Purpose The significance of positron emission tomography/computed tomography (PET–CT) in identifying patients with lymphoma-associated hemophagocytic lymphohistiocytosis (LAHLH) when pathological evidence is unavailable remains uncertain.MethodsIn this retrospective study, 44 HLH patients who underwent PET–CT before clinical treatment were enrolled, and 18 of them were highly suspected as LAHLH by PET–CT. We compared the PET–CT parameters between confirmed LAHLH and non-LAHLH patients. The efficacy of initial therapies for highly suspected LAHLH patients was analyzed as well.Results We found that the SUVSp, SUVBM, SUVLN, SUVmax, SUVLN/Li, and SUVmax/Li in LAHLH group were significantly higher than those in non-LAHLH group (p = 0.003, p = 0.034, p = 0.003, p < 0.001, p = 0.039, and p = 0.035, respectively). HLH patients with an SUVmax value >5.5, an SUVLN value >3.3, and an SUVSp value >4.8 were more likely to be LAHLH (p < 0.001, p = 0.003, and p = 0.003, respectively). And the incidence of multiple lymphadenopathy with increased FDG uptake or the incidence of multiple bone lesions in LAHLH patients was significantly higher than those in non-LAHLH group (92.9 vs. 35.7 %, p = 0.004; 42.9 vs. 0 %, p = 0.016, respectively). Furthermore, by comparing the efficacy of initial therapies for highly suspected LAHLH patients (n = 18), we indicated that the CR rate was significantly higher in lymphoma-chemotherapy group than in immunosuppressive therapy group (90 and 25 %, respectively; p = 0.013). OS analysis revealed that highly suspected LAHLH patients treated with lymphoma-chemotherapy had better prognosis (264 days) than those treated with immunosuppressive therapy (15 days) (p < 0.0001).ConclusionsWhen pathological evidence is absent, PET–CT may play an important role in identifying HLH patients underlying lymphoma. Once highly suspected as LAHLH by PET–CT, lymphoma-chemotherapies that directly treat the underling lymphoma may have a relatively favorable effect and better clinical outcomes than immunosuppressive therapy.
Read full abstract