The combination of radiomics and a blood marker-based prediction model may help to distinguish pseudoprogression from true progressive disease among patients with metastatic melanoma being treated with immune checkpoint inhibition, according to the results of a recent study.

According to authors, about one-half of patients with metastatic melanoma will not respond to immune checkpoint inhibition. Because of that, predictive biomarkers are needed to aid in patient selection.

In this study, researchers assessed PET/CT-based radiomics, lesion volume, and routine blood markers for use in the differentiation of psuedoprogression and true disease progression at 3 months. The study included 112 patients with a median follow-up of 22 months.

Continue Reading

Metastases were segmented on CT and FDG-PET at baseline, 3 months, and 6 months. LDH and S100 blood markers were measured at baseline and three months, with their relative change between the two time points calculated.

The median 2-year overall survival was 69%. At 3 months, 16% of lesions had progressed. Five percent of these lesions were identified as pseudoprogression at 6 months.

Related Articles

Those patients with true progression had a 15% 2-year overall survival. Comparatively, patients with pseudoprogression and those without true progressive lesions had a 100% 2-year overall survival rate (P =.002). Some patients had both progressive and pseudoprogressive lesions; these patients had a 2-year overall survival rate of 53%.

The researchers looked at 7 models to distinguish pseudoprogression from true progression and assessed their area under the receiver operating characteristic (AUC) curve:

  • Blood markers: AUC =0.71
  • Volume of metastases: AUC =0.72 at 3 months
  • Radiomics (including volume-related features): AUC =0.79
  • Radiomics (excluding volume-related features): AUC=0.78
  • Blood and volume: AUC =0.79
  • Blood and radiomics (excluding volume): AUC =0.82
  • Blood and radiomics (including volume): AUC =0.78

The combined blood radiomics model excluding volume-related features performed best.

“Noninvasive PET/CT-based radiomics and LDH/S100 are promising biomarkers for early differentiation of pseudoprogression from true progression at an early time-point of 3 months, which might help reduce added toxicity or delayed treatment switch in metastatic melanoma patients treated with immune checkpoint inhibitors,” the researchers concluded.


Basler L, Gabryes HS, Hogan S, et al. Radiomics, tumor volume and blood biomarkers for early prediction of pseudoprogression in metastatic melanoma patients treated with immune checkpoint inhibition [published online April 6, 2020]. Clin Can Research. doi: 10.1158/1078-0432.CC-20-0020.