Week one FLT-PET response predicts complete remission to R-CHOP and survival in DLBCL

// Ken Herrmann 1,2,* , Andreas K. Buck 1,2,* , Tibor Schuster 3 , Kathrin Abbrederis 4 , Christina Blümel 2 , Ivan Santi 1 , Martina Rudelius 5,6 , Hans-Jürgen Wester 1 , Christian Peschel 4 , Markus Schwaiger 1 , Tobias Dechow 4,7 and Ulrich Keller 4 1 Department of Nuclear Medicine, Technische Universität München, Munich, Germany 2 Department of Nuclear Medicine, Universitätsklinikum Würzburg, Würzburg, Germany 3 Department of Epidemiology, Biostatistics and Occupational Health, McGill University, Montreal, Canada 4 III. Medical Department, Technische Universität München, Munich, Germany 5 Institute of Pathology, Technische Universität München, Munich, Germany 6 Institute of Pathology, Universitätsklinikum Würzburg, Würzburg, Germany 7 Oncology Ravensburg, Ravensburg, Germany * These Authors contributed equally to this work Correspondence: Ulrich Keller, email: // Ken Herrmann, email: // Keywords : Lymphoma; DLBCL; Positron emission tomography; [18F]Fluorodeoxythymidine; FLT-PET Received : March 28, 2014 Accepted : May 17, 2014 Published : May 19, 2014 Abstract Despite improved survival in the Rituximab (R) era, a considerable number of patients with diffuse large B-cell lymphoma (DLBCL) ultimately die from the disease. Functional imaging using [ 18 F]fluorodeoxyglucose-PET is suggested for assessment of residual viable tumor very early during treatment but is compromised by non-specific tracer retention in inflammatory lesions. The PET tracer [ 18 F]fluorodeoxythymidine (FLT) as surrogate marker of tumor proliferation may overcome this limitation. We present results of a prospective clinical study testing FLT-PET as superior and early predictor of response to chemotherapy and outcome in DLBCL. 54 patients underwent FLT-PET prior to and one week after the start of R-CHOP chemotherapy. Repetitive FLT-PET imaging was readily implemented into the diagnostic work-up. Our data demonstrate that the reduction of FLT standard uptake value mean (SUV mean ) and SUV max one week after chemotherapy was significantly higher in patients achieving complete response (CR, n=48; non-CR, n=6; p<0.006). Martingale-residual and Cox proportional hazard analyses showed a significant monotonous decrease of mortality risk with increasing change in SUV. Consistent with these results, early FLT-PET response showed relevant discriminative ability in predicting CR. In conclusion, very early FLT-PET in the course of R-CHOP chemotherapy is feasible and enables identification of patients at risk for treatment failure.