The World's First Pre-Early Cancer Screening Company

We find cancer
before
tumors exist.

MCED blood tests detect what a tumor has already left behind. xTenure reads the biological trajectory that leads there — years earlier — from your medical history.

Latent disease discovery is not early detection. It is pre-early detection: identifying the trajectory of biological change before any tumor forms, before any marker spills into the bloodstream, before any symptom appears.

Scroll
The Core Distinction

There is a window that MCED cannot see.

Every cancer that MCED tests eventually detect began years earlier as a silent biological drift — a trajectory of micro-changes in how your body processes, signals, and adapts. xTenure reads that trajectory. MCED reads its consequences.

Existing MCED Approach

Detecting the signal a tumor leaves behind

Multi-cancer early detection tests — Galleri, OncoSeek, and others — are powerful tools. They detect protein markers, cell-free DNA fragments, and other biological signals that tumors shed into the bloodstream. But a tumor must already exist and be of sufficient size to shed detectable signal. Detection, however early, is detection of a disease already present.

xTenure's Approach

Reading the trajectory before the tumor forms

Your complete medical history — labs, diagnoses, prescriptions, vitals, procedures, over years — contains a latent biological story. Subtle deviations in CA 19-9 trends, a sequence of GI visits, a new-onset metabolic shift: individually unremarkable. As a trajectory, they point to where your biology is going. xTenure reads that trajectory years before any tumor would be detectable.

Where xTenure operates on the cancer timeline

Every cancer has a natural history — a progression from normal biology through latent trajectory to tumor formation to clinical presentation. xTenure operates at the earliest detectable stage of that journey.

Year −10 Year −8 Year −6 Year −4 Year −2 Diagnosis
◆ xTenure Detection Window
◆ MCED Detection Window
Latent trajectory
begins — no tumor
EHR signal patterns
become detectable
xTenure risk
stratification fires
Tumor forms —
MCED now possible
MCED detects
tumor markers
Clinical
diagnosis
xTenure Zone — Years Before Tumor

Latent Disease Discovery

Reading biological trajectories from longitudinal health records to identify where patient biology is heading — before any tumor exists to detect.

MCED Zone — Tumor Already Present

Multi-Cancer Early Detection

Blood-based tests detect signals shed by existing tumors. Powerful and important — but operating downstream of xTenure's detection window. xTenure identifies who needs MCED testing.

Clinical Zone — Late Presentation

Symptomatic Diagnosis

The majority of cancers are still diagnosed here. Stage III–IV diagnosis. Where survival outcomes are dramatically worse. Where the full burden of late-stage treatment falls.

The Platform

From medical history to clinical action.

A four-stage architecture that transforms longitudinal health records into disease trajectories, risk scores, and admissible clinical actions.

01

Latent Disease Discovery

xTenure's AI reads your complete longitudinal medical record — labs, diagnoses, medications, procedures, vitals — not as isolated events but as a biological trajectory. It infers the latent biological history that your observed records are a partial, fragmented reflection of. This is where the disease signal lives — in the pattern of the whole, not any single finding.

02

Disease Trajectory Modeling

From the latent history, xTenure projects forward: a disease-aware probability distribution over future biological states at a 3–5 year horizon. Not a single risk score — a structured distribution that specifies which disease branches are active, which are approaching transition thresholds, and how confident the model is in each trajectory. For cancer, it specifies which of nine cancer types the trajectory is consistent with.

03

Risk Stratification

The trajectory distribution feeds a three-gate architecture. Gate 1 asks: is the evidence strong enough to act? Gate 2 asks: what is the most economically justified action? Gate 3 asks: is the healthcare system able to absorb the recommended intervention load? Together, the gates produce an admissible action — the right intervention for the right patient at the right time, defensible to clinicians, payers, and regulators.

04

Targeted MCED & Clinical Decision Support

For patients whose trajectory reaches the pre-tumor threshold, xTenure recommends targeted MCED testing — directing blood-based cancer detection to exactly the individuals most likely to test positive. This makes MCED economically viable at population scale. For clinicians, xTenure provides the most informative next test, the most probable biological narrative, and the intervention with the highest expected clinical value.

Patient Trajectory Signal — Pancreatic Cancer Branch
xTenure MCED
Latent disease score over time (years before diagnosis)
CA 19-9 trend
0.78
New-onset DM
0.64
Weight change
0.55
GI visit pattern
0.49
Bilirubin trend
0.42
Clinical Cascade

From latent signal to admissible action.

01

Latent Disease Discovery

Complete longitudinal record ingested. Bayesian compressed sensing constructs the history cloud — the posterior over all plausible biological histories.

02

Trajectory Projection

Disease-aware probability distribution over 3–5 year biological futures. Nine cancer branches. Structured uncertainty.

03

Risk Stratification

Three-gate architecture. Decision sufficiency, economic justification, system feasibility. Admissible action produced.

04

Targeted Intervention

Directed MCED testing. Clinical decision support. Insurance underwriting. DTC health risk reporting.

>90%
5-year survival when cancer
detected at Stage I
9
High-mortality cancer types
targeted by xTenure's platform
5yr
Lead time advantage over
conventional MCED detection
150M
Patient longitudinal records
powering xTenure's AI backbone
The Science

Built on a principled theoretical foundation.

xTenure's platform is not a pattern-matching engine. It is grounded in Bayesian inference, stochastic dynamics, and information geometry — producing principled uncertainty quantification and theoretically defensible clinical recommendations.

History Cloud Inference
The observed medical record is an incomplete, fragmented projection of the true latent biological history. xTenure constructs a posterior distribution — the history cloud — over all plausible biological histories consistent with the observed data. This is Bayesian compressed sensing applied to medical records: the minimum-complexity biological explanation for the observed trajectory.
Bayesian CS
Latent Inference
Stochastic Trajectory Modeling
Disease progression follows the Onsager-Machlup stochastic action: biological trajectories that deviate from homeostatic drift are exponentially suppressed. The most probable disease trajectory — the instanton — is the least-action path from the current biological state to disease onset. This is the theoretical basis for lead-time estimation and biological plausibility constraints.
Stochastic Dynamics
Path Integrals
Disease-Aware Future Cloud
From the history cloud, xTenure derives a disease-aware distribution over future biological states at the prediction horizon. Each of the nine OncoSeek cancer types has a dedicated trajectory branch with specific biomarker inputs — AFP for HCC, CA 19-9 for pancreatic, CA125 for ovarian — enabling cancer-specific risk stratification rather than generic cancer risk scoring.
Trajectory Projection
9-Cancer Scope
Optimal Test Ordering (Gate 2)
When the history cloud is ambiguous, xTenure computes which additional observation — which lab test, which imaging study — would most efficiently resolve the uncertainty. This is optimal experiment design using Fisher information geodesics: the test that traverses the largest distance in the statistical manifold of biological hypotheses. Gate 2 produces the most informative next clinical action, not the most commonly ordered one.
Information Geometry
Active Learning
Foundation Model Backbone
A medical foundation model trained on 150 million longitudinal patient records across diagnoses, pharmacy, labs, procedures, clinical notes, and provider chains — substantially beyond published baselines. The backbone learns the nonlinear mapping from observed medical records to biological state space, enabling history cloud inference at population scale.
Medical AI
150M Patients
Who We Serve

Multiple channels. One platform.

xTenure's platform serves the full spectrum of healthcare decision-makers — from individual patients to population-scale health systems.

🏥

Health Systems & Payers

Population-level latent disease surveillance. Proactive identification of high-risk individuals years before expensive late-stage presentation. A direct path to reduced cancer mortality and lower long-term cost burden.

🛡️

Life & Health Insurers

Longitudinal risk stratification for underwriting. Admissible-action architecture designed for insurance portfolio constraints. Actuarially sound, individually precise, and gate-validated for regulatory compatibility.

🧬

MCED Test Partners

xTenure directs blood-based cancer detection to the individuals most likely to test positive — making MCED economically viable at national scale. A pre-screening funnel that multiplies the clinical and commercial impact of every MCED test ordered.

👤

Direct-to-Consumer Health

Partnering with leading consumer health platforms to give individuals access to their own biological trajectory — personalized cancer risk years in advance, with the most informative next step clearly identified.

🔬

Research Institutions

Latent disease discovery across nine cancer types using the largest longitudinal US patient dataset deployed for this purpose. A platform for pre-diagnostic signal research, lead-time analysis, and cancer trajectory modeling.

🌍

Global Health Programs

A low-cost, EHR-based pre-screening layer that makes MCED economically viable in settings where blanket blood-based screening is not. Designed for interoperability with existing health information infrastructure worldwide.

Start a conversation.

Whether you represent a health system, insurer, MCED partner, research institution, or investment organization — we welcome a briefing on how xTenure can transform cancer detection outcomes for your population.