BYSI — Q2 2026 Financial Report Analysis

Overview

BeyondSpring is a clinical-stage cancer drug developer with no product revenue. Its whole case rests on one drug, Plinabulin, used together with the chemotherapy drug docetaxel in lung cancer that has kept growing after immunotherapy. In the second quarter of 2026 (April–June), the core business lost $1.75 million, down from $1.88 million a year earlier. Spending stayed flat at a very low level: research and development (R&D) was $0.97 million and general and administrative costs were $0.76 million. Including SEED Therapeutics, the protein-degradation spin-out it is selling and now reports as "discontinued operations", the group lost $5.7 million. Reported loss per share was $0.02, half of last year's $0.04, but that improvement mostly comes from how the loss was split between BeyondSpring's shareholders and SEED's minority shareholders, not from lower spending (see below).

The quarter matters less than what happened after it. By June 30 the core business had only $6.5 million of cash and short-term deposits, and the filing carries a going-concern warning. In late September the company did three things. It sold its Chinese subsidiary to an investor that will pay for the China half of the planned Phase 3 trial in exchange for the shares. It received FDA Fast Track designation. And it set up a program to sell up to 9.2 million new shares on the market. Two Nasdaq listing notices followed in the same weeks.

At a glance

• $6.5M cash and short-term deposits (continuing operations, June 30), down from $12.6M at the end of 2025. That is about half the money gone in six months. Pro forma for the China sale, only $1.46M of cash stays with BeyondSpring.
• $0.56M clinical spend in Q2, up from $0.13M. Drug manufacturing "to prepare for potential future study initiation" is the only line that grew. In practice that means the new Phase 3 trial, DUBLIN-4.
• 9.2M shares on offer through the new at-the-market program, or 22% of the 41.1M shares outstanding. Buyers of the new shares are the company's main remaining source of cash, unless the $28M of pending SEED share sales arrives.

Q2 2026 results

*Derived: total operating cash outflow ($7.05M vs $10.07M) minus the amount the filing attributes to discontinued operations ($4.55M vs $5.79M).

The R&D total stayed flat, but what it pays for changed. Clinical costs rose by $0.43 million. The 10-Q attributes the increase to "drug manufacturing activities to prepare for potential future study initiation". Professional fees fell from $0.37 million to $0.09 million and staff costs fell from $0.35 million to $0.27 million, which offset it. G&A fell because of "lower legal fees and lower consulting costs". For the half year it was down 29%, to $1.91 million, on lower incentive and share-based pay. At about $1.7 million of operating costs a quarter, the company is running on a skeleton budget for a firm with a Phase 3 trial ahead of it.

What the headline numbers hide

The halved per-share loss is largely an accounting split, and part of that split was wrong. BeyondSpring owned only part of some subsidiaries, and outside owners (called "noncontrolling interests") absorb their share of those losses. In Q2 those outside owners were assigned $0.84 million of the continuing-business loss, against $0.07 million a year earlier. That is what cut the loss attributable to BeyondSpring to $0.85 million. In the pro forma financial statements filed on October 6, the company disclosed a $754,000 error in that split for the first half of 2026. Corrected, the half-year continuing loss attributable to BeyondSpring is $3.67 million ($0.09 per share), not the reported $3.13 million ($0.08). The total loss is unchanged. Only who bears it changes.

Cash: the real number is smaller than $6.5 million. In the pro forma balance sheet (a restated version showing how the June 30 balance sheet would have looked if the China sale had already happened), cash falls to $1.46 million and the $3.83 million of short-term deposits disappears. Those deposits were held by the entities that were sold. Continuing operations spent about $1.25 million a quarter in cash in the first half. That rate was held down by what the filing calls "a strategic extension of payment terms with vendors to manage short-term liquidity requirements", meaning the company paid suppliers more slowly. On those figures, the pro forma cash covers roughly one quarter of current spending, before DUBLIN-4 costs outside China ramp up. This is our estimate. The company gives no runway figure.

Going concern. The 10-Q states that losses, $4.3 million of net current liabilities and limited cash "raise substantial doubt regarding the Company's ability to continue as a going concern". In plain terms, the company itself cannot say it has enough money to keep operating for the next year without raising more.

The China sale fixes the balance sheet on paper, not in cash. Hengrui paid an RMB 200 million (about $29 million; carried at $29.5 million on June 30) upfront fee to the Chinese subsidiary, and that sat on the books as "deferred revenue", a liability. It leaves along with the subsidiary. The pro forma shows shareholders' equity swinging from a $36.0 million deficit to $17.3 million positive. That comes from a preliminary $53.7 million gain: $34.9 million of net liabilities removed, plus the retained SEED stake marked at an $18.8 million fair value. No cash changes hands. The buyer's commitment to fund the China portion of the trial is the price. BeyondSpring also gives up Plinabulin's Greater China rights. It keeps the rest of the world, plus a licence to the China trial data.

Dilution. The at-the-market (ATM) program lets the company sell up to 9.2 million new shares into the market through Citizens JMP, which takes up to a 3% commission. That would be 22% more shares. Nasdaq says the stock closed below $1.00 for 30 straight business days up to mid-September. At those prices, the full 9.2 million shares would raise less than $9.2 million before fees.

The SEED money is the big swing factor. The 10-Q still expects $28.07 million from the second and third closings of the SEED preferred-share sale agreed in January 2025. The second closing is "expected to be completed in 2026" and the third is due "no later than December 15, 2026". Both carry mutual termination rights if they slip. The October pro forma does not update that figure. Separately, Eli Lilly ended its research collaboration with SEED effective August 20, 2026.

Pipeline and strategy

• DUBLIN-4 is a global Phase 3 trial: the large, randomized study a regulator needs before approving a drug. About 442 patients are planned, split 1:1 between Plinabulin plus docetaxel and docetaxel alone. The main goal is overall survival. A prespecified interim analysis comes at 221 progression-free-survival events, meaning cases where the cancer worsens or the patient dies. The 10-Q (August) still described the trial as one the company "plan[s] to launch". The September agreement calls it the "Ongoing Trial". No enrollment numbers have been disclosed.
• Fast Track designation (announced September 29) covers this exact patient group. It allows more frequent FDA meetings and a rolling application. It does not make approval more likely, and the company's own release says so.
• The China partner is expected to enroll about 221 patients, roughly half the trial, within three years. If it falls short, it must return shares in proportion to the shortfall, and all of them if the shortfall is 90% or more.
• Supporting data: DUBLIN-3 (559 patients, published in Lancet Respiratory Medicine in 2024) showed a survival benefit in non-squamous patients (hazard ratio 0.72, meaning a 28% lower risk of death over the study period). A post-hoc look at patients who had already had immunotherapy showed median survival of 15.8 months versus 11.7 months. Post-hoc means the analysis was done after the fact, which is weaker evidence than a pre-planned one.
• Leadership: Min Qiu became CEO on July 1, 2026, succeeding co-founder Dr. Lan Huang.

Takeaway: The narrower Q2 loss is not the story. BeyondSpring has handed half of its Phase 3 trial, and its China rights, to a partner because it could not pay for that half itself. Pro forma cash of about $1.5 million covers roughly one quarter of its current, deliberately minimal spending. Whether DUBLIN-4 reaches its interim readout now depends on two outside sources of money: the $28 million SEED share sale closing by December 15, and selling new stock that trades below $1.

Outlook

Management gives no financial guidance. Its stated milestone is the DUBLIN-4 interim analysis at 221 PFS events, with no date attached. Three things to watch in the Q3 10-Q (likely mid-November):

1. The deconsolidation gain and the corrected equity figure. Removing the sold businesses from the accounts should take the company out of a shareholders' deficit. That matters for Nasdaq's equity-based listing tests even though no cash is involved.
2. Whether the second SEED closing happened. Without that cash, the company would have to lean far harder on the ATM.
3. ATM usage and share count. Any sales after September 29 will show up as share-count growth, and their price will show how costly this funding is.

Two Nasdaq deadlines also apply: March 16, 2027 for the $1 minimum bid price, and March 29, 2027 for the $35 million market-value-of-listed-securities test. A reverse stock split would fix the first but not the second. Our read: the trial design and Fast Track status give the drug a credible path, but the company's cash, not the science, is what decides the near term. A funding gap between now and the interim readout is the base case, not the risk case.