Possibility of a risk-adapted treatment strategy for untreated aggressive adult T-cell leukaemia-lymphoma (ATL) based on the ATL prognostic index: a supplementary analysis of the JCOG9801

JCOG9801, a randomized phase III trial, reported that vincristine, cyclophosphamide, doxorubicin and prednisone (VCAP); doxorubicin, ranimustine and prednisone (AMP); and vindesine, etoposide, carboplatin and prednisone (VECP) (VCAP-AMP-VECP; mLSG15) showed superior clinical outcomes when compared to cyclophosphamide, doxorubicin, vincristine and prednisone every 2 weeks (CHOP-14; mLSG19) in patients with untreated aggressive adult T-cell leukaemia-lymphoma (ATL). To identify patients who require VCAP-AMP-VECP, we conducted a supplementary analysis of JCOG9801. Overall, 105 patients were included and categorized into low- (n = 44), intermediate- (n = 54) and high-risk (n = 7) groups according to the age-adjusted ATL prognostic index (ATL-PI). We excluded the high-risk group due to small numbers of patients. VCAP-AMP-VECP did not show any superior trend for overall survival (OS) in the low-risk group (hazard ratio: 1·04; 95% confidence interval: 0·54-2·04). Better OS was observed in the intermediate-risk group treated with VCAP-AMP-VECP (hazard ratio: 0·65; 95% confidence interval: 0·36-1·19). In the intermediate-risk group, the VCAP-AMP-VECP arm showed higher complete response rates than the CHOP-14 arm (44·0% vs. 13·8%). The VCAP-AMP-VECP arm in both risk groups exhibited grade 4 thrombocytopenia, while grade 4 neutropenia was only observed in the intermediate-risk group. VCAP-AMP-VECP remains suitable for the intermediate-risk group, whereas its benefits appear modest in the low-risk group.