MNOV — MediciNova, Inc.

Company context · prepared 2026-08-13 · company sweep 2026-08-13 · USD · framework v5.6.1

How to read this. Written for a reader who has not studied pharmacology or finance. General terms are defined once in framework/04-glossary.md. Tags: [VERIFIED — source] / [UNVERIFIED] / [WEB ESTIMATE — source, date].

Program analyses for this company are the sub-folders listed in C.2.

C.0 Company-tier coverage — CLEARED

One row per mandatory tool in the company-tier table in framework/02-connectors.md, in the order that file lists them.

No mandatory row reads NOT CALLED, so the company tier is CLEARED. The regulatory tier of 02-connectors.md was not called at all, correctly: MNOV’s catalysts are trial readouts, not regulatory decisions, and that tier binds only a regulatory event.

C.1 What the company is

MediciNova is a drug-development company in La Jolla, California, incorporated in 2000 and listed on the Nasdaq Global Market. [VERIFIED — BPIQ fetch_company_info, read 2026-08-13] It does not discover its own molecules. It licenses late-stage-ready compounds from other companies and runs clinical trials in them, which is why almost the whole pipeline in C.2 is two molecules — MN-166 (ibudilast) and MN-001 (tipelukast) — both licensed from Kyorin Pharmaceutical of Japan, in October 2004 and March 2002 respectively. [VERIFIED — MediciNova 10-K for FY2025, filed 2026-03-10]

It has no approved product and no product revenue. Reported revenue for 2025 was $409,657, which is not product sales. [VERIFIED — 10-K FY2025, consolidated statements of operations] It is not a single-asset company in the usual sense — ten pipeline rows, two clinical molecules — but it is close to a single-molecule company: eight of the ten rows in C.2 are MN-166.

The share price at the sweep was $1.33. [VERIFIED — BPIQ fetch_company_info, read 2026-08-13]

C.2 Pipeline — every program

One row per row returned by fetch_company_drugs on 2026-08-13, including the ones with no catalyst. The catalyst is reported as catalyst_date_text; the “to a day?” column is yes only when that text names a day (01-rules.md rule 23).

[VERIFIED — BPIQ fetch_company_drugs, read 2026-08-13] for every field in the table.

No row is marked Failed. That is what the connector returns, and it is worth naming as an absence rather than a clean record: the BPIQ feed carries no Failed stage label for this company even though C.7 records a Phase 2b in alcohol dependence that missed its endpoint outright on 2023-06-29, and a Phase 2 in idiopathic pulmonary fibrosis that showed “no significant clinical trends” on 2021-08-12. Both those programs are still on the pipeline above, as Update rows with catalyst_date_text: TBA and no catalyst. [VERIFIED — BPIQ fetch_company_historical_catalysts, read 2026-08-13] A reader should treat “TBA, no catalyst” on this ticker as covering both “not scheduled yet” and “quietly stopped”, because the feed does not separate them.

Three rows carry a catalyst, and all three land inside the same half-year. MN-001 (16136) is guided to Q3 2026; COMBAT-ALS (14393) to year-end 2026; the OXTOX neuropathy trial (14184) to H2 2026. None of the three names a day, so all three are period placeholders (rule 23). The consequence for the analysed program is worked through in its own Attribution section, which the clustering pass computes rather than estimates.

One row is partnered rather than run by MediciNova. SAR444836 (17735) is a gene therapy for phenylketonuria developed by a collaborator, Sanofi’s Genzyme, with the first patient dosed 2023-10-26. [VERIFIED — BPIQ fetch_company_drugs note field, read 2026-08-13] MediciNova’s economics in it are not established by this sweep. [UNVERIFIED]

C.3 Financial position

What this check does and does not cover. It establishes that the runway comfortably clears every catalyst in C.2, on either burn figure. It does not establish the cash balance as of today: the newest figure is 2026-03-31, a Q2 2026 report was previewed in the press feed on 2026-08-08 (“MNOV Q2’26 Earnings: revenue estimate is 75.00K USD”), and that report had not been filed at the time of this sweep. [VERIFIED — BPIQ fetch_company_press_releases, read 2026-08-13; EDGAR submissions feed shows no 10-Q after 2026-05-14]

Dilution history. Every raise in the last 24 months, plus the one authorisation the company sought and did not get.

Is a raise before the next catalyst likely? Probably not, and the evidence is unusually clean for once. Two facilities together authorise $80.0 million of stock sales. Across the eighteen months they have existed, $0.2 million has actually been drawn — the share count moved from 49,046,246 to 49,221,246, a rise of 0.36%. [VERIFIED — EDGAR XBRL EntityCommonStockSharesOutstanding, the 2024-11-13 and 2026-05-12 readings] A company with $27.3 million of cash, a runway past 2027 and a $1.33 share price has strong reason not to sell stock here, and has behaved that way.

The June 2026 shareholder vote does not block the facilities, and it would be wrong to read it that way. Authorised stock stays at 100,000,000 shares against 49,221,246 outstanding, so roughly 50.8 million shares of headroom remain before options and reserves — about $67 million at $1.33, which is more than the $50 million at-the-market facility could sell. [UNVERIFIED — arithmetic on the verified authorised and outstanding counts] What the vote establishes is that shareholders declined to grant a much larger cushion, which is a governance fact rather than a financing constraint.

C.4 Valuation frame

Why this row was computed by hand and not from the price cache. data/prices/MNOV.json does not exist — MNOV is not yet in the committed daily cache, which covers twelve other tickers. C.4 of 03a-company-spec.md provides for exactly this: “Recompute (last − 52wk_low) / (52wk_high − 52wk_low) by hand only for a ticker the cache does not cover.” The consequence is that the 52-week range here is BPIQ’s, read at one moment, rather than derived from a series that can be recomputed for a past date. Populating the cache is scripts/fetch-prices.mjs’s job and sits outside an analysis run.

This is the run-up baseline for every program under this ticker. At 20.3% of its 52-week range the stock is nearer its low than its high, and it has drifted down through the year from $1.96. Nothing about the price says the market is positioned for a positive readout.

C.5 Ownership and flow

Short interest has tripled in six weeks, from a tiny base. The four most recent settlements read 102,466 shares (2026-06-15), 204,812 (2026-06-30), 310,377 (2026-07-15) and 346,440 (2026-07-31) — a 3.4× rise. [VERIFIED — FINRA consolidatedShortInterest, three queries, read 2026-08-13] Read that carefully in both directions. As a percentage of the company it is still trivial: 0.7% of shares outstanding is not a crowded short. As a multiple of the stock’s own liquidity it is not trivial at all: 8.75 days of average volume, up from 2.95 days six weeks earlier. Someone has taken a position against this stock into the guided readout window, and closing it would take almost two trading weeks at recent volumes.

Insider flow is grants and nothing else. Every one of the 250 returned transactions since 2022-11-28 is an award of stock options at a $0.00 transaction price, not a purchase. The most recent are three director grants of 44,500 options each on 2026-06-23 (Nagao, Lemerond, Beaver), and employee option grants on 2026-01-16 of 450,000 to Yuichi Iwaki, President and Chief Executive Officer, and 350,000 to Kazuko Matsuda, Chief Medical Officer. [VERIFIED — BPIQ fetch_company_insider_transactions, read 2026-08-13; the three 2026-06-25 Form 4 filings confirm the June grants in the EDGAR feed]

There has been no open-market purchase by any insider since 2020-03-13, when the Chief Medical Officer bought 2,000 shares at $3.62 and the Chief Executive bought 20,000 shares at $3.10 the day before. [VERIFIED — same source] There has been no open-market sale since 2022-11-28 either, when a director sold 6,754 shares at $2.25. [VERIFIED — same source] So there is no conviction buying to point at, and equally no insider distribution: the flow is simply absent. Every recent buying price on record is 2.3× to 2.7× the current $1.33.

The connector and the filing feed agree here, which is worth stating because 02-connectors.md records a case (TLSA) where fetch_company_insider_transactions returned zero rows while Form 4s were visible. On MNOV the connector’s June 2026 grants match three Form 4 accessions dated 2026-06-25 in the EDGAR submissions feed, and the January 2026 grants match two Form 4s dated 2026-01-20. [VERIFIED — EDGAR submissions, read 2026-08-13]

Concentration versus diffusion. With 2.9% inside, 11.2% institutional and zero hedge funds, this is a diffuse retail register. Nobody large is positioned either way. That cuts both ways for a catalyst: there is no big holder whose exit could break the stock, and equally no institutional base to absorb selling or to add on good news.

C.6 Options chain and its readability

Plain takeaway: this chain cannot price an event, and it is not a marginal call. The lowest listed strike on any expiry is $2.50, against a $1.33 share price. There is no strike within $1.17 of spot — the nearest one is 88% above it. Total open interest across all four expiries and all six strike/side combinations is 78 call contracts and 10 put contracts. Every put on every expiry reports implied volatility of exactly 0.01488, which 02-connectors.md identifies as the floor artifact and instructs be treated as null; every put also reports a delta of exactly −1, which is not a real reading for a strike 88% out of the money. No at-the-money straddle can be constructed because there is no at-the-money strike. Any expected-move figure taken from this chain would be invented.

What a program analysis should do instead. Give a bracket from this ticker’s own past catalyst moves in C.7, and say the chain is unusable. Do not quote a percentage from an option price on this name.

C.7 Reaction to past catalysts

Intraday moves from fetch_company_historical_catalysts, each cross-checked against the press-release feed for the same date.

[VERIFIED — BPIQ fetch_company_historical_catalysts, read 2026-08-13] for every date, event text and move figure. [VERIFIED — BPIQ fetch_company_press_releases, read 2026-08-13] for every same-day cross-check.

The pattern across eleven rows. Nine of the ten with a recorded move are positive, including both outright failures. Not one is worse than −4.5%. The largest moves attach to one-of-two-co-primary announcements (+25.2% open in 2018, +14.9% open in 2022) rather than to clean wins. The plainest reading is that this stock has been persistently under-owned and cheap enough that headlines of almost any kind produce a small bid, and bad news produces no meaningful selling because so little was priced in. [UNVERIFIED — interpretation across eleven rows] It is a reason to bracket a catalyst move rather than to point at one number, and a reason to be sceptical of a large negative scenario on this ticker.

C.8 Company data-quality flags

• market_cap is computed off the previous close, confirmed on this ticker. 02-connectors.md records this bug as found on TLSA (2026-08-07). MNOV is a second, exact confirmation: BPIQ’s market_cap of 64,972,048 divided by the EDGAR-filed share count of 49,221,246 gives $1.3200, precisely the reported previous_close, not the $1.33 last_price. Every market-capitalisation and enterprise-value figure in C.4 is therefore recomputed from the filed share count rather than read from the field. [VERIFIED — BPIQ fetch_company_info and EDGAR XBRL, both read 2026-08-13]
• max_52_week_position carries the same bug, and is not read. BPIQ reports 0.18987; ($1.32 − $1.17) ÷ ($1.96 − $1.17) = 0.1899 exactly, confirming it uses the previous close. C.4 reports 20.3% off the last price instead. [VERIFIED — same sources]
• No price cache exists for MNOV. data/prices/MNOV.json is absent, so C.4’s 52-week range is a point-in-time BPIQ reading rather than a figure derived from a committed daily series. This is the documented fallback in 03a-company-spec.md C.4, not an error, but it does mean the range cannot be recomputed for a past date and a run-up settlement on this ticker cannot be performed until the cache is populated.
• Cash and burn are 4.5 months stale. finance_updated_at is 2026-03-31 and a Q2 2026 report was previewed for 2026-08-08 but not yet filed at sweep time. The gap matters less than usual here because the runway is long on any figure, but no cash number in C.3 is current.
• insider_own and tute_own carry no timestamp. Only short_data_updated_at (2026-07-31) is dated, and it governs short_float alone. The ownership percentages in C.5 are therefore of unestablished vintage.
• Options implied volatility is unusable at every strike on every expiry. All ten put contracts sit on the documented 0.01488 floor artifact with delta exactly −1; the call implied volatilities of 0.84 to 4.91 are at strikes 88% to 464% out of the money. C.6 treats the whole chain as null rather than the artifact rows only.
• short_float units remain unconfirmed and are not relied on. BPIQ reads 0.0082; the FINRA count divided by the EDGAR share count gives 0.704%. The two are close, but only the second has known units, and C.5 uses it.
• The pipeline feed has no Failed label for a company with two failed programs. See the note under C.2. This is a limitation of what the connector returns, not a claim about the company, and it means the pipeline table cannot be read as a discontinuation record.
• Drug names are dirty on this ticker too. MN-001 (tipelukast) (row 14900) carries a trailing space while row 16136 does not, and eight rows share the identical name MN-166 (ibudilast). Nothing here keys on the name; bpiq_drug_id is the key throughout.