CMPX — Q2 2026 Financial Report Analysis

Spending rose ahead of a drug filing that the FDA has since pushed back on

Compass Therapeutics is a Boston biotech with no products on sale and no revenue: it is testing four antibody drugs against cancer and pays for that work with money raised from investors. In the second quarter of 2026 (April to June) its net loss widened 26.5% to $25.2 million, from $19.9 million a year earlier, because it spent more on its lead drug, tovecimig, and began paying for the groundwork to sell it. It ended June with $179.9 million in cash and short-term investments, which management says will fund the company "into 2028."

The quarter's numbers matter less than what happened after it closed. The 10-Q says the company expected FDA feedback in the third quarter "prior to a potential BLA filing later this year" (a BLA, or Biologics License Application, is the formal request to approve a drug for sale). On September 22, 2026, Compass disclosed that the FDA had recommended it run a trial showing tovecimig helps patients live longer before filing. Compass says it disagrees and still intends to prepare a filing, but the near-term path to a first product is now much less certain than the spending plan in this quarter assumed.

At a glance

• $27.0 million in operating expenses, up 28.1%. With no revenue, this is the cost of running the company for three months; most of the increase went to tovecimig and to preparing for a commercial launch.
• $179.9 million in cash and investments at June 30, down $29.0 million in six months. At the first half's pace of cash use ($32.0 million), that is roughly five and a half half-years of funding, consistent with management's "into 2028" estimate.
• Loss per share fell to $0.13 from $0.14 even though the total loss grew. That is only because there are about 35% more shares after an August 2025 stock sale; each shareholder owns a smaller slice of a larger loss.

The quarter in numbers

For the first half of the year, the net loss was $43.5 million versus $36.5 million, and loss per share was $0.23 versus $0.26.

Where the money went

Tovecimig took the larger share of R&D. R&D (research and development, the cost of running clinical trials, making trial drug supply and paying the scientists) rose $3.2 million. The company says the increase was "primarily driven by $2.6 million related to non-personnel project expenses for tovecimig." Spending on tovecimig climbed to $12.3 million from $8.1 million, and it now makes up 63% of all R&D, up from 49% a year earlier. Spending on two earlier-stage drugs fell: CTX-471 dropped to $1.5 million from $2.9 million and CTX-10726 to $1.3 million from $2.1 million. CTX-8371 was roughly flat at $1.7 million.

G&A jumped 59% as the company prepared to sell a drug. G&A (general and administrative: executives, finance, legal and other overhead) rose $2.8 million, "primarily due to an increase of $1.4 million of pre-commercialization expenses and an increase of $0.8 million of stock-based compensation expense." Over the six months, G&A included $1.8 million more in pre-commercialization spending and $1.3 million more in personnel costs "related to commercialization." In other words, Compass was hiring and spending for a launch it expected to follow a filing in late 2026.

Interest income partly offset the higher spending. The company earned $1.8 million on its investments, up from $1.2 million, because it held more cash after the August 2025 stock sale.

The pipeline: what the quarter's data showed

• Tovecimig in biliary tract cancer (BTC). BTC is a rare cancer of the bile ducts and gallbladder. In the Phase 2/3 COMPANION-002 trial in patients whose cancer had already progressed on earlier treatment, tovecimig plus the chemotherapy paclitaxel shrank tumors in 18.0% of patients (20 of 111) versus 5.3% on paclitaxel alone, which was the trial's main goal. Median time before the cancer grew again (progression-free survival) was 4.7 months versus 2.6 months. Overall survival, how long patients lived, was not statistically better. The company attributes this to 54% of control-arm patients switching ("crossing over") to tovecimig and then living notably longer, which blurs the comparison.
• CTX-8371 (a PD-1 x PD-L1 antibody, designed to release the immune system's brakes on two fronts). Early dose-finding data presented at ASCO in May 2026 showed 3 responses among 15 evaluable patients, including 2 of 6 at the two highest doses. Expansion groups in triple-negative breast cancer, Hodgkin lymphoma and non-small cell lung cancer are enrolling, with more data expected in the fourth quarter of 2026.
• CTX-10726 (a PD-1 x VEGF-A antibody). First patients have been dosed in a dose-finding study, with data expected in the fourth quarter of 2026.
• Investigator-run studies of tovecimig are under way in first-line BTC, second-line colorectal cancer and, combined with CTX-471, in glioblastoma.

What the headline numbers hide

• The per-share improvement is dilution, not progress. The net loss grew $5.3 million, but loss per share fell a cent because the weighted share count rose to 186.8 million from 138.3 million. That count includes 6.71 million pre-funded warrants (rights to buy shares for $0.0001 each, so they are treated as shares already issued) from the August 2025 offering, which raised $129.3 million net at $3.00 per share. The total loss is the better gauge of how fast the company is spending.
• Cash burn is smaller than the accounting loss, mainly because of stock pay. Over six months, cash used in operations was $32.0 million against a $43.5 million net loss. The biggest gap is $9.9 million of stock-based compensation, which costs no cash but does dilute shareholders; that is up from $5.3 million a year earlier. Stock pay was $4.8 million in the second quarter alone versus $2.8 million a year ago, so it accounts for about a third of the rise in operating expenses.
• Last year's cash flow was flattered by unpaid bills. In the first half of 2025, accrued expenses rose $4.7 million, holding back cash outflow; in 2026 that line rose only $0.2 million. So part of the 28% jump in cash burn is timing, not a step-change in activity.
• No new money was raised. Compass sold no shares through its at-the-market program (a facility that lets it sell stock gradually on the open market) in the first half. The only inflow was $3.9 million from employees exercising stock options. The $29.0 million fall in cash and investments therefore reflects actual spending.
• No one-off items, no debt. The loss is entirely operating spending minus interest income; total liabilities are just $24.1 million, mostly bills payable, accruals and an office lease.

Takeaway: Compass spent the second quarter building toward a 2026 drug filing that the FDA, a month and a half after the 10-Q was filed, said should wait for a new trial showing patients live longer. The $179.9 million cash pile buys time "into 2028," but the commercialization costs now running through G&A were sized for a launch that may be years away, so how quickly Compass trims or redirects that spending is the next thing to watch.

Outlook

Management's guidance. The 10-Q says existing cash "will enable us to fund our operating expenses and capital expenditure requirements into 2028," with the usual warning that the estimate rests on assumptions that "may prove to be wrong." The company also states it will need "substantial additional funding" to finish developing and commercializing its drugs. There is no going-concern warning (the formal accounting flag that a company may not survive the next 12 months).

Our read. The financial runway is not the immediate problem: at the first half's pace of about $16 million a quarter, $179.9 million lasts close to three years. The problem is what the money is buying. If the FDA's position holds, Compass faces three choices: run a new survival trial (more years and a larger budget, very likely requiring a new stock sale), file anyway against the FDA's advice (high risk of rejection), or cut back pre-launch spending and lean on the earlier-stage antibodies. Watch for:

• Further FDA discussions and whether Compass still files a BLA, or announces a new trial instead.
• The full COMPANION-002 presentation at the 2026 ESMO Congress, including the survival subset analyses management cited.
• Q3 2026 results (10-Q expected in early November): whether G&A pre-commercialization spending keeps rising or is cut, and any change to the "into 2028" runway.
• Fourth-quarter 2026 data from the CTX-8371 expansion cohorts and the CTX-10726 dose-finding study, which now carry more weight for the company's value.