Potent Efficacy of Combined PI3K/mTOR and JAK or SRC/ABL Inhibition in Philadelphia Chromosome-like Acute Lymphoblastic Leukemia

Background. Philadelphia chromosome-like acute lymphoblastic leukemia (Ph-like ALL) is associated with genomic alterations that activate JAK/STAT and PI3K/Akt/mTOR signal transduction and with poor clinical outcomes. Therapeutic disruption of PI3K pathway signaling in Ph-like ALL has been minimally investigated to date, however. We hypothesized that PI3K isoform-selective or dual PI3K pathway protein inhibition would robustly inhibit Ph-like ALL proliferation in vivo and abrogate aberrant signaling.

Methods. NOD.Cg-Prkdc^scid Il2rg^tm1Wjl/SzJ (NSG) mice were engrafted with primary CRLF2/JAK-mutant or ABL/PDGFR-mutant Ph-like ALL specimens (Table 1) and treated with inhibitors of PI3Kα (BYL719), PI3Kδ (idelalisib), PI3K/mTOR (gedatolisib), TORC1/TORC2 (AZD2014) or with vehicle. Treated patient-derived xenograft (PDX) models were assessed for pharmacodynamic inhibition of signal transduction phosphoproteins at 72 hours by phosphoflow cytometry and for residual ALL in murine spleens after 3-4 weeks of inhibitor or vehicle treatment by quantitative flow cytometry. Subsequent studies tested the efficacy of gedatolisib with the JAK1/2 inhibitor ruxolitinib (CRLF2/JAK-mutant models) or gedatolisib with the SRC/ABL inhibitor dasatinib (ABL/PDGFR-mutant models).

Table 1. Genomic characteristics of Ph-like ALL specimens utilized for PDX studies.

USI	Disease status	CRLF2/JAK  alterations	ABL/PDGFR alterations
PALTWS	D	IGH@-CRLF2*
PAMDKS	D	IGH@-CRLF2, JAK2R683G
PAMDRM	D	IGH@-CRLF2, JAK2GPinsR683
ALL121	R	IGH@-CRLF2, JAK2R683G
ALL4364	R	P2RY8-CRLF2, JAK2R683G
PAKMVD	D	JAK1S646F
PAKYEP	D	BCR-JAK2
PAKKCA	D		EBF1-PDGFRB
PAKVKK	D		NUP214-ABL1
PANSFD	D		ETV6-ABL1

USI = unique specimen identifier. D = de novo, R = relapse.

* non-Ph-like by prediction analysis of microarrays.                                                            

Results. All tested PDX models demonstrated inhibition of leukemia proliferation and abrogation of activated signaling with PI3K pathway inhibition. Gedatolisib treatment resulted in near-eradication of leukemia in CRLF2/JAK-mutant models (n=7) with mean 92.2% (range 86.0-99.4%) leukemia reduction vs vehicle treatment (p<0.0001), as well as prolonged animal survival (p<0.005). Gedatolisib also inhibited leukemia proliferation in ABL/PDGFR-mutant models (n=3) with mean 66.9 (range 42.0-87.6) leukemia reduction vs vehicle controls (p<0.0001). BYL719, idelalisib, and AZD2014 monotherapy decreased ALL burden in JAK-mutant models with mean 52.7% (range 27.5-72.9%), 41.6% (range 22.6-53.1%), and 56.3% (range 20.1-88.7%) reduction, respectively. These three inhibitors had variable potency in ABL/PDGFR-mutant models with 39.1% (range 11.4-71.2%) ALL reduction with BYL719, 0.4% (range -25.2-13.9%) reduction with idelalisib, and 14.5% (range -15.5-30.7%) reduction with AZD2014 treatment vs vehicle controls. Leukemias with greatest basal signaling activation measured by phosphoflow cytometry demonstrated greatest leukemia reduction with PI3K pathway inhibition in the treatment studies, suggesting that basal phosphoprotein levels may be predictive biomarkers of response.

Given the recently reported efficacy of JAK inhibition in CRLF2/JAK-mutant and SRC/ABL inhibition in ABL/PDGFR-mutant Ph-like ALL PDX models, we hypothesized that dual pathway inhibition would be more effective than monotherapy. Indeed, simultaneous treatment with gedatolisib and ruxolitinib additively or synergistically reduced ALL burden in all tested CRLF2/JAK-mutant models vs gedatolisib or ruxolitinib monotherapy (p<0.001). Similarly, combined gedatolisib and dasatinib treatment induced markedly greater anti-leukemia efficacy in ABL/PDGFR-mutant models vs. either inhibitor alone (p<0.001).

Conclusions. PI3K pathway inhibition is a biochemically relevant therapeutic approach for Ph-like ALL. Dual PI3K/mTOR inhibition with gedatolisib monotherapy potently inhibited leukemia proliferation and demonstrated additive or synergistic activity in combination with JAK or SRC/ABL inhibition in JAK-mutant or ABL/PDGFR-mutant Ph-like ALL, respectively. These data provide compelling rationale for testing combinations of signal transduction inhibitors without or with cytotoxic chemotherapy in children and adults with Ph-like ALL.