Key Points• BCR-ABL1-like signature and IKZF1 deletions are clinically important to identify high-risk acute lymphoblastic patients.Most relapses in childhood B-cell precursor acute lymphoblastic leukemia (BCP-ALL) are not predicted using current prognostic features. Here, we determined the cooccurrence and independent prognostic relevance of 3 recently identified prognostic features: BCR-ABL1-like gene signature, deletions in IKZF1, and high CRLF2 messenger RNA expression (CRLF2-high). These features were determined in 4 trials representing 1128 children with ALL: DCOG ALL-8, ALL9, ALL10, and Cooperative ALL (COALL)-97/03. BCR-ABL1-like, IKZF1-deleted, and CRLF2-high cases constitute 33.7% of BCR-ABL1-negative, MLL wild-type BCP-ALL cases, of which BCR-ABL1-like and IKZF1 deletion (co)occurred most frequently. Higher cumulative incidence of relapse was found for BCR-ABL1-like and IKZF1-deleted, but not CRLF2-high, cases relative to remaining BCP-ALL cases, reflecting the observations in each of the cohorts analyzed separately. No relapses occurred among cases with CRLF2-high as single feature, whereas 62.9% of all relapses in BCR-ABL1-negative, MLL wild-type BCP-ALL occurred in cases with BCR-ABL1-like signature and/or IKZF1 deletion. Both the BCR-ABL1-like signature and IKZF1 deletions were prognostic features independent of conventional prognostic markers in a multivariate model, and both remained prognostic among cases with intermediate minimal residual disease. The BCR-ABL1-like signature and an IKZF1 deletion, but not CRLF2-high, are prognostic factors and are clinically of importance to identify high-risk patients who require more intensive and/or alternative therapies. (Blood. 2013;122(15):2622-2629

SummaryMalignant cells infiltrating the bone marrow (BM) interfere with normal cellular behaviour of supporting cells, thereby creating a malignant niche. We found that CXCR4-receptor expression was increased in paediatric precursor B-cell acute lymphoblastic leukaemia (BCP-ALL) cells compared with normal mononuclear haematopoietic cells (P < 0Á0001). Furthermore, high CXCR4-expression correlated with an unfavourable outcome in BCP-ALL (5-year cumulative incidence of relapse AE standard error: 38Á4% AE 6Á9% in CXCR4-high versus 12% AE 4Á6% in CXCR4-low expressing cases, P < 0Á0001). Interestingly, BM levels of the CXCR4-ligand (CXCL12) were 2Á7-fold lower (P = 0Á005) in diagnostic BCP-ALL samples compared with non-leukaemic controls. Induction chemotherapy restored CXCL12 levels to normal. Blocking the CXCR4-receptor with Plerixafor showed that the lower CXCL12 serum levels at diagnosis could not be explained by consumption by the leukaemic cells, nor did we observe an altered CXCL12-production capacity of BM-mesenchymal stromal cells (BM-MSC) at this time-point. We rather observed that a very high density of leukaemic cells negatively affected CXCL12-production by the BM-MSC while stimulating the secretion levels of granulocyte colony-stimulating factor (G-CSF). These results suggest that highly proliferative leukaemic cells are able to down-regulate secretion of cytokines involved in homing (CXCL12), while simultaneously up-regulating those involved in haematopoietic mobilization (G-CSF). Therefore, interference with the CXCR4/ CXCL12 axis may be an effective way to mobilize BCP-ALL cells.

No abstract

Bosutinib is a tyrosine kinase inhibitor (TKI), approved for adults with Philadelphia chromosome-positive (Ph+) chronic myelogenous leukemia (CML); 400 mg in newly diagnosed (ND) patients (pts), and 500 mg in resistant/intolerant (R/I) pts, administered once daily (QD) with food. Its toxicity profile differs from other TKIs approved in children (imatinib, dasatinib and nilotinib), showing a higher incidence of gastrointestinal (GI) adverse events (AEs) but fewer toxicities such as musculoskeletal AEs (Cortes JE et al, 2016). Furthermore, in mice, bosutinib showed less growth impairment compared to other TKIs (Tauer JT et al, 2015). Study NCT04258943 is an international, open-label, phase I/II trial, initially enrolling pediatric Ph+ CML R/I pts (per 2013 European Leukemia Net criteria), in chronic, accelerated or blastic (AP/BP) phase, aged ≥1 and &lt;18 years, with normal liver and kidney function. Main exclusion criteria consisted of leptomeningeal leukemia in AP/BP pts, T315I or V299L BCR-ABL1 mutations, use of proton pump inhibitors and moderate/strong CYP3A inducers/inhibitors. The protocol was amended to enroll ND pts in the phase II. Results from the phase I in R/I pts, with a data cut-off 05 July 2021, are presented. The primary objective was to determine the recommended phase II dose (RP2D) of bosutinib in R/I and ND pts. Secondary objectives included safety and preliminary efficacy. A 6+4 dose-escalation design was used. The RP2D was defined as the dose resulting in 0/6 or 1/10 dose limiting toxicities (DLTs; see table 1) in pts who received ≥ 75% of the planned dose in cycle 1, and mean area under the curve (AUC) of 3150 (±20%) ng·hr/mL in R/I pts and 2270 (±20%) ng·hr/mL in ND pts (equivalent to the exposure of 500 mg and 400 mg QD in adults, respectively). Cycles consisted of 28 days of bosutinib QD, starting from 300 mg/m 2, with escalation to 350 mg/m 2 and 400 mg/m 2 (max 600 mg QD). Pharmacokinetic sampling was performed on day 14 of cycle 1, after achievement of steady-state. All protocol versions obtained ethics approval. The study is sponsored by Erasmus Medical Center in Europe and COG in the United States, it is open in 19 ITCC and 28 COG sites and is funded by Pfizer. Since November 15 2016, 23 pts with R/I CP CML were enrolled at 12 ITCC and 2 COG centers. Pts characteristics are provided in table 2. Six pts were enrolled at 300 mg/m 2 (median n. of cycles 18.5; range 11-57), without DLTs. The dose was escalated to 350mg/m 2 and 11 pts were enrolled (median n. of cycles 10, range 1-25); 1 DLT occurred (grade 3 GI toxicity). At 300 mg/m 2 (AUC: 2010 ng·hr/mL) and 350 mg/m 2 (AUC: 2385 ng·hr/mL) the AUC was under target for R/I pts; but the 300 mg/m 2 dose resulted in adequate and safe exposure for ND pts. For R/I pts the dose was increased to 400mg/m 2 and 6 pts were enrolled (median n. of cycles 3, range 0-7). One patient experienced a DLT (Grade 3 transaminase increase, rash and treatment interruption &gt; 7 days). Exposure at 400mg/m 2 (n=4) was within the target interval at 3064 ng·hr/mL. Overall, 22/23 pts had at least 1 AE, most common were GI toxicities (87%) and rash (39%). Grade 3 or 4 AEs were observed in 10/23 (43.5%) pts; GI were the most common (6/10 pts). Elevated transaminases occurred in 6 pts, 3 of which at grade 3 or 4. Median follow up was 9.9 months (mns). Including pts in Complete Cytogenetic Response/Major Molecular response (CCyR/MMR) at screening, cumulative incidence of CCyR at 3 and 6 mns was 68% (95% CI: 49-88%) and 82% (95% CI: 66-98%); the cumulative incidence of MMR at 3 and 6 mns was 27% (95% CI: 09-46%) and 36% (95% CI: 16-56%) respectively. Considering only pts that achieved CCyR/MMR during treatment, at 3 and 6 mns, 4/13 and 3/9 pts achieved CCyR (median time to response 3 mns; range 2.5-5.7) while 3/19 and 2/17 achieved MMR (median time to response 3 mns; range 2.5-16.5), respectively. All patients that achieved or entered the study in CCyR and/or MMR (table 2) maintained the response. Four pts stopped the treatment due to insufficient response. In ND pts, the RP2D was established at 300 mg/m 2 QD. In R/I pts, 400 mg/m 2 QD seems to result in comparable exposure as 500 mg QD in adults, however the cohort is to be completed to establish the RP2D. The AE profile of bosutinib was consistent with the known safety profile in adults. GI toxicities were the most common, mainly during the first cycles and tended to improve in subsequent cycles. Cumulative MMR and CCyR rates were comparable to those reported in literature for other TKIs (Hijiya N et al, 2019). Figure 1 Figure 1. Disclosures Durairaj: Pfizer Inc: Current Employment, Current equity holder in publicly-traded company. Viqueira: Pfizer: Current Employment, Current equity holder in publicly-traded company. Ingrosso: Pfizer: Current Employment, Current equity holder in publicly-traded company. Locatelli: Amgen: Membership on an entity's Board of Directors or advisory committees, Speakers Bureau; Novartis: Membership on an entity's Board of Directors or advisory committees, Speakers Bureau; Miltenyi: Speakers Bureau; Medac: Speakers Bureau; Jazz Pharamceutical: Speakers Bureau; Takeda: Speakers Bureau. Hijiya: Novartis: Consultancy; Stemline Therapeutics: Consultancy. Zwaan: SANOFI: Consultancy; NOVARTIS: Consultancy; ROCHE: Consultancy; INCYTE: Consultancy; PFIZER: Consultancy, Research Funding; JAZZ: Other: travel funding, Research Funding; BMS: Research Funding; Abbvie: Research Funding.

10017 Background: Bosutinib is a tyrosine kinase inhibitor (TKI), approved for adults with Philadelphia Chromosome (Ph+) CML; at the standard initial dose of 400 mg/day in ND patients, and 500 mg/day in resistant/intolerant (R/I) patients, administered orally once daily (QD) with food. Compared to the TKIs already approved in pediatrics, bosutinib has a different tolerability profile, and preclinical data suggest that longitudinal growth is potentially less impaired. Study NCT04258943 is an international, open-label, phase I/II trial, sponsored by the Erasmus Medical Center and the Children's Oncology Group, and funded by Pfizer Inc. The phase II arm enrolled ND patients, as well as R/I ones. The latter are not included in this analysis. Methods: ND patients aged 1-18 years with chronic phase CML, without evidence for organ toxicities, were enrolled. Main exclusion criteria included known T315I or V299L BCR-ABL1 mutations, and use of proton pump inhibitors and CYP3A inducers/inhibitors. The primary objectives of the phase II part of the study were to assess the safety and pharmacokinetics (PK) of bosutinib at the recommended phase II dose, which is body surface area adjusted, and consists of 300 mg/m2 QD for ND patients (max 500 mg/day), as determined in the phase I part of the study. Based on regulatory authorities requirements, at least 35 patients have to be enrolled in phase II, with a total of 50-60 patients in phase I and II combined. This allows for pooled AE rates to be estimated with a maximum standard error of 0.071 and 0.065, respectively. Results: On 20/12/2022, 25 ND patients were screened, and 24 were enrolled: 15 males, median age 13 years (range: 5-17). The median follow up was 14 (range: 0.9-31) months. The most common non-hematological adverse events (AEs) were diarrhea (n = 16, Grade (Gr) ≤ 2 = 13), abdominal pain (n = 10, Gr ≤ 2 = 10), and nausea/vomiting (n = 8/8, Gr ≤ 2 = 7/8). Most common hematological AEs included platelet count decrease (n = 11, Gr ≤ 2 = 5), and anemia (n = 10, Gr ≤ 2 = 4). At 6 and 12 months, the cumulative incidence of Major Cytogenetic Response (MCyR) was 85% (95%CI 49%-96%) and 92.5% (95%CI 39%-99%), while for Complete CyR it was 77% (47%-91%) and 88.3% (95%CI 55%-98%), and for Major Molecular Response it was 23% (95%CI 7%-46%) and 30% (95%CI 10%-54%), respectively. Three patients permanently discontinued bosutinib due to intolerance, four due to unsatisfactory response per investigator’s judgment, 17 were still on treatment at time of dataset lock. Conclusions: Bosutinib was well tolerated, despite common grade 1-2 gastrointestinal AEs. The preliminary efficacy seems comparable to published data from other second generation TKIs in children and to ND adult patients treated with bosutinib. Clinical trial information: NCT04258943 .

3528 Despite the relative high success-rate to treat children (<18 yrs) with precursor B-ALL (5-year event-free survival >80%), the prognosis of the subtype of BCRABL1- positive ALL remains poor (5-year event-free survival <50%). Patients carrying this translocation are treated according to intensified high-risk protocols which are, however, associated with therapy-related side-effects and mortality. A different treatment for those cases at high risk of treatment failure and less treatment for those at low risk of treatment failure is preferred to improve overall outcome results in children with BCRABL1 -positive ALL. Deletions in the gene encoding Ikaros (IKZF1) and an aberrant high expression level of the cytokine receptor-like factor 2 gene (CRLF2) have both recently been associated with a poor prognosis of children with newly diagnosed precursor B-ALL. We here studied whether these genomic abnormalities can also serve as prognostic factor within the subset of BCRABL1 -positive pediatric ALL. Methods; Leukemic cells of BCRABL1- positive patients enrolled in two consecutive treatment protocols of the Dutch Childhood Oncology Group (DCOG) were analyzed for IKZF1 and CRLF2 status. IKZF1 deletions were determined by Multiplex Ligation Probe-based Amplification assays (MLPA) and comparative genomic hybridization arrays (aCGH; Agilent, Sureprint G3 Human CG 180K). The p335 MPLA assay (MRC Holland) was used for discovery of IKZF1 deletions and data were validated by aCGH and a second MLPA assay (p202; MRC Holland) with higher resolution for affected regions. CRLF2 expression levels were determined by Affymetrix U133 plus 2.0 gene expression arrays. A large cohort of 459 newly diagnosed ALL cases served as reference. A high level of CRLF2 expression was assigned to 10% of all cases ranked according to CRLF2 expression level. Results; Deletions in IKZF1 were found in 65% of the cases (n=11); 10 cases had a partial deletion between exon 2 and 7 of IKZF1 resulting in a dominant negative variant and 1 case had a deletion of the complete IKZF1 gene. Analysis of patient characteristics revealed that cases with deletions in IKZF1 have a median 4.5-fold higher white blood cell count compared to unaffected cases (p=0.009). Analysis of the cumulative incidence of relapse (pCIR) with death as a competing risk indicated that children with IKZF1 -deleted BCRABL1 -positive ALL had a significantly increased risk of developing a relapse compared with those with an unaffected IKZF1 gene (73% versus 17%, p=0.02). As a consequence, the 5-year event-free survival was also lower for the IKZF1- deleted group compared to the unaffected group (18% versus 67% p=0.04, Log-Rank). The expression level of CRLF2 in BCRABL1- positive cases was median 91.4 arbitrary units (p25-p75: 84.2–97.2) compared to 91.3 arbitrary units (p25-p75: 82.6–107.3) for BCRABL1 -negative cases (p>0.05). Expression of none of the BCRABL1 -positive ALL cases was ranked within the highest 10% percentile of the reference group. Hence, high expression of CRLF2 cannot explain a poor prognosis of BCRABL1- positive cases. In conclusion, our data show that a deletion of IKZF1 but not a high expression of CRLF2 is a strong discriminative prognostic factor in BCRABL1- positive pediatric precursor B-ALL. We therefore advocate to implement the IKZF1 status as a new prognostic factor in BCRABL1- positive pediatric ALL. Disclosures: No relevant conflicts of interest to declare.

Malignant cells that infiltrate the bone marrow (BM) interfere with the normal cellular behavior of supporting cells, thereby creating an alternative malignant niche. This intercellular communication is mostly mediated by cytokines and their receptors. In this study, we find that expression of the CXCR4 receptor is significantly increased in pediatric precursor B-cell acute lymphoblastic leukemia (BCP-ALL) cells compared with normal mononuclear hematopoietic cells derived of the bone marrow (p=0.016). Furthermore, we show that high CXCR4 expression is correlated with an unfavorable clinical outcome in BCP-ALL (5-yr CIR ±SE: 38.4% ±6.9% in CXCR4-high versus 12.0% ±4.6% in CXCR4-low expressing patients, p<0.001). Interestingly, BM serum levels of the CXCR4 ligand (CXCL12) are 2.7-fold lower (p=0.005) in samples taken at initial diagnosis of BCP-ALL compared with the levels in samples taken of non-leukemic controls. We show that induction chemotherapy restores CXCL12 levels in the BM to normal levels. Blocking the CXCR4 receptor with Plerixafor (FDA-approved drug) showed that the lower CXCL12 serum levels at initial diagnosis could not be explained by consumption by the leukemic cells, nor did we observe an altered CXCL12-production capacity of BM-MSC at this time-point. We rather observed that a very high density of leukemic cells negatively affected CXCL12 production by the BM-MSC while stimulating the secretion levels of G-CSF. These results suggest that highly proliferative leukemic cells are able to down-regulate the production of cytokines involved in homing (CXCL12), while simultaneously up-regulating the production of cytokines involved in hematopoietic mobilization (G-CSF). This disbalance may stimulate the spreading of BCP-ALL outside the BM. The data presented here suggest that interference with the CXCR4/CXCL12 axis (for instance by using Plerixafor) may be an effective way to mobilize BCP-ALL cells; the more ALL cells become mobilized, the less ALL cells may escape from combination chemotherapy. In proof-of concept studies, this hypothesis needs to be validated to pave the way for implementation in future treatment protocols for children with ALL. Disclosures: No relevant conflicts of interest to declare.

No abstract