A researcher reaches out and sends us a paper.
The science looks interesting. The target is new. The experiments have been repeated, the results are consistent, and the biology makes sense. Then comes the question: what happens next?
This is where things usually get complicated.
A good paper is a starting point. Turning the finding in that paper into a drug is a very different exercise, and there’s a lot to answer before anyone should start spending serious time and money on chemistry. Is the target really doing what we think it’s doing? Can it be affected with a drug without causing problems elsewhere?
None of these questions are particularly mysterious. But they do require experience, the right experiments and, often, a willingness to stop and rethink the program when the data doesn’t support the original idea.
And this is where promising academic programs can sometimes get stuck.
The biology may be right. The drug may still fail.
A strong genetic result is encouraging. If removing a gene produces the expected effect, or if a genetic association is seen consistently in patients, there’s good reason to investigate the target further.
But genetics and pharmacology are not the same thing.
A genetic knockout can remove a protein completely and, in some cases, from very early in development. Everything downstream then grows up in its absence. A drug works differently. It might reduce perhaps seventy percent of the protein’s activity, for a few hours a day, reversibly, in an adult patient whose compensatory pathways are fully intact. The biology can respond quite differently to those two situations. Programs sometimes find this out after two years of optimizing against a phenotype that was never going to appear under pharmacological conditions in the first place.
That’s why one of the first questions we ask isn’t simply, “Is this target interesting?” It’s: “Can we show that changing this target with a drug produces the biology we want?”
That distinction can save a lot of time. Before moving too far into chemistry, it’s useful to have a tool compound, a model that resembles the disease, and some early sense of the therapeutic window.
A good assay is not necessarily a drug-discovery assay
This is another place where the transition from academia to drug discovery can be underestimated. An assay can be perfectly good for answering a research question and still not be suitable for screening thousands of compounds.
In discovery you need an assay that’s reliable day after day, plate after plate, with enough separation between signal and noise to trust a single well. You also need to know the compounds are acting on the target rather than interfering with the experiment. Molecules that aggregate, cycle redox, or carry reactive groups that modify whatever is nearest will give you clean dose-response curves and mean nothing at all.
So controls and counter-screens have to be designed in from the beginning. Otherwise it’s easy to spend six months on medicinal chemistry before discovering that most of the actives were detergent-sensitive.
Finding a hit is not the same as finding a useful starting point for a drug.
Then the chemistry starts
Suppose the biology is convincing and the screening works. Now you need molecules, and this is where things often take longer than expected.
An academic screen might give you one or two interesting compounds. That’s exciting, but a single active molecule doesn’t tell you whether there’s a chemical series behind it that can actually be developed. Can it be made more potent, and selective against the rest of the protein family? And has somebody already filed on that chemotype? Freedom to operate matters more than people expect. However good the pharmacology, a series with no clear IP position has no commercial path, and it’s much better to know that early than after two years of chemistry.
These questions lead into lead optimization, which is the part of drug discovery that nobody outside it ever really sees.
You make a molecule. You test it. You learn something. You make another one. Then you do it again, and again. A compound can look excellent in one experiment and disappoint in the next, because potency in an enzymatic assay tells you nothing about metabolism, permeability, solubility, plasma protein binding, hERG liability or CYP inhibition. A two-nanomolar inhibitor with no oral exposure isn’t a lead. Neither is a compound with good plasma levels that never reaches the tissue where the target sits.
So medicinal chemistry is really about the whole package rather than potency alone. You’re looking for a molecule with all the properties a drug needs, and trying to hold on to the activity and selectivity while you fix everything else.
Knowing when to stop
Not every program should continue. That isn’t necessarily bad news.
Sometimes the chemistry is the problem and the biology still looks strong. In that case a different molecule or chemical series may be worth pursuing. Sometimes the data starts telling you that the target itself may not be doing what you originally thought.
Those are very different situations, and the difficult part is knowing which one you’re looking at. A target can get written off because the compound tested against it had a five-minute half-life and never achieved meaningful exposure. That tells you something about the molecule. It doesn’t tell you the hypothesis was wrong.
Stopping the wrong program early can save years.
Stopping the right program early can be just as costly.
This is one reason experience matters so much in drug discovery. Teams that have been through the process before are often better equipped to recognize when a program needs another round of work and when it’s time to move on.
Declaring a candidate is a commitment
At some point, if the molecules hold up, a program reaches candidate selection. This deserves more attention than it usually gets, because it’s the point where the cost structure changes completely.
Nominating a development candidate commits the program to GLP toxicology, safety pharmacology, genotoxicity, a synthetic route that works at scale, formulation development and a projected human dose. That work is expensive and slow, and once you’ve started it you can’t really unwind it.
Call it too early, on data that isn’t there yet, and you find out in the clinic. Put it off indefinitely and the program just runs out of air. Either way the package has to be good enough to defend to a regulator, a partner and an investor.
And then there is the researcher
There’s something else that can easily get lost when an academic discovery moves into drug development.
The person who generated the original research often knows the biology better than anyone else. They know why a particular model was chosen. They know which results were surprising. They know which experiments were difficult to reproduce. They know the limitations of the system, including things that may never have made it into the final paper.
That knowledge can be extremely useful when the program starts making development decisions.
This is one of the reasons we believe in scientific continuity at Anveta Therapeutics. When we work with an academic or research partner, we don’t see the relationship as ending when the research moves out of the laboratory. The goal is to keep the originating researcher involved as the program develops, while bringing in the drug discovery, medicinal chemistry, pharmacology and development expertise needed to move the work forward.
The reason is simple. When difficult decisions come up, context matters.
A molecule may fail. An assay may need to change. A disease model may need to be reconsidered. A promising result may turn out to be less meaningful than it first appeared. Having the scientist who understands the original biology involved in those discussions can make a real difference.
So, what happens after the paper?
Ideally the paper isn’t the end of the story. It’s the beginning of a process.
First you need to understand the biology well enough to know whether there’s a real therapeutic hypothesis. Then you need the right experiments to test whether that hypothesis can be approached pharmacologically. Then come chemistry, optimization, pharmacology, safety and eventually the much larger decision of whether a program is ready to enter clinical development.
There will be failures along the way. There should be.
The important thing is learning from them quickly enough to know whether you’re looking at a problem with the molecule, the chemistry, the model or the underlying hypothesis. That’s what turns a research finding into a drug-development program.
What we’re looking for
At Anveta we’re interested in working with researchers who have strong science and a question worth pursuing. That could be a novel target, a compelling disease mechanism, a reproducible phenotype or an early small-molecule program, with particular interest in oncology, including immuno-oncology, and in longevity and age-related disease.
It doesn’t have to be a finished drug candidate.
If you have published research and believe there may be a path from the biology to a differentiated therapeutic, we’d be interested in having the conversation. Sometimes the most useful question is simply: what should we do next?
And it’s usually better to ask that question early.
Anveta Therapeutics is a drug discovery and early clinical development company translating academic science into differentiated small-molecule therapeutics. More on our model.