Cogent Biosciences, Inc. (COGT) Q2 2026 Earnings Analysis

Spending ramps ahead of two FDA decisions

Cogent Biosciences still has no product on the market and reported no revenue in the second quarter of 2026. What changed is where its money goes: general and administrative spending — the bucket that holds the new sales and marketing team — jumped 138% to $31.8 million as the company builds out a commercial organization for bezuclastinib, its drug for systemic mastocytosis (a rare disorder where mast cells, a type of white blood cell, build up in organs and bone marrow) and for gastrointestinal stromal tumors (GIST, a cancer of the digestive tract). Total operating expenses rose 36% to $102.6 million, and the net loss widened 31% to $96.4 million. Because the share count is about 49% higher than a year ago, the loss per share was essentially flat at $0.52.

The quarter matters less for its numbers than for the calendar it sets up: the FDA has set a target decision date of November 30, 2026 for bezuclastinib plus sunitinib in second-line GIST (under Priority Review) and December 30, 2026 for non-advanced systemic mastocytosis (NonAdvSM). A third application, for advanced systemic mastocytosis (AdvSM), was submitted in June 2026. Management says it expects a US launch in the second half of 2026, pending approval.

At a glance

$0 revenue, $96.4M net loss (+31% YoY): the company is still pre-commercial; the larger loss is mostly the cost of getting ready to sell a drug, not more expensive trials.
G&A $31.8M vs $13.4M a year ago (+138%): commercial-readiness hiring and infrastructure now make up nearly a third of total spending, up from under a fifth.
$792.3M in cash and marketable securities, plus $71.4M raised after quarter-end: management says this lasts into late 2028, covering both FDA decisions and the launch.

Key figures

Metric	Q2 2026	Q2 2025	YoY Change
Revenue	$0	$0	—
Research & development	$70.8M	$62.2M	+13.8%
General & administrative	$31.8M	$13.4M	+137.9%
Total operating expenses	$102.6M	$75.6M	+35.8%
Interest income	$7.5M	$2.4M	+214%
Net loss	$(96.4)M	$(73.5)M	loss 31.1% larger
Net loss per common share (basic & diluted)	$(0.52)	$(0.53)	loss 1.9% smaller per share
Weighted average common shares	171.0M	114.5M	+49.4%
Cash, equivalents & marketable securities (period-end)	$792.3M	$900.8M (Dec 31, 2025)	−12.0% vs year-end
Net cash used in operations (six months)	$164.2M	$121.1M	+35.6%

For the first half, the net loss was $193.8 million versus $145.5 million, with operating expenses of $206.2 million versus $150.5 million.

Where the money went

Research and development ($70.8M, +$8.6M). The filing attributes the rise to costs supporting the SUMMIT, PEAK and APEX trials, regulatory work for the pending approvals, the early-stage pipeline, and "pre-approval manufacturing costs that would otherwise be capitalized following anticipated FDA approval." That last item is worth understanding: until a drug is approved, the cost of manufacturing commercial supply is expensed immediately; after approval, similar costs would be held as inventory on the balance sheet and only expensed when product is sold. So part of today's R&D bill is really future inventory.

The breakdown is telling. Direct external spending on bezuclastinib actually fell slightly, to $25.6 million from $27.1 million, as the three registration trials have reported their results. The growth came from people: personnel costs rose $6.7 million to $26.2 million, of which $3.6 million was higher stock-based compensation. Early-stage programs — an ErbB2 (HER2) inhibitor and a PI3Kα inhibitor in Phase 1, plus KRAS and JAK2 programs targeting IND filings in 2026 — rose $1.8 million to $11.8 million.

General and administrative ($31.8M, +$18.4M). The 10-Q ties this directly to "continued investments in commercial readiness, including personnel and infrastructure to support the potential launch of bezuclastinib." This is the line that will keep rising if approvals come through.

Interest. Interest income tripled to $7.5 million on a much larger cash pile. Interest expense of $1.2 million relates to $230 million of 1.625% convertible notes due 2031, issued in November 2025, which replaced a $50 million term loan that was repaid in full.

Bezuclastinib: the three applications

Indication	Status	Key data cited in the 10-Q
GIST, 2nd line (with sunitinib)	NDA accepted with Priority Review, PDUFA date Nov 30, 2026	PEAK Phase 3: median progression-free survival 16.5 months vs 9.2 months on sunitinib alone; 50% lower risk of progression or death
Non-advanced systemic mastocytosis	NDA accepted Mar 2026, PDUFA date Dec 30, 2026	SUMMIT met all primary and key secondary endpoints
Advanced systemic mastocytosis	NDA submitted Jun 2026	APEX met all primary and key secondary endpoints

A PDUFA date is the FDA's target date to decide on an application; it is a goal, not a guarantee. Management's own market-size estimates — over $4 billion for second-line GIST, about $3.5 billion for NonAdvSM and about $500 million for AdvSM globally — come from internal analyses and market research, not from sales, and should be read as ceilings rather than forecasts. Bezuclastinib is licensed from Plexxikon (part of Daiichi Sankyo), which is owed tiered royalties from low- to high-single-digit percentages on net sales, and up to $15.0 million more in regulatory milestones that may become payable within the next twelve months.

What the headline numbers hide

Flat loss per share masks a 31% larger loss. The per-share figure barely moved ($0.52 vs $0.53) only because there are far more shares: weighted common shares rose to 171.0 million from 114.5 million. Each existing shareholder owns a smaller slice of a company that is losing more money. That dilution has continued: 1.34 million shares were sold through the at-the-market program (selling stock gradually into the market) in the first half for $45.7 million, and another 1.74 million after quarter-end for $71.4 million. The program had $88.6 million of capacity left at June 30, so less than $20 million remains after those post-quarter sales — a future raise would likely need a new program or offering.
Cash burn is close to the loss, but stock pay is growing fast. Operating cash outflow for the first half was $164.2 million against a $193.8 million net loss; most of the gap is $34.0 million of stock-based compensation, which costs no cash but dilutes shareholders. Stock-based compensation was $17.1 million in Q2 alone, up 76% from $9.7 million — about 18% of the quarter's net loss.
Interest income flatters the bottom line. $7.5 million of interest income offset about 7% of operating costs this quarter. That income shrinks as the cash is spent, so the net loss would widen even with flat spending.
Some R&D is temporary. The pre-approval manufacturing costs that run through R&D today should shift to inventory after approval, which may make R&D look like it is falling when the underlying spending has simply moved. The filing doesn't quantify this amount.
Debt is modest but real. The $230 million convertible notes (carried at $223.5 million net of issuance costs) are due in 2031 and can be settled in cash, shares or both. Against $792 million of cash and securities, they are manageable.
No one-offs. There is no restructuring charge, impairment or unusual gain in either year's quarter; the increase in the loss is driven by ordinary spending growth.

Takeaway: Cogent is now spending like a company about to launch a drug — G&A has more than doubled — while still having no revenue. The $864 million of cash after the post-quarter stock sales comfortably covers the November 30 and December 30 FDA decisions, so the near-term question is not money but approval and how fast prescriptions build once it arrives.

Outlook: what to watch

Runway. With $792.3 million at June 30 and $71.4 million raised since, management guides that cash lasts "into late 2028, including through potential FDA approval of bezuclastinib and commercial launch for SM and GIST." First-half operating cash use averaged about $82 million a quarter, and both R&D and G&A are expected to keep rising. At that pace the existing cash would last roughly ten quarters before any sales, so the late-2028 guidance looks consistent with modest early revenue; it would tighten if spending accelerates faster than prescriptions do.

The next two filings. The Q3 2026 10-Q (expected early November) should show G&A rising again as sales staff are hired before the first approval decision. The first possible revenue would be in the fourth quarter at the earliest, and only for a few weeks if the GIST application is approved on its November 30 target date.

What would change the picture. An approval on schedule for GIST, followed by NonAdvSM, would turn Cogent into a commercial company within about a month and make prescription uptake the key number to track. A delay or a request for more data from the FDA would leave the company carrying a launch-sized cost base with no offsetting revenue — the main risk in the current spending plan. Plexxikon's possible $15.0 million in regulatory milestones and the royalty on sales are known, small offsets to any launch revenue.

Our read: the trial data in the filing is strong on paper (a 7.3-month improvement in median progression-free survival in GIST), the balance sheet is sufficient, and the financial story for the next two quarters is entirely about the FDA calendar.