According to a study published this week in JCO Precision Oncology, researchers from the University of Florida have developed a predictive analytics tool that can identify acute lymphoblastic leukemia (ALL) patients’ risk for consequences from chemotherapy drug toxicity.
One of the most prevalent types of childhood leukemia, ALL, necessitates intense chemotherapy, which may result in toxicities from the chemotherapy drugs or other unfavorable side effects. These adverse effects, including as heart or nerve damage, are frequently significant or life-threatening and contribute to the burden of the disease because ALL survivors have a higher risk of developing chronic illnesses and dying young.
According to Jatinder Lamba, PhD, study leader, associate dean of research and graduate education, and professor in the University of Florida College of Pharmacy, “the survival rates for this type of leukemia are generally good, but one of the biggest clinical challenges for these patients is toxicity that affects their quality of life because of the intensive chemotherapy drugs they are given.” Our objective was to create a genetic toxicity score that can determine whether a patient is more or less likely to experience toxicity while taking these drugs.
The research team used numerous types of toxicities, such as neurological, gastrointestinal, and infections that required hospital admissions, detected within 100 days of the start of cancer treatment, to create these rankings. From 75 individuals who received treatment at UF Health between 2012 and 2020, toxicity and DNA data were collected.
The researchers showed a substantial correlation between several single-nucleotide polymorphisms (SNPs) and genes that affect toxicities and aid in risk prediction using this data for pharmacogenomic evaluation.
The researchers used artificial intelligence (AI) to model combinations of up to three SNPs and other genetic variants that are likely to indicate a certain type of toxicity to further enhance and refine prediction. The model’s outputs were utilized to create polygenic toxicity risk scores.
In univariate analysis, the researchers discovered numerous SNPs to be predictive of toxicity characteristics, and multivariable combination analysis indicated that several genes are probably responsible for a patient’s sensitivity to chemotherapy-induced toxicity. The most frequent toxicities during leukemia therapy were accurately anticipated by this combination-based strategy.
The findings, according to the researchers, show the potential of multivariable, SNP-based models to develop clinically applicable biomarkers for toxicity risk assessment. This could make cancer treatments more individualized and reduce side effects for patients. A risk-scoring tool like this one could direct professional judgement and enhance patient quality of life if used in the clinical context.
“With this score in a patient’s medical record, the clinician will know the patient’s risk of developing life-threatening toxicities and be able to make informed decisions,” Lamba said. “This score will be recorded in the patient’s medical record.” To prevent the toxicity, “they can preemptively enhance supportive therapy or change the drug dosages, or they can watch the patient more closely.”
