Distribution of adult T‐cell leukaemia/lymphoma subtypes and survival outcomes across geographical regions: An international retrospective cohort study Authors: Bryan Valcarcel, Atae Utsunomiya, Maki Otsuka, Masaharu Miyahara, Kenji Ishitsuka, Yasushi Takamatsu, Junji Suzumiya, Kazuo Tamura, Jule F Vasquez, Daniel Enriquez‐Vera, Brady E Beltran, Denisse Castro, Thanya Runciman, Henry Idrobo, Elizabeth Arrieta, Oriana Arias, Eduardo Gotuzzo, Gabriela Garrido‐Pinzás, Victor Rivera, Daniel La Rosa, Camila Peña, Lorena Fiad, Juan C Ramos, Jose Sandoval‐Sus, Alejandro Sica, Britney N Bell, Lucy B Cook, Veronica Leataud, Eliana C M Miranda, Carlos Chiattone, Hiroo Katsuya, Luis Malpica Published: 2026 ABSTRACT SummaryAlthough adult T‐cell leukaemia/lymphoma (ATL) has a higher burden in certain world regions, outcomes have not been directly compared across regions. We conducted a multicentre cohort study comparing the distribution of ATL subtypes and overall survival (OS) across 1090 adults (≥18 years) diagnosed with ATL during 2000–2023 in Japan (n = 366), South America (n = 364), the United States (n = 275) and the United Kingdom (n = 85). Survival was estimated using Kaplan–Meier methods. Acute ATL was more frequent in Japan (59%) than in the United States (45%), South America (36%) and the United Kingdom (24%) (p heterogeneity < 0.0001). Conversely, lymphomatous ATL predominated in the United Kingdom (53%), South America (51%) and the United States (45%) relative to Japan (20%) (p heterogeneity < 0.0001). With a median follow‐up of 38 months interquartile range (IQR 14–68), 3‐year OS across regions ranged from 10% to 25% for acute, 16%–34% for lymphomatous, 49%–86% for chronic and 60%–83% for smouldering ATL. Evidence of regional differences in OS was observed for acute, lymphomatous and chronic ATL but not for smouldering ATL. In conclusion, the distribution of ATL subtypes varies globally, and OS for aggressive subtypes (i.e. acute and aggressive) remains poor across regions. Our findings could guide the design of clinical studies aimed at developing accessible therapies through international collaboration.Keywords: adult T‐cell leukaemia/lymphoma, epidemiology, HTLV‐1, prevalence, survival INTRODUCTION Adult T‐cell leukaemia/lymphoma (ATL) is an aggressive malignancy caused by chronic infection with the human T‐cell lymphotropic virus type 1 (HTLV‐1), an oncogenic retrovirus. Although ATL is more prevalent in Japan and Latin America compared to other regions, 1 , 2 direct comparisons of clinical features and treatment outcomes across prevalent regions are lacking. Most available data come from Asian cohorts, 3 , 4 , 5 , 6 while studies describing the characteristics of patients managed in Latin America remain limited. 7 , 8 Cross‐study comparison of previous reports suggests that Latin American patients may present with a younger age at diagnosis, female predominance and a higher frequency of lymphomatous ATL compared to Japanese series, 3 , 7 , 9 but these observations remain unconfirmed. Additionally, a study has suggested distinct molecular features between patients from Japan and the Western region. 10 A better understanding of these regional differences in clinical presentation could provide valuable insights into the interaction between host factors and disease biology. The management of ATL in clinical practice remains heterogeneous, and the impact of regional variation on patient outcomes is poorly understood. Although the 2019 ATL consensus meeting report provides guidelines for clinical management, 11 the adoption of the recommendations is likely suboptimal. 7 For instance, the use of zidovudine plus interferon has been suggested to be effective for acute ATL, 12 but whether its use is consistent across regions is unknown. For instance, a cohort study suggested a higher rate of fungal infections among patients with acute ATL receiving zidovudine plus interferon. 13 Chemotherapy availability differs across regions, influencing treatment selection. 8 , 11 Previous studies have suggested that cyclophosphamide, doxorubicin, vincristine and prednisone (CHOP) is widely used in Latin America, 7 while series from Japan typically use the modified LSG15 regimen. 11 Additionally, haematopoietic stem cell transplantation has been shown to improve survival in acute ATL, 14 , 15 but limited accessibility in developing countries, such as those in Latin America, 16 may impair survival outcomes. To date, it is unknown to what extent the difference in clinical practice across endemic regions may affect survival. In the absence of high‐quality population‐based cancer registries in HTLV endemic regions, 17 hospital‐based data are a valuable effort to delineate ATL epidemiology. To address these gaps, we established the ‘ATL Consortium’, a multicentre international collaboration including patients managed in Japan, South America, the United States and the United Kingdom. This study describes differences in prevalence of ATL subtypes, treatment approaches and survival outcomes across diverse geographical regions. METHODS Population and design We conducted a multicentre retrospective cohort study of patients with ATL diagnosed between 2000 and 2023. Hospital‐based data were collected through medical chart review the Brazilian T‐cell project (BTCP, 2015–2022) in Brazil, the Epidemiology of Lymphomas in Latin America (ELLA) registry from the Grupo de Estudio Latinoamericano de Linfoproliferativos (GELL, 2000–2023 period), Japanese centres (2000–2009 period), US centres (2000–2023 period) and UK centres (2003–2023 period). The ELLA registry involved the participation of four countries (Argentina, Chile, Colombia and Peru). All patients were enrolled consecutively, although in the United Kingdom, only those who consented to data transfer were included in the study. We included all patients aged ≥18 years and diagnosed through pathology confirmation of a T‐cell lymphoma with a positive enzyme‐linked immunosorbent assay (ELISA) test for HTLV‐1. A confirmatory western blot test for HTLV‐1 infection was not uniformly performed due to its high cost. No central review was performed. For the prevalence estimation of ATL subtypes and clinical description, we excluded patients with unknown age, unknown diagnosis date, diagnosis before 2000, unknown ATL subtype or primary cutaneous ATL, given the small sample size (Figure 1). For survival analyses, we further excluded patients who had unknown vital status or last follow‐up date (Figure 1). Study flow chart. ATL, adult T‐cell leukaemia/lymphoma; BTCP, Brazilian T‐cell project; ELLA, Epidemiology of Lymphomas in Latin America; UK, United Kingdom; US, United States. Outcomes The outcomes of interest were the prevalence of ATL subtypes and overall survival (OS). We defined OS as the time between the date of diagnosis and death from any cause or last available follow‐up, whichever comes first. Outcomes are estimated and reported for each ATL subtype (i.e. acute, lymphomatous, chronic and smouldering), due to differences in clinical management and prognosis. 3 , 7 , 8 Exposure Patients were grouped by treatment location as follows: South America (ELLA and BTCP registries), Japan, the United States and the United Kingdom. Patients managed in US centres were either from the Caribbean or US‐born with Caribbean ancestry. Patients treated in the United Kingdom were generally Caribbean‐ or West African‐born, or born in the United Kingdom with Caribbean or West African ancestry. Variables We analysed the following variables: age, sex, skin involvement, Eastern Cooperative Oncology Group (ECOG) performance status, presence of B symptoms, nodal and/or extranodal involvement, Ann Arbor staging (for lymphomatous type), serum lactate dehydrogenase (LDH) levels and presence of hypercalcemia. We used the Shimoyama classification to classify ATL clinical subtypes. 18 Treatment approaches were grouped into (1) multiagent chemotherapy alone; (2) zidovudine‐interferon alpha (AZT/IFN) alone; (3) combined multiagent chemotherapy with AZT/IFN and (4) unknown. We grouped therapy regimens based on their frequency as follows: CHOP, CHOP‐like (i.e. addition of any agent to CHOP), etoposide‐based chemotherapy (i.e. CHOEP/EPOCH), LSG15 or its modified version (mLSG15, a modified combination of VCAP [vincristine, cyclophosphamide, doxorubicin and prednisolone], AMP [doxorubicin, ranimustine and prednisolone] and VECP [vindesine, etoposide, carboplatin and prednisolone]) and other chemotherapy. Data analysis Categorical variables are reported as frequencies and percentages, whereas continuous variables are reported as median and range. To compare clinical features across geographical regions, we used the chi‐square test for categorical variables and the Mann–Whitney U‐test for continuous age. The 95% confidence intervals (CIs) for ATL subtypes were estimated with the Sison and Glaz method for multinomial proportions. To assess whether the distribution of subtypes differs across regions, we modelled counts using a Poisson regression, with the log of the total population in each region as an offset. A likelihood ratio test was used to compare null and regional‐adjusted models. p‐values <0.05 indicated significant heterogeneity in the distribution of subtypes across regions. The lowest median follow‐up identified with the reverse Kaplan–Meier method (Latin America, 38 months) was used as the reference for comparisons. OS was estimated using the Kaplan–Meier method, and estimates were compared with the log‐rank test. To account for differences in age composition across regions and assess the relative magnitude of OS differences, Cox proportional hazards models adjusted for age were fitted, and results are reported as hazard ratios (HRs) with their respective 95% CIs. We further use these models to estimate age‐adjusted OS. Subtype‐specific Cox models were additionally constructed to account for clinical differences in disease presentation. These models were further adjusted to ECOG and Ann Arbor stage in lymphomatous ATL and to ECOG in acute and chronic ATL. Due to small sample sizes, we only report age‐adjusted models for smouldering ATL. Adjustment for additional factors was not possible because clinical information was incomplete in some registries. RESULTS Patient characteristics across regions We included 1090 patients with ATL (366 from Japan, 364 from South America, 275 from the United States and 85 from the United Kingdom) to describe clinical features (Figure 1). The median age at diagnosis was higher in Japan than in other regions (66 years vs. 52–57 years; p < 0.001) (Table 1). Females were more commonly diagnosed in the United States (61%) and South America (55%) than in Japan (48%; p = 0.0048). Additionally, patients in the United States had a significantly higher frequency of poor performance status (ECOG 2–4, 59% vs. 27%–38%; p < 0.001) and hypercalcemia (47% vs. 25%–33%; p < 0.0001). Although the distributions of advanced stage disease (range 88%–94%; p = 0.007) and high LDH values (range 77%–85%; p = 0.047) differed statistically, the percentages were high across all regions. Skin involvement (range 27%–30%; p = 0.673) was similar across regions. TABLE 1Clinical features of patients with adult T‐cell leukaemia/lymphoma according to the place where patients received treatment.CharacteristicsAll patients, N (%)Japan, N (%)South America, N (%)US, N (%)UK, N (%) p No. of patients109036636427585Year of diagnosis<0.0012000–2004274 (25)233 (64)7 (2)30 (11)4 (5)2005–2009210 (19)133 (36)25 (7)45 (16)7 (8)2010–2014146 (13)‐65 (18)62 (23)19 (22)2015–2019333 (31)‐210 (58)95 (35)28 (33)2020–2023127 (12)‐57 (16)43 (16)27 (32)Age at diagnosis in years, median (range)59 (18–95)66 (19–89)54 (18–95)57 (20–87)52 (31–83)<0.001Age >60 years501 (46)242 (66)119 (33)117 (43)23 (27)<0.001Females541 (54)175 (48)199 (55)167 (61)‐0.0048 a Missing88‐‐‐85ECOG score 2–4397 (40)134 (37)134 (38)107 (59)22 (27)<0.001Missing10629932B symptoms355 (47)84 (24)223 (69)48 (58)‐<0.0001 a Missing329124019285Skin involvement233 (29)105 (29)70 (27)58 (30)‐0.673 a Missing27451018385Extra‐nodal involvement593 (64)205 (56)201 (62)187 (79)‐<0.0001 a Missing163‐403885Extra‐nodal >1306 (33)113 (31)95 (29)98 (41)‐0.0061 a Missing1630403885Advanced cancer stage (III–IV)941 (92)344 (94)312 (88)228 (94)57 (90)0.007Missing65193322High LDH680 (81)280 (77)249 (82)151 (85)‐0.047 a Missing2462619885Hypercalcaemia286 (33)92 (25)89 (33)105 (47)‐<0.0001 a Missing2343955185First‐line treatment0.005Yes937 (86)294 (80)317 (87)248 (90)78 (92)No81 (7)45 (12)21 (6)8 (3)7 (8)Best supportive care72 (7)27 (7)26 (7)19 (7)0 (0)Abbreviations: ECOG, Eastern Cooperative Oncology Group; LDH, lactate dehydrogenase; UK, United Kingdom; US, United States. a Chi‐square test was estimated only between Japan, South America, and the US because of a lack of data from the UK cohort. Clinical features of patients with adult T‐cell leukaemia/lymphoma according to the place where patients received treatment. Abbreviations: ECOG, Eastern Cooperative Oncology Group; LDH, lactate dehydrogenase; UK, United Kingdom; US, United States. Chi‐square test was estimated only between Japan, South America, and the US because of a lack of data from the UK cohort. Distribution of ATL subtypes We observed a significantly lower prevalence of acute ATL in the United States (45%, 95% CI = 39%–51%), South America (36%, 95% CI = 31%–42%) and the United Kingdom (24%, 95% CI = 14%–35%) compared to Japan (59%, 95% CI = 54%–64%; p heterogeneity < 0.0001) (Figure 2; Table S1). Conversely, lymphomatous ATL was significantly more prevalent in the United Kingdom (53%, 95% CI = 44%–65%), South America (51%, 95% CI = 46%–57%) and the United States (45%, 95% CI = 39%–52%) than in Japan (20%, 95% CI = 15%–25%; p heterogeneity <0.0001). The prevalence of chronic ATL ranged from 8% in the United States to 16% in the United Kingdom without a statistical difference in its distribution across regions (p heterogeneity = 0.065). The percentage distribution of smouldering ATL was more pronounced in Japan (8%, 95% CI = 3%–13%) and the United Kingdom (7%, 95% CI = 0%–19%) compared to South America and the United States (2%, 95% CI = 0%–8% each) (p heterogeneity = 0.001). Percentage distribution across geographical regions among patients with acute (A), lymphomatous (B), chronic (C) and smouldering (D) ATL subtypes. UK, United Kingdom; US, United States. Treatment patterns First‐line treatment was assessed among 1068 patients (Figure 3; Table S2), showing considerable regional variability. For acute ATL, AZT/IFN alone or in combination with chemotherapy was commonly used in South America (41%), the United States (69%) and the United Kingdom (70%), but none in Japan (p < 0.001). Among chemotherapy approaches, CHOP was the most commonly used in Japan (59%), South America (55%) and the United Kingdom (69%), whereas CHOEP/EPOCH was most frequently used in the United States (42%); mLSG15/LSG15 was the second most frequent in Japan (26%). The use of allogeneic haematopoietic stem cell transplantation (allo‐HSCT) as consolidation ranged 11%–19% across regions, except in South America, where only one patient received this therapy. First‐line treatment approaches across geographical regions among patients with acute (A), lymphomatous (B), chronic (C) and smouldering (D) subtypes. AI, zidovudine‐interferon alpha; CT, chemotherapy; UK, United Kingdom; US, United States. Among patients with lymphomatous ATL, chemotherapy alone was the predominant approach in Japan (89%), the United States (75%) and South America (59%), while the United Kingdom frequently used chemotherapy associated with AZT/IFN, either concurrently or sequentially as a bridge to allo‐HSCT (58%) (p < 0.001). CHOP use was common in Japan (64%), South America (50%) and the United Kingdom (70%), compared to CHOEP/EPOCH in the United States (47%). HSCT was more common in the United Kingdom (38%) than in the United States (13%), Japan (8%) or South America (4%). Treatment approaches for chronic and smouldering ATL subtypes are depicted in Figure 3, but the limited sample sizes in each region precluded meaningful comparisons. Survival outcomes With a median follow‐up of 38 months (interquartile range 14–68 months), the median OS and the 3‐year OS in the whole cohort were 11 months (95% CI = 10–12) and 25% (95% CI = 22%–28%) respectively. Unadjusted Kaplan–Meier curves and estimates for each ATL subtype are shown in Figure 4 and Table S3. For acute ATL, 3‐year OS estimates were low across all regions (South America, 10%; Japan, 11%; the United States, 20%; and the United Kingdom, 25%; p = 0.0042). For lymphomatous ATL, 3‐year OS ranged from 16% in Japan to 34% in the United Kingdom, with intermediate estimates of 26% in South America and 20% in the United States (p = 0.065). Among patients with chronic ATL, 3‐year OS estimates were 49% in South America and 50% in Japan, while we observed 74% in the United States and 86% in the United Kingdom (p = 0.025). For smouldering ATL, 3‐year OS ranged from 60% to 83% across regions (p = 0.742). Kaplan–Meier curves for overall survival across geographical regions among patients with acute (A), lymphomatous (B), chronic (C) and smouldering (D) subtypes. ATL, UK, United Kingdom; US, United States. Age‐adjusted survival curves generally showed similar patterns to those observed in the unadjusted Kaplan–Meier analyses (Figures S1 and S2; Table S3). The magnitude of regional differences in OS persisted after adjusting for age and other clinical factors, although these findings should be interpreted descriptively given the heterogeneity of the cohorts (Table S4). DISCUSSION In our international cohort, we observed significant regional differences in the distribution and outcomes of ATL subtypes. Patients outside Japan were diagnosed about a decade younger and had a slight female predominance. As previously observed, lymphomatous ATL was more frequently diagnosed outside Japan, whereas acute ATL predominated in Japan; the reasons for this are unclear. Despite differences in treatment approaches, survival outcomes for acute and lymphomatous ATL remained poor. However, outcomes were relatively better in the United Kingdom, likely due to differences in cohort composition and underlying patient characteristics compared with other regions. Our descriptive findings provide a global perspective on ATL epidemiology and outcomes, underscoring the need to elucidate disease biology and to develop novel treatment approaches. Similarly, the reasons behind demographic and clinical differences are unclear. Consistent with previous series, the median age at diagnosis for regions outside Japan was approximately a decade earlier. 7 , 8 , 9 , 12 , 19 But unlike a US population‐based report showing 54% of patients aged >60, 20 our cohort found 42%. The slight observed female predominance is consistent with hospital‐based Western studies 7 , 12 , 21 but contrasts with US population‐based data showing no sex difference. 20 Interestingly, population‐based analyses in Japan have found a higher seroprevalence of HTLV‐1 among females, 9 but a slight male predominance in ATL. 3 , 9 Whether these findings reflect differences in disease biology, ATL subtype distribution, comorbidities, data source (hospital vs. registry) or differences in healthcare access remains uncertain. Further preclinical and population‐based studies are needed to confirm our findings. The regional variation in the prevalence of ATL subtypes may reflect differences in host–virus interactions, HTLV‐1 strain diversity, screening practices or classification criteria. The drivers of lymphomatous or acute presentations remain unclear. Previous research has identified distinct mutational profiles between North American, Caribbean and Japanese populations, 22 , 23 , 24 suggesting a complex interplay between environmental exposures, lifestyle factors and genetic predispositions. HTLV‐1 carriers, and hence those developing ATL, are generally from low socioeconomic settings and are likely vulnerable to accessing the healthcare system, 25 , 26 which could delay diagnosis, allowing the disease to advance. Another possibility is the criteria used for defining acute or lymphomatous ATL. For instance, patients with >5% lymphocytes (i.e. suggesting acute ATL) but with a bulky presentation may have been classified as lymphomatous ATL in our database, partly explaining the variation in ATL subtype prevalence across regions. Our findings should be validated in standardized prospective studies. Additional research into viral genomics, host environment and social factors is needed to understand ATL subtype development across populations. Despite variability in therapy approaches, survival outcomes for acute and lymphomatous ATL remained poor across all regions. These findings are in line with prior studies reporting 4‐year OS rates of 4%–16% for acute and lymphomatous ATL in North American and Asian populations. 3 , 8 The higher OS observed in the United Kingdom is likely attributable to improved access to therapies (e.g. allo‐HSCT) or cohort composition. The UK cohort included patients referred to the national HTLV‐1 service for specialized management, who were fitter (27% ECOG score ≥ 2 vs. 37%–59%) and younger (52 years vs. 54–66 years), and if in remission, more likely to be eligible for allo‐HSCT. The sickest patients in the United Kingdom generally did not survive induction therapies and were not routinely captured. Despite recommendations to use allo‐HSCT for treating acute or lymphomatous ATL, 11 , 27 only one individual in South America underwent this procedure. Several barriers limit access to HSCT in this region, including high cost, a shortage of specialized personnel, poor infrastructure, limited access to timely care, challenges in identifying a suitable donor and inadequate treatment responses that preclude eligibility for transplantation. 28 Greater access to AZT/IFN in the United States and the United Kingdom was likely associated with improved OS in chronic ATL, consistent with prior evidence supporting this approach where available, 11 but findings need validation in prospective studies. Effective and accessible therapies are urgently needed for managing patients with ATL, particularly for aggressive subtypes. Our study has limitations to address. HTLV‐1 confirmation by western blot was not uniformly performed due to its high out‐of‐pocket cost, and data on which patients had access to this test were not captured. We relied on hospital‐based data to estimate prevalence because population‐based cancer registries likely capture ATL as peripheral T‐cell lymphoma not otherwise specified, 29 lack ATL subtype classification and lack follow‐up data. Although generalizability is possible, US patients are managed in specific geographical locations, and those in Latin America are treated at large academic centres, which supports generalizability to patients treated in metropolitan areas, though this would not represent patients in rural communities. Clinical features beyond cancer stage and performance status were unavailable for the UK cohort. Changes in HTLV‐1 screening practices over time among patients undergoing differential diagnosis for a haematological malignancy may also have influenced prevalence estimates. Similarly, missing data on ATL subtype may have influenced ATL prevalence estimates. For survival outcomes, this study includes Japanese patients treated prior to the widespread adoption of the mLSG15 regimen, which should be taken into consideration when interpreting the results. Additionally, the japan clinical oncology group (JCOG) 9801 trial did not identify a difference in survival compared to CHOP, possibly suggesting stable survival outcomes over time. 30 In contrast, therapy for ATL outside Japan has largely been consistent over time. 8 The relatively low sample size for each country and ATL subtype limited the estimation of relative survival. To note, estimation of treatment effectiveness of chemotherapy approaches or HSCT was outside the scope of this paper. In conclusion, this international cohort represents one of the largest global efforts demonstrating geographical heterogeneity in the distribution of ATL subtypes. Acute ATL was common in Japan, while lymphomatous ATL was more frequent in the other regions. Despite the different treatment approaches, survival outcomes remained poor for aggressive ATL subtypes. The difference in cohort methodology and composition urges the need to develop an ATL‐based cohort to improve data collection practices. Therefore, the establishment of the ‘ATL consortium’ is a step forward, strengthening partnerships among international institutions to improve clinical research. A better understanding of ATL biology across populations, including patient ancestry, could lead to the development of more effective therapies to improve global outcomes. AUTHOR CONTRIBUTIONS Masaharu Miyahara: Writing – review and editing; resources; data curation. Jule F. Vasquez: Writing – review and editing; data curation; resources. Kenji Ishitsuka: Data curation; resources; writing – review and editing. Kazuo Tamura: Writing – review and editing; resources; data curation. Maki Otsuka: Writing – review and editing; resources; data curation. Yasushi Takamatsu: Writing – review and editing; resources; data curation. Junji Suzumiya: Data curation; resources; writing – review and editing. Brady E. Beltran: Writing – review and editing; data curation; resources. Denisse Castro: Data curation; resources; writing – review and editing. Daniel Enriquez‐Vera: Data curation; resources; writing – review and editing. Oriana Arias: Data curation; resources; writing – review and editing. Atae Utsunomiya: Resources; writing – review and editing; data curation. Henry Idrobo: Data curation; resources; writing – review and editing. Elizabeth Arrieta: Writing – review and editing; resources; data curation. Thanya Runciman: Writing – review and editing; resources; data curation. Daniel La Rosa: Writing – review and editing; resources; data curation. Victor Rivera: Data curation; writing – review and editing. Camila Peña: Writing – review and editing; resources; data curation. Lorena Fiad: Data curation; resources; writing – review and editing. Bryan Valcarcel: Conceptualization; investigation; methodology; validation; visualization; software; formal analysis; data curation; writing – original draft; writing – review and editing. Eduardo Gotuzzo: Writing – review and editing; resources; data curation. Gabriela Garrido‐Pinzás: Writing – review and editing; resources; data curation. Hiroo Katsuya: Data curation; resources; writing – review and editing; conceptualization. Juan C. Ramos: Writing – review and editing; resources; data curation. Alejandro Sica: Writing – review and editing; data curation; resources. Jose Sandoval‐Sus: Data curation; resources; writing – review and editing. Britney N. Bell: Writing – review and editing; resources; data curation. Carlos Chiattone: Writing – review and editing; resources; data curation. Lucy B. Cook: Writing – review and editing; data curation; resources. Eliana C. M. Miranda: Data curation; resources; writing – review and editing. Veronica Leataud: Writing – review and editing; resources; data curation. Luis Malpica: Conceptualization; funding acquisition; writing – review and editing; methodology; validation; project administration; resources; supervision; data curation. FUNDING INFORMATION Funding was provided by the Amos Medical Faculty Development Program from the American Society of Hematology, the Robert A. Winn Diversity in Clinical Trials and the Birdies Fore Logan Foundation. CONFLICT OF INTEREST STATEMENT JS‐S: Speaker: Pfizer; Advisory Board: AstraZeneca, Acrotech, ADC therapeutics, BeOne, BMS, Genentech, Genmab, Janssen, Kite, Novartis, Roche, Pfizer. No conflict of interest for the remaining authors. Supporting information Figures S1–S2. Tables S1–S4. ACKNOWLEDGEMENTS We would like to thank the Grupo de Estudios Latinoamericano de Linfoproliferativos (GELL) and the Registro Brasileiro de Linfomas T for their data management and statistical support. DATA AVAILABILITY STATEMENT The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions. REFERENCES Associated Data Supplementary Materials Figures S1–S2. Tables S1–S4. Data Availability Statement The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.