The date that matters most: 14 November
The single most concrete event in this story is a regulatory deadline. Summit Therapeutics' Biologics License Application for ivonescimab, based on its HARMONi trial results, has an FDA goal action date of November 14, 2026.1 That is the date by which the US regulator has committed to act on the application. It is not a promise of approval.
Ivonescimab is a bispecific, a drug engineered to hit two cancer targets at once. It is one of three kinds of next-generation programme now reaching decision points. The others are antibody-drug conjugates (ADCs), which deliver a toxic payload to cancer cells, and personalised mRNA vaccines. On the vaccine side, Moderna and Merck plan to present detailed intismeran results at the ESMO cancer congress on October 24 and to engage regulators about possible filings.1 The dossier does not say what those results contain, so we do not characterise them.
What one big bet costs: Summit's verified numbers
Headlines about a possible approval say little about what it takes to get there. Summit's SEC filings, which Via News has checked against the source, show a company that has spent heavily and earned nothing from sales. Revenue was zero in both FY2023 and FY2024.2
The losses have grown. Net loss was $614.9 million in FY2023, $221.3 million in FY2024 and $1.08 billion in FY2025.2 To picture the latest figure, that is more than the company lost in the previous two years combined. Loss per share was $0.99 in FY2023, $0.31 in FY2024 and $1.44 in FY2025.2 The operating loss, which strips out items outside day-to-day business, was $609.7 million in FY2023 and $226.0 million in FY2024.2
Actual cash leaving the business is a different measure from the headline loss, and the two diverge. Operating cash flow was negative $76.8 million in FY2023, negative $142.1 million in FY2024 and negative $322.9 million in FY2025.2 So the FY2025 loss of $1.08 billion is far larger than the $322.9 million of cash used in operations. The dossier does not explain the gap, and we will not guess. Readers should note that the loss figure and the cash-burn figure tell different stories. Capital spending on equipment and facilities stayed small throughout: $128,000 in FY2023, $139,000 in FY2024 and $657,000 in FY2025.2
The cash balance: a company that keeps needing to be refilled
Summit's cash balance has swung widely. It ended FY2023 at $71.4 million and FY2024 at $104.9 million.2 It was $211.5 million at the end of the first quarter of 2025 and $297.9 million at the end of the second.2 It then fell to $238.6 million in the third quarter and $225.3 million at the end of FY2025.2 By the first quarter of 2026 it was $106.5 million, and by the second quarter of 2026 it had risen to $419.4 million.2
The jumps upward cannot come from sales, because there are none. The dossier does not say how the money arrived, so we cannot tell you whether it was new share sales, borrowing or something else. The lesson for an ordinary investor is simpler: this kind of company can look well funded one quarter and thin the next. An approval decision lands against a balance sheet that has repeatedly needed topping up. Anyone who owns such a stock, directly or through a healthcare or biotech fund, should know that.
The second front: AI-designed medicines
The second theme is how drugs get invented. In an article published by MIT Technology Review, Puja Sapra describes AstraZeneca's vision for a 'lab of the future'. She uses a plain analogy: “Where a self-driving car uses sensors and models to navigate its environment, this system uses AI to make predictions, robotic systems to execute experiments, and instruments to generate data.”3
She names AstraZeneca's data as its edge: “Data is our differentiator. We've built an intentionally diverse portfolio across multiple disease areas and drug types. All of that data empowers us to fine-tune frontier AI models with richer, more representative training sets.”3 The company's datasets span molecular structures, binding measurements, safety profiles and manufacturing outcomes.3
She is candid about the limits. “One of the hardest problems in de novo design is predicting whether a computationally generated molecule will be safe in the human body.”3 She also says human oversight stays central: “With more autonomous systems, human oversight remains at the heart of this approach—ensuring explainable and ethical AI for the benefit of patients.”3 The practical hope is for models that help pick “which two or three targets to prioritize,” then balance potency, stability, manufacturability and safety, a direct link to the multi-target bispecifics above.3
A reliability note. Via News measures how often claims from each source hold up when checked. For this MIT Technology Review item, none of the 11 claims checked held up.3 We quote Sapra accurately, but these are statements of vision and ambition, not results. The article offers no evidence in this dossier that an AI-designed cancer drug has yet reached patients.
The third front: where patients are treated, and who owns it
The third theme is the business of delivering care. McKesson and the private-equity firm Clayton, Dubilier & Rice announced a take-private acquisition of Option Care Health on October 6, 2026. They will pay $32.05 per share, a 37% premium to the previous closing price, valuing the company at $5.8 billion including debt. CD&R would own 51% and McKesson 49%.1 The dossier gives no wider comparison for that price, so we offer none. In plain terms, shareholders are being offered a clear premium to get out, and the company would leave the stock market.
Meanwhile, the existing drugs business keeps growing. Novocure reported second-quarter 2026 net revenues of $184 million, up 16% on a year earlier, with 18% growth in global active patients.4 It also won a CE Mark, which allows sale in the EU, for Optune Pax in locally advanced pancreatic cancer, with Germany the first market to launch.4 Frank Leonard said: “This was our strongest quarter to date, with record net revenues and active patients on therapy.”4
Evidence is arriving too
Several results landed this month. In the EXCALIBER-RRMM Phase 3 trial in multiple myeloma, a regimen abbreviated ZDd reached a median progression-free survival (time before the disease worsens or the patient dies) of 42 months, against 20 months for the DVd comparison. That is a hazard ratio of 0.49 and a 51% reduction in the risk of progression or death, with p below 0.000001.1 The median follow-up was 23 months, shorter than the 42-month median in the winning arm, so the long-term picture will matter as more data mature.1 Safety was reported as consistent with prior findings.1
The FDA also approved Tecentriq, and its Hybreza version, with chemotherapy for adjuvant treatment of stage III colon cancer with a mismatch-repair deficiency, based on the Phase III ATOMIC study.1 Our dossier describes it as the twelfth US indication for Tecentriq.1
In Europe, the EMA's CHMP gave a positive opinion on July 24 for Gilead's Trodelvy plus Keytruda in first-line, PD-L1-positive metastatic triple-negative breast cancer.5 Evandro de Azambuja said patients “need effective treatment options as early as possible in the course of their disease, particularly when their tumors express PD-L1.”5 A CHMP opinion is a recommendation; formal authorisation is a separate step.
Telix dosed the first patient in its Phase 3 LUTEON trial of TLX250-Tx in relapsed or recurrent clear cell renal cell carcinoma, a first for this type of radiation-delivering drug in that cancer.6
The cost side: not every company is flush
Not every company in this story is expanding. ADC Therapeutics announced a workforce reduction it expects to deliver “$10 million in annualised cost savings,” while saying it “maintains a cash runway into 2028.”7 Everest Medicines, for its part, hired two senior translational scientists from Moderna's oncology and cancer-vaccine teams on October 9.1 Money and talent are moving, but unevenly.
How much to trust the feeds
This story draws on press releases and news feeds, and their track records differ. Via News's measurement shows that 56% of 4,957 claims checked from the NewsEOD feed held up, the source of the Gilead and Novocure items.4,5 For the Telix release, 31% of 2,928 checked claims held up.6 For the weekly biotech roundup, it was 46% of 520.7 These are company announcements describing their own news, and they are best read as claims. Only the Summit figures above are checked against SEC filings.
What to watch
- October 24: Moderna and Merck's detailed intismeran data at ESMO, and whether regulatory filings follow.1
- November 14: the FDA's goal date on ivonescimab. Compare it with Summit's next cash disclosures, since the balance has swung from $225.3 million to $106.5 million to $419.4 million in three reporting dates.1,2
- EXCALIBER-RRMM: whether longer follow-up supports the 42-month figure.1
- Option Care: the deal terms, with CD&R at 51% and McKesson at 49%, as the transaction proceeds.1
