アラマー・バイオサイエンシズ（ALMR）2026年第3四半期決算説明会：売上高は68%増、ガイダンスを上方修正

主要トピックス

• アラマー・バイオサイエンス（Alamar Biosciences）が発表した2026年第3四半期の売上高は、前年同期比68%増の3,350万ドルとなり、消耗品売上高が同79%増の1,780万ドルに拡大したことが牽引した。
• 売上高粗利益率は前年同期の59.7%から過去最高の61.5%に達し、装置コストの低減、消耗品の価格上昇、TAPサービスの利益率改善、および良好な製品ミックスが寄与した。
• 経営陣は2026年通期の売上高ガイダンスを従来の1億1,600万〜1億2,000万ドルから1億2,100万〜1億2,300万ドルに引き上げた。これは中間値で64%の成長に相当する。
• 同社はUKバイオバンクの8万件のサンプルをプロファイリングする事業に選定されたほか、6万件以上の血清サンプルを対象とするGSKとの提携を締結した。
• アラマーは、2026年に設置台数を少なくとも100台拡大し、装置1台当たりの平均年間プルスルーを40万ドル以上に維持する計画を順調に進めている。
• 第3四半期の純損失は前年同期の800万ドルから1,330万ドルに拡大した一方、四半期末時点の現金および現金同等物、短期投資、制限付き現金の合計は2億3,150万ドルとなった。

主要財務データ

事業・業績動向

マルチプレックスパネルキットの需要に支えられ、消耗品は引き続きアラマーの最大の売上カテゴリーとなった。経営陣は、装置の設置台数増加に伴い、継続的な消耗品売上が拡大し続けていると述べている。

第3四半期の装置設置実績は、学術機関向けとバイオ医薬品企業向けがほぼ均等であった。大半は新規顧客向けであったが、今四半期は既存顧客の新たな研究室への設置台数としても過去最高を記録した。

第3四半期の売上高構成比は、バイオ医薬品企業が49%、学術・研究機関が44%、代理店が7%となった。バイオ医薬品企業が最大の顧客カテゴリーとなったのは今四半期が初めてである。地域別では、南北アメリカが売上高の70%、EMEAが20%、APACが10%を占めた。

TAPサービスは、大型のカスタムアッセイ開発プロジェクトや人口集団研究に牽引され、四半期として過去最高の売上高を達成した。経営陣は、一部案件の収益認識時期が第4四半期から第3四半期へ前倒しされたことが要因であり、第4四半期に同様の水準が繰り返されることは見込んでいないと説明した。

アラマーのUKバイオバンクプロジェクトでは、NULISAを用いて8万件の血液サンプルから神経に関連する220種類のタンパク質をプロファイリングする予定である。解析処理は2026年第4四半期に開始され、2027年に本格化し、2027年末までに完了する見通しである。

GSKとの提携では、NULISAseq Immune 340パネルおよびNeuro 220パネルを使用し、呼吸器、肝臓、腎臓、神経、自己免疫疾患に関する6万件以上の血清サンプルをプロファイリングする。同研究は2027年第1四半期に開始される予定だが、経営陣は完了時期を明らかにしていない。

第3四半期中に35件以上が追加されたことで、同社の科学論文・プレプリント掲載数は累計200件を超えた。また、経営陣はエモリー大学の研究を強調し、検証コホートにおいてALS（筋萎縮性側索硬化症）と健常対照群を98%のバランス正確度で識別する10種類のタンパク質シグネチャーが特定されたことを挙げた。これはNFL単独の約90%と比較して高い水準である。

業績見通し（ガイダンス）

アラマーは、2026年通期の売上高見通しを従来の1億1,600万〜1億2,000万ドルから1億2,100万〜1億2,300万ドルに引き上げた。新たな見通し範囲の中間値は、前年比64%の成長に相当する。

第4四半期について、経営陣は消耗品売上高の前期比増加がTAP売上高の減少を相殺すると見込んでいる。通期見通しの上限を達成した場合、売上高は第3四半期から前期比で増加することになる。

同社は当初2027年に計画していた特定の研究開発プロジェクトを前倒しで実施するため、第4四半期の営業費用は増加する見通しである。これらの投資は、装置およびコンテンツのロードマップを推進するものである。

経営陣は、2026年中に少なくとも100台の装置を追加設置し、装置1台当たりの平均年間プルスルーを40万ドル以上に維持するという目標を改めて表明した。

2027年に向けて、アラマーは設置台数の拡大を継続し、新たな疾患領域におけるマルチプレックスパネルの投入、診断パートナーシップの確立、および臨床用装置「ARGO HT/DX」の年末までのFDA販売承認申請を計画している。

リスクと注視すべき事項

• 複数の大型プロジェクトが第3四半期に完了したことや、プロジェクトの時期が一部顧客のスケジュールに依存することから、第4四半期のTAP売上高は前期比で減少する見通しである。
• 経営陣は、第3四半期の粗利益率61.5%を底値（下限）とみなさないよう警告した。利益率は製品ミックスやTAPの稼働状況によって変動する可能性がある。
• 経営陣はパイプラインが堅調であると説明しているものの、装置の販売サイクルは比較的長いままであり、個別の設置時期を予測することは困難である。
• 研究開発の加速に伴い、第4四半期の営業費用は増加する見込みである。
• アラマーは、UKバイオバンクおよびGSKとのコラボレーションについて、四半期ごとのサンプル処理予測やプロジェクト別の売上寄与度を開示していない。

アナリスト質疑応答の要点

保守的な第4四半期見通しについて質問された際、経営陣は第3四半期が第4四半期から前倒しされたTAP案件の恩恵を受けたためであると説明した。第4四半期は消耗品が前期比で増加する一方、TAPは減少すると予想している。

装置の需要について、経営陣はここ数四半期で設置台数が増加しており、2027年に向けたパイプラインも順調に推移していると述べた。設置の時期には依然として不確実性があるものの、顧客の資金調達や購買動向に重大な変化は見られないとしている。

大規模研究に関して、経営陣はUKバイオバンクがフォローアップ研究、トランスレーショナル研究、そして最終的な臨床応用をサポートする可能性があると強調した。第4四半期に見込まれる寄与は改定後のガイダンスに含まれており、将来のプロジェクトの前提条件はリスク調整後に組み込まれている。

「ARGO HT/DX」に関して、経営陣は計画中のFDA申請に、同プラットフォームの超高感度およびマルチプレックス機能を実証するための検査が含まれると述べた。同社は、パートナー企業が最終的に複数タンパク質シグネチャーを用いたLDTおよびIVD検査を開発することを見込んでいる。

決算説明会トランスクリプト全文

決算説明会の完全なトランスクリプト

経営陣による説明

Operator

Good day, and thank you for standing by. Welcome to Alamar Biosciences' Third Quarter 2026 Financial Results Webcast Call. [Operator Instructions] Please be advised that today's conference is being recorded.

I would now like to hand the conference over to your first speaker today, Carrie Mendivil, VP of Investor Relations and Corporate Communications.

Good afternoon, everyone, and thank you for joining us today to review Alamar Biosciences' Third Quarter 2026 Financial Results.

Joining me on the call are Yuling Luo, our Chief Executive Officer; and Justin McAnear, Chief Financial Officer. Our President, Todd White, will join for Q&A.

Earlier this afternoon, we issued a press release detailing our third quarter financial results, and we posted an accompanying presentation to the Investors section of our website.

Before we begin, I'd like to remind you that statements we make during this call will include forward-looking statements as defined under applicable securities laws. Forward-looking statements are subject to risks and uncertainties, and the company can give no assurance they will prove to be correct.

Additionally, we are not under any obligation to provide further updates on our business trends or our performance during the quarter. To better understand the risks and uncertainties that could cause actual results to differ, we refer you to the periodic reports that Alamar Biosciences files with the Securities and Exchange Commission.

I'll now turn the call over to Yuling Luo, Alamar's CEO.

Yuling Luo

Thank you, Carrie. And thank you all for joining us this afternoon.

I will start our call with a brief overview of Alamar and our platform, then walk through our Q3 business highlights and our strategy to unlock the opportunity ahead. I will then turn it over to Justin to cover our financials and update full-year outlook.

Starting on Slide 3. We founded Alamar with a bold mission to enable the earliest detection of disease through precision proteomics. That purpose still guides every decision we make. Proteomics is one of the largest opportunities in life science, and it remains largely untapped. Researchers have long wanted to measure the proteins that signal disease, but the tools to sufficiently detect them have not existed. Alamar was built to solve that problem.

Turning to Slide 4. Precision proteomics depends on 5 capabilities working together, ultra-high sensitivity, high specificity, flexible multiplexing, broad dynamic range and seamless automation. Other technologies have forced researchers to choose, giving up sensitivity to gain multiplexing or giving up steps to keep the workflow simple. Alamar is the only platform that delivers all 5 without compromise. Since we launched our precision proteomics platform in January 2024, customers have adopted rapidly, and that momentum continued this quarter.

Turning to our top line results on Slide 5. Total revenue for the third quarter grew 68% year-over-year, led by consumables, which grew 79% and made up 53% of total revenue. We also continue to improve our gross margin, which reached 61.5% for the first time in Q3. Based on this performance, we are raising our full-year revenue guidance.

Turning to Slide 6. Our near-term growth strategy is powered by 3 drivers. The first is expanding our installed base into new institutions and geographies. The second is building new content, both to deepen our position in neurology and inflammation, where we first establish the platform and move into adjacent disease areas with significant unmet need. The third is working with leading institutions and biopharma partners on new applications and large studies, which in turn grew our publication base and drives awareness and adoption.

Turning to Slide 7. We continue to make important progress across our business. It is increasingly clear that NULISA is the platform of choice to reach the low-abundance proteome, and that differentiation drove 2 significant recent wins. We were selected to deliver U.K. Biobank's Vanguard study of blood-based biomarkers for neurodegeneration, profiling 80,000 samples, and we were selected for a large-scale partnership with GSK, profiling more than 60,000 samples to advance its drug discovery program.

Our literature footprint grew to more than 200 publications and preprints including a study in The Lancet Neurology that is the first longitudinal evaluation of a broad plasma biobank panel in patients receiving anti-amyloid therapy in real-world practice.

We also strengthened our Board and leadership team. Bob Ragusa, the former CEO of GRAIL and the former Chief Operations Officer of Illumina, joined our Board of Directors. Bob helped scale 2 of the defining companies in genomics and early cancer detection and believes the future of disease detection lies in multi-marker testing using proteins. His experience will be invaluable as we build towards clinical applications. We also welcome John Brottem as General Counsel; and Amelia Merrill as Chief People Officer as we scale the organization for our next phase of growth.

Turning to Slide 8. We continue to be inspired by the data and the biologic insight being generated by our customers. At our User Group Meeting in October, customers shared exciting new data on treatment response in prostate cancer, healthy aging and on protein quantitative trait loci or pQTLs, which link genetic variation to protein levels. Much of this work is possible because of our ability to reach the low-abundance proteome.

Many of the proteins most relevant to disease such as cytokines, signaling molecules and proteins released from the brain, circulate in blood at extremely low concentrations. Conventional approaches often cannot detect them reliably or at all and the platform that measure many proteins advance typically give us sensitivity as they scale. That gap matters most in the applications with greatest potential impact, including early detection of disease, disease risk prediction and minimal residual disease where the signals are faint and the ultra-high sensitivity is essential.

NULISA was built for exactly this problem, measuring hundreds of low-abundance proteins together from a small blood sample. That is why researchers and biopharma partners who need to reach this part of proteome at scale are turning to NULISA. This matters even more as the field turns to machine learning and AI models built on proteomic data. These models learn what disease looks like by comparing it to health. So they need to see low-abundance proteins measured accurately in healthy samples where levels are often at their lowest.

When a platform cannot detect a protein, that missing value is typically filled in statistically, and a model trained on that data risk learning patterns that reflect the gaps in the measurement rather than real biology. Reliable detection across healthy and disease sample gives these models a more accurate foundation, and we believe that will make data generated on our platform increasingly valuable as AI becomes central to biomarker discovery.

Turning to Slide 9. The U.K. Biobank study is an excellent example of the field recognizing the value of our platform. In October, together with Nightingale Health, we were selected to deliver UK Biobank's Vanguard study on blood-based biomarkers for neurodegenerative disease. Running on NULISA in Nightingale's laboratories, the study will profile 220 neurologically relevant proteins across 80,000 blood samples from individuals who were healthy when they enrolled.

Founded by The Michael J. Fox Foundation, LifeArc and ARUK, this study will create one of the world's largest open access resources for understanding the earliest molecular signs of neurodegeneration. This will be the first time brain-derived and other low-abundance proteins are measured from the UK Biobank one of the most widely used health research resources in the world. It will also create the first definitive proteomics data set for Parkinson's disease, the fastest-growing neurological disorder, which affects more than 12 million people today.

Together with Alzheimer's disease, which affects more than 55 million people worldwide, these diseases are usually diagnosed only after years or decades of molecular damage. Profiling people long before symptoms appear, give us the chance to find the earliest signals when intervention is still possible. The study is expected to be complete by the end of 2027, and the data will be integrated with UK Biobank's imaging, genetic and longitudinal health data and are made available to approved researchers worldwide.

Turning to Slide 10. Today, we announced a strategic scientific partnership with GSK, another strong validation of our platform by biopharma. Together, we will study how circulating protein signatures relate to disease progression, biological subtypes and treatment response across respiratory, liver, kidney, neurological and autoimmune conditions using real-world health care data that is far more representative of patients than clinical trials. The work will focus on linking protein signals to clinical outcomes, identifying new biomarkers, uncovering the biological subtype beneath a single diagnosis and understanding how patients respond to existing therapies.

Patients who share the diagnosis often have biologically distinct disease, and that remains one of the central challenges in drug development. We believe NULISA can help stratify patients at the earliest possible stage and improve how disease is classified and treated. To support the collaboration, we will profile more than 60,000 serum samples using NULISAseq Immune 340 Panel and Neuro 220 Panel. This partnership will create one of the most comprehensive biomarker data set applied to GSK's drug discovery program and allow Alamar to potentially develop and commercialize novel protein signatures in precision prevention, treatment and earlier disease detection.

Turning to Slide 11. Our scientific footprint continues to grow. We added more than 35 publications and preprints in Q3, bringing our cumulative total to more than 200 across neurodegenerative disease, oncology and autoimmune conditions. Several of these studies show our newest product already at work.

Just 3 months after launch, the first pre-print using our eMTBR-Tau assay supports its use in staging Alzheimer's disease. A publication in Nature Medicine also supports new applications for our platform in women's health. And an independent study from Emory University published in Nature Communications identified a 10-protein signature that distinguish ALS from healthy controls with 98% balanced accuracy in the validation cohort compared with about 90% for NFL alone. It's a clear example of why measuring many proteins at once matters. Each new study adds independent evidence of what NULISA can do and helps bring new researchers to our platform.

Turning to Slide 12. One of the most exciting things for us is seeing NULISA generate new insight in translational setting. A study from Dr. Carlos Cruchaga's lab at Washington University published this quarter in the Lancet Neurology is a great example. It is the first longitudinal study to evaluate a broad panel of plasma biomarkers in patients receiving anti-amyloid therapy in routine clinical practice.

Using our NULISAseq CNS 120 Panel, researchers measured 130 proteins across nearly 2,400 samples from patients treated with lecanemab and from untreated comparison group using just 25 microliter of plasma per sample.

The findings addressed one of the biggest open questions in Alzheimer's care. Why clearing amyloid does not deliver the same clinical benefit for every patient? The protein associated with amyloid clearance was largely different from those associated with later cognitive decline, which suggests two reflect distinct biologic processes. Among treated patients, 34 of the 130 biomarkers changed significantly over the course of therapy, and they did not all move in the same direction.

The study also shows why sensitivity and specificity matter. Brain-derived form of tau consistently outperformed the same protein circulating in the rest of the body, accounting for 4 of the top 5 markers that distinguish the treated patients from the controls. Separating brain-derived tau from its periphery form and measuring along the more than 100 other proteins from a single, small sample is exactly what our platform was built to do.

As more therapy move into routine care and late-stage trial, measuring one or 2 markers is not enough to understand how patients respond to treatment. This study shows the richer picture our platform can provide, and it speaks directly to what our pharmaceutical partners need as they work to understand the treatment response in their own programs.

Turning to Slide 13. The data we continue to see from our customers, including these studies I just described, reinforce our belief that the future of testing will be powered by multi-marker protein panels. Patients with the same diagnose can differ significantly in their underlying biology.

In complex disease like Alzheimer's, researchers and drug developers also need to understand which pathologies are present, how the disease is progressing and how a patient is responding to treatment. No single protein can answer all those questions. This becomes even more important as testing moves towards population scale. When screening broad populations where disease is relatively rare, even small gap in accuracy can lead to missed cases or large number of false positives.

We believe single marker tests will not be sensitive or specific enough for that setting, and that testing, measuring many proteins together, will become the standard. 20 years ago, the market moved from single marker PCR testing to next-generation sequencing, and that transition transformed oncology and ushered in a new era of precision medicine. We expect a similar shift in proteomics from single marker immunoassays to multi-marker protein signatures that can guide diagnose, prognosis, treatment selection and disease monitoring across the entire patient journey. NULISA was built to lead that shift.

Turning to Slide 14. That shift towards multi-marker testing is what makes the opportunity ahead of us so large. Today, we serve the research market, which is substantial on its own and estimated to reach $11 billion over the next decade. We believe our platform can also support clinical diagnostics, early detection and health monitoring at population scale. What makes this possible is the combination of deep multiplexing and sensitivity that clinical use requires.

Our initial approach is to provide the platform for partners to build upon. We will provide the hardware and operating system, and a partner with disease expertise, clinical development and regulatory experience, reimbursement pathways and commercial channels will build applications, which, in our case, are diagnostic tests.

To get there, we're in the process of developing ARGO HT/DX, our clinical instrument and plan to submit for FDA marketing authorization by the end of next year. We expect this to support use in late-phase clinical trial, open the door for partners to develop LDT and IVD tests and drive incremental demand in research since our customers value a tool with a clear path to the clinic. Once authorized, we plan to work with IVD and LDT partners to develop differentiated tests, either stand-alone or as part of multiomic solutions.

We are at an extraordinary moment for proteomics, and I believe Alamar is uniquely positioned to shape the future of the field. I'm deeply proud of what this team has built and energized by what lies ahead.

With that, I will turn the call over to Justin.

Justin McAnear

Thanks, Yuling. Good afternoon, everyone.

I'm pleased to walk through our third quarter 2026 financial results in more detail. Unless otherwise noted, all growth rates reflect year-over-year comparisons. And as a reminder, we plan to update the installed base and average instrument pull-through metrics on an annual basis, and we'll be sharing those on our year-end 2026 earnings call.

Turning to Slide 15. Total revenue for the third quarter was $33.5 million, up 68% compared to $19.9 million in the third quarter of 2025. Consumables continued to be the standout performer, accounting for more than half of the total revenue in the third quarter. Consumables revenue was $17.8 million, up 79% year-over-year, driven by strong demand for our multiplex panel kits.

Instrument revenue in the third quarter was $8.5 million, up 50%, driven by continued growth in new instrument placements. We are pleased with the pace of placements and the installed base continues to support a strong and growing recurring consumables revenue stream.

Service and other revenue was $7.2 million in the quarter. As a reminder, this line consists mostly of TAP services with maintenance contracts making up a smaller portion since much of our instrument installed base is still under its initial 1 year warranty. TAP is designed to bring in new customers, support custom assay development and beta test new products before launch.

I'd like to take a moment to commend our TAP team for delivering our largest TAP revenue quarter-to-date. This was driven by some large custom assay development projects and population studies that the team was able to complete within the quarter, and we don't expect to repeat at this level in Q4. Given this, we expect TAP revenue to decline sequentially in Q4. And over time, we continue to expect the rest of the business to grow more quickly than TAP.

Turning to Slide 16. We'll now look at the geographic breakdown for the third quarter. The Americas represented 70% of revenue, EMEA represented 20% and APAC represented 10%. We continue to build out our international presence, and we see meaningful runway in both EMEA and APAC as we deepen our distribution partnerships and direct commercial efforts in those regions.

From a customer mix perspective, 49% of third quarter revenue came from biopharmaceutical companies, 44% from academic and research institutions and the remaining 7% from distributors. This is the first quarter in which biopharma customers made up the largest share of revenue, up from 42% in the second quarter. Biopharma customers tend to run larger, more sustained studies that drive meaningful consumables pull-through over time, and we are encouraged by the growing traction of our platform in drug development.

Turning to Slide 17. Gross profit for the third quarter was $20.6 million compared to $11.9 million in the prior year period.

Gross margin was 61.5% in the third quarter of 2026, up from 59.7% in the third quarter of 2025. Gross margins were again at an all-time high. The improvement was driven by several factors: lower instrument costs, higher average selling prices for consumables, improved margins on TAP services and a favorable shift in product mix.

Total operating expenses in the third quarter were $35.8 million compared to $19.8 million in the prior year period. The increase reflects investments in people, infrastructure and consumable content as well as incremental public company costs.

Research and development expenses were $16.2 million in the quarter, up 58% from $10.2 million a year ago. The increase reflects higher lab supply costs to support our expanding consumable panel menu and an increase in personnel costs.

Selling, general and administrative expenses were $19.6 million in the quarter, up 105% from $9.6 million in the prior year period. This was primarily driven by an increase in personnel costs as well as a higher professional services cost for legal and accounting services.

Loss from operations was $15.1 million in the third quarter compared to $7.9 million in the third quarter of 2025. This includes $3.9 million of stock-based compensation for the third quarter of 2026 as compared to $0.7 million for the third quarter of 2025.

Net loss was $13.3 million in the quarter compared to $8 million in Q3 2025.

We ended the third quarter in a strong financial position. And as of September 30, 2026, we had $231.5 million in cash, cash equivalents, short-term investments and restricted cash.

We are well capitalized to execute our plan, and we intend to deploy this capital strategically to scale our organization and capabilities in support of future growth by expanding our commercial sales and support function, increasing manufacturing capacity and enhancing our research and development organization to expand product content.

Turning to Slide 18, our outlook for the rest of the year. We are raising our full-year 2026 revenue guidance and now expect revenue in the range of $121 million to $123 million, representing annual growth of 64% at the midpoint. This compares to our previous guidance of $116 million to $120 million. For the fourth quarter, we expect an increase in consumables revenue to offset the decrease we expect in TAP.

On the expense side, we expect operating expenses to increase in the fourth quarter as we pull forward certain R&D projects that were planned for 2027. Accelerating this work supports our content and instrument road map, and we remain focused on a disciplined path to profitability.

With that, I will turn the call back to Yuling.

Yuling Luo

Thanks, Justin.

Before I turn to our goals, I want to take a moment to recognize Todd White. As we announced earlier today, Todd plans to retire at the end of the year, after which he will continue to serve as an advisor to the company through June 2027. Todd and I have worked together for more than 25 years. And when he joined Alamar as our Chief Financial Officer and Chief Business Officer in 2021, he brought the same trust and partnership to build this company. Todd helped build the foundation for our growth from finance organization and operations to guiding us through our IPO.

Personally, I'm deeply grateful for his friendship, his counsel and everything he has given to Alamar. On behalf of our Board and the entire team, thank you, Todd.

We have a strong leadership team in place, and we're well positioned to continue executing on our plan. Todd is with us today and will join us for Q&A.

Before we open the call to questions, I want to come back to the goals we set out last quarter. Turning to Slide 19. On goals one and 2, we remain on track to grow our installed base by at least 100 instruments in 2026 and to maintain per instrument pull-through above $400,000. On our third goal, advancing into more large-scale studies, the UK Biobank and the GSK collaborations are exactly the kind of progress we set out to make.

Looking to 2027, we plan to keep expanding our installed base, launch a multiplex panel in a new disease area, submit ARGO HT/DX for FDA marketing authorization and establish partnerships to support our diagnostic strategy.

I want to thank the Alamar team for another strong quarter. We have the capital, the road map and the people to capture the opportunity ahead.

With that, we will now open the call for questions. Operator?

Operator

At this time we will conduct a Question-and-Answer Session. [Operator Instructions] Our first question comes from the line of Dan Brennan of TD Cowen.

質疑応答

Daniel Brennan

Congrats on the quarter. Maybe just to start off with the near-term strength on TAP and kind of what's implied for the fourth quarter, I heard Justin and Yuling discussed it appears more maybe one time. I don't know if there was a pull forward, but maybe just discuss a little bit about why you think the strength you saw in Q3 won't continue? I mean you came off a really strong Q2 as well.

And then kind of just related to that, just consumables, the pull-through was better than expected, too. Just any color you can provide on kind of what you're seeing there and what's driving the upside?

Justin McAnear

Yes. Thanks, Dan, for the question. This is Justin. I'll take that one.

As far as looking at TAP results for Q3, Q3 benefited from some large custom assay development projects, which we don't expect to repeat at the same level in Q4. TAP projects are normally partially driven by customer time lines. So the quarter that they land in can vary. And so when we look at the second half of 2026, we see that playing out in whole, slightly above where we had described it last quarter with some timing shifts into Q3 from Q4.

As far as the pipeline and what we're seeing in consumables, I will say, overall, our pipeline remains strong. Our underlying business remains on track. As far as instruments go, we expect to place 100 or more instruments this year. And as far as consumables go, there's no meaningful changes in trends to call out from what we've experienced before. And we are still tracking to keep pull-through above 400,000 average annual instrument pull-through for the year.

Daniel Brennan

Terrific. And then maybe just on the Glaxo and the U.K. Biobank deal. Maybe could you give a little color, I mean, both meaningful in their own rights, but just how do we think about the revenue potential? I know they're important kind of validation of the platform. But just from a revenue basis, any color on how that might come into the model over the next, call it, 18 months or so? And anything you could speak to on the pipeline for other large deals that might be kind of working?

Yuling Luo

Maybe I'll answer first and Justin can chime in.

We're very pleased to announce that NULISA was selected for the 80,000 sample study. I think the big opportunity is where this project leads to. UK Biobank is one of the most widely used data sets in the biomedical research and thousands of researchers around the world will work with the NULISA data. This gives the field a large, high-quality data set that shows the value of measuring low-abundance proteins at the population scale. It supports our position in other large cohort studies that we are pursuing.

Over time, we expect funding from this study will create demand for follow-up studies, translational work and eventually clinical applications, all of which runs on our platform. Justin?

Justin McAnear

And as far as the timing on this, Dan, you're right, this plays out over multiple quarters. For UK Biobank, we expect the sample processing to begin in Q4, and we have contemplated that in our updated guide and expect that to ramp up in 2027.

As far as breaking out the specific revenue contribution from any single project, we won't typically do that. But like I said, the anticipated contributions for Q4, those were included in what we just updated the guide to.

Yuling Luo

Just to add to Justin, that we just announced our collaboration with GSK. This is a really exciting collaboration for us. It shows that the major pharma recognize the value of platforms beyond the neurodegeneration. It is expanding into a number of new disease applications such as respiratory, liver, kidney and autoimmune and is used in real-world patient samples. It's also a great validation of the value biopharma is seeing in our platform.

Operator

Our next question comes from the line of Puneet Souda of Leerink Partners.

Puneet Souda

First one on the guide. I mean, you beat by nearly $4 million, you raised by similar. I appreciate the comments on TAP. But even when reaching the top end of the guide for the year, it implies flat sequential revenue. You're talking about U.K. Biobank also coming in, you're talking about conversion from TAP to more to consumables. Fourth quarter is generally stronger in placements. I don't know if there's a budget flush in there, too.

So just given all those drivers, I mean, just sort of maybe walk us through what are the considerations why flat sequentially only?

Justin McAnear

Yes. Thanks for the question, Puneet. We did raise our full-year guide to reflect a strong third quarter. We're taking a prudent approach to Q4. As I mentioned, a step-up occurred in the back half of the year earlier than we had originally contemplated it. Like we said, Q3 did benefit from some large custom assay development projects within TAP.

We do expect that top line to decrease from Q3 to Q4. And then we expect consumables revenue to increase sequentially from Q3 to Q4. And then at the top end of the range would be a sequential increase from Q3 to Q4.

Puneet Souda

Got it. But one of the things I think we were hoping for by this time would be the number of instrumentations that have been in the pipeline and also may be part of the TAP program, some of these customers coming in somewhat more strongly versus the 100 that you've outlined for the year. That's a strong number. But I'm just wondering what's been the feedback on instrumentation? Is there anything that's holding them back in terms of funding or just timing?

And just maybe just talk to us about ARGO. ARGO clearly has a lot of value proposition. So just trying to understand that the instrumentation placements, when should that number start to inflect meaningfully?

Justin McAnear

I'll start with that, Puneet, and then Yuling can add after.

But I would say that our instrument pipeline is robust. It remains quite strong. We've been increasing the number of placements each quarter over the last few quarters. We have a number of instruments in upside as we look out over the quarters in the near term and our 2027 instrument funnel is taking shape nicely.

As far as placement dynamics overall, when you look at this quarter, it's been roughly 50-50 between academic and biopharma, roughly mirroring our overall revenue. It is a longer sales cycle due to its cost, but I would not call out any meaningful trends that have changed from previous quarters. And we still feel really excited about the opportunity that we have ahead.

Yuling Luo

Given that, I think even though the pipeline remains to be strong and robust, I think the timing of replacement is harder to predict. But overall, we certainly see a continued increase in installed base.

Operator

Our next question comes from the line of Dan Arias of Stifel.

Daniel Arias

Alzheimer's is a beachhead therapeutic area for you guys pretty clearly. But you mentioned that Parkinson's is actually the fastest-growing neurological disease, I believe. I'm curious whether the degree to which biomarker research is keeping pace with disease occurrence is something that you think is like a driver for placements going forward and then project work going forward?

How would you compare Parkinson's to Alzheimer's momentum at this point? Would just be helpful to get some context there.

Yuling Luo

Yes. So clearly, over the last couple of years, the largest adoption of our platform is in the Alzheimer's disease space, where we have unique offerings and high-value biomarkers that were able to measure those -- such as pTau217 MTBR markers in blood. And that really is a differentiation that we have created.

And now with the recent launch of Neuro 220 we have expanded our content from covering Alzheimer's disease to Parkinson's as well as other disease areas. And that's why the research community are very excited about our Neuro 220, and that's one of the major reasons why UK Biobank has selected us to do this large-scale 80,000 sample studies.

Remember, we disclosed in the last quarter, we had a collaboration with Michael J. Fox Foundation, where they have funded us to deliver over a dozen proprietary unique biomarkers for Parkinson's disease. And that really is one of the major reasons that we're able to carry out this large-scale study within UK Bio and covering pretty much all the other disease patients in the sample cohort.

I also noticed that we're able to expand beyond the Parkinson to other neurodegenerative disease areas such as ALS with a recent publication come from Emory University, where they have found a disease signature that can accurately diagnose Alzheimer's disease at over 98% accuracy compared with a single marker test.

So we're quite excited about the adoption of our platform and our panels in the neurodegenerative disease areas. And certainly, we are seeing the application beyond the neurodegenerative disease with the recent announcement of collaboration with GSK and noticed that the highlight of this collaboration really is about the incorporation of Immune 340 covering other disease areas such as respiratory, liver metabolic disease and autoimmune disease.

Daniel Arias

Maybe just as a follow-up, you mentioned clinical applications and Justin mentioned pulling forward some R&D dollars. So is that related to the diagnostic system that you guys are working on? Can you just talk maybe about the extent to which that's the research focus that's getting brought into the picture versus more panel content for research? Maybe it's both. Just sort of wondering about where you guys are thinking on the R&D front.

Yuling Luo

I think certainly within our company, our goal is to submit for FDA clearance for our ARGO HT/DX platform by the end of next year. Along with this submission, there will be a test that goes through the process. And what we are aiming for is a test that can demonstrate our platform's ultra-high sensitivity as well as the multiplex capabilities.

And as you can see that there are many studies out there that are highlighting a number of different multi-marker signatures that provide a differentiated performance compared with single marker test. And we certainly expect over time that some of those signatures could turn into LDT and IVD tests in the coming years.

Justin McAnear

Yes. And just to add to that, Dan, on the rest of the pull forward for R&D projects, we see significant growth opportunities in the future and are choosing to pull some of those forward now. These are multiyear programs that we expect to add to revenue growth over the longer term as well. And then we're also deliberately making investments in the commercial org and the public company infrastructure to support growth as well.

Operator

Our next question comes from the line of Casey Woodring of JPM.

Casey Woodring

Just a quick clarification on the GSK, the 60,000 sample study, I don't think I caught a time line on that. When do you expect that to be completed?

Yuling Luo

We announced this strategic sort of collaboration today. This study will begin in Q1 2027. We will use our Immune 340 and Neuro 220 Panel to provide more than 60,000 serum samples across respiratory, liver, kidney, neurological and autoimmune conditions. We don't break out revenue contributions from individual projects or partnership. These studies are including our internal forecast on a risk-adjusted basis. We aren't providing a time line on its completion as of today.

Casey Woodring

I guess my follow-up to that would just be on the 2 studies. Just how should we think about how many samples you guys will be running per quarter and just kind of spread across like the 80,000 sample study from UK Biobank, for example, like you're going to complete that by the end of '27. Should we just assume something like 16,000 samples a quarter, considering you're going to start processing in 4Q? Just any sort of sense on contribution per quarter from a sample volume perspective would be helpful to understand.

Justin McAnear

This can fluctuate quarter-to-quarter. There can also be ramp-up time within the project for additional analysis. So we aren't guiding to projects or samples per quarter on this. But when it does come time to guide for full-year, we will be contemplating this in our full-year guide, and we could potentially share additional detail at that time.

Operator

Our next question comes from the line of Avantika Dhabaria of BofA.

Avantika Dhabaria

Congratulations on the quarter. How should we think about 4Q gross margins, especially if TAP steps down and consumables grow? And then I have a follow-up.

Justin McAnear

Yes. So I'll take that one on gross margin. So it was a great quarter for gross margin. When you look at the different variables that impact the margin, I would say most all of those went the right direction. So it was driven overall by lower instrument costs, higher average selling price for consumables, improved margins on TAP services and a favorable shift in product mix towards consumables.

When we think about the TAP services, too, you do have a relatively fixed base within a quarter as far as the team size in which to perform those projects. And when they overdeliver like they did in Q3, you can get a benefit to the gross margin.

So I would be careful about treating the margin that we had in Q3 as a floor. I will say that this can continue to fluctuate quarter-to-quarter, and it's going to be highly sensitive to product mix in future quarters as well. But I'd also like to add that over time, we do expect the gross margin to trend upward as consumables scale and continue to become a larger portion of our overall revenue.

Avantika Dhabaria

And then on the placement pipeline, are you seeing these placements like within new customers, or are you seeing existing customers purchase more instruments?

Justin McAnear

Yes. So for Q3, the majority of placements are still coming from new customers. I'll also add that this has been our biggest quarter-to-date of instruments going to existing customers in new labs. So basically demonstrating our ability to expand to new teams and application areas.

I mentioned before, instrument placements have been about 50-50 between academic and biopharma, roughly mirroring our overall revenue split as well.

Operator

Thank you. This concludes the question-and-answer session. Thank you for your participation in today's conference. This does conclude the program. You may now disconnect.