Received: 17 September 2019 | Revised: 12 December 2019 | Accepted: 16 December 2019 DOI: 10.1002/cam4.2814 ORIGINAL RESEARCH A pediatric regimen for adolescents and young adults with Philadelphia chromosome-negative acute lymphoblastic leukemia: Results of the ALLRE08 PETHEMA trial Josep-Maria Ribera1 | Mireia Morgades1 | Pau Montesinos2 | Mar Tormo3 | Daniel Martínez-Carballeira4 | José González-Campos5 | Cristina Gil6 | Pere Barba7 | Raimundo García-Boyero8 | Rosa Coll9 | María Pedreño10 | Jordi Ribera1 | Santiago Mercadal11 | Susana Vives1 | Andrés Novo12 | Eulàlia Genescà1 | Jesús-María Hernández-Rivas13 | Juan Bergua14 | María-Luz Amigo15 | Ferran Vall-Llovera16 | Pilar Martínez-Sánchez17 | María Calbacho18 | Irene García-Cadenas19 | Antoni Garcia-Guiñon20 | María-José Sánchez-Sánchez21 | Marta Cervera22 | Evarist Feliu1 | Alberto Orfao13 | on behalf of the PETHEMA Group, Spanish Society of Hematology 1 Departments of Clinical Hematology, ICO-Hospital Germans Trias i Pujol. Josep Carreras Research Institute, Universitat Autònoma de Barcelona, Barcelona, Spain 2 Departments of Clinical Hematology, Hospital Universitari i Politècnic La Fe, CIBERONC, Instituto Carlos III, Valencia, Spain, Madrid, Spain 3 Departments of Clinical Hematology, Hospital Clínico Valencia, Valencia, Spain 4 Hospital Central de Asturias, Oviedo, Spain 5 Departments of Clinical Hematology, Hospital Universitario Virgen del Rocío, Sevilla, Spain 6 Departments of Clinical Hematology, Hospital General Universitario de Alicante, Alicante, Spain 7 Departments of Clinical Hematology, Hospital Universitari Vall d’Hebron, Universitat Autònoma de Barcelona, Barcelona, Spain 8 Departments of Clinical Hematology, Hospital General Universitario de Castellón, Castelló, Spain 9 Departments of Clinical Hematology, ICO-Hospital Josep Trueta, Girona, Spain 10 Departments of Clinical Hematology, Hospital Dr. Peset, Valencia, Spain 11 Departments of Clinical Hematology, ICO-Hospital Duran i Reynals, L’Hospitalet de Llobregat, Barcelona, Spain 12 Departments of Clinical Hematology, Hospital Son Espases, Palma de Mallorca, Spain 13 Departments of Clinical Hematology, Hospital Clínico Universitario, Cancer Research Center (IBMCC-CSIC/USAL), Cytometry Service CIBERONC, Salamanca, Spain 14 Departments of Clinical Hematology, Hospital San Pedro de Alcántara, Cáceres, Spain 15 Departments of Clinical Hematology, Hospital Morales Meseguer, Murcia, Spain 16 Departments of Clinical Hematology, Hospital Mútua de Terrassa, Terrassa, Spain 17 Departments of Clinical Hematology, Hospital Doce de Octubre, Madrid, Spain 18 Departments of Clinical Hematology, Hospital Ramón y Cajal, Madrid, Spain 19 Departments of Clinical Hematology, Hospital de Sant Pau, Barcelona, Spain 20 Departments of Clinical Hematology, Hospital Arnau de Vilanova, Lleida, Spain 21 Departments of Clinical Hematology, Hospital Lucus Augusti, Lugo, Spain 22 Departments of Clinical Hematology, ICO-Hospital Joan XXIII, Tarragona, Spain Preliminary results were presented at the 59th Annual Meeting of the American Society of Hematology (Atlanta, 9-12 December 2017). Abstract 1273. This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited. © 2020 The Authors. Cancer Medicine published by John Wiley & Sons Ltd. Cancer Medicine. 2020;9:2317–2329.  wileyonlinelibrary.com/journal/cam4 | 2317 2318 |    RIBERA et al. Correspondence Josep-Maria Ribera, Clinical Hematology Abstract Department, ICO-Hospital Germans Trias Background: Pediatric-based or -inspired trials have improved the prognosis of ado- i Pujol, Josep Carreras Research Institute, lescents and young adults (AYA) with Philadelphia chromosome-negative (Ph-neg) Badalona, Spain. Email: jribera@iconcologia.net acute lymphoblastic leukemia (ALL). Methods: This study reports the results of treatment of the ALLRE08 trial, a full Funding information Supported in part by grants PI14/01971 pediatric trial for AYA aged 15-30 years with standard-risk (SR) ALL. from Fondo de Investigaciones Sanitarias, Results: From 2008 to 2018, 89 patients (38 adolescents [15-18 years] and 51 Instituto Carlos III, SGR288 (GRC) young adults [YA, 19-30 years], median age: 20 [15-29] years) were enrolled in the Generalitat de Catalunya, Fundació La Caixa and CIBERONC (JMHR and AO). ALLRE08 trial. The complete response (CR) was 95%. Twenty-two patients were transferred to a high-risk (HR) protocol because of poor marrow response on day 14 (n = 20) or high-level of end-induction minimal residual response (MRD ≥ 0.25%, n = 2). Cumulative incidence of relapse (CIR) at 5 years was 35% (95%CI: 23%- 47%), with significant differences between adolescents and YA: 13% (4%-28%) vs 52% (34%-67%), P = .012. No treatment-related mortality was observed in 66/66 patients following the ALLRE08 trial vs 3/23 patients moved to a HR trial. The esti- mated 5-year overall survival (OS) was 74% (95%CI: 63%-85%), with significantly higher rates for adolescents vs YA: 87% (95%CI: 74%-100%) vs 63% (46%-80%), P = .021. Although CIR or OS were lower in patients who were transferred to a HR trial, the differences were not statistically significant (CIR: 34% [21%-47%] vs 37% [14%-61%]; OS: 78% [66%-90%] vs 61% [31%;91%]). Conclusion: A full pediatric trial is feasible and effective for AYA with Ph-neg, SR- ALL, with better results for adolescents than for YA. Outcome of patients with poor early response rescued with a HR trial was not significantly inferior. KEYWORDS acute lymphoblastic leukemia, adolescents and young adults, pediatric treatment two types of trials relate to the greater experience observed 1 | IN TRO D U C T IO N in pediatric centers and higher adherence of AYA treated in pediatric vs adult cancer centers in some countries.7-10 In recent years, the results of treatment of adolescents and However, a recently published study by three adult US co- young adults (AYA) diagnosed with ALL have significantly operative groups (CALGB, ECOG, and SWOG) demon- improved with the use of pediatric-based or -inspired pro- strated that the use of a full pediatric regimen for young tocols.1-3 This was first observed when retrospective stud- adults (YA) with ALL up to the age of 40 years was feasi- ies demonstrated clear differences in outcomes for AYA ble and effective, resulting in improved survival rates com- depending on enrollment in pediatric vs adult cooperative pared to historical controls.11 group studies, despite most adult treatment regimens evolv- In 1996 the Spanish PETHEMA (Programa Español ing from a pediatric background. These improvements de Tratamientos en Hematología) Group conducted a have also been confirmed in population-based studies.4 full pediatric trial (ALL96) to treat Philadelphia chromo- However, the design of pediatric trials significantly differs some-negative (Ph-neg) AYA up to the age of 30 years, from that of adult trials.5 Pediatric trials involve more in- showing no significant differences in outcome between tensive dosing of key therapeutic agents in ALL, including adolescents (15-18 years) and YA (19-30 years).12 At the vincristine, steroids, and asparaginase, while fewer mye- time of the design of the trial the assessment of minimal losuppressive drugs such as anthracyclines and cytarabine residual disease (MRD) for tailored treatment was not ex- are used in these trials. In addition, hematopoietic stem cell tensively employed. In 2007, the PETHEMA Group devel- transplantation (HSCT) is less used in pediatric trials.6,7 oped another trial (ALLRE08) incorporating MRD to the Other factors contributing to the differences between the decision-making process without changing the inclusion RIBERA et al.    | 2319 criteria and the chemotherapy schedule. This study reports required to define a normal karyotype. Fluorescence in situ the outcomes of the Ph-neg AYA ALL patients included in hybridization (FISH) studies for BCR-ABL1, KMT2A, MYC, this latter trial. and TCF3-PBX1 were performed in cases without a valid cy- togenetic study. 2 | PATIE N TS A N D ME T HOD S 2.3 | Minimal residual disease assessment 2.1 | Eligibility Bone marrow MRD levels were assessed at each participat- From August 2008 to April 2018, adolescents (age range: ing center in CR patients at: i) end of induction (weeks 5-6) 15 to 18 years) and YA age range: 19 to 30 years) with in CR patients, ii) end of consolidation/reinduction (weeks standard-risk (SR) Ph-neg ALL were enrolled in the 19-20), iii) end of the fourth reinduction cycle (weeks 25-26), ALLRE08 PETHEMA trial (clinicaltrials.gov identifier: and iv) end of maintenance (week 106-108) using a 4-color NCT02036489). SR-ALL was defined for both B-cell pre- multiparameter flow cytometry (FCM) approach with a limit cursor ALL or T-ALL patients who fulfilled all the follow- of detection of 10-4. ing criteria: a white blood cell count ≤ 30×109/L, absence of t(9;22)/BCR-ABL rearrangements or t(4;11), or any other 11q23/KMT2A gene rearrangement. This definition 2.4 | Treatment and response criteria base on a minimal set of criteria was identical to that of the ALL-96 trial and allowed a homogeneous recruitment Treatment (Table 1) consisted of a prephase with pred- of patients by the participating centers. Patients were not nisone (maximum of 1 week) and one dose of triple in- eligible if they had previously received antileukemic treat- trathecal therapy (TIT) with methotrexate, cytarabine, and ment, or had uncontrolled or severe cardiovascular (history hydrocortisone given while ALL was being fully character- of congestive heart failure [New York Heart Association ized. Subsequently, the patients received induction therapy. class III or IV] or left ventricular ejection < 40%), he- Patients with morphologic CR and lower MRD levels (de- patic (total serum bilirubin > 1.5 × the upper normal limit fined as < 0.25% in this trial) received consolidation therapy [ULN], not attributable to ALL) or renal disease (serum followed by reinduction. Maintenance therapy with monthly creatinine > 1.5 × ULN and estimated creatinine clear- reinforcement cycles (M-1) was administered up to 1 year ance < 40 mL/minute [Cockcroft-Gault formula] not at- after ALL diagnosis. Then a second phase of maintenance tributable to ALL) or had a severe psychiatric condition therapy without reinforcement (M-2) with mercaptopurine (mental illness that does not allow to follow the protocol plus methotrexate was administered up to 2 years to patients according to the physicians criteria). Patients with mature in continuous CR. CNS prophylaxis included the admin- B ALL, Ph-positive ALL, T or B-cell lymphoblastic lym- istration of 14 courses of TIT over the 2 years of therapy. phoma or those with mixed phenotype acute leukemia were Cumulative doses of the main cytotoxic drugs were as fol- not included in the trial. Patients were centrally registered lows: 24 mg vincristine, 5,150 mg/m2 prednisone, 175 mg/ after informed consent was obtained from adult patients or m2 dexamethasone, 320,000 IU/m2 native E coli asparagi- from parents or guardians for patients under 18 years. The nase (pegylated formulation of asparaginase was not licensed study was approved by the Institutional Review Boards of in Spain at the time of this protocol), 240 mg/m2 daunoru- all the participating centers. bicin, 2,200 mg/m2 cyclophosphamide, and 9,000 mg/m2, 24-h continuous IV infusion methotrexate plus 1,560 mg/ m2 IM methotrexate during maintenance. Hospitalization, 2.2 | Diagnostic procedure prophylaxis and management of infections, transfusion, and other supportive care therapies were carried out according ALL was defined based on bone marrow (BM) infil- to institutional protocols. Therapeutic dose monitoring of tration of > 20% of blasts with lymphoid morphology. asparaginase was not performed. Immunophenotyping was performed by flow cytometry with Patients with poor early cytological response (defined as monoclonal antibodies against with B-cell, T-cell, myeloid, BM blast cells ≥ 10% on day 14 of induction), or MRD and precursor cell-associated antigens, and B or T lineage level ≥ 0.05% at the end of reinduction or after the 4th commitment was defined based on the WHO 2008 crite- monthly reinforcement cycle were considered as high-risk ria. Chromosomal analyses (using direct methods and un- (HR) patients and entered in the ALL-HR PETHEMA trial stimulated short-term cultures with G-banding) of BM or active at that time (ALLHR0313 [NCT 00853008] until peripheral blood (PB) samples performed at diagnosis were 2010 and ALLHR1114 [NCT01540812] from 2011 to 2019) centrally reviewed. A minimum of 20 metaphase cells were (Table S1). 2320 |    RIBERA et al. TABLE 1 PETHEMA ALLRE08 protocol. Chemotherapy schedule Phase Drugs Week Route Dose Days a Prephase Prednisone 1 IV 60 mg/m2 1-7 Triple IT therapy 1 IT 1 MTX 1 IT 15mg 1 ARA-C 1 IT 30 mg 1 Hydrocortisone 1 IT 20 mg 1 Remission induction Vincristine 1-4 IV 2 mg (absolute) 1, 8, 15, 22 Prednisone 1-5 IV/PO 60 mg/m2 1-27 IV/PO 30 mg/m2 28-35 Asparaginase (E coli, native) 2-4 IV 10,000 IU/m2 1-12, 17-19, 24-26 5 IV 1,000 mg/m2 35 Cyclophosphamide 1, 5 IT 1, 29 Triple IT Therapy Consolidation Mercaptopurine 1 PO 50 mg/m2 1-7 Methotrexate 1, 4, 8 IV (24h) 3 g/m2 b 1, 28, 56 Teniposide 150 mg/m2/12h 14, 42 ARA-C 2,6 IV 14, 42 Triple IT therapy 2,6 IV 500 mg/m2/12h 1, 28, 56 1, 4, 8 IT Reinduction Dexamethasone 1, 2 IV/PO 10 mg/m2 1-14 Vincristine 1-3 IV/PO 5 mg/m2 15-21 IV 2 mg (absolute) 1, 8, 15 Daunorubicin 1, 2 IV 30 mg/m2 1-2, 8-9 Asparaginase (E coli, native) 1, 3 IM/IV 10,000 IU/m2 1-3, 15-17 Triple IT Therapy 1, 3 IT 1, 15 Maintenance with reinforcement Maintenance IM 20 mg/m2/week Until week 52 Methotrexate PO Until week 52 Mercaptopurine 50 mg/m2/d Reinductions 1 IV 1, until week 52 Vincristine 1 IV/PO 2 mg (absolute) 1-7, until week 52 Prednisone 1 IV 60 mg/m2 1, until week 52 Asparaginase (E coli, native) 20,000 IU/m2 Maintenance without reinforcement Methotrexate IM 20 mg/m2/week Until week 105 Mercaptopurine PO 50 mg/m2/d Until week 105 Abbreviations: ARA-C, cytarabine; IM, intramuscular; IT, intrathecal; IV, intravenous; MTX, methotrexate; PO, oral. a Duration of less than 1 week if ALL has been well characterized. b With folinic acid rescue. 2.5 | Outcome assessment (OS). Secondary objectives included assessment of toxicity and comparison with ALL96 trial outcomes (updated for this The primary endpoints of this study were: i) CR rate, ii) cu- study). Planned enrollment was of 90 patients to allow com- mulative incidence of relapse (CIR), and iii) overall survival parison with the ALL96 trial. RIBERA et al.    | 2321 CR was defined as the absence of clinical manifes- The absence or insufficient number of metaphases was reg- tations of ALL, neutrophil count > 1.5×109/L, plate- istered in 28 (32%) cases. No significant differences were let count > 150×109/L, and hemoglobin level > 100 g/L, observed in karyotypic results on comparison between ad- with < 5% of blast cells in BM. Patients with ≥ 5% BM blast olescents and YA (Table 2). cells at the end of the induction phase were considered as induction failures. Two patterns of early response were con- sidered in BM aspirate performed on day 14 of treatment: 1) 3.2 | Treatment outcome slow, defined as the presence of ≥ 10% of blast cells, and 2) standard, defined as < 10% BM blast cells or hypoplastic BM. There were no deaths in induction. On day 14 of induction Overall survival was defined as the time from study entry to therapy 68 patients (76%) showed standard BM response death or last follow-up. Event-free survival was defined as the while 21 patients (24%) showed slow response, and 20 of time from diagnosis to failure, relapse or death by any cause or these latter patients were moved to a HR protocol; the re- last follow-up. The Common Terminology Criteria for Adverse maining patient withdrew the study. All patients with stand- Events (CTCAE v 3.0) was used for analysis of toxicity. ard early response achieved CR, and 16 of the remaining 20 patients with slow response achieved CR with the induction therapy of the HR protocol (three patients were resistant 2.6 | Statistical analysis and one was not evaluable due to withdrawal from the pro- tocol during induction), for a total of 84 CR patients (95%) Patients’ follow-up was updated on September 2018. The (Figure 1). Two CR patients with standard early cytologic median test was used to compare quantitative variables, while response showed high MRD levels (0.8% and 0.75%, re- the Chi-square or Fisher's exact tests were employed to assess spectively) at the end of induction and were then moved to differences in proportions. All comparisons were two-tailed. a HR-ALL protocol. Thus, a total of 66 patients continued Curves for OS and event-free survival (EFS) were plotted ac- with the ALL RE08 trial, and 22 (10 adolescents and 12 YA) cording to the Kaplan-Meier method and were compared by were transferred to a HR-ALL trial (7 to ALLHR03 and 15 to the log-rank test. The CIR was calculated using cumulative ALLHR11) (Figure 1). incidence functions by competing risks analysis, nonrelapse At the time of the analysis 9/66 patients who followed mortality (NRM) being the competing event. Data collection the ALLRE08 trial were receiving consolidation or main- and statistical analyses were performed at the PETHEMA tenance therapy at the time of the analysis, 8 patients re- Data Center for ALL using SPSS v.24 and R v.3.5.2 software. lapsed during consolidation (n = 3) or maintenance (n = 5) therapy and 11 patients relapsed off therapy. Seven patients abandoned the protocol due to excess toxicity (n = 4), their 3 | R E S U LTS own decision (n = 2) or a major deviation from the pro- tocol (n = 1). No therapy-related deaths were registered. 3.1 | Patients features at diagnosis Regarding those 18 patients in CR after inclusion in the HR protocol, 13 received chemotherapy and 5 allogeneic One-hundred and two patients were enrolled in 40 Spanish HSCT. Six patients relapsed (5 during chemotherapy and hospitals, 89 of whom were eligible for the study. The re- 1 after HSCT), 1 died by toxicity of chemotherapy, and maining 13 patients were excluded because age < 15 years 11 are in CR (7 after chemotherapy and 4 after HSCT). (n = 1), ALL with high-risk features at baseline (n = 9), Fourteen patients remain alive (11 in first CR and 3 in sec- lymphoblastic lymphoma (n = 2), and Philadelphia chromo- ond CR). some-positive ALL (n = 1). Overall, 25 patients relapsed in either BM (n = 19) or Table 2 shows the main clinical and biological fea- BM plus CNS (n = 2), BM and testicle (n = 1) BM and tures of the whole series including adolescents (n = 38) lymph nodes (n = 1), CNS (n = 2) and testicular (n = 1). vs YA (n = 51). Median age at diagnosis was 20 years Ten patients were alive in second CR, one was alive in (range 15-30 years) and 54 patients (61%) were males. relapse, seven dead by treatment-related mortality (either Extramedullary disease was present at diagnosis in four by alloHSCT [n = 2] or by rescue chemotherapy [n = 5]) cases (5%). Eighty-four patients (94%) had B-cell precur- and seven dead by disease progression. The CIR at 5 years sor ALL. Twenty-three patients (26%) showed a normal for the whole series was 35% (95%CI: 23%-47%) (Figure karyotype. Abnormal karyotypes included high hyper- 2A). With a median follow-up of 4.19 years (range 0.04- diploidy (n = 8, 9%), t(1;19) (n = 4, 5%), t(12;21)(ETV6/ 9.47), 18 patients have died, with the OS probability at RUNX1) (n = 1, 1%) and del(9p) (n = 3, 3%), among oth- 5 years being 74% (95%CI: 63%-85%) for the whole series ers. Four cases (5%) had a complex karyotype, an abnor- (Figure 2B). The main causes of death were disease pro- mality not considered as HR at the time of protocol design. gression (n = 15), transplant-related (n = 2) and toxicity 2322 |    RIBERA et al. TABLE 2 Patient characteristics in the Whole series Adolescents Young adults whole series and according to the age group Characteristic (n = 89) (n = 38) (n = 51) P value Age at diagnosis, years, median 20 (15-29) 17 (15-18) 23 (19-29) <0.001 (range) Gender 0.197 Male 54 (61%) 26 (68%) 28 (55%) Female 35 (39%) 12 (32%) 23 (45%) Performance status (ECOG 0.746 scale) 0 26 (30%) 13 (34%) 13/48 (27%) 1 52 (61%) 22 (58%) 30/48 (63%) 2 7 (8%) 3 (8%) 4/48 (8%) 3 1 (1%) 0 1/48 (2%) Extramedullary disease NA CNS 3 (3%) 0 3 (3%) CNS + other 1 (1%) 0 1 (1%) WBC count, x109/L, median 6.83 (0.4-30) 6.82 7.19 (0.8-30) 0.736 (range) (0.4-24.8) Phenotype 0.884 Early pre-B 0 0 0 Common 68 (76%) 30 (79%) 38 (74%) Pre-B 16 (18%) 6 (16%) 10 (20%) T 5 (6%) 2 (5%) 3 (6%) Karyotype 0.452 No growth 28 (32%) 9 (24%) 19 (37%) Normal 23 (26%) 10 (26%) 13 (25%) Hyperdiploidy > 50 chr 8 (9%) 5 (13%) 3 (6%) t(12;21)/TEL-AML1 1 (1%) 1 (3%) 0 t(1;19) 4 (5%) 2 (5%) 2 (4%) Other abnormalities 25 11 (29%) 14 (28%) Abbreviations: chr, chromosomes; CNS, central nervous system; ECOG, Eastern Cooperative Oncology Group; NA, not available; WBC, white blood cell. of HR chemotherapy (n = 1). Fifty-one patients remained 3.3 | MRD status alive and off therapy (41 in first CR and 10 in second CR). Figure 2C shows the EFS for the whole series (probability MRD at the end of the induction was evaluable in 61/68 pa- at 5 years 62% [95%CI: 49%-73%]). Table 3 shows the main tients with standard response. Of these, 53 showed MRD outcomes for both the whole series and for those patients level < 0.1%, and 8 had MRD levels ≥ 0.1% (≥0.5% in two, who remained in the ALLRE08 protocol, those who were who were moved to a HR protocol). Forty-eight patients moved to a HR protocol, and for adolescents and YA con- showed end-induction MRD levels < 0.05%. MRD levels sidered separately. As shown, significant differences in OS postconsolidation were evaluable in 54/66 patients who began and CIR were observed in adolescents vs YA, with 5 years. consolidation. The reasons for lack of MRD data in the re- OS rates of 87% (95%CI: 74%-100%) vs 63% (46%-80%), maining 12 patients were: early relapse (n = 3), withdrawal respectively, P = 0.021 and CIR rates of 13% (4%-28%) vs from protocol (n = 1), protocol deviation (n = 1), still on 52% (34%-67%), P = 0.012, respectively (Figures 3A,B). therapy (n = 1), and no test performed (n = 6). No patient Interestingly, no significant differences were observed in was moved to a HR protocol at this time point, because the the CIR in patients who remained in the ALLRE08 proto- MRD level remained < 0.05% in all evaluable patients. From col vs those who were moved to a HR protocol (5-year CIR those 60 patients who began the maintenance therapy 47 had rate of 34% [21%-47%] vs 37% [14%-61%], respectively, MRD assessed, while 13 did not because of relapse (n = 1), P = 0.588) (Figure 4). withdrawal from protocol (n = 5), still on therapy (n = 1) and RIBERA et al.    | 2323 FIGURE 1 Flow chart of the patients in the ALLRE08 PETHEMA trial MRD testing not performed (n = 6). Again, no patient was The remaining 24 patients had no MRD data because they had transferred to a HR protocol due to MRD levels < 0.05%. relapsed (n = 3), have just finished M-2 therapy (n = 6) or MRD testing at end-of-therapy was performed in 27 patients. had missing data (n = 15). In 6/11 patients who relapsed off 2324 |    RIBERA et al. 1.0 A (13%) evaluable patients during C1, in 9/55 (16%) evalu- able patients during C2, and in 34/52 (65%) patients during maintenance. No differences were found in the frequency of Cumulative incidence of relapse 0.8 delays or dose modifications of the chemotherapy between adolescents and YA. Of note, OS and CIR were not signifi- 0.6 cantly different in patients who showed delays or dose mod- ifications vs those who received the schedule timely and at full dose (5-year OS: 76% [62%-90%] vs 84% [64%-100%], 0.4 P = 0.326; 5-year CIR: 32% [14%-46%] vs 27% [0%-47%], P = 0.920). Main grade III-IV toxicities observed in the trial 0.2 (Table 4) were: i) hematologic, particularly in induction, C1 and C2. Despite this relatively high frequency, the number of severe infections was low, ii) hepatic, and iii) hypersen- 0.0 sitivity to native E coli asparaginase. As expected, the lat- 0 2 4 6 8 ter was more frequent after its reexposure in C2 cycle or in Years after first CR M1. Erwinia asparaginase was given to 9 of 13 patients with Number at risk 84 56 34 17 3 grade III-IV allergic reactions, and removal of asparaginase was done in the remaining four cases. B 1.0 3.5 | Patients outcome in the ALLRE08 vs the ALL-96 PETHEMA trials 0.8 Overall Survival (Probability) Table S1 shows the comparison of the main outcomes be- 0.6 tween these protocols. As expected, the outcomes were not significantly different, although a significantly lower OS was observed in YA vs adolescents in the current study and not in 0.4 the ALL96 trial. 0.2 4 | DISCUSSION 0.0 This study shows that treatment with a full pediatric protocol 0 2 4 6 8 is feasible and effective for AYA with Ph-neg, SR-ALL, with Years after diagnosis better results for adolescents than for YA. Of note, one third Number at risk 89 66 40 20 3 of the patients showed poor early response, but their outcome was not significantly inferior than that observed for the good F I G U R E 2 Cumulative incidence of relapse (panel A), and responders after being rescued with a HR therapy. overall survival (panel B) in the whole series Adolescent and YA patients with ALL represent a popu- lation with specific characteristics and needs.1-3 At present it therapy MRD levels at the end of maintenance were < 0.05% is well established that adolescents aged 15 to 20 years are while were missing in the remaining five cases. best treated with full pediatric protocols; in parallel, growing evidence suggests that this might also be true for YA (with an upper age limit of 30 to 40 years in most studies or even 3.4 | Dose modifications and up to 50-55 years in some studies),15-18 with 5-year survival treatment toxicity rates around 70%, despite HSCT is less used in pediatric trials. Since MRD has emerged as a powerful prognostic The dose of chemotherapy was delayed in 1/68 (1.5%) eval- marker useful for risk stratification in ALL,19 the PETHEMA uable patients during induction, in 22/62 (35%) evaluable ALLRE08 trial added MRD assessment to the early cytologic patients in C1, 10/55 (18%) evaluable patients in C2 and response to select among AYA patients with SR-ALL those in 13/52 (25%) evaluable patients during maintenance. In who were poor responders and transfer them to a HR proto- addition, there were dose modifications of cytotoxic drugs col, while maintaining the full pediatric chemotherapy sched- in 2/68 (3%) evaluable patients during induction, in 8/62 ule for good responders. RIBERA et al.    | 2325 TABLE 3 Main outcomes of the OS (5-yrs, EFS (5-yrs, CIR (5-yrs, patients from the whole series, for those Patient group N 95% CI) 95% CI) 95% CI) who remained in the ALLRE08 protocol and those who were moved to high-risk Whole series 89 74% 62% 35% protocol, and for adolescents and young (63%;85%) (49%;73%) (23%;47%) adults separately Patients that remained in the 66 78%a 67%e 34%ab ALLRE08 trial (66%;90%) (51%;78%) (21%;47%) Patients moved to HR trials 22 61%a 48%e 37%b (31%;91%) (25%;68%) (14%;61%) Adolescents 38 87%c 78%f 13%d (74%;100%) (59%,;89%) (4%;28%) Young adults 51 63%c 49%f 52%d (46%;80%) (31%;65%) (34%;67%) Abbreviations: CIR, cumulative incidence of relapse; EFS, event-free survival; HR, high risk; OS, overall survival. a p = 0.343 b p = 0.588 c p = 0.021 d p = 0.012 e p = 0.028 f p = 0.151 Overall, the CR rate observed was high and similar to that presenting chemoresistance. As expected for HR ALL, one reported for most full pediatric or pediatric-inspired proto- third of the patients relapsed.7 It could be speculated that cols focused on AYA with ALL.11,16-21 and those without the proportion of relapses would have been lower if all pa- a pediatric basis.21 Of note, one third of patients were poor tients had undergone allogeneic HSCT, as the frequency of responders and were transferred per protocol to a HR trial, relapses was lower in transplanted than in nontransplanted which allowed CR to be attained in 80% of patients. No toxic patients (1/5 for allogeneic HSCT vs 5/13 for chemotherapy), toxicity-related deaths occurred among AYA who followed but the low number of cases in this series does not allow to the ALLRE08 trial and the overall adherence to the protocol draw definitive conclusions. Despite the OS for these patients being good. Despite this a few patients abandoned the trial was lower (and the CIR slightly higher) than that observed due to excess toxicity, their own decision or a major devia- for patients showing good early response, the differences did tion from the predefined therapeutic schedule. In contrast, a not reach statistical significance, indicating that a substantial significant rate of relapses off therapy (11/19) occurred de- proportion of these patients can be successfully rescued if spite the MRD levels were low in all tested cases. Because no they are transferred early to a HR protocols. However, the regular MRD assessment was performed off therapy, early re- absence of statistically significant differences in OS between emergence of MRD could not be identified and consequently patients on trial on those who were rescued in HR trials does early therapeutic intervention at an MRD positive CR status not mean that the rescue therapy was really very effective was not feasible in these patients. Whether or not prolonged (40% of relapses). It is clear that other approaches are nec- maintenance therapy for 2.5 or 3 years from CR, as done in essary to improve the prognosis of this subset of patients, as some pediatric protocols, would result in a lower frequency of discussed later in this section. late relapses among these patients remains to be investigated. Although the OS rates observed in the ALLRE08 and Among all treated patients those with poor early response the ALL96 trials were equivalent, some differences should were more difficult to treat, showing a higher rate of resistant be pointed out. Thus, while no differences existed in the leukemia and deaths by toxicity. It might well be that a more inclusion criteria and the baseline characteristics of the pa- extensive work-up of ALL at diagnosis would allow identifi- tients enrolled in both trials, significantly lower OS rates cation of patients with poor genetic risk (eg, BCR-ABL1-like were observed in the here reported trial for YA vs adoles- ALL) among this group of patients.22 However, still 80% of cents, not being observed in the ALL96 trial, where the cases with slow response treated with a HR trial achieved two groups of patients showed similar outcome.12 This CR. For these patients allogeneic HSCT was performed ac- might be due to a better OS of adolescents in the ALLRE08 cording to physicians’ criteria, resulting in only 5/18 CR pa- trial vs the ALL96 study (87% [74%-100%] vs 80% [67%- tients (including the two patients who had an excess of MRD 93%]) and a lower OS for YA (as a consequence of a higher at the end of the first induction) being transplanted, a propor- CIR) in this vs the previous trial (63% [46%-80%] vs 71% tion that is clearly lower than that expected for ALL patients [58%-84%]) emphasizing in the ALLRE08 trial the overall 2326 |    RIBERA et al. A 1.0 1.0 SR YA HR Adolescent Cumulative incidence of relapse 0.8 0.8 Overall Survival (Probability) 0.6 0.6 0.4 0.4 0.2 0.2 P = .021 P = .588 0.0 0.0 0 2 4 6 8 0 2 4 6 8 Years after diagnosis Years after first CR Number at risk Number at risk YA 51 35 17 8 1 SR 66 46 29 15 2 Adolescent 38 31 23 12 2 HR 18 10 5 2 1 B F I G U R E 4 Cumulative incidence of relapse of patients who 1.0 remained in the ALLRE08 protocol vs those who were moved to a YA Adolescent high-risk (HR) protocol Cumulative incidence of relapse 0.8 especially during maintenance, there were no differ- 0.6 ences between adolescents and YA, and these modifi- cations did not show impact on patients’ outcome. Dose 0.4 modifications and delays are important in the pediatric setting, but their relevance in AYA and adults has been less studied. 0.2 Despite the long follow-up of the patients here reported, some limitations should be pointed out for this trial. First, 0.0 P = .012 the definition of risk factors did not take into account im- 0 2 4 6 8 portant features such as deep molecular characterization of Years after first CR the patients at baseline aimed to detect specific gene dele- Number at risk tions/mutations, as well as other important characteristics YA 49 29 14 7 1 with recognized prognostic impact such as the BCR-ABL Adolescent 35 27 20 10 2 like signature or the early pre-T signature, that are frequent F I G U R E 3 Comparison between the overall survival (panel A), among AYA,22-24 although the latter subgroups were identi- and the cumulative incidence of relapse (panel B) between adolescents fied when the trial was in an advanced phase of recruitment. and young adults It is clear that at present the risk definition should include the aforementioned studies. Second, combined cytological and MRD criteria were used to identify poor early respond- difference in OS observed between adolescents and YA, ers, but only two patients were transferred to a HR protocol that was already present at a nonsignificant level in the because of poor MRD clearance. Despite the early cytolog- ALL96 trial.13 ical response has been extensively used in the past for early Overall, the tolerability of the ALLRE08 trial was identification of this subset of patients resistant to induction good, probably due to the relatively conservative upper therapy, and has prognostic significance,25 most modern age limit for inclusion, set at 30 years. Main toxic trials use end-induction MRD as the only tool to assess the events were due to cytopenias and infections, followed quality of response, given that there is not a precise correla- by liver toxicity and hypersensitivity to E coli aspar- tion between poor early cytological response and the pattern aginase. Although the frequency of delays and dose of MRD clearance.11 In turn, the end-induction MRD cut-off modification of the chemotherapeutic drugs were high, (0.25%) selected for transferring the patients to a HR trial RIBERA et al.    | 2327 TABLE 4 Main grade III-IV toxicities observed per treatment phases of the ALLRE08 trial Maintenance with Induction Consolidation Reinduction reinforcements Neutropenia 55/67 (82%) 34/58 (59%) 26/56 (46%) 5/52 (10%) Days, median [min;max] 15 [1; 41] 6 [1; 30] 3,50 [1 ;13] Thrombocytopenia 26/65 (40%) 8/58 (14%) 1/55 (2%) 0 Days, median [min;max] 8 [1; 35] 3 [1; 13] 3 Coagulation disorders 9/67 (13%) 1/59 (2%) 1/56 (2%) 0 Infection 21/65 (32%) 14/59 (24%) 5/56 (9%) 1/52 (2%) Hypersensitivity 0 1/59 (2%) 8/56 (14%) 4/52 (8%) Neurologic 2/68 (3%) 0 2/56 (4%) 1/52 (2%) Hepatic 13/68 (19%) 8/58 (14%) 5/56 (9%) 6/52(12%) Renal 0 2/59 (3%) 0 0 Gastrointestinal 0 0 1/56 (2%) 0 Vascular 2/67 (3%) 0 1/56 (2%) 0 Pulmonary 0 0 1/56 (2%) 1/52 (2%) Metabolic 2/67 (3%) 0 0 0 can be considered too high, since most modern trials con- ACKNOWLEDGMENTS sider lower MRD cut-offs (>0.1% or even > 0.01%) to define We thanks to the following institutions from Spain their sup- poor MRD response after induction. Third, MRD was not port for this paper: ICO-Hospital Germans Trias i Pujol. centrally assessed, and consequently introduced a potential Josep Carreras Research Institute. Universitat Autònoma de bias in the study, potential impact on the reproducibility of Barcelona. Hospital Universitari i Politècnic La Fe, Valencia. the results. In addition, the MRD was assessed by 4-color Hospital Clínico Valencia. Hospital Central de Asturias, Oviedo. FCM, a technique known now to provide insufficient sensi- Hospital Universitario Virgen del Rocío, Sevilla. Hospital tivity. Finally, the study was restricted to AYA with SR fea- General Universitario de Alicante. Hospital Universitari Vall tures, which might limit direct comparison with other trials d'Hebron, Universitat Autònoma de Barcelona. Hospital that include AYA with Ph-negative ALL, only selected by General Universitario de Castellón. ICO-Hospital Josep Trueta, age. Girona. Hospital Dr Peset, Valencia. ICO-Hospital Duran i At present, there is room for improvement of patients Reynals, L'Hospitalet de Llobregat. Hospital Son Espases, outcome after pediatric type therapy in AYA with Ph-neg Palma de Mallorca. Hospital Clínico Universitario, Cancer ALL, and according to our results, this improvement is Research Center (IBMCC-CSIC/USAL), Cytometry Service, especially necessary for YA.26 The more extensive use Salamanca. Hospital San Pedro de Alcántara, Cáceres. Hospital of conventional drugs such as PEG-asparaginase11,17 to- Morales Meseguer, Murcia. Hospital Mútua de Terrassa, gether with the early incorporation of immunotherapeutic Terrassa. Hospital Doce de Octubre, Madrid. Hospital Ramón approaches with naked (anti CD20),27 immunoconjugated y Cajal, Madrid. Hospital de Sant Pau, Barcelona. Hospital (inotuzumab ozogamicin),28 or bispecific (blinatum- Arnau de Vilanova, Lleida. Hospital Lucus Augusti, Lugo. omab)29,30 monoclonal antibodies which have proven ef- ICO-Hospital Joan XXIII, Tarragona. fective in relapsed/refractory or MRD positive patients might contribute to improve the quality of remission and CONFLICT OF INTEREST decreased the relapse rate observed in our study. In addi- The authors made no disclosures. tion, early use of genetically engineered T cells that rec- ognize ALL epitopes (CAR T cells) will also provide an AUTHOR CONTRIBUTIONS opportunity to improve the outcomes of these patients.31-33 JMR was primarily responsible for the design of the trial and Finally, incorporation of targeted kinase inhibitors to wrote the paper. MM analyzed the data. The remaining authors pediatric type chemotherapeutic regimens for specific (PM, MT, DMC, JGC, CG, PB, RGB, RC, MP, JR, SM, SV, subtypes of ALL such as BCR-ABL like might further con- AN, EG, JMHR, JB, MLA, FV, PMS, MC, IGC, AGG, MJSS, tribute to overcome the resistance to therapy and improve MC, EF, and AO) qualified for authorship according to the survival of this particularly difficult to treat subgroup of WAME criteria, reported the patients and followed them clini- ALL patients. cally. These contributions explain the order of authors. 2328 |    RIBERA et al. DATA AVAILABILIT Y STATEMENT Tratamiento en Hematologia pediatric-based protocol ALL-96. J The data that support the findings of this study are available Clin Oncol. 2008;26:1843-1849. 13. Ribera J-M, Oriol A, Morgades M, et al. Treatment of high-risk from the corresponding author upon reasonable request. Philadelphia chromosome-negative acute lymphoblastic leukemia in adolescents and adults according to early cytologic response and ORCID minimal residual disease after consolidation assessed by flow cy- Josep-Maria Ribera https://orcid. tometry: final results of the PETHEMA ALL-AR-03 trial. J Clin org/0000-0003-1042-6024 Oncol. 2014;32(15):1595-1604. Pau Montesinos https://orcid.org/0000-0002-3275-5593 14. Ribera JM, Morgades M, Ciudad J, Barba P, Martino R, Montesinos Santiago Mercadal https://orcid. P, et al. Post-remission treatment with chemotherapy or allogeneic hematopoietic stem cell transplantation (alloHSCT) of High-risk org/0000-0003-4741-7885 (HR) Philadelphia chromosome-negative (Ph-neg) Adult Acute Alberto Orfao https://orcid.org/0000-0002-0007-7230 lymphoblastic leukemia (ALL) according to minimal residual dis- ease (MRD). Preliminary results of the pethema ALL-HR-11 trial. R E F E R E NC E S Blood. 2015;126:1333 [abstract]. 1. Curran E, Stock W. How I treat acute lymphoblastic leukemia in older 15. Rijneveld AW, van der Holt B, Daenen SMGJ, et al. Intensified adolescents and young adults. Blood. 2015;125(24):3702-3710. chemotherapy inspired by a pediatric regimen combined with al- 2. Boissel N, Baruchel A. Acute lymphoblastic leukemia in ad- logeneic transplantation in adult patients with acute lymphoblastic olescent and young adults: treat as adults or as children? Blood. leukemia up to the age of 40. Leukemia. 2011;25(11):1697-1703. 2018;132(4):351-361. 16. Toft N, Birgens H, Abrahamsson J, et al. Results of NOPHO 3. Siegel SE, Stock W, Johnson RH, et al. Pediatric-inspired treatment ALL2008 treatment for patients aged 1–45 years with acute lym- regimens for adolescents and young adults with philadelphia chro- phoblastic leukemia. Leukemia. 2018;32(3):606-615. mosome-negative acute lymphoblastic leukemia: a review. JAMA 17. DeAngelo DJ, Stevenson KE, Dahlberg SE, et al. Long-term Oncol. 2018;4(5):725-734. outcome of a pediatric-inspired regimen used for adults aged 4. Björkholm M, Edgren G, Dickman P. Trends in survival of young 18–50 years with newly diagnosed acute lymphoblastic leukemia. adult patients with acute lymphoblastic leukemia in Sweden and Leukemia. 2015;29(3):526-534. USA. Blood. 2019. https​://doi.org/10.1182/blood.20190​01168​ 18. Huguet F, Chevret S, Leguay T, et al. Group of research on adult 5. Stock W, La M, Sanford B, et al. What determines the outcomes for ALL (GRAALL). Intensified therapy of acute lymphoblastic leu- adolescents and young adults with acute lymphoblastic leukemia kemia in adults: report of the randomized GRAALL-2005 clinical treated on cooperative group protocols? a comparison of children's trial. J Clin Oncol. 2018;36(24):2514-2523. cancer group and cancer and leukemia group B studies. Blood. 19. Berry DA, Zhou S, Higley H, et al. Association of minimal residual 2008;112(5):1646-1654. disease with clinical outcome in pediatric and adult acute lympho- 6. Muffly L, Li Q, Alvarez E, et al. Hematopoietic cell transplan- blastic leukemia: a metaanalysis. JAMA Oncol. 2017;3(7):e170580. tation in young adult acute lymphoblastic leukemia: a United 20. Ribera JM, Ribera J, Genescà E. Treatment of adolescent and States population-level analysis. J Adolesc Young Adult Oncol. young adults with acute lymphoblastic leukemia. Mediterr J 2019;8(3):254-261. Hematol Infect Dis. 2014;6(1):e2014052. 7. Seftel MD, Neuberg D, Zhang M-J, et al. Acute leukemia committee 21. Rytting ME, Jabbour EJ, Jorgensen JL, et al. Final results of a of the CIBMTR and the Dana Farber ALL Consortium. Pediatric- single institution experience with a pediatric-based regimen, the inspired therapy compared to allografting for Philadelphia chro- augmented Berlin-Frankfurt-Münster, in adolescents and young mosome-negative adult ALL in first complete remission. Am J adults with acute lymphoblastic leukemia, and comparison to the Hematol. 2016;91(3):322-329. hyper-CVAD regimen. Am J Hematol. 2016;91(8):819-823. 8. Smith AW, Keegan T, Hamilton A, Lynch C, Wu XC, Schwartz SM, 22. Roberts KG. Genetics and prognosis of ALL in children vs adults. et al, AYA HOPE Study Collaborative Group. Understanding care Hematology Am Soc Hematol Educ Program. 2018;2018(1):137-145. and outcomes in adolescents and young adult with Cancer: A review 23. Roberts KG, Gu Z, Payne-Turner D, et al. High frequency and poor of the AYA HOPE study. Pediatr Blood Cancer. 2019;66(1):e27486. outcome of Philadelphia chromosome-like acute lymphoblastic 9. Wolfson JA, Richman JS, Sun C-L, et al. Causes of inferior leukemia in adults. J Clin Oncol. 2017;35(4):394-401. outcome in adolescents and young adults with acute lympho- 24. Bond J, Graux C, Lhermitte L, et al. Early response-based therapy blastic leukemia: across oncology services and regardless of stratification improves survival in adult early thymic precursor acute clinical trial enrollment. Cancer Epidemiol Biomarkers Prev. lymphoblastic leukemia: a group for research on adult acute lym- 2018;27(10):1133-1141. phoblastic leukemia study. J Clin Oncol. 2017;35(23):2683-2691. 10. Parsons HM, Penn DC, Li Q, et al. Increased clinical trial enroll- 25. Short NJ, Kantarjian HM, Sasaki K, et al. Prognostic significance ment among adolescent and young adult cancer patients between of day 14 bone marrow evaluation in adults with Philadelphia 2006 and 2012–2013 in the United States. Pediatr Blood Cancer. chromosome-negative acute lymphoblastic leukemia. Cancer. 2019;66(1):e27426. 2016;122(24):3812-3820. 11. Stock W, Luger SM, Advani AS, et al. A pediatric regimen for 26. Friend BD, Schiller GJ. Closing the gap: novel therapies in treat- older adolescents and young adults with acute lymphoblastic leu- ing acute lymphoblastic leukemia in adolescents and young adults. kemia: results of CALGB 10403. Blood. 2019;133(14):1548-1559. Blood Rev. 2018;32(2):122-129. 12. Ribera JM, Oriol A, Sanz MA, et al. Comparison of the results of 27. Maury S, Chevret S, Thomas X, et al. Rituximab in the treatment of adolescents and young adults with standard-risk B-lineage adult acute lymphoblastic leukemia. N Engl J Med. acute lymphoblastic leukemia with the Programa Español de 2016;375(11):1044-1053. RIBERA et al.    | 2329 28. Kantarjian HM, Vandendries E, Advani AS. Inotuzumab ozo- SUPPORTING INFORMATION gamicin for acute lymphoblastic leukemia. N Engl J Med. Additional supporting information may be found online in 2016;375(21):2100-2101. the Supporting Information section. 29. Kantarjian H, Stein A, Gökbuget N, et al. Blinatumomab versus chemotherapy for advanced acute lymphoblastic leukemia. N Engl J Med. 2017;376(9):836-847. 30. Gökbuget N, Dombret H, Bonifacio M, et al. Blinatumomab for How to cite this article: Ribera J-M, Morgades M, minimal residual disease in adults with B-cell precursor acute lym- Montesinos P, et al; on behalf of the PETHEMA phoblastic leukemia. Blood. 2018;131(14):1522-1531. Group, Spanish Society of Hematology. A pediatric 31. Maude SL, Laetsch TW, Buechner J, et al. Tisagenlecleucel in chil- regimen for adolescents and young adults with dren and young adults with B-cell lymphoblastic leukemia. N Engl Philadelphia chromosome-negative acute J Med. 2018;378(5):439-448. lymphoblastic leukemia: Results of the ALLRE08 32. Park JH, Rivière I, Gonen M, et al. Long-term follow-up of CD19 PETHEMA trial. Cancer Med. 2020;9:2317–2329. CAR therapy in acute lymphoblastic leukemia. N Engl J Med. https​://doi.org/10.1002/cam4.2814 2018;378(5):449-459. 33. Hay KA, Gauthier J, Hirayama AV, et al. Factors associated with du- rable EFS in adult B-cell ALL patients achieving MRD-negative CR after CD19 CAR T-cell therapy. Blood. 2019;133(15):1652-1663.