Second-Line Bosutinib Is Effective and Feasible in Patients Failing Second Generation TKI (2G-TKI) in Chronic Phase Chronic Myeloid Leukemia (CML) – an Analysis of a Subcohort of the Bodo Trial (CML-VII-study of the German CML Study Group)

Introduction: Currently, six tyrosine kinase inhibitors (TKIs) are approved for chronic phase chronic myeloid leukemia (CML). Unfortunately, treatment sequence has never been tested prospectively, and, therefore, it remains unclear which TKI should be chosen, especially after first-line treatment with more potent second generation (2G)-TKIs. The original goal of the “Bosutinib Dose Optimization Study” (BODO) study was to evaluate whether a bosutinib step-in dosing regimen decreases gastrointestinal (GI) toxicity while maintaining optimal efficacy in patients (pts) with CML after failure or intolerance to 2G-TKIs. Despite its premature study end, it thereby contains one of the largest cohorts of 2^nd generation TKI treatment after failure/intolerance of a 2G-TKI in first line.

Methods: This is a sub-analysis of the BODO trial (NCT03205267), a multicenter, open-label single arm phase II study testing tolerability and efficacy of 2^nd and 3^rd line bosutinib step-in dosing in chronic phase CML pts intolerant and/or refractory to previous imatinib and/or nilotinib, and/or dasatinib therapy. Bosutinib was commenced with 300 mg QD and was (in the absence of > grade 1 toxicities) dose-increased by increments of 100 mg daily dosing every 14 days up to a maximum dose of 500 mg QD. 127 pts were planned to be recruited. However, due to slow recruitment, the trial had to be stopped prematurely after inclusion of 57 pts. For this analysis, we focused on the 45 patients treated with either nilotinib (n=23) or dasatinib (n=22) as first-line therapy. Cumulative incidence curves of response to bosutinib were calculated and compared using Gray’s test.

Results: 45 pts (n=28 males; median age 51 years, range: 19 - 77) were included in the analysis. The ECOG status was 0 (n=36, 80%) or 1 (n=9, 20%). Prior to study inclusion, a median daily dose of 600 mg nilotinib (range: 300 – 800) and 100 mg of dasatinib (range: 57 - 200) had been administered. The median duration of previous nilotinib and dasatinib therapy was 20.0 months (range: 23 days - 8.0 years) and 16.6 months (range: 3.7 - 34.3 months, p = 0.184), respectively. 18 (40%) pts were intolerant, 17 resistant (38%), and 10 (22%) both intolerant and resistant to previous TKI treatment.

17 (38%) pts entered the study in molecular response (at least major molecular remission (MMR) at screening). The odds of being in MMR at screening were higher in patients with longer pretreatment duration (OR = 1.9; CI: 1.1, 3.7; p = 0.038 per additional year of pretreatment). The corresponding probabilities of MMR were 40% (CI: 26, 54%), 58% (CI: 42, 71%), 68% (CI: 51, 80%) at 3, 6, and 12 months, and 79% (CI: 62, 89%) at 18 and 24 months, respectively. Median time to MMR was 4.9 months. MMR probabilities by pretreatment did not differ between nilotinib and dasatinib (p=0.788).

3 out of 4 intolerant pts without MMR at baseline reached MMR or a better molecular response level with bosutinib. 24 pts refractory to previous therapy (16 resistant; 8 both resistant and intolerant) were lacking baseline MMR, of which 16 pts achieved MMR or better (1 pt with MR4.5, 2 with MR4 and 13 with MMR). In the 24 refractory pts without MMR at baseline, the cumulative MMR and MR4 rates by 1 year were 48% (CI: 26, 68%) and 10% (CI: 2, 27%), respectively. One patient reached MR4.5 after 15.2 months of bosutinib therapy. Response to bosutinib was not significantly different in patients pretreated with either dasatinib or nilotinib. 33 pts received bosutinib for at least 6 months and also had a 6-months molecular response evaluation. At 6 months, median dose levels of bosutinib were not significantly different between responders and non-responders. No disease progressions were reported during the study and follow up.

All pts had ≥1 any grade treatment emergent adverse event (TEAE), occuring a median of 15 days after starting bosutinib (range: 1 - 225 days). Cumulative probability of grade 3/4 TEAEs, and serious adverse events by 1 year was 69% (53, 81%) and 26% (CI: 13, 40%), respectively. The most common TEAE was diarrhea with a cumulative probability of 63% (CI: 47, 76%) by 1 month.

Conclusion: Despite of the limitations of a single-arm study with incomplete recruitment, bosutinib was able to induce optimal responses in two thirds of pts previously resistant to 2G-TKIs while GI toxicity rarely led to treatment discontinuation. We conclude that bosutinib is a safe option after intolerance/failure of a 2G-TKI in first-line.