DNTH · Dianthus Therapeutics, Inc. · Stock research

Current Thesis

The narrative leg on offer is a single-molecule complement re-rate that has now cleared every structural level that previously defined it. Claseprubart (DNTH103), a selective antibody binding only activated C1s to shut the classical complement pathway while sparing the lectin and alternative arms, carries CIDP, gMG and MMN on one fuse. Since the last refresh the stock has extended from the July breakout to a 52-week high of $112.64, closing at $108.92 on 2026-08-07, 3.3% below that high, with a three-month return of +25.6% and RSI(14) at 59.5.

What changed in August is the newsflow around the move, not the data underneath it. The 2026-08-04 Q2 print reaffirmed the pipeline and added two things the July note did not have: MoMeNtum (MMN) enrollment completed at 46 patients against a 36-patient target, with top-line "on track for December '26," and a new candidate, DNTH312, a bifunctional fusion of claseprubart and TACI, aimed at Phase 1 readiness by year-end 2027. Between 2026-08-03 and 2026-08-06 four brokers raised targets into the high: Evercore ISI to $134, Wedbush to $122, Truist to $128, HC Wainwright to $158. All four post-date the June–July leg off roughly $82.

One piece of guidance language deserves precision. The Q2 release states Part B "top-line guidance expected by YE'26" — that is guidance about when the randomized read lands, expected by year-end 2026, which is not the same object as top-line data by year-end 2026. Until the company dates it, the binary the whole valuation rests on has no calendar entry.

Bullish and bearish views on Dianthus Therapeutics, Inc.

The model's bull view on Dianthus Therapeutics, Inc. (DNTH), in brief: New 52-week high, structure intact. $112.64 print, $108.92 close on 2026-08-07, extending above the July breakout zone (2026-07-09 high $102.24, close $100.03) and far above the $81.00 March offering print. CAPTIVATE Part A interim held up in the Q2 restatement. 75% response… The bear view: The mechanism overhang has not been retired. Both cases follow in full.

Bull Case

• New 52-week high, structure intact. $112.64 print, $108.92 close on 2026-08-07, extending above the July breakout zone (2026-07-09 high $102.24, close $100.03) and far above the $81.00 March offering print.
• CAPTIVATE Part A interim held up in the Q2 restatement. 75% response rate in the first 40 Part A completers versus a ≥50% target, "consistent and clinically meaningful across multiple efficacy measures," with no related serious infections, no clinical symptoms of drug-induced lupus, no related SAEs and no safety discontinuations (2026-08-04 release).
• A nearer readout now exists. MoMeNtum in multifocal motor neuropathy finished enrolling at 46 patients (target 36) with top-line guided to December 2026 — the first same-molecule randomized-era datapoint before Part B.
• Funded through everything. ~$1.2B cash at 2026-06-30, runway guided into 2030, against Q2 R&D of $48.7M and G&A of $13.6M.
• Pipeline optionality added, not subtracted. DNTH312 announced 2026-08-04 (aC1s plus BAFF/APRIL via TACI); DNTH212 Phase 1 healthy-volunteer data anticipated by year-end 2026; EMERGE gMG Phase 3 initiated June 2026.
• Sell-side raising into strength. Four targets lifted 08-03 through 08-06, all at $122 or above, versus the $117.82 average that stood on 2026-07-09.

Bear Case

• The mechanism overhang has not been retired. Sanofi's riliprubart, also a selective anti-activated-C1s antibody, had MOBILIZE halted on 2026-06-10 for insufficient efficacy in refractory CIDP. CAPTIVATE's GO decision came from open-label responder data with no placebo arm. Only randomized Part B settles it, and Part B has no date.
• Guidance about guidance. The Q2 wording promises Part B timing by year-end 2026. A November or January update that dates top-line in 2H'27 would push the binary a full year without a single negative datapoint.
• Q2 P&L deteriorated. Net loss $50.2M, or $(0.90) per share, versus $31.6M / $(0.88) in Q2 2025, and $(0.90) missed the $(0.88) consensus reported on 2026-08-04. Revenue of $0.761M against a $0.259M estimate is collaboration income, not a commercial base.
• Shelf take-down risk is unresolved. The March 2026 raise printed at $81.00 (8,470,989 shares plus 402,468 pre-funded warrants, ~$719M gross). No follow-on or ATM prospectus supplement has surfaced since. With the stock 34% above that print, an equity take-down near highs remains an open possibility rather than a closed chapter.
• Aggregator consensus is unreliable here. Provider averages for DNTH diverge sharply as of 2026-08-08 — ChartMill shows $128.30 across 23 analysts while MarketBeat shows $60.88 across 11. The individually dated broker actions are the usable read; any single "average PT" cited for this name should be checked against its analyst count.
• gMG is not a 2026 driver. EMERGE top-line is guided to 2H'28.

Setup & Price Structure

Last close $108.92 on 2026-08-07, 52-week high $112.64, 3.3% below it. RSI(14) 59.5. Three-month return +25.6%.

The reference levels are stacked and dated. $81.00 is the March 2026 institutional cost basis and the shelf the stock based on through Q2. Roughly $82 was the 2026-06-18 pre-breakout base. $100.03/$102.24 is the 2026-07-09 breakout close and high. Above that, August has been extension without acceleration — an RSI of 59.5 while sitting 3% under a fresh 52-week high is a flatter, lower-slope advance than the June–July leg produced, which is an inference from the momentum reading rather than a measured flow figure.

Life-cycle placement: MATURING. The narrative is fully identified — 23 analysts on ChartMill's count as of 2026-08-08, four target raises compressed into 08-03 through 08-06, all landing after the price move rather than before it. Price is still making highs, so it is working; the incremental headline supply is broker catch-up and a preclinical candidate with a 2027 Phase 1 target, not new clinical evidence. That is a well-known story still functioning on moderating fuel, which is the MATURING profile rather than the ACCELERATING one that described the 2026-06-18 to 2026-07-09 breakout.

Crowding and positioning observables, stated as observables: four PT raises in four consecutive sessions into a new high; the highest of them ($158, HC Wainwright, 2026-08-06) sits 45% above the 2026-08-07 close; price 3.3% below the 52-week high with RSI under 60; a cash-burning issuer with $1.2B on hand and an unexercised capacity to sell stock near round-number highs; and a completed Q2 print, meaning the next scheduled company disclosure is roughly three months out.

Catalyst Calendar (next 30 days)

• No dated company event falls between 2026-08-08 and 2026-09-07. The Q2 print cleared on 2026-08-04; the next scheduled disclosure is the Q3 report. This window is a data vacuum by the company's own guidance.
• ~2026-11-05 (est.) — Q3 2026 results. Q1 landed 2026-05-05 and Q2 on 2026-08-04, so early November is the pattern-implied date. The material question is whether Part B top-line timing gets a calendar entry.
• ~2026-12-31 (est.) — CAPTIVATE Part B top-line guidance expected by YE'26 per the 2026-08-04 release.
• ~2026-12 (est.) — MoMeNtum MMN top-line, 46 patients, "on track for December '26."
• ~2026-12-31 (est.) — DNTH212 Phase 1 healthy-volunteer data "anticipated by year end."
• 2H'28 — EMERGE gMG Phase 3 top-line. Not a 2026 or 2027 driver.

What Would Change Our Mind

The structure that defines this leg is the July breakout. Losing it takes the stock back inside the $82–$100 pre-breakout range and turns the August extension into a failed high. The gradeable form of that: a weekly close below $100.

Three non-price conditions would independently break the frame. First, a company update that dates CAPTIVATE Part B top-line in 2H'27 or later — the "guidance by YE'26" phrasing leaves room for a push and the market is currently pricing a randomized win it cannot yet see. Second, the December MoMeNtum MMN readout coming and going without a clean primary hit; MMN is a small indication but the molecule is the same one carrying CIDP and gMG, so a miss reprices all three. Third, a 424B5 prospectus supplement or ATM sales notice priced near current levels, which would confirm that the run was treated as a financing window.

Conversely, a Part B date announced inside 2026 with an unchanged design, or a December MMN primary hit, would convert this from a pre-data re-rate into a partially de-risked one and reset the level structure higher.

Correlation Notes

• Sanofi read-through remains the dominant single-name correlation. The 2026-06-10 MOBILIZE halt in refractory CIDP is the closest mechanistic analogue; DNTH sold off on it and had recovered to $102.24 by 2026-07-09. Any further riliprubart disclosure — subgroup data, program discontinuation, or a competing classical-pathway result — moves DNTH regardless of Dianthus newsflow.
• The approved-FcRn complex sets the bar. CIDP already has an approved FcRn option in argenx's Vyvgart Hytrulo; peak-sales models for claseprubart in CIDP are built off share assumptions against that class, so FcRn commercial datapoints feed directly into DNTH estimate revisions.
• Clinical-stage biotech beta. With no product revenue and $1.2B of cash, the name trades as a duration asset: sector risk appetite and the rate path drive the multiple between data events, and the next data event is guided to December at the earliest.
• Single-molecule concentration. CIDP, gMG and MMN are three indications on one antibody. The correlation between the three programs inside this company is effectively one, which removes the diversification cushion that a multi-asset clinical biotech would carry into a randomized readout.