What if your patients can be accurately characterized and stratified after all?
Most conventional approaches to patient characterization find limited signal — not because the biology isn't there, but because they look at predefined targets, or they can't process data at the scale where the real signals live.
Our new technology, COMPASS, analyzes an average of one million molecular datapoints per blood sample. Unfiltered and unbiased, it uses machine learning to identify the blood signatures that separate your clinical groups. If signals exist in your patients' blood, we believe COMPASS will find them — even if others couldn't.
Start with a Foundation Study for a low-risk entry point to high-value information.
The Problem: Poorly Characterized Trial Populations
Most clinical trials entail significant uncertainty:
-
Patient populations are heterogeneous
-
Response signals are often diluted or missed
-
Critical biological differences remain hidden
Most trials don't fail because the drug doesn't work. They fail because the patient population was never fully understood.
Why Current Approaches Fall Short
Modern approaches to patient stratification rely on imperfect biological signals.
Whether based on targeted assays, predefined biomarkers, or constrained datasets, most methods:
-
Capture only a small fraction of the biological information present in a sample
-
Focus on expected or known markers
-
Lack the depth needed to fully differentiate complex patient populations
As a result:
-
Important biological differences between patients remain unresolved
-
Subtle but meaningful signals are missed
-
Patient stratification is often incomplete or imprecise
When the underlying data lacks depth, the resulting decisions are inherently limited.
Blood as a Source of Patient Information
In theory, using mass spectrometry to analyze blood samples is one of the most powerful ways to understand patients. In practice, that potential is rarely realized.
The bottleneck in blood-based molecular analysis is not the sample, nor the instrument. It is the software — which cannot process data at the scale that modern mass spectrometers generate.
As a result, researchers must:
-
Simplify sample preparation to limit the number of detectable molecules
-
Adjust instrument methods to reduce data complexity
-
Pre-select or filter data before analysis, and/or
-
Focus on known or expected molecular targets
These decisions are not driven by biology—they are driven by computational limitations.
Blood contains deep patient information—but most workflows capture only part of it.
How Magellan is Fundamentally Different
Magellan overcomes data scale limitations.
Our suite of analytical tools, COMPASS, is designed to process the full scale and complexity of mass spectrometry data—without requiring data reduction, filtering, or pre-selection.
Because of this, the entire workflow changes:
-
Samples can be prepared to release the broadest possible range of molecular information
-
Instruments can be run to capture all detectable signals
-
All resulting data can be retained and analyzed in full
The result is a fundamentally different kind of analysis. Not a larger panel. A complete and unbiased view of the molecular information present in blood — analyzed at the depth and scale where meaningful biological signals actually live.
Conventional approaches find what they are designed to look for. Our COMPASS platform looks at everything detectable in blood— giving you the best chance of finding what matters.
Magellan is Designed to Help Clinical Development:
-
Identify treatment-response signals
-
Stratify patients
-
Predict treatment response
-
Detect clinically relevant subgroups
-
Inform enrollment and trial design decisions
Make better-informed decisions about your patients and your trial.
How It Fits Into Your Trial
Using standard serum or plasma samples, Magellan integrates into existing studies without changing workflows.
-
No specialized sample collection required
-
Applicable to retrospective or prospective studies
-
Scales from small pilot studies to large trials
Works with prospective or banked serum or plasma samples—no additional collection required.