Poster Board III-856
Accurate staging is critical in patients with lymphoma. Despite the lack of functional information CT remains the standard imaging technique based on anatomic criteria. PET with FDG provides functional evaluation but lacks of specificity due to the absence of anatomic landmarks. We have prospectively compared the accuracy of combined PET/CT with that of CT and PET alone at initial staging in lymphoma patients. Material and Methods. Patients with newly diagnosed lymphoma were prospectively included in the study. All patients underwent conventional staging studies which included clinical history, physical examination, laboratory work-up, biopsy of enlarged lymph node and of the iliac crest bone marrow, and PET/CT (either with full-dose contrast enhanced CT (FD PET/CT) and with low-dose non-enhanced CT (LD-PET/CT)) before starting therapy. The combined results of the clinical and imaging studies and the results of biopsies of suspected sites involved with lymphoma, when possible, formed the basis of our reference standard. The primary objective of the study was to address the accuracy of the PET/CT examination in clinical staging. For each patient, staging was assessed according to the Ann Arbor classification system on the basis of CT, PET images alone and the PET/CT images. The results were then compared with the true clinical stage based on the reference standard. Secondary objectives were: evaluation of the influence of clinical staging with PET/CT in the treatment approach, assessment of the efficacy of PET in evaluating bone marrow (BM) involvement, and definition of the accuracy of PET/CT imaging in the staging of different histologic subgroups. Results. 108 patients (64 female and 44 male; mean age 50 years; range 18-75 years) were included in the study. 76 patients had NHL and 32 had HL. 36% of the patients (28) had DLBCL whereas 18% (14 patients) had Follicular lymphoma. Other subtypes of NHL accounted for less than 10%. 31 of HL patients had classic variants. True clinical stage for NHL patients was I (7), II (11), III (6) and IV (52), and for HL was II (17), III (4) and IV (10). Agreement in staging was statistically significant between the reference standard and the staging algorithms by CT (k=0.493 p<0.001), LD-PET/CT (k=0.65 p=0.007) and FD-PET/CT (k=0.645 p=0.035). Moreover, we found an almost perfect agreement between LD-PET/CT and FD-PET/CT staging algorithms (k=0.97 p<0.001). PET staging algorithm did not show a statistically significant agreement with the reference standard (k=0.467 p=0.29). Discordant staging by diagnostic CT and reference standard was found in 39 patients, nine of which (8.3%) would have received either more (8) or less (1) aggressive therapy. PET/CT accurate staged those 9 patients allowing a correct therapy. In 5 (4.9%) patients FD-PET/CT staging was discordant with reference standard. FD-PET/CT would assign higher stages due to extranodal findings not confirmed either in biopsy or follow-up. All 5 patients would have received more aggressive therapy. PET efficacy in evaluating BM led to 27% sensitivity (CI 95% 13-42) and 85% specificity (CI95% 77-93) in our study with a likelihood ratio positive (LR+) of 1.86 and LR- of 0.85. PET with FDG accurate detected 88.8% of the patients (96% of DLBCL, 78% of Follicular lymphoma and 96% of HL). Comparing CT, PET and PET/CT staging accuracy for different histologic subtypes, CT accurate staged 67% of DLBCL, 81% of HL and 57% of Follicular lymphoma; PET accurate staged 57% of DLBCL, 78% of HL and 35% of Follicular lymphoma; FD–PET/CT accurate staged 82% of DLBCL, 87% of HL and 86% of Follicular lymphoma. Conclusions. 1.PET/CT is the most accurate imaging modality for initial staging of lymphoma patients in our study. 2.Accurate staging with PET/CT would induce therapy modifications in 8.3% of the patients. 3.FDG-PET is not superior to BM biopsy in detecting lymphomatous infiltration of BM. 4.Combined PET/CT accurate stages more than 80% of patients among the most frequent histologic subtypes (DLBCL, Follicular lymphoma and HL).
No relevant conflicts of interest to declare.
Asterisk with author names denotes non-ASH members.