FLUDARABINE, CYTARABINE, GRANULOCYTE COLONY-STIMULATING FACTOR, AND IDARUBICIN (FLAG-IDA) FOR THE TREATMENT OF CHILDREN WITH POOR-PROGNOSIS ACUTE LEUKEMIA: The Hacettepe Experience
The prognosis of relapsed and/or primary resistant acute leukemia (acute lymphoblastic leukemia [ALL] and acute myeloblastic leukemia [AML]) and secondary AML is quite poor, and an effective reinduction reg- imen for these types of leukemias is quite rare.
Allogeneic and autologous hematopoietic stem cell transplantation (HSCT) offers a chance for long- term survival in these childhood cases with poor-prognosis acute leukemia. The fludarabine, cytarabine, granulocyte colony-stimulating factor (G-CSF), and idarubicin (FLAG-IDA) regimen has been proven to be a potentially useful chemotherapy regimen for relapsed or poor-prognosis childhood leukemia [1–3].
Fludarabine (9-β-D-arabinofuranosyl-2 fluoroadenine) is a synthetic purine analogue. Its antitumor activity is conducted through inhibition of DNA and RNA synthesis, DNA and RNA polymerases, DNA primase, DNA ligase, and ribonucleated reductase, and by potentiation of deoxycytidine kinase activity. It also induces apoptosis and potentiates the activity of cy- tosine arabinoside (Ara-C).
High-dose Ara-C is one of the most effective drugs in the treatment of acute leukemia. The efficacy of Ara-C has been related to the higher intracellular concentration of the active metabolite Ara-C 5r-triphosphate (Ara-CTP).
An increase in intracellular Ara-CTP can be achieved upon pretreatment with fludarabine. Administration of fludara- bine prior to Ara-C may enhance the cytotoxicity and clinical efficacy of Ara-C [4]. The addition of G-CSF to the FLAG regimen has shortened the dura- tion of neutropenia and reduced infection rates following the chemotherapy regimen.
Furthermore, G-CSF increases the efficacy of the chemotherapeu- tic agents by recruitment of leukemic cells in to the S-phase. IDA, an an- thracycline, has also been combined with the FLAG regimen to increase antileukemic activity [1–3].
The FLAG regimen has been extensively used in adult patients with poor- prognosis acute and chronic leukemia since 1996 [5, 6]. However, a series of children with poor-prognosis acute leukemia treated with the FLAG-IDA regimen has rarely been reported in the literature [1–3]. FLAG-IDA has been used in relapsed and/or primary resistant acute leukemia (ALL and AML) and secondary AML cases in Hacettepe University Faculty of Medicine, Pedi- atric Hematology Unit, since 2002.
The aim of the study was to evaluate the complete remission (CR) rate, toxicity, and overall survival of children with poor-prognosis acute leukemia who received the FLAG-IDA regimen. Fur- thermore, we aimed to investigate the children who achieved CR following FLAG-IDA treatment regarding their eligibility for allogeneic HSCT.
MATERIAL AND METHODS
Between January 2002 and April 2007, a total of 25 children with poor-prognosis acute leukemia were treated with the FLAG-IDA regimen at our center. Among them, 11 (44%) were girls, and 14 (56%) were boys. The median age of the patients was 12 years, with an age range spanning from 2 to 17 years.
Prior to administering the regimen, informed written consent was obtained from all children and/or their parents. Within this cohort, 9 children were diagnosed with acute lymphoblastic leukemia (ALL), including 2 cases of resistant ALL and 7 cases of second- or third-relapsed ALL. The remaining 16 children had acute myeloid leukemia (AML), comprising 3 cases of resistant AML, 8 cases of first-relapsed AML, and 5 cases of secondary AML.
The patient group was heterogeneous regarding the extent of prior chemotherapy. All patients had undergone intensive pretreatment with a variety of chemotherapeutic agents.
All 25 children had bone marrow infiltration and 7 had additional ex- tramedullary disease at the time of FLAG-IDA administration. Among the 7 children with extramedullary disease, 2 had central nervous system (CNS) relapse, 2 had bone involvement, 1 had testicle relapse, 1 had orbital involve- ment, and 1 had orbital and CNS involvement.
Cardiac evaluation was performed by echocardiography and left ventri- cle ejection fraction (EF) >60% was the eligibility criterion for all of the children before administration of the regimen. Renal and hepatic func- tions were also evaluated before the FLAG-IDA regimen. According to the National Cancer Institude (Common Toxicity Criteria), the patients with transaminases (serum oxaloacetic transaminase [SGOT], serum glutamic pyruvic transaminase [SGPT]) >2.5× normal and creatinine >1.5× normal were not received FLAG-IDA regimen [7].
Routine Procedures and Prophylactic Agents
The implantation of a central venous catheter or a previously implanted port was necessary before administration of the regimen. Hydration, alka- linization, and allopurinol were advised for patients at risk of tumor lysis syndrome. Corticosteroid eye drops starting prior to the first Ara-C infusion were given every 6 hours and discontinued 12 hours after the last Ara-C infusion.
An antiemetic drug (ondansetron or granisetron) was used throughout the chemotherapy regimen. Trimethoprim-sulfamethoxazole (TMP-SMX) at a dosage of 5 mg/kg body weight daily, 3 times a week, and fluconazole at a dosage of 3 mg/kg body weight daily were administered as antibacterial and antifungal prophylaxis, respectively.
Treatment Regimen
The FLAG-IDA regimen includes 30 mg/m2/day of fludarabine by a 30-minute intravenous infusion daily for 4 consecutive days (days 1–4); 2000 mg/m2/day of Ara-C by a 3-hour intravenous infusion daily for 4 consec- utive days (days 1–4), starting 4 hours after the beginning of fludarabine; and 12 mg/m2/day IDA by a 1-hour intravenous infusion daily for 3 consecutive days (days 2–4), starting 1 hour prior to Ara-C infusion. G-CSF was admin- istered at a dosage of 400 µg/m2/day from day 0 to the day of absolute neutrophil count of >1.0 × 109/L.
Of the total 25 children with poor-prognosis acute leukemia, 7 (28.0%) children received 1 cycle, 17 (68.0%) received 2 cycles, and 1 (4%) received 3 cycles of FLAG or FLAG-IDA regimen. Among the total of 44 cycles of the regimen, 35 (79.5%) cycles were administered as FLAG-IDA and 9 (20.5%) cycles were administered as FLAG regimen without IDA. When cardiotoxicity was observed on echocardiography after the first FLAG-IDA regimen, only FLAG regimen was allowed to avoid further cardiotoxicity.
Overall Survival
Allogeneic HSCT was performed in 10 (40%) patients who had HLA 6/6 identical sibling donors. However, 5 patients who underwent allogeneic HSCT were lost due to infection (n = 1) or relapse of the primary disease (n = 4). One patient (Case 3) had relapse of the primary disease after al- logeneic HSCT and she received FLAG regimen after the transplantation.
The other 9 children received FLAG-IDA or FLAG regimen to achieve re- mission state before transplantation. The overall survival of patients who underwent allogeneic HSCT (mean: 40.6 ± 4.7, median: 40, range: 34–46 months) was longer than that of patients (mean: 5.5 ± 4.3, median: 4, range: 1–15 months) who did not. (All of the survivors were AML and underwent HSCT.)
Thus, after FLAG-IDA or FLAG regimen, 5 (20.0%) children were alive and 20 (80.0%) children were lost due to infection or relapse of the primary diseases. Only 2 children (Cases 6 and 11) were lost with infection before marrow recovery. Of those 5 children who are alive, all underwent allogeneic HSCT and are still in remission. The overall survival of the 25 patients was 28.6 ± 17.7 months (median: 36, range: 2–48 months).
DISCUSSION
Children with refractory and relapsed leukemia, as well as children with a secondary AML, have a poor prognosis, and alternative induction therapies to achieve repeat remission are very limited [8, 9]. High-dose Ara-C is one of the most effective drugs in reinduction therapy of poor-prognosis acute leukemia, and fludarabine applied prior to Ara-C increases the Ara-CTP accumulation in the blast cells. Furthermore, high-dose Ara-C acts syner- gistically with fludarabine, anthracyclines, and G-CSF.
On the basis of this information, fludarabine and high-dose Ara-C with or without G-CSF were used in poor-prognosis adult AML cases and produced CR rates of 50% to 60% [10–12]. After the initial usage of fludarabine and high-dose Ara-C, IDA was added to this combination chemotherapy. That application increased the CR rate to 67% to 80% in pediatric cases with refractory or relapsed AML [12, 13].
The number of studies in which the FLAG-IDA regimen has been used in children with poor-prognosis acute leukemia, especially in ALL, is very limited [1–3, 13–15]. In this study, we aimed to evaluate the CR rate, toxicity, and overall survival of children with poor-prognosis acute leukemia who received the FLAG-IDA regimen in our center.
In a study conducted by Fleischhack et al [2], the FLAG-IDA regimen was administered to children with refractory, relapsed, or secondary AML (age range: 1.2–17.5 years), and 17/23 (73.9%) children achieved a CR with a median duration of 13.5 months; 9 (39.1%) children remained in continuous CR with a median duration of 17.5 months. Of these 23 children, 11 (47.8%) underwent bone marrow transplantation (BMT) or peripheral blood SCT.
In their series, most of the children (17/23, 73.9%) were on the first relapse; there were no cases of post-BMT relapse, and only 1 child was secondary AML. However, in our series, there were only 8 (32.0%) children who were on the first relapse. Additionally, there were 5 secondary AML cases and 1 post-BMT relapse case, which have extremely poor prognosis.
McCarthy et al [15] preferred FLAG without anthracyclines in heavily pretreated children, and 13/16 (70%) children achieved CR in that study. Thirteen (70%) patients underwent BMT, and 7 (43.8%) patients were alive 12 months after the FLAG regimen. The rate of patients who underwent BMT was higher than that of our series, and this might explain the increased number of patients who were alive.
Yalman et al [3] from Turkey used the FLAG-IDA regimen in 17 children (age range: 3–18 years) with relapsed acute leukemia or chronic leukemia with acute blastic crisis. They reported that only 3/17 (17.6%) children achieved CR and 1/17 (5.9%) was disease-free in the 30th month of the sec- ond remission. The only surviving child also had donor lymphocyte infusion (DLI) on the 7th day of the second FLAG-IDA course. Seven of 25 courses resulted in fatal infections. In that study, induction failure was high and the outcome was disappointing.
They concluded that the FLAG-IDA regi- men was not a cost-effective therapy in a developing country. In our study, after all cycles of FLAG-IDA or FLAG regimen, 60.0% of children showed CR. Identical dosing and schedule of Flu, Ara-C, and G-CSF was used in all patients in our series. Only 2 children (Cases 6 and 11) were lost with infec- tion before marrow recovery. Although our patients carried very poor risk criteria, the overall response rate was acceptable.
Yalman et al’s paper was written in 2000, during that period the supportive care possibilities might have been limited in Turkey. We still suggest that the FLAG-IDA regimen can be attempted in children with poor-prognosis acute leukemia in Turkey.
Fleischhack et al [2] observed cardiac toxicity in 3 of 24 (12.5%) FLAG- IDA courses. Nine children (9/17, 52.9%) had pulmonary involvement and 2 (2/17, 11.8%) were suspected to have fungal infections in that study. In our series, cardiotoxicity was observed in 9/25 (36.0%) children on echocar- diography, and these 9 children received the second cycles of the regimen as FLAG, without IDA. A mild or asymptomatic decrease in left ventricle EF was observed in these children and none developed congestive heart failure.
Additionally, on the follow-up echocardiographic evaluation, the decrease in left ventricle EF was improved. During the follow-up, 8 (32.0%) children had pulmonary fungal infections. Mucositis and febrile neutropenia were other toxic side effects of the regimen observed in our series. Long-term neutropenia, toxic injury of the lung epithelial cells by cytotoxic drugs, and long-term T-helper cell depletion produced by fludarabine are among the causes of infectious complications, including pulmonary fungal infections, mucositis, and febrile neutropenia [16, 17].
In our series, only 2 courses resulted in fatal infections (both fungal infections) before marrow recovery, and we thus consider the toxicity of the FLAG-IDA regimen to be acceptable for children with poor-prognosis acute leukemia.
In conclusion, the complete remission rate using the FLAG-IDA regimen was quite high in the present study, and we thus suggest that this regimen is a possible option, especially prior to allogeneic HSCT, in children with poor-prognosis acute leukemia.