were respectively six of 13 (46%) for the full dose, six of 15 for the low-dose regimen (40%, P = 1). Fifteen patients (54%) experienced at least one serious AE (SAE). SAEs occurred in seven of 13 patients with the fulldose regimen (54%), and eight of 15 (53%, P = 1) with low doses (see Table 1). After a median follow-up of 11 (range 1-69) months, nine patients were still alive (CR, n = 5, including four patients who received alloHSCT) and 19 had died. The causes of death were progression (n = 12), sepsis (n = 4), heart failure (n = 2) and HHV6B hepatitis (n = 1). I+E treatment was discontinued in all patients. Reasons to stop the treatment were: progression (n = 10, 36%), AE n = 8, 29%), both (n = 1, 4%), PR with another treatment line needed for pretransplant CR (n = 3, 11%), CR/PR allowing maintenance treatment (n = 4, 14%), PR allowing alloHSCT (n = 1, 4%). In this study, I+E showed antitumor activity in these heavily pretreated patients with CTCL, with an ORR of 43%, and no significant difference between the full-dose and reduced-dose regimens in terms of ORR or SAE frequencies, indicating that the reduced regimen may be a suitable option in fragile patients. Eleven patients (39%) achieved an ORR4, lower than the ORR4 of brentuximab vedotin in the ALCANZA trial (56%). 7 However, patients had received a median number of two previous systemic treatments vs. six in our study, and 31% of the patients in the ALCANZA trial had early-stage disease. The ORR of 43% in the present study is quite similar to that of romidepsin (36%). 8 Although the myelosuppressive effects of I+E must be taken into account, neutropenia can be prevented by the use of granulocyte colony-stimulating factor, and I+E can be used as a bridge to alloHSCT or to maintenance treatment. I+E is cheaper than romidepsin or brentuximab vedotin. Some patients experienced long-term remissions, but as described by Hughes et al., 6 most responses were short lived, like other chemotherapy regimens, and SAEs were frequent (54%). This treatment should be reserved for patients with advanced, mostly transformed, and refractory disease, either as a bridge to alloHSCT, or in patients with refractory disease and functional consequences, after having carefully evaluated the risks and benefits, and discussed them with the patient. It must be noted that most of our patients had large-cell transformation (68%) and had received monochemotherapy (mostly gemcitabine, 79% or liposomal doxorubicin, 75%) before I+E. In conclusion, I+E is a valuable therapeutic option in patients with advanced, refractory CTCL.