Stephanie Dobersch

Persistence, mostly, about finishing things. Discovery is the fun part and I would not have stayed in this field without it, but a lot of good science stops at the publication, because that is where the reward is. The question I keep asking is what would have to be true for this to reach a patient, and then doing the work that answer implies.

And I let myself have ideas that sound unreasonable, and then test them. Most do not survive contact with an experiment, which is fine, that is what an experiment is for. But the ones that do are the only place real innovation has ever come from, and you cannot get to them by only asking questions you already suspect the answer to.

Choose the people, not the project. It came in pieces rather than as one line, from several people over years, and I did not appreciate it until I had seen both versions: a brilliant project in an environment that taught me nothing, and an ordinary starting question with people who changed how I think.

Projects fail or shift or get scooped. What you take with you is how you were taught to reason, what standard you saw held, and who is still willing to pick up the phone a decade later. That compounds and the project does not. Later, once you know what you are actually working toward, the question inverts and you choose the people who can help you get there. But early on you cannot know that yet, and the environment is what teaches you.

Learn to do the thing yourself before you learn to delegate it. Early on there is pressure to specialize narrowly and hand off everything else, and it feels efficient. But the people who can generate data and analyze it, or who understand both the biology and what a clinical laboratory needs, are the ones who end up leading, because they can see the whole problem. Breadth is not indecision. It is leverage later.

And find the rooms where decisions actually get made, then get into them. Committees, editorial work, grant panels, technology transfer, whatever the version is in your setting. The science alone does not put you there and nobody hands you an invitation. The first time you walk into one of those rooms you will feel underqualified, and that feeling is not information. Go anyway, and stay long enough to be useful.

The biggest challenge in pancreatic cancer is that treatment has barely changed in two decades while our understanding of the biology has changed enormously. We know these tumors are heterogeneous, that they switch identity under therapy, and that basal and classical tumors respond very differently to the same drug. Almost none of that reaches the clinic. There is no assay in routine use that tells an oncologist which subtype they are treating, and there are no drugs designed against a specific subtype. Those two gaps are the same gap, because a test only changes care if there is a different decision to make, and a subtype-specific drug only works if you can identify the patients who should receive it.

That is also where the opportunity sits. Single-cell and spatial methods have shown us what the tumor actually looks like, but they are research tools, too expensive and too slow for routine pathology. Their value is in telling us what to measure. The work I find most interesting is what comes after, distilling those findings down to a handful of markers that can be read on a stained slide in any hospital laboratory, and then designing the trials that test whether a subtype-directed therapy beats the standard of care.

Knowing what you are actually for, and then protecting it. There are endless things you could work on and most of them are interesting. Being clear about your purpose is what lets you say no to the rest, and that clarity matters more than working harder. Mine is getting the right treatment to the right patient, and it decides what I take on and what I let go.

Underneath that is caring about the work itself. I do not think you sustain twelve years of experiments that mostly fail on discipline alone. It has to be something you would still want to know the answer to on a bad week.

And kindness, which sounds soft next to the rest and is not. Science is small and careers are long. I want the work to serve something larger than my own record, and how you treat the people around you is part of that rather than separate from it.