Patients who suffer from chronic myeloid leukemia are treated by tyrosin kinase inhibitors (TKI) saying imatinib, nilotinib, dasatinib, bosutinib and ponatinib. These drugs are highly efficient with excellent response allowing some patients to definitely stop their cancer treatment. However, in 30% of cases, when the treatment is stopped, pains could arise in shoulders, hips, joints… These symptoms occurring after the withdrawal of a drug are odd and biologically unexplained so far. This study seeks to discover the biological factors behind these symptoms called 'TKI withdrawal syndrome' by the scientific community.

The success of TKI in the treatment of chronic myeloid leukemia is undeniable. Indeed, since the advent of imatinib in 2000 and the subsequent 4 other TKIs, this once fatal disease has become just another chronic disease with patient life expectancy close to the general population. Lifelong TKI therapy offers an 'operational cure' which means quiescent BCR-ABL leukemic stem cells are not eradicated but only controlled.

This statement was challenged in 2008 by the French 'STIM' study in which patients with undetectable BCR-ABL were proposed to discontinue their imatinib: 40% of these patients successfully achieved treatment free remission (TFR) with no recurrence of the disease. These figures are confirmed by the Australian 'TWISTER' study. The resumption of Imatinib therapy was triggered by the positivity of BCR-ABL. In the A-STIM study, the loss of major molecular response is the main criteria to resume imatinib. Accordingly, TFR was a success in 60% of patients. Similar figures was published with second generation TKIs with the same criteria.

In the light of these studies, it is clear that the fields have moved to a more ambitious goal than the operational cure, without the need for remaining on TKI.

Paradoxically, a new syndrome emerged from the Sweden group in 2014: 'the TKI withdrawal syndrome. It consisted of musculoskeletal pains resembling the Polymyalgia Rheumatica in CML patients after discontinuation of Imatinib. It occurred within 6 weeks in 30% of the patients who went on stopping imatinib. Patients with severe pain responded to 10-20 mg of prednisolone. Similar manifestations were described in the European 'EuroSki' study. No explanation was found.

It is well known that TKIs have off-target effects. In addition to BCR-ABL, they inhibit c-Kit, SRC, PDGFR, BTK, TEC, NFKB pathways resulting in anti-inflammatory effects. They can also cause disturbances in electrolyte balance (hypophosphatemia) and bone metabolism. Some research teams have correlated the success of TFR to the increase in cytotoxic NK cells and production IFN and TNFα.

Is the TKI withdrawal syndrome a rebound phenomenon from these off-target effects with excessive release of some cytokines? Of note, an Italian team showed in 4 out of 8 patients with WS an increase of PDGFβ (in 1 patient there was also increase in IL6, TGFβ and VEGF). But no tests were performed in the other patients who stopped TKI and did not presented with the WS as comparison.

Here, the proposed study concerns any CML patient candidate for TKI discontinuation. Will prospectively be checked a number of cytokines and biological factors susceptible to contribute to the clinical features taking into account the off-target effects of TKIs. Therefore, the investigators could compare the variation in a potentially involved cytokine within patients with or without withdrawal syndrome. In addition, the study will prospectively provide a broader description of TKI WS using Brief pain Inventory (BPI) questionnaire.

Eventually, cytokines, inflammatory and bone markers will be dosed at diagnosis and every month during 3 months corresponding to the timeframe in which the withdrawal syndrome is due to occur. A variation of 20% (increase or decrease) in whatever marker(s) will be considered significant if it is (or they are) observed in the group with "TKI withdrawal (TKI WS)" but not in the group without "TKI WS". Pathophysiological explanation of this syndrome could be assumed later taking into account the clinical presentation partly explored by the pain questionnaires

Inclusion Criteria:

• Adult CML patient candidate for TKI discontinuation.
• Performans status : 0-2
• Patient provided with information

Exclusion Criteria:

• Patient unable to fulfill the questionnaires
• Patient suffering from demantia
• Patient treated with any cytokine antagonist

This study concerns patients suffering from Chronic myeloid leukemia treated with TKI (Imatinib, Nilotinib, dasatinib, bosutinib or Ponatinib) for at least 5 years with deep molecular response (MR4) for at least 2 years who agree to stop their drug. Other situation of TKI discontinuation. Five cities are participating in the study: CHU de Caen, CHU d'Amiens, CHU de Clermont-Ferrand, Centre Henri Becquerel de Rouen et le CH de Valenciennes.

Not apllicable