Paracelsus taught us that “the dose makes the poison.” A new analysis of respiratory-virus infection suggests that biology requires an important qualifier: where an exposure enters the body and which tissues encounter it first can matter even more than how much arrives. Dose describes quantity; route helps determine what that quantity actually does.
While COVID’s arrival and the pandemic are happily in our rearview mirror, cultural and scientific questions surrounding COVID continue to reverberate. A recent study of the virus's spread in animal models, reported in Scientific Advances, calls into question Paracelsus's oft-repeated words that the dose makes the poison. As it turns out, dose is only part of the story; where a respiratory virus lands — in our nose, airway, lungs, or gut — can matter more than how much virus arrives.
A Better Experiment—With Important Caveats
The study reviews 107 non-human primate challenge studies in which researchers can control dose and route. Although the individual studies are small, the aggregate includes a little over 22,000 observations. It examines how route, dose, age, sex, and species affect whether infection occurs, where it occurs, when it peaks, and how long it persists.
Before we jump in, let’s address some limitations. While the model closely reproduced the observed data, it reflects our “best guess” of a biological system. The study’s strength in combining multiple animal studies is also its main limitation. The data do not form a tidy grid that an ideal experiment would provide. The available experiments were unevenly distributed across the route-dose space. Low-dose aerosol exposures are especially underrepresented.
The authors tried to compensate statistically. Their model was adjusted to generate predictions across combinations of route, dose, age, sex, and species. It cannot replace a “cleaner” prospective experimental study, nor can it reproduce natural human infection. The analysis does not eliminate uncertainty so much as organize it, leaving the decisive experiments for the future.
Where the Virus Lands Changes What Happens
Researchers exposed three monkey species to a broad range of doses via five major routes of transmission: intranasal, intratracheal, a combination of both, aerosol, and intragastric. Their model estimated several salient pandemic outcomes, including the probability of infection, viral loads, and time to clearance. [1]
- Dose mattered, but route mattered more. Paracelsus was right: dose does matter, but the route of exposure was the dominant influence in 35 of 37 tissue-level measures. Dose ranked second overall.
- Where the virus entered determined where infection began. The virus appeared earlier and more often in tissues directly exposed by the inoculation route. For example, intranasal exposure favored the nose.
- Spread followed anatomy. Infection tended to move upward and downward between adjacent tissues in the respiratory tract and from the throat toward the GI tract, consistent with mechanisms such as aspiration, swallowing, and the mucociliary escalator. The chance of spread generally declined with anatomical distance.
- Upper-airway exposure changed the subsequent course of infection. When the URT was exposed, nasal infection tended to reach higher peaks and last longer, while lung infection often peaked earlier and cleared sooner. GI infection tended to peak later and persist longer.
- Aerosol exposure was biologically distinct from liquid inoculation. It produced nearly synchronous infection in the nose and lungs and a different pattern of persistence and peak viral load. Importantly, the commonly used combined intranasal/intratracheal model did not reproduce aerosol infection well.
- There was no single ID50. The dose required to infect 50% of subjects varied by exposure route, the tissue sampled, and whether infection was defined by PCR or by recoverable viable virus. PCR often yielded much lower ID50 estimates because it could detect viral RNA even when productive infection had not been established.
- Dose still mattered, but its effect depended on route. Larger doses increased the probability of infection across routes, but their influence on timing, peak viral load, and duration was much greater with upper respiratory tract exposure. For intratracheal and intragastric exposure, higher doses produced relatively few additional effects beyond those seen in directly exposed or adjacent tissues.
Beyond Paracelsus: Biology Has a Geography
The authors' principal argument is that infectious-disease research has probably overemphasized dose while underappreciating route. Their analysis shows that route explains far more of the variation in COVID infection kinetics than dose, particularly with respect to whether the upper respiratory tract is exposed. They place special emphasis on aerosol exposure. Aerosol particles distribute the virus through the respiratory tract differently from liquid inoculation, resulting in a distinctive combination of earlier lung infection, earlier pulmonary clearance, and greater nasal shedding. The paper also challenges the idea of ID50 as a single characteristic of a pathogen. Infection depends on the route, the tissue examined, and whether the assay measures RNA or a replicating virus.
Paracelsus gives us essentially a quantitative model: how much of an agent reaches the organism determines what happens. This paper adds a spatial component: where that agent enters, where it is initially deposited, and which tissues it encounters first can determine what happens next. Dose mattered, but its effect was often conditional on the route.
Biology is both spatial and quantitative. Medicine and toxicology have traditionally privileged magnitude—dose, concentration, viral load—while biological systems also care enormously about geometry, location, sequence, and connectivity. The aerosol findings make the point concrete. Aerosol exposure did not simply function as another way to deliver the same amount of virus. It produced a distinctly different spatiotemporal pattern that the conventional combined intranasal/intratracheal model could not reproduce.
The Principle Goes Beyond Viruses
Silica makes the spatial argument especially clear. Crystalline silica is chemically the same whether swallowed or inhaled, but the biological consequences differ greatly. Ingested particles largely pass through the gastrointestinal tract. Fine respirable particles that reach the distal lung trigger persistent inflammation, cell injury, and fibrosis.
Fluoride applied to the surface of a tooth strengthens enamel and helps prevent dental caries. Fluoride that is swallowed becomes a systemic exposure; during tooth development, excessive chronic intake can produce dental fluorosis, and at much higher sustained exposures it can affect bone. Again, the same chemical and the same nominal quantity can have very different biological effects depending on where it acts.
The paper points to a broader principle: the same quantity presented to a biological system at a different place can be a different biological event. Dose measures quantity, not biological context. Where an exposure lands can determine what that dose means
[1] The complete list included estimating probability of infection, time to detection, time and magnitude of peak viral load, time to clearance, duration, and overall viral burden.
Source: Exposure route, not dose, is the primary driver of infection patterns for a respiratory virus Science Advances DOI: 10.1126/sciadv.aei4853
Category
Chuck Dinerstein, MD, MBA
Director of Medicine
Dr. Charles Dinerstein, M.D., MBA, FACS is Director of Medicine at the American Council on Science and Health. He has over 25 years of experience as a vascular surgeon.
