Prediction of Response to Immune Checkpoint Inhibitor Therapy Using Early-Time-Point 18F-FDG PET/CT Imaging in Patients with Advanced Melanoma

Abstract

The purpose of this study was to evaluate 18F-FDG PET/CT scanning as an early predictor of response to immune checkpoint inhibitors (ICIs) in patients with advanced melanoma. Methods: Twenty patients with advanced melanoma receiving ICI prospectively underwent 18F-FDG PET/CT at 3 scan intervals: before treatment initiation (SCAN-1), at days 21-28 (SCAN-2), and at 4 mo (SCAN-3). This study was approved by the institutional review board, and informed consent was received from all patients who were enrolled between April 2012 and December 2013. Tumor response at each posttreatment time point was assessed according to RECIST 1.1, immune-related response criteria, PERCIST (PERCIST 1.0), and European Organization for Research and Treatment of Cancer (EORTC) criteria. Performance characteristics of each metric to predict best overall response (BOR) at ≥ 4 mo were assessed. Results: Twenty evaluable patients were treated with ipilimumab (n = 16), BMS-936559 (n = 3), or nivolumab (n = 1). BOR at ≥ 4 mo included complete response (n = 2), partial response (n = 2), stable disease (n = 1), and progressive disease (n = 15). Response evaluations at SCAN-2 using RECIST 1.1, immune-related response criteria, PERCIST, and EORTC criteria demonstrated accuracies of 75%, 70%, 70%, and 65%, respectively, to predict BOR at ≥ 4 mo. Interestingly, the optimal PERCIST and EORTC threshold values at SCAN-2 to predict BOR were >15.5% and >14.7%, respectively. By combining anatomic and functional imaging data collected at SCAN-2, we developed criteria to predict eventual response to ICI with 100% sensitivity, 93% specificity, and 95% accuracy. Conclusion: Combining functional and anatomic imaging parameters from 18F-FDG PET/CT scans performed early in ICI appears predictive for eventual response in patients with advanced melanoma. These findings require validation in larger cohorts.

Trial registration: ClinicalTrials.gov NCT01666353.

Keywords: FDG; PET/CT; immune checkpoint inhibitor; melanoma; response assessment.

© 2017 by the Society of Nuclear Medicine and Molecular Imaging.

Figures

FIGURE 1.

Scatterplot comparing early CT- and PET-based changes with response to ICI at ≥ 4 mo. Each dot represents a single patient, color coded according to best overall response at ≥ 4 mo. Two horizontal dashed lines on y-axis (+20% and −30%) correspond to thresholds for PD and PR, respectively, using RECIST 1.1, in absence of appearance of new tumor lesions. Vertical dashed line at +15.5% on x-axis represents a threshold associated with eventual response according to criteria proposed in Figure 2.

FIGURE 2.

Patients whose CT scans performed 3–4 wk into therapy demonstrate an objective response (PR or CR by RECIST 1.1) are predicted to maintain a response at 4 mo. Similarly, PD detected at that same interval predicts continued disease progression at 4 mo. In patients with stable disease by RECIST 1.1 at 3–4 wk, an increase > 15.5% in SULpeak of hottest lesion by PET is associated with eventual clinical benefit (PR or CR at 4 mo or stable disease ≥ 6 mo). Sensitivity, specificity, and accuracy of algorithm to predict response at 4 mo were 100%, 93.3%, and 95.0%, respectively. CR = complete response; PD = progressive disease; PR = partial response; SD = stable disease; SULpeak = average SUV corrected by lean body mass within a 1-cm3 spheric volume of interest.

FIGURE 3.

PET/CT images demonstrating representative changes in melanoma inguinal lymph node metastasis (red arrowheads) at 4 wk and 4 mo after initiation of ipilimumab. At about 4 wk (SCAN-2), sum of target lesion diameters assessed by CT scan (top) increased by 18.6% (stable disease by RECIST 1.1). During that same interval, PET imaging revealed 25.1% increase in SULpeak (average SUV corrected by lean body mass within a 1-cm3 spheric volume of interest) (PERCIST). Imaging at approximately 4 mo revealed a marked improvement in 18F-FDG avidity of inguinal lymph node metastasis. Similar pattern was observed in this patient’s other sites of disease, including hepatic, nodal, and soft-tissue metastases. Patient’s metastases outside of brain remained stable for 51 wk.