| Background and purposeAngioimmunoblastic T-cell lymphoma(AITL),originated from follicular helper T cells,is a kind of peripheral T-cell lymphoma(PTCL)with special pathological and clinical characteristics.AITL is more common in the elderly,with an aggressive course,poor prognosis and the median overall survival(OS)of less than 3 years.At present,the prognostic indicators of AITL are still largely controversial.Some clinical characteristics,such as thrombocytopenia,complicated serous cavity effusion,elevated serumβ2 microglobulin level,elevated Ig A level and extranodal involvement at>1 site are considered to be important adverse prognostic factors for AITL.Similarly,the use of the International Prognostic Index(IPI)and the Prognostic Index for peripheral T-cell lymphoma(PIT)of PTCL has been limited in assessing the prognosis of AITL,and there have been different models for evaluating the prognosis of AITL established at home and abroad.This study retrospectively analyzed 75 patients of AITL from the Affiliated Cancer Hospital of Zhengzhou University,studied the clinical characteristics and prognostic factors of AITL patients,and compared the evaluation effects of different prognosis evaluation models in AITL patients,so as to provide reference for the diagnosis,treatment and prognosis evaluation of AITL.Methods1.The data of 75 newly diagnosed AITL patients admitted to the Affiliated Cancer Hospital of Zhengzhou University from September 2011 to April 2020 were collected.All patients underwent histomathological and immunohistochemical examinations according to the lesion invasion site,and all patients met the WHO diagnostic criteria for hematopoietic and lymphoid tissue tumors in 2016,and all patients underwent imaging and bone marrow cytology examinations to accurately assess Ann Arbor staging.The therapeutic effect was evaluated using Cheson’s lymphoma evaluation criteria.The basic data of patients included gender,age,etc.,and laboratory examination indexes included blood routine,LDH,β2 microglobulin,serum Ig A,albumin,Ki-67,etc.The conditions of the patients in this study were evaluated by IPI,PIT,PIAI and ATPI scores.2.SPSS25.0 software was used to process and analyze the data.Kaplan-Meier method was used to draw the survival curve and Log-rank test was applied.Cox regression model was used for multivariate analysis.Receiver operating characteristic curve(ROC)was used to evaluate the predictive value of different prognostic stratification models.P<0.05 indicated statistically significant difference.Results1.Among the 75 patients,39 were male patients and 36 were female patients.The median age was 62 years and the age of onset was 21~83 years.17patients(22.7%)presented with rash/pruritus initially,40 patients(53.3%)with B symptoms,35 patients(46.7%)with serous cavity effusion,73 patients(97.3%)with Ann Arbor stage III~IV,13 patients(17.3%)with ECOG PS score 2~4 points,10patients(13.3%)with bone marrow involvement,and 27 patients(36.0%)with extranodal involvement at>1 site.In the lab,15 patients(20.0%)patients presented with white blood cell count>10×10~9/L before treatment,42 patients(56.0%)with hemoglobin<120g/L,32 patients(42.7%)with platelet count<150×10~9/L,42patients(56.0%)with LDH level>245u/L,30 patients(40.0%)with albumin level<35g/L,35 patients(46.7%)withβ2-microglobulin level>4 mg/L,and 38 patients(50.7%)with Ki-67>50%.And in 61 patients,8 patients(13.1%)presented with Ig A level>400mg/d L.2.Among the 75 patients,in the IPI score,9 patients(12.0%)were evaluated with 0~1 point,29 patients(38.7%)with 2 points,23 patients(30.7%)with 3 points,and 14 patients(18.7%)with 4~5 points.In the PIT score,10 patients(13.3%)were evaluated with 0 point,38 patients(50.7%)with 1point,17 patients(22.7%)with 2points,and 10 patients(13.3%)with 3~4 points.In the PIAI score,29 patients(38.7%)were evaluated with 0~1 point,21 patients(28.0%)with 2 points,14 patients(18.7%)with 3 points,and 11 patients(14.7%)with 4~5 points.Among the 61 patients,in the ATPI score,24 patients(39.3%)were evaluated with 0~1 point,15 patients(24.6%)with 2 points,14 patients(23.0%)with 3 points,and 8 patients(13.1%)with 4~6points.3.Among the 75 patients,20 patients were treated with CHOP regimen(cyclophosphamide,doxorubicin,vincristine,prednisone)in the first-pass course,2 patients with CHOP regimen combined with chidamide,31 patients with CHOPE regimen(cyclophosphamide,doxorubicin,vincristine,prednisone,etoposide),6 patients with CHOPE regimen combined with chidamide,7 patients with EPOCH regimen(etoposide,prednisone,vincristine,cyclophosphamide,doxorubicin),4 patients with DICE regimen(dexamethasone,ifosfamide,cisplatin,etoposide),3 patients with GDP regimen(gemcitabine,cisplatin,dexamethasone),and 2 patients with oral chemotherapeutic agent.The relapsed or refractory patients were treated with DICE,GDP,GLD(gemcitabine,oxaliplatin,dexamethasone),ESHAP(etoposide,cisplatin,cytarabine,prednisone),hyper-CVAD(high dose of cyclophosphamide,vincristine,doxorubicin,dexamethasone),etc.Of all patients,38patients received chidamide in treatment.All 75 patients were evaluated.After holistic treatment,22 patients(29.3%)achieved complete remission(CR)and 27patients(36.0%)achieved partial remission(PR).The overall response rate(including CR and PR)was 65.3%.In first-line treatment,in patients treated with CHOP regimen,CR and PR rates were 20.0%and 35.0%,and ORR 55.0%,and in patients treated with CHOPE regimen,CR and PR rates were 38.7%and 35.5%,and ORR74.2%,respectively.There was no significant difference in ORR between CHOP and CHOPE(P=0.156).4.The median time of OS was 27 months,and the median time of PFS was 10months.The 3-year OS rate was 43.3%,and the 5-year OS rate was 28.9%.The 3-year PFS rate was 27.7%,and the 5-year PFS rate was 22.2%.Univariate analysis showed that age,B symptoms,bone marrow involvement,albumin level,and ECOG PS score were prognostic factors affecting patients’OS,while Age,B symptoms,bone marrow involvement,serous cavity effusion,β2-microglobulin level,albumin level,ECOG PS score and initial treatment regimens were prognostic factors affecting patients’PFS.Multivariate analysis concluded that age>60 was the independent risk factor for OS.ECOG PS score>1 was the independent risk factors for PFS.5.The scores of the four prognostic stratification models were taken as the test variables,and the OS status of AITL patients was taken as the state variable,and the value of the state variable was defined as death.The area under ROC(AUC)of PIT score was 0.705(95%CI,0.589~0.820),which was better than the other three prognostic stratification models.The scores of the four prognostic stratification models were used as test variables,and the PFS status of AITL patients was used as the state variable,and the value of the state variable was defined as progression,recurrence or death.The AUC value of PIAI score was 0.726(95%CI,0.613~0.839),which was better than that of the other three prognostic stratification models.Conclusion1.AITL tends to occur in the elderly,and most patients have a late stage of diagnosis,short median OS time and PFS time,and poor prognosis.2.Age,B symptoms,bone marrow involvement,albumin level,and ECOG PS score were prognostic factors affecting OS,among which age>60 was the independent risk factor for OS.Age,B symptoms,bone marrow involvement,serous cavity effusion,β2-microglobulin level,albumin level,ECOG PS score and initial treatment regimens were prognostic factors affecting PFS,among which ECOG PS score>1 was the independent risk factors for PFS.3.If the AUC value was used as the evaluation standard,PIT score had a better predictive effect on the OS status of AITL patients,and PIAI score had a better predictive effect on the PFS status of AITL patients. |