A-Portfolio

The company in one breath

Iambic Therapeutics, ticker IAM on the Nasdaq Global Select Market, uses artificial intelligence and automated laboratories to design cancer drugs, then develops the most promising molecules itself.

Drug companies also pay Iambic to use the platform. The company has no approved medicine and no product sales, which means every drug dollar still sits years and several clinical trials away. Good would look like the platform repeatedly producing useful human medicines while partnerships pay part of the research bill.

Why now

IAM finished its first trading day at $18.91 on October 15, up 18.2% from its $16 initial public offering price. An initial public offering, or IPO, is the first sale of company shares to public investors. The jump means buyers immediately paid a premium for the story. Good entry quality would look like the science improving faster than the valuation, not a second-day chase because the letters “AI” walked into a laboratory coat.

The company sold 9.375 million shares and raised $150 million before fees. That means Iambic added enough money to push three cancer programs forward. Good would look like that cash funding decisive human evidence before another stock sale makes each current share own less.

The real event sits inside the lead drug, IAM1363. It is a pill built to attack HER2, a protein that can drive several cancers, while reaching tumors in the brain and avoiding EGFR, a related protein whose inhibition can cause side effects. Good would look like meaningful, durable tumor shrinkage in larger groups, especially patients with brain metastases.

Another catalyst is close. Iambic filed to begin human testing of IAM217, its second wholly owned drug candidate, on September 30. That filing asks the Food and Drug Administration for permission to start a trial. Good would look like clearance, first-patient dosing, and tolerable early safety without distracting management from IAM1363.

The numbers, translated

• First-half collaboration revenue was $12.8 million, up 225% from $3.9 million. Collaboration revenue is money from drug-company partners, not patients buying medicine. Good would look like this funding source growing toward $30 million a year without one partner supplying most of it.
• First-half net loss was $50.1 million, up from $33.2 million. Net loss is the final amount the business lost after every expense, and the hole widened by roughly half. Good would look like losses growing slower than the pipeline’s clinical progress and partner payments.
• Research spending was $57.1 million in six months, up 74%. That is the money spent on scientists, laboratories, and trials. Good would look like the extra spending producing new trial starts, more patients, and stronger data, not merely a more expensive science fair.
• Operating cash use was $24.9 million in the first half. That is the cash that left while running the business. Good would look like yearly cash use staying below $100 million until the lead drug reaches a pivotal trial, the study meant to support approval.
• Cash was $207.9 million on June 30, and estimated IPO proceeds add $135 million after fees. That puts rough pro forma cash, meaning cash as if the deal had already closed, near $342.9 million before spending after June. Good would look like at least $200 million still available when the next major IAM1363 data arrive.
• Management says that money funds operations through the fourth quarter of 2028. Runway is how long the bank account can carry the plan. Good would look like clear clinical proof and a regulator-approved late-stage path arriving at least a year before the runway ends.
• At the $18.91 close, 47.4 million basic shares value the company near $896.4 million. Market value is the price of every current share. Subtracting rough post-IPO cash leaves about $553.5 million for the drug pipeline, platform, and partnerships. Good would look like paying below $450 million for that operating business until larger human datasets confirm the platform’s promise.
• Potential options, restricted shares, and warrants lift the fully diluted share count near 55.1 million. Fully diluted means counting shares that can appear later. At the same price, that produces a value near $1.04 billion. Good would look like future share growth staying below 5% a year after the IPO awards settle.
• Among 53 evaluable patients at the chosen 960-milligram daily dose, 36% had a best partial response and 23% had a confirmed response. A partial response means measurable tumors shrank by the trial’s rules. Confirmed means a later scan proved the shrinkage held. Good would look like the confirmed rate climbing above 30% in a larger, cleaner group.
• Among 22 patients with measurable brain metastases, 41% had an intracranial partial response. That means tumors in the brain shrank for roughly four of every ten evaluable patients. Good would look like at least 40% confirmed, with responses lasting six months or longer in a much larger sample.
• Four of 68 monotherapy patients had a severe treatment-related side effect, with no life-threatening or fatal treatment-related event reported. That early safety profile looks manageable, but the sample is small. Good would look like severe drug-related events staying below 10% as enrollment grows.

What I like

The machine has produced something more useful than a slide deck. IAM1363 moved from program launch to human testing in roughly two years, while Iambic says the industry often needs much longer. That means the platform may compress early discovery. Good would look like IAM217 and IAM-C1 repeating the speed without sacrificing safety or quality.

Big drug companies have paid to look under the hood. AbbVie (ABBV), Takeda Pharmaceutical (TAK), Bayer (BAYRY), and Revolution Medicines (RVMD) have signed collaborations. Takeda’s agreement carries potential future payments above $1.7 billion, but that giant figure depends on years of success. Good would look like recurring upfront payments and near-term milestones before anyone counts fantasy royalties.

The lead drug attacks a hard problem. Existing HER2 treatments help many patients, but brain metastases and treatment resistance still create openings. IAM1363 enters the brain, covers several HER2 mutations, and aims to spare EGFR. Good would look like those laboratory advantages becoming longer responses in people who already failed approved drugs.

The team fits the assignment. Co-founders Tom Miller and Fred Manby built careers in computational chemistry, while the medical team includes veterans of oncology drug development. Good would look like that scientific depth turning into disciplined trial design, honest disclosure, and one approved medicine. Brilliant chemistry without clinical execution is a Ferrari with no steering wheel.

What worries me

The AI claim remains unproven where it counts. One candidate reaching Phase 1 does not show that the platform improves the odds of approval. Good would look like several internally designed molecules reaching humans, then beating normal industry survival rates across successive trial stages.

The lead data are promising and messy. The trial combines several cancers, HER2 alterations, prior treatments, and doses. That makes a broad signal useful for finding where the drug works, but weak for valuing a future product. Good would look like tumor-specific cohorts with at least 30 to 50 patients each, confirmed responses, and duration data.

Competition is already in the pharmacy. Boehringer Ingelheim (private) won approval for zongertinib in previously treated HER2-mutant lung cancer, while established therapies already own major HER2 markets. Good would look like IAM1363 winning on brain activity, mutation coverage, tolerability, or combinations. “Also works” will not build a category leader.

Revenue can fool a beginner. The 225% increase came from collaborations and milestones, which can jump one quarter and vanish the next. Good would look like several independent partners paying repeatedly. One fat milestone is dessert, not a meal plan.

Dilution is loaded into the cap table. Existing options, restricted shares, warrants, and the new employee plan create a gap between basic and fully diluted ownership. Good would look like management using those awards to retain irreplaceable scientists while keeping annual dilution below 5%.

Public enthusiasm is too thin to trust. Thirty-day research found four Reddit threads with 25 upvotes and two YouTube videos with 74 views, and neither video supplied a transcript. That means there is no useful crowd consensus yet. Good would look like oncologists debating response durability and patients discussing trial access, not traders repeating “Nvidia-backed” like it cures cancer.

What would make this a buy

My position size is zero. This is research and commentary, never a buy signal.

I want the next IAM1363 update to show at least a 30% confirmed response rate in a larger group at the selected dose, at least a 40% confirmed brain-response rate, and a median response lasting at least six months. Those tests would show the drug works often enough, reaches the brain, and keeps working. Good data must do all three.

I want the Food and Drug Administration to agree on a registrational path, meaning a trial plan that can support approval, before I give management credit for a 2027 pivotal study. I also want IAM217 cleared and dosed without derailing the lead program. Good would look like two clinical assets with one clear priority.

Then price gets a vote. At $16, the IPO price, basic market value would be about $758.4 million and rough operating-business value would be about $415.5 million after cash. That is a more rational price for early data and an unproven platform. I would rather pay the offer price after stronger evidence than pay $20 for a clean opening-day chart.

What makes me walk away

I walk if the confirmed response rate falls below 20% in expanded IAM1363 cohorts or if brain activity disappears in a larger sample. That would mean the most valuable clinical claim weakened when more patients entered the room.

I walk if severe treatment-related side effects exceed 15%, if interstitial lung disease becomes a real signal, or if frequent dose reductions prevent patients from staying on therapy. A cancer drug can carry side effects, but it still needs a usable therapeutic window, the space between an effective dose and an intolerable one.

I walk if regulators reject the proposed late-stage path, if the pivotal start slips beyond 2028, or if cash falls below $150 million before decisive data. That combination would put the scientific schedule behind the financing clock.

I walk if the company sells another large block of shares before proving IAM1363, unless a partnership or acquisition adds more value than the dilution takes. The IPO already funded the exam. Shareholders should not pay the tuition twice before seeing the grade.

IPO Watch

• City Therapeutics (CTY), expected to begin Nasdaq trading October 16 after pricing at $18. It develops RNA-interference drugs, medicines designed to silence disease-driving genetic instructions. The company sold 11.1 million shares and raised roughly $200 million before fees, which means demand supported a larger deal at the midpoint of its range. I would write it up if its lead drug shows a clear advantage over existing gene-silencing medicines and the valuation leaves room for clinical failure.
• Web3Labs Global (private), proposed ticker MDAT, expected the week of October 19 at $4 to $5. It sells technology and services tied to digital assets and Web3. The range is what IPO buyers may pay before public trading. I would write it up only if audited revenue, cash flow, customer concentration, and legal exposure survive a hard reading. “Web3” without economics is a nightclub wristband from 2021.
• Bitari (private), proposed ticker BIAI, expected the week of October 19 at $7. It is pursuing a small U.S. listing with roughly $30 million of expected gross proceeds, meaning the float may trade thinly and violently. I would write it up if the final prospectus shows a real operating business, clean ownership, and enough public shares for price discovery rather than a carnival ride.

The verdict

Score: 69 out of 100. Bucket: watch. Iambic earned points for a differentiated lead drug, early brain activity, a fast discovery engine, multiple pharma partners, founder-level scientific depth, roughly $342.9 million of pro forma cash, and several clear clinical catalysts.

It lost points because no drug is approved, product revenue is zero, losses are widening, IAM1363 data mix cancers and tiny subgroups, HER2 competition is fierce, collaboration revenue is lumpy, dilution is substantial, and the first-day pop made the entry worse. Candidate status starts at 80. IAM misses by eleven. The company deserves a deep file. The stock needs more patients and less perfume.

The DBTech read

Best response and confirmed response are not the same score. A best response counts the smallest tumor measurement seen on any scan. A confirmed response requires a later scan to show the shrinkage held, so it carries more weight. IAM1363 posted 36% best partial responses but 23% confirmed responses in its main dose group. When a drug company leads with the bigger number, check whether the smaller confirmed number tells the sturdier truth.

Sources and as-of date

• Price: $18.91 at 4:00 p.m. Eastern on October 15, 2026, the first completed regular-session close, with a $16.15 open, $19.70 high, $16.15 low, and about 2.0 million shares traded, Yahoo Finance and Seeking Alpha. Yahoo showed a thin $20.07 premarket indication at 7:00 a.m. Eastern on October 16 with zero reported volume, so every valuation uses the liquid $18.91 close. Retrieved at 7:07 a.m. Eastern on October 16, 2026. Read-only Robinhood price and option-chain tools were unavailable in this run, so no contract, fill, bid, ask, or Greek is assumed.
• IPO terms, share count, dilution, cash, revenue, losses, operating cash use, clinical data, partnerships, management, risks, and runway: Iambic final prospectus, dated October 14 and filed October 15, 2026.
• Platform, pipeline, and partnership claims: Iambic platform, Iambic pipeline, Takeda collaboration, and IAM217 filing announcement, retrieved October 16, 2026. Company claims require independent clinical proof.
• Trial design and enrollment: ClinicalTrials.gov study NCT06253871, updated September 2026 and retrieved October 16, 2026.
• HER2 competition: Pharmaphorum on approved zongertinib and BioWeek on IAM1363, retrieved October 16, 2026.
• IPO calendar: City Therapeutics pricing announcement and IPOScoop calendar, retrieved October 16, 2026. Dates can move until trading starts.
• Community research covered September 16 through October 16, 2026. It returned four Reddit threads and two YouTube videos. X was unavailable, and YouTube transcripts failed, so this evidence is thin. Raw research: workspace/newsletters/a-portfolio/research/iambic-therapeutics-iam-raw-v3.md.
• Calculations use decimal arithmetic, the $18.91 close, 47,400,796 basic shares, roughly 55.1 million fully diluted shares, $207.9 million of June cash, and $135 million of estimated IPO proceeds after fees. Spending since June makes the cash estimate optimistic. Research and commentary only, never a buy signal.