ASH-Report #1: Stopping CML treatment for therapy-free remission

EUROSKI[Translations: German - Russian]

Treatment-free remission, or stopping treatment in stable long-term deep molecular remission, has probably been the most reported topic at this year’s ASH.

The European Stop Tyrosine Kinase Inhibitor study (EURO-SKI), being the largest ongoing STOP study, has just completed recruitment of 700 participants. An interim analysis of the first 200 patients from 8 European countries with a follow-up of minimum 12 months was now presented. The study aims to define prognostic markers to increase the rate of patients in durable deep molecular response after stopping TKI treatment, evaluation of molecular monitoring procedures, and the assessment of quality of life.

Other than earlier studies like the French STop IMatinib (STIM) studies or the Novartis ENEST Stop trials, this study features slightly more relaxed eligibility criteria, with requiring only MR4 (<0,01% BCR-ABL) for more than 1 year on a minimum 3 years of  TKI treatment, and a restart of therapy when major molecular response (MMR, BCR-ABL >0,1%) is lost. In comparison: The first French STIM studies required patients to have a deeper molecular remission (MR4.5, <0,0032% BCR-ABL) and required immediate re-initiation of therapy when MR4.5 was lost. The 200 patients on EURO-SKI had a median treatment duration of 8 years (range 3-13 years), and 5 years of MR4 before stopping TKI (range 1-12 years).

As a result of stopping, 111 of the 200 patients (56%) remained therapy-free after stopping TKIs , and 89 relapsed. Like in the STIM study, most relapses occurred very quickly within the first 6 months. Amongst those 89 patients that relapsed, 76 regained MMR and 70 returned to MR4 by the time of the report. There have been no progressions to advanced phases. In terms of prognostic factors for relapse, the duration of prior TKI treatment, and having had MR4 for more than 5 years before stopping seemed to have a positive influence, while the depth of molecular response (MR4 vs MR4.5 vs MR5) prior to stopping did not influence the risk of relapse. 31 of the 200 patients experienced stopping-related side effects, mostly pain in muscles, joints or bones, but also sweating, skin problems, depressive episodes, tiredness or weight loss. None of the side effects were very severe (grade 4). Debates are still going on how to manage and prevent those “therapy withdrawal symptoms”. The estimated drug-related savings to healthcare systems in the 8 EURO-SKI countries was estimated to be 7 million Euro.

The study will further investigate prognostic factors, quality of life and other factors that might influence the feasibility to stop and the more interesting findings will follow - of note this year's ASH report was only an interim analysis of the first 200 of in total 700 patients.

2GSTOPThe presentation of the STOP-2G-TKI study, conducted by the French CML study group, recruited 52 patients. The inclusion criteria of this study were a bit more strict, requiring prior 2nd generation TKI treatment (Dasatinib, Nilotinib, Bosutinib) with a MR4.5 remission for at least 2 years. 52 patients with a median of 6.5 years of TKI treatment and a median duration of MR4.5 of 2.3 years have been evaluated in this ASH report, and they were followed 12-60 months. Like in EURO-SKI, this study defined the loss of MMR (BCR-ABL 0.1%) as criterium for reinitiation of therapy. 24 of the 52 patients (46%) lost MMR and had to restart treatment.

However, one issue led to debates at the end of the presentation: While most relapses occurred very quickly (median 3.7 months), one relapse occurred 24 months and one 37 months after stopping treatment. The patient who relapsed in month 37 was treated with a 2nd generation TKI as a first line treatment and remained in very deep molecular remission until just before the relapse. In conclusion, it seems obvious that patients need to be in regulator monitoring even years after stopping treatment. The good news though is that the patient, after restarting treatment, got back to deep molecular response. Overall, none of the patients progressed to advanced CML phases. All patients regained MMR within 1-6 months and MR4.5 within 1-21 months.

Certainly the prognostic factors around stopping treatment remain unclear: EURO-SKI had revealed that duration of prior TKI treatment and the duration of MR4 before stopping seemed to increase the chance for treatment-free remission. The French STOP-2G-TKI study found no association with age, gender, Sokal risk at the time of diagnosis, prior Interferon therapy, duration of TKI treatment, duration of MR4.5 and type of TKI prior, while suboptimal response or resistance to Imatinib reduced the probability of successful stopping under Dasatinib/Nilotinib therapy. The third STOP study presented at this ASH, the Imatinib Suspension and Validation Study (ISAV), demonstrated similar relapse rates like EURO-SKI and STOP-2G-TKI, but found that age <45 years or detection of BCR-ABL in a more sensitive “digital PCR” test was increasing the risk of relapse. And finally, the Korean Imatinib Discontinuation Study (KIDS) concluded that both duration of deep molecular response and Imatinib treatment were the most important predictors for successful therapy stop. There is no common ground in prognostic factors yet.

To better understand whether specific characteristics of a patients’ individual immune system has an influence on the risk of relapse after stopping, a EURO-SKI sub-study investigates on how the immune system keeps a residual CML under control using T-Cells and “Natural Killer cells” (NK-cells), and whether TKI influence the immune system. The findings suggest that those patients that relapse have a lower proportion and absolute number of NK-cells at the time of TKI discontinuation. The presentation raised the question whether the measurement of those might help to predict relapse, and suggests to further investigate the biological mechanisms and patient-specific factors.

Summary and thoughts

In the education session, CML expert Dr Michael Mauro concluded that achieving “treatment-free remission seems feasible” and that “concern over long-term risk of attempting treatment free remission has not increased”, albeit in a very tightly monitored framework.

This is supported by the data from four different STOP studies which were presented at ASH this year. Tight monitoring within a clinical study, very sensitive PCR diagnostic in the best labs with a reliable sensitivity to measure MR4.5, and immediate re-initiation of therapy when PCR rises above major molecular response (MMR), seems to provide a considerably safe framework to stop treatment. Depending on study and re-start criteria, the proportion of patients remaining in deep molecular remission varies between 40-60%. Both the ENESTnd studies (Nilotinib in first line) and DASISION studies (Dasatinib in first line) suggest more than half of patients on 2nd generation TKI achieve MR4.5 within a couple of years, and only around half of them would relapse when stopping treatment. This means approximately every fourth CML patient may be able to stop CML treatment after years of adherent treatment – and still remain in stable, therapy-free remission. 

However, the late relapse in the STOP-2G-TKI study in month 37 suggests that even years after maintaining therapy-free remission, regular molecular monitoring and quick re-initiation of therapy in case of loss of MMR may be essential. None of the patients who relapsed on the four studies has progressed, but clearly only within a framework of tightly controlled clinical studies with very frequent and sensitive PCR monitoring and immediate TKI restart on loss of MMR. There is yet no way to predict which patient may stop treatment without experiencing a relapse. Understanding the mechanisms why some patients relapse under minimal residual disease while others don’t is still in infancy. Studies like EURO-SKI and their sub-studies will help to understand these points. Until we have the results and until standardised diagnostic to reliably detect MR4.5 and early molecular relapse is more widely available, experts stick to their recommendation at ASH not to stop treatment outside of clinical trials. 

See our other detailed reports from ASH 2014:

Related ASH abstracts:

See the ASH-Abstracts database:

  • Sunday Presentation #151: Interim Analysis of a Pan European Stop Tyrosine Kinase Inhibitor Trial in Chronic Myeloid Leukemia : The EURO-SKI  study
  • Monday Poster #4547: Attitudes and Perceptions of Patients (pts) with Chronic Myeloid Leukemia in Chronic Phase (CML-CP) Toward Treatment-Free Remission (TFR)
  • Tuesday Presentation #812: Early Disease Relapse after Tyrosine Kinase Inhibitor Treatment Discontinuation in CML Is Related Both to Low Number and Impaired Function of NK-Cells
  • Sunday Poster #3155: Results from the Korean Imatinib Discontinuation Study (KIDS): Updated Data with 15-Month Median Follow up
  • Tuesday Presentation #811: Dasatinib or Nilotinib Discontinuation in Chronic Phase (CP)-Chronic Myeloid Leukemia (CML) Patients (pts) with Durably Undetectable BCR-ABL Transcripts: Interim Analysis of the STOP 2G-TKI Study with a Minimum Follow-up of 12 Months
  • Tuesday Presentation #813: The Risk of Relapse in CML Patients Who Discontinued imatinib Can Be Predicted Based on Patients Age and the Results of dPCR Analysis
  • Saturday Poster #1797: BCL2L11 (BIM) Deletion Polymorphism Is Associated with Molecular Relapse after ABL Tyrosine Kinase Inhibitor Discontinuation in Patients with Chronic Myeloid Leukemia with Complete Molecular Response
  • Monday Poster #4561: Achieving the Deep Molecular Response Levels Required for an Imatinib Discontinuation Trial Is Strongly Associated with the BCR-ABL Level at the First Qualifying Timepoint
  • Monday Poster #4553: Long-Term Follow-up Results after Imatinib Discontinuation in Chronic Myeloid Leukemia Patients with Undetectable Minimal Residual Disease

Written by Jan Geissler and Giora Sharf, CML Advocates Network, 12 Dec 2014


Please donate!

Please donate!


Tweet Feed

@jlsteeg Thank you so much for sharing our study! For more info please visit #CML Education & Science Adherence page

Thursday, 23 March 2017 in reply to jlsteeg

Factors influencing adherence in CML Results of a patient-driven survey of 2546 patients in 63 countries.

Retweeted Thursday, 23 March 2017

Call for Abstracts: CML Advocacy Best Practice presentations at CML Horizons 2017.Submit your abstract here……

Wednesday, 22 March 2017