FDG PET parameters differ for predicting advanced colon cancer survival

With a host of guidelines for tumor response presented by the now well known Response Evaluation Criteria in Solid Tumors (RECIST) and PET Response Criteria in Solid Tumors (PERCIST), clinicians need to know that not all parameters are alike for predicting patient survival. Case in point, metabolic volume and lesion glycolysis using F-18 FDG PET has been found to be more predictive of survival in patients with metastatic colorectal tumors treated with radioembolization than standardized uptake guidelines, according to a study published May 31 in the Journal of Nuclear Medicine.

Wolfgang Peter Fendler, from the department of nuclear medicine at Ludwig-Maximilians-University of Munich, in Germany, and colleagues compared criteria for predicting survival after FDG PET therapy monitoring of colorectal metastases treated with Yttrium-90.

“Despite major advances in systemic treatment of metastatic disease, the overall five-year patient survival in the United States has remained lamentably low, below 10 percent,” wrote Fendler et al. “In response to this, several liver-focused treatment strategies have been developed aimed at controlling local metastatic growth. Among these treatments, radioembolization using Y-90 microspheres, also known as selective internal radiation therapy (SIRT), has emerged as a palliative treatment option for hepatic metastases of CRC. Clinical trials with or without concomitant chemotherapy have shown that SIRT can reduce the mass of hepatic metastases sufficiently to permit their surgical resection.”

Colorectal cancer (CRC) is estimated to be the third leading malignant cancer in women and the fourth in men worldwide. Approximately 25 percent of patients being diagnosed with CRC already have liver metastases, the most common secondary site of cancer. The challenge in predicting patient outcomes is defining the appropriate criteria for therapy monitoring, which is important given dismal general prognoses for this type of cancer, with median survival after SIRT currently at about 65 weeks and overall disease control from 35 to 88 percent. Both RECIST and PERCIST were set up to standardize response stratification based on standardized uptake values.

“To evaluate the prognostic value of metabolic parameters, most F-18 FDG PET studies have endeavored to correlate tracer uptake in the tumor with CT or MR imaging volumes, with duration of posttreatment survival being the standard endpoint,” wrote the authors. “However, there is yet no standardized approach for response stratification by means of metabolic imaging, although the maximum F-18 FDG standardized uptake value within the tumor (SUVmax) is traditionally taken as an indicator of tumor vitality. More recently, [PERCIST] led to the adoption of a new metric, SUVpeak, defined as the mean F-18 FDG uptake within a spheric 1-cm3 region around the tumor voxel with the highest SUV, which is intended to provide a more reproducible parameter of maximum lesional uptake. Other authors have proposed a visual response index or changes in total lesion glycolysis as indices of metabolic response to therapy.”

For this study, 80 subjects with hepatic metastases of colorectal cancer underwent SIRT and had F-18 FDG PET/CT performed at the outset and then three months after the treatment. Researchers compared metabolic volume and total lesion glycolysis with SUVmax and SUVpeak in line with current guidelines for every patient, looking at three liver lesions in each subject. All percent changes from baseline to three-month follow up were determined, with tumor response defined as a more than 30 percent decrease in any one of these parameters. Additionally, response was evaluated using RECIST guidelines.

Results for this study showed overall median survival of 60 weeks following SIRT and that metabolic volume or total lesion glycolysis was a better predictor of patient survival. Patients who were seen to have a change in metabolic volume or total lesion glycolysis had significantly better predicted survival. For metabolic volume it was 92 weeks instead of 49 weeks (P = 0.006) and for lesion glycolysis it was 91 instead of 48 weeks (P = 0.025). RECIST 1.1 and changes in SUVmax SUVpeak after treatment predicted outcome at P = 0.086 for RECIST, P = 0.155 for changes in SUVmax and P = 0.310 for changes in SUVpeak.

“Changes in metabolic volume and total lesion glycolytic rate as measured by F-18 FDG PET predicted survival in patients with hepatic metastases from colorectal cancer, whereas changes in SUVpeak or SUVmax and RECIST 1.1 criteria did not predict survival,” wrote the authors.