Combination of cladribine plus topotecan for recurrent or refractory pediatric acute myeloid leukemia

Abstract

BACKGROUND:

The prognosis after recurrence of pediatric acute myeloid leukemia (AML) is poor, and effective salvage regimens are urgently needed.

METHODS:

In phase 1 and pilot studies, the authors evaluated the maximum tolerated dose (MTD) and dose-limiting toxicities (DLTs) of a 5-day course of cladribine followed by topotecan in pediatric patients with recurrent/refractory AML. The cladribine dose was escalated as follows: 9.1, 13.6, 16.3, and 19.5 mg/m² per day (8.9 mg/m² per day in the pilot study). Outcome was analyzed according to the absence (Stratum 1) versus presence (Stratum 2) of previous allogeneic hematopoietic stem cell transplantation. Twenty-six patients (20 in Stratum 1 and 6 in Stratum 2) were treated.

RESULTS:

The MTD was not reached in Stratum 1, but a DLT occurred at the lowest cladribine dosage (9.1 mg/m² per day) in Stratum 2. Febrile neutropenia was common in both strata. Nine (34.6%) of 26 patients experienced a complete response, and 7 (30.4%) achieved a partial response; 5 (19.2%) were long-term survivors at the time of last follow‒up. Clinical outcome was not associated with cladribine or topotecan systemic exposure.

CONCLUSIONS:

The combination was well tolerated in Sratum 1, and the response rate was encouraging. This regimen offers a postrecurrence treatment alternative for patients, especially those who have received anthracycline-containing chemotherapy. Cancer 2010. © 2010 American Cancer Society.

Substantial progress in the management of childhood acute myeloid leukemia (AML) has been achieved over the past 30 years.1-3 However, approximately 30% to 40% of patients with AML experience disease recurrence despite intensive chemotherapy and the use of allogeneic hematopoietic stem cell transplantation (HSCT). Patients with FLT3 internal tandem duplication, M6 or M7 French‒American‒British (FAB) subtype, myelodysplastic syndrome (MDS)-related AML, monosomy 7 karyotype, or persistent disease after 2 courses of conventional induction chemotherapy have a particularly poor outcome. Because most frontline AML protocols feature high cumulative doses of anthracyclines, treatment regimens for recurrent diseases should limit or avoid the use of this class of agents, particularly if salvage treatment includes HSCT.

Cladribine is a deoxyadenosine analog. Intracellular cladribine is rapidly phosphorylated to the triphosphate, which resists degradation by adenosine deaminase and therefore accumulates to cytotoxic levels.4 Cladribine triphosphate competes with deoxyadenosine triphosphate as a substrate of DNA polymerases, and its incorporation interferes with subsequent chain elongation. Cladribine has been extensively evaluated at our institution in pediatric patients with recurrent and refractory AML.5, 6 In a phase 2 study, a 5-day continuous infusion of single-agent cladribine at a dose of 8.9/mg/m² per day induced a high rate of response in patients with recurrent AML.6 Subsequently, cladribine has been incorporated into upfront therapy for newly diagnosed patients with AML as a single agent or as combination therapy.7, 8

Topotecan, a semisynthetic analog of camptothecin, is a specific inhibitor of topoisomerase I.9, 10 Topotecan stabilizes the topoisomerase-I/DNA complex, causing DNA single-strand breaks and cell death, particularly during S-phase. Daily 30-minute infusions of topotecan reportedly induced significant responses in a Children's Oncology Group phase 1 study of children with refractory leukemia.11

The functional DNA interference caused by topotecan and by cladribine may result in enhanced cytotoxicity when these agents are given in combination. Because both agents can be administered as short bolus infusions,8, 12 we studied various schedules of short exposure to these 2 drugs in vitro. On the basis of a preclinical trial demonstrating an inhibitory effect of cladribine followed by topotecan (given after a drug-free period) (unpublished data), we developed a phase 1 study to determine the maximum tolerated dose (MTD) and dose-limiting toxicities (DLTs) of short infusions of escalating doses of cladribine followed by topotecan targeted to achieve a predetermined systemic exposure in pediatric patients with recurrent or refractory AML.

MATERIALS AND METHODS

Patients

Eligible patients with recurrent or refractory AML who were aged <21 years were treated at St. Jude Children's Research Hospital and at the University of Michigan Comprehensive Cancer Center from 1998 to 2003. They had a life expectancy ≥6 weeks, an Eastern Cooperative Oncology Group performance status ≤3, and adequate renal (creatinine <2 × normal) and hepatic (bilirubin ≤3 mg/dL, alanine aminotransferase and aspartate aminotransferase ≤500 U/dL) function. Three patients previously treated in a pilot study at St. Jude Children's Research Hospital with the same combination were included. Patients were grouped for analysis as those who had not previously undergone allogeneic HSCT (Stratum 1) and those who had (Stratum 2). This study and the inclusion of the 3 additional patients were approved by institutional review board. Signed informed consent was obtained from patients, parents, or legal guardians, with assent from the patients, if appropriate.

Study Design

Both drugs were administered for 5 consecutive days per course. Cladribine was given intravenously over 3 hours at a dosage of 9.1, 13.6, 16.3, or 19.5 mg/m² per day; 3 patients treated in the pilot study received 8.9 mg/m² per day. Topotecan was administered intravenously over 30 minutes, starting 8 hours after initiation of cladribine. The topotecan dosage was 4.0 mg/m² on Day 1 and was then individualized on the basis of topotecan clearance to target a systemic exposure of 140 ± 20 ng/mL/h. In each stratum, the treatment and dose escalation were performed separately. All patients received filgrastim (granulocyte–colony-stimulating factor) at a dose of 5 μg/kg/d for a minimum of 10 days, beginning 24 hours after completion of the last dose of topotecan.

A modified phase 1 design proposed by Storer13 was adopted. At least 1 patient was treated at a starting dose level of cladribine with topotecan for 5 days. If no DLT was observed, the next patient received the next higher cladribine dosage. If a DLT occurred, the next patient was treated at the previous dose level, and the traditional phase 1 design was adopted thereafter. Patients treated with this combination at the University of Michigan Comprehensive Cancer Center were evaluated for toxicity, response, and pharmacokinetics, but did not participate in the dose escalation scheme.

Toxicity

Toxicity was graded according to the National Cancer Institute Common Toxicity Criteria (version 2.0). A DLT was defined as any nonhematologic toxicity grade ≥3, with the exception of nausea, vomiting, and grade 3 febrile neutropenia. Any prolonged (>35 days) hematologic toxicity that included neutropenia (absolute neutrophil count <300/μL) and thrombocytopenia (platelets <30,000/μL) in the absence of persistent or progressive disease was considered a DLT.

Response

A bone marrow aspirate and core needle biopsy were assessed before and on Day 15 after the initiation of the first course. A complete response (CR) was indicated by an aspirate containing <5% blast cells and a partial response (PR) by 15% to 50% blast cells. Bone marrow studies were repeated after 1 week if the aspirate contained 5% to 15% blast cells. Patients were removed from the study if they achieved less than a PR in the first course or experienced unacceptable toxicity. Patients who had a CR or PR were eligible to receive additional courses at the same dosage level. There was no intrapatient dose escalation or de-escalation.

Pharmacokinetically Guided Topotecan Dosing

Each patient's topotecan dosage was individualized to achieve a topotecan lactone systemic exposure of 140 ± 20 ng/mL/h.14 Briefly, plasma samples obtained before and 0.25, 0.5, 1, 3, and 6 hours after completion of the first topotecan infusion were immediately processed and analyzed. If the single-day topotecan lactone area under the concentration-time curve (AUC) was within the target range, no dose adjustment was required. Otherwise, the topotecan dosage was adjusted linearly on the basis of the patient's topotecan lactone clearance estimated from the pharmacokinetic model, and repeat pharmacokinetic studies were performed on Day 2. This procedure continued until the patient's topotecan AUC was within the target range. A 2-compartment pharmacokinetic model was fitted to the topotecan lactone plasma concentration as previously described.14

Cladribine Pharmacokinetic Studies

On Day 1 of the first course of treatment, blood samples were obtained before the cladribine infusion; 30, 60, and 90 minutes after the start of the infusion; at the end of the infusion; and 2, 4, 10, and 24 hours after completion of the infusion. The plasma cladribine concentration was determined using a sensitive and specific bigradient, high-performance liquid chromatography assay with ultraviolet detection.7 At 264 nanometers (nm), the range of the standard curve was 5 to 120 nM.

A 2-compartment model was fit to the cladribine concentration-versus-time data by nonlinear regression with a Bayesian algorithm implemented in ADAPT II and using published values (mean and variance) as the Bayesian priors.7 Model parameters estimated for each patient included the volume of the central compartment, elimination rate constant, and intercompartment rate constants kcp (rate constant from central to peripheral compartment) and kpc (rate constant from peripheral to central compartment). These parameters were used to simulate each patient's plasma concentration-time profile, from which the AUC from time 0 to hour 24 (AUC_0→24) was calculated.

Statistical Analysis

Overall survival was defined as the time between study enrollment and death; data for survivors were censored at the time of last follow-up. Survival was estimated by the Kaplan-Meier method, and survival estimates were compared according to stratum and cladribine dose level by the exact log-rank test. Fisher exact test was used to examine the relation between clinical response and cladribine dose level, the proportion of topotecan AUC values within the target range after the initial dose, and the proportion of topotecan AUC values within the target range after dose adjustment. Logistic regression was used to assess the relation between clinical response and cladribine dose level, cladribine AUC, and topotecan AUC. The Kruskal-Wallis test was used to examine the association between cladribine AUC and cladribine dose. All tests were 2-tailed. Results with a P value <.05 were considered statistically significant. No adjustment was made for multiple tests.

RESULTS

Patient Characteristics

The characteristics of the 26 patients are summarized in Table 1. Twenty-three patients (17 in Stratum 1 and 6 in Stratum 2) were enrolled on the phase 1 study. Three additional patients treated with 8.9 mg/m² per day of cladribine during a pilot study were assigned to Stratum 1. There were 4 cases of AML with prior MDS. Of the 20 patients in Stratum 1, 14 patients had recurrent leukemia, with a median initial remission duration of 11.4 months, and 6 had refractory disease. All 6 patients in Stratum 2 had recurrent disease, with a median duration of initial remission of 18.5 months.

Table 1. Characteristics of the Study Group (n = 26)

Characteristics	Total	Stratum
		1	2
WBC indicates white blood cell count; AML, acute myeloid leukemia; HSCT, hematopoietic stem cell transplantation; CR, complete response; FAB, French-American-British classification; AML/MDS, acute myeloid leukemia with prior myelodysplastic syndrome; CNS, central nervous system; 2-CDA, cladribine. * At the time of study enrollment. † Chromosomal abnormality was present in addition to trisomy 8.
Sex
Female	13	10	3
Male	13	10	3
Age, y
Median	10.0	10.0	6.5
Range	1.0-19.0	1.0-18.0	2.0-19.0
WBC, ×10^9/L*
Median	3.4	2.3	5.5
Range	0.2-116.2	0.4-116.2	0.2-16.5
Therapy for initial AML
Chemotherapy only	15	15	0
Autologous HSCT	5	5	0
Allogeneic HSCT	6	0	6
Duration of first CR, mo
Median	11.4 (n = 20)	11.4 (n = 14)	18.5 (n = 6)
Range	3.0-65.0	3.0-23.7	9.3-65.0
FAB subtype
M0	1	1	0
M1	3	2	1
M2	9	8	1
M4	4	3	1
M5	3	3	0
M7	2	1	1
AML/MDS	4	2	2
Karyotype
Trisomy 8	4	1	3
7q−	3	2	1†
11q23	3	2	1†
t(8;21)	3	3	0
inv(16)	1	1	0
Normal	6	5	1
Other	8	6	2
CNS at recurrence
Involvement	1	1	0
No involvement	18	14	4
Traumatic tap	3	2	1
Traumatic tap with blasts	2	1	1
Not done	2	2	0
2-CDA dose, mg/m^2
8.9	3	3	0
9.1	6	2	4
13.6	5	4	1
16.3	6	5	1
19.5	6	6	0

Toxicity

In Stratum 1, the regimen was generally well tolerated (Tables 2 and 3). Febrile neutropenia was common (10 patients), and 2 patients had documented bacteremia (α-streptococcus and Pseudomonas aeruginosa, respectively). After a patient treated at a dose of 13.6 mg/m² per day of cladribine had prolonged bone marrow suppression of undetermined etiology, the cladribine dosage was assigned according to the traditional phase 1 design. One patient receiving 19.5 mg/m² per day of cladribine developed grade 4 renal failure and grade 4 respiratory failure with bilateral pulmonary effusions. Only this patient experienced a DLT; the MTD was not reached in this stratum.

Table 2. Observed Grade 3 and 4 Toxicities* by 2-CDA Dose Level

	2-CDA Dose, mg/m^2/day
	Stratum 1 (n = 20)					Stratum 2 (n = 6)
2-CDA indicates cladribine. * Toxicities were graded according to the National Cancer Institute Common Toxicity Criteria (version 2.0). † Numbers in parentheses indicate number of patients who experienced dose-limiting toxicity.
	8.9	9.1	13.6	16.3	19.5	9.1	13.6	16.3
Toxicity
Nausea/vomiting						1
Pulmonary					1 (1)†
Nephrotoxicity					1 (1)†
Liver enzyme abnormalities						1 (1)†
Bone marrow aplasia							1 (1)†
Fever and neutropenia		1	2	3	4	3 (2)†	1	1
Evaluable patients	3	2	4	5	6	4	1	1

Table 3. Individual Patient, Therapy, and Outcome Data

Patient	Sex	Age at Recurrence, Years	FAB	Duration of First CR, Months	Prior Treatment	2-CDA Dose, mg/m^2/day	Day 15 Response	Dose- Limiting Toxicity	Subsequent Therapy	Outcome
FAB indicates French-American-British classification; CR, complete response; 2-CDA, cladribine; Auto, autologous hematopoietic stem cell transplantation (HSCT); Allo, allogeneic HSCT; CCR, continuous CR; NR, no response; Chemo, chemotherapy; DOD, died of disease; PR, partial response; TRM, treatment-related mortality; AML/MDS, acute myeloid leukemia with prior myelodysplastic syndrome; GO, gemtuzumab ozogamicin; VOD, veno-occlusive disease; DLI, donor lymphocyte infusion; IL-2, interleukin-2. * Patients did not participate in the dose escalation scheme. † Patient was inadvertently given a higher dose.
Stratum 1
1	Male	3	M5	23.7	Auto	8.9	CR	None	Allo	CCR
2	Female	15	M2	7.9	Auto	8.9	CR	None	Allo	CCR
3	Female	7	M2	8.8	Auto	8.9	NR	None	Chemo	DOD
4	Male	1	M7	6.2	Chemo	9.1	CR	None	Allo	DOD
5	Female	14	M2	21.7	Auto	9.1	NR	None	Allo	DOD
6	Female	10	M2	Refractory	Chemo	13.6	PR	Undetermined	Chemo	DOD
7	Male	11	M5	13.9	Auto	13.6	NR	None	None	DOD
8	Male	10	M4	Refractory	Chemo	13.6	NR	None	None	DOD
9	Male	10	M5	5.6	Chemo	13.6	PR	None	Allo	CCR
10	Female	13	M2	11.9	Chemo	16.3	NR	None	None	DOD
11	Female	13	M4	9.3	Chemo	16.3	CR	None	Allo	Died, TRM
12	Male	10	AML/MDS	3	Chemo	16.3	PR	None	Allo	DOD
13*	Male	16	M2	Refractory	Chemo	16.3	CR	None	Allo	CCR
14*	Female	4	M0	12.2	Chemo	16.3	PR	None	Allo	CCR
15	Male	10	M1	16.2	Chemo	19.5	CR	None	Allo	Died, TRM
16	Female	9	M1	15.5	Chemo	19.5	NR	None	GO/Allo	Died, TRM
17	Female	10	M5	10.9	Chemo	19.5	NR	None	GO	DOD
18	Female	15	AML/MDS	Refractory	Chemo	19.5	PR	Pulmonary/renal	None	DOD
19	Male	18	M4	Refractory	Chemo	19.5	NR	None	None	DOD
20	Male	18	M2	Refractory	Chemo	19.5	NR	None	None	DOD
Stratum 2
1	Female	2	M7	9.3	Allo	9.1	NR	None	None	DOD
2	Female	8	AML/MDS	29.2	Allo	9.1	CR	Infection	Allo	DOD
3	Male	4	AML/MDS	9.4	Allo	9.1	PR	Infection/VOD	None	DOD
4*	Male	5	M1	9.6	Allo	9.1	CR	None	DLI	DOD
5	Female	19	M2	65	Allo	13.6	PR	Bone marrow aplasia	None	DOD
6*	Male	18	M4	27.4	Allo	16.3†	CR	None	IL-2	Died, TRM

In Stratum 2, 5 patients developed febrile neutropenia. Two patients receiving 9.1 mg/m² per day of cladribine had grade 4 febrile neutropenia; 1 of these developed sepsis (Enterobacter sp.) with an episode of hypotension, and 1 had documented disseminated aspergillosis complicated by severe hepatic veno-occlusive disease. One patient treated with a dose of 13.6 mg/m² per day of cladribine had prolonged neutropenia and thrombocytopenia (>35 days) associated with fever. A DLT was observed at the initial cladribine dose of 9.1 mg/m² per day in Stratum 2, and further patient enrollment was discontinued.

Response

Nine (34.6%) of the 26 patients experienced a CR and 7 (26.9%) achieved a PR; the overall response rate was 61.5% (Table 3). In Stratum 1, 11 (55%) of the 20 patients achieved responses, and 6 (30%) had CRs. Three patients (Patients 1, 11, and 14 in Table 3) had 1 additional course of cladribine/topotecan before subsequent therapy. Eleven patients proceeded to allogeneic HSCT, and 4 were changed to therapy with gemtuzumab ozogamicin (n = 2) or methotrexate and L-asparaginase (n = 2). Five (83%) of the 6 patients in Stratum 2 responded. The 3 patients who had achieved CRs subsequently underwent treatment with allogeneic HSCT, donor lymphocyte infusion, or interleukin-2 (IL-2 ).

Of the 16 responders overall, 9 (56.3%) remained in disease remission and received additional therapy (allogeneic HSCT in 8 patients and IL-2 in 1 patient), but 4 of these patients died of treatment-related toxicity during remission. Five (19.2%) of the 26 patients survived in continuous complete remission at the last follow-up date (Fig. 1). We were unable to demonstrate a statistically significant difference in survival according to stratum (P = .38), cytogenetics (t[8;21] or inv[16] versus others [P = .63]), or duration of first remission (>1 year vs ≤1 year [P = .84]).

Figure 1. The overall survival of patients treated with cladribine and topotecan is shown. Tick marks denote those patients still alive at the time of last follow-up.

Pharmacokinetically Guided Topotecan Dosing

Table 4 summarizes the topotecan pharmacokinetic parameters derived from 54 pharmacokinetic studies in the 26 patients. Of the 26 studies performed after the initial topotecan dose of 4.0 mg/m², only 6 (23%) demonstrated a topotecan AUC within the target range. Of the remaining 28 pharmacokinetic studies (those performed after targeted dose adjustment), 20 (72%) were within the target range. The median dosage achieving the target AUC in these patients was 4.0 mg/m² (range, 1.7-6.0 mg/m²). In univariate and multivariate logistic regression models, clinical outcome was not found to be significantly associated with the proportion of topotecan AUC values within the target range after the initial dose or after dose adjustment.

Table 4. Summary of Topotecan Pharmacokinetic Parameters (n = 26)

	Mean	Median	Minimum	Maximum
Vc indicates volume of the central compartment; Kel, elimination rate constant; T1/2 beta, beta half-life; CL, systemic clearance; Vdss, steady-state volume of distribution.
Vc, L/m^2	39.4	39.4	9.6	316.7
Kel, h^−1	0.84	0.73	0.43	2.04
T1/2 beta, h	2.47	2.23	1.55	4.94
CL, L/h/m^2	30.5	31.45	12.7	54.3
Vdss, L/m^2	68.3	66.9	28.6	124.3

Cladribine Pharmacokinetics

Cladribine pharmacokinetics were studied in 16 of the 26 patients. The necessary blood samples for 3 patients treated before study initiation and for 7 other patients were not available. The median cladribine systemic clearance rate was 18.1 L/h/m² (range, 10.6-27.5 L/h/m²). The AUC varied widely at different cladribine dose levels, but the median systemic exposure was found to be significantly associated with the dose level (P = .031) (Fig. 2A). Response occurred at all dose levels and was not significantly associated with the cladribine AUC (P = .19) (Fig. 2B).

Figure 2. (A) The correlation between the cladribine (2-CDA) area under the plasma concentration-time curve (AUC) and the 2-CDA dose is shown. (B) The correlation between the 2-CDA AUC and clinical response is shown. Patients received 2-CDA daily for 5 days in combination with topotecan doses adjusted to achieve a targeted AUC of 140 ± 20 ng/mL/h. CR indicates complete response; PR, partial response; NR, no response.

DISCUSSION

Because of the risk of organ toxicity posed by frontline chemotherapy regimens for childhood AML, we sought to develop an anthracycline-free regimen for the treatment of recurrent or refractory AML. Our preclinical studies demonstrated that when cladribine is combined with topotecan, its activity is schedule-dependent, especially at clinically relevant concentrations (ie, 75 nM topotecan15 and 290 nM cladribine).16 In these experiments, inhibition of leukemia cell growth was optimal when cladribine was added to the culture first and topotecan was then added after a drug-free interval. It is possible that cladribine exacerbates the disruption of normal DNA replication caused by topotecan's inhibition of topoisomerase, with a consequent increase in cytotoxicity. On the basis of these in vitro results, we designed the combination regimen to include a 5-hour delay between the end of the cladribine infusion and the start of the topotecan infusion. This interval allowed the plasma concentration of cladribine to fall to approximately 40 nM, which was the optimal cladribine concentration in the preclinical studies.

The combination of cladribine and topotecan in the current study appeared to be well tolerated by patients who had not undergone allogeneic HSCT; the MTD was not reached in this group. Because we found no association between efficacy and the cladribine dose or the cladribine/topotecan pharmacokinetics, we suggest that cladribine at a dose of 9.1 mg/m² per day and topotecan at a dose of 4.0 mg/m² per day are a reasonable dose regimen in this patient population.

Patients who had received allogeneic HSCT before disease recurrence (Stratum 2) were found to have a high incidence of regimen-related toxicity in the current study, with a DLT observed at the initial dose of cladribine (9.1 mg/m²). However, 5 of 6 patients in this stratum achieved a response; therefore, lower doses of cladribine and/or topotecan may be considered for this patient population.

We would suggest using prophylactic antibacterial and antifungal agents, because fever with neutropenia was the most common toxicity of this regimen, and we recently reported that prophylactic treatment during intensive treatment for pediatric AML reduced morbidity because of septicemia and dramatically decreased the incidence of septicemia and of inpatient days.17

After recurrence of childhood AML, chemotherapy alone is unlikely to be curative because of drug resistance.18 Every effort should be made to induce remission before HSCT, but it is crucial to reduce the toxicity of reinduction therapy to a level that allows HSCT to proceed. Various regimens, including fludarabine plus cytarabine and cytarabine plus mitoxantrone/idarubicin, appear to give similar results.19, 20 The addition of liposomal daunorubicin to fludarabine plus cytarabine is currently being tested in an effort to improve CR rates while minimizing cardiotoxicity. Our regimen induced an overall response rate of 61.5% and a CR rate of 34.6%. Nine (34.6%) patients remained in remission, although 4 died of toxicity associated with subsequent HSCT or IL-2 treatment. These patients are at high risk of transplant-related mortality and morbidity after second remission.18, 21 Therefore, innovative HSCT approaches are needed. Donor-recipient mismatch of killer immunoglobulin-like receptors and ligand in the presence of T-cell depletion can facilitate natural killer cell-mediated cytotoxicity to leukemia cells.22 Reduced-intensity regimens have been used in elderly adults or those with comorbidities and could be considered for this patient population.23

In summary, on the basis of our extensive preclinical data, we designed and conducted what to the best of our knowledge is the first clinical trial of cladribine and topotecan in children with recurrent/refractory AML. The combination of short intravenous cladribine and topotecan infusions was found to be safe and well tolerated among patients who had not undergone allogeneic HSCT. The response rate in this very drug-resistant group is encouraging. This regimen offers a treatment alternative for patients previously treated with anthracycline-containing regimens.

Acknowledgements

We thank Sharon Naron for editorial advice.

CONFLICT OF INTEREST DISCLOSURES

Supported in part by Cancer Center Core Grant CA 21765 from the National Cancer Institute and by the American Lebanese Syrian Associated Charities.

Dr. Ching–Hon Pui is an American Cancer Society Professor.