"We're halfway through Phase I and our biomarker isn't working like we thought."

That's the call I get more often than I'd like to admit.

The panic in their voice is always the same. Months of planning, millions invested, patients enrolled—and now the data isn't telling the story they expected.

Sometimes it's the biomarker itself. What looked predictive in preclinical studies falls apart when faced with the messy reality of clinical samples. Different handling, varying quality, real-world variability that controlled lab conditions never prepared them for.

Other times it's operational. Sample collection protocols that sounded straightforward on paper become impossible to execute across multiple sites. Turnaround times stretch from days to weeks. The "simple" tissue biopsy requirement becomes a major enrollment bottleneck.

The worst cases?
When they discover too late that their biomarker strategy doesn't align with regulatory expectations for their therapeutic context. All that clinical data, but it won't support the companion diagnostic they actually need.

Most biomarker emergencies aren't actually emergencies—they're the inevitable result of decisions made months or years earlier without considering downstream consequences.

The companies that avoid these calls? They bring me in during preclinical development, when there's still time to stress-test assumptions and build robust strategies.

Because by the time you're in crisis mode, your options have already been limited by choices you made when the stakes felt lower.

----------------
I help biotech teams build biomarker strategies that survive first contact with clinical reality. Better to plan for complexity upfront than manage disasters later.