Gilead grew revenue 10% to $7.8 billion on 12% HIV growth and a scaling Yeztugo launch, but $11.2 billion of acquired-IPR&D charges and a $1.75 billion Trodelvy write-off produced a $10.5 billion net loss and drained cash from $10.6 billion to $3.2 billion.
Revenue
$7.8B
+10.2% YoY
Net income
-$10.5B
Diluted EPS
$-8.45
Operating margin
-133.2%
A $10.5 billion loss on top of a 10% revenue quarter
Gilead Sciences' second quarter of 2026 has to be read on two separate tracks, because the headline number and the underlying business point in opposite directions.
The business itself grew. Total revenues rose 10% to $7.80 billion from $7.08 billion a year earlier, and the HIV portfolio — which supplies roughly three-quarters of Gilead's sales — grew 12% to $5.69 billion.
The reported result was a net loss of $10.50 billion, or $(8.45) per diluted share, against a $1.96 billion profit ($1.56 per share) in the same quarter of 2025. Essentially all of that swing comes from two accounting charges tied to deals and a failed trial, not from anything that happened in the commercial business:
$11.18 billion of "acquired in-process research and development" (IPR&D) expense, up from $61 million a year ago. When Gilead buys a company whose value is entirely unapproved, still-experimental drugs, accounting rules make it expense the whole purchase price immediately rather than carry it on the balance sheet. Three deals closed in the quarter: $7.0 billion for Arcellx (the anito-cel cell therapy for multiple myeloma), $3.1 billion for Tubulis (antibody-drug conjugates), and a net $1.0 billion for Ouro Medicines — $1.9 billion gross, reduced by an $860 million payment from partner Lakefront Biotherapeutics for half the upfront cost.
A $1.75 billion write-off of an intangible asset originally acquired with Immunomedics. In June 2026 Gilead stopped the Phase 3 EVOKE-03 study of Trodelvy plus Keytruda in first-line lung cancer; with that program dead, the filing states the company "determined that no future cash flows were expected to be generated in relation to the NSCLC IPR&D intangible asset," so the remaining balance went to zero.
Gilead's own non-GAAP loss per share — the figure that strips out amortization and similar non-cash items but not the IPR&D charges — was $(6.75) versus $2.01 a year ago. Management quantifies the acquisition-related hit at $(9.08) per share including the related tax effects. Back that out and the quarter was profitable on an operating basis.
Key figures
Metric
Q2 2026
Read 0 community reports on Gilead Sciences, or write your own.Write a report
"n.m." means not meaningful — a percentage change between a profit and a loss doesn't describe anything useful, and the filing labels these the same way. Gross margin is the share of product revenue left after the cost of manufacturing the drugs; operating margin is what's left after all operating costs, here dominated by the one-off charges above.
Takeaway: Strip out $12.9 billion of deal and write-off charges and this was one of Gilead's better quarters — HIV up 12%, gross margin steady, base-business guidance raised twice this year. What the loss actually signals is a deliberate, very expensive bet: Gilead spent more on unapproved pipeline assets in one quarter ($11.2 billion) than it earned in revenue ($7.8 billion), drained its cash and securities from $10.6 billion to $3.2 billion, and now has to convert anito-cel, the Tubulis ADCs and the Ouro T-cell engager into approved products to justify it.
HIV: price and switching, not just volume
HIV sales of $5.69 billion (+12%) came, per the filing, "primarily due to higher average realized price and demand" — worth noting because price and volume are different in quality. Price gains can reverse with payer mix or policy; demand gains are stickier.
Biktarvy, the single biggest product at $3.77 billion (+7%), grew on "higher average realized price, favorable inventory dynamics and higher demand, including patients switching from Genvoya and other Gilead HIV products." Two flags there. "Favorable inventory dynamics" means wholesalers stocked up — that borrows sales from a future quarter rather than creating them. And the switching is partly cannibalization: Genvoya fell 23% to $289 million and Odefsey fell 20% to $239 million in the same quarter. Biktarvy's $242 million gain is roughly matched by the $147 million those two lost.
Descovy jumped 48% to $967 million on "higher average realized price and demand." Descovy is the established pill for pre-exposure prophylaxis (PrEP) — medication healthy people take to avoid catching HIV.
Yeztugo (lenacapavir), the twice-yearly injectable PrEP shot launched in mid-2025, reached $232 million, versus $15 million a year ago — $223 million of it in the US. Half-year sales are $397 million. This is the launch to watch: it is the first product in years with a plausible path to changing how HIV prevention is delivered, and at this run-rate it is already Gilead's fourth-largest HIV product.
Importantly, currency did almost none of the work. About 25% of product sales are in non-dollar currencies, and the filing puts the net-of-hedges foreign exchange benefit at just $23 million in the quarter — about 0.3 percentage points of the 8% product-sales growth. The growth is real, not a weak-dollar artifact.
Oncology: one franchise climbing, one sliding
Total oncology sales were nearly flat at $873 million (+3%), but that masks two opposite trends.
Trodelvy grew 26% to $457 million on higher demand, and got meaningful regulatory wins in the quarter: FDA approval in first-line metastatic triple-negative breast cancer and European Commission authorization in the same setting. That is the growth engine.
Cell Therapy fell 14% to $417 million — Yescarta down 12% to $346 million, Tecartus down 24% to $70 million — which the filing attributes to "lower demand reflecting ongoing competitive headwinds," both in-class and out-of-class. This is a franchise losing share, and it is precisely why Gilead paid $7.0 billion for Arcellx's anito-cel: buying a next-generation cell therapy because the current one is being out-competed.
The Trodelvy picture is not uniformly good either. The same molecule that won breast cancer approvals failed in lung cancer, and that failure is what triggered the $1.75 billion impairment.
Costs: the acquisitions are expensive twice over
Beyond the upfront purchase prices, integrating three companies inflated the recurring expense lines:
R&D rose 18% to $1.76 billion, but the composition matters. Personnel and infrastructure costs jumped 34% to $1.15 billion, including $229 million of stock-based compensation from cashing out Arcellx, Ouro and Tubulis employee equity. Actual clinical study spending fell 3% to $618 million on lower oncology trial activity. On Gilead's own non-GAAP basis, R&D was $1.4 billion — below last year's $1.5 billion.
SG&A rose 41% to $1.92 billion. The split is telling: selling and marketing rose 17% to $1.01 billion on "higher HIV promotional expenses" — i.e. real money behind the Yeztugo launch — while general and administrative costs rose 82% to $910 million, including $332 million of acquisition-related stock compensation.
So roughly $560 million of the combined R&D and SG&A increase is one-off deal compensation that should not repeat at this scale. Product gross margin, meanwhile, held at 79.3% versus 78.7%, so nothing is eroding on the manufacturing side.
The effective tax rate — tax as a share of pre-tax profit — was (2.4)% versus 19.3%. That odd negative figure exists because Gilead still owed $242 million of tax despite a $10.25 billion pre-tax loss: most of the acquired IPR&D expense is not deductible. It is a real cash cost of structuring these deals as asset acquisitions.
Balance sheet: the war chest is spent
Cash, equivalents and marketable debt securities fell to $3.2 billion from $10.6 billion at year-end 2025. The half-year flows: $11.3 billion out for acquisitions, $2.8 billion of debt repaid, $2.1 billion of dividends and $774 million of buybacks, against $6.1 billion of operating cash flow and $4.1 billion of net new debt (a $3.0 billion senior note issue plus a $1.1 billion one-year term loan).
Operating cash generation is genuinely strong — $6.12 billion in the first half versus $2.58 billion a year ago, helped by lower tax payments, collections on higher sales and the $860 million Lakefront receipt. Gilead also declared a quarterly dividend of $0.82 per share payable September 29, 2026. But the capacity for another multi-billion-dollar deal without more borrowing is now limited, and buybacks have already been throttled ($355 million in the quarter).
Guidance and what to watch
Management raised the underlying revenue outlook on August 4 while cutting the reported earnings outlook — a combination that only makes sense once you separate the charges from the business:
Full-year 2026 guidance
August 4, 2026
Prior (May 7, 2026)
Product sales
$30,100M – $30,400M
$30,000M – $30,400M
Product sales excluding Veklury
$29,800M – $30,100M
$29,400M – $29,800M
Veklury
~$300M
~$600M
GAAP diluted loss per share
$(3.75) – $(3.40)
$(3.25) – $(2.85)
Non-GAAP diluted loss per share
$(0.65) – $(0.30)
$(1.05) – $(0.65)
The base business (everything except Veklury, the COVID-19 antiviral) was raised by $400 million at the low end, while Veklury guidance was halved to about $300 million. Veklury sales fell 81% to $23 million in the quarter as COVID-19 hospitalizations declined; at this point it has stopped being a swing factor for Gilead in either direction.
The GAAP and non-GAAP guidance moving in opposite directions is itself informative: non-GAAP loss per share improved by roughly $0.37 on the stronger base business, while GAAP worsened by about $0.52. That gap is consistent with the $1.75 billion impairment (roughly $1.41 per share) falling into GAAP only — our reading, not a breakdown the company provides.
Our view on trajectory. The commercial base is in better shape than the loss suggests: HIV growing 12% with a genuinely new PrEP product scaling from nothing to $232 million a quarter, Trodelvy up 26% with fresh first-line approvals, gross margin stable, and roughly $560 million of this quarter's expense growth clearly one-off. Three specific things determine whether the $11.2 billion of pipeline spending was worth it:
Yeztugo's trajectory, including the once-weekly oral lenacapavir formulation with an FDA decision date of February 2, 2027. Success there would extend the HIV franchise well past Biktarvy's eventual patent expiry — the risk every large-molecule pharma faces when a top seller goes generic.
Whether anito-cel can reverse the cell-therapy decline. Yescarta and Tecartus are shrinking double digits; a $7.0 billion purchase needs to do more than stabilize that line.
Whether Biktarvy's growth survives without price and inventory help. A 7% gain built partly on realized price, wholesaler stocking and switching from Gilead's own Genvoya and Odefsey is lower-quality than a 7% gain from new patients.
The caution is concentration and timing: Gilead has converted a liquid balance sheet into clinical-stage assets whose payoff is years away, while the products funding it all sit in one therapeutic area. Nothing in this quarter is alarming on its own — but the company has far less financial cushion than it had six months ago, and the burden of proof has moved to the pipeline.