AVIR — Q2 2026 Financial Report Analysis

Q2 2026: a smaller loss as the first Phase 3 trial wound down, with the trial result itself landing in July

Atea Pharmaceuticals has no product on the market and reported no revenue in the second quarter of 2026, so the numbers that matter are how much it spent, how much cash is left, and whether its hepatitis C drug is working. The quarter's net loss narrowed to $32.9 million from $37.2 million a year earlier, mainly because spending on C-BEYOND, the North American Phase 3 trial of its bemnifosbuvir/ruzasvir (BEM/RZR) combination, fell as patients finished their final follow-up visits. In July, after the quarter closed, Atea reported that C-BEYOND met its main goal: an 8-week course of BEM/RZR cured about as many patients without cirrhosis as the standard 12-week Epclusa regimen.

At a glance

• $219.5 million in cash and investments at June 30, down $82.4 million in six months. Management says this funds the company "through 2027", which covers the remaining Phase 3 trial and a planned FDA filing, but not a full commercial launch.
• 93.9% vs 94.8% cure rate for BEM/RZR vs Epclusa (sofosbuvir/velpatasvir) across 905 C-BEYOND patients. That cleared the trial's "non-inferiority" test, meaning the gap was inside the pre-agreed 5-point margin, so BEM/RZR counts as no worse than the existing drug.
• R&D spend of $28.2 million, down 12.8% from a year ago and down 32% from $41.1 million in Q1 2026, as C-BEYOND costs fell away. The second trial, C-FORWARD, is fully enrolled and reports early in Q1 2027.

The numbers

All figures are from Atea's Form 10-Q for the quarter ended June 30, 2026. The company has no adjusted (non-GAAP) figures in this filing.

Where the money went

Atea's cost base is almost entirely clinical trials. In Q2 2026, external spending on the hepatitis C (HCV) program was $17.9 million, down from $24.3 million a year earlier. The 10-Q attributes the drop to "the completion of the Week 24 post treatment visits by patients in our C-BEYOND Phase 3 clinical trial": once the last patients in that trial finished their follow-up, the per-patient costs stopped.

Two things partly offset that:

• Hepatitis E (HEV) spending started. AT-587, Atea's second drug candidate, cost $2.9 million in external spending in the quarter ($6.9 million in the first half), against nothing a year earlier. It entered its first human trial, a Phase 1 safety study in healthy volunteers, in July 2026.
• Internal R&D costs (Atea's own staff and overhead) were $7.3 million, slightly below $7.8 million, mostly because of lower stock-based compensation.

General and administrative costs (head office, legal, accounting) fell 23.4% to $6.9 million, which the filing puts down to lower stock-based compensation and lower professional fees. That is partly the payoff from the roughly 25% workforce cut in Q1 2025, which management expects to save about $15.0 million through 2027.

On a six-month view, the picture reverses: R&D was up 12.0% to $69.3 million, because both Phase 3 trials (C-BEYOND and C-FORWARD) were running at once in Q1 2026. Q1 alone carried $41.1 million of R&D and a $45.4 million net loss. So Q2 shows spending coming down after a peak, not a company that has generally become cheaper to run.

What the headline numbers hide

• Last year's quarter included a one-off $5.0 million payment. Atea licenses ruzasvir, the second drug in its combination, from Merck. The first milestone payment of $5.0 million came due when the first patient enrolled in C-BEYOND, and Atea booked it as R&D expense in Q2 2025. Without it, Q2 2025 R&D would have been about $27.3 million, so underlying R&D actually rose about 3% this quarter rather than falling 12.8%. Take the milestone out and last year's loss would have been about $32.2 million, slightly smaller than this quarter's $32.9 million: lower overhead roughly offset the drop in interest income, and the year-on-year improvement is the one-off comparison, not cheaper trials.
• Cash went out faster than the loss suggests. Six-month operating cash burn was $83.5 million, $5.1 million more than the $78.4 million net loss. Two items explain the gap: Atea paid down $11.3 million of bills and accrued costs, and $7.1 million of stock-based compensation counted as an expense but cost no cash. A year earlier the pattern was the other way round, with burn ($63.4 million) well below the loss ($71.4 million). The faster H1 2026 burn mostly reflects timing of payments, but it is the figure that sets the runway, not the loss.
• Interest income halved. Interest income fell from $4.4 million to $2.2 million "primarily due to lower investment balances". The cash pile is smaller after last year's spending and $25.5 million of share buybacks, so it earns less and no longer offsets as much of the spending.
• The buyback makes loss per share look slightly worse, not better. Atea bought back about 7.7 million shares in 2025, so the average share count fell from 83.7 million to 80.1 million. Spread over last year's share count, this quarter's loss would have been $0.39 per share instead of $0.41. Fewer shares mean each one carries a bigger slice of the loss.
• No new shares sold yet. Atea has a $200 million "at-the-market" program with Jefferies, which lets it sell new shares gradually into the market. As of June 30 it had sold none, so current shareholders have not been diluted to fund the trials so far.

The trial result is the real news

C-BEYOND is what the share price will trade on, more than any line in the income statement. Results by patient group, from the 10-Q:

"Cure" here means sustained virologic response: no detectable virus 12 weeks after treatment ends. Atea's argument for BEM/RZR is not a higher cure rate (it was about one point lower overall) but a shorter course for most patients. The filing says 80–90% of US hepatitis C patients do not have cirrhosis, and those patients took BEM/RZR for 8 weeks rather than 12. The company also points to fewer interactions with common medicines such as acid reflux drugs, and no need to take it with food. The 10-Q reports no drug-related serious side effects and no drug-related early discontinuations.

The market Atea is targeting is large but already well served: the filing puts 2025 global sales of the two leading branded regimens, Epclusa (including its authorized generic) and Mavyret, at more than $2.5 billion, about half in the US. Atea would also owe Merck royalties of high single digits to mid-teens percent of net sales.

Takeaway: The quarter's smaller loss is mostly timing (one Phase 3 trial finished, last year's quarter carried a $5.0 million Merck milestone), not a lasting cost reset. What changed Atea's position is C-BEYOND's July result, and the $219.5 million in the bank is enough to reach the second trial's readout and an FDA filing, but not to launch a drug on its own.

What to watch next

• C-FORWARD topline results, early Q1 2027. This second Phase 3 trial enrolled more than 880 patients at about 120 sites in 17 countries outside North America, with a wider mix of hepatitis C genotypes than C-BEYOND. Good results in those genotypes are what support a global label.
• FDA filing, targeted for Q2 2027, if C-FORWARD is positive. FDA acceptance of the application would trigger the next $10.0 million milestone payment to Merck, which Atea expects in Q3 2027.
• Cash runway. At the first half's burn rate (about $42 million a quarter), $219.5 million lasts roughly five quarters, which lines up with management's "through 2027" guidance. Q2's lower spending helps, but the company also plans manufacturing scale-up, pre-launch work and further AT-587 development. A launch (on a standard FDA review timeline, approval would come in 2028 at the earliest) would need either new money (share sales through the at-the-market program, or debt) or a commercial partner. Management says it remains "open to consideration of strategic transactions"; it ended its formal partnership-search mandate with Evercore in November 2025, saying Phase 3 data would drive those talks.
• Next quarter's spending. With C-BEYOND finished, HCV trial costs should keep falling while C-FORWARD completes follow-up, but AT-587's Phase 1 and pre-launch manufacturing work will add costs. The Q3 10-Q, expected around mid-November 2026, will show which way that nets out.

Our read: the business risk has shifted from "does the drug work" to "will the second trial confirm it, and who pays for the launch". The balance sheet covers the first question, not the second.