Pixantrone Instead of Doxorubicin in CHOP-R Regimen Has Comparable Efficacy, Reduced Cardiotoxicity in DLBCL

SAN DIEGO, CA—For patients with untreated diffuse large B-cell lymphoma (DLBCL), CHOP-R (cyclophosphamide, doxorubicin, vincristine, prednisolone + rituximab) is standard therapy. In a study in which pixantrone dimaleate replaced doxorubicin (CPOP-R), comparable efficacy with reduced cardiotoxicity was observed, investigators reported during the 53rd American Society of Hematology Annual Meeting and Exposition.

Pixantrone, a novel aza-anthracenedione, forms stable DNA adducts and, in preclinical lymphoma models, has enhanced efficacy when compared with doxorubicin, with substantially less cardiotoxicity—as determined by overall and serious declines in left ventricular ejection fraction (LVEF), troponin T elevations, and congestive heart failure (CHF) occurrence—than either doxorubicin or mitoxantrone, due to its inability to bind iron or form alcohol metabolites.

Previously, a Phase 1/2 study in 65 patients with aggressive B-cell non-Hodgkin's lymphoma who had relapsed following treatment with CHOP±R found that subsequent treatment in which pixantrone replaced doxorubicin in CHOP led to a complete remission/unconfirmed complete remission (CR/CRu) rate of 52%. Overall response was 80%, median progression-free survival (PFS) was 8.2 months, and overall survival (OS), 17.9 months.

Raoul Herbrecht, MD, Hôpital  de Hautepierre, Strasbourg, France, and colleagues evaluated the efficacy and safety of CPOP-R vs CHOP-R as first-line therapy in patients with DLBCL (CD20+ stage II–IV) in a Phase 2 open-label study conducted in France, Canada, Germany, and the United States. Patients (n=124) were randomized 1:1 to CHOP-R at standard doses (n=63) or CPOP-R, in which pixantrone 150mg/m2 was substituted for doxorubicin (n=61). After four cycles, patients with a CR received two additional cycles; those with a partial response (PR) received four more cycles. All patients were subsequently followed for 36 months, an independent panel-assessed response. Evaluation of noninferiority of CPOP-R to CHOP-R, measured by rates of CR/CRu, required 280 patients; however, due to resource constraints, enrollment stopped after 124 patients and the study was therefore not powered to detect noninferiority.

Secondary endpoints included objective response rate, OS, time to progression (TTP), and safety. A total of 122 patients (98%) were treated. Baseline demographics and disease characteristics were balanced between treatment arms. Median age was 68 years in both arms; 78% of patients in the CHOP-R and 79% in the CPOP-R arms were Ann Arbor stage 3–4; 54% vs 48% had an International Prognostic Index (IPI) score ≥3; and 11% vs 20%, respectively, had 3 or more comorbid conditions.

Median number of cycles delivered was eight in the CHOP-R arm and eight in the CPOP-R arm. In the intent-to-treat population, CR/CRu was 84.1% for CHOP-R and 75.4% for CPOP-R (95% CI: -5.4–22.8). Overall response rate (CR/CRu  + PR) was 90.5% in the CHOP-R arm and 82.0% in the CPOP-R arm. When both arms were compared, PFS was similar (HR 1.02), as was TTP (median not reached in either arm). Median OS was also not reached in either arm (HR=2.37, 95% CI: 1.07–5.28, P=0.029).

In the CHOP-R arm, the three-year survival rate was 85% vs 69% in the CPOP-R group. When stratified by IPI, those with a score ≤2 had three-year survival rates of 87% for CHOP-R and 84% for CPOP-R; for those with an IPI ≥3, survival was 84% vs 51%, respectively. The investigators noted that in the CHOP-R arm, patients with an IPI score of 3 (n=25) had a better survival rate than those with an IPI ≤2, 8% mortality at three years, which they described as unusual. “This historically high survival rate for CHOP-R patients with IPI=3 appeared responsible for the disparate survival rates between CPOP-R and CHOP-R,” they stated.

A history of coronary artery disease, CHF, or myocardial infarction was reported in two patients in the CHOP-R arm and eight in the CPOP-R arm. Grade 3/4 adverse events (AEs) occurred in approximately 85% of patients in both arms. Incidence of Grade 3/4 AEs was similar for neutropenia (61.9% vs 61.0%), febrile neutropenia (15.9% vs 15.3%), thrombocytopenia (6.3% vs 5.1%), and infections (17.5% vs 16.9%). Stomatitis occurred less frequently in the CPOP-R arm, 6.8%, compared with 17.5% in the CHOP-R arm.

LVEF, measured prospectively at intervals through 24 months, was generally stable in patients in the CPOP-R arm; however, they declined significantly from baseline in those in the CHOP-R arm (P<0.05). In the CHOP-R group, eight patients had LVEF decline ≥20% vs one patient in the CPOP-R group (P<0.05).

Elevated troponin T levels were observed in 34.5% of patients in CHOP-R vs 8.5% in CPOP-R (P<0.05); CHF developed in four patients in CHOP-R vs none in CPOP-R. Within 30 days of the last dose, no patient had died in the CHOP-R arm compared with 3 patients who died in the CPOP-R arm. In both arms, the majority of deaths after treatment were related to disease progression.

While noninferiority of CPOP-R to CHOP-R could not be established, the pixantrone-containing regimen as first-line therapy for DLBCL showed comparable efficacy and reduced cardiotoxicity. “Larger studies are required to establish the potential role of the CPOP-R regimen as first-line therapy for DLBCL,” the researchers concluded on their poster.