"The way we dose chemotherapy is crude. Cancer patients can end up with bad reactions that put them off continuing treatment."
Chemotherapy is one of the most widely used treatments in cancer care – yet the way the doses are determined has adapted very little in the last several decades. Despite advances in imaging, genomics, and precision medicine, most chemotherapy regimens still begin with the same calculation:
Body Surface Area (BSA) - a formula from the 1970s that estimates dosage using only a patient's height and weight.
BSA was never designed to capture the full complexity of the human body. Let’s take a look at the bodies of a sumo wrestler and a bodybuilder. While their height and weight may be similar, our sumo wrestler would have relatively more fat in their body, while the bodybuilder would have more muscle. Yet under the system of BSA-based dosing, both would receive roughly the same dose. Since their body compositions are vastly different, they will develop different degrees of toxicity.
And the consequences of these toxicities are incredibly severe.
They are so severe that at least 20% of chemotherapy patients stop treatment early, and a small proportion may even progress to death.
They also mean increased hospitalizations, poorer quality of life, greater health economic impact, and an increase in cancer treatment failure.
A more individualized approach
At PredicTx Health, we are developing ML-driven tools designed to analyze patient-specific biomarkers (including imaging-derived measures of body composition) to better understand how an individual might tolerate chemotherapy.
Instead of relying solely on an archaic method, our algorithm examines deeper physiological signals that correlate with treatment tolerance. Early validation shows promising potential:
Chemotherapy will always require clinical decision-making, but the data behind those decisions should evolve. PredicTx aims to move dosing away from one-size-fits-all formulas and towards individualized care.