Evaluating Volumetric FDG-PET Parameters for Identifying Microsatellite Instability in Colorectal CancerPET scans help identify specific markers in colorectal cancer

This meta-analysis evaluated the diagnostic utility of various FDG-PET/CT parameters in distinguishing between high microsatellite instability (MSI-H) and low microsatellite instability (MSI-L) in patients with colorectal cancer. The study included a cohort of 1,579 patients, aiming to determine if imaging biomarkers could provide non-invasive insights into the molecular profile of the primary tumor. Identifying MSI status is clinically critical, as it significantly influences treatment selection and prognosis in the colorectal cancer landscape.

Analysis of intensity-based metrics, specifically SUVmax and SUVmean, demonstrated no statistically significant difference between the MSI-H and MSI-L cohorts. The confidence intervals for these parameters overlapped significantly, suggesting that peak and mean uptake values alone are insufficient for reliably distinguishing microsatellite instability. These findings indicate that while FDG-PET is a standard imaging modality, its standard quantitative metrics may lack the specificity required for molecular subtyping.

In contrast, volumetric parameters demonstrated significant discriminatory power. The Metabolic Tumor Volume (MTV) and Total Lesion Glycolysis (TLG) showed statistically significant differences between MSI-H and MSI-L groups. Specifically, MTV exhibited a standard mean difference (SMD) of 0.67, while TLG showed an SMD of 0.37. These results suggest that the spatial extent and total metabolic activity of the tumor provide a more nuanced reflection of the underlying biology associated with microsatellite instability.

Further investigation into methodology revealed that the detection method for SUVmax influenced results, with studies utilizing PCR showing larger effect sizes compared to those using immunohistochemistry (IHC). However, the correlation between SUVmax and high-grade tumor proportions remained marginally non-significant. This highlights the complexity of integrating morphological features with metabolic imaging to predict molecular characteristics.

While the volumetric metrics show promise as adjunct biomarkers, the evidence remains limited. The diagnostic accuracy, while statistically significant in this meta-analysis, is not sufficient to replace gold-standard tissue-based assessments. Clinicians should view these PET-derived metrics as supplementary tools that may assist in characterizing tumor burden and metabolic activity rather than standalone diagnostic tests. In clinical practice, the integration of MTV and TLG into the diagnostic workflow could potentially assist in identifying patients who might benefit from specific immunotherapies associated with MSI-H status. However, the current data underscores the necessity of maintaining tissue-based confirmation. Future studies should focus on larger, multi-center cohorts to refine the sensitivity and specificity of these volumetric parameters in routine clinical settings.