Antibody therapies are medicines built to bind to selected biological targets. Some block signals that drive disease; others help immune cells recognise and attack abnormal cells. Antibody–drug conjugates, usually shortened to ADCs, add a different function: an antibody carries a linked drug payload to target cells. For investors, the distinction changes what needs to be proved. An effective antibody must produce a useful biological effect at a tolerable dose. An ADC must also deliver and release its payload in a way that creates a workable balance between tumour killing and damage to healthy tissue. A convincing target is the starting point, not a complete investment case.
The investment question is whether the medicine’s design produces a clinical advantage that a company can develop, manufacture and sell on attractive terms. That requires connecting the science with patient selection, trial evidence and ownership rights. Within biotech science and drug technologies, antibodies deserve their own assessment framework because apparently similar products can differ substantially in how they work and where their risks sit.
How antibody therapies work
A monoclonal antibody is a laboratory produced protein designed to recognise a chosen target, often called an antigen. Binding can have different consequences. An antibody may interrupt a disease related signal or mark a cell for destruction by the immune system. Rituximab, for example, binds to CD20 on B cells, including some cancerous B cells. Engineered antibodies can also connect immune cells with cancer cells. Blinatumomab binds to CD19 on certain leukaemia cells and CD3 on T cells, bringing them into contact. These mechanisms require different evidence, even though all sit under the antibody treatment umbrella. The National Cancer Institute’s monoclonal antibody resource documents these distinctions.
Antibodies also treat diseases outside oncology. Adalimumab, sold under the Humira brand and through biosimilar products, binds to tumour necrosis factor, or TNF, and blocks its activity to reduce inflammation. Its mechanism illustrates why an antibody investment thesis need not involve killing cells. The desired effect can instead be sustained control of a disease pathway, as documented in the EMA assessment of Humira’s mechanism and uses. An investor assessing that type of medicine should frame the commercial questions around lasting symptom control, treatment persistence and the practicality of repeated administration, rather than importing an oncology response rate framework.
| Format | Basic function | Question for investors |
|---|---|---|
| Unconjugated antibody | Binds a target without carrying an attached drug payload. | Does target binding produce enough clinical benefit at an acceptable dose? |
| Bispecific antibody | Has two binding specificities, potentially connecting different cells or acting on different targets. | Does the combined action improve treatment without creating an impractical safety burden? |
| Antibody–drug conjugate | Links an antibody to a drug payload through a chemical linker. | Does the complete construct deliver a better balance of efficacy and toxicity? |
These categories can overlap. “Monoclonal” does not mean unconjugated, and “bispecific” does not automatically mean a T cell engager. Establish the molecule’s actual function before comparing it with another company’s programme. A useful research note should identify what the antibody binds, what happens after binding, and which observations would show that the proposed mechanism works in patients. The broader differences between small molecules and biologics provide context, but neither format deserves a valuation premium simply because it sounds more sophisticated. Complexity must earn its place through evidence.
What determines whether an ADC design works?
For the oncology ADCs considered here, the main components are the antibody, a chemical linker and a cytotoxic payload. The antibody provides target recognition; the payload supplies cell killing activity. The linker connects them and helps govern when the active drug becomes available. Many ADCs enter a cell after binding, then release their payload through intracellular processing. Treat the complete construct as the product. A strong antibody paired with unsuitable release chemistry can disappoint, while a powerful payload has little investment value if patients cannot tolerate the exposure needed for useful activity.
Target expression and delivery
An ADC target needs to be accessible to the antibody. Investors should examine how much target is present, how consistently it appears across tumour cells, and whether healthy tissues also express it. Ask how those measurements were made. A target positive cell line is a narrower piece of evidence than a reproducible testing method applied to patient tissue. For an ADC that depends on cellular uptake, binding alone also leaves an unanswered question: does enough active payload reach the relevant cells? The framework for evaluating drug targets and mechanisms of action helps separate biological plausibility from evidence that the chosen delivery strategy can work.
Target expression should also be treated as a measurement problem. When reviewing a programme, check the test used, the threshold for eligibility and whether samples came from recent or archived biopsies. Then ask whether the commercial forecast assumes the same definition. A company can generate encouraging results in a carefully selected group while presenting a market estimate based on a much broader population. That gap deserves attention before the revenue model becomes elaborate. The distinction matters, a tumour marker used to select trial participants is not automatically a dependable estimate of the number of patients who will receive treatment.
Linker and payload choices
The linker needs to support useful delivery and release. “More stable” is not a complete product claim: investors need evidence about stability during circulation and release under the intended conditions. Payload choice raises a related set of questions. What cellular process does the drug disrupt? How much exposure is needed for activity? Does the clinical programme show that patients can receive enough treatment for long enough? Avoid assessing linker chemistry and payload potency in isolation. A presentation can make each component look attractive separately while leaving the performance of the assembled ADC uncertain. Human exposure, efficacy and tolerability data should eventually replace design claims as the centre of the investment case.
Drug to antibody ratio
The drug to antibody ratio, or DAR, describes payload loading. It is often reported as an average, which can conceal a distribution of molecules carrying different amounts of drug. A higher number does not establish a superior product. In a preclinical study of an anti-CD30 ADC, constructs carrying four payload molecules produced antitumour activity comparable with constructs carrying eight at equal antibody doses; the more heavily loaded construct also cleared faster. The original study of ADC drug loading demonstrates the trade-off in that experimental system. It does not establish an ideal DAR for every antibody, linker or payload.
Bystander activity
Some released payloads can affect neighbouring cells, including cells with little or none of the antibody’s target. This is called bystander activity. Preclinical research on DS-8201a, subsequently known as trastuzumab deruxtecan, demonstrated activity against neighbouring HER2 negative cells in mixed tumour models. The original study of DS-8201a bystander killing provides experimental support for that construct’s behaviour. Investors should resist turning the finding into a blanket claim about ADCs. Ask whether the company has demonstrated bystander activity for its own molecule, and keep evidence from laboratory and animal studies separate from proof of clinical benefit.
Clinical evidence matters more than format labels
Early ADC announcements often emphasise tumour shrinkage. Assess the denominator before the percentage. How many patients received treatment, how many were included in the response analysis, and how many had enough follow-up for their responses to be confirmed? Check whether the selected dose has adequate evidence behind it, rather than pooling every dose level into one attractive result. Patient history matters too. A cohort with fewer previous treatments may be a poor comparison for a heavily pretreated population. Use a consistent method for reading clinical trial results, then add the questions raised by the antibody’s target and the ADC’s delivery system.
DESTINY-Breast04 provides a concrete example of evidence beyond an early response signal. The initial 2022 report enrolled 557 patients with previously treated HER2-low metastatic breast cancer. Among all randomised patients, median progression-free survival was 9.9 months with trastuzumab deruxtecan versus 5.1 months with the physician’s choice of chemotherapy. Median overall survival was 23.4 months versus 16.8 months. The primary endpoint concerned progression-free survival in the hormone receptor positive cohort, rather than the entire population. Safety remained material: adjudicated drug related interstitial lung disease or pneumonitis occurred in 12.1% of treated patients, including fatal events in 0.8%. These are results from the original DESTINY-Breast04 trial publication, not universal efficacy or safety rates for ADC treatment.
The investment lesson is to assess the whole result. A randomised comparison can provide evidence that a medicine improves outcomes against the treatments used in that trial. It does not establish the outcome for a different tumour type, an earlier treatment setting or a changed comparator. Response rate, progression-free survival and overall survival answer different questions; the distinction between clinical endpoints belongs directly in the investment model. Give more weight to an outcome that answers the intended commercial question than to an impressive percentage that leaves it unresolved. Also record the analysis date. An initial publication should not be mistaken for the most mature follow-up available.
Dose selection and safety can change the investment case
For ADCs, dose selection involves both the intact conjugate and its constituent parts. Their exposure can affect efficacy and safety differently. The FDA’s March 2024 guidance encourages evaluation of multiple dose levels or regimens early in development, together with analysis of the relationships between exposure, activity and adverse effects. It also addresses measurement of the ADC and relevant components. These are recommendations in the FDA clinical pharmacology guidance for ADC development, not evidence that any particular programme has found an appropriate dose. Investors should look for the supporting data rather than treating a selected development dose as a settled question.
A useful safety review goes beyond the headline rate of severe adverse events. Examine treatment related deaths, permanent discontinuations and dose reductions. Look at when problems occur and how long patients have been exposed. Ask whether the proposed management plan requires frequent testing, specialist assessment or repeated interruptions. Then connect those demands to the intended treatment setting. An acceptable trade-off in patients with few remaining options may face a different commercial test where effective alternatives exist. The assessment of adverse events and treatment discontinuations should therefore sit beside efficacy analysis, rather than appearing as a footnote after the revenue forecast.
Be cautious when a company describes toxicity as manageable without showing what management required. Establish whether patients resumed treatment, remained at a reduced dose or stopped permanently. Those outcomes have different implications for treatment duration and adoption. Compare the safety population with the proposed customer population as well. If trial eligibility excluded patients whose medical conditions would complicate treatment, ask how much confidence the available data provide about broader use. This is a question about the evidence’s reach. It should not become an unsupported assumption that every excluded patient is either safely treatable or permanently outside the market.
Approval scope is narrower than a target’s name
On April 5, 2024, the FDA granted accelerated approval to fam-trastuzumab deruxtecan-nxki for adults with unresectable or metastatic HER2 positive, IHC 3+, solid tumours after prior systemic treatment and with no satisfactory alternative treatment options. That historical decision illustrates how broad tumour coverage can still carry tightly defined eligibility conditions. The approval relied on objective response rate and duration of response; continued approval could depend on confirmatory evidence of clinical benefit. The FDA’s April 2024 tumour agnostic approval notice records the population and evidence supporting that decision.
For investment research, separate the approved population from populations still being studied. A tumour type, biomarker threshold or treatment setting mentioned in a pipeline presentation should not enter the base revenue forecast on the same terms as an established indication. Programmes using accelerated or conditional approval pathways also need a clear account of remaining evidence obligations. Check the current approval notice and prescribing information when updating a model. A historical regulatory milestone is useful evidence, but it cannot establish the product’s complete label years later or resolve the risk of every planned expansion.
Manufacturing quality belongs in the scientific assessment
ADC manufacturing must control more than the antibody alone. The antibody intermediate, linker and payload need appropriate quality controls before conjugation, followed by evidence that the assembled product has consistent properties. Drug loading, purity and biological activity all deserve attention. The WHO guidelines on monoclonal antibody production and quality control address these requirements and the importance of consistent conjugation. This makes manufacturing evidence relevant to the product thesis itself. A process that produces material reliably at one scale still needs to support the quantities and consistency expected at later development stages.
Investors do not need to reproduce a manufacturing dossier. They do need to identify unresolved dependencies. Who supplies the antibody and payload? Which company performs conjugation? Has management explained whether the proposed commercial process differs from the one used to generate clinical evidence? What work remains before launch supply is ready? These questions belong in the milestone schedule and funding plan. Keep manufacturing scale-up, capacity and production costs separate from clinical success assumptions, so a positive trial does not automatically erase an unfinished production programme. Outsourcing can transfer tasks to a contractor; the research note should still show where delays or quality problems would affect the sponsor.
Commercial potential depends on the treatment pathway
A target appearing across several cancers can support a broad research programme without supporting one undifferentiated market forecast. Build the commercial assessment around each proposed indication. Identify the relevant disease stage, biomarker threshold and previous treatment requirements, then estimate how many patients could reach treatment. Include testing as a practical step rather than assuming that every potentially eligible patient has already been identified. The method for estimating a drug’s addressable market is particularly useful here: disease prevalence is a starting population, while treated patients are the output of several clinical and commercial filters.
Resistance also deserves a place in the assessment. Cancer cells can become resistant to targeted treatment when the target changes or when growth no longer depends on the blocked pathway. Targeted medicines can also cause serious adverse effects despite their intended selectivity. These constraints are documented in the National Cancer Institute resource on targeted therapy. For an ADC developer, ask what its evidence establishes after earlier treatments and what remains hypothetical. Do not assume that a different antibody target automatically solves every resistance problem, or that changing the delivery system removes the need to demonstrate a tolerable clinical regimen.
Competition should be framed around the decision a physician would face. The relevant alternative may be another ADC, an unconjugated antibody or a treatment from a different drug class. Compare outcomes in similar populations and identify where direct comparisons are missing. Convenience and monitoring requirements also belong in the assessment, even when the efficacy story looks favourable. A structured review of standards of care and competing pipelines helps prevent a familiar error: comparing a development programme with yesterday’s treatment options while forecasting sales into a future market. Record the date of the competitive assessment and the evidence that could change it.
Ownership rights determine what reaches shareholders
Product success and shareholder economics are separate questions. An antibody company may own the complete medicine, receive royalties, share profits or hold rights only in certain territories. A platform provider may supply part of the technology without owning the resulting product. Map those rights before assigning value to sales estimates. The research note should distinguish revenue that the company records from revenue earned by a partner, then identify costs and payments attached to those rights. A medicine with a large potential market can still contribute a much smaller amount to the company being valued, depending on the contract.
The original 2019 trastuzumab deruxtecan collaboration illustrates why deal structure matters. AstraZeneca agreed to a $1.35 billion upfront payment to Daiichi Sankyo, with up to $5.55 billion in contingent payments. The announced agreement specified equal sharing of development and commercialisation costs and profits outside Japan, with Daiichi Sankyo responsible for manufacturing and supply. Those are historical terms from the original AstraZeneca and Daiichi Sankyo collaboration announcement. They illustrate the need to distinguish committed consideration from payments dependent on future events. The maximum announced amount was not equivalent to cash received at signing.
Apply the same discipline when valuing licensing deals and royalties elsewhere in the sector. Check territories, development contributions, milestone conditions and termination provisions. For companies presenting several ADC programmes as a platform, map their shared dependencies too. Different targets do not establish independent investment risks if programmes reuse the same linker, payload or production approach. Conversely, success with one construct does not prove every future construct. The distinction between platform businesses and single-asset companies should rest on transferable evidence and retained economics, rather than the number of molecules shown on a pipeline slide.
Build the investment case around the next unresolved question
A practical research file should make clear what the next result can settle. A dose expansion might improve confidence about activity and tolerability without resolving comparative efficacy. A randomised trial might establish clinical benefit while leaving launch supply unfinished. A partnership might fund development while reducing the sponsor’s share of future economics. Identify the remaining question, the evidence expected and the consequences of an unfavourable result. Then align the financing assessment with that schedule. Cash burn and runway matter because the company needs enough capital to reach a decision point, analyse the evidence and finance the work that follows.
Make the valuation respond to evidence rather than enthusiasm for the format. In a hypothetical ADC model, an effective but poorly tolerated regimen might justify shorter treatment duration and slower adoption. A narrower biomarker population might reduce eligible patients. An additional study could delay launch and increase development spending. Model these changes separately before combining them in a downside case. A risk-adjusted pipeline valuation becomes more useful when its assumptions map to observable questions. For each major assumption, write down the result that would cause it to change. That turns the next clinical or regulatory announcement into a test of the investment case, rather than an invitation to rewrite it after the event.