TRAQinform IQ provides spatial and temporal information for each lesion from serial PET/CT scans, helping to inform the next steps for cancer treatment.
Just like every patient responds differently to a treatment, each individual lesion within a patient responds differently as well. Appreciation of treatment response heterogeneity within an individual provides a higher level of understanding, allowing the clinician to make more informed clinical decisions.
CancerNetwork® spoke with Glenn Liu, MD, professor of medicine and medical physics, leader of Genitourinary Oncology at the University of Wisconsin Carbone Cancer Center, and founder of AIQ Solutions. He discussed what makes TRAQinform IQ, an algorithm-based service that provides quantitative information based on PET/CT images, unique when compared with other technologies in the cancer space.
Unlike imaging tools developed for use in diagnostics, which are optimized for accuracy in metastatic cancer, TRAQinform IQ uses imaging to quantify change, thus the priority is precision. While a trained radiologist can provide detailed information for select lesions, patients may have hundreds of lesions.As a result, there is a limitation on quantified information that can be provided within existing workflows with precision. This means that traditional radiology reports often provide limited quantified metrics on select lesions or a descriptive statement regarding the findings.
If the outcome is a binary (progressing/not progressing), we know that a patient who is “progressing” may still be benefitting from treatment (e.g. majority of lesions are improving), and a patient who is “not progressing” may have low benefit (e.g. minority of non-responding lesions may be increasing rapidly). By providing a change in spatial-temporal metrics for all lesions, oncologists can begin to appreciate nuances like oligoprogression or oligoresistance in the setting of widespread metastatic disease.
TRAQinform IQ analysis has specific methods to automatically identify, quantify, and align all lesions (and normal organs) obtained at different time points, which allows the technology to track each individual lesion changes over time, accounting for complex dynamics such as lesions coming and going or splitting or merging. From this, a change in overall disease burden and volume in a patient can be provided to the clinician, as well as a change in each individual lesion. This provides spatial-temporal information (overall and each individual lesion) that oncologists can use to prognosticate overall treatment response and determine future treatments for the patient, Liu said.
In addition, TRAQinform IQ technology can analyze two different imaging modalities (e.g. PET vs CT) and multiple imaging tracers (e.g. FDG, PSMA, or DOTATATE) and match these different scans together, providing a report of lesion concordance between the scans. This information is useful in theragnostics and is being used today to provide information on patient and lesion level SUVmean, as well as the percent and volume of discordant lesions that would not respond to radioligand therapies.
In a study published in the European Journal of Nuclear Medicine and Molecular Imaging that evaluated the effectiveness of overall survival prognosis between TRAQinform IQ technology and established standard of care approaches in 241 patients with diffuse large B-cell lymphoma and non–small cell lung cancer, models trained with all lesion-ROI (TRAQinform IQ) had significantly superior performance than those trained with features extracted from only a few lesion-ROI (e.g., RECIST, PERCIST). This underscores the importance of understanding how every lesion responds.
TRAQinform IQ is a cloud-based software solution and does not require any software to be installed at the site. This tool is FDA cleared and currently being used as part of an Early Access Program at many institutions across the US and in Australia. “Oncologists who see these reports already know how to use them,” Liu said. “They know treatment response heterogeneity exists. Now they can see it. It opens up opportunities to truly individualize treatment decisions.”
Transcript:
There are a lot of tools that are being used, from an imaging standpoint. The vast majority are in diagnostics. Those are different because they are optimized for accuracy. In our case, we are dealing with metastatic cancer, we are [primarily] focused on treatment response, so that technology is optimized for precision. The technology is unique because we have patented methods––how we align scans at different time points, how we match lesions, [or] track how they change over time––because these lesions are complicated. They are coming, they are emerging, they are splitting.
It is [quite] challenging. When we are able to track them over time, we are able to quantify the change in each individual lesion, capturing multiple metrics, and all of them are complementary information. The bottom line is we provide spatial and temporal information for all lesions and the patient, as well as each individual lesion. Ultimately, [this] gives the provider more information to let them know what to do with their patient.
Lokre O, Perk TG, Weisman AJ, et al. Quantitative evaluation of lesion response heterogeneity for superior prognostication of clinical outcome. Eur J Nucl Med Mol Imaging. 2024;51(12): 3505-3517. doi:10.1007/s00259-024-06764-0