Combination chemotherapy with clofarabine, cyclophosphamide, and etoposide in children with refractory or relapsed haematological malignancies

Clofarabine-containing combination regimens are under evaluation for patients with relapsed or refractory leukaemia (Faderl et al, 2005; Hijiya et al, 2009; Locatelli et al, 2009; O’Connor et al, 2011). We report our single-institution experience with clofarabine, cyclophosphamide, and etoposide (CCE) combination therapy in children with relapsed or refractory haematological malignancies. Informed consent for treatment was obtained, and the study was approved by our institutional review board.

Fifteen patients (11 acute lymphoblastic leukaemia [ALL], 3 acute myeloid leukaemia [AML], and 1 hepatosplenic gamma/delta T-cell lymphoma and bone marrow involvement) were treated between October 2006 and May 2009 (Table I). The median age of the patients was 11·4 years (including three infants). The median number of regimens administered before CCE was two; four patients had received allogeneic haematopoietic stem cell transplantation (HSCT). Thirteen patients had morphologically detectable disease and two had persistent minimal residual disease (MRD). Drugs were administered intravenously once daily for five consecutive days: clofarabine at 30–40 mg/m², cyclophosphamide at 300–440 mg/m², and etoposide at 100 mg/m². All patients received all planned doses of CCE. Patients were eligible for HSCT after CCE if a suitable donor was available. Otherwise, subsequent courses of CCE or alternative salvage therapy were given. All patients received prophylactic antibiotics (either vancomycin plus fluoroquinolone or single-agent cefepime) and an antifungal agent (voriconazole or micafungin) during periods of neutropenia. Voriconazole was held from 48 h before to 24 h after the CCE course.

Table I.   Clinical course of 15 patients.

Pt	Age at CCE (years)	Sex	Disease	Cytogenetic abnormality	No of prior regimens	Pre-CCE transplant	Response to most recent therapy	Most recent therapy	Doses of CCE	No. of courses	Response to CCE	Post-CCE transplant	Duration of remission (weeks)	Outcome	Cause of death
CCE, clofarabine, cyclophosphamide, etoposide; M, male; F, female; pre B, B-precursor; ALL, acute lymphoblastic leukaemia; AML, acute myeloid leukaemia; MUD, matched unrelated donor; Haplo, haploidentical; CR, complete remission; CRp, CR without platelet recovery; PR, partial remission; NA, not applicable; CCR, continuous complete remission; MRD, minimal residual disease; Dex, dexamethasone; VP, etoposide; Ara-C, cytarabine; ASP, asparaginase; Mito, mitoxantrone; Flu, fludarabine; Ida, idarubicin; VCR, vincristine; Dox, doxorubicin; MTX, methotrexate; MP, mercaptopurine; DNR, daunorubicin; GO, gemtuzumab ozogamicin; PJP, Pneumocystis jiroveci pneumonia; MOF, multi-organ failure; GVHD, graft-versus-host disease. *Cytogenetic abnormalities other than those described in other patients. †CRMRD indicates MRD below detection limit (<0·01%). ‡PRMRD indicates more than 10-fold decrease in MRD but still above detection limit.
1	5·1	M	Pre-B ALL	Others*	2	None	Persistent MRD	Dex/VP/Ara-C	40/400/100	2	CRMRD†	MUD	27·7	Death	Disease
2	7·8	M	Pre-B ALL	Others*	2	None	CR	Dex/VP/Ara-C/ASP/Mito	40/440/100	1	CR	Cord	61	Death	PJP
3	13·5	M	Pre-B ALL	Others*	3	None	Refractory	Flu/Ara-C/Ida	30/440/100	2	Refractory	MUD	NA	Death	Disease
4	15·5	M	Pre-B ALL	Hypodiploid	2	None	Refractory	Dex/VP/Ara-C/ASP	40/400/100	1	Refractory	MUD	NA	Death	Acute GVHD, MOF
5	16·3	F	Pre-B ALL	t(9;22)	1	None	CR	Dex/VCR/Dox/ASP/Dasatinib	40/400/100	1	CRp	MUD	91+	CCR	NA
6	17·2	F	Pre B ALL	Hypodiploid	4	MUD	Refractory	Anti-CD22 immunotoxin	30/300/100	1	PR	None	0·6	Death	Mucor infection, MOF
7	0·5	M	Infant ALL	t(9;11)	2	Haplo	Refractory	MTX/MP/Ara-C	40/440/100	2	Refractory	Haplo	NA	Death	Massive lipid pneumonia
8	0·7	M	Infant ALL	t(11;11)	3	None	Refractory	MTX/Ara-C/MP/ASP	40/400/100	1	PR	Haplo	15	Death	Disease
9	1·0	M	Infant ALL	t(4;11)	2	None	Refractory	Dex/VCR/DNR/ASP	40/440/100	2	Refractory	None	NA	Death	Disease
10	5·0	F	T-ALL	Others*	2	None	CR	Dex/VCR/Mito/ASP	30/440/100	3	PR	None	11·4	Death	Disease
11	15·0	M	T-ALL	Others*	2	None	Persistent MRD	Dex/VP/Ara-C	40/400/100	1	PRMRD‡	MUD	16·3	Death	Acute GVHD, MOF
12	1·1	M	AML M7	Others*	1	None	Refractory	VP/Ara-C/GO	40/400/100	1	Refractory	None	NA	Death	Disease
13	11·4	F	AML M1	t(11;19)	2	MUD	CR	Ara-C/ASP and transplant	40/400/100	1	CR	MUD	101·1+	CCR	NA
14	18·4	M	AML M4	Monosomy 5	4	Haplo	CR	Flu/Ara-C/Ida and transplant	40/440/100	1	Refractory	None	NA	Death	Disease
15	15·8	M	Hepatosplenic gamma/delta T-cell lymphoma	Others*	2	None	Refractory	Pentostatin/Alemtuzumab	40/440/100	1	Refractory	None	NA	Death	Disease

Table I summarizes the clinical course of patients. Of the 13 patients with morphologically detectable disease, two achieved complete remission (CR) and one complete remission without platelet recovery (CRp). Three patients achieved partial remission, and seven had refractory disease. Of the two patients with persistent MRD, MRD became undetectable in one patient (MRD level 0·086% to <0·01%) and the disease decreased by more than 10-fold in the other (0·547–0·032%). Following CCE, 9 patients received HSCT at a median time interval of 85 days (range, 28–288 days). Of these, two who underwent HSCT after achieving CR or CRp with CCE are alive in CR (follow-up; 91 and 101 weeks, respectively), three died of disease relapse after HSCT, and four died from post-HSCT complications while in CR. Five of the six patients who did not receive HSCT died of disease, and one died of treatment-related complications.

Adverse events during cycle 1 are listed in Table II. Transient elevation of liver enzymes, febrile neutropenia, anorexia, and nausea were common. The only death that could be attributable to therapy was of a 17-year-old patient with B-precursor ALL and a germline TP53 mutation who had relapsed after HSCT 7·8 months earlier and whose disease was refractory to anti-CD22 immunotoxin. She experienced grade 4 hyperbilirubinaemia on day 15 of treatment without clinical features of veno-occlusive disease and died of a mucormycosis and multi-organ failure on day 30 (Tables I and II). Two additional patients had fungal infections: candidaemia (Candida albicans) and a skin mould infection (Macrophomina phaseolina).

Table II.   Toxicity of clofarabine, cyclophosphamide, and etoposide combination therapy in 15 patients with refractory or relapsed haematological malignancies*.

Adverse event	Total events N (%)	Grades
		1 N	2 N	3 N	4 N	5 N
ALT, alanine transaminase; AST, aspartate transaminase. *The National Cancer Institute’s Common Terminology Criteria for Adverse Events (version 3·0) (http://ctep.cancer.gov/protocolDevelopment/electronic_applications/docs/ctcaev3.pdf) was used to determine grades.
Gastrointestinal
Nausea	7 (46·7)	1	4	2
Anorexia	8 (53·3)	1	5	2
Diarrhoea	3 (20·0)			3
Mucositis	1 (6·7)			1
Metabolic/laboratory
AST elevation	9 (60·0)	3	4	2
ALT elevation	10 (66·7)	5	4	1
Hyperbilirubinemia	4 (27·0)	1	2		1
Infection
Febrile neutropenia	8 (53·3)			7		1
Pulmonary
Pneumonia	3 (20·0)			1	1	1
Dyspnea	2 (13·3)				1	1
Pleural effusion	1 (6·7)		1
Cardiac
Hypertension	1 (6·7)			1
Hypotension	1 (6·7)		1
Renal/genitourinary
Acute renal failure	2 (13·3)			2
Haemorrhagic cystitis	1 (6·7)		1
Neurology
Seizure	1 (6·7)		1
Encephalopathy	1 (6·7)			1
Dermatology/skin
Rash	1 (6·7)		1

Pretreatment with clofarabine may suppress the repair of DNA damage caused by cyclophosphamide and etoposide, and these three drugs may act synergistically (Yamauchi et al, 2001). CCE at 40, 440, and 100 mg/m², respectively, once daily for 5 days, was recommended in a Phase I study for children with refractory or relapsed leukaemia (Hijiya et al, 2009). All 25 patients received granulocyte-colony stimulating factor (G-CSF). Use of prophylactic antibiotics was left to the physician’s discretion. Four patients had dose reductions or interruptions. Febrile neutropenia, increased aspartate transaminase and alanine transaminase, and hyperglycaemia were common, and one patient died of Micrococcus meningitis. In the Phase II study, four of the first eight patients experienced severe hepatotoxicity (three veno-occlusive disease and one hyperbilirubinaemia), which was considered to be associated with prior HSCT or concurrent infections (Hijiya et al, 2009). Locatelli et al (2009) used CCE at 40, 400, and 150 mg/m², respectively. All 25 patients were hospitalized for the duration of treatment, and 23 remained in-patients until haematological recovery. All patients received prophylaxis with acyclovir, ciprofloxacin, or amoxicillin conjugated with clavulanic acid, and voriconazole or fluconazole. The most common side effects were reversible hepatotoxicity and febrile neutropenia. In that report, there was no discontinuation of treatment due to adverse events, death from treatment-related complications, or cases of veno-occlusive disease.

The toxicity profile observed in our study is similar to that reported by Locatelli et al (2009). In both studies, the majority of patients did not receive G-CSF. However, stringent supportive care, including prophylactic antibiotics and antifungals, is advised. We have previously reported that anti-bacterial and -fungal prophylaxis during intensive treatment of children with AML reduces episodes of infection and hospitalization (Kurt et al, 2008), which was successfully applied in the current study. Although there were three cases of fungal infections, there was no case of bacteraemia. After administration of CCE, our patients were hospitalized only for medical reasons.

Recently, O’Connor et al (2011) reported three CRs in five infants with MLL rearrangement treated with single-agent clofarabine (three patients) or CCE (two patients). In their study, CCE was dosed based on infant weight. In our current frontline ALL protocol, Total XVI, infants with MLL rearrangement receive CCE at doses based on body surface area, as demonstrated feasible in this study. We reduce CCE doses by 50% in only infants <1 month of age and those who are <3 months of age and born significantly prematurely. In addition, patients with persistent or pre-HSCT MRD have worse prognosis (Coustan-Smith et al, 2000; Bader et al, 2009) and may benefit from non-cross-resistant chemotherapeutic agents. We are also studying the efficacy of CCE to reduce MRD prior to HSCT for high-risk ALL patients in first remission. For patients with relapsed or refractory leukaemia, we are testing the safety of a conditioning regimen with CCE followed by Natural Killer-cell infusion from a haploidentical donor.

In summary, CCE is well tolerated and effective in children with relapsed or refractory haematological malignancies. Appropriate supportive care should be implemented.

Acknowledgements

The authors would like to thank Cherise Guess, PhD, for editorial assistance in the development of this manuscript.

Funding

The study is supported in part by Cancer Center Support Grant P30 CA021765, National Institutes of Health Grant R01 115422, and the American Lebanese Syrian Associated Charities (ALSAC). Ching-Hon Pui is an American Cancer Society Professor.

Conflict of interest

The authors declare no conflict of interest.