BLSM — Q2 2026 Financial Report Analysis

Headline: spending on one lung-cancer drug nearly doubled the loss, and the IPO arrived just in time

BlossomHill Therapeutics (Nasdaq: BLSM) filed its first quarterly report as a public company on 18 September 2026, covering the three months ended 30 June 2026 — a period that ended before the company went public on 10 August. That timing is the single most important thing to understand about this filing: the balance sheet and the loss-per-share figure describe a private company, while the forward-looking discussion describes a newly funded public one.

BlossomHill is a clinical-stage biopharmaceutical company — meaning it has no approved products and no product revenue, and won't for years, if ever. Its entire income statement is expenses. In the filing's own words: "We have never generated any revenue from product sales and do not expect to generate any revenue from product sales unless and until we successfully complete development of and obtain regulatory approval for one or more of our product candidates, which will not be for several years, if ever."

So the question for a company like this is not "did it grow?" but "what is it spending money on, how fast is the cash going out, and how long until it runs out?"

The numbers

All figures from the unaudited condensed statements of operations, balance sheets and cash flows in the 10-Q. Percentage changes and runway figures marked (calculated) are derived from those line items, not disclosed by the company.

For the first six months of 2026, R&D was $41.3 million (vs. $22.1 million), G&A $5.5 million (vs. $3.5 million), and the net loss $44.8 million (vs. $23.7 million). Accumulated deficit — the running total of every loss since the company started in 2020 — reached $179.8 million at 30 June.

Where the money actually went: one drug, one trial

BlossomHill discloses R&D by program, which makes the story unusually legible. External costs (payments to contract research organizations, trial sites, and contract manufacturers) versus internal costs (salaries, facilities):

The split matters. External R&D spending more than doubled year over year (roughly $14.0 million from $6.6 million, calculated from the table above), while internal spending rose only about 25%. This is not a company that got expensive by hiring — it got expensive by running trials. Management attributes the BH-30643 increase to "significant growth in clinical trial enrollment and sites activated and an increase in manufacturing costs related to the clinical development of BH-30643."

BH-30643 alone consumed 42% of all R&D spending in the quarter (calculated). It is a small-molecule inhibitor aimed at EGFR-mutant non-small cell lung cancer. In plain terms: EGFR is a protein on the surface of cells that, when mutated, tells a tumour to keep growing. Existing drugs block it, but tumours eventually develop further mutations that stop those drugs working — the filing singles out "C797S-mediated resistance for third-generation EGFR TKIs, such as osimertinib." Patients whose cancer acquires C797S currently have, per the filing, "no currently approved targeted therapy." That gap is BlossomHill's wedge: a small, well-defined, underserved patient group is the fastest route to a first approval.

The second-largest swing is BH-501284, a preclinical KRAS program where external spend went from essentially nothing ($61 thousand) to $2.0 million, which the filing attributes to "manufacturing-related expenses to support IND-enabling studies" — the toxicology and manufacturing work required before the FDA will permit human testing. Notably, spending on "other programs and discovery" fell $0.4 million, which management explains directly: "we focused investment toward the development of our most advanced product candidates." That is a deliberate narrowing, not a broad expansion — earlier-stage research is being defunded to pay for the two lead programs.

The 111% jump in G&A looks alarming in percentage terms but is small in absolute terms (+$1.7 million) and is what pre-IPO preparation costs: for the six-month period, management breaks the $2.1 million increase into "$1.0 million increase in personnel-related expenses, a $0.5 million increase in legal expenses, and an increase in general overhead and outside services expense of $0.6 million."

The loss-per-share figure is not what it looks like

The reported net loss per share of $(8.75) is technically correct and practically meaningless as a forward indicator. It divides the quarter's loss by 2,713,696 weighted-average common shares — the count for a private company whose investors held convertible preferred stock (a senior class of shares that converts into ordinary shares at IPO) which is excluded from that denominator. All 18,258,960 preferred shares converted to common stock when the IPO closed, and the company sold a further 10,516,240 shares. The cover page reports 31,784,909 shares outstanding as of 16 September 2026.

Run the same $23.8 million quarterly loss over that post-IPO share count and it is roughly $(0.75) per share (calculated) — about one-twelfth of the printed figure. Anyone comparing BLSM's headline EPS against a peer's should not use the number in this filing.

A 1-for-4.6855 reverse stock split on 31 July 2026 has been applied retroactively to every share figure in the filing, so prior-period share counts and per-share figures are internally consistent, just not comparable to anything published before that date.

Cash: the part that actually constrains the business

Cash fell from $136.7 million at the end of 2025 to $95.4 million at 30 June — a $41.3 million decline in six months. Operating activities consumed $44.2 million, offset by $2.9 million from employees exercising stock options. Capital spending was negligible at $13 thousand, down from $1.0 million a year earlier, so nearly the entire outflow is trial and payroll spending.

The loss and the cash burn track each other closely — $44.8 million of net loss against $44.2 million of cash out — because there is almost nothing non-cash inside the loss. Stock-based compensation was only $0.9 million for the half-year, depreciation $0.25 million, and non-cash lease expense $0.7 million, totalling $1.9 million. That is unusually clean for a biotech: reported loss here is close to real money leaving the building, which is not always the case at companies with large equity-compensation charges.

At the 30 June cash balance alone, the first-half burn pace of about $7.4 million a month gave roughly 13 months of runway (calculated) — versus about 24 months on the same basis a year earlier. That compression is the whole reason the IPO happened when it did.

The IPO closed on 10 August 2026 at $16.00 per share: 9,375,000 shares for $150.0 million gross, less $15.3 million of underwriting discounts, commissions and offering expenses, for $134.7 million net; underwriters then partially exercised their option on 25 August for another 1,141,240 shares and roughly $17.0 million net. Total net proceeds: $151.7 million. Combined with the 30 June cash balance, that is about $247 million of pro forma resources.

The runway guidance implies burn goes up, not sideways

Management's stated expectation is that existing cash plus IPO proceeds "will be sufficient to fund our projected operating expenses and capital expenditures into the second quarter of 2028."

That guidance is worth arithmetic. From 30 June 2026, "into the second quarter of 2028" is roughly 24 months. Spending $247 million over 24 months implies about $10.3 million a month (calculated) — roughly 40% above the $7.4 million monthly pace of the first half of 2026. Management is not guiding to a company that coasts on its new cash; it is guiding to one that keeps accelerating, and the filing says so plainly: it expects "to continue to incur significant and increasing operating losses for the foreseeable future," with G&A rising "substantially" on top of public-company costs — "audit, legal, regulatory and tax-related services associated with maintaining compliance with exchange listing and SEC requirements, director and officer insurance premiums, board of director costs and investor relations costs."

The more conservative reading cuts the other way: if burn instead held at the first-half pace, $247 million would last closer to 34 months (calculated). The gap between those two scenarios — roughly Q2 2028 versus early 2029 — is essentially the cost of the pivotal trial described below. Investors should treat the Q2 2028 date as the planning assumption that includes it.

Takeaway: The reported 79% jump in net loss is not cost inflation — it is one drug moving into expensive late-stage testing, with BH-30643 taking 42% of all R&D while earlier discovery work gets cut. The consequence is that BlossomHill's $247 million post-IPO cash pile funds roughly one shot: management's own "into the second quarter of 2028" runway ends before a pivotal Phase 2 trial starting in Q1 2027 would plausibly read out, so the company will almost certainly need to raise again — and the terms of that raise will be set by the C797S data it reports in the first half of 2027.

What to watch next

The 10-Q sets out an unusually specific catalogue of milestones:

Two regulatory items already landed. In August 2026 the FDA granted BH-30643 Fast Track designation for advanced or metastatic EGFR C797S-positive NSCLC after prior third-generation EGFR TKI treatment — a status that gives more frequent FDA contact and eligibility for faster review, though it is not itself evidence the drug works. And in September 2026 the company presented updated SOLARA trial data in C797S-positive patients; the 10-Q references that presentation but discloses no efficacy figures from it, so the results are not assessable from this filing.

Our read on trajectory. Three things stand out.

First, the durability readout in H1 2027 is the real test. The filing says BH-30643 has "confirmed radiographic responses" — tumours measurably shrinking on scans — but the entire commercial thesis for EGFR inhibitors rests on how long responses last, since the whole reason this patient population exists is that prior drugs stopped working. The company has scheduled that specific disclosure for H1 2027, and it is the data point on which the next financing will be priced.

Second, the accelerated approval pathway question in Q4 2026 is higher-stakes than a routine FDA meeting. If the FDA agrees a single-arm trial measuring tumour response rate can support approval in C797S, BlossomHill's capital requirement stays roughly within its stated runway. If the FDA instead requires a randomised trial with survival endpoints — the filing itself flags this risk for earlier-line settings, noting such trials "are likely to require randomized controlled designs with active comparator arms and primary endpoints based on progression-free or overall survival" — the cost and duration rise materially and the Q2 2028 runway becomes tight.

Third, concentration risk is real and increasing. Spending on "other programs and discovery" fell year over year while BH-30643 more than doubled. The pipeline is technically three programs, but only one is funded like a priority, and only one has a defined path to a first approval. BH-30236 (an inhibitor of CLK, an enzyme involved in how cells assemble their genetic instructions, being tested in relapsed leukaemia) is still in dose-escalation — the earliest stage of human testing, establishing tolerable dosing rather than benefit — and BH-501284 has not yet entered humans at all. Neither is close enough to provide a fallback if BH-30643 disappoints.

One smaller item worth noting: the balance sheet carries $23.4 million of operating lease liabilities against $19.2 million of right-of-use assets for San Diego office and laboratory space, with rent expense of $1.3 million in the quarter. For a company with $95 million of cash at period-end, a multi-year facilities commitment of that size is a fixed cost that cannot be cut quickly if a program fails — a modest but genuine constraint on how fast the company could shrink its burn if it needed to.

---

Source: BlossomHill Therapeutics, Inc. Form 10-Q for the quarterly period ended 30 June 2026, filed with the SEC on 18 September 2026 (accession number 0001193125-26-395043). All figures are from the unaudited condensed financial statements and Management's Discussion and Analysis in that filing; items marked (calculated) are derived from disclosed line items. This analysis is for information only and is not investment advice.