Oral Arsenic Trioxide in the Frontline Treatment of Newly-Diagnosed Acute Promyelocytic Leukemia: A 5-Year Prospective Cohort Study

Introduction and objectives

Intravenous arsenic trioxide (As₂O₃) has been tested in the frontline treatment of newly-diagnosed acute promyelocytic leukemia (APL). We have formulated an oral preparation of As₂O₃ (oral-As₂O₃), and shown that it is efficacious for APL in when used in maintenance and re-induction protocol. In this study, we evaluated the use of oral-As₂O₃ in the frontline treatment of APL. The main objective of the study was to define the impact of upfront use of oral-As₂O₃ in newly-diagnosed patients.

Methods

Patients.Patients with newly-diagnosed APL in eight regional hospitals in Hong Kong were identified.

Arsenic-induction cohort. Induction comprised oral-As₂O₃ (10 mg/day, reduced to 5 mg/day in patients with serum creatinine >upper reference value; as 1 mg/ml solution), ATRA (45 mg/m²/day in 2 divided doses) and ascorbic acid (1 g/day) (AAA regimen), administered for six weeks. Three days of daunorubicin (50 mg/m²/day) were administered to patients aged <70 years, and omitted for those aged ≥70 years. For patients not receiving daunorubicin, if the white blood cell count (WBC) rose to >5 x 10⁹/L, hydroxyurea (2-3 g/day) was used for cytoreduction.

Consolidation regimen. On confirmation of CR1, two monthly consolidations comprising daunorubicin (50 mg/m²/day x 2) and cytarabine (100 mg/m²/day x 5) were administered to patients aged <70 years. For patients aged ≥70 years, consolidation with two monthly cycles of AAA (2 weeks/cycle) was instead administered. Bone marrow examination was performed two weeks after completion of consolidation.

Maintenance regimen. All patients received maintenance AAA, given two weeks every eight weeks for two years.

Results

In a 5-year period (January 1, 2013-March 31, 2018), there were 104 consecutive cases of newly-diagnosed APL. Five patients died at presentation due to disease complications.

Arsenic-induction cohort. 62 newly-diagnosed patients (24 men, 38 women) at a median age of 52 (22-85) years, 36% having high-risk features, consented to and received oral-As₂O₃ induction. Complete remission (CR) rate was 100%. After a median of 29 (5-69) months, there were no relapses, so that conventional risks (age, leucocyte and platelet counts, FLT3 mutations) were not relevant. The leukemia-free survival (LFS) and overall survival (OS) at 3 years were both 100%. There was no leukemia- or treatment-related mortality. The most common non-hematologic AE was hepatotoxicity, presenting as transaminitis in all cases (grade 1/2, N=19, 30.6%; grade 3/4, N=16, 25.8%). Hyperbilirubinemia did not occur. In the sixteen cases with grade 3/4 transaminitis, oral-As₂O₃ was temporarily withheld for a median of 2 (1-5) days. With improvement in transaminases, oral-As₂O₃ was successfully re-administered without recurrence of transaminitis. Three patients developed QTc prolongation to a median maximum of 490 (480-496) ms. It was transient and self-limiting without need for oral-As₂O₃ dose interruption. Arrhythmias and heart failure were not observed. Headache, nausea and vomiting were other minor and less common AEs, which responded to symptomatic treatment.

Non-arsenic-induction cohort. A contemporaneous cohort of 37 newly-diagnosed patients (15 men, 22 women; median age: 51, 23-78 years), 46% having high-risk features, did not consent to oral-As₂O₃ induction, and received similar induction (without oral-As₂O₃) and consolidation. CR rate was 100%. Most of them (89%) then received oral-As₂O₃ maintenance. After a median of 44 (6-69) months, there were three relapses (8%), all achieving remission again with oral-As₂O₃-containing salvage regimens.

Arsenic induction versus non-arsenic induction. To define the role of oral-As₂O₃ during induction, we selected patients from the arsenic-induction cohort, who received daunorubicin induction, daunorubicin/cytarabine consolidation and AAA maintenance (N=50), and compared them with patients from the non-arsenic-induction cohort, who received the same induction, consolidation and AAA maintenance (N=32). The 5-year OS was comparable (arsenic-induction subgroup, 100% versus non-arsenic-induction subgroup, 96.8%; P=0.25). However, the 5-year LFS of the arsenic-induction subgroup was significantly superior to that of the non-arsenic-induction subgroup (100% versus 89.7%, P=0.04).

Conclusion

In conclusion, oral-As₂O₃ induction for newly-diagnosed APL was safe and reduced relapses.