Single Knock-In PML-RARα Expressing Murine APL Tumors Provide a Unique Model for Studying In Vivo Sensitivity and Resistance to ATRA and Arsenic Trioxide.

Although approximately 70% of patients with acute promyelocytic leukemia (APL) are cured after treatment with all-trans retinoic acid (ATRA) and anthracycline based chemotherapy, those who relapse cannot be cured even with chemotherapy or arsenic trioxide salvage therapy. In order to study the molecular mechanisms of resistance to ATRA, liposomal ATRA, and arsenic trioxide, we have utilized a murine model of APL generated by “knocking in” the PML/RARα gene into the murine cathepsin G locus (Westervelt et al. Blood. 102(5):1857). Of the 76 single knock-in (SKI) mice followed for a median time of 368 days, 30 (39%) suffered from rapid-onset leukocytosis, anemia, thrombocytopenia, and hepato-splenomegaly. Spleens from these leukemic animals were removed and the tumors cells (> 90% CD34⁺/Gr-1⁺ APL cells) were frozen in aliquots and banked for future passive transfer into genetically compatible recipients. The SKI mice exhibited the expected median time to death of 222 days. Attempts to expand these tumors in vitro failed, thus precluding the use of in vitro methods to generate ATRA and arsenic resistant SKI APL tumor cells. An alternate in vitro model of identifying de novo resistant SKI APL tumors to the drugs was established by incubating individual SKI APL tumors with ATRA, liposomal ATRA, and arsenic in vitro for only 3 days and using ELISA to measure upregulation of MMP-9 (gelatinase B), a surrogate marker for neutrophil differentiation. ATRA, liposomal ATRA, and arsenic induced 6.2 ± 4.4, 5.7 ± 3.5, 1.7 ± 1.0 fold increase in accumulated levels of MMP-9, respectively. These data suggest that ATRA and liposomal ATRA are more potent inducers of terminal differentiation than arsenic in these APL tumors. More importantly, the SKI APL tumors were also assessed for in vivo resistance to these agents by injecting 1 x 10⁶ banked APL tumor cells into the peritoneal cavity of wild-type genetically compatible mice. The mice were then treated i.p. with either arsenic trioxide (0.4 mg q.d.), liposomal ATRA (0.4 mg q.d.), or diluent control. Doses were administered daily starting on day +5 until death. Tumors in the recipient mice were tracked by PCR for the transgene and FACS. The tumors proliferated in the peritoneal cavity (days 1–14) followed by migration to the bone marrow and spleen (days 14–28). By the eighth week post-injection, all mice suffered from leukocytosis and eventually died. Of the 22 banked tumors studied, only two tumors showed de novo resistance to ATRA both in vitro and in vivo. We will present data on preliminary RNA profiling and DNA sequencing of the PML-RARα transgenes in these APL tumors and attempt to correlate these results and the results of in vitro induction of MMP-9. The SKI APL tumors will provide us with unique reagents to study the biology of resistance to ATRA and arsenic trioxide.