One molecule, three markets
A monoclonal antibody against an infectious disease can be used before exposure to protect people who cannot rely on a vaccine, after a known exposure to prevent infection taking hold, or as treatment for people who are already sick. Those are three different markets. They have different patients, different ways of finding those patients, different settings of care, different payers and different timing. A market model that adds them into one number before understanding each will be wrong in ways that are hard to see afterward.
The investor, program manager or business development lead asking for a market size usually wants one figure. The useful answer is a small set of figures, one per use case, each with a range and the assumptions that drive it, rolled up only at the end.
The three use cases
- Pre-exposure prophylaxis. Given ahead of a risk period to people at high risk of severe disease: infants in their first season, people with weakened immune systems who respond poorly to vaccines, older adults in care facilities. The population is identifiable in advance, which makes it the most forecastable use.
- Post-exposure prophylaxis. Given after a known exposure to prevent infection or illness: household contacts, outbreak settings in care facilities, laboratory accidents, and in biodefense scenarios, people exposed in a deliberate release. Demand depends entirely on someone knowing the exposure happened.
- Treatment of illness. Given to people who are sick, either early to prevent progression or later for severe disease. Demand depends on diagnosis, timing and eligibility, and is limited by all three.
- Government reserve. Purchased in advance and held against an outbreak or a deliberate event. Not a clinical use case, but often the largest early channel for threats where routine demand is near zero.
Each use case needs its own row in the model. A prophylaxis indication and a treatment indication for the same pathogen can differ in volume by an order of magnitude, and they respond to different evidence.
Pre-exposure prophylaxis: a population you can count
Prophylaxis markets start from a population that can be counted before the season begins. A birth cohort is known from birth records. People with specific immune-compromising conditions can be counted from claims or registries. Residents of long-term care facilities can be counted from facility census data.
The questions that size the market are then practical. How often does a person need a dose to stay protected through the risk period, which depends on how long the antibody lasts in the body? Where is it given, and does that setting have a routine visit at the right time? Will guidance from clinical bodies recommend it, and for whom? And who pays: a public program, a commercial plan, a hospital budget? The precedent of seasonal antibody prophylaxis given to infants against a common respiratory virus has shown that a prophylaxis product can be offered to an entire birth cohort, and it gives analysts a recent template for how quickly a prophylaxis program can scale when guidance and coverage line up.
Post-exposure use depends on knowing about the exposure
Post-exposure demand is bounded by exposure identification. Someone has to know a person was exposed, recognize they are eligible and get them a dose within the window that matters. For household transmission of a common respiratory virus, most exposures are never identified as such. For a laboratory needlestick or a confirmed case in a nursing home, they are.
The sizing approach is to list exposure settings where identification is reliable, count how often each occurs, estimate how many people are exposed per event, and apply a realistic share who receive the product within the window. That share is usually lower than the team expects, because it depends on the product being on hand at the facility when the event happens. A post-exposure product that has to be ordered after the exposure is reported often arrives too late.
Why diagnosis timing limits treatment markets
Treatment markets look large at the top of the funnel and shrink sharply at every step below it. The reason is timing. Many therapies against viral infections are expected to work best early in illness, and clinical guidance and product labels often reflect that by limiting use to a window after symptoms begin. The patient has to notice symptoms, decide to seek care, get to a clinician, be tested, get a result, be judged eligible and receive the product, all inside that window.
Walk the funnel explicitly. Start with symptomatic infections in the target group, then apply each step as its own assumption.
From symptomatic cases to treated patients
Start with 100,000 symptomatic infections in a high-risk group over a season. Suppose 40 percent seek care, half of those are tested, 60 percent of tests return a result inside the treatment window, 70 percent of positives meet the eligibility criteria, and clinicians prescribe to half of the eligible. The treated population is 4,200, a little over 4 percent of the starting figure. Every percentage here is illustrative. The point is the shape: five reasonable-looking steps remove more than 95 percent of the top line.
Treatment of severe disease in hospitalized patients avoids the care-seeking problem, because the patient is already identified. But it moves the patient later in the course of illness, and whether a product has a role there is a clinical question the trials answer. The market model should carry the early-treatment and severe-disease cases separately and show what each depends on, rather than assuming either.
Two levers change the funnel and belong in scenarios. Faster diagnosis, such as rapid tests at the point of care or at home, moves more patients inside the window. Prescriber familiarity moves the last step, and it is lowest in the first seasons after launch. Neither lever is free, and neither should be assumed at full strength from year one.
Site of care changes who can use it
An antibody given by intravenous infusion needs an infusion chair, a nurse, an observation period and a site that can handle the scheduling. An antibody given by injection into a muscle or under the skin can be given in a clinician's office, a pharmacy where rules allow, or a long-term care facility. That difference can matter as much as anything in the clinical profile.
Site of care also changes payment. In the hospital inpatient setting, Medicare generally pays a fixed amount per stay, so the drug's cost sits inside a bundle the hospital already receives unless a separate add-on payment applies. In the outpatient setting, many physician-administered drugs are bought by the provider and reimbursed under Medicare Part B at a rate tied to the average sales price plus a percentage. In the pharmacy, the drug benefit applies. A model that ignores site of care will get the price, the volume and the speed of uptake wrong at once.
Pricing and reimbursement analogs
Before launch there is no price, so the model borrows from analogs. Choose them by use case and setting, not by molecule type. A prophylaxis antibody given once a season to infants competes, in a payer's mind, with vaccines and other preventive products for that population. A hospital treatment for severe disease is judged against other inpatient therapies and the cost of an extra day in intensive care. An outpatient early treatment is judged against oral antivirals for the same indication, if any exist.
Do not use antibody prices from oncology or immunology as analogs. They reflect chronic dosing, different payer dynamics and different value arguments. Use two or three analogs per use case to set a range, and pair the price assumption with the recommendation or coverage assumption it implies. A high price that narrows coverage to the highest-risk patients may produce less revenue than a moderate price with broad coverage, and the model should be able to show that.
Stockpile and government purchase channels
For pathogens with outbreak or biodefense relevance, government purchase can be the largest channel, especially early. In the United States, the Biomedical Advanced Research and Development Authority funds advanced development of medical countermeasures, Project BioShield supports procurement of countermeasures against chemical, biological, radiological and nuclear threats, and the Strategic National Stockpile holds supplies for public health emergencies. Other governments run their own reserves.
Model this channel as contracts, not as demand. Government purchases come in large, irregular orders with option years, and they depend on appropriations, threat assessments and competing priorities. Two features are worth modeling explicitly. Product held in a reserve expires, so a stockpile that is maintained creates replacement orders on a cycle tied to shelf life. And government contracts often include development funding alongside procurement, which matters to an investor even though it is not product revenue.
Seasonality, outbreaks and the risk of escape
Seasonal respiratory pathogens give a recurring market with a predictable rhythm and an unpredictable size: some seasons are mild, some severe, and the timing of the peak shifts. Outbreak pathogens give a market that is near zero most years and very large in a few. Averaging those years produces a number that never occurs.
There is also a risk specific to antibodies against fast-changing viruses: the pathogen can change so that the antibody no longer binds well. During the COVID-19 pandemic, several authorized antibody products stopped being used when new variants made them unlikely to work. A market model for an antibody against a variable pathogen should carry a scenario in which the product loses relevance partway through the horizon, and should ask whether the developer has a plan for updated versions.
Top-down versus bottom-up sizing
Two approaches are standard, and a credible estimate uses both and explains the gap between them.
| Top-down | Bottom-up | |
|---|---|---|
| Starting point | Total disease burden or total spending in a category | Countable patients in a specific use case and setting |
| Method | Apply shares: addressable, treated, captured | Walk the funnel step by step: identified, eligible, reached, treated, paid |
| Strength | Fast, easy to explain, good for a first sense of scale | Every assumption is visible and testable against data |
| Weakness | The shares are guesses that hide the funnel, so it usually overstates | Slow, needs data for each step, can miss segments no one listed |
| Best use | Early screening of whether an opportunity is worth a closer look | Investment, pricing, capacity and program decisions |
| Typical error | Treating burden as demand | Optimistic step rates in the first seasons after launch |
When the two disagree by a large factor, the gap is information. It usually means the top-down shares assumed a funnel that does not exist, or the bottom-up model left out a segment. Reconcile them explicitly and say which one the decision should rely on.
The sizing build
Market Sizing Build, Illustrative Six to Eight Weeks
Framing scenarios for the decision
Build three to five scenarios that are internally consistent: a narrow scenario with prophylaxis only for the highest-risk group and slow uptake, a base scenario with the most likely guidance and coverage, a broad scenario with wider guidance and faster diagnosis, and, where relevant, an outbreak scenario and an escape scenario. Each scenario should state its assumptions in words before it states numbers.
Then test each against the threshold that matters. An investor wants to know whether the narrow scenario still returns capital. A program office funding development wants to know whether the product has a sustainable commercial market after the government purchase, or whether it will depend on reserve orders indefinitely. A manufacturer wants to know the peak volume it might need to supply and how fast. The same model serves all three if it is built per use case and reports the scenarios against the decision rather than a single headline number.
Frequently asked questions
Because each patient has to pass a series of steps inside a limited time window: notice symptoms, seek care, get tested, receive a result, qualify and be prescribed. Each step loses people. A bottom-up funnel makes those losses visible, while a top-down estimate based on total cases usually hides them.
Yes, but as a separate channel modeled from contracts rather than clinical demand. Stockpile orders are large, irregular and dependent on appropriations and threat priorities. Include replacement purchases tied to shelf life if the reserve is maintained, and keep development funding separate from product revenue.
Use both. Top-down is a fast screen for scale. Bottom-up is what supports an investment or program decision, because every assumption can be checked. A large gap between them usually means the top-down shares assumed a patient pathway that does not exist, or the bottom-up model missed a segment.
It determines where the product can be given, which determines who can reach it and how it is paid for. An infusion needs a chair, staff and observation time, which limits capacity. An injection can be given in an office, a pharmacy where allowed, or a care facility, which widens access and changes the payer channel.