Open in interactive viewer → charts, metric popovers & call review

Harrow’s Record Quarter Comes With a Profit Haircut

Q3 revenue hit $99.3M and VEVYE finally converted demand into dollars — but a halved EBITDA guide and a missed QUELL endpoint leave the stock down 34% in 90 days.
HROW · Earnings Call · 2026-11-10

The quarter that broke the old model

Harrow’s third quarter is a study in divergence. Revenue was a record — “Third quarter revenue was a record $99.3 million, up 41% sequentially and 39% year-over-year.” — Andrew Boll, Executive, likely CFO or Head of Finance · 2026-11-10 — yet the stock is down 34.2% over 90 days and 47.2% below its October 2024 peak, with a market cap near $1.14 billion. The reason is the profit line. Management cut full-year adjusted EBITDA guidance to more than $40 million from $80–100 million, even while reaffirming revenue of $350–365 million. That revenue guide implies a Q4 of $136–151 million, or 37–52% sequential growth, a step-up that would be extraordinary even for a company with Harrow’s recent momentum. Unlike the AI-infrastructure and data-center keywords crowding the global tape, HROW’s fresh keywords are company-specific: a trial readout, an NDA timing shift, and a profit reset.

We're updating our full year adjusted EBITDA outlook to more than $40 million, including more than $33 million in the fourth quarter compared with our previous guidance of $80 million to $100 million.

Andrew Boll, Executive, likely CFO or Head of Finance · 2026-11-10
The explanation is familiar from the first half: lower gross margins, a revenue shortfall, and now accelerated spending on the G MELT NDA, QUELL, the TRIESENCE Phase 3, YOCHIL, and the TYRVAYA acquisition. What is less familiar is the scale of the reset. To hit the low end of the old EBITDA range, Q4 alone would have needed roughly $73 million. The new language is a floor — more than $33 million in Q4 — with “room to outperform.” Investors are being asked to underwrite a hockey-stick revenue quarter and a much lower profit base at the same time. The fundamentals show why the margin for error is thin. In the last filed quarter, revenue was $71 million, operating income was negative $17 million, and effective net cash was negative $216 million. Q3 improved the picture — gross margin rose to 75% from 71% in Q2, and adjusted EBITDA turned positive at $21.2 million — but the prior-year Q3 call had already celebrated “Q3, we had great adjusted EBITDA just under $23 million.” — Andrew Boll, Chief Financial Officer · 2025-11-11 This Q3 delivered less EBITDA on 39% more revenue. That is the credibility gap in one sentence. The balance sheet is also nearly fully liability-funded: liabilities to assets stood at 96.4%.

VEVYE: from cash pay to covered, with the algorithm still hidden

VEVYE is the clearest evidence that demand is converting. Revenue reached $39.7 million, up 35% sequentially, while total prescriptions grew 17% sequentially and 83% year-over-year. The company reorganized its commercial force into four teams, with the largest now focused on Ocular Surface Disease, and it added a major national plan covering roughly 10 million lives effective January — its third coverage win this year. That is the bull case: more reps, more covered lives, and better economics on every VEVYE prescription. But the economics remain a black box. When asked about the business-rule changes that drove the sequential revenue out-performance, Mark Baum declined: “we don't want to reveal the specific algorithm that we use in order to improve VEVYE economics.” — Mark Baum, CEO · 2026-11-10 That is a notable shift. In the prior quarter, Andrew Boll said the company would get the full benefit of the amended rules in Q3 and that it should add additional positive momentum to VEVYE ASP. Earlier, he had quantified the expected lift at roughly 30%. Now the language is qualitative — Covered prescriptions are rising, cash pay prescription mix is shifting, and management says the economics are improving — but the size of the net-price improvement is no longer on the table. The stock’s reaction suggests investors wanted more than a directional answer.

QUELL: a miss dressed as medical messaging

The QUELL readout is the other fresh company-specific event. The Phase 4 study randomized 231 patients and compared IHEEZO with routine anesthesia — proparacaine drops plus subconjunctival lidocaine — in intravitreal injection. The primary endpoint failed: “QUELL did not meet its primary endpoint of non-inferiority.” — Amir Shojaei, Executive, likely Head of Clinical or Medical Affairs · 2026-11-10 The success rate was 55.2% for IHEEZO versus 59.7% for routine anesthesia, and the confidence interval exceeded the prespecified 12-point margin. Management argues the study was designed assuming an 88% success rate, while the observed rate was about 57%, making the endpoint inconclusive rather than a finding of inferiority. The secondary data are more supportive: cumulative pain through 24 hours was 2.9 versus 3.7, ocular surface staining increased 1.7 versus 3.2, and adverse events were 27% versus 47%, with no serious adverse events. Harrow now plans to use the data for medical messaging, not a label expansion, and has no plans for a second study. Full results are slated for Hawaiian Eye in January 2027. For IHEEZO — $25.3 million of Q3 revenue, unit demand up 21% year-over-year but down 5% sequentially — this is a commercial support asset, not the clean win the bulls wanted. It may help in the historically strong Q4, but it does not remove the competitive and reimbursement questions around the franchise.

Pipeline arithmetic: earlier NDA, later profits

The pipeline is moving faster than expected. G-MELT’s NDA is now slated for late November or early December 2026, earlier than the prior call’s “let's think about a Q1 2027 submission.” — Mark Baum, Chief Executive Officer · 2026-05-12 That is a genuine positive — a potential on-label, IV-free, opioid-free sedation product for cataract surgery — but the filing fee and preparation costs land in Q4, which is precisely why the EBITDA guide moved down. TRIESENCE is also inflecting: unit demand reached 20,022, up 37% sequentially and 110% year-over-year, and management says “our penetration is certainly in the low single digits.” — Mark Baum, CEO · 2026-11-10 The Phase 3 readout is expected late Q1 or early Q2 2027. BYOOVIZ had a broad-but-shallow launch, with follow-through purchasing now showing up in Q4; the OPUVIZ launch is set for 2027. TYRVAYA should contribute more than $30 million next year, but the 10-Q flags a potential breach of the VEVYE licensing contract with Novaliq. Baum’s answer was not reassuring: “we are under confidentiality. We're in active discussions with them.” — Mark Baum, CEO · 2026-11-10 That is an unquantified overhang on the dry-eye franchise.

What to watch

The next quarter is a referendum on execution, not strategy. Watch whether Q4 revenue actually lands in the $136–151 million range; whether VEVYE’s covered mix lifts net price enough to matter; whether QUELL’s full data changes retina prescribing; whether the G-MELT NDA is accepted with a 2027 PDUFA date; and whether the Novaliq discussion ends in a stronger relationship or a structural constraint. The company has more products and a larger commercial organization than at any point in its history. But the market is no longer paying for the platform build — it is demanding proof that the platform converts into profit. The EBITDA reset is the story, and the revenue record is the test.