|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
© 2000 American Society for Clinical Oncology Use of Recombinant Human Granulocyte Colony-Stimulating Factor to Increase Chemotherapy Dose-Intensity: A Randomized Trial in Very High-Risk Childhood Acute Lymphoblastic LeukemiaFrom the University Hospital Centers at Marseille, Paris-Trousseau, Paris-St Louis, Rennes, Nancy, Tours, Clermont-Ferrand, Bordeaux, Brest, Bayonne, Amiens, and Rouen, France. Address reprint requests to Gérard Michel, MD, Pédiatrie et Hématologie Pédiatrique, Hôpital d’Enfants La Timone, 264 Rue St Pierre, 13385, Marseille Cedex 05, France; email gmichel{at}ap-hm.fr
PURPOSE: To determine whether the use of a recombinant human granulocyte colony-stimulating factor ([G-CSF] lenogastrim) can increase the chemotherapy dose-intensity (CDI) delivered during consolidation chemotherapy of childhood acute lymphoblastic leukemia (ALL). PATIENTS AND METHODS: Sixty-seven children with very high-risk ALL were randomized (slow early response to therapy, 55 patients; translocation t(9;22) or t(4;11), 12 patients). Consolidation consisted of six courses of chemotherapy; the first, third, and fifth courses were a combination of high-dose cytarabine, etoposide, and dexamethasone (R3), whereas the second, fourth, and sixth courses included vincristine, prednisone, cyclophosphamide, doxorubicin, and methotrexate (COPADM). G-CSF was given after each course, and the next scheduled course was started as soon as neutrophil count was > 1 x 109/L and platelet count was > 100 x 109/L. CDI was calculated using the interval from day 1 of the first course to hematologic recovery after the fifth course (100% CDI = 105-day interval). RESULTS: CDI was significantly increased in the G-CSF group compared with the non–G-CSF group (mean ± 95% confidence interval, 105 ± 5% v 91 ± 4%; P < .001). This higher intensity was a result of shorter post-R3 intervals in the G-CSF group, whereas the post-COPADM intervals were not statistically reduced. After the R3 courses, the number of days with fever and intravenous antibiotics and duration of hospitalization were significantly decreased by G-CSF, whereas reductions observed after COPADM were not statistically significant. Duration of granulocytopenia was reduced in the G-CSF group, but thrombocytopenia was prolonged, and the number of platelet transfusions was increased. Finally, the 3-year probability of event-free survival was not different between the two groups. CONCLUSION: G-CSF can increase CDI in high-risk childhood ALL. Its effects depend on the chemotherapy regimen given before G-CSF administration. In our study, a higher CDI did not improve disease control.
THE PROGRESSIVE intensification of chemotherapy regimens has improved the cure rate of children with acute lymphoblastic leukemia (ALL) and unfavorable presenting features such as elevated WBC count, adolescent age, or a T-cell lineage phenotype.1-6 However, the risk of disease recurrence remains persistently high for a subgroup of patients who respond slowly to initial therapy7 or who have t(4;11) or t(9;22).8 Hematopoietic colony-stimulating factors (CSFs) were initially introduced into clinical practice as supportive measures to reduce the likelihood of neutropenic complications caused by chemotherapy. By reducing these complications, administration of a CSF might also allow delivery of a higher chemotherapy dose-intensity (CDI), but this approach currently remains experimental.9 The goal of this study was to determine whether CSFs can increase CDI delivered to children with ALL who have a particularly high risk of treatment failure and to evaluate the impact of such a strategy on disease control. For this purpose, the use of a recombinant human granulocyte colony-stimulating factor ([G-CSF] lenogastrim) was randomized during consolidation therapy of children with either slow response to initial therapy or with t(4;11) or t(9;22).
Patient Selection Criteria Children enrolled onto the French pediatric ALL protocol FRALLE 93 were eligible for the G-CSF study if they were initially treated in the high-risk branch of the protocol, they achieved a complete remission after the 4-week induction chemotherapy, they lacked an HLA-identical sibling donor, and if they had, in addition to their high-risk features, at least one of the following unfavorable prognosis factors: slow early response to prednisone or chemotherapy or t(9;22) or t(4;11) (Table 1). Of 982 children who were registered in the FRALLE 93 protocol from June 1993 through January 1998, 67 (6.8%) fulfilled all these eligibility criteria and were randomly assigned to receive G-CSF during consolidation therapy.
The slow response to prednisone was defined by a peripheral blast cell count greater than 1 x 109/L at day 7 of induction therapy after exposure to prednisone 60 mg/m2 for 7 days and intrathecal administration of methotrexate, cytarabine, and corticosteroids on day 0. Slow response to chemotherapy was defined as more than 25% marrow blasts on day 21, after exposure to the 7-day steroid prephase, and through the next 2 weeks of induction chemotherapy (two vincristine infusions at 1.5 mg/m2, two daunorubicin infusions at 40 mg/m2, and 14 days of prednisone therapy at 40 mg/m2).
Therapeutic Schedule
G-CSF Study Design Patients were randomized to receive or not to receive recombinant human G-CSF (lenogastrim) after each consolidation course. All children were randomized together, without stratification according to eligibility criteria. G-CSF 5 µg/kg/d was administered subcutaneously, starting 24 hours after chemotherapy, until achieving an absolute neutrophil count (ANC) of greater than 1 x 109/L. In the G-CSF group, the next scheduled course of chemotherapy was started 24 hours after G-CSF was discontinued but with a sustained ANC greater than 1 x 109/L. Children randomized to not receive G-CSF did not receive placebo. For this non–G-CSF group, an ANC greater than 1 x 109/L was required before chemotherapy commenced. A platelet count greater than 100 x 109/L and absence of major nonhematologic toxicity was required before chemotherapy in both groups. In addition, a 14-day minimum interval after day 1 of the preceding course was necessary before beginning a new consolidation course. The primary end point of this study was CDI during consolidation courses. CDI was calculated using the interval from day 1 of the first course of therapy to hematologic recovery after the fifth course. A 100% dose-intensity was defined as an interval equal to 105 days, and for each patient, the CDI was calculated as ([105 divided by the interval] x 100). Therefore, CDI was higher than 100% if the interval was less than 105 days and  100% in the other cases. Secondarily, we studied infectious toxicity (number of days with fever, number of days with intravenous [IV] antibiotics, and risk of septicemia), the number of days that patients were hospitalized, the number of days with aplasia, transfusion requirements, and mucosal toxicity. The 3-year probabilities of relapse and disease-free survival were calculated in both groups.
Supportive Care, Management of Fever, and Neutropenia
Statistical Methods
Patient Characteristics Sixty-seven patients were randomized, 34 with G-CSF and 33 without. Overall, 55 patients (82%) were enrolled onto this study because of slow response to therapy (slow prednisone response only, 36 patients; slow response to chemotherapy only, 10 patients; and slow response to both prednisone and chemotherapy, nine patients), whereas in 12 patients (18%), the main inclusion criterion was based on cytogenetic features (t(4;11), seven patients or t(9;22), five patients). As listed in Table 3, the number of children with t(4;11) was higher in the non–G-CSF group than in the G-CSF group (six patients v one patient, respectively; P = .05). This difference explained the higher mean WBC count at diagnosis in the non–G-CSF group (223 x 109/L v 136 x 109/L, respectively) and the higher incidence of infants less than 1 year old in the same group (six patients v one patient, respectively). The frequency of T-cell phenotype, meningeal involvement, and the distribution of different types of slow initial response were similar.
CDI Overall, CDI was higher in the G-CSF group than in the non–G-CSF group (mean ± 95% CI, 105% ± 5% v 91% ± 4%; P < .001). When intervals between each consolidation course were separately analyzed (Table 4), it was shown that the duration of intervals after course 1, 3, and 5 (R3 regimen) was significantly shortened in the G-CSF group when compared with the non–G-CSF group. More precisely, the mean duration of the intervals was 4.1 days shorter after course 1 (first R3), 2.6 days shorter after course 3 (second R3), and 3.8 days shorter after course 5 (third R3). The observed shortening was less and not statistically significant after courses 2 and 4 (COPADM regimen). Because of the imbalance in the t(4;11) distribution between groups, we also calculated CDI as well as intervals between each consolidation course after excluding data from children with this translocation. The same conclusions were drawn (mean ± 95% CI: CDI was 106% ± 5% in the G-CSF group v 92% ± 4% in the non–G-CSF group; P < .001). Post-R3 intervals were significantly shortened in the G-CSF group, whereas the shortening was not statistically significant after COPADM.
Chemotherapy-Related Toxicity Neutropenia, fever, and hospitalization. As listed in Table 5, duration of neutropenia was reduced in the G-CSF group when compared with the non–G-CSF group. This reduction was statistically significant after COPADM as well as after R3 regimens but was less pronounced after COPADM than after R3. Number of days with fever, number of days with IV antibiotics, and duration of hospitalization was reduced in the G-CSF group, but the difference was only observed after R3 regimens and not after COPADM regimens. The risk of septicemia per patient and per course was 4% in the G-CSF group and 11% in the non–G-CSF group (P = .075).
Thrombocytopenia and transfusion requirement. Table 6 demonstrates that duration of thrombocytopenia was slightly but significantly longer in the G-CSF group, particularly after COPADM regimens. This longer thrombocytopenia translated to greater platelet transfusion requirements in the G-CSF group, whereas the number of packed RBC transfusions was the same in both groups.
Oral mucositis. Clinical mucositis occurred more frequently after the COPADM regimen than after the R3 regimen (Table 7). Use of G-CSF slightly decreased the incidence of mucositis after the R3 regimen, but when analysis was restricted to severe mucositis requiring opioid analgesics, the difference was no longer statistically significant. G-CSF did not significantly influence the frequency of mucositis after the COPADM regimen.
Outcome of Antileukemic Therapy Thirty-eight children (17 in the G-CSF group and 21 in the non–G-CSF group) are alive in first complete remission, with a median follow-up of 27 months. The 3-year probability of disease-free survival of the two groups did not statistically differ (mean ± 95% CI: 47% ± 9% with G-CSF v 55% ± 10% without G-CSF) (Fig 1A). When patients with either t(9;22) or t(4;11) were excluded, the 3-year disease-free survival (mean ± 95% CI) was 54% ± 10% in the G-CSF group versus 62% ± 11% in the non–G-CSF group (Fig 1B). Twenty-six children relapsed, 15 in the G-CSF group and 11 in the non–G-CSF group. Three patients in each group relapsed before their planned graft. The 3-year probabilities of relapse (mean ± 95% CI) were 49% ± 9% in the G-CSF group and 41% ± 10% in the non–G-CSF group. Three patients died from therapy-related toxicity. One child in the G-CSF arm developed pneumonia caused by respiratory syncytial virus at day 14 after his second COPADM, remained profoundly neutropenic thereafter, and died from septic shock at day 34. Two children (one in each randomized arm) with t(4;11) died from transplant-related complications after an unrelated allogeneic bone marrow transplantation.
In the study presented here, we developed a strategy to increase postinduction CDI in a subgroup of children with ALL who had a particularly high risk of relapse under conventional chemotherapy. Criteria used to define this subgroup of patients was either slow response to therapy associated with another high-risk feature or the presence of t(4;11) or t(9;22). The prognostic significance of these translocations is clearly established. Similarly, slow early response to therapy is a consistent and independent poor prognostic feature in childhood ALL.7 Various criteria have been used in different trials to assess the rapidity of response. The persistence of more than 25% blasts in the bone marrow (M3 marrow rating) on day 7 or day 14 after the beginning of polychemotherapy has been identified as a poor prognostic criteria in numerous studies.6,10-13 German investigators also reported poor outcomes for children with slow early response to corticotherapy, defined as persistence of more than 1,000 blasts/µL in peripheral blood after a 7-day course of prednisone and one intrathecal administration of methotrexate.4,14 In our study, we considered children with either poor response to corticotherapy at day 7 or an M3 day-21 marrow as being slow early responders, which corresponds to the marrow response observed after a 7-day exposure to corticotherapy alone followed by a 14-day polychemotherapy. Of note, this cohort of slow responders was drawn from the high-risk group of the FRALLE 93 protocol and, therefore, constituted a subgroup of highly selected patients with regard to their risk of treatment failure. In this group of children, we found that use of G-CSF during the consolidation phase resulted in a 14% increase of CDI. Our data also show that the effects of G-CSF depend on the chemotherapeutic regimen given before G-CSF therapy. More precisely, the increase in CDI observed in this study was mainly caused by shortening of the post-R3 intervals, whereas post-COPADM intervals were not significantly affected. Similarly, the beneficial effects of reducing the number of days with fever, number of days with IV antibiotics, and duration of hospitalization was statistically significant after R3 courses but not after COPADM courses. This discrepancy could be because of differences in hematologic as well as in nonhematologic toxicity induced by the two chemotherapy regimens. Myelosuppression was less prolonged after the COPADM regimen than after the R3 regimen; the mean number of days with ANC < 0.5 x 109/L was 9.7 days after R3 versus 7.6 days after COPADM for children treated without G-CSF. Interestingly, the granulocytopenic period shortening observed in the G-CSF arm was more pronounced after R3 than after COPADEM. It has been previously observed that the effects of cytokines become more significant when myelosuppression is prolonged. Of four published studies that tested the potential benefit of G-CSF after chemotherapy for childhood ALL, two demonstrated that supportive therapy with G-CSF may be unnecessary in children with neutropenia of short duration.15,16 The other two studies demonstrated that G-CSF treatment had some clinical benefit in children who receive intensive chemotherapy during induction17 or consolidation phases.18 Pui et al17 found that G-CSF therapy, as compared with placebo, accelerated recovery from neutropenia after myelosuppressive remission induction in children with ALL. The beneficial effects of reducing the period of neutropenia included decreases in incidence of documented infections, duration of hospitalization, and likelihood of delay in starting consolidation chemotherapy on schedule. However, G-CSF did not result in decreased rate of hospitalization for febrile neutropenia, higher probability of disease-free survival, or lower supportive care costs. Welte et al18 showed that prophylactic G-CSF, given after each of nine intensive consolidation courses, significantly reduced neutropenia, febrile neutropenia, culture-confirmed infections, and use of parenteral antibiotics. Although a tighter adherence to treatment schedule was facilitated by G-CSF when compared with the non–G-CSF group, disease control was not improved. The fact that two types of chemotherapeutic regimens can lead to different outcomes of G-CSF therapy is probably not completely explained by differences in the myelosuppressive effects of these regimens. In some instances, differences in nonhematologic toxicity, such as more severe oral mucositis, may play a role. In our study, severe oral mucositis occurred much more frequently after the COPADM regimen than after the R3 regimen. This high frequency of severe mucositis was not influenced by use of G-CSF and could have jeopardized the beneficial effect of reducing the duration of neutropenia. Similarly, the prolonged thrombocytopenia duration induced by G-CSF use might be a limiting factor for dose intensification. However, this delayed platelet recovery was observed after both R3 and COPADEM and is, therefore, unlikely to explain the difference in dose intensification between the two regimens. In our study, CDI increased with the use of G-CSF, but this effect did not translate into improved disease-free survival. One explanation for this result could be that the magnitude of CDI increase was only 14%, which might have been insufficient to decrease relapse risk and improve outcome. On the other hand, authors from the Children’s Cancer Group recently reported a large randomized trial of augmented postinduction therapy for children with high-risk ALL and a slow response to initial therapy (slow response being defined as > 25% blasts in bone marrow at day 7).19 They found that the outcome at 5 years was significantly better in the augmented-therapy group than in the standard-therapy group. When compared with the standard arm, the augmented therapy included both increased dose-intensity and prolonged therapy duration. The relative contributions of these two components of the augmented therapy could not be distinguished. Our study suggests that increased dose-intensity does not have a crucial role, at least with the schedule used in our protocol and in the setting of a relatively short intensive therapy duration. In a new French therapeutic trial for children with ALL and a slow response to initial therapy, we plan to study dose intensification in the setting of a more prolonged intensive chemotherapy duration. We conclude that G-CSF can increase CDI in high-risk childhood ALL. However, in our study, this higher CDI did not translate into improved disease-free survival. G-CSF can also decrease the number of days with fever and IV antibiotics and duration of hospitalization. The different effects of G-CSF depend on intensity and schedule of the chemotherapy regimen.
1. Steinherz PG, Gaynon P, Miller DR, et al: Improved disease-free survival of children with acute lymphoblastic leukemia at high risk for early relapse with the New York regimen: A new intensive therapy protocol—A report from the Children’s Cancer Study Group. J Clin Oncol 4:744-752, 1986 2. Gaynon PS, Steinherz PG, Bleyer WA, et al: Improved therapy for children with acute lymphoblastic leukemia and unfavorable presenting features: A follow-up report of the Children’s Cancer Group Study CCG-106. J Clin Oncol 12:2234-2242, 1993
3.
Rivera GK, Pinkel D, Simone JV, et al: Treatment of acute lymphoblastic leukemia: A 30 years’ experience at St. Jude Children’s Research Hospital. J Med 329:1289-1295, 1993
4.
Reiter A, Schrappe M, Ludwig WD, et al: Chemotherapy in 998 unselected childhood acute lymphoblastic leukemia patients: Results and conclusions of the multicenter trial ALL-BFM 86. Blood 84:3122-3133, 1994 5. Schorin MA, Blattner S, Gelber RD, et al: Treatment of childhood acute lymphoblastic leukemia: Results of Dana-Farber Cancer Institute/Children’s Hospital Acute Lymphoblastic Leukemia Consortium Protocol 85-01. J Clin Oncol 12:740-747, 1994[Abstract] 6. Chessels JM, Bailey C, Richards SM: Intensification of treatment and survival in all children with lymphoblastic leukaemia: Results of the UK Medical Research Council Trial UKALL X. Lancet 345:143-148, 1995[Medline] 7. Gaynon PS, Desai AA, Bostrom BC, et al: Early response to therapy and outcome in childhood acute lymphoblastic leukemia: A review. Cancer 80:1717-1726, 1997[Medline] 8. Chessels JM, Swansbury GJ, Reeves B, et al: Cytogenetics and prognosis in childhood lymphoblastic leukaemia: Results of MRC UKALL X—Medical Research Council Working Party in Childhood Leukaemia. Br J Haematol 99:93-100, 1997[Medline]
9.
American Society of Clinical Oncology: Recommendations for the use of hematopoietic colony-stimulating factors: Evidence-based, clinical practice guidelines. J Clin Oncol 12:2471-2508, 1994 10. Gaynon PS, Bleyer WA, Steinherz PG, et al: Day 7 marrow response and outcome for children with lymphoblastic leukemia and unfavorable presenting features. Med Pediatr Oncol 18:273-279, 1990[Medline] 11. Miller DR, Leikin S, Albo V, et al: Prognostic factors and therapy in acute lymphoblastic leukemia of childhood: CCG-141—A report from Children’s Cancer Study Group. Cancer 51:1041-1049, 1983[Medline] 12. Miller DR, Coccia PF, Bleyer WA, et al: Early response to induction therapy as a predictor of disease-free survival and late recurrence of childhood acute lymphoblastic leukemia: A report from Children’s Cancer Study Group. Oncol 7:1807-1815, 1989
13.
Steinherz PG, Gaynon PS, Breneman JC, et al: Cytoreduction and prognosis in acute lymphoblastic leukemia: The importance of early marrow response—Report from Children’s Cancer Group. J Clin Oncol 14:389-398, 1996 14. Riehm H, Feickert HJ, Schrappe M, et al: Therapy results in five ALL-BFM studies since 1970: Implications of risk factors for prognosis. Hamatology Blutttransfusion 30:139-146, 1987 15. Dibenedetto SP, Ragusa R, Ippolito AM, et al: Assessment of the value of treatment with granulocyte colony-stimulating factor in children with acute lymphoblastic leukemia: A randomized clinical trial. Eur J Haematol 55:93-96, 1995[Medline] 16. Laver J, Amylon M, Desai S, et al: Randomized trial of r-metHu granulocyte colony-stimulating factor in an intensive treatment for T-cell leukemia and advanced-stage lymphoblastic lymphoma in childhood: A pediatric oncology group study. J Clin Oncol 16:522-526, 1998[Abstract]
17.
Pui CH, Boyett JM, Hughes WT, et al: Human granulocyte colony-stimulating factor after induction chemotherapy in children with acute lymphoblastic leukemia. N Engl J Med 336:1781-1787, 1997
18.
Welte K, Reiter A, Mempel K, et al: A randomized phase-III study of the efficacy of granulocyte colony-stimulating factor in children with high-risk acute lymphoblastic leukemia. Blood 87:3143-3150, 1996
19.
Nachman JB, Sather HN, Sensel MG, et al: Augmented post-induction therapy for children with high-risk acute lymphoblastic leukemia and a slow response to initial therapy. N Engl J Med 338:1663-1671, 1998 Submitted September 9, 1999; accepted December 6, 1999. This article has been cited by other articles:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||
|
|||||||||||
|
|
Copyright © 2000 by the American Society of Clinical Oncology, Online ISSN: 1527-7755. Print ISSN: 0732-183X
|
TOP
ABSTRACT
INTRODUCTION
