Treatment-Free Remission in CML Patients Discontinuing Post-First Line Therapy

Background: Recommendations for treatment discontinuation in patients with Ph+ Chronic Myeloid Leukemia (CML) stipulate that it should be proposed in the first line of Tyrosine Kinase Inhibitors (TKIs) treatment or in the second line if the reason for change was intolerance. There are reports in literature of patients (pts) who discontinued therapy in second line also for resistance, with outcome comparable or inferior to those with intolerance depending on the studies. Aims: This study aims to identify prognostic factors for maintaining response in the context of post-first line suspensions. We proposed a focused sub-analysis within the context of the Italy-TFR observational study (NCT04769947) which is ongoing and currently includes 673 pts. Methods: We performed descriptive statistics and survival analysis with Kaplan-Meier. Bayesian Model Averaging (BMA) was used to assess potential risk factors. BMA averages all data-supported models to estimate effect sizes of variables on the endpoint. The percentage of inclusion (PI) indicates how often a variable appears across models, suggesting its importance. Standard rules of thumb for interpreting this posterior probability are: P<50% evidence against the effect; 50-75% weak evidence; 75-95% positive evidence; >95% strong evidence. Results: We collected the data from 144 pts who attempted TFR in second or later lines of treatment with switch due to resistance (33) or intolerance (53) or unknown reason (58). Median age at discontinuation was 64 years (IQR 53-74); 49% of pts were male. Among resistant pts the majority discontinued a 2nd generation TKI, 1 discontinued ponatinib and 1 discontinued imatinib. This pattern was similar to what we observed in intolerant pts. First line treatment in resistant and intolerant pts was imatinib in 87% of pts. Median duration of total treatment was 110 months (IQR 80-147). In the resistant group, 43.8% of pts showed less than Complete Cytogenetic Response (CCyR) as best response to first line therapy, defined as low response, while 56.3% achieved a CCyR, a major molecular response (MMR) or deep molecular response (DMR), defined as high response; in the intolerant group 24.5% and 75.5% of pts had a low and high response, respectively (p<0.001). Thus, we stratified pts who discontinued in second or later lines by best response to first TKIs: Low Responders (LoRes, 31 pts) were resistant in 53.8% and intolerant in 46.2%, High Responders (HiRes, 65 pts) were resistant in 32.7% and intolerant in 67.3% (p 0.09); there was no significant difference in Sokal score. Combining HiRres, LoRes and resistant or intolerant, we distinguished 4 subgroups of pts, whom showed statistically significant differences in rates on Kaplan-Meier curves: at 24 month, 35% of resistant-LoRes maintained TFR vs 71% of resistant-HiRes, vs 82% of intolerant-LoRes, vs 92% of intolerant-HiRes. (p<0.001). Using BMA, longer duration of the last TKI (PI=69%), better response to prior treatment (PI=40%), better response at 3 months (PI=40%), and longer DMR duration from first treatment start (PI=30%) emerged as the strongest factors associated with TFR persistence, whereas resistance to treatment (PI=55%) was associated to treatment restart. Resistant compared to intolerant had an increased risk to restart treatment (HR=4.94, P=91%). Among intolerant pts, each additional month of TKI treatment was associated with a risk reduction of 4% (HR = 0.96, P=93%). The same effect was not observed among resistant pts (HR=1.04, P=84%). Pts with a response at 3 months that was at least MMR had a risk reduction of 63% (HR=0.37, P=91%). Conclusions: We have shown that TFR may be feasible for some pts treated in second-line, not only for intolerance. Duration of TKI treatment seems to be the most relevant prognostic factor also in this category of pts. More observations are needed to safely include second line resistant pts in an elective discontinuation process.