Adagene Inc. (ADAG) H1 2026 Earnings: Revenue $1.6M
ADAG — H1 2026 Financial Report Analysis
H1 (Interim) · Fiscal year 2026 · Published by Pham Hop
Adagene's H1 2026 net loss widened 21.5% to $16.4M, mostly on FX and warrant revaluation, while an April share sale lifted cash to $127.9M and runway into late 2028, past its 1H 2027 Phase 2 colorectal-cancer readout.
Revenue
$1.6M
Net income
-$16M
-21.5% YoY
Diluted EPS
$-0.23
0.0% YoY
Overview
Adagene, a clinical-stage cancer-drug developer based in Suzhou and San Diego, reported a net loss of $16.4 million for the six months to June 30, 2026, up from $13.5 million a year earlier. It has no approved products, so the loss is the price of running clinical trials. What changed the company's position was the April 2026 share sale: cash rose to $127.9 million from $74.5 million at the end of 2025, and management now says that money lasts "into late 2028". That is after its main drug's randomized Phase 2 colorectal-cancer readout, which it expects in the first half of 2027.
Adagene files as a foreign private issuer, so it reports half-years on Form 6-K instead of quarterly 10-Qs. This analysis uses the unaudited interim financial statements (Exhibit 99.2) and the results press release (Exhibit 99.1) furnished on August 12, 2026.
At a glance
$127.9M in cash. This includes about $65.8M net from an April share sale at $3.75 per ADS. It is enough to get past the 2027 Phase 2 readout without raising money again first.
$14.7M of cash used by operations, up 32% from $11.1M. That is faster than the growth in reported losses, and the gap would be bigger without an extra $2.3M of unpaid supplier bills.
$1.6M of revenue, all from partners. It came from Sanofi ($1.28M) and Exelixis ($0.34M). The Sanofi portion is cash received in 2025 being counted as revenue now, not new money.
The numbers
Metric
H1 2026
H1 2025
YoY Change
Licensing & collaboration revenue
$1.6M
nil
n/m
Research & development expense
$14.0M
Read 0 community reports on Adagene Inc., or write your own.Write a report
$12.0M
+16.2%
Administrative expense
$4.2M
$3.7M
+15.5%
Loss from operations
$16.6M
$15.7M
+5.7%
Net loss
$16.4M
$13.5M
-21.5% (loss widened)
Net loss per ordinary share (basic & diluted)
$(0.23)
$(0.23)
0.0%
Non-GAAP net loss (excl. share-based pay)
$14.2M
$11.4M
+24.5% (loss widened)
Net cash used in operating activities
$14.7M
$11.1M
+32.3%
Cash & cash equivalents (period-end)
$127.9M
$62.8M
+103.6%
Bank borrowings (period-end, RMB loans)
$5.7M
n/a (Dec-25: $6.1M)
n/a
Ordinary shares outstanding (period-end)
83.9M
59.2M (Dec-25)
+41.7% vs Dec-25
Operating margin (the share of revenue left after running the business) isn't shown. With $1.6M of revenue against a $16.6M operating loss, it would be about -1,000%, which says nothing useful about a company at this stage. One ADS (the US-traded share) represents 1.25 ordinary shares, so the loss works out to roughly $0.29 per ADS (our conversion).
Takeaway: The half-year's main event was financing, not operations. The April raise nearly doubled the cash pile and moved the funding question past the 1H 2027 randomized Phase 2 readout for muzastotug, so that result is now the only thing that matters much. It came at a cost: 42% more shares than at year-end. And a runway "into late 2028" on $127.9M implies spending much faster than the first half's roughly $2.5M a month.
Where the money went
Research and development (spending on lab work and clinical trials) rose 16.2% to $14.0M. Management attributes the increase to the "continued clinical focus and development of muzastotug". Muzastotug (ADG126) is the company's lead drug: a "masked" antibody against CTLA-4, a brake on the immune system. The mask is meant to keep the drug inactive until it reaches the tumor. Administrative costs rose 15.5% to $4.2M because of higher personnel and office costs.
Revenue of $1.6M (nil a year ago) came from partners that license Adagene's SAFEbody masking technology:
Sanofi: $1.28M. This is an option-exercise payment Adagene received in late 2025. It is being counted as revenue as the related work gets done. Contract liabilities (cash received for work not yet done) fell by exactly $1.28M, to $2.18M, so this revenue brought in no new cash during the half.
Exelixis: $0.34M, recognized when some side projects were completed.
Revenue offset most of the $2.5M rise in operating expenses. That is why the operating loss grew only 5.7%, while the net loss grew 21.5%.
What the headline numbers hide
Most of the jump in net loss came from items unrelated to the business. Net loss grew by $2.9M, and only $0.9M of that came from operations. The rest: a $1.42M swing in currency effects (a $1.25M foreign-exchange gain in H1 2025 became a $0.17M loss) and a $0.97M non-cash charge from revaluing warrants (rights to buy shares) issued in 2025 to a consultant for investor-relations and business-development work. Warrants like these are counted as liabilities and marked to market, so when their value rose, the increase counted as a loss. Lower interest expense ($0.14M vs $0.32M) and higher interest income ($1.38M vs $1.21M) partly offset these.
Cash burn grew faster than the loss, and it was flattered. Operating cash outflow was $14.7M, compared with a non-GAAP loss of $14.2M. That gap looks small only because unpaid bills grew. Accounts payable jumped from $2.9M to $15.5M, but $10.3M of that is a relabeling. In April, WuXi AppTec's stake fell below 5%, so WuXi AppTec and WuXi Biologics stopped counting as related parties. Their bills moved out of "amounts due to related parties" (which went from $10.3M to zero) and into ordinary payables. Taken together, payables rose from $13.2M to $15.5M, about $2.3M of spending not yet paid. Without that, operating cash burn would have been about $17M. Payables of $15.5M are more than the half-year's entire R&D bill. That is worth watching as trial activity grows.
Per-share figures hide the dilution. Loss per ordinary share was flat at $0.23 only because the average share count grew about 20% (70.8M vs 58.9M). The loss itself grew 21.5%. Shares outstanding at period-end were 83.9M, up from 59.2M in December, after 23.3M shares from the April offering and 1.0M from the at-the-market program (share sales made gradually at market prices).
The GAAP vs adjusted gap is modest and steady. Non-GAAP net loss excludes only share-based compensation: $2.13M, compared with $2.03M a year ago. No other adjustments are made.
The financing terms. The offering sold 18.67M ADSs at $3.75, which the filing says was the 30-day volume-weighted average price, so there was no discount to the recent trading price. Gross proceeds were about $70M, net about $65.8M.
Balance-sheet items to note. There are $16.55M of Series A contingently redeemable convertible preferred shares, issued to Sanofi in July 2025. They sit outside shareholders' equity, in "mezzanine equity". Bank debt in China was $5.7M. Cash minus bank debt is about $122M.
Pipeline: what the cash is paying for
The core question is whether muzastotug, combined with Merck's Keytruda (pembrolizumab), works in microsatellite-stable colorectal cancer (MSS CRC). This is the most common form of colorectal cancer, and it has largely not responded to existing immunotherapies. Updated Phase 1b/2 data from April 2026, in patients without liver metastases:
Cohort
Overall response rate (ORR)
Median progression-free survival (PFS)
10 mg/kg, every 3 weeks
17% (5/29)
4.8 months
10 mg/kg, every 6 weeks
0% (0/10)
4.5 months
20 mg/kg, every 6 weeks
25% (3/12)
4.9 months
20 mg/kg loading dose, then 10 mg/kg every 3 weeks
36% (5/14)
15.4 months
ORR is the share of patients whose tumors shrank meaningfully. PFS is how long patients lived before the cancer worsened. Response rates rose with dose. Higher doses also brought more serious side effects: grade 3 treatment-related adverse events hit 38% of patients in the 20 mg/kg cohorts, compared with 15% at 10 mg/kg. There were no grade 4 or 5 events, and 4% of all 67 patients discontinued. The best cohort had only 14 patients, so these figures are a signal, not proof.
The randomized Phase 2 compares two dosing regimens: 10 mg/kg induction, and 20 mg/kg induction followed by 15 mg/kg maintenance. Its purpose is to pick the dose for a registration (Phase 3) trial. Results are due in 1H 2027, and a registration trial is expected to start in 2027 once the dose is set. Other data are early. In first-line liver cancer (HCC), the triple combination showed a 50% ORR by standard RECIST v1.1 criteria in six patients, against 17.5% in a 40-patient control arm. In late-line MSS CRC, adding fruquintinib produced responses in 1 of 4 and 2 of 5 patients.
Did last time's read hold up?
This is our first published analysis of Adagene, so there is no earlier outlook to check.
Outlook
Management's guidance: cash runway into late 2028, Phase 2 MSS CRC results in 1H 2027, a potential registration trial in 2027, and an Incyte-run Phase 1 combination study starting later in 2026.
After the period, Adagene announced a $2.0M preclinical milestone from Exelixis for XB404, plus an undisclosed payment for a second ADC (antibody-drug conjugate) program (September 9). Third Arc Bio selected its first lead candidate under an expanded agreement and will fund its clinical development (September 10). China's regulator cleared a Phase 1 trial of ADG138, a double-masked HER2×CD3 T-cell engager, due to start in Q4 2026 (September 22). Separately, the Executive Director of Finance resigned for personal reasons, effective September 30, 2026, and the company said there was no disagreement with the board.
Our view: Partner milestones like Exelixis's $2M help at the margin, but they are small next to a burn rate of roughly $30M a year. The company's value depends on the 1H 2027 dose-selection readout. Two things to watch in the full-year 20-F (expected around April 2027):
Whether operating cash burn moves toward the pace implied by the runway guidance. $127.9M lasting to late 2028 means averaging roughly $4M a month, against about $2.5M in H1 2026.
Whether payables keep growing faster than spending, which would mean the cash figures are flattered by delayed payments to contract research and manufacturing suppliers.
A clean, well-tolerated dose that keeps the loading-dose cohort's response rate would justify the registration trial the raise was sized for. A weaker result would leave Adagene with plenty of cash but a much thinner story.