Personalizing Cancer Therapy For Complex Molecular Blueprints: Insights from the I-PREDICT Trial

One of the persistent challenges in precision oncology is the fact that most advanced cancers are complex and highly individualized. A recent study published by Jason K. Sicklick, MD and colleagues (J. Clinical Oncology) tackles and provides a possible solution to this problem. Traditional precision medicine approaches tend to match patients with a single drug aimed at a single biomarker, often defined by tumor type. But reality shows that a typical advanced cancer harbors multiple driver alterations, frequently five or more, that do not align neatly with histology. This genomic complexity, combined with differences between patients in age, frailty, organ function, metabolism, gender, and race, means that the one-size-fits-all approach to drug choice and dosing often misses the mark, either by under-treating or by causing unnecessary side effects. Compounding the problem is the historical barrier that most combination therapies require phase I trials to establish safe dosing, a process that may be too slow and impractical for patients with life-threatening disease.

In response, Sicklick and his colleagues used I-PREDICT (Investigation of Profile-Related Evidence Determining Individualized Cancer Therapy; NCT02534675) to study an innovative N-of-1 design in which each patient’s treatment was matched to their unique constellation of pathogenic tumor alterations, often with novel drug combinations. The team used various advanced genetic sequencing technologies on tumor tissue and circulating tumor DNA to create a comprehensive molecular profile for each patient. These profiles were reviewed by a multidisciplinary molecular tumor board, which recommended the best possible drug combination for that specific genomic profile. A key metric, the “matching score”, quantified how well the administered regimen targeted the patient’s known pathogenic alterations. Because many drug regimens had no prior dosing data, physicians started at lower doses and adjusted upward or downward within the same patient based on tolerance, rather than using conventional interpatient dose escalation models.

The study enrolled 210 evaluable patients with unresectable or metastatic cancers from various tissue origins.  The median number of pathogenic alterations was five and treatment diversity reflected this complexity. 157 different drug regimens were administered, 103 of which had no established safety or recommended dose at the time. Frequent monitoring during the early weeks allowed for individualized dose titration. Interestingly, serious toxicity rates were lower in first-in-human combinations than in regimens with established dosing, likely due to conservative starting doses.

The results showed that the better the molecular match, the better the outcome. Patients with a matching score above 50 percent had significantly higher disease control rates, longer progression-free survival, and longer overall survival compared to those with lower scores. In other words, the degree of matching of drugs to tumor molecular alterations (reflected by the matching score) was found to associate significantly, independently, and linearly with disease control rates and longer progression-free survival and overall survival.

This work matters because it shows a viable path forward for truly personalized cancer care, moving beyond tumor type or single biomarker models to address the full spectrum of actionable alterations. It integrates patient-specific clinical factors into dosing decisions and uses modern genomic tools to build treatment from the molecular blueprint up. While the findings are hypothesis-generating and require confirmation in randomized controlled trials, they offer a blueprint for how oncology could evolve toward routinely delivering the right drugs, at the right doses, at the right time—for each patient.

N-of-1 matched combination therapy, including with previously unstudied regimens, can be safely managed, and that better matching of tumor molecular profiles to drugs administered, enabled by personalized dosing established via intrapatient dose titration, correlates with activity in patients with lethal, advanced/metastatic cancer.