Imatinib Suspension and Validation (ISAV) Study: Results at 24 Months

Introduction. A substantial proportion of patients (pts) affected by Chronic Myeloid Leukemia (CML) achieve complete negativity in Q-RT-PCR. In this situation, as already demonstrated in other STOP trials, it is possible to safely discontinue imatinib treatment but it is still not clear how to discriminate subjects who will relapse. In fact even undetectable BCR-ABL may not equate to eradication of the disease because of the sensitivity of  Q-RT-PCR. A new diagnostic method, the digital-PCR (dPCR), able to detect 1 BCR-ABL+ cell out of 10⁷ cells, has been developed. Therefore, dPCR by assessing the presence of minimal residual disease with higher sensitivity, could potentially identify pts in whom CML has been eradicated.

Aims. The ISAV study is aimed at validating the capability of dPCR to predict relapses after imatinib discontinuation in CML pts with negative Q-RT-PCR results and to evaluate relapse rate and timing of recurrence, survival and the impact of imatinib treatment on Quality of Life (QoL).

Methods. This study involves 15 sites, 10 in Italy and 1 in each of the following countries: Germany, Spain, The Netherlands, Canada and Israel. CML pts (Chronic or Accelerated Phase) under imatinib therapy since more than 2 years and in complete molecular remission (CMR) were eligible. Patients had to be in CMR for at least 18 months (mts), with a minimum of 3 Q-RT-PCR performed at their own sites. After signing the informed consent, blood samples were obtained for dPCR and the pts discontinued imatinib therapy. Standard Q-RT-PCR is performed monthly (mts 1-6) and then bimonthly for 36 mts to assess the maintenance of the molecular remission. The loss of molecular remission is defined as two consecutive positive Q-RT-PCR tests with at least one BCR-ABL/ABL value above 0.1%. Patients losing molecular remission resume imatinib treatment at the same dosage used before interruption. Patients’ QoL during imatinib discontinuation/resumption is evaluated through the EORTC QLQ–C30 questionnaire.

Results. The study enrolled 112 pts with a median follow-up (FUP) time of 28.0 mts [95% CI: 25.5-30.1]. The 59.3% of pts were male and 37.0% were aged 65 or older; median duration of imatinib treatment was 103.1 mts with median duration of CMR of 25.7 mts before imatinib discontinuation. The cumulative probability of survival is 97.8% [95% CI: 91.4-99.5]. dPCR results showed that 23.1% of pts were positive and 75.9% negative, with a significant Negative Predictive Value ratio (dPCR/Q-RT-PCR) of 1.112 [95% CI: 1.009-1.225]. At 24 mts from imatinib discontinuation, 53 pts (49.1%, 95% CI: 39.3-58.9) of the 108 eligible pts relapsed and resumed imatinib; 73.6% of them relapsed in the first 9 mts and the last relapse occurred 21.8 mts after imatinib discontinuation. A loss of CCyR occurred in 13 pts (34.2% of those tested): 10/13 CCyR losses were recovered, the remaining 3 were not assessed for response. No case of CML progression or resistance to imatinib was observed. After the resumption of imatinib the median time to MMR/CMR was 1.9 [95% CI: 1.2-2.4] mts. Of the 55 not-relapsed pts, 42 (38.9% of the total) regained Q-RT-PCR positivity but never lost MMR. The median time to Q-RT-PCR positivity was 2.9 mts [95% CI: 2.0-3.1] in the relapsed pts and 4.5 mts [95% CI: 2.9-6.9] in pts who developed only PCR positivity. No significant correlation between relapse and previous duration of imatinib treatment, use of interferon, time to CCyR, Sokal score or duration of CMR was identified, while an inverse relationship between pts age and risk of relapse is evident. Moreover, age and dPCR results together can predict the risk of relapse: pts with less than 45 years and with a positive dPCR had the highest risk of relapse (100%) as opposed to pts ≥ 45 years and with negative dPCR (36.1%). The analysis of QoL evidenced a statistically significant improvement in the general well-being and symptoms scales at 1 month after imatinib discontinuation and in particular nausea, diarrhea and fatigue (p<0.01). An inverse and transient trend toward increased pain emerged at mts 1 and 3.

Conclusions. At 45 mts from the beginning of the study, with a median FUP of 28.0 mts, 49.1% of pts relapsed; the majority of relapses developed in the first 9 mts after imatinib discontinuation. Age < 45 years and dPCR positivity are significantly associated with relapses. QoL analysis showed a significant decrease in symptoms after imatinib discontinuation. Funded by Regione Lombardia.