Letter Open Access Impact of Center-related Characteristics and Macroeconomic Factors on the Outcome of Adult Patients With Acute Lymphoblastic Leukemia Treated With Pediatric-inspired Protocols Pere Barba1, Mireia Morgades2, Pau Montesinos3, Jose Gonzalez-Campos4, Anna Torrent2, Cristina Gil5, Teresa Bernal6, Mar Tormo7, Santiago Mercadal8, Sandra Novoa1, Irene García-Cadenas9, M. Paz Queipo de Llano10, Marta Cervera11, Rosa Coll12, Arancha Bermudez13, M. Luz Amigo14, Silvia Monsalvo15, Jordi Esteve16, Raimundo Garcia-Boyero17, Andres Novo18, Jesús Maria Hernandez Rivas19, Antonia Cladera20, Pilar Martinez-Sanchez21, Josefina Serrano22, Maria Teresa Artola23, Beatriz Soria24, Eugenia Abella25, Ferran Vall-Llovera26, Juan Bergua27, Pilar Herrera28, Daniel Barrios29, Josep Maria Ribera2, on behalf of the Spanish PETHEMA Group Correspondence: Pere Barba (pbarba@vhio.net). D iagnosis and treatment of hematological can- In most European countries, ALL treatment protocols are cers is usually provided in many healthcare facil- based on pediatric-inspired regimens which include a large ities including large but also middle size centers.1 number of immune-chemotherapeutic agents and several key Providing cancer care in local institutions might be decision points to allocate patients to distinctive treatment arms advantageous for patients and caregivers in terms of financial based on genetics and treatment response.3,4 Several patient and burden and quality of life. However, it might carry potential disease characteristics have been identified as prognostic factors risks derived of the limited experience of smaller centers and for outcomes including age, white blood cell count (WBC) at differences in accessibility to complex therapies including diagnosis, central nervous system (CNS) infiltration, clearance allogeneic hematopoietic cell transplantation (allo-HCT) and of measurable residual disease (MRD),4 and disease genetics.5–7 chimeric antigen receptor (CAR) T-cells. These risks might be The outcome of patients with ALL may also depend on external especially relevant in infrequent cancers as adult acute lym- factors including center experience, access to cellular therapies phoblastic leukemia (ALL). and economic variables.8–10 The impact of these center-related 1 Hematology Department, Hospital Universitari Vall d’Hebron, Barcelona, Spain 18 Hematology Department, Hospital Universitari Son Espases, Palma de Mallorca, 2 Hematology Department, Institut Català d’Oncologia-Hospital Germans Trias Spain i Pujol, Josep Carreras Research Institute, Badalona, Universitat Autònoma de 19 Hematology Department, IBSAL, IBMCC, Centro de Investigación del Cáncer, Barcelona, Spain CIBERONC, Universidad de Salamanca-CSIC, Hospital Universitario de 3 Hematology Department, Hospital Universitari i Politècnic La Fe, Valencia, Spain Salamanca, Spain 4 Hematology Department, Hospital Universitario Virgen del Rocío, Sevilla, Spain 20 Hematology Department, Hospital Universitari Son Llàtzer, Palma de Mallorca, 5 Hematology Department, Hospital General Universitario de Alicante, Spain Spain 6 Hematology Department, Hospital Universitario Central de Asturias, Oviedo, Spain 21 Hematology Department, Hospital Universitario 12 de Octubre, Madrid, Spain 7 Hematology Department, Hospital Clínico Universitario de Valencia, Spain 22 Hospital Universitario Reina Sofía, IMIBIC, Córdoba, Spain 8 Hematology Department, Institut Català d’Oncologia-Hospital Duran i Reynals, 23 Hematology Department, Hospital Universitario de Donostia, Spain L’Hospitalet de Llobregat, Spain 24 Hematology Department, Hospital Universitario de Canarias, Santa Cruz de 9 Hematology Department, Hospital de la Santa Creu i Sant Pau, Barcelona, Spain Tenerife, Spain 10 Hematology Department, Hospital Universitario Virgen de la Victoria, Málaga, Spain 25 Hematology Department, Hospital del Mar, Barcelona, Spain 11 Hematology Department, Institut Català d’Oncologia-Hospital Joan XXIII, 26 Hematology Department, Hospital Universitari Mútua de Terrassa, Spain Tarragona, Spain 27 Hematology Department, Hospital San Pedro de Alcántara, Cáceres, Spain 12 Hematology Department, Institut Català d’Oncologia-Hospital Josep Trueta, 28 Hematology Department, Hospital Universitario Ramón y Cajal, Madrid, Spain Girona, Spain 29 Hematology Department, Hospital Regional Universitario de Málaga, 13 Hematology Department, Hospital Universitario Marqués de Valdecilla, Spain Santander, Spain Supplemental digital content is available for this article. 14 Hematology Department, Hospital General Universitario Morales Meseguer, Copyright © 2022 the Author(s). Published by Wolters Kluwer Health, Inc. on Murcia, Spain behalf of the European Hematology Association. This is an open access article 15 Hematology Department, Hospital General Universitario Gregorio Marañón, distributed under the Creative Commons Attribution License 4.0 (CCBY), which Madrid, Spain permits unrestricted use, distribution, and reproduction in any medium, provided 16 Hematology Department, Hospital Clínic de Barcelona, IDIBAPS, Universitat de the original work is properly cited. Barcelona, Spain HemaSphere (2023) 7:1(e810). 17 Hematology Department, Hospital General Universitario de Castellón, Castellón http://dx.doi.org/10.1097/HS9.0000000000000810. de la Plana, Spain Received: July 20, 2022 / Accepted: November 3, 2022 1 Barba et al Center and Macroeconomic Variables in ALL and macroeconomic variables on the outcome of patients has Table 1 been scarcely studied. Thus, the aim of this study was to ana- Patients, Disease, Center, and Macroeconomic Characteristics lyze the potential impact of center-related and macroeconomic variables on the outcome of newly diagnosed adult ALL patients Characteristic N = 816 included in 4 consecutive trials of the Spanish Program for Age, y, median (range) 35.53 (15; 60) Treatment of Hematological Malignancies (PETHEMA) Group. Gender, n (%) Male 475 (58) Patients with Philadelphia chromosome (Ph) positive or neg- Female 341 (42) ative ALL enrolled in one of the 4 consecutive protocols of the ECOG PS, n (%) 0–1 648/772 (84) PETHEMA group by Spanish institutions adhered to the public 2–3 124/772 (16) health system from 2003 to 2018 were included in this study. WBC, ×109/L, median (range) 16 (0; 842) The 4 protocols have been closed and reported elsewhere.4,11–13 CNS infiltration, n (%) No 707/769 (92) Centralized analysis of MRD was performed in 5 centers in Yes 62/769 (8) the ALL-AR-03 and in a single institution in the ALL-HR-11 Ph+ ALL, n (%) No 688 (84) trial. The PH-08 protocol included adults with newly diagnosed Yes 128 (16) Ph-positive ALL up to the age of 60 years. Patients received an Precursor phenotype, n (%) B 603/803 (75) induction therapy with daunorubicin, vincristine, and steroids T 200/803 (25) in combination with imatinib 600 mg/d followed by a 12-week Treatment period, n (%) 2003–2010 341 (42) consolidation chemotherapy based on alternated cycles of high- 2011–2018 475 (58) dose methotrexate and cytarabine in combination with imatinib. Protocol, n (%) ALL-AR-03 323 (40) Allo-HCT was offered to all fit patients with a suitable donor. ALL-RE-08 86 (10) Treatment protocols used in the ALL-RE-2008, ALL-AR-03, PH-08 128 (16) and ALL-HR-11 trials were pediatric-inspired.4,11–13 The ALL-HR-11 279 (34) ALL-RE-2008 trial included intermediate risk patients based on Allo-HCT center, n (%) No 292 (36) age (<30 years), WBC count (<25,000 cells/μL), and cytogenet- Yes 524 (64) ics. The ALL-AR-03 and ALL-HR-11 protocols included high- Allo-HCT center in the same province, n (%) No 168/292 (57) risk patients up to the age of 60 years diagnosed from 2003 to Yes 124/292 (43) 2011 and 2011 to 2019, respectively. In both trials, bone mar- ALL cases reported to PETHEMA, median 30 (1; 74) row MRD assessment by flow cytometry was performed at the (range) end of induction (week 5) and at the end of the third consolida- Protocol deviation in ≥5% of reported No 445 (55) tion cycle (weeks 16–18). Only patients with slow clearance of patients from this center, n (%) MRD in both trials were allocated to allo-HCT, while patients Yes 371 (45) with good MRD clearance continued with chemotherapy for up Beds in the hospitala, median (range) 832 (248; 1525) to 2 years. City populationa, median (range) 409,661 [29,288; Clinical variables analyzed in this study included age, gen- 3,223,334] der, ECOG performance status, WBC, CNS infiltration, precur- AC populationa, median (range) 6,578,079 [580,229; sor lineage (B or T) presence or absence of Ph and treatment 8,384,408] period (2003–2010 versus 2011–2018). The Allo-HCT center Relative Health investment GDPa (%), 6.4 [3.9; 9.5] was defined as centers having authorization by the Spanish gov- median (range) ernment to perform allo-HCT in the same institution where the Health investment/inhabitant €a, median 1,312 [1,090; 1,631] patient was treated for the ALL. The Allo-HCT center in the (range) same province was defined as having a designated allo-HCT GDP capita/AC, median (range)a 22,700 [17,554; center in the same province where the patient was treated. 33,824] Reported ALL referred to the number of ALL patients reported to the PETHEMA database by a particular center and served as a Source: Ministry of Health and Social Services (https://www.sanidad.gob.es/en/estadE- a surrogate marker of center experience in treating ALL. Nine studios/estadisticas/sisInfSanSNS/tablasEstadisticas/InfAnSNS.htm) and Spanish National centers reporting at least 30 ALL patients each and around Statistics Institute (https://www.ine.es/dyngs/INEbase/en/categoria.htm?c=Estadisti- half of the patients together in this data set were considered as ca_P&cid=1254734710984). “experienced centers.” Protocol deviation center referred to cen- AC = autonomous community; Allo-HCT = allogeneic hematopoietic cell transplantation; CNS = central neurological system; ECOG PS = Eastern Cooperative Oncology Group Performance Status; ters with identified protocol deviations in key treatment deci- GDP = growth domestic product; Ph = Philadelphia chromosome; WBC = white blood cell count. sions (allo-HCT versus chemotherapy allocation or autologous HCT instead of allo-HCT when not indicated in the protocol) in at least 5% of the patients. Gray test was used for comparisons.14 The effects of center-re- Other demographic and macroeconomic variables are self-ex- lated and macroeconomic indicators on OS and DFS were ana- planatory and listed in Table 1. Demographic and economic lyzed using Cox proportional hazards regression models, and variables were obtained from the Spanish Government (See the Fine and Gray model was used for analyzing these effects on footnote in Table 1). CIR and NRM.15 For each main effect, a separate multivariable Clinical endpoints included overall survival (OS), disease-free model was built, adjusted for other potential risk factors (listed survival (DFS), cumulative incidence of relapse (CIR), and non- in table 1). All statistical analyses were performed using SPSS relapse mortality (NRM) were defined as previously described.11 (v.24) and R (v.4.1.0). Two-sided values of P < 0.05 were con- Infection-related mortality (IRM) was considered an explor- sidered statistically significant. atory endpoint and was defined as patients after the first CR Eight hundred sixteen patients were included in the study. dying of infectious causes during ALL treatment or after allo- Main characteristics at diagnosis appear in Table 1. Median HCT without previous ALL relapse. follow-up for patients alive was 3 years (range 0.1–9.5). Two- All numerical variables were summarized and categorized by hundred twenty-four (38%) of the evaluable patients received medians. Spearman’s rank coefficient or median test was used to allo-HCT in the first CR. In line with protocol recommenda- analyze correlations between factors. OS and DFS curves were tions, allo-HCT in CR1 ranged from 94% of the patients in the plotted by the Kaplan–Meier method and compared by the log- PH-08 trial to 6% patients in the ALL-RE-08 trial. The remain- rank test. CIR, NRM, and IRM were estimated using cumu- ing patients were allocated to further chemotherapy consolida- lative incidence functions by competing risks analysis, and the tion and maintenance regimens according to each protocol. 2 (2023) 7:1 Table 2 Multivariable Adjusted Model for Center-related and Macroeconomic Variables for Patients Outcomes Factor N OS, HR (IC 95%) P N DFS, HR (IC 95%) P N CIR, HR (IC 95%) P N NRM, HR (IC 95%) P Allo-HCT center  No 285 1 0.803 257 1 0.475 274 1 0.450 259 1 0.610  Yes 488 1.029 (0.822;1.288) 442 1.087 (0.864; 1.368) 462 1.111 (0.847; 1.456) 449 1.114 (0.736; 1.688) ALL reported cases  ≤30 407 1 0.543 370 1 0.623 394 1 0.320 374 1 0.930  >30 366 1.069 (0.862; 1.326) 329 1.056 (0.850; 1.313) 342 1.137 (0.882; 1.465) 334 0.983 (0.664; 1.457) Number of beds in the hospital  >824 384 1 0.707 346 1 0.504 359 1 0.670 353 1 0.640  ≤824 389 1.042 (0.839; 1.295) 353 1.078 (0.865; 1.342) 377 1.057 (0.818; 1.365) 355 0.909 (0.612; 1.352) Region population  ≤6,578,079 442 1 0.302 406 1 0.450 429 1 0.210 410 1 0.930 3  >6,578,079 331 1.122 (0.901; 1.398) 293 1.090 (0.872; 1.363) 307 1.177 (0.913; 1.517) 298 1.019 (0.681; 1.524) City population  ≤409,661 396 1 0.446 354 1 0.176 368 1 0.010 360 1 0.240  >409,661 377 1.087 (0.878: 1.345) 345 1.162 (0.935; 1.444) 368 1.399 (1.084; 1.804) 348 0.792 (0.537; 1.169) Region GDP per capita (€)  >22,700 380 1 0.690 341 1 0.660 355 1 0.310 344 1 0.930  ≤22,700 393 1.045 (0.842; 1.297) 358 1.050 (0.844; 1.306) 381 1.143 (0.884; 1.477) 364 0.982 (0.660; 1.462) Health investment (%)  ≥6.4% 419 1 0.919 380 1 0.859 403 1 0.660 386 1 0.850  <6.4% 354 1.011 (0.814; 1.256) 319 1.020 (0.819; 1.270) 333 1.060 (0.820; 1.371) 322 1.038 (0.670; 1.547) Health investment (per capita)  ≥1312 393 1 0.766 350 1 0.963 362 1 0.460 356 1 0.440  <1312 380 0.968 (0.779; 1.202) 349 0.995 (0.799; 1.238) 374 1.100 (0.582; 1.418) 352 0.855 (0.576; 1.270) AC = autonomous community; Allo-HCT = allogeneic hematopoietic cell transplantation; CNS = central neurological system; CIR = cumulative incidence of relapse; DFS = disease-free survival; ECOG PS = Eastern Cooperative Oncology Group Performance Status; GDP = growth domestic product; HR = hazard ratio; NRM = nonrelapse mortality; OS = overall survival; Ph = Philadelphia chromosome; WBC = white blood cell count. www.hemaspherejournal.com Barba et al Center and Macroeconomic Variables in ALL Fifty-five centers in 40 different cities located in 15 Spanish with standardized disease evaluations in reference center labora- Autonomous Communities included at least 1 patient in one tories allowed adult patients with ALL to have similar outcomes of the 4 protocols. Twenty-three (42%) centers were accred- irrespective of demographic, social and economic characteris- ited for allo-HCT and 47% of nontransplant institution had an tics of the centers, cities, and regions where they were treated. allo-HCT center in the same province. Median number of ALL These results support the work done by PETHEMA and other reported by center to PETHEMA database during the study National Cooperative Groups in Europe as they assure quality period was 10 (interquatile range 4–20). Approximately half of of care and equity to all citizens. the patients (n=424, 52%) were reported by ten centers which To the best of our knowledge, this is the first study focused on were considered experienced centers for this study. Patient a wide variety of center-related and macroeconomic variables in characteristics were similar between experienced and less-expe- adult ALL. Contrary to ours, other studies have either focused rienced centers, except for treatment period (Suppl.Table S1). in one or very few center-related variables (eg, distance to ALL Twenty centers (36%) had at least a major deviation in more treating center) or have not included well known clinical and than 5% of their reported patients. Details on center-related and genetic factors in their analyses.16 Impact of insurance status has macroeconomic variables are summarized in Table 1. also been identified as a prognostic factor for adults with ALL,17 As expected, there were some associations among center-re- although this would not apply to most European public health lated and macroeconomic variables. Hence, being a transplant systems where the great majority of ALL patients are treated in center was correlated with the number of ALL cases reported fully accessible public hospitals. (P = 0.002), the number of beds in the hospital (P < 0.001), Regarding efficacy, we did not observe differences across cen- and with major protocol deviations (P = 0.039). Health invest- ters in terms of center-related and macroeconomic variables. The ment relative to GDP was inversely correlated with the number only variables associated with DFS and OS were those depen- of inhabitants in the Autonomous Community (P = 0.032) and dent on patient and disease characteristics including age, ECOG directly associated with its GDP per capita (P = 0.001). Other performance status and WBC at diagnosis, all of them being explored associations are listed in Suppl. Table S2. previously identified as prognostic factors for ALL patients.7,11,18 Probability of OS at 5 years for the whole cohort was 46% Noteworthy, none of the variables associated with the experi- (95% confidence interval [CI] 42-50). Variables associated with ence of the center or size of the city or region where the patient lower OS in the univariable analysis were older age, higher was treated associated with DFS or OS. Patients treated in larger ECOG PS score, higher WBC at diagnosis, Ph-negative ALL and cities had a higher CIR which did not translated into lower OS diagnosis in the earlier period (Suppl. Figure S1). Probability or DFS. Reasons for this finding are uncertain and might reflect of DFS at 5 years for the whole cohort was 43% (95% CI: early referrals of more complex cases at diagnosis to larger cen- 38-47). In the univariable analysis, ECOG PS > 1, higher WBC ters located in highly populated cities. at diagnosis, Ph-negative ALL, and earlier diagnosis were asso- Center experience has been identified as a prognostic factor in ciated with worse DFS (Suppl. Table S3). In the multivariable other high-risk procedures performed in hematological patients model adjusted for macroeconomic variables, none of the cen- such as allo-HCT.19 However, we did not find any impact of ter-, region- and economic-related variables was associated with center experience on the outcome of patients. Transplantation lower OS or DFS (Table 2; Suppl. Figure S2). is probably a more complex and less well standardized proce- Of the 816 patients, 738 (90%) achieved CR after one or dure than first-line therapy for ALL, in which physician’s exper- two induction cycles. Cumulative incidence of relapse among tise might be more relevant on patient outcomes. Conversely, them was 41% (95% CI: 36-45). In the univariable analysis, PETHEMA ALL protocols include very detailed information in factors associated with an increased risk of relapse were older terms of dosing and modification of chemotherapeutic agents age, higher WBC at diagnosis, Ph-negative ALL, and T-cell phe- and transplant indications which might have contributed to mit- notype. The only center-, region- and economic-related variable igate the potential impact of center experience and assure the associated with a higher CIR in the multivariable model was a equity and access to complex therapies to all the population. higher number of inhabitants in the city where the treatment Limitations of our study include a local diagnosis of ALL center was located hazard ratio (HR) 1.399 (95% CI: 1.084- in most patients, a relatively limited sample size and the fact 1.804, P = 0.01) (Table 2). that our results are based on data reported to our cooperative One hundred ten patients died of non-relapsing causes, 49 of group database. Prospective reporting of data to a cooperative them (45%) after allo-HCT. Cumulative incidence of NRM at group might indicate certain commitment on scientific studies 5 years was 17% (95% CI: 14-20). In the univariable analysis, of these centers. While using data prospectively reported to a older age, higher EOCG PS score at diagnosis, Ph-negative ALL, cooperative group might harbor a selection bias toward well and diagnosis in the earlier period were associated with higher structured and organized centers, the use of national cancer sta- NRM. In the multivariable adjusted model, none of the center-, tistics including all patients diagnosed with ALL in our country region- and economic-related variables were associated with would not have allowed us to include homogeneous population higher NRM (Table 2). of patients treated under the same protocols and with central- Finally, of the 738 patients achieving CR, 110 died without ized MRD determination, which constitutes a strength of our previous ALL relapse. Of them, causes of death could be identi- study. fied in 104 (94.5%). Seventy-three patients (70%) died of infec- In summary, center-related and macroeconomic variables do tious causes (44 during ALL treatment and 29 after allo-HCT) not seem to have an impact on outcomes in newly diagnosed and were considered to have IRM. Factors associated with adult patients with ALL in the setting of a public health system higher IRM in the univariable analysis were older age (HR = with clearly defined and structured treatment protocols. Based 1.031 [95% CI: 1.012-1.050], P = 0.001) and earlier treatment on our results, these protocols can be safely administered to period (HR = 2.061 [95% CI: 1.289-3.297], P = 0.003). Again, ALL patients in all hospitals adhered to a national cooperative none of the center-related and macroeconomic variables were ALL group. These data should be considered for nation-wide associated with higher IRM (not shown). planning of healthcare infrastructures for cancer patients. The objective of this study was to analyze the potential impact of center-related and macroeconomic variables on the outcome of adult patients with ALL treated within 4 consecutive AUTHOR CONTRIBUTIONS PETHEMA protocols. These variables were considered along Conception and Design: PB, JMR. Collection of data: All authors. Data with patient and disease factors to adjust the impact of each analysis and interpretation: PB, MM, JMR. Manuscript writing: All authors. of them. Our results indicate that detailed treatment protocols Final approval of manuscript: All authors. 4 (2023) 7:1 www.hemaspherejournal.com DISCLOSURES 8. Makkeyah S, Manzour A, Tantawy A, et al. Treatment outcomes for childhood acute lymphoblastic leukemia in low-middle income country PB declares having received honoraria from Allogene, Amgen, BMS, before minimal residual disease risk stratification. Cancer Epidemiol. Gilead, Incyte, Jazz Pharmaceuticals, Miltenyi Biomedicine, Novartis and 2021;75:102040. Roche. J.M.R. has received honoraria and grants from Amgen, Pfizer, Incyte, 9. Mock J, Meyer C, Mau LW, et al. Barriers to access to hematopoietic cell Servier, Takeda and Novartis. All the other authors have no conflicts of inter- transplantation among patients with acute myeloid leukemia in Virginia. est to disclose. Transplant Cell Ther. 2021;27:869.e1–869.e9. 10. Giri S, Pathak R, Aryal MR, et al. Impact of hospital volume on out- SOURCES OF FUNDING comes of patients undergoing chemotherapy for acute myeloid leuke- mia: a matched cohort study. Blood. 2015;125:3359–3360. PB received research funding from the Carlos III Health Institute 11. Ribera JM, Oriol A, Morgades M, et al. Treatment of high-risk (PI16/01433, PI21/0199), Asociación Española contra el Cáncer (Ideas Philadelphia chromosome-negative acute lymphoblastic leukemia Semilla 2019) and a PERIS 2018-2020 grant from the Generalitat de in adolescents and adults according to early cytologic response and Catalunya (BDNS357800). J.M.R. has received grants from Asociacion minimal residual disease after consolidation assessed by flow cytom- Española contra el Cáncer (GC16173697BIGA) and HARMONY Alliance etry: final results of the PETHEMA ALL-AR-03 trial. J Clin Oncol. (HORIZON 2020, European Union). 2014;32:1595–1604. 12. Ribera JM, García O, Moreno MJ, et al. Incidence and outcome after first molecular versus overt recurrence in patients with Philadelphia REFERENCES chromosome-positive acute lymphoblastic leukemia included in 1. Ssenyonga N, Stiller C, Nakata K, et al. Worldwide trends in popula- the ALL Ph08 trial from the Spanish PETHEMA Group. Cancer. tion-based survival for children, adolescents, and young adults diag- 2019;125:2810–2817. nosed with leukaemia, by subtype, during 2000-14 (CONCORD-3): 13. Ribera JM, Morgades M, Montesinos P, et al. A pediatric regimen for analysis of individual data from 258 cancer registries in 61 countries. adolescents and young adults with Philadelphia chromosome-negative Lancet Child Adolesc Health. 2022;6:409–431. acute lymphoblastic leukemia: results of the ALLRE08 PETHEMA trial. 2. Ribera JM, García O, Montesinos P, et al. Treatment of young patients Cancer Med. 2020;9:2317–2329. with Philadelphia chromosome-positive acute lymphoblastic leukaemia 14. Cheng SC, Fine JP, Wei LJ. Prediction of cumulative incidence function using increased dose of imatinib and deintensified chemotherapy before under the proportional hazards model. Biometrics. 1998;54:219–228. allogeneic stem cell transplantation. Br J Haematol. 2012;159:78–81. 15. Prentice RL, Zhao S. Regression models and multivariate life tables. J 3. Gökbuget N, Kneba M, Raff T, et al. Adult patients with acute lympho- Am Stat Assoc. 2021;116:1330–1345. blastic leukemia and molecular failure display a poor prognosis and are 16. Rotz SJ, Wei W, Thomas SM, et al. Distance to treatment center is associ- candidates for stem cell transplantation and targeted therapies. Blood. ated with survival in children and young adults with acute lymphoblas- 2012;120:1868–1876. tic leukemia. Cancer. 2020;126:5319–5327. 4. Ribera JM, Morgades M, Ciudad J, et al. Chemotherapy or allogeneic 17. Krakora R, Shih W, Popli P, et al. Impact of insurance status on transplantation in high-risk Philadelphia chromosome–negative adult survival outcomes in adults with acute lymphoblastic leukemia lymphoblastic leukemia. Blood. 2021;137:1879–1894. (ALL): a single-center experience. Clin Lymphoma Myeloma Leuk. 5. Testi AM, Canichella M, Vitale A, et al. Adolescent and young adult 2020;20:e890–e896. acute lymphoblastic leukemia. Final results of the phase II pediatric-like 18. Bassan R, Pavoni C, Intermesoli T, et al. Updated risk-oriented strategy GIMEMA LAL-1308 trial. Am J Hematol. 2021;96:292–301. for acute lymphoblastic leukemia in adult patients 18-65 years: NILG 6. Della Starza I, Nunes V, Lovisa F, et al. Droplet Digital PCR Improves ALL 10/07. Blood Cancer J. 2020;10:119. IG-/TR-based MRD risk definition in childhood B-cell precursor acute 19. Giebel S, Labopin M, Ibatici A, et al. Association of macroeconomic lymphoblastic leukemia. HemaSphere. 2021;5:e543. factors with nonrelapse mortality after allogeneic hematopoietic cell 7. Huguet F, Chevret S, Leguay T, et al. Intensified therapy of acute lym- transplantation for adults with acute Lymphoblastic Leukemia: an anal- phoblastic leukemia in adults: report of the randomized GRAALL-2005 ysis from the acute Leukemia working party of the EBMT. Oncologist. clinical trial. J Clin Oncol. 2018;36:2514–2523. 2016;21:377–383. 5