Second-Line Treatment with Imatinib (IM) after Crossover from Interferon Alfa Plus Cytarabine Yields High Rates of Durable Response: Results from the International Randomized Study of Interferon vs STI571 (IRIS).

Imatinib (IM) and interferon alfa plus cytarabine (IFN+Ara-C) were compared in the phase 3 IRIS trial, which enrolled 1106 patients (pts) with newly diagnosed CML-CP. Pts were allowed to cross over to the alternative treatment arm provided intolerance to treatment was indicated and/or efficacy landmarks were not met. All crossover requests required approval by the Study Management Committee according to strict criteria. Reasons for crossover were amended during the trial to include reluctance to continue IFN+Ara-C. The status of the 553 pts randomly assigned to receive 1st-line IFN+Ara-C and the reasons for crossover to IM are summarized in Table 1.

The median time on IFN+Ara-C treatment before crossover was 9 months (mos), the median time since diagnosis was 13 mos at start of 2nd-line IM. The median time on 2nd-line IM was 45 mos (average 40 mos), with a maximum of 63 mos as per data cutoff 31-Jan-06. About one third of pts discontinued 2nd-line therapy: reasons were unsatisfactory therapeutic effect (13%), adverse events (4%), death (1%) and others (12%) including withdrawal of consent, BMT, loss to follow-up. A total of 251 pts remain on 2nd-line IM.

Among 359 pts treated with 2nd-line IM, the best observed CHR rate was 93%, the best observed MCyR rate was 86%, and the best observed CCyR rate was 80%. The CCyR rate was 95% in pts who were reluctant to continue on IFN+Ara-C arm, 82% in intolerant pts, 78% in pts with lack of response on IFN+AraC and 71% in pts who progressed on 1st-line IFN+Ara-C. After 18 mos of 2nd-line IM therapy, the estimated rate of freedom from progression to AP/BC was 94%, and the estimated overall survival rate was 97%. These data are consistent with a previous trial of IM in pts with CML-CP after failure of prior IFN in which an estimated 89% of pts were free from progression to AP/BC, and an estimated 95% were alive, after 18 mos (Kantarjian NEJM 2002). In this trial, median time from diagnosis and median duration of prior IFN therapy were 34 and 14 mos, respectively. Table 2 compares 1st-line and 2nd-line IM results in the IRIS trial.

The 48-mos rates of freedom from progression to AP/BC (overall survival) were 97% (96%) in pts who were reluctant to continue on IFN+Ara-C arm, 93% (90%) in intolerant pts, 90% (88%) in pts with lack of response on IFN+Ara-C and 80% (85%) in pts who progressed 1st-line IFN+Ara-C.

The adverse event profile was similar between 1st- and 2nd-line IM therapy. CML-CP pts treated with 2nd-line IM after IFN+Ara-C achieved high response rates that are durable; the vast majority of pts remain free from progression to AP/BC after median follow-up of nearly 4 years. However, the overall efficacy results were better for 1st-line IM pts in newly diagnosed CML-CP.