Introduction: Successful TKI discontinuation is feasible in approximately 40-60% of CML patients with stable and deep molecular response, but the persistence of quiescent CML stem cells (LSC) may be responsible for relapse. Pioglitazone, a PPARgamma agonist may reduce STAT5 activity and cause an erosion of CML LSC.

Objectives: to evaluate safety and efficacy of pioglitazone in treatment-free remission (TFR) after imatinib discontinuation in patients with CML treated with pioglitazone for 3 months before imatinib discontinuation.

Patients and methods: prospective, phase II trial. Inclusion criteria: CML in chronic phase, treated with imatinib for at least 3 years, with CCyR and stable deep molecular response (MR4.5) for at least 2 years. Patients received pioglitazone 30 mg/day, orally, for 3 months before imatinib discontinuation. BCR-ABL levels were measured by real-time quantitative PCR monthly for 12 months, every two months for 12 months, and then every 3 months during the subsequent follow-up. Imatinib was reinitiated at molecular relapse, defined as any single sample with a value of >0.1% (loss of major molecular response - MMR) or two consecutive positive samples with a value >0.01% (loss of MR4.0). Treatment-free remission (TFR) was calculated from imatinib discontinuation until molecular relapse, resumption of imatinib by any cause, progression to advanced phases or death to any cause.

Results: Recruitment is ongoing. Between June 2016 and July 2017, 21 chronic phase CML patients were included. Median age was 48 years (23-65) at CML diagnosis and 56 years (29-76) at trial initiation; 52.4% male; Sokal: 62.5% low risk, 25% intermediate, 12.5% high risk; 14% treated previously with IFN. Two patients had diabetes and were using hypoglycemic drugs. The median time between diagnosis and imatinib start was 2 months (1-39); the median duration of imatinib treatment was 8.2 years (4-12). Nineteen patients received pioglitazone for 3 months and discontinued imatinib and two are still using both. There were no grade 2 or 3 adverse events related to pioglitazone; one patient did not complete the first month of pioglitazone treatment due to worsening of edema and went to discontinuation phase. The median follow-up of the 20 patients that discontinued imatinib was seven months (0-11). One patient dropped out of the trial and was censured in the analysis. Seven/18 patients (38%) presented newly or worsening musculoskeletal pain or arthralgia grades 1-2 occurring within 1-6 months after imatinib discontinuation. The two diabetic patients presented increase in glucose levels (grade 2 and 3), seven and eight months after imatinib discontinuation. There was no loss of hematological response or transformation to advanced phases. However, 4/18 (22%) patients (all male) reinitiated imatinib after molecular relapse: 1 lost MMR and CCyR response; 2 lost MMR, one had confirmed loss of MR4. Two of them recovered RM4.0 in 2 and 3 months after imatinib reintroduction. TFR was 74% (95% CI: 52-96%) in 9 months. TFR was higher in female patients (100% vs. 45%, P=0.019).

Conclusion: Pioglitazone was well tolerated in the majority of the patients. The rate of withdraw syndrome was similar to the reported in other discontinuation trials. Increase in glucose levels was observed in two patients with diabetes after imatinib discontinuation. TFR was higher in female patients. A longer follow-up is necessary to evaluate the duration of TFR.

This trial is registered in ClinicalTrials.gov (NCT02852486)


Pagnano: Bristol-Meirs Squibb: Consultancy, Speakers Bureau; Roche: Speakers Bureau; Amgen: Consultancy; Novartis: Consultancy.

Author notes


Asterisk with author names denotes non-ASH members.