Impact of Philadelphia chromosome‐like alterations on efficacy and safety of blinatumomab in adults with relapsed/refractory acute lymphoblastic leukemia: A post hoc analysis from the phase 3 TOWER study

Philadelphia chromosome (Ph)-like acute lymphoblastic leukemia (ALL) is a high-risk subgroup of precursor B-cell (BCP) ALL. Patients with Ph-like ALL have a gene expression profile (GEP) similar to those with Ph+ ALL, but lack the characteristic BCR-ABL1 fusion. Among these patients, inferior outcomes are observed compared with those without Ph-like ALL.1 A standard treatment approach does not yet exist for the treatment of patients with Ph-like ALL. Blinatumomab, a BiTE® (bispecific T-cell engager) molecule, is effective in adults and children with relapsed/refractory (R/R) BCP ALL and persistent minimal residual disease (MRD).2 Results from a randomized, multicenter, phase 3 trial in adult patients with R/R BCP ALL (TOWER; NCT02013167) demonstrated significantly longer median overall survival (OS) with blinatumomab compared with standard of care chemotherapy (SOC; 7.7 months vs. 4.0 months, p = 0.01).3 Here, we report the results of a post hoc analysis of the TOWER trial that evaluated the impact of the Ph-like alterations on the efficacy and safety of blinatumomab and SOC in patients with R/R BCP ALL. The previously reported TOWER study conducted in adult patients with R/R BCP ALL compared the efficacy and safety of blinatumomab with SOC3. Of the 405 patients randomly assigned in a 2:1 ratio to receive blinatumomab (n = 271) or SOC (n = 134), stored and available samples from 142 patients were analyzed at the MD Anderson Cancer Center (MDACC; MDACC analysis set) in this post hoc study. A comprehensive targeted next-generation sequencing assay, which identifies fusions, point mutations, expression levels in 81 genes associated with ALL, and all recurrent Ph-like ALL fusion events, was performed on RNA extracted from samples of these patients (n = 142) using a multiplex fusion and mutation detection assay (Archer® FusionPlex® ALL, ArcherDX, Inc., Colorado, USA; Table S1). Details of targeted Ph-like ALL testing and RNA sequencing are described in the supplement. Efficacy analyses and baseline characteristics were performed on the MDACC analysis set and safety analyses were performed on patients included in the MDACC analysis set, who received at least one dose of blinatumomab or SOC. Descriptive analysis was used to summarize demographics and safety data. The proportion of patients who achieved responses, including complete remission (CR), CR with partial hematologic recovery (CRh), CR with incomplete hematologic recovery (CRi), and MRD remission, was summarized with exact 95% confidence intervals (CIs). The Kaplan–Meier method was used to estimate OS. The hazard ratio (HR) estimates were obtained from the Cox proportional hazard model with stratification factors of age (<35 vs. ≥35 years), prior salvage therapy (yes vs. no), and prior allogeneic hematopoietic stem cell transplantation (alloHSCT; yes vs no). Of the 142 samples analyzed for the Ph-like ALL alterations, 101 (71%) were from patients treated with blinatumomab and 41 (29%) from patients with SOC (Table S2). As analyzed by RNA sequencing, a total of 15 (11%) patients had Ph-like ALL; nine were treated with blinatumomab and six with SOC. Demographics and baseline characteristics of patients in this post hoc study were similar to those of patients enrolled in the phase 3 TOWER study.3 Among the 15 patients identified with Ph-like ALL, RCSD1-ABL1, SQSTM1-JAK2, ZMYM2-FGFR1, ATF7IP-JAK2, ETV6-ABL1, JAK2-CCDC171, PAX5-JAK2, and IGH-CRLF2 were observed in one (7%) patient each, EPOR-IGH and EBF1-PDGFRB were observed in two (13%) patients each, and P2RY8-CRLF2 was observed in three (20%) patients. Remission rates within 12 weeks of treatment initiation with blinatumomab were similar in patients with (n = 9) and without (n = 92) Ph-like ALL with respect to CR (33% vs. 36%, respectively) and CRh (11% vs. 10%, respectively; Table S3). Among patients receiving blinatumomab who had CR, CRh, or CRi, two of four (50%) patients with Ph-like ALL and 18 of 43 (42%) patients without Ph-like ALL achieved a negative status for MRD (Table S4). In contrast, none of the patients with Ph-like ALL receiving SOC achieved CR, CRh, CRi, or a negative status for MRD (Tables S3 and S4). Among nine patients without Ph-like ALL receiving SOC who had CR, CRh, or CRi, a negative status for MRD was achieved in two (22%) patients (Table S4). The median OS in patients with and without Ph-like ALL treated with blinatumomab was comparable (Figure 1). The median OS in patients with and without Ph-like ALL treated with blinatumomab was longer than those treated with SOC (Figure 1). The OS HRs from a Cox proportional hazard model in patients with and without Ph-like ALL were 0.39 (95% CI, 0.04–3.78) and 0.73 (95% CI, 0.43–1.23), respectively, in favor of blinatumomab. Overall, 44% of patients with Ph-like ALL and 22% of patients without Ph-like ALL receiving blinatumomab underwent alloHSCT, including 33% and 9% of patients, respectively, who achieved remission without the use of another anticancer therapy (Table S5). In contrast, a greater proportion of patients without Ph-like ALL treated with SOC underwent alloHSCT (26%) than those with Ph-like ALL (17%). None of the patients with Ph-like ALL in blinatumomab or SOC group who achieved remission with the use of an intervening anticancer treatment underwent alloHSCT (Table S5). Overall, the toxicity profile was similar in patients with and without Ph-like ALL in both treatment groups (Table S6). A majority of patients with or without Ph-like ALL treated with blinatumomab or SOC experienced treatment-emergent adverse events (AEs), most of which were treatment-related. A fatal AE, deemed not related to blinatumomab, occurred in one (11%) patient with Ph-like ALL. In patients without Ph-like ALL treated with blinatumomab, fatal AEs were reported in 16 (18%) patients, three (3%) of which were related to treatment. No fatal AEs occurred in patients with Ph-like ALL treated with SOC. Of the four (13%) fatal AEs that occurred in patients without Ph-like ALL treated with SOC, one (3%) was related to treatment. Central neuropsychiatric events due to direct neurotoxicities, cytopenia, and infections were the most frequent AEs of interest observed in patients with and without Ph-like ALL treated with blinatumomab and SOC (Tables S7 and S8). Key findings from this study indicated that efficacy and safety of blinatumomab were similar in patients with and without Ph-like ALL, thereby negating the effect of Ph-like alterations. In contrast, SOC did not prove to be efficacious in the six patients with Ph-like ALL. Overall, the efficacy and safety findings from this post hoc analysis were consistent with those reported in the phase 3 TOWER study.3 The targeted lysis of normal and malignant B-lineage cells by blinatumomab, achieved by engaging CD3+ cytotoxic T cells with CD19 expressing B-lineage cells, is independent of the presence or absence of the mutations associated with Ph-like ALL. This could provide a plausible explanation for the similar efficacy exhibited by blinatumomab in patients with or without Ph-like ALL. In comparison, none of the patients with Ph-like ALL treated with SOC achieved CR/CRh/CRi in the present study, thereby demonstrating blinatumomab's superiority and potential utility in treating patients with Ph-like ALL. In this study, a negative status for MRD was reported in 50% of the patients with Ph-like ALL who achieved CR/CRh/CRi within 12 weeks of blinatumomab treatment initiation. Similarly, results from a study evaluating outcomes of adult patients with Ph-like ALL (n = 148) treated with a hyperfractionated cyclophosphamide, vincristine, doxorubicin, and dexamethasone or an augmented Berlin-Frankfurt-Münster regimen demonstrated that the likelihood of achieving MRD negativity in these patients was significantly less than those with Ph+ or B-other ALL.4 Central neuropsychiatric events due to direct neurotoxicities, cytopenia, and infections were the most frequent AEs of interest in this post hoc analysis compared with neutropenia, infections, and elevated liver enzymes observed in the TOWER study. Several studies have indicated that Ph-like ALL is more common in patients with Hispanic or Latino ethnicity.4-6 Although the phase 3 TOWER study was a global study conducted across 21 countries, a majority of the study centers were located in European countries, which have a very low Hispanic or Latino population. This could provide a possible explanation for the low number of patients with Ph-like ALL identified from the TOWER study. The study had a few limitations. Firstly, the number of samples available for RNA sequencing to identify patients with Ph-like ALL in this study were limited, thus reducing our ability to draw firm conclusions from the data. Secondly, additional patients with Ph-like ALL could have been identified if GEP was performed. Finally, although the efficacy and safety of blinatumomab was demonstrated in this post hoc analysis in patients with Ph-like alterations, further validation of these results in a clinical trial with a large cohort of patients diagnosed with Ph-like ALL needs to be conducted. In conclusion, the efficacy and safety of blinatumomab appeared to be similar in patients with and without Ph-like ALL, thus potentially abrogating the poor prognosis associated with Ph-like alterations in these patients. Overall, findings from this post hoc analysis in patients with Ph-like ALL treated with blinatumomab were consistent with that reported in the TOWER study. The authors want to acknowledge contributions by Jessica (Chieh) Lan and Ronald Abraham for performing targeted Ph-like ALL Fusion testing and the ATGC core facility for performing RNA-seq testing. The authors received medical writing and editing support from Advait Joshi, PhD, of Cactus Life Sciences (part of Cactus Communications), which was supported by Amgen Inc. The study was funded by Amgen Inc. The ATGC core is supported by CA016672(ATGC) and NIH 1S10OD024977-01 grants. Elias Jabbour receives research grants from Amgen, Abbvie, Spectrum, BMS, Takeda, Pfizer, Adaptive, and Genentech. Keyur Patel, Rajyalakshmi Luthra, Anthony San Lucas have nothing to disclose. Dr Dzifa Duose receives honorarium for lectures, presentations, speakers' bureaus, manuscript writing or educational events from Chrysallis Biomedical Advisors. Nicholas J. Short receives research grants from Takeda Oncology and Astellas; consulting fees from Takeda Oncology, AstraZeneca, Amgen, NGMBio, and Novartis; honorarium for lectures, presentations, speakers' bureaus, manuscript writing or educational events from Amgen. Nitin Jain receives research grants from Pharmacyclics, AbbVie, Genentech, AstraZeneca, BMS, Pfizer, Servier, ADC Therapeutics, Cellectis, Adaptive Biotechnologies, Incyte, Precision Biosciences, Aprea Therapeutics, Fate Therapeutics and payment or honoraria from Pharmacyclics, Janssen, AbbVie, Genentech, AstraZeneca, BMS, Adaptive Biotechnologies, Servier, Precision Biosciences, Beigene, Cellectis, TG Therapeutics, and ADC Therapeutics. Gerhard Zugmaier is an inventor or co-inventor of patent applications/patents belonging to the families of international patent applications published as WO2010/052014, WO2010/052013, WO2011/051307, WO2012/062596, and WO2015/181683. Qui Tran is an employee of Amgen, Inc. Faraz Zaman and Kelly Velasco are employees and holds stocks of Amgen, Inc. Marina Konopleva reports grants and other from AbbVie, grants and other from F. Hoffman La-Roche, grants and other from Stemline Therapeutics, grants and other from Forty-Seven, grants from Eli Lilly, grants from Cellectis, grants from Calithera, grants from Ablynx, grants from Agios, grants from Ascentage, grants from Astra Zeneca, other from Reata Pharmaceutical, grants from Rafael Pharmaceutical, grants from Sanofi, other from Janssen, grants and other from Genentech, outside the submitted work. In addition, Marina Konopleva has a patent US 7,795,305 B2 CDDO-compounds and combination therapies with royalties paid to Reata Pharm., a patent on combination therapy with a mutant IDH1 Inhibitor and a BCL-2 licensed to Eli Lilly, and a patent 62/993,166 titled combination of a mcl-1 inhibitor and midostaurin, uses and pharmaceutical compositions thereof pending to Novartis. Hagop Kantarjian receives research grants from Amgen, Ascentage, BMS, Daiichi-Sankyo, Immunogen, Jazz, Pfizer, and Sanofi and honoraria from AbbVie, Actinium, Adaptive Biotechnologies, Amgen, Aptitude Health, BioAscend, Daiichi-Sankyo, Delta Fly, Janssen Global, Novartis, Oxford Biomedical, Pfizer, and Takeda Oncology. Elias Jabbour and Hagop Kantarjian conceptualized and designed the study, collected patient data, and analyzed and interpretated the data. Faraz Zaman conceptualized and designed the study, and analyzed and interpretated the data. Keyur Patel, Dzifa Duose, Rajyalakshmi Luthra, Anthony San Lucas, and Qui Tran analyzed and interpretated the data. Gerhard Zugmaier and Kelly Velasco conceptualized and designed the study. Nicholas J. Short, Nitin Jain, and Marina Konopleva collected patient data. Other than these, all the authors contributed in writing the manuscript. Qualified researchers may request data from Amgen clinical studies. Complete details are available at the following: http://www.amgen.com/datasharing Table S1 List of 81 genes associated with ALL Table S2. Baseline patient and disease characteristics Table S3. Best response within 12 weeks of treatment initiation Table S4. MRD remission within 12 weeks of treatment initiation Table S5. Patients with alloHSCT Table S6. Incidence of AEs in patients treated with blinatumomab and SOC Table S7. Incidence of AEs of interest by category in patients treated with blinatumomab Table S8. Incidence of AEs of interest by category in patients treated with SOC Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.