JAKing Up Targeted Therapy for Ph-like Acute Lymphoblastic Leukemia

A Phase 2 Study of the JAK1/JAK2 Inhibitor Ruxolitinib with Chemotherapy in Children, Adolescents, and Young Adults with De Novo High-Risk Cytokine Receptor-Like Factor 2-Rearranged and/or JAK Pathway-Mutant Acute Lymphoblastic Leukemia

NCT02723994

Incyte Corporation in collaboration with the Children’s Oncology Group (COG)

108 COG sites

180 patients

AALL1521 is a nonrandomized, two-part phase II clinical trial designed to study the safety and efficacy of combining the JAK1/JAK2 inhibitor ruxolitinib with post-induction chemotherapy for pediatric patients with Philadelphia chromosome–like (Ph-like) B-cell acute lymphoblastic leukemia (B-ALL). This study is enrolling children, adolescents, and young adults ages one to 21 years with newly diagnosed National Cancer Institute (NCI) high-risk (HR) Ph-like B-ALL and CRLF2 rearrangements or other JAK pathway alterations following completion of a four-drug induction chemotherapy on or as per the COG AALL1131 phase III trial (NCT02883049).

Patients with HR B-ALL are screened for the Ph-like gene expression signature during induction via a low-density microarray (LDA) assay.¹  Specimens identified as Ph-like undergo further genetic testing; those with CRLF2 overexpression are assessed for CRLF2 rearrangements (i.e., P2RY8-CRLF2 by reverse-transcriptase polymerase chain reaction [PCR], IGH-CRLF2 by fluorescence in-situ hybridization [FISH]) and for mutations in JAK1, JAK2, and IL7R by PCR, while those with normal CRLF2 expression undergo Archer fusionplex testing for ABL class, JAK2, EPOR, and other translocations. Rare Ph-like cases without identified lesions are allocated for RNA sequencing (RNA-seq) to identify novel mutations/fusions (Figure). Patients with the Ph-like signature and CRLF2 rearrangements, JAK2 or EPOR fusions, IL7R indels, or SH2B3 deletions are eligible for AALL1521 and are stratified into cohorts based on genetic lesion and end-induction minimal residual disease (MRD) status.

In the part 1 safety phase, investigators evaluated the safety of six dosing strategies in patients concomitantly receiving standard augmented Berlin-Frankfurt-Munster postinduction chemotherapy as per AALL1131. No dose-limiting toxicities were identified. The recommended phase II dose (RP2D) of ruxolitinib was determined to be 50 mg/m²/dose 14-days-on/14-days-off (dose level 2), and these interim results were presented at the 2018 ASH Annual Meeting.² Part 2 of the trial is currently accruing to evaluate the efficacy of combination therapy. The primary endpoint of this trial is three-year event-free survival (EFS) of patients treated with ruxolitinib and chemotherapy versus 65 percent three-year EFS of historic control patients treated with an identical chemotherapy backbone. Secondary and exploratory outcomes include pharmacokinetic and pharmacodynamic assessments and rate of end-consolidation MRD negativity.

Despite high cure rates achievable in most children with B-ALL using risk-adapted chemotherapy, certain HR subsets have persistently very poor outcomes. Recent genomic profiling of patients with HR B-ALL led to identification of the Ph-like subtype,^3,4  which comprises 15 to 20 percent of HR B-ALL in children and adolescents and nearly 30 percent in adults.^5-9  Ph-like ALL is defined by the lack of BCR-ABL1 fusion but has a kinase-activated gene expression signature similar to that of Ph+ ALL, frequent deletions of IKZF1, and mutations in other kinase and cytokine receptor signaling pathway genes. Patients with Ph-like ALL have extremely high relapse rates and poor overall survival (OS). There are several main classes of genetic alterations described in Ph-like ALL to date: 1) ABL-class rearrangements involving ABL1, ABL2, CSF1R, and PDGFRB, and 2) JAK/STAT pathway alterations of CRLF2, JAK2, EPOR, IL7R, and SH2B3, as well as additional uncommon lesions in other kinase-associated genes.¹⁰  Approximately 50 percent of patients with Ph-like ALL have rearrangements in CRLF2 (often co-occurring with JAK2 or JAK1 point mutations) that result in constitutive JAK/STAT activation. An additional 8 to 16 percent of patients have JAK2 or EPOR translocations also characterized by hyperactive JAK/STAT pathway signaling. Preclinical studies have demonstrated in vitro and in vivo activity of ruxolitinib in Ph-like ALL models.^11,12  The COG ADVL1011 phase I clinical trial (NCT01164163) previously demonstrated safety and tolerability of ruxolitinib monotherapy in children with relapsed/refractory cancers. These observations provide the rationale for frontline testing of JAK inhibitors in newly diagnosed patients with relevant genetic lesions and in combination with chemotherapy.¹³  Of note, patients with Ph-like B-ALL enrolled on AALL1131 with identified ABL-class fusions are nonrandomly allocated to receive dasatinib with post-induction chemotherapy.

Addition of imatinib to chemotherapy for patients with BCR-ABL1-rearranged (Ph+) ALL dramatically improved EFS and revolutionized the paradigm for precision medicine in patients with acute leukemias.^14,15  AALL1521 is a first-in-child investigation of the potential efficacy of combining JAK inhibition and chemotherapy to decrease relapse risk and improve OS in the relatively recently recognized Ph-like subtype of HR B-ALL. This trial is unique in its rigorous, real-time clinical characterization of Ph-like genetic lesions via gene expression analyses, FISH, PCR, unbiased fusionplex testing, and RNA-seq.

While a small number of anecdotal reports have described patients with Ph-like B-ALL and JAK pathway alterations who were treated with commercially available ruxolitinib and chemotherapy,16, the AALL1521 trial is systematically investigating key questions of whether the addition of a JAK inhibitor can decrease relapse and improve survival in patients with Ph-like B-ALL, as well as identify which genetic subgroups may or may not benefit from combination therapy. This trial represents a promising precision medicine strategy for targeting oncogenic drivers in a common and high-risk subset of B-ALL that spans the age spectrum.