DESTINY trial: sustained recurrence free survival in the second stopping year

In the British study of De-Escalation and Stopping Treatment with Imatinib, Nilotinib or sprYcel (DESTINY) the dose of tyrosine kinase inhibitor (TKI) was de-escalated to 50% of the standard dose for 1 year after which treatment was stopped. During EHA 2018 it was shown that during the second stopping year only 5 new recurrences occurred which suggest that de-escalating prior to stopping does not simply delay recurrence.

The DESTINY study included patients who were aged  ≥ 18 years, in first chronic phase, on the same TKI for at least 3 years since original diagnosis (except that 1 change was permitted if intolerant to the initial TKI), and with all BCR-ABL qPCR transcript levels (minimum of 3) < 0.1% in the 12 months before entry. Patients of whom all the BCR-ABL results were <0.01% were assigned to the so called MR4 group, which is studied in many other stopping trials. Patients with ≥1 result between 0.1% and 0.01% were allocated to the MMR but not MR4 group, not previously studied in a stopping trial.

For the 174 patients (imatinib: 148, nilotinib: 16, and dasatinib: 10) that entered the DESTINY trial, TKI treatment was de-escalated to 50% of the standard dose (imatinib 200mg daily, dasatinib 50mg daily or nilotinib 200mg twice daily) for 12 months, after which the treatment was stopped for a further 24 months. Centralised PCR monitoring was carried out 1-2 monthly and molecular recurrence was defined as the first of 2 consecutive samples >0.1%; this required recommencement of the relevant TKI at full dose.

The 174 patients that entered the trial were treated for a median duration of 6.8 years. Previously it was reported that after 24 months on study, molecular recurrence was lower in patients with stable MR4 at entry (29 of 125 patients; 23.2%) than in those in MMR but not MR4 (29 of 49 patients; 59.2%) (p<0.001). During EHA 2018 Clark et al showed that during the subsequent 12 months of complete treatment cessation (i.e. months 25-36 of study), only 5 further recurrences occurred, all in patients with stable MR4 at entry, giving a recurrence free survival (RFS) of 72% (90%CI: 65-79%) at 36 months follow-up in this group of patients. The overall recurrence rate is higher in the MMR but not MR4 group (20 of 36 patients during cessation), resulting in a RFS of 39% (90% CI: 29-52%); p = < 0.001, although no new recurrences have been seen in months 25-36 in this group.

Multivariable Cox proportional hazards modelling revealed that the baseline entry PCR result did not add to the predictive effect on RFS of the PCR pattern in the 12 months prior to trial entry; however the duration of TKI treatment was an additional predictive factor (p = 0.047; HR 0.93), as shown in other studies. The RFS probability remains unrelated to age, gender, performance status or prior TKI (imatinib versus second generation). No progression to advanced phase was seen; 2 deaths occurred due to unrelated causes, and 1 case lost haematological response. All relapsing patients have regained MMR within 4 months of resuming full dose TKI. No difference in RFS was seen between patients with a PCR level <0.0032% (MR4.5) at entry and those not.

The present finding of only 5 recurrences in the second stopping year, thus sustaining the previously reported RFS after 24 months of stopping, suggests that initial de-escalation is not simply delaying recurrence, although the mechanism of its benefit is not yet clear. Possibilities include gradual mobilisation of leukemic stem cells into cycle and/or gradual improvement in the anti-leukemic immune response at a time when TKI is still present. To elucidate the mechanism further study is required.

Reference

Clark R, Polydoros F, Apperley J, et al. Final results of the DESTINY study of de-escalation and stopping treatment in chronic myeloid leukaemia. Presented at EHA 2018; abstract S809.