Consolidation outcomes in CPX-351 versus cytarabine/daunorubicin-treated older patients with high-risk/secondary acute myeloid leukemia

Abstract

CPX-351 is a dual-drug liposomal encapsulation of cytarabine/daunorubicin. In a phase 3 study (ClinicalTrials.gov Identifier: NCT01696084), patients aged 60–75 years with newly diagnosed, high-risk/secondary AML received 1–2 induction cycles with CPX-351 or 7 + 3 chemotherapy; those achieving complete remission (including with incomplete platelet or neutrophil recovery) could receive up to 2 consolidation cycles with CPX-351 or 5 + 2 chemotherapy, respectively. In this exploratory analysis of the subgroup of patients who received consolidation, median overall survival was prolonged among patients receiving CPX-351 induction/consolidation versus 7 + 3/5 + 2 (25.43 vs. 8.53 months; HR = 0.44 [95% CI: 0.25–0.77]). The safety profile of CPX-351 consolidation was consistent with that of the overall study. Outpatient administration of CPX-351 consolidation occurred in 51%–61% of patients and did not diminish overall survival. These findings suggest consolidation with CPX-351 in this patient population contributed to the prolonged overall survival versus 7 + 3/5 + 2, building upon findings from the overall study population, and provide evidence that, with careful monitoring, some patients can successfully receive CPX-351 as outpatients.

Keywords:

• CPX-351
• consolidation
• daunorubicin
• cytarabine
• acute myeloid leukemia

Introduction

Outcomes for patients with acute myeloid leukemia (AML) remain poor, particularly for older adults and those who are diagnosed with secondary AML (sAML; i.e. AML that evolves from a prior hematologic disorder or as a late complication of chemotherapy, radiation, or immunotherapy [1]) or have high-risk biologic features (e.g. adverse or complex cytogenetics or phenotype associated with multidrug resistance) [2–8]. For decades, the standard-of-care induction therapy for patients with newly diagnosed AML, including those with high-risk/sAML, who are deemed fit enough for intensive therapy has been continuous, multiple-day cytarabine infusion in combination with an anthracycline (such as daunorubicin), commonly known as the 7 + 3 regimen [9,10]. Patients who achieve complete remission (CR) or CR with incomplete platelet or neutrophil recovery (CRi) to induction therapy require post-remission consolidation with continued intensive chemotherapy and/or hematopoietic cell transplantation (HCT) in order to prolong remission and prevent recurrence of AML. Chemotherapy consolidation has generally consisted of multiple cycles of high- or intermediate-dose cytarabine (HiDAC and IDAC, respectively), or lower-dose cytarabine combined with an anthracycline (e.g. 5 + 2 regimen).

CPX-351 (Vyxeos^® [daunorubicin and cytarabine liposome for injection]; Jazz Pharmaceuticals, Palo Alto, CA) is approved by the US Food and Drug Administration and the European Medicines Agency for the treatment of adults with newly diagnosed therapy-related AML or AML with myelodysplasia-related changes [11,12]. The CPX-351 liposome encapsulates cytarabine and daunorubicin at a synergistic 5:1 molar drug ratio, which is maintained in human plasma for ≥24 hours after CPX-351 administration, with drug exposure persisting for approximately 7 days [13]. In the pivotal randomized phase 3 study of CPX-351 versus conventional cytarabine and daunorubicin (7 + 3 regimen for induction and 5 + 2 regimen for consolidation), CPX-351 was associated with significantly prolonged overall survival (OS; 9.56 vs. 5.95 months; 1-sided p = .003) and higher rates of induction CR + CRi (47.7% vs. 33.3%; 2-sided p = .016) versus 7 + 3 in older adults with newly diagnosed high-risk/sAML [14]. The overall safety profile of CPX-351 in this study was consistent with the known profile of 7 + 3 chemotherapy [14]. In addition, CPX-351 is administered as one 90-minute infusion (on Days 1, 3, and 5 for induction and Days 1 and 3 for consolidation) and thus has the potential to be given in the outpatient setting [14]. Here we report the results of an exploratory post hoc analysis of the phase 3 study that assessed outcomes in patients who achieved CR or CRi with induction and continued CPX-351 from induction through consolidation therapy versus those who received 7 + 3 induction followed by 5 + 2 consolidation. This analysis also evaluated the impact of administering CPX-351 consolidation in the outpatient setting, as a large proportion of patients from the phase 3 study received CPX-351 consolidation in the outpatient setting.

Materials and methods

Study design

In this open-label, controlled, multicenter, phase 3 study (ClinicalTrials.gov Identifier: NCT01696084), patients were randomized 1:1 to receive induction therapy with CPX-351 or 7 + 3 chemotherapy and stratified by age (60–69 vs. 70–75 years) and AML subtype, as described previously [14]. Patients could receive up to 2 cycles of induction with 100 units/m² CPX-351 (i.e. 100 mg/m² cytarabine plus 44 mg/m² daunorubicin) administered as a 90-minute infusion on Days 1, 3, and 5 (Days 1 and 3 for second induction) or with the 7 + 3 regimen, consisting of 100 mg/m²/day continuous infusion of cytarabine for 7 days (5 days for second induction) plus 60 mg/m² daunorubicin on Days 1, 2, and 3 (Days 1 and 2 for second induction). Patients who achieved CR or CRi could receive up to 2 cycles of consolidation with 65 units/m² CPX-351 (i.e. 65 mg/m² cytarabine plus 29 mg/m² daunorubicin) administered as a 90-minute infusion on Days 1 and 3 or with the 5 + 2 regimen, consisting of 100 mg/m²/day continuous infusion of cytarabine for 5 days plus 60 mg/m² daunorubicin on Days 1 and 2. The choice of inpatient or outpatient administration for induction and consolidation was left to the discretion of the investigators. Outpatient administration of CPX-351 was feasible due to the 90-minute duration of infusion. Patients who achieved a CR or CRi could also be considered for HCT, per institutional criteria. Use of prophylactic anti-infectives, per institutional guidelines, was highly recommended for profound neutropenia (until absolute neutrophil count ≥500/μL). Use of growth factors, per institutional guidelines and according to American Society of Clinical Oncology criteria [15], was also permitted.

This study was conducted in accordance with the principles of the Declaration of Helsinki and the International Conference on Harmonization Good Clinical Practice guidelines. The study protocol and all modifications were approved by the institutional review board or independent ethics committee at each study site. All patients provided written informed consent prior to study participation, and the data presented are fully anonymized.

Eligibility criteria

Study eligibility criteria have been described in detail elsewhere [14]. Briefly, patients were aged 60 to 75 years and newly diagnosed with a pathological diagnosis of AML according to World Health Organization 2008 criteria (>20% blasts in peripheral blood or bone marrow) [1] and had confirmed high-risk/sAML based on prior cytotoxic treatment (i.e. therapy-related AML), a history of myelodysplastic syndrome (MDS) or chronic myelomonocytic leukemia (CMML), or de novo AML with cytogenetic changes linked to MDS. Patients must have been deemed to have the ability to tolerate intensive AML chemotherapy. Key exclusion criteria included acute promyelocytic leukemia t(15;17) or favorable cytogenetics at screening, prior treatment intended as induction therapy for AML (except hydroxyurea was permitted), and any active secondary malignancy or central nervous system leukemia.

Endpoints and assessments

As reported previously [14], the primary study endpoint was OS. Additional key efficacy endpoints included remission rate (CR, CR + CRi; assessed according to the 2003 International Working Group Criteria for AML [16]) and the rate of patients undergoing HCT. Safety outcomes included adverse events reporting and laboratory assessments.

Statistical analyses

The efficacy and safety exploratory post hoc analyses described herein were performed in the subset of patients who were treated with consolidation therapy in this study. Any patient who spent ≥1 day in the hospital during the treatment cycle (i.e. during the chemotherapy administration time frame [Days 1–3 for CPX-351 and Days 1–5 for 5 + 2]) for the indicated cycle was classified as an ‘inpatient’ for that treatment cycle, regardless of the actual setting at the start or end of the given consolidation cycle administration. Patients who started and completed consolidation treatment without hospitalization during the treatment cycle were classified as an ‘outpatient’ for that treatment cycle, even if, after completion of treatment, management of adverse events led to hospital admission.

The statistical approach used in this study has been described previously [14]. Briefly, the Kaplan-Meier method was used to estimate time-to-event endpoints. Hazard ratios (HRs) and 95% confidence intervals (CIs) for time-to-event endpoints were estimated using a Cox proportional hazards regression model. Odds ratios (ORs) and 95% CIs for remission rates and other binary measures were also calculated.

Results

Patients

A total of 309 patients were enrolled in the study, and 304 patients received 1 or 2 cycles of induction with CPX-351 (n = 153) or 7 + 3 (n = 151). Seventy-three (48%) patients achieved CR or CRi with CPX-351 induction, and 52 (33%) patients achieved CR or CRi with 7 + 3 induction. CPX-351 consolidation was administered to 49 patients, including 1 patient who had not achieved a CR or CRi during induction, and 5 + 2 consolidation was administered to 32 (62%) patients. The majority of patients who achieved CR or CRi but did not receive consolidation went on to receive an HCT instead (CPX-351: 16/25 [64%]; 7 + 3: 12/20 [60%]). The remaining patients who achieved CR or CRi but did not receive consolidation discontinued treatment (CPX-351: 6/25 [24%]; 7 + 3: 4/20 [20%]), relapsed (CPX-351: 2/25 [8%]; 7 + 3: 4/20 [20%]), or completed treatment (CPX-351: 1/25 [4%]; 7 + 3: 0).

Demographic and clinical characteristics of patients who received consolidation were generally balanced between treatment arms (Table 1) and generally consistent with the overall study population [14], with the exception that there appeared to be slightly fewer patients with AML with antecedent MDS who had received prior treatment with hypomethylating agents (HMA) among the consolidation population. Among patients who achieved a CR or CRi but did not receive consolidation, higher proportions of patients were between 60 and 69 years of age (CPX-351: 18/25 [72%]; 7 + 3: 17/20 [85%]) and had a history of MDS with prior hypomethylating agent treatment (CPX-351: 10/25 [40%]; 7 + 3: 9/20 [45%]) compared with patients who received consolidation.

Table 1. Baseline demographic and clinical patient characteristics.

	All consolidation		Inpatient^a		Outpatient
	CPX-351 (n = 49)	5 + 2 (n = 32)	CPX-351 (n = 24)	5 + 2 (n = 30)	CPX-351 (n = 25)	5 + 2 (n = 2)
Age
Median (min-max), years	68 (61–75)	68.5 (61–75)	67.5 (61–75)	68.5 (61–75)	69 (61–75)	66 (63–69)
60–69 years, n (%)	30 (61)	20 (63)	17 (71)	18 (60)	13 (52)	2 (100)
70–75 years, n (%)	19 (39)	12 (38)	7 (29)	12 (40)	12 (48)	0
Male sex, n (%)	30 (61)	18 (56)	16 (67)	16 (53)	14 (56)	2 (100)
Race, n (%)
Asian	3 (6)	1 (3)	2 (8)	1 (3)	1 (4)	0
Black or African American	0	2 (6)	0	2 (7)	0	0
White	43 (88)	29 (91)	20 (83)	27 (90)	23 (92)	2 (100)
Other	3 (6)	0	2 (8)	0	1 (4)	0
ECOG PS, n (%)^b
0	15 (31)	14 (44)	6 (25)	13 (43)	9 (36)	1 (50)
1	30 (61)	16 (50)	15 (63)	15 (50)	15 (60)	1 (50)
2	4 (8)	2 (6)	3 (13)	2 (7)	1 (4)	0
AML subtype, n (%)
Therapy-related AML	10 (20)	7 (22)	5 (21)	6 (20)	5 (20)	1 (50)
Antecedent MDS
With prior HMA treatment	9 (18)	9 (28)	7 (29)	9 (30)	2 (8)	0
Without prior HMA treatment	8 (16)	7 (22)	4 (17)	7 (23)	4 (16)	0
Antecedent CMML	4 (8)	3 (9)	3 (13)	3 (10)	1 (4)	0
De novo AML with MDS karyotype	18 (37)	6 (19)	5 (21)	5 (17)	13 (52)	1 (50)
Bone marrow blast percentage, n (%)^c
40%	28 (57)	21 (70)
>40–60%	14 (29)	6 (20)
>60%	7 (14)	3 (10)
Cytogenetic risk by NCCN, n (%)^d
Favorable/intermediate	23 (49)	18 (64)
Unfavorable	24 (51)	10 (36)
FLT3 mutation, n (%)	9 (20)	3 (11)
White blood cell count, n (%)^b
<20 × 10^9/L	46 (94)	29 (91)
≥20 × 10^9/L	3 (6)	3 (9)
Platelet count, n (%)^b
≤50 × 10^9/L	28 (57)	16 (50)
>50 × 10^9/L	21 (43)	16 (50)

Min: minimum; max: maximum; ECOG PS: Eastern Cooperative Oncology Group performance status; AML: acute myeloid leukemia; MDS: myelodysplastic syndrome; HMA: hypomethylating agent; CMML: chronic myelomonocytic leukemia; NCCN: National Comprehensive Cancer Network.

^aIf a patient spent ≥1 day in the hospital during the treatment cycle, that patient was classified as an ‘inpatient’ regardless of the actual setting of treatment administration.

^bAt baseline.

^cData are missing for 2 patients in the 5 + 2 arm.

^dData are missing for 2 patients in the CPX-351 arm and 4 patients in the 5 + 2 arm.

Efficacy

Consistent with results for the overall study population [14], median OS was prolonged among patients who received CPX-351 through induction and consolidation versus 7 + 3/5 + 2 (25.43 vs. 8.53 months, respectively; HR = 0.44 [95% CI: 0.25–0.77]; Figure 1). Further, there was a trend toward longer median OS with CPX-351 versus 7 + 3/5 + 2 in patients who received consolidation followed by HCT (not reached vs. 9.82 months; HR = 0.28 [95% CI: 0.11–0.72]). A similar trend was observed in patients who received consolidation without HCT (median OS: 13.67 vs. 8.44 months; HR = 0.62 [95% CI: 0.31–1.26]). Among responders who did not receive consolidation, median OS was 19.15 months with CPX-351 and 13.75 months with 7 + 3 (HR = 0.59 [95% CI: 0.26–1.32]). No patient experienced disease progression/relapse or died during consolidation in either treatment arm. Of the patients who achieved CR or CRi prior to receiving consolidation, relapses were observed following consolidation with both CPX-351 (Consolidation 1: 4/25 [16%]; Consolidation 2: 12/23 [52%]) and 5 + 2 (Consolidation 1: 7/20 [35%]; Consolidation 2: 3/12 [25%]). Of the 40 patients who had achieved CR or CRi with CPX-351 induction and later underwent HCT, 60% received consolidation; of the 24 patients who had achieved CR or CRi with 7 + 3 induction and later underwent HCT, 50% received consolidation with 5 + 2.

Figure 1. OS in patients who received consolidation. Median OS was prolonged in patients who received CPX-351 throughout induction and consolidation (black line) versus those who received 7 + 3/5 + 2 (grey line). This trend was observed in patients who received consolidation with or without HCT (table).

Safety

The safety profile of CPX-351 consolidation was generally consistent with trends observed during the overall study [14]. There were no clinically meaningful differences between treatment arms in the frequencies of treatment-emergent adverse events (TEAEs; all grades and grades 3–5) during consolidation, with variations likely attributable to the relatively small numbers of patients, particularly for the second consolidation cycle (Table 2). During consolidation, cardiac TEAEs occurred in 12% and 22% of patients in the CPX-351 and 7 + 3/5 + 2 arms, respectively, with no individual cardiac TEAE reported in >1 patient in the CPX-351 arm. The most frequently reported grade 3 to 5 TEAEs during consolidation were febrile neutropenia (CPX-351: 29%; 5 + 2: 25%), pneumonia (8%; 6%), cellulitis (8%; 3%), fatigue (6%; 3%), sepsis (4%; 6%), staphylococcal bacteremia (0%; 6%), and peripheral edema (0%; 6%). Febrile neutropenia (CPX-351: 16%; 5 + 2: 19%) was also the most frequently reported serious TEAE. No patient died due to any cause during the first consolidation cycle in either treatment arm.

Table 2. Summary of TEAEs in patients during consolidation treatment.

Frequency of TEAEs, n (%)	During Consolidation 1		During Consolidation 2
	CPX-351 (n = 49)	5 + 2 (n = 32)	CPX-351 (n = 23)	5 + 2 (n = 12)
Any TEAE^a	42 (86)	29 (91)	20 (87)	12 (100)
Febrile neutropenia	10 (20)	6 (19)	6 (26)	3 (25)
Nausea	8 (16)	3 (9)	3 (13)	5 (42)
Constipation	7 (14)	7 (22)	5 (22)	3 (25)
Fatigue	6 (12)	9 (28)	6 (26)	2 (17)
Pyrexia	6 (12)	2 (6)	2 (9)	2 (17)
Epistaxis	6 (12)	1 (3)	2 (9)	1 (8)
Diarrhea	4 (8)	9 (28)	2 (9)	1 (8)
Peripheral edema	4 (8)	7 (22)	4 (17)	1 (8)
Decreased appetite	4 (8)	3 (9)	4 (17)	1 (8)
Chills	3 (6)	3 (9)	7 (30)	1 (8)
Rash	3 (6)	1 (3)	1 (4)	2 (17)
Arthralgia	3 (6)	0	3 (13)	0
Dizziness	2 (4)	6 (19)	5 (22)	0
Headache	2 (4)	3 (9)	2 (9)	2 (17)
Oropharyngeal pain	1 (2)	2 (6)	2 (9)	3 (25)
Skin lesion	0	4 (13)	1 (4)	0
Atrial fibrillation	0	3 (9)	0	2 (17)
Rhinorrhea	0	1 (3)	0	2 (17)
Mucosal inflammation	0	0	3 (13)	1 (8)
Grade 3–5 TEAE^a	19 (39)	12 (38)	13 (57)	6 (50)
Febrile neutropenia	10 (20)	6 (19)	6 (26)	3 (25)
Serious TEAE^a	15 (31)	11 (34)	10 (43)	5 (42)
Febrile neutropenia	6 (12)	4 (13)	2 (9)	3 (25)

TEAE: treatment-emergent adverse event.

^aTable includes all TEAEs reported in ≥10% of patients in either treatment group during Consolidation 1 or 2.

Similar to findings for the overall population [14], CPX-351 consolidation was associated with slower median time to recovery of neutrophils and platelets compared with 5 + 2 chemotherapy (Table 3). This trend was observed among subpopulations who had achieved CR or CR + CRi to induction therapy and irrespective of whether patients received 1 or 2 consolidation cycles.

Table 3. Median time to neutrophil and platelet recovery in patients who achieved remission with 1 and 2 consolidation cycles.

	1 consolidation cycle		2 consolidation cycles^a
	CPX-351	5 + 2	CPX-351	5 + 2
CR, n	19	18	21	12
Median (Q1–Q3) time to ANC recovery ≥1000 neutrophils/µL, days	41 (35–48)	32.5 (29–43)	35 (28–41)	33.5 (27–39)
Median (Q1–Q3) time to platelet recovery ≥100,000 platelets/µL, days	43 (34–48)	30 (29–45)	37 (33–49)	35.5 (31–41)
CR + CRi, n	25	20	23	12
Median (Q1–Q3) time to ANC recovery ≥500 neutrophils/µL, days	35 (34–44)	32.5 (29–43)	36 (28–48)	33.5 (27–39)
Median (Q1–Q3) time to platelet recovery ≥50,000 platelets/µL, days	40 (35–43)	30 (28–41)	36 (28–47)	31.5 (27–37)

CR: complete remission; Q: quartile; ANC: absolute neutrophil count; CRi: CR with incomplete platelet or neutrophil recovery. One patient received consolidation but had no CR or CRi and was not included in this analysis.

^aMeasured from the treatment start date.

Outpatient administration

CPX-351 consolidation was administered completely in the outpatient setting for 51% of patients during the first consolidation cycle and 61% of patients during the second consolidation cycle (Table 4). In contrast, 5 + 2 chemotherapy was rarely administered completely in the outpatient setting (6% of patients during the first consolidation cycle and no patients during the second cycle). Of note, for this analysis, patients were only classified as outpatients if they started and completed treatment without hospitalization during the treatment cycle, so some patients classified as inpatients may have received a part of their treatment in the outpatient setting.

Table 4. Hospitalization during consolidation.

	CPX-351	5 + 2
Consolidation 1, n	49	32
Outpatient administration, n (%)	25 (51)	2 (6)
Inpatient administration, n (%)^a	24 (49)	30 (94)
Median (min–max) days of treatment cycle spent as inpatient	4.0 (0–28)	7.0 (0–31)
Median (min–max) days of treatment cycle spent as inpatient PPY	33.2 (0–310)	71.0 (0–315)
Median (min–max) days of treatment cycle spent in the ICU	0 (0–0)	0 (0–11)
Median (min–max) days of treatment cycle spent in the ICU PPY	0 (0–0)	0 (0–112)
Consolidation 2, n	23	12
Outpatient administration, n (%)	14 (61)	0 (0)
Inpatient administration, n (%)^a	9 (39)	12 (100)
Median (min–max) days of treatment cycle spent as inpatient	4.0 (0–31)	7.0 (5–28)
Median (min–max) of treatment cycle spent as inpatient PPY	44.3 (0–343)	73.1 (52–292)
Median (min–max) days of treatment cycle spent in the ICU	0 (0–10)	0 (0–17)
Median (min–max) days of treatment cycle spent in the ICU PPY	0 (0–111)	0 (0–177)

Min: minimum; max, maximum; PPY: per patient-year; ICU: intensive care unit.

^aIf a patient spent ≥1 day in the hospital during the treatment cycle, that patient was classified as an ‘inpatient’ regardless of the actual setting at the start of consolidation treatment administration.

Baseline characteristics were generally similar between patients who received CPX-351 consolidation in the outpatient versus inpatient setting (Table 1). For instance, the median age fell between 66 and 69 years for all consolidation groups. However, among patients who received CPX-351 consolidation, there appeared to be a greater proportion of patients with de novo AML with MDS karyotype and fewer patients with antecedent MDS with prior HMA treatment among those who received outpatient versus inpatient consolidation. Additionally, a greater proportion of patients who received outpatient CPX-351 consolidation were female and began treatment with an Eastern Cooperative Oncology Group performance status of 0 to 1 than patients who received inpatient CPX-351 consolidation.

The median number of days spent in the hospital was lower in patients treated with CPX-351 versus 5 + 2 (Table 4). This difference between CPX-351 and 5 + 2 was notable when number of hospital days per patient-year (PPY) was calculated to normalize for the length of the treatment cycle and was observed in both Consolidation 1 (33.2 vs. 71.0 days, respectively) and Consolidation 2 (44.3 vs. 73.1 days). In both treatment arms, the median number of days spent in the ICU during consolidation was low (Table 4).

Among patients who received Consolidation 1 in the inpatient setting, median OS was 14.72 months with CPX-351 (n = 24) and 9.26 months with 5 + 2 (n = 30; HR = 0.55 [95% CI: 0.25–1.21]; Table 5). Similarly, among those who received Consolidation 2 in the inpatient setting, median OS was not reached with CPX-351 (n = 9) versus 14.31 months with 5 + 2 (n = 12; HR = 0.45 [95% CI: 0.09–2.36]). Median OS was 25.43 months among patients who received Consolidation 1 with CPX-351 in the outpatient setting (n = 25) and was 26.32 months among patients who received Consolidation 2 with CPX-351 in the outpatient setting (n = 14). Only 2 patients received 5 + 2 as outpatients in Consolidation 1 (median OS: 6.87 months); no patient received Consolidation 2 with 5 + 2 in the outpatient setting.

Table 5. OS by setting of consolidation administration.

	Inpatient^a		Outpatient
	CPX-351	5 + 2	CPX-351	5 + 2
Consolidation 1, n/N (%)^b	24/49 (49)	30/32 (94)	25/49 (51)	2/32 (6)
Median OS, mo	14.72	9.26	25.43	6.87
HR (95% CI)	0.55 (0.25–1.21)		0.10 (0.01–1.11)
Consolidation 2, n/N (%)	9/23 (39)	12/12 (100)	14/23 (61)	0/12 (0)
Median OS, mo	NR	14.31	26.32	–
HR (95% CI)	0.45 (0.09–2.36)		–

OS: overall survival; HR: hazard ratio; CI: confidence interval; NR: not reached.

^aIf a patient spent ≥1 day in the hospital during the treatment cycle, that patient was classified as an ‘inpatient’ regardless of the actual setting of treatment administration.

^bHere, n denotes the number of patients receiving a specific consolidation cycle in a given setting (inpatient vs. outpatient), and N denotes the number of patients in that treatment arm who received that consolidation cycle.

The safety profile of CPX-351 was generally similar when administered in the inpatient versus outpatient settings. There was a slightly higher rate of serious TEAEs in the outpatient (n = 9/25 [36%]) versus inpatient (n = 6/24 [25%]) setting, although no difference in the frequency of febrile neutropenia (12% vs. 13%, respectively) was seen. In the inpatient setting, 16 of 24 (67%) patients treated with CPX-351 and 19 of 30 (63%) patients treated with 5 + 2 were rehospitalized following the first consolidation cycle. In the outpatient setting, 11 of 25 (44%) patients treated with CPX-351 were rehospitalized following the first consolidation cycle, and 7 of these patients were rehospitalized due to a serious TEAE.

Discussion

This exploratory post hoc analysis of a large, multicenter, randomized controlled study was performed to evaluate post-remission consolidation outcomes for CPX-351 versus 5 + 2 following induction with CPX-351 and 7 + 3, respectively. Consistent with the overall study population of older patients with newly diagnosed high-risk/sAML [14], median OS was longer for patients who received CPX-351 consolidation compared with those who received 5 + 2 consolidation. This finding is noteworthy because, with the exception of 1 patient in the CPX-351 arm, the patients included in this analysis had already achieved a CR or CRi during induction. Although minimal residual disease (MRD) results are not available for this phase 3 study, these data suggest patients in the CPX-351 arm may have achieved a deeper level of molecular response than those in the 7 + 3/5 + 2 arm. This hypothesis is supported by data from a recent study of MRD using next-generation sequencing in French patients who received CPX-351 for the treatment of AML, in which 50% of patients evaluated for MRD achieved MRD-negative status [17]. Furthermore, a trend toward a benefit was observed both in patients who subsequently underwent HCT and those who did not. These results highlight the importance of the continuum of treatment with CPX-351 for 1 to 2 induction cycles followed by continued CPX-351 consolidation treatment.

The safety profile of CPX-351 during consolidation was generally comparable with that of conventional 5 + 2 consolidation and was consistent with the safety profile of CPX-351 observed in the overall study population [14]. As expected and observed previously during induction, median time to recovery of neutrophils and platelets was longer with CPX-351 consolidation compared with 5 + 2 chemotherapy, regardless of the number of consolidation cycles received. The prolonged myelosuppression that is consistently observed with CPX-351 treatment may be attributable to the prolonged exposure of both malignant myeloblasts and normal hematopoietic precursors. The accumulation of cytarabine and daunorubicin within the bone marrow and the resulting prolonged exposure to chemotherapy results in both a relatively selective cytotoxic effect on leukemia myeloblasts and a delay in recovery from neutropenia and thrombocytopenia [13,18]. The level of persistent drug exposure in the bone marrow at a synergistic drug ratio that is achieved with CPX-351 cannot be achieved with the administration of free cytarabine and daunorubicin due to the differences in their independent pharmacokinetic properties [13,18,19]. Knowing that the timing for neutrophil and platelet count recovery is expected to differ with CPX-351 versus traditional chemotherapeutic regimens, the optimal timing for monitoring patients who receive CPX-351 may also differ. In the outpatient setting, patients receiving CPX-351 may require longer and/or more diligent follow-up given prolonged cytopenia.

Anthracyclines such as daunorubicin have been shown to induce cardiotoxicity, particularly with increasing exposure. However, the pharmacokinetics of CPX-351 suggest off-target tissue exposure is generally limited [20]. In a recently published phase 2 study that characterized cardiac repolarization in patients with acute leukemia who received CPX-351, no clinically meaningful changes in corrected QT interval were observed, even among those patients who received multiple cycles of induction/consolidation [21]. Similarly, no clinically significant effects on heart rate, QRS interval, or PR interval were observed. In this phase 3 study, cardiac disorders occurred with a similar frequency in both treatment arms during induction (CPX-351: 37%; 7 + 3: 38%) but were observed in 12% versus 22% of patients during CPX-351 and 5 + 2 consolidation, respectively, suggesting the liposomal design and pharmacokinetics of CPX-351 may mitigate cardiac toxicity.

Although the 5 + 2 chemotherapy regimen is typically administered in an inpatient hospital setting due to the 5-day continuous cytarabine infusion, consolidation regimens that are composed of discreet periodic infusions, such as HiDAC and IDAC, can be administered in an outpatient setting. Several studies involving relatively small sample sizes have demonstrated that outpatient consolidation using HiDAC or IDAC is feasible for some patients [22–24], although at least one study reported that hospital admission was required by 21% of patients [24]. CPX-351 consolidation is administered as one 90-minute infusion on each of Days 1 and 3, and therefore has the potential to be administered in the outpatient setting. The current analysis of consolidation outcomes thus also evaluated outcomes among subsets of patients who remained in the outpatient setting throughout the consolidation treatment cycle versus those who were inpatients for all or part of consolidation treatment. In this study, approximately half of CPX-351 patients received consolidation entirely in the outpatient setting. In contrast, very few patients received 5 + 2 consolidation entirely in the outpatient setting, preventing any meaningful comparison of outcomes for patients receiving outpatient consolidation between treatment arms.

Importantly, median OS was not diminished when CPX-351 consolidation was administered in the outpatient versus inpatient setting. Administration of CPX-351 consolidation to some patients in the outpatient setting contributed to lower the overall number of days spent in the hospital during the treatment cycle with CPX-351 versus 5 + 2. The proportion of patients with serious TEAEs was slightly higher among patients who received CPX-351 consolidation in the outpatient versus inpatient setting, and this may be due, in part, to the occurrence of TEAEs requiring a move from the outpatient setting to the hospital, which is one of the criteria for classifying a TEAE as a serious TEAE. Taken together, these results establish outpatient administration of CPX-351 as a feasible alternative to consolidation with CPX-351 in the inpatient setting.

CPX-351 consolidation has potentially important implications for patient and caregiver convenience, particularly when given in the outpatient setting, as patients can stay in their home (with outpatient visits). Unlike 5 + 2 consolidation, patients receiving CPX-351 consolidation also do not receive continuous multiple-day infusions, even if they receive treatment in the inpatient setting. Quality of life data for CPX-351 treatment are not currently available; however, a recent study used a vignette-based, time tradeoff utility valuation interview approach with members of the United Kingdom general population to evaluate the perceived disutility associated with AML treatment. Although all chemotherapy regimens were associated with disutility, regimens with shorter hospitalization and less time-intensive infusion (i.e. CPX-351) were generally perceived as preferable [25]. In this study, the ability to administer CPX-351 consolidation in the outpatient setting likely led to reduced hospitalization days with CPX-351 versus 7 + 3 (Consolidation 1: 4.0 vs. 7.0 days; Consolidation 2: 4.0 vs. 7.0 days), which could potentially offset the cost of CXP-351.

In summary, among patients who received both induction and consolidation with CPX-351, OS was significantly improved versus conventional 7 + 3/5 + 2 chemotherapy in this population of older adults with newly diagnosed high-risk/sAML. These findings build upon the longer OS and higher rate of CR observed with CPX-351 versus 7 + 3 in the overall study population [14]. In addition, outpatient administration of CPX-351 consolidation did not diminish the efficacy of CPX-351 or adversely affect its safety profile, providing evidence that, with careful monitoring, some patients may be able to successfully receive CPX-351 in the outpatient setting, with potential quality of life and cost implications.

Acknowledgments

The authors thank all of the patients who participated in the study and their families, as well as the investigators, nurses, coordinators, and other research staff at each study site. Medical writing and editorial support were provided by William Perlman, PhD, and Kimberly Brooks, PhD, CMPP, of SciFluent Communications under the direction of the authors and were financially supported by Jazz Pharmaceuticals.

Disclosure statement

J Kolitz has received honoraria from Gilead, Magellan, and Novartis; consulting fees from Gilead, Magellan, Novartis, Pharmacyclics, and Seattle Genetics; institutional research funding from Boehringer Ingelheim, Cantex, Erytech, and Millennium; and travel support from Gilead, Novartis, and Seattle Genetics. S Strickland has received consulting fees from Astellas, Baxalta, Boehringer Ingelheim, CTI BioPharma, Daiichi-Sankyo, Sunesis, and Tolero; and institutional research funding from Boehringer Ingelheim and Sunesis. J Cortes has received consulting fees from Amphivena, Ariad, BiolineRX, Bristol-Myers Squibb, Janssen, Novartis, and Pfizer; and institutional research funding from Ambit, Araid, Arog, Astellas, Bristol-Myers Squibb, Celgene, ImmunoGen, Incyte, Jazz Pharmaceuticals, Novartis, Pfizer, and Teva. D Hogge has received consulting fees from Novartis, Roche, and Sanofi. J Lancet has received consulting fees from Biopath, Biosight, Boehringer Ingelheim, Jazz Pharmaceuticals, Celgene, Janssen, Karyopharm, and Novartis; and institutional research funding from Pfizer. S Goldberg has received consulting fees from Bristol-Myers Squibb and Novartis; participated in spearkers’ bureaus for Ariad, Bristol-Myers Squibb, Celgene, and Novartis; received institutional research funding from Ambit, Astellas, Bristol-Meyers Squibb, Celgene, Jazz Pharmaceuticals, Novartis, and Pfizer; provided expert testimony for Novartis; and is an employee of and holds stock ownership in COTA. K Villa, M Chiarella, and A Louie are former employees of and hold stock ownership in Jazz Pharmaceuticals; A Louie additionally holds patents/royalties with Jazz Pharmaceuticals. R Ryan is an employee of and holds stock ownership in Jazz Pharmaceuticals. E Ritchie has received consulting fees from Incyte, Celgene, Pfizer, and Novartis; research funding from Pfizer, Novartis, Astellas Pharma, Bristol-Myers Squibb, and NS Pharma; has participated in speakers’ bureaus for Incyte, Celgene, Novartis, and ARIAD Pharmaceuticals; and received travel support from Celgene and Novartis. R Stuart has received research funding from Sunesis Pharmaceuticals, Agios, Astellas Pharma, Bayer AG, Jazz Pharmaceuticals, and Incyte; and travel support, honoraria, and consulting fees from Sunesis Pharmaceuticals.

Data availability statement

All relevant data are provided within the manuscript and supporting files. The study protocol and statistical analysis plan are available as online supplemental content for Lancet JE, et al. JCO. 2018;36(26):2684-2692.

Additional information

Funding

This study was funded by Jazz Pharmaceuticals, Inc.