CRDF — Q2 2026 Financial Report Analysis

Cardiff Oncology's second-quarter net loss shrank to $9.2 million from $13.9 million a year earlier, but almost all of that improvement came from its mid-stage colorectal cancer trial winding down, not from the business getting cheaper to run. The quarter's real story is money: the company ended June with $34.5 million in cash and short-term investments, the filing still carries a "going concern" warning, and the Phase 3 trial it wants to start in early 2027 is explicitly "subject to securing additional financing." Cardiff has no product on the market; the $104,000 of revenue is a royalty on an old diagnostics license unrelated to its drug.

At a glance

• $34.5 million in cash and short-term investments at June 30, down from $58.3 million at the end of 2025. A July share sale raised another $10.05 million (before fees), and management now says its money lasts "into the third quarter of 2027" — but the 10-Q still says that is not enough to cover the next 12 months.
• R&D spending fell 49% to $5.9 million, because the Phase 2 trial is mostly finished and fewer patients are still on treatment. That spending will have to rise sharply once the planned ~640-patient Phase 3 trial starts.
• Shares outstanding rose 14% in about five weeks, from 68.3 million at June 30 to 77.8 million on August 6, after the July offering — and that offering also handed out warrants for about 9.8 million more.

What Cardiff Oncology does

Cardiff is developing one drug, onvansertib, a pill that blocks PLK1 — a protein cancer cells rely on to divide. Its lead use is first-line treatment (the first therapy a patient receives) for metastatic colorectal cancer in patients whose tumors carry a RAS mutation, a group that responds poorly to some existing targeted drugs. Onvansertib is added on top of standard chemotherapy plus bevacizumab (an antibody that cuts off tumor blood supply), not used instead of it.

Q2 2026 results

Operating margin isn't shown: with revenue of about $0.1 million against $9.7 million of costs, a margin percentage carries no information for a company at this stage.

Where the R&D cut came from. Almost all of it is one line: "clinical trials, outside services, and lab supplies" fell to $3.2 million from $8.3 million. The 10-Q attributes the drop to "the completion of clinical trials, as well as fewer patients still on treatment in our Phase 2 mCRC trial, and a reduction in pre-clinical activities as we focus on our upcoming Phase 3 mCRC trial." Salaries in R&D fell only modestly ($1.85 million vs $2.13 million).

Where the overhead increase came from. Selling, general and administrative (SG&A — the cost of running the company outside the lab) rose $0.5 million in the quarter. Outside services and professional fees nearly doubled to $2.0 million from $1.1 million, which the filing ties to "attorney costs for our intellectual property dispute with NMS" (more on that below). For the first half, SG&A was up $2.6 million, driven mainly by severance: salaries and staff costs in SG&A more than doubled to $3.8 million from $1.8 million after the former CEO (Mark Erlander) and former CFO (James Levine) left under separation agreements signed March 27, 2026. Changes to their stock options added another $0.5 million of non-cash expense.

The trial data behind the plan

At the ASCO cancer conference in June, Cardiff presented results from CRDF-004, a randomized Phase 2 trial with 110 patients (data cut March 18, 2026). In the arm Cardiff picked for Phase 3 — 30 mg onvansertib plus the FOLFIRI chemotherapy regimen and bevacizumab:

• Confirmed objective response rate (the share of patients whose tumors shrank by a set amount, confirmed on a later scan) was 72.2% (13 of 18 patients) versus 42.1% (8 of 19) with standard treatment alone.
• Progression-free survival (time before the cancer worsens) had a hazard ratio of 0.55 by blinded independent review — meaning roughly 45% lower risk of progression at any point — but the 95% confidence interval runs from 0.15 to 2.09. Because that range includes 1.0 (no difference), the result is consistent with a real benefit but does not prove one. Median PFS had not yet been reached in the onvansertib arm, while it had been in both standard-care arms.
• In the arms using the other chemotherapy regimen, FOLFOX, the 10-Q states plainly: "No meaningful differences in efficacy were observed."
• Safety: no additive adverse events and no unexpected toxicities were reported; neutropenia (low white blood cells) was the most common serious side effect across both onvansertib and standard-care arms.

The honest read: a large response-rate gap in small groups (18 vs 19 patients), a favorable but statistically inconclusive PFS signal, and one of the two chemotherapy backbones showing no benefit. That is the kind of result that justifies a Phase 3 trial, which is exactly what Phase 3 exists to confirm.

After an End-of-Phase 2 meeting with the FDA in April, Cardiff designed the Phase 3 trial (CRDF-005) as a global randomized study of about 640 patients, comparing onvansertib 30 mg plus FOLFIRI/bevacizumab against FOLFIRI/bevacizumab alone. Management plans to start it in the first quarter of 2027, "subject to securing additional financing."

Cash, runway and dilution

This is the part of the filing that matters most for shareholders.

• Cash: $34.5 million at June 30, down from $46.1 million at March 31 — an $11.6 million decline in one quarter.
• Going concern: Note 1 of the 10-Q says the cash "is not sufficient to meet its funding requirements for at least the next 12 months" and that "there exists a substantial doubt about the Company's ability to continue as a going concern." That's accountant language for: without more funding, the company can't be sure it will still be operating a year from now.
• Runway guidance moved out by about two quarters. In May, management said cash would last "into the first quarter of 2027." In August, after the July raise, it said "into the third quarter of 2027." Neither figure includes the cost of actually running the Phase 3 trial at full speed, and the filing gives no cost estimate for that trial.
• The July raise: On July 14, Cardiff sold 8,571,429 shares to institutional investors at $1.05 per share, each with a warrant (a right to buy another share later) exercisable at $1.31. Officers and directors bought 721,649 shares at $1.455, with warrants at $1.33. Gross proceeds were about $10.05 million, of which 7% goes to the placement agent, H.C. Wainwright, which also received warrants for 465,157 shares at $1.3125.
• Share count: 68.3 million shares at June 30, 77.8 million on August 6. If every new warrant is exercised, that adds roughly 9.8 million more. There are also 11.8 million stock options outstanding, but with a weighted-average exercise price of $3.29 they had almost no in-the-money value at June 30 ($14,239 of intrinsic value).
• More authorized shares: The new warrants can't be exercised until shareholders approve raising the cap on common shares from 150 million to 300 million. A special meeting is set for October 22, 2026; the July purchase agreement requires Cardiff to keep calling meetings every 60 days until it passes. Doubling the authorized share count also clears room for future offerings, which the Phase 3 plan will need.

The licensing dispute — now settled

Cardiff doesn't own onvansertib outright; it licenses it from Italy's Nerviano Medical Sciences (NMS) under a 2017 agreement. In February 2026, NMS claimed Cardiff was in material breach (over whether an NMS scientist should be named as co-inventor on two Cardiff patents, and later over whether Cardiff was developing the drug diligently), and in May NMS said it was terminating the license. Cardiff sued in federal court. As of the 10-Q, a preliminary-injunction motion was "fully briefed and awaiting decision," and the filing warned that losing would mean losing the rights to its only drug.

After the quarter, on September 14, 2026, the two companies settled: the lawsuit is dismissed with prejudice, no money changes hands, and the license continues under an amendment. The amendment is not free, though. Per the 8-K, Cardiff takes on:

• Development milestone and "performance-related" payments tied to the Phase 3 program (amounts redacted);
• An additional license fee on net sales of any product covered by a Cardiff patent, payable instead of (not on top of) the existing royalty;
• A share of the proceeds to NMS if Cardiff is sold or transfers the license before Phase 3 data read out, with the percentage depending on when the trial's patients were dosed — this makes a pre-readout takeover more expensive for a buyer;
• A board observer seat and a Scientific Advisory Board seat for NMS, plus a requirement that any future termination for cause first be backed by a court or arbitrator's finding of breach.

The settlement removes the single biggest existential risk from the second half of 2026 and should end the legal bills that pushed up SG&A. The price is new, partly undisclosed payment obligations and less freedom to sell the company early.

Takeaway: The narrower loss is a side effect of a trial ending, not a sign of a leaner company. Cardiff has roughly a year of cash, a going-concern warning, and a 640-patient Phase 3 trial it cannot start without raising much more money — so the next thing shareholders should expect is another share sale, likely at a time when the share count has already grown 14% since June.

What the headline numbers hide

• Cash burn went up even as expenses went down. Operating expenses for the first half fell $6.8 million year over year, yet cash used in operations rose to $24.1 million from $21.1 million. The difference is unpaid bills being settled: accounts payable fell to $4.0 million from $8.1 million and accrued liabilities to $6.4 million from $7.6 million since December, which the filing reflects as $5.3 million of extra cash used. Some of that is catching up on bills from the busier trial period last year, so burn should ease — but the income statement understates how fast cash left this half.
• The lower loss is temporary by design. R&D is low because Cardiff is between trials. Once Phase 3 enrollment starts, clinical costs will rise well above today's $3.2 million a quarter.
• One-offs inflated overhead. First-half SG&A carried severance for the departed CEO and CFO, $0.5 million of option-modification expense, and legal fees from the NMS fight. With the dispute settled in September and the executive transition done, underlying SG&A should be lower than the first-half run rate.
• Interest income is shrinking with the cash pile. It fell to $0.38 million from $0.84 million — less cash earning interest — which slightly widens the net loss going forward.
• Watch the dollar loss, not just the per-share loss. The weighted-average share count rose 2.8% year over year, and the July offering will lift it by roughly 14% in Q3 — spreading the same dollar loss over more shares makes the per-share figure look better without the company losing any less money.

Outlook

Management's stated path: start the Phase 3 CRDF-005 trial in Q1 2027 if financing is secured, keep running the remaining Phase 2 patients (12 were still on trial as of a June 23 data cut), and continue supporting investigator-run studies in pancreatic, small cell lung and triple-negative breast cancer and a blood cancer (CMML). There is no revenue guidance, which is normal for a company with no approved product.

Our read: the clinical case for a Phase 3 trial is reasonable, the license is secure again, and the company has trimmed its spending to buy time. But "into the third quarter of 2027" is a runway that ends before a 640-patient trial could produce data, so the investment case now hinges on how much money Cardiff can raise and at what share price. Things to watch in the Q3 10-Q (expected around mid-November 2026): the outcome of the October 22 vote on authorized shares, whether quarterly cash burn falls below Q2's roughly $11.6 million decline now that legal fees and severance are behind it, any update to the runway or Phase 3 start date, and any disclosure of the dollar size of the new NMS milestone payments.