With a median follow-up of more than 6 years after treatment discontinuation, the STIM1 study demonstrates that [imatinib] can safely be discontinued in patients with a sustained deep molecular response with no late molecular recurrence.— Gabriel Etienne, MD, PhD, and colleagues
As reported in the Journal of Clinical Oncology by Gabriel Etienne, MD, PhD, of the Institut Bergonié, Bordeaux, and colleagues, long-term follow-up in the French Stop Imatinib (STIM1) study indicates that imatinib can be safely stopped in patients with chronic myeloid leukemia (CML) with undetectable minimal residual disease sustained for at least 2 years.1
In STIM1, 100 patients aged ≥ 18 years with Philadelphia chromosome–positive CML in chronic phase at diagnosis who had been treated with imatinib at any dose for at least 3 years and were in sustained undetectable minimal residual disease for at least 2 years were enrolled at 19 French centers. Sustained undetectable minimal residual disease was defined as at least six undetectable BCR-ABL transcripts on quantitative reverse transcriptase–polymerase chain reaction.
After discontinuation of imatinib, molecular follow-up was performed once a month during the first year, once every 2 months during the second year, and once every 3 months thereafter. Molecular recurrence was defined as at least two positive reverse transcriptase–polymerase chain reaction results showing significant increase (by 10 times—ie, 1 log) at two consecutive measurements or loss of major molecular response. Of the 100 patients, 50 had previously received interferon-α treatment.
An interim analysis showed that molecular remission was maintained in approximately 40% of patients after stopping imatinib despite the persistence of leukemic cells at low levels in most patients. No clinical disease progression was observed after imatinib was stopped, and retreatment in patients with molecular recurrence resulted in a second, deep molecular response. The current analysis presents outcomes in the study population after a median follow-up of 77 months (range = 9–95 months) after discontinuation of imatinib.
In total, 61 patients (61%) lost undetectable minimal residual disease after a median of 2.5 months (range = 1–22 months), and 1 patient died with undetectable minimal residual disease at 10 months; 49 molecular recurrences (80% of patients with relapse) occurred between months 1 and 3; 9 (15%), between months 4 and 7; and 3 (5%), between months 18 and 22, with no molecular recurrences observed thereafter. Median BCR-ABL:ABL International Scale ratios at the time of first and second consecutive assessments were 0.008% and 0.032%.
A total of 17 patients had loss of major molecular response at the time of confirmed molecular recurrence, with a median BCR-ABL:ABL International Scale ratio of 0.28%. The molecular recurrence–free survival was 43% (95% confidence interval [CI] = 33%–52%) at 6 months, 40% (95% CI = 30%–49%) at 18 months, and 38% (95% CI = 29%–47%) at 60 months. Among the 4 patients with molecular recurrence more than 6 months after stopping imatinib (at 7, 18, 20, and 22 months), 2 had a low and 2 had an intermediate Sokal risk score and 3 had received interferon-α for 4, 16, and 53 months.
Effect of Retreatment
Treatment was restarted at a median of 2.1 months after molecular recurrence (range = 0.7–15.6 months) in 57 of 61 patients with molecular recurrence, consisting of imatinib in 56 patients and dasatinib (Sprycel) in 1 patient. Of the four patients not restaring treatment, three refused to restart and one received treatment for a second neoplasm.
A second undetectable minimal residual disease was achieved in 55 of 57 patients (96%) restarting treatment at a median of 4.3 months (range = 1.5–21 months). Median time to second undetectable minimal residual disease was 4.2 months (range = 1.5–15.8 months) in those with loss of major molecular response at the time of molecular recurrence and 5 months (range = 2.3–21 months) in those without loss of major molecular response at the time of molecular recurrence.
Over a median molecular follow-up of 73 months, none of the patients with molecular recurrence exhibited CML progression. Four patients died due to CML-unrelated causes; two died of second malignancies (pleural mesothelioma, metastatic gastric adenocarcinoma) and two died of other causes (acute renal failure and cerebral hemorrhage).
On multivariate analysis including age, sex, Sokal and Eutos risk scores, previous interferon-α therapy, duration of imatinib therapy, time to undetectable minimal residual disease, and undetectable minimal residual disease duration until imatinib discontinuation, only Sokal risk score (subdistribution hazard ratio [SHR] = 2.22 for high vs low plus intermediate risk, P = .024) and imatinib treatment duration (SHR = 0.54 for less than median of 58.8 months vs ≥ median, P = .024) were significant independent prognostic factors for molecular recurrence. The cumulative incidence of molecular recurrence was significantly lower among the 62 patients with imatinib treatment duration of ≥ 54 months vs 37 patients with imatinib treatment duration < 54 months (P = .009) and significantly lower among 51 patients with imatinib treatment duration ≥ 54 months plus low or intermediate Sokal risk score compared with 37 patients with imatinib treatment duration < 54 months plus low or intermediate Sokal risk score and 11 patients with imatinib treatment duration ≥ 54 months plus high Sokal risk score (P = .001 overall).
The investigators concluded: “With a median follow-up of more than 6 years after treatment discontinuation, the STIM1 study demonstrates that [imatinib] can safely be discontinued in patients with a sustained deep molecular response with no late molecular recurrence.”
They noted: “We can now hypothesize that the proportion of patients who will achieve a sustained deep molecular response and will be eligible for treatment discontinuation according to the STIM1 criteria will increase over time. Whether a shorter exposure to second-generation [tyrosine kinase inhibitors] together with the achievement of a sustained deep molecular response will be associated with similar rates of molecular recurrence remains a key issue…. With respect to the heterogeneous characteristics of the patients, Sokal risk score and [imatinib] duration before treatment discontinuation have to be considered from the perspective of [imatinib] discontinuation. These results make treatment-free remission legitimate as a criterion of treatment evaluation in the future.” ■
Disclosure: The study was supported by grants from the French Ministry of Health Programme Hospitalier de Recherche and the Institut National du Cancer. For full disclosures of the study authors, visit www.jco.ascopubs.org.
We could say that next to being able to give tyrosine kinase inhibitors to patients with CML to improve their life expectancy, the best thing we can do now is to take them away…sometimes.— Jorge Cortes, MD
Sir Donald Munger: “You have been on holiday, I understand....!-->!-->