Abstract

Imatinib is remarkably effective in treating patients with newly diagnosed chronic myeloid leukemia (CML) in chronic phase (CP) but somewhat less effective in treating patients who have previously received interferon-alfa (IFN), most of whom can be classified as being in ‘late CP’. For such patients who have failed IFN, a proportion achieve complete cytogenetic responses (CCyR) and subsequently major molecular responses (MMolR) on imatinib, but the durability of these responses is not yet established. We analyzed long-term outcomes for 216 consecutive CML-CP patients who started on imatinib after failing IFN at our institution. Their Sokal risk score was ‘low’ in 58 (27%) patients, ‘intermediate’ in 75 (35%) and ‘high’ in 83 (38%). At the time of starting imatinib 73 (34%) patients were IFN intolerant, 40 (18%) were hematologically resistant to IFN and 103 (48%) were cytogenetically resistant; of this last group 58 (27%) had primary resistance and 45 (21%) secondary resistance. The median interval between diagnosis and start of imatinib was 38 months (range 6 to 217). Ninety-two patients (42.6%; 95CI, 36.0–49.5%) achieved CCyR during follow-up; the estimated cumulative incidence of CCyR at 5 years was 46.8% (95CI 40.0–53.7%). Forty-five patients (20.8%; 95CI, 15.6–26.9%) achieved a MMolR; the estimated cumulative incidence of MMolR at 5 years was 23.7% (95CI 16.6–32.8%). The independent factors predicting achievement of MMolR were prior response to interferon and favorable Sokal category: the relative risks (RR) for achievement of a MMolR response were 3.34 for patients with secondary cytogenetic resistance to IFN (p< 0.0001) and 0.5 and 0.2 for the Sokal intermediate and high risk groups respectively (p=0.012). For the 45 patients who achieved a MMolR the median follow-up was 68 months (range, 32–85 months); 24 (53%) patients achieved a 4-log reduction in the BCR-ABL transcript level, and in 10 (22%) cases the transcripts became undetectable. By intention-to-treat analysis the estimated progression-free survival (PFS, defined as loss of complete hematologic response or progression to advanced phase) for this group at 72 months was 100%. At latest follow-up 7 patients (16%) had lost their MMolR but only 2 (4%) of these had lost their CCyR. When comparing those 45 patients with 76 CML-CP patients who received front-line imatinib and achieved a MMolR (out of 207 patients at our institution), we found no differences in terms of cumulative loss of CCyR (p=0.60) or of MMolR (p= 0.94), suggesting a comparable durability of the responses in these two patient groups. In conclusion, although patients who receive imatinib in late CP generally fare worse than patients starting imatinib soon after diagnosis, patients in the two groups who achieve a MMolR have equivalent outcomes.

Author notes

Disclosure: No relevant conflicts of interest to declare.