and 72% overall response in the RALLY-MF Phase 2 in myelofibrosis anaemia, consistent across transfusion cohorts and with or without background JAK inhibitor, with hepcidin down more than 75%; presented at both ASCO and EHA, granted EU Orphan Drug Designation, with an end-of-Phase-2 FDA meeting due by year-end and further data in Q4 2026. DISC-3405 (anti-TMPRSS6) completed RESTORE-PV Phase 2 enrolment in polycythemia vera with initial data PULLED FORWARD into Q3 2026, plus a Phase 1b sickle cell readout in Q4. An expanded access programme for bitopertin launched in June. APOLLO itself enrolled 183 patients (expanded from 150) in under a year. - THE PRICE HAS UN-PRICED THE REJECTION: IRON closed at a 12-month high of $94.11 on November 24, 2025. In October 2025 the company sold 2,619,049 shares at $84.00 (plus 59,523 pre-funded warrants) for $211.0M net — four months before the CRL. On February 13 the stock fell 21.9% to $55.95 on 5.0M shares, roughly fourteen times normal volume. It has since recovered 38% to $77.00, leaving it just 8% below that pre-rejection financing price. Beta 2.07; 50-day average $72.48, 200-day $74.77. - VALUATION AND THE IMPLIED PROBABILITY: net cash is $717.7M less $60.0M of debt and the $4.1M end-of-term charge = $654M, or $16.84 a share. At $77.00 the market value is ~$2,991M, so the market pays ~$2,337M for the pipeline. We value the programmes unrisked at $1,150M (bitopertin), $1,000M (selcodebart) and $850M (DISC-3405); at 50%, 32% and 22% odds that is $1,082M risk-adjusted — the market is paying 2.2x our number. Base case: $654M cash + $1,035M bitopertin (unrisked x 90% approval) + $507M for the other two, less $150M corporate overhead PV and $112M expected dilution = $1,934M, or ~$50 a share. THE KEY TEST: assume APOLLO succeeds with CERTAINTY and blend the two success cases 3-to-1 and you still only get ~$58 — 25% below the price. To justify $77 you must put roughly 85% weight on the bull case in which all three programmes work. Our peak-sales assumption is not the disagreement: Morgan Stanley models peak bitopertin sales above $1B by 2035 and our $700M sits inside the Street's range. We differ on probability and discount rate. What to watch: Bullish: a clean RESTORE-PV haematocrit-control signal for DISC-3405 in Q3, a clean end-of-Phase-2 outcome for selcodebart by year-end, or an APOLLO result that hits the time-in-sunlight co-primary outright. Bearish: APOLLO missing the sunlight co-primary even if it hits the PPIX one, an equity raise before the readout, or a further draw on the Hercules facility while $717.7M of cash sits on the balance sheet. Also on YouTube: @ChargedAlpha DISCLAIMER: For informational and educational purposes only. Not financial advice. Do your own research before any investment decision.