Abstract
Introduction: High standardized uptake value (SUV) on FDG PET scan in follicular lymphoma (FL) suggests aggressive disease and possible transformation to diffuse large B-cell lymphoma. Schoder et al, J Clin Oncol, 2005, reported that SUV >10 predicted aggressive lymphoma with >80% certainty and SUV >13 with >90% certainty. However, it is unknown whether the maximum SUV (SUVmax) on FDG PET scan at baseline, suggesting the possibility of focal aggressive or transformed disease, has prognostic value in FL. Here, we determined the prognostic value of SUVmax on baseline FDG PET scan in patients with advanced stage FL treated uniformly with R-CHOP chemoimmunotherapy at initial diagnosis.
Methods: We reviewed medical records of all patients with stage III or IV FL who had FDG PET scan at initial diagnosis and were treated with R-CHOP chemoimmunotherapy at MD Anderson Cancer Center between January 2001 and December 2012. Patients with histological diagnosis of concurrent diffuse large B-cell lymphoma were excluded. Results: For the 225 patients studied, the median age was 57 years (range, 20-82). 83 (37%) patients were >= age 60, 137 (61%) had grade 1 or 2 FL, and 88 (39%) had grade 3A (n=57, 25%) or 3B (n=31, 14%). The Ki-67 score was <=40% for 85 (60%) patients and >40% for 56 (40%) patients. FLIPI risk groups were 54 patients (24%) low, 74 (33%) intermediate, and 97 (43%) high. GELF criteria were met in 133 (59%) patients. Tumor bulk of >= 6 cm was seen in 97 (43%) patients. The absolute lymphocyte count (ALC) was normal or high in 155 (69%) patients and low in 70 (31%). Sixty-nine (31%) patients received rituximab maintenance. There was no correlation between baseline SUVmax on FDG PET scan and Ki-67 score (Pearson correlation co-efficient of 0.168). The overall and complete response rates were 96% and 87%, respectively. The median follow-up time was 66 months. At 5 years, progression-free survival (PFS) was 85% and overall survival (OS) was 90%. Male gender, stage IV, high risk FLIPI, presence of GELF criteria, high beta-2 microglobulin, and low ALC were associated with significantly inferior PFS and OS (p<0.05). Histological grade was not associated with PFS or OS. Age >= 60 was associated with inferior OS but not PFS. Rituximab maintenance was associated with improved PFS but not OS. On baseline FDG PET scan, median SUVmax was 13.7 and the SUVmax range was 1.5-42.1. 105 (47%) patients had SUVmax <=13 and 120 (53%) had SUVmax >13. Patient characteristics including age, gender, histological grade, Ki-67 score, and FLIPI risk groups were not significantly different between the two SUVmax populations (p>0.05). The overall response rates were 94% and 96% for the SUVmax <=13 and SUVmax >13 groups, respectively. The complete response rate was 87% in both groups. At 5 years, the PFS and OS were not significantly different between the low and high SUVmax groups (61% vs 63% for PFS, p=0.98 and 90% vs 89% for OS, p=0.63). PFS and OS were not significantly different even when the patients were grouped into SUVmax <=10 vs >10 (p=0.9 and 0.61, respectively) or when other cut-offs were used. SUVmax was also not predictive of PFS and OS when only the patients meeting GELF criteria were analyzed.
Conclusions: In this large cohort of advanced stage FL patients treated uniformly with R-CHOP chemoimmunotherapy, SUVmax on baseline FDG PET scan was not predictive of clinical outcome or correlated with other features. It is possible that the doxorubicin-based chemotherapy regimen may have benefited patients with high SUVmax who may have underlying aggressive or undiagnosed transformed disease. It remains to be determined whether SUVmax is predictive of clinical outcome in FL patients treated with other commonly used therapies such as rituximab monotherapy, rituximab and bendamustine, or R-CVP.
Wang:Pharmacyclics, Janssen: Honoraria, Research Funding. Westin:Novartis: Research Funding.
