Original Article

Augmented Berlin-Frankfurt-M€ unster Therapy in Adolescents and Young Adults (AYAs) With Acute Lymphoblastic Leukemia (ALL) Michael E. Rytting, MD1,2; Deborah A. Thomas, MD2; Susan M. O’Brien, MD; Farhad Ravandi-Kashani, MD2; Elias J. Jabbour, MD2; Anna R. Franklin, MD1; Tapan M. Kadia, MD2; Naveen Pemmaraju, MD2; Naval G. Daver, MD2; Alessandra Ferrajoli, MD2; Guillermo Garcia-Manero, MD2; Marina Y. Konopleva, MD, PhD2; Jorge E. Cortes, MD2; Gautham Borthakur, MD2; Rebecca Garris, BA2; Maria Cardenas-Turanzas, MD2; Kurt Schroeder, RN2; Jeffrey L. Jorgensen, MD, PhD3; Steven M. Kornblau, MD2; and Hagop M. Kantarjian, MD2

BACKGROUND: Various trials have reported improved outcomes for adolescents and young adults (AYAs) with acute lymphoblastic leukemia (ALL) who received treatment with pediatric-based regimens. Those reports prompted the current investigation of the pedi€nster (ABFM) regimen in AYA patients. The results were compared with those from a similar popatric augmented Berlin-Frankfurt-Mu ulation that received the hyperfractionated cyclophosphamide, vincristine, doxorubicin, and dexamethasone (hyper-CVAD) regimen. METHODS: Eighty-five patients ages 12 to 40 years who had Philadelphia chromosome (Ph)-negative ALL received the ABFM regimen from October 2006 through April 2012. Their outcome was compared with outcomes in 71 historic AYA patients who received hyper-CVAD from the authors’ institution. Patient and disease characteristics, as well as minimal residual disease status, were analyzed for their impact on outcomes. RESULTS: The complete response rate with ABFM was 94%. The 3-year complete remission duration (CRD) and overall survival (OS) rates were 70% and 74%, respectively. For patients aged 21 years, the 3-year CRD and OS rates were 72% and 85%, respectively; and, for patients ages 21 to 40 years, the respective rates were 69% and 60%. The initial white blood cell count was an independent predictive factor of OS and CRD. The minimal residual disease status on days 29 and 84 of therapy also were predictive of long-term outcomes. Severe regimen toxicities included transient hepatotoxicity in 35% to 39% of patients, pancreatitis in 11% of patients, osteonecrosis in 11% of patients, and thrombosis in 22% of patients. The 3-year OS rate was 74% in the ABFM group versus 71% in the hyper-CVAD group, and the corresponding 3-year CRD rate was 70% versus 66%, respectively. CONCLUSIONS: ABFM was tolerable in AYA patients with ALL but was not associated with significant improvements in CRD and OS C 2014 American Cancer Society. compared with hyper-CVAD. Cancer 2014;000:000-000. V €nster, KEYWORDS: pediatric-based therapy, acute lymphoblastic leukemia, pediatric-based regimens, augmented Berlin-Frankfurt-Mu Philadelphia chromosome-negative ALL.

INTRODUCTION In retrospective comparisons, the outcomes of adolescent and young adults (AYA) with acute lymphocytic leukemia (ALL) treated on pediatric protocols have been superior to the outcomes of similar patients treated on adult protocols.1-4 An exception is the study by Usvasalo et al,5 who observed no difference in outcomes between adult and pediatric-based treatments. Several studies have investigated pediatric-based regimens in adults with ALL up to age 55 years.6-8 Older adults (ages 40 to 45 years) had significantly worse toxicities with pediatric-based regimens.7 Nachman and colleagues demonstrated that the augmented Berlin-Frankfurt-M€ unster (ABFM) regimen could be administered successfully to patients up to age 21 years and was associated with very favorable survival, particularly in patients who had a rapid response to induction therapy.9 To improve the outcome of AYAs with ALL at our institution, ABFM therapy was investigated in newly diagnosed patients aged 40 years who had Philadelphia (Ph) chromosome-negative ALL. This regimen was chosen because of the success of ABFM therapy in treating AYAs ages 16 to 21 years and its acceptable toxicity profile. Herein, we report our results using ABFM and compare the outcomes of patients who received ABFM with those from a similar historic AYA

Corresponding author: Michael E. Rytting, MD, Department of Pediatrics, The University of Texas MD Anderson Cancer Center, 1515 Holcombe Boulevard, Houston, TX 77030; Fax: (713) 792-0608; [email protected] 1 Department of Pediatrics-Patient Care, The University of Texas MD Anderson Cancer Center, Houston, Texas; 2Department of Leukemia, The University of Texas MD Anderson Cancer Center, Houston, Texas; 3Hematopathology, The University of Texas MD Anderson Cancer Center, Houston, Texas

DOI: 10.1002/cncr.28930, Received: April 10, 2014; Revised: May 20, 2014; Accepted: May 21, 2014, Published online Month 00, 2014 in Wiley Online Library (wileyonlinelibrary.com)

Cancer

Month 00, 2014

1

Original Article TABLE 1. Treatment Induction (4 wk) Daunorubicin 25 mg/m2 weekly for 4 doses Vincristine 2 mg weekly for 4 doses Prednisone 60 mg/m2 po daily for 28 d PEG-asparaginase 2500 IU on d 4 of induction Intrathecal cytarabine on d 1 Intrathecal methotrexate 12 mg on d 8 and d 29 Consolidation 1 (8 wk) Cyclophosphamide 1 g/m2 in wk 1 and wk 5 Cytarabine 75 mg/m2 sc on d 1-4 and d 8-11 Mercaptopurine 60 mg/m2 po daily on d 1-14 Vincristine 2 mg in wk 3 and wk 4 PEG-asparaginase 2500 IU in wk 3 and wk 4 Intrathecal methotrexate 12 mg weekly in wk 1-4 Consolidation 2 Vincristine 2 mg every 10 d for 5 doses Methotrexate iv every 10 d starting at 100 mg/m2 and increasing by 50 mg/m2 as tolerated PEG-asparaginase 2500 IU in wk 1 and wk 4 Intrathecal methotrexate 12 mg in wk 1 Consolidation 3A Doxorubicin 25 mg/m2 weekly for 3 doses Dexamethasone 10 mg/m2 po on d 1-7 and d 15-21 Vincristine 2 mg iv weekly for 3 doses PEG-asparaginase 2500 IU in wk 1 Intrathecal methotrexate 12 mg in wk 1 Consolidation 3B Cyclophosphamide 1 g/m2 in wk 1 Cytarabine 75 mg/m2 sc on d 1-4 and d 8-11 Thioguanine 60 mg/m2 daily for 14 d Intrathecal methotrexate 12 mg in wk 1 and wk 2 Vincristine 2 mg in wk 3 and wk 4 PEG-asparaginase 2500 IU in wk 3 Maintenance (24 mo) Mercaptopurine 75 mg/m2 po nightly Methotrexate 20 mg/m2 po weekly, hold on days of intrathecal methotrexate Dexamethasone 6 mg/m2 po daily on d 1-5 every 28 d Vincristine 2 mg every 28 d Intrathecal methotrexate 12 mg every 3 mo for 4 doses Slow responders: Repeat consolidation 2 and consolidation 3A/3B before maintenance Testicular disease at diagnosis: If resolution by end of induction, then therapy continued; if abnormality persisted, then a biopsy was required and radiation administered if positive CNS disease at diagnosis: Intrathecal therapy given twice weekly until cleared, then weekly for 6 doses, then resume intrathecal treatments as per protocol; radiation was recommended for overt CNS disease but was per investigator choice Abbreviations: CNS, central nervous system; iv, intravenously; PEG, pegylated; po, orally; sc, subcutaneously.

population that received the hyperfractionated cyclophosphamide, vincristine, doxorubicin, and dexamethasone (hyper-CVAD) regimen. MATERIALS AND METHODS Study Group

Patients with Ph chromosome-negative ALL ages 12 to 30 years were initially enrolled. After evaluating toxicities in the first 10 patients, the upper age limit was expanded to 40 years. The diagnosis of ALL was confirmed by bone 2

marrow morphology and flow cytometry. Other eligibility criteria included an Eastern Cooperative Oncology Group performance status 3 and adequate renal and hepatic functions (unless the abnormalities were attributed to leukemia). Steroid treatment before enrollment initially was limited to 3 days, but this restriction was removed later. The protocol was approved by the MD Anderson Cancer Center Institutional Review Board, and informed consent for therapy was obtained according to the Declaration of Helsinki and our institutional guidelines. Treatment

Treatment details are provided in Table 1 and have been previously published.10 Chemotherapy started within 72 hours of the first intrathecal chemotherapy. Bone marrow was reassessed at day 15. Patients who had 5% bone marrow blasts on day 29 received 2 weeks of extended induction. At the end of the extended induction, patients who had >5% bone marrow blasts were taken off study. Early responders received 12 intrathecal therapies (ITs), whereas slow responders received 16 ITs. Patients with overt leukemia in the spinal fluid received treatment with intensified ITs (see Table 1). Radiation for overt central nervous system (CNS) leukemia was recommended. Wright-Giemsa staining was used to assess routine morphology and the percentage of bone marrow blasts. Myeloperoxidase immunohistochemistry and 4-color flow cytometry were performed. Rapid fluorescence in situ hybridization (FISH) was used to establish preliminary Ph chromosome status using breakpoint cluster region-Abelson murine leukemia viral oncogene (BCRABL) probes. Rapid Ph chromosome testing was corroborated with conventional cytogenetics and polymerase chain reaction testing. Bone marrow morphology and minimal residual disease (MRD) status were assessed on day 29 and on approximately day 84 of treatment. B-cell markers included cluster of differentiation 9 (CD9), CD10, CD13, CD15, CD19, CD20, CD22, CD34, and CD58 (sensitivity, 1024). MRD status in the patients with T-cell morphology was followed using a panel of T-cell markers that included CD1, CD2, CD3 (surface and cytoplasmic), CD4, CD5, CD7, CD8, and CD 10. Standard cytogenetic studies were evaluated at diagnosis and on days 29 and 84 of therapy. Polymerase chain Cancer

Month 00, 2014

Pediatric-Based Therapy: AYAs With ALL/Rytting et al

TABLE 2. Patient Characteristics at the Time of Enrollment and Comparison With Historic Patients Who Received Hyperfractionated Cyclophosphamide, Vincristine, Doxorubicin, and Dexamethasone Characteristic Total no. of patients No. of males/females Age: Median [range], y No. with performance status of 2 (%) WBC at diagnosis: Median [range], 3109/L No. with pre-B phenotype (%) Percentage blasts in blood: Median [range] No. with karyotype (%) Diploid Hyperdiploid Hypodiploid Pseudodiploid MLL ND/IM No. positive for CNS disease at diagnosis (%) No. transplanted in first CR (%) Treatment period

ABFM

HCVAD

P

85 49/36 21 [13-39] 8 (9) 14 [0.4-494.2] 69 (81) 30 [0-93]

71 42/29 26 [16-40] 4 (6) 6.7 [0.6-334.2] 62 (87) 28 [0-96]

.85 21 years; P value not significant). Isolated CNS relapse occurred in 6 patients (7%), and 3 patients (4%) had a relapse in the spinal fluid and bone marrow. Sixteen patients (19%) patients had a bone marrow relapse alone. The median time to relapse was 21.9 months in patients aged 21 years and 11.5 months in patients aged >21 years.

Sixty patients (71%) remained alive with at a median follow-up of 40 months (range, 4-75 months). The overall 3-year OS and CRD rates were 74% and 70%, respectively. The 3-year OS and CRD rates were 85% and 72%, respectively, for patients aged 21 years and 60% and 69%, respectively, for patients aged >21 years (Fig. 1A,B). The difference in the 3-year OS rates for the 2 age groups had borderline significance (P5.055). Outcome by MRD Status

The outcome of patients according to MRD status is detailed in Table 3. The 3-year OS rate was 82% for patients who had MRD-negative status on day 29 versus 65% for those who had MRD-positive status on day 29 (P5.02) (Fig. 1C). The 3-year CRD rate was 75% for patients who had MRD-negative status on day 29 versus 59% for those who had MRD-positive status on day 29 (P5.2) (Fig. 1D). The 3-year OS rate was 83% for patients who had MRD-negative status on day 84 versus 66% for those who had MRD-positive status on day 84 (P5.02), and the corresponding CRD rates were 82% versus 20%, respectively (p5.01) (Table 3). The results within age groups are provided in Table 3. Prognostic Factors

On univariate analysis, the following factors were adverse for OS (P.10): day-29 and day-84 MRD-positive status, leukocytosis >503109/L, older age (continuous variable) (Fig. 1A), and slow responder status. There were interactions between age and MRD status. On multivariate analysis, only leukocytosis (P.001) and day-28 MRD-positive status (P5.05) retained an adverse, independent association with survival. In a second model, we included MRD status on day 84, age, white blood cell (WBC) count at diagnosis, response type (rapid vs slow), and cytogenetics (diploid vs not diploid). Only the WBC count at diagnosis was independently associated with OS (P5.003). On univariate analysis, the following factors were adverse for CRD (P.10): day-84 MRD-positive status Cancer

Month 00, 2014

Pediatric-Based Therapy: AYAs With ALL/Rytting et al

Figure 1. (A-D) Kaplan-Meier curves illustrate overall survival (OS) and complete remission duration (CRD) according to age and day-29 (D29) minimal residual disease. Neg indicates negative; Pos, positive; ns, nonsignificant.

and leukocytosis >503109/L (Fig. 2B). On multivariate analysis, only leukocytosis remained independently adverse (P5.04). Comparison of ABFM and Hyper-CVAD

Details of the hyper-CVAD regimen and results from its use were published previously.15 In the current comparison, we included only patients aged 40 years. Among 71 such patients, 37 (52%) had CD20-positive expression on 20% leukemia cells (median, 77% of leukemia cells; range, 25%-99% of leukemia cells) and received rituximab 375 mg/m2 for 8 doses in the first 4 inductionconsolidation cycles. In a comparison between ABFM and hyper-CVAD, there were no statistically significant differences in the CR rate (ABFM, 94% CR rate; hyper-CVAD, 99% CR rate [70 of 71 patients]) or in CRD or OS rates between the 2 treatment regimens (Fig. 2C,D). The 3-year OS rate was 74% with ABFM versus 71% with hyper-CVAD, and the Cancer

Month 00, 2014

3-year CRD rate was 70% with ABFM versus 66% with hyper-CVAD. When OS and CRD were further analyzed in multivariate analyses that included age, presenting WBC count, and MRD status at the end of induction therapy, no statistically significant differences in outcome were detected between the treatment regimens (data not shown). Regimen Toxicities

Toxicities associated with the ABFM regimen, particularly hepatic toxicity, were significant but expected (Table 4). Grade 3 and 4 hyperbilirubinemia was observed in 39% of patients, and grade 3 and 4 liver enzyme elevations were observed in 35%. Liver toxicity resolved in most patients but resulted in chemotherapy dose decrements and omissions according to protocol guidelines. Hypofibrinogenemia was prominent (36%) but did not result in excessive bleeding. Thrombosis, mostly in the upper extremities in which central lines were located, occurred in 22% of patients, and stroke-like events 5

Original Article

Figure 2. (A-D) Kaplan-Meier curves illustrate overall survival (OS) and complete remission duration (CRD) according to white blood cell (WBC) count at presentation (503109/L or >503109/L) and protocol. HCVAD indicates hyperfractionated cyclophos€nster therapy; ns, phamide, vincristine, doxorubicin, and dexamethasone; Aug BFM, augmented Berlin-Frankfurt-Mu nonsignificant.

TABLE 4. Toxicities Overall and by Age No. of Patients (%)

Toxicity Allergic reaction, asparaginase Grade 3-4 hypofibrinogenemiaa Pancreatitis Grade 3-4 elevated liver enzymes Grade 3-4 elevated bilirubin Osteonecrosis Thrombosis Stroke-like event Grade 3-4 neuropathy

Overall, n585

Aged 21 y, n544

Aged >21 y, n541

18 (21)

9 (20)

9 (22)

31 (36)

10 (23)

21 (51)

9 (11) 30 (35)

4 (9) 12 (27)

5 (12) 18 (44)

33 (39) 9 (11) 19 (22) 3(4) 4(5)

17 (39) 6 (14) 8 (18) 1(2) 2 (4)

16 (39) 3 (7) 11 (27) 2(5) 2 (5)

a P values were not significant by age group except for hypofibringogenemia (P5.006).

6

developed in 3 patients. Toxicities that led to permanent changes in therapy consisted of osteonecrosis (11% of patients), severe allergy to PEG-asparaginase (21% of patients), and pancreatitis (11% of patients). The incidence of osteonecrosis was comparable to that reported previously in adolescents who were treated on pediatric trials.10 A higher incidence of severe asparaginase allergy was noted on study compared with that reported in the pediatric literature.16 Neuropathy was not prominent: only 4 patients had vincristine held because of grade 3 neuropathy. Infections and febrile episodes during induction and later in therapy were common and did not differ significantly between the age groups. Fever or documented infections were noted in 22% of patients during induction and in 63% of patients during consolidations. Cancer

Month 00, 2014

Pediatric-Based Therapy: AYAs With ALL/Rytting et al

DISCUSSION In this experience in AYA patients with newly diagnosed ALL, the ABFM regimen resulted in a CR rate of 94%, a 3-year OS rate of 74%, and a CRD rate of 70%. Severe toxicities of the regimen were significant but expected and mostly were related to PEG-asparaginase–based therapy. These included hepatotoxicity in 35% to 39% of patients, pancreatitis in 11% of patients, osteonecrosis in 11% of patients, and thrombotic events in 22% of patients. The results from this trial were compared retrospectively with historic data on the hyper-CVAD regimen in a similar population at our institution in relation to CR, OS, and CRD. The 2 populations—those who received ABFM and those who received hyper-CVAD—were well matched (Table 2). The patients who received hyperCVAD were older (age range, 16-40 years; median age, 26 years; P

Augmented Berlin-Frankfurt-Münster therapy in adolescents and young adults (AYAs) with acute lymphoblastic leukemia (ALL).

Various trials have reported improved outcomes for adolescents and young adults (AYAs) with acute lymphoblastic leukemia (ALL) who received treatment ...
396KB Sizes 0 Downloads 5 Views