Surveillance imaging with FDG-PET in the follow-up of melanoma patients at high risk of relapse.

Melanoma/Skin Cancers
Session Type and Session Title: 
Oral Abstract Session, Melanoma/Skin Cancers
Abstract Number: 


J Clin Oncol 33, 2015 (suppl; abstr 9003)
Jeremy Howard Lewin, Alexandra Sanelli, Imogen Walpole, Damien Kee, Michael A. Henderson, David Speakman, John Spillane, David E. Gyorki, Vanessa Estall, Luc Te Marvelde, Kathy Pope, Margaret Chua, Grant A. McArthur, Rodney J Hicks, Mark J. Shackleton; Peter MacCallum Cancer Centre, East Melbourne, Australia

Abstract Disclosures


Background: In the modern era of melanoma treatment, approaches to imaging surveillance following surgery require reconsideration. The aim of this study was to evaluate disease sub-stage specific schedules of positron emission tomography (PET) surveillance for resected stage III melanoma. Methods: Between 2009-2013, patients at the Peter MacCallum Cancer Centre with fully resected AJCC stage III melanoma underwent serial whole body PET/CT scans according to schedules based on Bayesian disease sub-stage relapse probabilities. Schedules were stage IIIA: 6, 18 months; IIIB: 6, 12, 18, 24, 36, 48, 60 months; IIIC: 6, 12, 18, 24, 36 months. Descriptive statistics and contingency table analyses were used to evaluate outcomes for each schedule. Results: Eighty-six patients underwent PET surveillance according to schedule (IIIA: 11; IIIB: 50; IIIC: 25). In total, 232 PET scans were performed over a median follow-up of 28 months after surgery. Relapses were identified in 25 (29%) patients (IIIA: 4%; IIIB: 56%; IIIC: 40%), of which 20 (80%) were asymptomatic at the time of scanning. Incidental secondary malignancies were found in 6 (6.5%) patients. Stage IIIA/B relapses were more likely than stage IIIC to be loco-regional (IIIA/B: 42%; IIIC: 10%; p = NS). Nine (36%) relapsed patients underwent potentially curative resection (IIIA: 1; IIIB: 6; IIIC: 2), with 5 (IIIA: 1; IIIB: 4) free of disease after a median 32 months follow-up. The positive and negative predictive values (PPV, NPV) of an individual PET scan for detecting disease relapse at the same time point were: stage IIIB – PPV 69% (CI: 43-87%) and NPV 99% (CI: 95-100%), stage IIIC – PPV 73% (CI: 39-94%) and NPV 97% (CI: 90-100%). The PPV and NPV of each surveillance protocol for detecting any disease relapse were: stage IIIB – PPV 68% (CI: 43-87%) and NPV 97% (CI: 83-99%), stage IIIC – PPV 73% (CI: 39-94%) and NPV 86% (CI: 57-98%). The sensitivity and specificity of the overall approach of sub-stage specific PET/CT surveillance for detecting disease relapse were 88% (CI: 69-97%) and 84% (CI: 72-92%), respectively. Conclusions: FDG-PET is effective in detecting asymptomatic metastases and thus facilitating early treatment in patients who relapse after resection of stage III melanoma.