Every cancer specimen is genetically unique, yet all share common traits like uncontrolled division and co-opting normal cell survival mechanisms. The key to treatment is finding pathways that are different enough from normal cells to target and exploit.

Because cancer cells can be genetically different even a centimeter apart within the same tumor, a single targeting agent will inevitably miss some malignant tissue. The solution is a 'cocktail' of multiple tumor-targeted dyes, each targeting a different marker, to ensure visualization of virtually all cancer variants in a patient.

An individual tumor can have hundreds of unique mutations, making it impossible to predict treatment response from a single genetic marker. This molecular chaos necessitates functional tests that measure a drug's actual effect on the patient's cells to determine the best therapy.

Instead of creating therapies for hundreds of specific driver mutations, which vary widely between patients, Earli's platform targets downstream commonalities—the "hallmarks of cancer" like rapid cell proliferation. These pathways are where diverse mutations converge, creating a more universal and reliable target across different cancers.

The same cancer-driving mutation behaves differently depending on the cell's internal "wiring." For example, a drug targeting a mutation works in melanoma but induces resistance in colorectal cancer due to a bypass pathway. This cellular context is why genetic data alone is insufficient.

Many blood cancers are better understood as "regulatory problems" driven by epigenetic failures—the systems controlling which genes are turned on or off. This shifts the therapeutic focus from targeting DNA mutations to developing drugs, like IDH inhibitors, that correct these underlying control mechanisms.

Despite billions invested over 20 years in targeted and genome-based therapies, the real-world benefit to cancer patients has been minimal, helping only a small fraction of the population. This highlights a profound gap and the urgent need for new paradigms like functional precision oncology.

The progress of AI in predicting cancer treatment is stalled not by algorithms, but by the data used to train them. Relying solely on static genetic data is insufficient. The critical missing piece is functional, contextual data showing how patient cells actually respond to drugs.

By training on data across many cancer types ("pan-cancer"), AI models learn universal biological principles. This approach allows them to generalize learnings from large, common cancer datasets to significantly improve prediction accuracy for rare cancers, which often suffer from a lack of specific data for training effective models.