Phase 3 results for vosaroxin/cytarabine in the subset of patients ≥60 years old with refractory/early relapsed acute myeloid leukemia

Refractory/early relapsed (Ref/eRel) acute myeloid leukemia (AML) in patients ≥60 years old is the most important unmet medical need in the salvage setting, where outcomes are exceptionally poor and no standard of care exists.1 Vosaroxin is a first-in-class anticancer quinolone derivative that intercalates DNA and inhibits topoisomerase II, but has chemical and pharmacologic characteristics that differ from other topoisomerase II inhibitors.2 The randomized phase 3 VALOR study (clinicaltrials.gov Identifier: 01191801) evaluated vosaroxin (90 mg/m cycle 1 [70 mg/m subsequent cycles] by short intravenous [IV] infusion on days 1 and 4) plus cytarabine (1 g/m IV over 2 hours on days 1–5) (vos/cyt) versus placebo plus cytarabine (pla/cyt) in 711 patients ≥18 years old with Ref or first relapsed (Rel) AML. A detailed methodology has been published previously.3 In the primary efficacy analysis, overall survival (OS) was 7.5 months with vos/cyt versus 6.1 months with pla/cyt (unstratified P=0.061; stratified P=0.024).3 The addition of vosaroxin to cytarabine significantly improved the complete response (CR) rate (30% with vos/cyt vs. 16% with pla/cyt; P<0.0001). Prespecified subgroup analyses according to randomization strata (age [<60/≥60 years], disease status [Ref/eRel [within 12 months]/late Rel], and geographic location [USA/outside USA]) demonstrated that the treatment benefit was similar by geographic region, but varied considerably by age and disease status. The OS benefit with vos/cyt was most substantial in patients ≥60 years of age (n = 451; hazard ratio [HR] =0.75 [95% confidence interval [CI]: 0.62–0.92]; P=0.0030) and patients with eRel disease (n = 256; HR = 0.77 [95% CI: 0.59–1.00]; P=0.039), patient groups that are typically treatment resistant.3

These results prompted further analyses in patient subgroups as defined by age and disease status. When results in patients ≥60 years of age were analyzed by disease status, a substantial improvement in the OS and CR rate was observed with vos/cyt vs. pla/cyt in patients age ≥60 years with Ref/eRel disease. In contrast, an improvement of OS in patients ≥60 years with late Rel disease was not demonstrated (n = 87; median OS 9.2 months vs. 9.8 months, respectively; HR = 1.06; P=0.82), despite a significant improvement in the CR rate (vos/cyt: 56.8% vs. pla/cyt: 27.9%; P=0.0064); small patient numbers and high rates of subsequent therapy, including transplant, may have confounded the OS analysis in this group. Based on the OS and CR benefit observed in patients ≥60 years with Ref/eRel disease, an exploratory analysis of the risk-benefit profile in patients ≥60 years of age with duration of first CR <12 months or no initial CR was conducted and is reported herein. The VALOR intent-to-treat (ITT) population included 364 patients ≥60 years old with Ref/eRel AML; of these, 361 patients received treatment (Figure 1). Baseline characteristics in this subset were generally well-balanced between treatment arms (Table 1). The median age was 68 years (range 60–78) in both treatment arms. Due to stratified randomization, disease status was evenly distributed: 57.7% Ref and 42.3% eRel disease in both treatment arms. The majority of patients (>75%) in both treatment arms had received only one prior induction cycle; the median number of prior induction and consolidation cycles was two (range 1–9). Five patients (2.7%) in each treatment arm had received prior transplant.

Figure 1.CONSORT diagram of patients ≥60 years old with Ref/eRel AML. AML: acute myeloid leukemia. ITT: intent-to-treat; PI: primary investigator; pla/cyt: placebo plus cytarabine; Ref/eRel: refractory and early relapsed disease; vos/cyt: vosaroxin plus cytarabine.

Table 1.Baseline characteristics in patients ≥60 years of age with Ref/eRel disease (ITT population).

The proportion of patients ≥60 years old with Ref/eRel disease who received at least one cycle of study treatment was similar across treatment arms (vos/cyt: 99.5%; pla/cyt: 98.9%). Of these, 31 patients (17.0%) treated with vos/cyt and 39 (21.8%) treated with pla/cyt received a second induction cycle, and 48 (26.4%) and 16 (8.9%) patients, respectively, received at least one consolidation cycle. A median of one treatment cycle was received (range, 1–4 cycles) in both treatment arms. Fewer patients treated with vos/cyt discontinued due to treatment failure or relapse (52.7%) compared with pla/cyt (76.5%) (Figure 1).

The CR rate in patients ≥60 years with Ref/eRel disease was substantially higher with vos/cyt than pla/cyt. The CR rate was 25.8% (47/182; 95% CI: 19.6–32.8) in the vos/cyt arm and 10.4% (19/182; 95% CI: 6.4–15.8) in the pla/cyt arm (P=0.0001). The overall response rate (ORR) was also improved (vos/cyt: 34.1% [62/182] vs. pla/cyt: 12.6% [23/182]; P<0.0001).

The addition of vosaroxin to cytarabine substantially increased OS in patients ≥60 years old with Ref/eRel disease (Figure 2). Median OS was 6.5 months (95% CI: 4.4–7.8) with vos/cyt and 3.9 months (95% CI: 3.3–5.2) with pla/cyt (HR = 0.69 [95% CI: 0.55–0.86]; P=0.0009). When patients with subsequent transplant were censored from the OS analysis at the time of transplant, median OS was 6.2 months (95% CI: 4.4–7.4) and 3.9 months (95% CI: 3.3–5.0), respectively (HR = 0.71 [95% CI: 0.56–0.90]; P=0.0047). Vosaroxin plus cytarabine also improved leukemia-free survival (HR = 0.50 [95% CI: 0.25–0.99]; P=0.0424) and event-free survival (HR = 0.59 [95% CI: 0.47–0.74]; P<0.0001) compared with pla/cyt (Figure 2). At the time of database lock, 17.0% of patients in the vos/cyt arm and 7.1% in the pla/cyt arm remained alive and in continued follow-up.

Figure 2.Kaplan-Meier estimates of overall (A), leukemia-free (B), and event-free (C) survival in patients ≥60 years old with Ref/eRel AML. AML: acute myeloid leukemia; HR: hazard ratio; CI: confidence interval; pla/cyt: placebo plus cytarabine; Ref/eRel: refractory and early relapsed disease; vos/cyt, vosaroxin plus cytarabine.

Post-treatment transplantation rates were identical in both treatment arms (17.0% [31/182]; [95% CI: 11.9–23.3]). However, a greater proportion of transplanted patients achieved CR with study therapy prior to transplant in the vos/cyt arm (48.4% [15/31]) than in the pla/cyt arm (32.3% [10/31]) and the 100-day mortality rate after transplant was lower in patients treated with vos/cyt (19.4% [6/31]) than with pla/cyt (25.8% [8/31]). Among transplanted patients, median OS was 18.3 months (95% CI: 11.9-NE) with vos/cyt and 9.9 months (95% CI: 7.7–12.2) with pla/cyt (HR = 0.46 [95% CI: 0.25–0.86]; P=0.0125).

Importantly, the addition of vosaroxin did not increase 30-or 60-day mortality (30-day: 9.9% [18/182] vs. 10.6% [19/179]; 60-day: 20.9% [38/182] vs. 24.6% [44/179] for vos/cyt vs. pla/cyt, respectively). Most patients in both treatment arms experienced at least one adverse event (AE) of any grade (vos/cyt: 99.5% vs. pla/cyt: 100%) or grade ≥3 AE (93.4% vs. 86.6%, respectively); however, the incidence of discontinuations due to AEs was low (<3.0%) in both treatment arms. Serious AEs (SAEs) were more common with vos/cyt (53.8%) than with pla/cyt (32.4%). Serious AEs leading to death occurred in 15.9% of patients in the vos/cyt arm and 11.2% in the pla/cyt arm. The rates of treatment-related AEs, treatment-related grade ≥3 AEs, and treatment-related SAEs were 91.8%, 72.0%, and 31.3% in the vos/cyt arm compared with 86.0%, 60.9%, and 15.1% in the pla/cyt arm, respectively.

Myelosuppression, infections, and gastrointestinal (GI) toxicities were the most common AEs and SAEs in both treatment arms. Grade ≥3 febrile neutropenia was more common in the vos/cyt arm (40.7%) than the pla/cyt arm (29.1%); however, other grade ≥3 hematologic events occurred with similar frequency in both arms, including (for vos/cyt vs. pla/cyt, respectively) thrombocytopenia (23.1% vs. 26.8%), anemia (22.5% vs. 25.7%), and neutropenia (17.6% vs. 16.2%). Differences between treatment arms in grade ≥3 non-hematologic events included higher rates for vos/cyt vs. pla/cyt, respectively, of hypokalemia (15.4% vs. 7.3%), stomatitis (15.4% vs. 4.5%), and sepsis (12.1% vs. 5.6%). Vos/cyt therapy was not associated with a higher incidence of other end organ toxicities, such as hepatic, neurologic, renal, and cardiac toxicities.

Overall, this analysis demonstrated that vos/cyt produced clinically meaningful improvements in response and survival compared with pla/cyt in patients ≥60 years old with Ref/eRel AML, without increasing early mortality. The AE profile of vos/cyt in older patients was consistent with the AE profile of the overall VALOR population reported previously.3 Rates of SAEs were higher in the vos/cyt arm compared with the pla/cyt arm, though this would be expected with the addition of a second cytotoxic agent and has been observed in other trials of cytarabine combination regimens compared to cytarabine alone.54 The toxicities seen with vos/cyt therapy were similar in type and severity to those commonly seen with currently available therapies used to treat Ref/Rel AML patients, and physicians who treat leukemia are accustomed to managing these types of toxicity.

Vosaroxin’s activity in patients ≥60 years old with Ref/eRel AML, a generally treatment-resistant population, may be due in part to its ability to evade common drug resistance mechanisms. In older AML patients, there is a higher incidence of unfavorable cytogenetics, resulting in higher resistance to chemotherapy.86 P-glycoprotein (P-gp) expression levels are higher in older patients and those with relapsed disease.76 Similarly, older patients are more likely to have alterations in the TP53 gene, increased frequency of AML driver gene mutations, and increased probability of of RAS, Src, and tumor necrosis factor (TNF) pathway dysregulation.109 Vosaroxin is a broadly active cytotoxic agent that has demonstrated activity in a number of drug-resistant models, including those with breast cancer resistance protein (BCRP) or P-gp transporter-mediated efflux,1211 and has activity independent of TP53 status,1413 characteristics that may make it particularly useful in the older Ref/eRel AML population.15

As this is a post-hoc subset analysis, a confirmatory study is required. However, the results of this analysis suggest that vosaroxin plus cytarabine represents a potential new treatment option for poor prognosis AML patients ≥60 years old with Ref/eRel AML that merits confirmation with a randomized clinical trial.