AI Software Tackles the Hidden Bottleneck Derailing One in Five Cancer Molecular Tests

When cancer patients undergo biopsies for molecular testing, up to one in five samples fails to yield results—not because the technology fails, but because the tissue itself contains insufficient tumor cells. Now, Aignostics has launched an AI software suite designed to catch these inadequate samples before they enter expensive molecular workflows, potentially saving laboratories money while sparing patients repeat procedures.

Addressing a Costly Quality Control Gap

The newly launched HETA Clinical suite analyzes H&E-stained digital pathology slides to assess tumor content and tissue quality before molecular testing proceeds. According to pathologists at academic medical centers and precision medicine laboratories cited by CLP Magazine, 10–20% of molecular tests fail due to insufficient tumor cellularity in tissue specimens—a problem that triggers repeat biopsies, extends diagnostic timelines, and wastes laboratory resources on samples destined to fail.

The four-product suite quantifies slide quality, tumor presence, mitotic counts, and tumor cellularity across whole-slide images. Built on Aignostics' HETA platform (formerly Atlas H&E-TME), a foundation model for spatial H&E tissue analysis, the pan-cancer system is designed to work across diverse oncology indications and pathology subspecialties. Dr. Rosemarie Krupar, Vice President of Pathology at Aignostics, positions the technology as a solution to inter-observer variability in tissue adequacy assessment—a subjective judgment that currently relies on pathologist experience and can vary between reviewers.

From Research Validation to Laboratory Integration

A recent study demonstrated that HETA's cell classification matched or exceeded expert pathologist benchmarks across multiple tumor types, according to the company. The software suite carries research-use-only (RUO) designation and is not cleared for diagnostic procedures, limiting its current application to research and laboratory workflow optimization rather than direct clinical decision-making.

Aignostics plans to integrate HETA Clinical with Techcyte Fusion AP, a digital pathology image management platform, suggesting a strategy to embed the quality control tools directly into existing laboratory information systems. The company has also released PathoSearch, a visual search engine for historical case reference, in early access—a complementary tool that could help pathologists compare current samples against archived cases when assessing adequacy.

Economic and Clinical Stakes

For pathology laboratories operating on tight margins, the economic case is straightforward: identifying inadequate samples before molecular testing prevents wasted reagents, technician time, and instrument capacity on tests destined to fail. For patients awaiting molecular profiling to guide precision oncology treatment decisions, successful implementation could mean fewer repeat biopsies and faster time to diagnosis.

The launch positions Aignostics in the pre-analytical quality control segment of computational pathology, a space distinct from diagnostic AI tools that interpret pathology for disease detection. As digital pathology adoption accelerates and molecular testing becomes standard of care for more cancer types, the ability to predict—and prevent—sample failures before expensive genomic workflows begin represents a pragmatic application of AI to laboratory operations rather than clinical interpretation.

Whether HETA Clinical can deliver on its promise to reduce the 10–20% failure rate remains to be demonstrated in real-world laboratory settings, but the company's focus on a well-documented operational pain point suggests a clear value proposition for pathology labs navigating the economics of precision medicine.

Sources

1. New AI Tools Aim to Identify Tissue Samples at Risk of Failed Molecular Testing — CLP Magazine