PET/CT in lymphoma surveillance: A large single-center experience.

Lymphoma and Plasma Cell Disorders
Session Type and Session Title: 
General Poster Session, Lymphoma and Plasma Cell Disorders
Abstract Number: 
J Clin Oncol 32:5s, 2014 (suppl; abstr 8569)
Lourdes M. Mendez, Anish V. Sharda, Kelly A. Bodio, Robin Joyce; Beth Israel Deaconess Medical Center, Boston, MA; Beth Israel Deaconess Medical Center/Harvard Medical School, Boston, MA

Abstracts that were granted an exception in accordance with ASCO's Conflict of Interest Policy are designated with a caret symbol (^).

Abstract Disclosures


Background: The role of FDG PET-CT in the modern management of lymphoma continues to evolve. Current guidelines are largely based on studies with limited numbers of patients. The aim of this study was to define the utility of PET-CTs in identifying asymptomatic relapsed disease during post-treatment follow-up. Methods: We performed a large, retrospective cohort study of all PET-CTs performed for patients with pathologically proven lymphoma at our institution from 2003-2005. 1,085 consecutive PET-CTs from 370 patients with lymphoma and accompanying clinical data were reviewed. 81 patients had Hodgkin lymphoma (HL), 154 aggressive Non-Hodgkin lymphoma (a-NHL) and 135 indolent Non-Hodgkin lymphoma (i-NHL). Most scans were obtained in the post-treatment setting (861: either at completion of therapy or during follow-up); 110 and 114 were performed as part of initial staging and during therapy respectively and were excluded from the following analysis. Results: 44% of all scans were positive of which 75.7% were associated with progressive disease and 24.3% with disease remission. PET-CT had a sensitivity of 97.5%, specificity of 90.3%, PPV of 75.7% and NPV of 99.2% for relapsed disease in our cohort (Table 1 shows sub-group statistics). The false positive rate for PET-CT was 24.3% most often leading to further imaging. In comparison, the presence of symptoms had a sensitivity of 43.6%, specificity of 88.5%, PPV of 53.8% and NPV of 83.7% without significant variation among HL, a-NHL and i-NHL. Notably, 52% of a-NHL patients with relapsed disease documented by PET-CTs were asymptomatic. Conclusions: Our results indicate that PET-CT is highly sensitive for detecting progressive lymphoma in the context of post-treatment follow-up offset by a high rate of false-positives. This highlights the need for algorithms incorporating additional parameters for risk stratification. In this series, a significant portion of a-NHL recurrences detected by PET-CTs were clinically silent. Analysis of how detection of asymptomatic relapse affects patient outcomes is underway.

Measures of performance of PET-CT by lymphoma subgroup.
Lymphoma Sensitivity (%) Specificity (%) PPV (%) NPV (%)
HL 100 94 80.5 100
a-NHL 97.8 76.1 50.9 99.3
i-NHL 96.3 90.2 82.3 98.1