Prognostic Value of End of Treatment FDG-PET Scan in T-Cell Lymphoma, a 20-Year Single Institution Study

T-cell lymphomas represent a rare and heterogeneous group of non-Hodgkin lymphomas (NHLs) characterized by dismal prognosis. Current first line chemotherapeutic approaches have not significantly changed during the last 20 years, and several efforts have been made for early risk stratification of T-cell NHL patients with unsatisfactory results. Functional imaging with 2-deoxy-2-[fluorine-18]fluoro- D-glucose positron emission tomography scan (FGD-PET) has demonstrated improved outcome stratification and response evaluation in Hodgkin lymphoma and B-cell NHL, in comparison with computed tomography (CT) scan. Most T-cell lymphomas are FDG avid and FDG-PET is routinely used in clinical practice, however limited and conflicting data are available on the value of FDG-PET in response assessment at the end of treatment (FDG-PETend). With the aim of evaluating the prognostic value of FDG-PETend in T-cell lymphomas we performed a retrospective study evaluating all T-cell NHL patients treated at the European Institute of Oncology (IEO, Milan) over the past 20 years. Clinical data of all T-cell NHL patients treated at IEO from 1995 to 2015 were retrospectively collected. The back bone of first line therapy did not significantly change over time, being based on the administration of CHOP/CHOP-like chemotherapy eventually followed by treatment intensification with autologous stem cell transplantation (ASCT). Post first-line therapy FDG-PET scans were visually evaluated according to the criteria of the international Harmonization Project (positive vs negative). Ninety-eight consecutive patients (58 males, 40 females) with complete clinical data were considered in this analysis. FDG-PET was used for response evaluation starting from 2002, but only 18 patients considered in the present analysis were treated before 2002. Median age at diagnosis was 54 years (range 14-82). Fifty-two patients (53.1%) had Peripheral T-cell lymphomas not otherwise specified (PTCL-NOS), 23 (23.5%) had anaplastic large T-cell lymphoma (ALCL), 15 (15.3%) had acute lymphoblastic T-cell lymphoma (LAL-T), 8 (8.1%) had NK/T-cell lymphoma. Ten patients were diagnosed in stage I, 21 in stage II, 17 in stage III, 50 patients in stage IV. Thirty-two patients underwent ASCT during the study period. Median follow-up was 16 months (range 1-182). 42 patients were evaluated with FDG-PET at the end of first line therapy: 25 had negative (60%), and 17 (40%) positive PET scans. In 56 patients response was evaluated with CT scan only, and complete responses or complete responses unconfirmed (CR/CRu) were detected in 29 cases (51%). The 10-year progression free survival (PFS) and overall survival (OS) of the whole patient cohort were 22% and 32% respectively. To determine the prognostic value of FDG-PETend we first assessed the PFS of PET positive vs PET negative patients. Those patients who were PET positive at the end of treatment had a statistically inferior PFS compared to PET negative ones (5-year PFS: 29% vs 47% respectively; p<0.01). These data indicate that although a negative PET after initial therapy is a good prognostic predictor, about half of all patients still relapse after achievement of a negative PET. To better define the prognostic value of FDG-PETend scan in T-cell lymphomas, we restricted our analysis to those patients who obtained a CR after first line therapy, comparing the outcome of PET negative patients (n=25) with the outcome of patients who did not have a PET scan but were deemed in CR after a negative CT scan (n=17). Interestingly we did not observe significant differences between the 2 groups (5-year PFS was 47% in both groups). These data were also confirmed when the analysis was restricted to the PTCL-NOS and ALCL subgroups, suggesting unsatisfactory improvements in the quality of CR assessment with a limited negative predictive value of negative PET scans after first line therapy. In conclusion, although our data confirm the prognostic value of end of treatment FDG-PET scan in T-cell lymphoma, they also highlight the limits of current prognostic stratification tools suggesting that different timing of response evaluation or alternative methods to detect minimal residual disease in those patients in CR after first line therapy are warranted.