Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Clinical Outcomes of Patients With B-Cell Non-Hodgkin Lymphoma in Real-World Settings: Findings From the Hemato-Oncology Latin America Observational Registry Study

Clinical Outcomes of Patients With B-Cell Non-Hodgkin Lymphoma in Real-World Settings: Findings... original reports abstract HEMATOLOGIC MALIGNANCIES Clinical Outcomes of Patients With B-Cell Non-Hodgkin Lymphoma in Real-World Settings: Findings From the Hemato-Oncology Latin America Observational Registry Study 1 2 3 4 1 Miguel Pavlovsky, MD ; Daniel Cubero, MD, PhD ; Gladys Patricia Agreda-Vasquez, ´ MD ; Alicia Enrico, MD ; Maria J. Mela-Osorio, MD ; 5 6 7 8 Jorge Armenta San Sebastian, ´ MD ; Laura Fogliatto, MD, PhD ; Roberto Ovilla, MD ; Oscar Avendano, MD ; 9 9 10 10 Gerardo Machnicki, MD, PhD ; Paula Barreyro, MD ; Damila Trufelli, MD, MSc ; and Pamella Villanova, MD PURPOSE Real-world evidence on non-Hodgkin lymphoma (NHL) management in Latin America is currently lacking. The objective of this study was to describe treatment characteristics and outcomes of NHL in Latin America. METHODS A total of 2,967 patients with NHL with aggressive and indolent subtypes, including diffuse large B-cell lymphoma (DLBCL), follicular lymphoma (FL), mantle-cell lymphoma (MCL), and mucosa-associated lymphoid tissue (MALT) lymphoma, with incident or prevalent diagnosis between 2006 and 2015, were retrospectively identified using clinical charts registered in the Hemato-Oncology Latin America Observational Registry. As- sociations between treatment regimen and age at diagnosis with clinical outcomes within each subtype were estimated using Cox proportional hazard regression. RESULTS Most patients with NHL received 1L chemoimmunotherapy, most commonly cyclophosphamide, doxorubicin, vincristine, and prednisone (CHOP) with/without rituximab. Five-year survival rates were higher for MALT lymphoma (90.8%) and FL (87.6%) versus DLBCL (69.0%) and MCL (57.1%), with variations between countries. The median overall survival from first relapse for patients with DLBCL was 6.6 years, with lower risk of death for those diagnosed at age , 65 years (hazard ratio = 0.732; P = .0161). Patients achieved a longer median progression-free survival with 1L rituximab-CHOP (R-CHOP) versus CHOP or rituximab, cyclophosphamide, vincristine, and prednisone (RCVP) (7.7 v 3.0 or 1.8 years, respectively). Use of regimens other than R-CHOP was associated with a higher risk of death/progression for patients with DLBCL (rituximab, ifosfamide, carboplatin, and etoposide/ifosfamide, carboplatin, and etoposide) and FL (CHOP). There was no relationship between treatment prescribed and age at diagnosis with outcomes from first/second relapse in DLBCL and FL. CONCLUSION Differences in treatment outcomes between NHL subtypes were observed, reflecting variations in NHL management and barriers to treatment access in Latin America. These data provide necessary evidence to understand NHL management in this region and highlight the need to improve treatment outcomes for these patients. JCO Global Oncol 8:e2100265. © 2022 by American Society of Clinical Oncology Licensed under the Creative Commons Attribution 4.0 License ASSOCIATED CONTENT Appendix INTRODUCTION By 2030, numbers of NHL incidence and deaths are Author affiliations and support projected to increase by 60% (to 43,000 cases and Non-Hodgkin lymphoma (NHL) is a heterogeneous information (if 24,000 deaths) in the region. group of hematologic malignancies resulting from applicable) appear at uncontrolled proliferation of lymphoid tissues. NHL In the past 15 years, there have been paradigm shifts the end of this article. has diverse genotypes, immunologic phenotypes, in diagnosis, staging, and treatment of hematologic Accepted on March molecular biology, morphology, and clinical charac- malignancies around the world. However, there is a 14, 2022 and teristics. Central and South America shoulder 7% of paucity of real-world evidence (RWE) on NHL in Latin published at the world’s burden of NHL, including 27,000 incident America, so it is unclear how these changes have been ascopubs.org/journal/ 2 2-4 cases and 14,000 deaths each year. Based on these translated to clinical practice in the region. The WHO go on April 29, 2022: data, NHL is the 9th-ranked cancer diagnosis and the estimates that only 8% of Latin American populations DOI https://doi.org/10. 1200/GO.21.00265 11th-ranked cause of cancer death in Latin America. are covered by cancer registries. 1 Pavlovsky et al CONTEXT Key Objective There is a lack of evidence on real-world outcomes of diagnosis, staging, and treatment of patients with non-Hodgkin lymphoma (NHL) in Latin America. The aim of this study was to describe the treatment characteristics and clinical outcomes of patients diagnosed with NHL who received systemic therapy in this region. Knowledge Generated Nearly 3,000 patients with NHL from seven Latin American countries were included and variations in disease management and differences in 5-year survival rates between different NHL subtypes and countries were found. The majority of patients received first-line chemoimmunotherapy, with newer treatment options such as rituximab-based regimens showing improved clinical outcomes. Relevance The findings from our study are important for informing ongoing efforts and future strategies in improving NHL management in Latin America, highlighting potential differences in patient access to appropriate therapy between different regions. The Hemato-Oncology Latin America Observational practice type representativeness, and willingness to comply (HOLA) Registry study was conducted with the primary aim with study requirements. Data were collected via retro- of quantifying demographic and clinical characteristics of spective chart reviews conducted by trained medical ab- patients who had diagnoses of multiple myeloma (MM), stractors using standardized data collection forms. chronic lymphocytic leukemia (CLL), or NHL and received Study Participants care at tertiary care hospitals. MM- and CLL-related epi- An inception cohort of patients with NHL who satisfied the demiology and treatment outcome data from the HOLA 6-8 following criteria were included in the analysis: (1) incident study have been published elsewhere. This study reports or prevalent NHL diagnosis between January 1, 2006, and the treatment patterns and clinical outcomes of patients December 31, 2015; (2) patient age ≥ 18 years at the time who were diagnosed with NHL and received systemic of the first observed diagnosis; (3) 1 year or more of patient therapy. data after the first observed diagnosis (except in the event of METHODS patient death within 1 year of being diagnosed); and (4) patient (or legal representative) able and willing to provide Study Design informed consent (except in the event of obtaining a waiver This multicenter, observational study retrospectively of informed consent). Participants in clinical trials were reviewed medical records from January 1, 2006, to De- excluded. cember 31, 2015, of adult Latin American patients diag- Study Bias nosed with MM, CLL, or NHL (NCT02559583) from 30 oncology hospitals in seven countries: Argentina (five), Selection bias was minimized by applying broad eligibility Brazil (nine), Chile (one), Colombia (five), Mexico (six), criteria, recruiting patients from a diverse pool of clinical Panama (three), and Guatemala (one). Participating hos- sites, and screening and identifying eligible patients con- pitals were selected based on their experience in providing secutively. To minimize measurement bias associated with clinical care for hematologic patients, geographic and inaccurate assessments and variable content of medical TABLE 1. Outcome Measure Definition Outcome Measure Beginning Date End Date PFS/OS from diagnosis The date of initial diagnosis PFS: first date of disease progression or death OS: date of death PFS/OS from frontline treatment The date of frontline treatment initiation PFS: first date of disease progression or death OS: date of death PFS/OS from first relapse The date of first relapse PFS: first date of disease progression or death OS: date of death PFS/OS from second relapse The date of second relapse PFS: first date of disease progression or death OS: date of death NOTE. These defined time points (beginning date and ending date) were used to measure OS or PFS from start of diagnosis, frontline treatment, first relapse, or second relapse. Abbreviations: OS, overall survival; PFS, progression-free survival. 2 © 2022 by American Society of Clinical Oncology Clinical Outcomes of B-cell NHL in Latin America TABLE 2. Baseline Characteristics of Patients by Non-Hodgkin Lymphoma Subtype Characteristic DLBCL (N = 1,457) FL (N = 578) MCL (N = 183) MALT Lymphoma (N = 84) Others (N = 180) IPI used IPI FLIPI MIPI NA NA Total, No. 599 208 50 NA NA Low, n (%) 189 (31.6) 75 (36.1) 14 (28.0) NA NA Intermediate, n (%) 67 (32.2) 12 (24.0) Intermediate low 145 (24.2) NA NA Intermediate high 156 (26.0) NA NA High, n (%) 109 (18.2) 66 (31.7) 24 (48.0) NA NA Bone marrow infiltration Total tested, No. 1,052 447 133 55 107 Negative for infiltration, n (%) 871 (82.8) 269 (60.2) 56 (42.1) 47 (85.5) 23 (21.5) Positive for infiltration, n (%) 181 (17.2) 178 (39.8) 77 (57.9) 8 (14.5) 30 (28.0) Age at diagnosis, years—categorical Total, No. 1,455 576 183 84 NA , 50, n (%) 448 (30.8) 147 (25.5) 32 (17.5) 28 (33.4) NA 50 to , 60, n (%) 314 (21.6) 188 (32.6) 45 (24.6) 12 (14.3) NA 60 to , 70, n (%) 373 (25.6) 139 (24.1) 57 (31.1) 19 (22.6) NA 70 to , 80, n (%) 236 (16.2) 76 (13.2) 35 (19.1) 20 (23.8) NA ≥ 80, n (%) 84 (5.8) 26 (4.5) 14 (7.7) 5 (6.0) NA Median age at diagnosis, years (range) 58 (18-95) 57 (18-92) 61 (27-93) 62 (18-88) NA ECOG score at diagnosis Total, No. 516 170 52 25 43 0, n (%) 242 (46.9) 117 (68.8) 33 (63.5) 16 (64.0) 27 (62.8) 1, n (%) 165 (32.0) 38 (22.4) 11 (21.2) 5 (20.0) 11 (25.6) 2, n (%) 63 (12.2) 13 (7.6) 8 (15.4) 3 (12.0) 2 (4.7) 3, n (%) 30 (5.8) 1 (0.6) 0 (0.0) 1 (4.0) ≥ 4, n (%) 16 (3.1) 1 (0.6) 0 (0.0) 0 (0.0) 1 (2.3) Stage (Ann Arbor) Total, No. 1,212 520 152 50 117 I, n (%) 216 (17.8) 69 (13.3) 12 (7.9) 18 (36.0) 26 (22.2) II, n (%) 296 (24.4) 102 (19.6) 10 (6.6) 13 (26.0) 14 (12.0) III, n (%) 295 (24.3) 177 (34.0) 23 (15.1) 7 (14.0) 19 (16.2) IV, n (%) 405 (33.4) 172 (33.1) 107 (70.4) 12 (24.0) 58 (49.6) Not specified, n (%) 97 21 16 15 NA Hemoglobin, g/dL Total, No. 1,177 478 141 69 156 , 8, n (%) 61 (5.2) 11 (2.3) 5 (3.5) 2 (2.9) 18 (11.5) 8-12, n (%) 443 (37.6) 107 (22.4) 58 (41.1) 21 (30.4) 59 (37.8) . 12, n (%) 673 (57.2) 360 (75.3) 78 (55.3) 46 (66.7) 79 (50.6) NOTE. Baseline characteristics of patients classified as DLBCL (n = 1,457), FL (n = 578), MCL (n = 183), MALT lymphoma (n = 84), and other types of lymphoma (others; n = 180) at diagnosis are as listed, including score based on IPI used, presentation of bone marrow infiltration, age, ECOG rating, stage (Ann Arbor), and hemoglobin level. Abbreviations: DLBCL, diffuse large B-cell lymphoma; ECOG, Eastern Cooperative Oncology Group; FL, follicular lymphoma; FLIPI, follicular lymphoma International Prognostic Index; IPI, International Prognostic Index; MALT, mucosa-associated lymphoid tissue; MCL, mantle-cell lymphoma; MIPI, Mantle- Cell Lymphoma International Prognostic Index; NA, not available. records, chart abstractors underwent comprehensive assessment of desired data elements. Missing values did central training in performing reviews, and clear definitions not contribute to denominators used to estimate of variables of interest were provided to ensure accurate percentages. JCO Global Oncology 3 Pavlovsky et al Statistical Methods age at diagnosis on OS and PFS for DLBCL and FL, es- timated using the Cox proportional hazard model. Sta- Demographics and clinical variables were characterized tistical analyses were conducted using SAS version 9.4 using descriptive statistics, including measures of central (SAS Institute, Cary, NC). tendency (mean and median) and spread (variance, range, minimum, and maximum) for continuous variables (eg, age RESULTS at diagnosis) and frequency distributions (No., %) for Overview of Patients With NHL Included in the Study categorical variables (eg, sex). Outcome measures for overall survival (OS) and progression-free survival (PFS) A total of 2,967 patients with NHL were included from were defined as summarized in Table 1. Patients who were Mexico (28.4%) Argentina (20.4%), Colombia (16.6%), lost to follow-up or alive at the last available data extraction Brazil (15.5%), Chile (15.3%), and Panama/Guatemala were censored at the date of last consult. PFS was censored (3.8%). The median follow-up duration was 2.20 years at the time of the last consult date available. (range , 0.1-11.8; n = 2,821). Overall, 691 (23.3%) pa- tients died during follow-up and 1,571 (52.9%) remained OS and PFS were assessed following Kaplan-Meier under follow-up at the end of the study period. methodology, with results described when the number of patients at risk at the beginning of the observation Of these 2,967 patients, 2,948 had accurate NHL subtype period was ≥ 10. Outcome analyses were conducted in information: 2,518 (85.4%) were classified as B-cell NHL four B-cell NHL subtypes: diffuse large B-cell lymphoma including 1,457 (57.9%) DLBCL, 578 (23.0%) FL, 183 (DLBCL), follicular lymphoma (FL), mantle-cell lymphoma (7.3%) MCL, 90 (3.6%) CLL/small lymphocytic lymphoma, (MCL), and mucosa-associated lymphoid tissue (MALT) 84 (3.3%) MALT lymphoma, 60 (2.4%) Burkitt, 35 (1.4%) lymphoma but were not further analyzed by country. lymphoplasmacytic, and 31 (1.2%) B-lymphoblastic lym- Because of the limitation of sample sizes, associations phoma. Of the remainder, 250 (8.5%) were diagnosed with were only examined for different treatment regimens and T-cell NHL and 180 (6.1%) as other types of lymphoma FIG 1. Five-year cumulative survival rate of patients by NHL subtypes from diagnosis. The 5-year cumulative sur- vival rates of patients with DLBCL (n = 1,475), FL (n = 578), MCL (n = 183), MALT lymphoma (n = 84), Burkitt lymphoma (n = 60), lympho- plasmacytic lymphoma (n = 35), B- lymphoblastic lymphoma (n = 31), T- cell lymphoma (n = 250), and other types of lymphoma (others, n = 180) from diagnosis are illustrated in the top figure. The survival rate for each NHL subtype further stratified by country, NHL Subtypes expressed as percentage (%), are de- tailed in the bottom table. CLL, chronic 5-Year Lympho- B lymphocytic leukemia; DLBCL, diffuse Survival Rate DLBCL FL MCL MALT Burkitt plasmac- Lympho- T-cell Other (%) ytic blastic large B-cell lymphoma; FL, follicular lymphoma; MALT, mucosa-associated Overall 69 87.6 57.1 90.8 55.7 65.9 35.1 49.3 83 lymphoid tissue; MCL, mantle-cell Argentina 72.6 86.6 66.5 100 69.3 NA NA 40.3 94 lymphoma; NA, not available; NHL, Brazil 50.8 71.3 35.4 92.3 57.1 52.6 0 37.5 76.7 non-Hodgkin lymphoma; SLL, small Chile 81.7 87.3 38.1 75 NA NA NA 83.9 85.4 lymphocytic lymphoma. Colombia 62.1 96.1 75.1 91.7 47.6 100 60 49.2 63.5 Mexico 71.7 95.3 75.7 92.9 0 NA NA 47.6 86.1 Panama 93.8 93.3 85.7 NA NA NA NA 77.8 100 and Guatemala 4 © 2022 by American Society of Clinical Oncology DLBCL (n = 1,457) FL (n = 578) MCL (n = 183) MALT (n = 84) Burkitt (n = 60) Lymphoplasmacytic (n = 35) B-Lymphoblastic (n = 31) T-Cell (n = 250) Others (n = 180) 5-Year Cumulative Survival Rate (%) Clinical Outcomes of B-cell NHL in Latin America TABLE 3. Cumulative Survival Rate for DLBCL Stratified by IPI chemotherapy-only regimens, typically CHOP (n = 70 of 5-Year Cumulative 546, 12.8%) or cyclophosphamide, vincristine, and IPI Risk Group Total, No. Survival Rate, % prednisone (CVP; n = 35 of 546, 6.4%). The median Low IPI 189 87.3 duration from initial diagnosis to treatment was 1.1 months Intermediate low IPI 145 75.2 (range 0-74.7; n = 467 of 578), which could be interpreted Intermediate high IPI 156 57.6 as the median watch and wait period. High IPI 109 45.6 For MCL, the majority of patients received chemother- apy (n = 172 of 183, 94.0%). Commonly prescribed reg- NOTE. The 5-year cumulative survival rates of patients with DLBCL imens included R-CHOP (n = 71 of 172, 41.3%); CHOP (n = 588) from diagnosis stratified by their IPI score at diagnosis are as (n = 28 of 172, 16.3%); hyper cyclophosphamide, vin- tabulated. cristine, and doxorubicin (hyper CVAD; n = 18 of 172, Abbreviations: DLBCL, diffuse large B-cell lymphoma; IPI, 10.5%); and RCVP (n = 9 of 172, 5.2%). The median International Prognostic Index; NHL, non-Hodgkin lymphoma. duration from initial diagnosis to treatment was 0.9 months (range 0-63.3; n = 130 of 183). (which may include subtypes of marginal zone lymphoma other than MALT; Table 2). A similar pattern was observed among patients with MALT lymphoma, with 79.8% (n = 67 of 84) receiving a chemo- Treatment Characteristics therapy regimen (R-CHOP [n = 24 of 67, 35.8%]; CHOP Most patients with DLBCL received chemoimmunotherapy [n = 12 of 67, 17.9%]; RCVP [n = 9 of 67, 13.4%]; CVP [n = 6 regimens (n = 1,420 of 1,457; 97.5%), of which the most of 67, 9.0%]; and rituximab monotherapy [n = 6 of 67, commonly reported was rituximab, cyclophosphamide, 9.0%]). The median duration from initial diagnosis to treat- doxorubicin, vincristine, and prednisone (R-CHOP; n = 998 ment was 2.3 months (range, 0.1-42.1; n = 57 of 84), which of 1,420; 70.3%) and CHOP (excluding rituximab; n = 208 could be interpreted as the median watch and wait period. of 1,420; 14.6%). The median duration from initial diag- Relapse or progression since start of treatment was re- nosis to treatment for DLBCL was 0.8 months (range ported for 29% (n = 392 of 1,351) of patients with DLBCL, 0-23.1; n = 1,198). 30% (n = 157 of 524) with FL, 46.8% (n = 72 of 154) with The majority of patients with FL received chemoimmu- MCL, and 24.2% (n = 15 of 62) with MALT lymphoma. The notherapy (n = 546 of 578, 94.5%). Approximately 70% use of transplant was reported for 6.2% (n = 88 of 1,420) of received CHOP-like regimens combined with rituximab: patients with DLBCL, 5.7% (n = 31 of 546) with FL, 13.4% R-CHOP (n = 280 of 546, 51.3%) or rituximab, cyclo- (n = 23 of 172) with MCL and none with MALT lymphoma. phosphamide, vincristine, and prednisone (RCVP; n = 96 of Among patients who received a transplant, most had an 546, 17.6%). A further 19.2% of patients received autologous transplant (97.7%, 93.5%, and 90.9% in 1.0 0.8 FIG 2. OS of patients by NHL subtypes from start of frontline treatment. Kaplan- Meier survival curves depicting the OS of 0.6 patients with DLBCL (n = 1,387), FL (n = 544), MCL (n = 172), and MALT lymphoma (n = 77) from the start of 0.4 first-line therapy; the results are de- Censored scribed when the number of patients at DLBCL risk at the beginning of the observation 0.2 FL period was ≥ 10. DLBCL, diffuse large MCL B-cell lymphoma; FL, follicular lym- MALT lymphoma phoma; MALT, mucosa-associated 0 50 100 150 200 lymphoid tissue; MCL, mantle-cell lym- phoma; NHL, non-Hodgkin lymphoma; Time (months) OS, overall survival. No. at risk: DLBCL 1,387 497 100 0 FL 544 284 65 1 1 MCL 172 70 8 0 MALT lymphoma 77 34 3 0 JCO Global Oncology 5 OS (probability) Pavlovsky et al TABLE 4. PFS of Patients by NHL Subtypes From Start of Frontline Treatment for 0.7-57.6; n = 74) for CHOP, and 5 months (range 0.6-21.6; the Three Most Commonly Used Regimens n = 14) for RCVP. For FL, patients treated with R-CHOP NHL Subtype/Treatment Total Patient 5-Year Cumulative Median PFS, achieved a median time from treatment initiation to relapse/ Regimen Number, No. PFS Rate (%) Years (95% CI) progression of 17.6 months (range , 0.1-74.4; n = 61), DLBCL whereas patients treated with CHOP or RCVP achieved R-CHOP 948 56.2 7.7 (5.8 to n.e.) medians of 14.5 months (range 0.5-83.3; n = 29) and 9.8 months (range 0.1-70.6; n = 26), respectively. CHOP 199 44.9 3.0 (1.9 to 6.1) RCVP 38 48.2 1.8 (0.8 to n.e.) For DLBCL, there were no statistically significant differ- ences in risk of death between patients treated with CHOP FL versus R-CHOP (HR = 1.330; P = not significant) or RCVP R-CHOP 259 65.1 NA versus R-CHOP (HR = 1.118; P = not significant). However, RCVP 91 51.0 5.9 (4.2 to n.e.) higher risk of death was observed for patients with FL CHOP 66 42.5 2.9 (1.7 to 6.9) treated with CHOP versus R-CHOP (HR = 2.508; P = .001), MCL although no differences were seen for RCVP versus R-CHOP 66 30.5 2.4 (1.3 to 3.0) R-CHOP (HR = 1.066; P = not significant). CHOP 28 20.4 1.9 (1.6 to 4.1) For patients with DLBCL, 5-year PFS rates for the three most common frontline regimens ranged from 44.9% to NOTE. The 5-year cumulative PFS and median PFS of patients with DLBCL 56.2%, with a median PFS of 7.7 years (95% CI, 5.8 to not (n = 1,185), FL (n = 416), MCL (n = 98), and MALT lymphoma (n = 42) from the estimable (n.e.); n = 948) for patients receiving R-CHOP start of first-line therapy treated with the top three most used regimens in each NHL (Table 4). Similarly, 5-year PFS rates among patients with subtype are as tabulated. Note that data for patients with MCL treated with hyper FL for the three most commonly used regimens were CVAD and for patients with MALT lymphoma are not available because of low sample sizes. 42.5%-65.1%. There was no relationship between type of Abbreviations: CHOP, cyclophosphamide, doxorubicin, vincristine, and frontline regimen prescribed and age at diagnosis with PFS, prednisone; CVAD, cyclophosphamide, vincristine, and doxorubicin; DLBCL, for either DLBCL or FL. diffuse large B-cell lymphoma; FL, follicular lymphoma; MALT, mucosa-associated OS/PFS From First Relapse lymphoid tissue; MCL, mantle-cell lymphoma; NA, not available; n.e., not estimable; NHL, non-Hodgkin lymphoma; PFS, progression-free survival; R-CHOP, Overall survival of patients with DLBCL, FL, MCL, or MALT rituximab, cyclophosphamide, doxorubicin, and vincristine; RCVP, rituximab, lymphoma from first relapse is shown in Figure 3. The cyclophosphamide, vincristine, and prednisone. median OS from first relapse was 6.6 years (95% CI, 3.1 to n.e.) for DLBCL, whereas median OS from first relapse was not reached for FL, MCL, or MALT lymphoma. DLBCL, FL, and MCL, respectively); however, data on the line of therapy were not available. The median PFS from first relapse was 2.3 years (95% CI, 0.4 to n.e.) for DLBCL and 2.6 years (95% CI, 1.1 to 3.9) for OS from Diagnosis by NHL Subtypes MCL. Median PFS was not reached in patients with FL or Analysis of cumulative survival rates from diagnosis for MALT lymphoma. Additional analyses on OS/PFS from NHL subtypes found that 5-year survival rates were highest second relapse are shown in Appendix Figs A1 and A2. for patients with MALT lymphoma (90.8%), followed by FL For DLBCL, the three most common treatments at first (87.6%), DLBCL (69.0%), and MCL (57.1%). Survival relapse were ifosfamide, carboplatin, and etoposide with/ rates for each NHL subtype were further stratified by without rituximab (RICE/ICE; 21.1%, n = 86 of 407); eto- country (Fig 1). poside, methylprednisolone, cytarabine, and cisplatin Among patients for whom International Prognostic Index (10.1%, n = 41 of 207); and R-CHOP (8.8%, n = 36 of (IPI) was available at diagnosis (n = 599), the 5-year survival 407). Patients treated with RICE/ICE after first relapse were rate was highest in patients with low IPI (87.3%) and lowest at significantly higher risk of death versus patients treated in patients with high IPI scores (45.6%; Table 3). with R-CHOP (HR = 3.115; P = .0035). Similarly, the risk of disease progression after first relapse was also higher with OS/PFS From Start of Frontline Treatment RICE/ICE versus R-CHOP (HR = 3.846; P = .0011). Overall survival of patients with DLBCL, FL, MCL, and MALT However, we were not able to identify the first-line treat- lymphoma were analyzed from the start of frontline therapy ments prescribed to those patients who used R-CHOP as (Fig 2). Patients with DLBCL age , 65 years at diagnosis second-line rescue therapy. were at lower risk of death compared with patients ≥ 65 years (hazard ratio [HR] = 0.732; P = .0161). DISCUSSION Of patients who experienced relapse or progression, the median time from treatment initiation to relapse/ Management of patients with NHL has changed sub- progression was 8.4 months (range , 0.1-92.6; stantially over the past two decades with improved diag- n = 236) for those treated with R-CHOP, 7.6 months (range nostic techniques, newer treatment options, and 6 © 2022 by American Society of Clinical Oncology Clinical Outcomes of B-cell NHL in Latin America 1.0 0.8 FIG 3. OS of patients by NHL subtypes from first relapse. Kaplan-Meier survival 0.6 curves depicting the OS of patients with DLBCL (n = 393), FL (n = 204), MCL (n = 94), and MALT lymphoma (n = 17) 0.4 from first relapse; the results are de- scribed when the number of patients at Censored risk at the beginning of the observation DLBCL 0.2 FL period was ≥ 10. DLBCL, diffuse large MCL B-cell lymphoma; FL, follicular lym- MALT lymphoma phoma; MALT, mucosa-associated lymphoid tissue; MCL, mantle-cell lym- 07 25 50 5 100 125 phoma; NHL, non-Hodgkin lymphoma; OS, overall survival. Time (months) No. at risk: DLBCL 393 149 72 26 8 0 FL 204 131 67 31 10 0 MCL 94 49 16 3 0 MALT lymphoma 17 9 2 0 consequently better treatment outcomes. However, RWE rate of 87.6%, which was higher than observed for other generation is necessary to better understand how these NHL subtypes (except MALT lymphoma). These observa- treatment improvements are implemented in individual tions are in line with previously reported improvements in countries. As RWE for NHL in Latin America is generally OS for patients with FL treated with rituximab-based 2-4 9,10 lacking, data presented here from the observational regimens, as observed by comparing OS of patients HOLA study provide a significant step in better under- diagnosed with FL during the rituximab era (1997-2003, standing treatment pathways and outcomes for patients in median OS not reached) with earlier decades (median OS of the region. Of note are the cohort size of the patients with 13.6-18.5 years). The results from our study also highlight NHL from seven Latin American countries (n = 2,967) and the necessity of detecting FL earlier in Latin America as the the long-term 8-year follow-up period. median time from diagnosis to treatment was 1.1 months (range 0-74.7) for nearly 80% of the FL patient cohort in the Among all NHL subtypes, DLBCL is the most common HOLA study, who were mostly diagnosed at advanced subtype reported globally (1,457, 57.9% in our study) and stages of their disease and required systemic therapy. is typically treated using combination chemotherapy in- cluding CHOP and R-CHOP. In our study, the 5-year MCL is an aggressive form of B-cell NHL. Relapsed MCL survival rates for DLBCL ranged from 45.6% in patients has been associated with poor responses and OS of , 3 with high IPI to 87.3% in patients with low IPI. Improved years. Little has been reported on treatment outcomes of outcomes were seen in our study when comparing patients with MCL in Latin America, with one retrospective R-CHOP versus CHOP, including 5-year PFS rates and study conducted in Mexico identifying only 12 patients. median PFS; however, there was no statistical difference The majority of patients with MCL in our study received in risk of death or disease progression between the reg- chemotherapy, where 41.3% of patients received R-CHOP imens. The reason for this is unclear, although may be due and 16.3% received CHOP. The 5-year survival rate for to the limitations of analyzing data from retrospective MCL was 57.1%, which compared poorly with rates for observational studies. Nonetheless, we found that pa- DLBCL and FL. The MCL cohort studied in the HOLA tients with DLBCL treated with RICE/ICE after first relapse registry is smaller compared with other subtypes, as would were at significantly higher risk of death and/or disease be expected given that MCL is a rarer form of NHL. Despite progression versus R-CHOP, thus signifying the impor- this, our study provides useful insights into the manage- tance of newer therapies in improving survival outcomes ment of the disease in this region, and greater international for patients. efforts should be considered to further understand out- comes of MCL in Latin America. FL accounts for around 20-30% of NHL cases in our study, with 51.3% of patients treated with chemotherapy receiving A cohort of 84 patients with MALT lymphoma were ana- R-CHOP. Patients with FL achieved a 5-year overall survival lyzed in our study. Given the indolent nature of the disease, JCO Global Oncology 7 OS (probability) Pavlovsky et al specifically identifying the watch and wait subgroup proved Collectively, caution should be exercised when interpreting challenging. However, the median time from diagnosis to our PFS and OS estimates, especially in comparison with treatment initiation reported for MALT lymphoma other studies. Finally, censored and not censored patients (2.3 months) could potentially be interpreted as such. may not have presented homogeneous characteristics at Additionally, we were unable to categorize further subsets the beginning of the study. of marginal zone lymphoma lymphomas, such as splenic or Overall, this study summarizes the treatment outcomes for extranodal subtypes, which may have been classified as patients with NHL in Latin America and highlights some others (n = 180). differences between participating countries which should Maintenance therapy is a known effective strategy in be considered with care because of sample numbers and prolonging remission for patients with NHL after induction heterogeneity of patient characteristics. Nonetheless, dis- therapy or stem-cell transplantation. For example, the parities in survival outcomes between the seven partici- clinical value of rituximab maintenance in treatment of FL pating countries (for example, survival rates for DLBCL, FL, has been demonstrated in several real-world studies in the and MCL were lowest in Brazil), potentially reflect regional 14 15 16 United States, Taiwan, and Czech Republic. Unfor- differences in patient access to timely treatment. Socio- tunately, data on the use of maintenance therapy in low- economic, geographic, and cultural disparities in the region grade lymphomas were not gathered in the HOLA study, can affect outcomes for hematological malignancies, with with the exception of Argentina where rituximab mainte- some of the key challenges including limited numbers of nance was reported as the most common treatment for hematologists and inadequate access to bone marrow DLBCL at first or at second relapse (n = 11 of 74, 14.9% transplant centers. Furthermore, although R-CHOP is the and n = 4 of 18, 22.2%, respectively). standard of care for DLBCL globally, nearly 15% of patients with DLBCL in our study were prescribed CHOP, which may There are certain limitations of this study, some of which are be indicative of late adoption of novel treatments in this common to retrospective observational studies. First, region. Furthermore, we found that management of missing data were observed for some variables because of second- and third-line patients was inconsistent, reflecting limited information available in patient records. Second, the lack of standard of care. although the total sample size was adequate, smaller sample sizes for some subgroups may have influenced In conclusion, the HOLA study is the first large scale, real- findings, particularly those related to outcomes from Cox world observational study to report treatment pathways and models. Third, PFS and OS analyses were affected by a clinical outcomes for patients with NHL in multiple countries large degree of loss to follow-up, especially from first and across Latin America. These data are important for informing second relapse, limiting the precision of those analyses. local improvements in NHL management and for under- PFS and OS analyses also assumed that patients with standing how treatment practices differ compared with other continued follow-up were representative of the original total regions in the world. Data gaps from this study highlight the patient cohort and may also be overly optimistic if, for real-world challenges of providing treatment and recording example, an out-of-hospital death resulted in loss to follow- data across the region, indicating a need for further studies up but was not documented in the patient’s medical chart. and support for treatment of NHL in Latin America. AFFILIATIONS SUPPORT ´ ´ ´ ´ Servicio de Hematologıa e Investigacion Clınica, Fundacion para The HOLA registry and this report were sponsored by Janssen-Cilag Combatir la Leucemia (FUNDALEU), Buenos Aires, Argentina Farmaceuti ˆ ca Ltda (São Paulo, Brazil). Assistance in manuscript CEPHO/ABC School of Medicine, Santo Andre, ´ Brazil preparation, according to the Strengthening the Reporting of Instituto Nacional de Ciencias Medicas y Nutricion Salvador Zubiran, ´ Observational Studies in Epidemiology guidelines, was provided by Wen Mexico City, Mexico Chean Lim (VMLY&R Health) with support from the study sponsor. Hospital Italiano La Plata, Buenos Aires, Argentina Centro Oncologico ´ Estatal ISSEMYM, Toluca, Mexico AUTHOR CONTRIBUTIONS Hospital de Cl´ınicas de Porto Alegre, Brazil Conception and design: Miguel Pavlovsky, Laura Fogliatto, Gerardo Hospital Angeles Lomas, Huixquilucan, Mexico Machnicki, Pamella Villanova Medical Solutions S.A. Guatemala City, Guatemala Provision of study materials or patients: Miguel Pavlovsky, Daniel Cubero, Janssen-Cilag Farmaceutica Ltda, Buenos Aires, Argentina Gladys Patricia Agreda-Vasquez, Alicia Enrico, Maria J. Mela-Osorio, Janssen-Cilag Farmaceutica Ltda, São Paulo, Brazil Jorge Armenta San Sebastian, Laura Fogliatto, Roberto Ovilla, Oscar Avendano CORRESPONDING AUTHOR Collection and assembly of data: Daniel Cubero, Gladys Patricia Agreda- Pamella Villanova, MD, Av. Presidente Juscelino Kubitscheck, 2041 São Vasquez, Alicia Enrico, Maria J. Mela-Osorio, Jorge Armenta San Paulo, CEP 04543-011 Brazil Complexo JK - Bloco B, Brazil; Sebastian, Laura Fogliatto, Roberto Ovilla, Oscar Avendano, Paula e-mail: pvillano@its.jnj.com. Barreyro Data analysis and interpretation: Miguel Pavlovsky, Gladys Patricia Agreda- ´ ´ Vasquez, Jorge Armenta San Sebastian, Laura Fogliatto, Gerardo Machnicki, Damila Trufelli, Pamella Villanova 8 © 2022 by American Society of Clinical Oncology Clinical Outcomes of B-cell NHL in Latin America Manuscript writing: All authors Gladys Patricia Agreda-Vasquez Final approval of manuscript: All authors Honoraria: Janssen, Roche Accountable for all aspects of the work: All authors Consulting or Advisory Role: Asofarma Gerardo Machnicki AUTHORS’ DISCLOSURES OF POTENTIAL CONFLICTS OF Employment: Janssen INTEREST Stock and Other Ownership Interests: Janssen The following represents disclosure information provided by authors of Paula Barreyro this manuscript. All relationships are considered compensated unless Employment: Janssen-Cilag otherwise noted. Relationships are self-held unless noted. I = Immediate Stock and Other Ownership Interests: Janssen-Cilag Family Member, Inst = My Institution. Relationships may not relate to the subject matter of this manuscript. For more information about ASCO’s Damila Trufelli conflict of interest policy, please refer to www.asco.org/rwc or ascopubs. Employment: Janssen Oncology org/go/authors/author-center. Leadership: Janssen Oncology Open Payments is a public database containing information reported by Stock and Other Ownership Interests: Janssen Oncology companies about payments made to US-licensed physicians (Open Honoraria: Janssen Oncology Payments). Pamella Villanova Miguel Pavlovsky Employment: Janssen Honoraria: AbbVie/Pharmacyclics, Roche, Novartis, Janssen, Stock and Other Ownership Interests: Janssen AstraZeneca Honoraria: Janssen Consulting or Advisory Role: AbbVie/Pharmacyclics, Roche, Novartis, Travel, Accommodations, Expenses: Janssen Janssen Speakers’ Bureau: AbbVie/Pharmacyclics, Roche, Novartis, Janssen No other potential conflicts of interest were reported. Travel, Accommodations, Expenses: Roche, Sanofi, AbbVie, Raffo REFERENCES 1. Curado MP, de Souza DL: Cancer burden in Latin America and the Caribbean. Ann Glob Health 80:370-377, 2014 2. Diumenjo MC, Abriata G, Forman D, et al: The burden of non-Hodgkin lymphoma in Central and South America. Cancer Epidemiol 44:S168-S177, 2016 (suppl 1) 3. Ortega V, Verastegui E, Flores G, et al: Non-Hodgkin’s lymphomas in Mexico. A clinicopathological and molecular analysis. Leuk Lymphoma 31:575-582, 1998 4. Muller AM, Ihorst G, Mertelsmann R, et al: Epidemiology of non-Hodgkin’s lymphoma (NHL): Trends, geographic distribution, and etiology. Ann Hematol 84:1-12, 2005 5. Bray F, Colombet M, Mery L, et al: Cancer Incidence in Five Continents, Vol. XI. IARC Scientific Publication No. 166. Lyon: International Agency for Research on Cancer. https://publications.iarc.fr/597 6. Chiattone C, Gomez-Almaguer D, Pavlovsky C, et al: Results from Hemato-Oncology Latin America Observational Registry (HOLA) providing real world outcomes for the treatment of patients with CLL. Blood 128:5578, 2016 7. Tietsche De Moraes Hungria V, Chiattone C, Pavlovsky M, et al: Epidemiology of hematologic malignancies in real-world settings: Findings from the Hemato- Oncology Latin America Observational Registry study. JCO Glob Oncol 5:1-19, 2019 8. de Moraes Hungria VT, Martınez-Baños DM, Peñafiel CR, et al: Multiple myeloma treatment patterns and clinical outcomes in the Latin America Haemato- Oncology (HOLA) Observational Study, 2008–2016. Br J Haematol 188:383-393, 2020 9. Tan D, Horning SJ, Hoppe RT, et al: Improvements in observed and relative survival in follicular grade 1-2 lymphoma during 4 decades: The Stanford University experience. Blood 122:981-987, 2013 10. Provencio M, Sabin P, Gomez-Codina J, et al: Impact of treatment in long-term survival patients with follicular lymphoma: A Spanish Lymphoma Oncology Group Registry. PLoS One 12:e0177204, 2017 11. Schieber M, Gordon LI, Karmali R: Current overview and treatment of mantle cell lymphoma. F1000Res 7:F1000 Faculty Rev-1136, 2018 ´ ´ ´ ´ 12. Cordova-Ramırez AC, Sanchez-Valledor LF, Colon-Otero G, et al: Mantle cell lymphoma may have a different clinical course in Mexican Mestizos: Real-world data from a single center. Rev Invest Clin 73(2):94-9, 2021 13. Raderer M, Kiesewetter B, Ferreri AJM: Clinicopathologic characteristics and treatment of marginal zone lymphoma of mucosa-associated lymphoid tissue (MALT lymphoma). CA Cancer J Clin 66:152-171, 2016 14. Nastoupil LJ, Sinha R, Byrtek M, et al: The use and effectiveness of rituximab maintenance in patients with follicular lymphoma diagnosed between 2004 and 2007 in the United States. Cancer 120:1830-1837, 2014 15. Huang H-H, Wen Y-C, Chen H-M, et al: Rituximab maintenance improves overall survival in follicular lymphoma: A retrospective nationwide real-world analysis from Taiwan Cancer Registry Database. Cancer Med 7:3582-3591, 2018 16. Belada D, Prochazka V, Janikova A, et al: The influence of maintenance therapy of rituximab on the survival of elderly patients with follicular lymphoma. A retrospective analysis from the database of the Czech Lymphoma Study Group. Leuk Res 73:29-38, 2018 17. Goss PE, Lee BL, Badovinac-Crnjevic T, et al: Planning cancer control in Latin America and the Caribbean. Lancet Oncol 14:391-436, 2013 18. Tilly H, Gomes da Silva M, Vitolo U, et al: Diffuse large B-cell lymphoma (DLBCL): ESMO clinical practice guidelines for diagnosis, treatment and follow-up. Ann Oncol 26:v116-v125, 2015 (suppl 5) nn n JCO Global Oncology 9 Pavlovsky et al APPENDIX 1.0 0.8 0.6 0.4 Censored 0.2 DLBCL FL MCL 08 20 40 600 100 Time (months) No. at risk: DLBCL 147 44 24 9 1 1 FL 66 37 19 10 4 0 MCL 36 14 10 3 0 FIG A1. OS of patients by NHL subtypes from second relapse. Kaplan-Meier survival curves depicting the OS of patients with DLBCL (n = 147), FL (n = 66), and MCL (n = 36) from second relapse; the results are described when the number of patients at risk at the beginning of the observation period was ≥ 10. In total, 152 patients with DLBCL, 70 patients with FL, 38 patients with MCL, and 3 patients with MALT lymphoma received treatment for second relapse. For DLBCL, the median OS from second relapse was 1.9 years (95% CI, 0.43 to n.e.), whereas the median OS from second relapse for FL or MCL was 5.0 years (95% CI, 1.9 to n.e.) or 4.4 years (95% CI, 0.8 to n.e.), respectively. Outcome analysis was not conducted for MALT lymphoma because of small patient sample size. DLBCL, diffuse large B-cell lymphoma; FL, follicular lymphoma; MALT, mucosa-associated lymphoid tissue; MCL, mantle-cell lymphoma; n.e., not estimable; NHL, non-Hodgkin lymphoma; OS, overall survival. 10 © 2022 by American Society of Clinical Oncology OS (probability) Clinical Outcomes of B-cell NHL in Latin America 1.0 0.8 0.6 0.4 Censored 0.2 DLBCL FL MCL 020 40 60 80 Time (months) No. at risk: DLBCL 145 27 12 3 0 FL 64 29 12 5 0 MCL 35 8 2 0 FIG A2. PFS of patients by NHL subtypes from second relapse. Kaplan-Meier survival curves depicting the PFS of patients with DLBCL (n = 145), FL (n = 64), and MCL (n = 35) from second relapse; the results are described when the number of patients at risk at the beginning of the observation period was ≥ 10. The median PFS from second relapse was 1.3 years (95% CI, 0.4 to 6.2) in DLBCL, 4.3 years (95% CI, 1.8 to n.e.) in FL, and 1.5 years (95% CI, 0.8 to 4.4) in MCL. DLBCL, diffuse large B-cell lymphoma; FL, follicular lymphoma; MCL, mantle-cell lymphoma; n.e., not estimable; NHL, non-Hodgkin lymphoma; PFS, progression-free survival. JCO Global Oncology 11 PFS (probability) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png JCO: Global Oncology Wolters Kluwer Health

Clinical Outcomes of Patients With B-Cell Non-Hodgkin Lymphoma in Real-World Settings: Findings From the Hemato-Oncology Latin America Observational Registry Study

Download PDF
11 pages

Loading next page...
 
/lp/wolters-kluwer-health/clinical-outcomes-of-patients-with-b-cell-non-hodgkin-lymphoma-in-real-OZDAA6FH2D

References (36)

Publisher
Wolters Kluwer Health
Copyright
© 2022 by American Society of Clinical Oncology
eISSN
2687-8941
DOI
10.1200/go.21.00265
Publisher site
See Article on Publisher Site

Abstract

original reports abstract HEMATOLOGIC MALIGNANCIES Clinical Outcomes of Patients With B-Cell Non-Hodgkin Lymphoma in Real-World Settings: Findings From the Hemato-Oncology Latin America Observational Registry Study 1 2 3 4 1 Miguel Pavlovsky, MD ; Daniel Cubero, MD, PhD ; Gladys Patricia Agreda-Vasquez, ´ MD ; Alicia Enrico, MD ; Maria J. Mela-Osorio, MD ; 5 6 7 8 Jorge Armenta San Sebastian, ´ MD ; Laura Fogliatto, MD, PhD ; Roberto Ovilla, MD ; Oscar Avendano, MD ; 9 9 10 10 Gerardo Machnicki, MD, PhD ; Paula Barreyro, MD ; Damila Trufelli, MD, MSc ; and Pamella Villanova, MD PURPOSE Real-world evidence on non-Hodgkin lymphoma (NHL) management in Latin America is currently lacking. The objective of this study was to describe treatment characteristics and outcomes of NHL in Latin America. METHODS A total of 2,967 patients with NHL with aggressive and indolent subtypes, including diffuse large B-cell lymphoma (DLBCL), follicular lymphoma (FL), mantle-cell lymphoma (MCL), and mucosa-associated lymphoid tissue (MALT) lymphoma, with incident or prevalent diagnosis between 2006 and 2015, were retrospectively identified using clinical charts registered in the Hemato-Oncology Latin America Observational Registry. As- sociations between treatment regimen and age at diagnosis with clinical outcomes within each subtype were estimated using Cox proportional hazard regression. RESULTS Most patients with NHL received 1L chemoimmunotherapy, most commonly cyclophosphamide, doxorubicin, vincristine, and prednisone (CHOP) with/without rituximab. Five-year survival rates were higher for MALT lymphoma (90.8%) and FL (87.6%) versus DLBCL (69.0%) and MCL (57.1%), with variations between countries. The median overall survival from first relapse for patients with DLBCL was 6.6 years, with lower risk of death for those diagnosed at age , 65 years (hazard ratio = 0.732; P = .0161). Patients achieved a longer median progression-free survival with 1L rituximab-CHOP (R-CHOP) versus CHOP or rituximab, cyclophosphamide, vincristine, and prednisone (RCVP) (7.7 v 3.0 or 1.8 years, respectively). Use of regimens other than R-CHOP was associated with a higher risk of death/progression for patients with DLBCL (rituximab, ifosfamide, carboplatin, and etoposide/ifosfamide, carboplatin, and etoposide) and FL (CHOP). There was no relationship between treatment prescribed and age at diagnosis with outcomes from first/second relapse in DLBCL and FL. CONCLUSION Differences in treatment outcomes between NHL subtypes were observed, reflecting variations in NHL management and barriers to treatment access in Latin America. These data provide necessary evidence to understand NHL management in this region and highlight the need to improve treatment outcomes for these patients. JCO Global Oncol 8:e2100265. © 2022 by American Society of Clinical Oncology Licensed under the Creative Commons Attribution 4.0 License ASSOCIATED CONTENT Appendix INTRODUCTION By 2030, numbers of NHL incidence and deaths are Author affiliations and support projected to increase by 60% (to 43,000 cases and Non-Hodgkin lymphoma (NHL) is a heterogeneous information (if 24,000 deaths) in the region. group of hematologic malignancies resulting from applicable) appear at uncontrolled proliferation of lymphoid tissues. NHL In the past 15 years, there have been paradigm shifts the end of this article. has diverse genotypes, immunologic phenotypes, in diagnosis, staging, and treatment of hematologic Accepted on March molecular biology, morphology, and clinical charac- malignancies around the world. However, there is a 14, 2022 and teristics. Central and South America shoulder 7% of paucity of real-world evidence (RWE) on NHL in Latin published at the world’s burden of NHL, including 27,000 incident America, so it is unclear how these changes have been ascopubs.org/journal/ 2 2-4 cases and 14,000 deaths each year. Based on these translated to clinical practice in the region. The WHO go on April 29, 2022: data, NHL is the 9th-ranked cancer diagnosis and the estimates that only 8% of Latin American populations DOI https://doi.org/10. 1200/GO.21.00265 11th-ranked cause of cancer death in Latin America. are covered by cancer registries. 1 Pavlovsky et al CONTEXT Key Objective There is a lack of evidence on real-world outcomes of diagnosis, staging, and treatment of patients with non-Hodgkin lymphoma (NHL) in Latin America. The aim of this study was to describe the treatment characteristics and clinical outcomes of patients diagnosed with NHL who received systemic therapy in this region. Knowledge Generated Nearly 3,000 patients with NHL from seven Latin American countries were included and variations in disease management and differences in 5-year survival rates between different NHL subtypes and countries were found. The majority of patients received first-line chemoimmunotherapy, with newer treatment options such as rituximab-based regimens showing improved clinical outcomes. Relevance The findings from our study are important for informing ongoing efforts and future strategies in improving NHL management in Latin America, highlighting potential differences in patient access to appropriate therapy between different regions. The Hemato-Oncology Latin America Observational practice type representativeness, and willingness to comply (HOLA) Registry study was conducted with the primary aim with study requirements. Data were collected via retro- of quantifying demographic and clinical characteristics of spective chart reviews conducted by trained medical ab- patients who had diagnoses of multiple myeloma (MM), stractors using standardized data collection forms. chronic lymphocytic leukemia (CLL), or NHL and received Study Participants care at tertiary care hospitals. MM- and CLL-related epi- An inception cohort of patients with NHL who satisfied the demiology and treatment outcome data from the HOLA 6-8 following criteria were included in the analysis: (1) incident study have been published elsewhere. This study reports or prevalent NHL diagnosis between January 1, 2006, and the treatment patterns and clinical outcomes of patients December 31, 2015; (2) patient age ≥ 18 years at the time who were diagnosed with NHL and received systemic of the first observed diagnosis; (3) 1 year or more of patient therapy. data after the first observed diagnosis (except in the event of METHODS patient death within 1 year of being diagnosed); and (4) patient (or legal representative) able and willing to provide Study Design informed consent (except in the event of obtaining a waiver This multicenter, observational study retrospectively of informed consent). Participants in clinical trials were reviewed medical records from January 1, 2006, to De- excluded. cember 31, 2015, of adult Latin American patients diag- Study Bias nosed with MM, CLL, or NHL (NCT02559583) from 30 oncology hospitals in seven countries: Argentina (five), Selection bias was minimized by applying broad eligibility Brazil (nine), Chile (one), Colombia (five), Mexico (six), criteria, recruiting patients from a diverse pool of clinical Panama (three), and Guatemala (one). Participating hos- sites, and screening and identifying eligible patients con- pitals were selected based on their experience in providing secutively. To minimize measurement bias associated with clinical care for hematologic patients, geographic and inaccurate assessments and variable content of medical TABLE 1. Outcome Measure Definition Outcome Measure Beginning Date End Date PFS/OS from diagnosis The date of initial diagnosis PFS: first date of disease progression or death OS: date of death PFS/OS from frontline treatment The date of frontline treatment initiation PFS: first date of disease progression or death OS: date of death PFS/OS from first relapse The date of first relapse PFS: first date of disease progression or death OS: date of death PFS/OS from second relapse The date of second relapse PFS: first date of disease progression or death OS: date of death NOTE. These defined time points (beginning date and ending date) were used to measure OS or PFS from start of diagnosis, frontline treatment, first relapse, or second relapse. Abbreviations: OS, overall survival; PFS, progression-free survival. 2 © 2022 by American Society of Clinical Oncology Clinical Outcomes of B-cell NHL in Latin America TABLE 2. Baseline Characteristics of Patients by Non-Hodgkin Lymphoma Subtype Characteristic DLBCL (N = 1,457) FL (N = 578) MCL (N = 183) MALT Lymphoma (N = 84) Others (N = 180) IPI used IPI FLIPI MIPI NA NA Total, No. 599 208 50 NA NA Low, n (%) 189 (31.6) 75 (36.1) 14 (28.0) NA NA Intermediate, n (%) 67 (32.2) 12 (24.0) Intermediate low 145 (24.2) NA NA Intermediate high 156 (26.0) NA NA High, n (%) 109 (18.2) 66 (31.7) 24 (48.0) NA NA Bone marrow infiltration Total tested, No. 1,052 447 133 55 107 Negative for infiltration, n (%) 871 (82.8) 269 (60.2) 56 (42.1) 47 (85.5) 23 (21.5) Positive for infiltration, n (%) 181 (17.2) 178 (39.8) 77 (57.9) 8 (14.5) 30 (28.0) Age at diagnosis, years—categorical Total, No. 1,455 576 183 84 NA , 50, n (%) 448 (30.8) 147 (25.5) 32 (17.5) 28 (33.4) NA 50 to , 60, n (%) 314 (21.6) 188 (32.6) 45 (24.6) 12 (14.3) NA 60 to , 70, n (%) 373 (25.6) 139 (24.1) 57 (31.1) 19 (22.6) NA 70 to , 80, n (%) 236 (16.2) 76 (13.2) 35 (19.1) 20 (23.8) NA ≥ 80, n (%) 84 (5.8) 26 (4.5) 14 (7.7) 5 (6.0) NA Median age at diagnosis, years (range) 58 (18-95) 57 (18-92) 61 (27-93) 62 (18-88) NA ECOG score at diagnosis Total, No. 516 170 52 25 43 0, n (%) 242 (46.9) 117 (68.8) 33 (63.5) 16 (64.0) 27 (62.8) 1, n (%) 165 (32.0) 38 (22.4) 11 (21.2) 5 (20.0) 11 (25.6) 2, n (%) 63 (12.2) 13 (7.6) 8 (15.4) 3 (12.0) 2 (4.7) 3, n (%) 30 (5.8) 1 (0.6) 0 (0.0) 1 (4.0) ≥ 4, n (%) 16 (3.1) 1 (0.6) 0 (0.0) 0 (0.0) 1 (2.3) Stage (Ann Arbor) Total, No. 1,212 520 152 50 117 I, n (%) 216 (17.8) 69 (13.3) 12 (7.9) 18 (36.0) 26 (22.2) II, n (%) 296 (24.4) 102 (19.6) 10 (6.6) 13 (26.0) 14 (12.0) III, n (%) 295 (24.3) 177 (34.0) 23 (15.1) 7 (14.0) 19 (16.2) IV, n (%) 405 (33.4) 172 (33.1) 107 (70.4) 12 (24.0) 58 (49.6) Not specified, n (%) 97 21 16 15 NA Hemoglobin, g/dL Total, No. 1,177 478 141 69 156 , 8, n (%) 61 (5.2) 11 (2.3) 5 (3.5) 2 (2.9) 18 (11.5) 8-12, n (%) 443 (37.6) 107 (22.4) 58 (41.1) 21 (30.4) 59 (37.8) . 12, n (%) 673 (57.2) 360 (75.3) 78 (55.3) 46 (66.7) 79 (50.6) NOTE. Baseline characteristics of patients classified as DLBCL (n = 1,457), FL (n = 578), MCL (n = 183), MALT lymphoma (n = 84), and other types of lymphoma (others; n = 180) at diagnosis are as listed, including score based on IPI used, presentation of bone marrow infiltration, age, ECOG rating, stage (Ann Arbor), and hemoglobin level. Abbreviations: DLBCL, diffuse large B-cell lymphoma; ECOG, Eastern Cooperative Oncology Group; FL, follicular lymphoma; FLIPI, follicular lymphoma International Prognostic Index; IPI, International Prognostic Index; MALT, mucosa-associated lymphoid tissue; MCL, mantle-cell lymphoma; MIPI, Mantle- Cell Lymphoma International Prognostic Index; NA, not available. records, chart abstractors underwent comprehensive assessment of desired data elements. Missing values did central training in performing reviews, and clear definitions not contribute to denominators used to estimate of variables of interest were provided to ensure accurate percentages. JCO Global Oncology 3 Pavlovsky et al Statistical Methods age at diagnosis on OS and PFS for DLBCL and FL, es- timated using the Cox proportional hazard model. Sta- Demographics and clinical variables were characterized tistical analyses were conducted using SAS version 9.4 using descriptive statistics, including measures of central (SAS Institute, Cary, NC). tendency (mean and median) and spread (variance, range, minimum, and maximum) for continuous variables (eg, age RESULTS at diagnosis) and frequency distributions (No., %) for Overview of Patients With NHL Included in the Study categorical variables (eg, sex). Outcome measures for overall survival (OS) and progression-free survival (PFS) A total of 2,967 patients with NHL were included from were defined as summarized in Table 1. Patients who were Mexico (28.4%) Argentina (20.4%), Colombia (16.6%), lost to follow-up or alive at the last available data extraction Brazil (15.5%), Chile (15.3%), and Panama/Guatemala were censored at the date of last consult. PFS was censored (3.8%). The median follow-up duration was 2.20 years at the time of the last consult date available. (range , 0.1-11.8; n = 2,821). Overall, 691 (23.3%) pa- tients died during follow-up and 1,571 (52.9%) remained OS and PFS were assessed following Kaplan-Meier under follow-up at the end of the study period. methodology, with results described when the number of patients at risk at the beginning of the observation Of these 2,967 patients, 2,948 had accurate NHL subtype period was ≥ 10. Outcome analyses were conducted in information: 2,518 (85.4%) were classified as B-cell NHL four B-cell NHL subtypes: diffuse large B-cell lymphoma including 1,457 (57.9%) DLBCL, 578 (23.0%) FL, 183 (DLBCL), follicular lymphoma (FL), mantle-cell lymphoma (7.3%) MCL, 90 (3.6%) CLL/small lymphocytic lymphoma, (MCL), and mucosa-associated lymphoid tissue (MALT) 84 (3.3%) MALT lymphoma, 60 (2.4%) Burkitt, 35 (1.4%) lymphoma but were not further analyzed by country. lymphoplasmacytic, and 31 (1.2%) B-lymphoblastic lym- Because of the limitation of sample sizes, associations phoma. Of the remainder, 250 (8.5%) were diagnosed with were only examined for different treatment regimens and T-cell NHL and 180 (6.1%) as other types of lymphoma FIG 1. Five-year cumulative survival rate of patients by NHL subtypes from diagnosis. The 5-year cumulative sur- vival rates of patients with DLBCL (n = 1,475), FL (n = 578), MCL (n = 183), MALT lymphoma (n = 84), Burkitt lymphoma (n = 60), lympho- plasmacytic lymphoma (n = 35), B- lymphoblastic lymphoma (n = 31), T- cell lymphoma (n = 250), and other types of lymphoma (others, n = 180) from diagnosis are illustrated in the top figure. The survival rate for each NHL subtype further stratified by country, NHL Subtypes expressed as percentage (%), are de- tailed in the bottom table. CLL, chronic 5-Year Lympho- B lymphocytic leukemia; DLBCL, diffuse Survival Rate DLBCL FL MCL MALT Burkitt plasmac- Lympho- T-cell Other (%) ytic blastic large B-cell lymphoma; FL, follicular lymphoma; MALT, mucosa-associated Overall 69 87.6 57.1 90.8 55.7 65.9 35.1 49.3 83 lymphoid tissue; MCL, mantle-cell Argentina 72.6 86.6 66.5 100 69.3 NA NA 40.3 94 lymphoma; NA, not available; NHL, Brazil 50.8 71.3 35.4 92.3 57.1 52.6 0 37.5 76.7 non-Hodgkin lymphoma; SLL, small Chile 81.7 87.3 38.1 75 NA NA NA 83.9 85.4 lymphocytic lymphoma. Colombia 62.1 96.1 75.1 91.7 47.6 100 60 49.2 63.5 Mexico 71.7 95.3 75.7 92.9 0 NA NA 47.6 86.1 Panama 93.8 93.3 85.7 NA NA NA NA 77.8 100 and Guatemala 4 © 2022 by American Society of Clinical Oncology DLBCL (n = 1,457) FL (n = 578) MCL (n = 183) MALT (n = 84) Burkitt (n = 60) Lymphoplasmacytic (n = 35) B-Lymphoblastic (n = 31) T-Cell (n = 250) Others (n = 180) 5-Year Cumulative Survival Rate (%) Clinical Outcomes of B-cell NHL in Latin America TABLE 3. Cumulative Survival Rate for DLBCL Stratified by IPI chemotherapy-only regimens, typically CHOP (n = 70 of 5-Year Cumulative 546, 12.8%) or cyclophosphamide, vincristine, and IPI Risk Group Total, No. Survival Rate, % prednisone (CVP; n = 35 of 546, 6.4%). The median Low IPI 189 87.3 duration from initial diagnosis to treatment was 1.1 months Intermediate low IPI 145 75.2 (range 0-74.7; n = 467 of 578), which could be interpreted Intermediate high IPI 156 57.6 as the median watch and wait period. High IPI 109 45.6 For MCL, the majority of patients received chemother- apy (n = 172 of 183, 94.0%). Commonly prescribed reg- NOTE. The 5-year cumulative survival rates of patients with DLBCL imens included R-CHOP (n = 71 of 172, 41.3%); CHOP (n = 588) from diagnosis stratified by their IPI score at diagnosis are as (n = 28 of 172, 16.3%); hyper cyclophosphamide, vin- tabulated. cristine, and doxorubicin (hyper CVAD; n = 18 of 172, Abbreviations: DLBCL, diffuse large B-cell lymphoma; IPI, 10.5%); and RCVP (n = 9 of 172, 5.2%). The median International Prognostic Index; NHL, non-Hodgkin lymphoma. duration from initial diagnosis to treatment was 0.9 months (range 0-63.3; n = 130 of 183). (which may include subtypes of marginal zone lymphoma other than MALT; Table 2). A similar pattern was observed among patients with MALT lymphoma, with 79.8% (n = 67 of 84) receiving a chemo- Treatment Characteristics therapy regimen (R-CHOP [n = 24 of 67, 35.8%]; CHOP Most patients with DLBCL received chemoimmunotherapy [n = 12 of 67, 17.9%]; RCVP [n = 9 of 67, 13.4%]; CVP [n = 6 regimens (n = 1,420 of 1,457; 97.5%), of which the most of 67, 9.0%]; and rituximab monotherapy [n = 6 of 67, commonly reported was rituximab, cyclophosphamide, 9.0%]). The median duration from initial diagnosis to treat- doxorubicin, vincristine, and prednisone (R-CHOP; n = 998 ment was 2.3 months (range, 0.1-42.1; n = 57 of 84), which of 1,420; 70.3%) and CHOP (excluding rituximab; n = 208 could be interpreted as the median watch and wait period. of 1,420; 14.6%). The median duration from initial diag- Relapse or progression since start of treatment was re- nosis to treatment for DLBCL was 0.8 months (range ported for 29% (n = 392 of 1,351) of patients with DLBCL, 0-23.1; n = 1,198). 30% (n = 157 of 524) with FL, 46.8% (n = 72 of 154) with The majority of patients with FL received chemoimmu- MCL, and 24.2% (n = 15 of 62) with MALT lymphoma. The notherapy (n = 546 of 578, 94.5%). Approximately 70% use of transplant was reported for 6.2% (n = 88 of 1,420) of received CHOP-like regimens combined with rituximab: patients with DLBCL, 5.7% (n = 31 of 546) with FL, 13.4% R-CHOP (n = 280 of 546, 51.3%) or rituximab, cyclo- (n = 23 of 172) with MCL and none with MALT lymphoma. phosphamide, vincristine, and prednisone (RCVP; n = 96 of Among patients who received a transplant, most had an 546, 17.6%). A further 19.2% of patients received autologous transplant (97.7%, 93.5%, and 90.9% in 1.0 0.8 FIG 2. OS of patients by NHL subtypes from start of frontline treatment. Kaplan- Meier survival curves depicting the OS of 0.6 patients with DLBCL (n = 1,387), FL (n = 544), MCL (n = 172), and MALT lymphoma (n = 77) from the start of 0.4 first-line therapy; the results are de- Censored scribed when the number of patients at DLBCL risk at the beginning of the observation 0.2 FL period was ≥ 10. DLBCL, diffuse large MCL B-cell lymphoma; FL, follicular lym- MALT lymphoma phoma; MALT, mucosa-associated 0 50 100 150 200 lymphoid tissue; MCL, mantle-cell lym- phoma; NHL, non-Hodgkin lymphoma; Time (months) OS, overall survival. No. at risk: DLBCL 1,387 497 100 0 FL 544 284 65 1 1 MCL 172 70 8 0 MALT lymphoma 77 34 3 0 JCO Global Oncology 5 OS (probability) Pavlovsky et al TABLE 4. PFS of Patients by NHL Subtypes From Start of Frontline Treatment for 0.7-57.6; n = 74) for CHOP, and 5 months (range 0.6-21.6; the Three Most Commonly Used Regimens n = 14) for RCVP. For FL, patients treated with R-CHOP NHL Subtype/Treatment Total Patient 5-Year Cumulative Median PFS, achieved a median time from treatment initiation to relapse/ Regimen Number, No. PFS Rate (%) Years (95% CI) progression of 17.6 months (range , 0.1-74.4; n = 61), DLBCL whereas patients treated with CHOP or RCVP achieved R-CHOP 948 56.2 7.7 (5.8 to n.e.) medians of 14.5 months (range 0.5-83.3; n = 29) and 9.8 months (range 0.1-70.6; n = 26), respectively. CHOP 199 44.9 3.0 (1.9 to 6.1) RCVP 38 48.2 1.8 (0.8 to n.e.) For DLBCL, there were no statistically significant differ- ences in risk of death between patients treated with CHOP FL versus R-CHOP (HR = 1.330; P = not significant) or RCVP R-CHOP 259 65.1 NA versus R-CHOP (HR = 1.118; P = not significant). However, RCVP 91 51.0 5.9 (4.2 to n.e.) higher risk of death was observed for patients with FL CHOP 66 42.5 2.9 (1.7 to 6.9) treated with CHOP versus R-CHOP (HR = 2.508; P = .001), MCL although no differences were seen for RCVP versus R-CHOP 66 30.5 2.4 (1.3 to 3.0) R-CHOP (HR = 1.066; P = not significant). CHOP 28 20.4 1.9 (1.6 to 4.1) For patients with DLBCL, 5-year PFS rates for the three most common frontline regimens ranged from 44.9% to NOTE. The 5-year cumulative PFS and median PFS of patients with DLBCL 56.2%, with a median PFS of 7.7 years (95% CI, 5.8 to not (n = 1,185), FL (n = 416), MCL (n = 98), and MALT lymphoma (n = 42) from the estimable (n.e.); n = 948) for patients receiving R-CHOP start of first-line therapy treated with the top three most used regimens in each NHL (Table 4). Similarly, 5-year PFS rates among patients with subtype are as tabulated. Note that data for patients with MCL treated with hyper FL for the three most commonly used regimens were CVAD and for patients with MALT lymphoma are not available because of low sample sizes. 42.5%-65.1%. There was no relationship between type of Abbreviations: CHOP, cyclophosphamide, doxorubicin, vincristine, and frontline regimen prescribed and age at diagnosis with PFS, prednisone; CVAD, cyclophosphamide, vincristine, and doxorubicin; DLBCL, for either DLBCL or FL. diffuse large B-cell lymphoma; FL, follicular lymphoma; MALT, mucosa-associated OS/PFS From First Relapse lymphoid tissue; MCL, mantle-cell lymphoma; NA, not available; n.e., not estimable; NHL, non-Hodgkin lymphoma; PFS, progression-free survival; R-CHOP, Overall survival of patients with DLBCL, FL, MCL, or MALT rituximab, cyclophosphamide, doxorubicin, and vincristine; RCVP, rituximab, lymphoma from first relapse is shown in Figure 3. The cyclophosphamide, vincristine, and prednisone. median OS from first relapse was 6.6 years (95% CI, 3.1 to n.e.) for DLBCL, whereas median OS from first relapse was not reached for FL, MCL, or MALT lymphoma. DLBCL, FL, and MCL, respectively); however, data on the line of therapy were not available. The median PFS from first relapse was 2.3 years (95% CI, 0.4 to n.e.) for DLBCL and 2.6 years (95% CI, 1.1 to 3.9) for OS from Diagnosis by NHL Subtypes MCL. Median PFS was not reached in patients with FL or Analysis of cumulative survival rates from diagnosis for MALT lymphoma. Additional analyses on OS/PFS from NHL subtypes found that 5-year survival rates were highest second relapse are shown in Appendix Figs A1 and A2. for patients with MALT lymphoma (90.8%), followed by FL For DLBCL, the three most common treatments at first (87.6%), DLBCL (69.0%), and MCL (57.1%). Survival relapse were ifosfamide, carboplatin, and etoposide with/ rates for each NHL subtype were further stratified by without rituximab (RICE/ICE; 21.1%, n = 86 of 407); eto- country (Fig 1). poside, methylprednisolone, cytarabine, and cisplatin Among patients for whom International Prognostic Index (10.1%, n = 41 of 207); and R-CHOP (8.8%, n = 36 of (IPI) was available at diagnosis (n = 599), the 5-year survival 407). Patients treated with RICE/ICE after first relapse were rate was highest in patients with low IPI (87.3%) and lowest at significantly higher risk of death versus patients treated in patients with high IPI scores (45.6%; Table 3). with R-CHOP (HR = 3.115; P = .0035). Similarly, the risk of disease progression after first relapse was also higher with OS/PFS From Start of Frontline Treatment RICE/ICE versus R-CHOP (HR = 3.846; P = .0011). Overall survival of patients with DLBCL, FL, MCL, and MALT However, we were not able to identify the first-line treat- lymphoma were analyzed from the start of frontline therapy ments prescribed to those patients who used R-CHOP as (Fig 2). Patients with DLBCL age , 65 years at diagnosis second-line rescue therapy. were at lower risk of death compared with patients ≥ 65 years (hazard ratio [HR] = 0.732; P = .0161). DISCUSSION Of patients who experienced relapse or progression, the median time from treatment initiation to relapse/ Management of patients with NHL has changed sub- progression was 8.4 months (range , 0.1-92.6; stantially over the past two decades with improved diag- n = 236) for those treated with R-CHOP, 7.6 months (range nostic techniques, newer treatment options, and 6 © 2022 by American Society of Clinical Oncology Clinical Outcomes of B-cell NHL in Latin America 1.0 0.8 FIG 3. OS of patients by NHL subtypes from first relapse. Kaplan-Meier survival 0.6 curves depicting the OS of patients with DLBCL (n = 393), FL (n = 204), MCL (n = 94), and MALT lymphoma (n = 17) 0.4 from first relapse; the results are de- scribed when the number of patients at Censored risk at the beginning of the observation DLBCL 0.2 FL period was ≥ 10. DLBCL, diffuse large MCL B-cell lymphoma; FL, follicular lym- MALT lymphoma phoma; MALT, mucosa-associated lymphoid tissue; MCL, mantle-cell lym- 07 25 50 5 100 125 phoma; NHL, non-Hodgkin lymphoma; OS, overall survival. Time (months) No. at risk: DLBCL 393 149 72 26 8 0 FL 204 131 67 31 10 0 MCL 94 49 16 3 0 MALT lymphoma 17 9 2 0 consequently better treatment outcomes. However, RWE rate of 87.6%, which was higher than observed for other generation is necessary to better understand how these NHL subtypes (except MALT lymphoma). These observa- treatment improvements are implemented in individual tions are in line with previously reported improvements in countries. As RWE for NHL in Latin America is generally OS for patients with FL treated with rituximab-based 2-4 9,10 lacking, data presented here from the observational regimens, as observed by comparing OS of patients HOLA study provide a significant step in better under- diagnosed with FL during the rituximab era (1997-2003, standing treatment pathways and outcomes for patients in median OS not reached) with earlier decades (median OS of the region. Of note are the cohort size of the patients with 13.6-18.5 years). The results from our study also highlight NHL from seven Latin American countries (n = 2,967) and the necessity of detecting FL earlier in Latin America as the the long-term 8-year follow-up period. median time from diagnosis to treatment was 1.1 months (range 0-74.7) for nearly 80% of the FL patient cohort in the Among all NHL subtypes, DLBCL is the most common HOLA study, who were mostly diagnosed at advanced subtype reported globally (1,457, 57.9% in our study) and stages of their disease and required systemic therapy. is typically treated using combination chemotherapy in- cluding CHOP and R-CHOP. In our study, the 5-year MCL is an aggressive form of B-cell NHL. Relapsed MCL survival rates for DLBCL ranged from 45.6% in patients has been associated with poor responses and OS of , 3 with high IPI to 87.3% in patients with low IPI. Improved years. Little has been reported on treatment outcomes of outcomes were seen in our study when comparing patients with MCL in Latin America, with one retrospective R-CHOP versus CHOP, including 5-year PFS rates and study conducted in Mexico identifying only 12 patients. median PFS; however, there was no statistical difference The majority of patients with MCL in our study received in risk of death or disease progression between the reg- chemotherapy, where 41.3% of patients received R-CHOP imens. The reason for this is unclear, although may be due and 16.3% received CHOP. The 5-year survival rate for to the limitations of analyzing data from retrospective MCL was 57.1%, which compared poorly with rates for observational studies. Nonetheless, we found that pa- DLBCL and FL. The MCL cohort studied in the HOLA tients with DLBCL treated with RICE/ICE after first relapse registry is smaller compared with other subtypes, as would were at significantly higher risk of death and/or disease be expected given that MCL is a rarer form of NHL. Despite progression versus R-CHOP, thus signifying the impor- this, our study provides useful insights into the manage- tance of newer therapies in improving survival outcomes ment of the disease in this region, and greater international for patients. efforts should be considered to further understand out- comes of MCL in Latin America. FL accounts for around 20-30% of NHL cases in our study, with 51.3% of patients treated with chemotherapy receiving A cohort of 84 patients with MALT lymphoma were ana- R-CHOP. Patients with FL achieved a 5-year overall survival lyzed in our study. Given the indolent nature of the disease, JCO Global Oncology 7 OS (probability) Pavlovsky et al specifically identifying the watch and wait subgroup proved Collectively, caution should be exercised when interpreting challenging. However, the median time from diagnosis to our PFS and OS estimates, especially in comparison with treatment initiation reported for MALT lymphoma other studies. Finally, censored and not censored patients (2.3 months) could potentially be interpreted as such. may not have presented homogeneous characteristics at Additionally, we were unable to categorize further subsets the beginning of the study. of marginal zone lymphoma lymphomas, such as splenic or Overall, this study summarizes the treatment outcomes for extranodal subtypes, which may have been classified as patients with NHL in Latin America and highlights some others (n = 180). differences between participating countries which should Maintenance therapy is a known effective strategy in be considered with care because of sample numbers and prolonging remission for patients with NHL after induction heterogeneity of patient characteristics. Nonetheless, dis- therapy or stem-cell transplantation. For example, the parities in survival outcomes between the seven partici- clinical value of rituximab maintenance in treatment of FL pating countries (for example, survival rates for DLBCL, FL, has been demonstrated in several real-world studies in the and MCL were lowest in Brazil), potentially reflect regional 14 15 16 United States, Taiwan, and Czech Republic. Unfor- differences in patient access to timely treatment. Socio- tunately, data on the use of maintenance therapy in low- economic, geographic, and cultural disparities in the region grade lymphomas were not gathered in the HOLA study, can affect outcomes for hematological malignancies, with with the exception of Argentina where rituximab mainte- some of the key challenges including limited numbers of nance was reported as the most common treatment for hematologists and inadequate access to bone marrow DLBCL at first or at second relapse (n = 11 of 74, 14.9% transplant centers. Furthermore, although R-CHOP is the and n = 4 of 18, 22.2%, respectively). standard of care for DLBCL globally, nearly 15% of patients with DLBCL in our study were prescribed CHOP, which may There are certain limitations of this study, some of which are be indicative of late adoption of novel treatments in this common to retrospective observational studies. First, region. Furthermore, we found that management of missing data were observed for some variables because of second- and third-line patients was inconsistent, reflecting limited information available in patient records. Second, the lack of standard of care. although the total sample size was adequate, smaller sample sizes for some subgroups may have influenced In conclusion, the HOLA study is the first large scale, real- findings, particularly those related to outcomes from Cox world observational study to report treatment pathways and models. Third, PFS and OS analyses were affected by a clinical outcomes for patients with NHL in multiple countries large degree of loss to follow-up, especially from first and across Latin America. These data are important for informing second relapse, limiting the precision of those analyses. local improvements in NHL management and for under- PFS and OS analyses also assumed that patients with standing how treatment practices differ compared with other continued follow-up were representative of the original total regions in the world. Data gaps from this study highlight the patient cohort and may also be overly optimistic if, for real-world challenges of providing treatment and recording example, an out-of-hospital death resulted in loss to follow- data across the region, indicating a need for further studies up but was not documented in the patient’s medical chart. and support for treatment of NHL in Latin America. AFFILIATIONS SUPPORT ´ ´ ´ ´ Servicio de Hematologıa e Investigacion Clınica, Fundacion para The HOLA registry and this report were sponsored by Janssen-Cilag Combatir la Leucemia (FUNDALEU), Buenos Aires, Argentina Farmaceuti ˆ ca Ltda (São Paulo, Brazil). Assistance in manuscript CEPHO/ABC School of Medicine, Santo Andre, ´ Brazil preparation, according to the Strengthening the Reporting of Instituto Nacional de Ciencias Medicas y Nutricion Salvador Zubiran, ´ Observational Studies in Epidemiology guidelines, was provided by Wen Mexico City, Mexico Chean Lim (VMLY&R Health) with support from the study sponsor. Hospital Italiano La Plata, Buenos Aires, Argentina Centro Oncologico ´ Estatal ISSEMYM, Toluca, Mexico AUTHOR CONTRIBUTIONS Hospital de Cl´ınicas de Porto Alegre, Brazil Conception and design: Miguel Pavlovsky, Laura Fogliatto, Gerardo Hospital Angeles Lomas, Huixquilucan, Mexico Machnicki, Pamella Villanova Medical Solutions S.A. Guatemala City, Guatemala Provision of study materials or patients: Miguel Pavlovsky, Daniel Cubero, Janssen-Cilag Farmaceutica Ltda, Buenos Aires, Argentina Gladys Patricia Agreda-Vasquez, Alicia Enrico, Maria J. Mela-Osorio, Janssen-Cilag Farmaceutica Ltda, São Paulo, Brazil Jorge Armenta San Sebastian, Laura Fogliatto, Roberto Ovilla, Oscar Avendano CORRESPONDING AUTHOR Collection and assembly of data: Daniel Cubero, Gladys Patricia Agreda- Pamella Villanova, MD, Av. Presidente Juscelino Kubitscheck, 2041 São Vasquez, Alicia Enrico, Maria J. Mela-Osorio, Jorge Armenta San Paulo, CEP 04543-011 Brazil Complexo JK - Bloco B, Brazil; Sebastian, Laura Fogliatto, Roberto Ovilla, Oscar Avendano, Paula e-mail: pvillano@its.jnj.com. Barreyro Data analysis and interpretation: Miguel Pavlovsky, Gladys Patricia Agreda- ´ ´ Vasquez, Jorge Armenta San Sebastian, Laura Fogliatto, Gerardo Machnicki, Damila Trufelli, Pamella Villanova 8 © 2022 by American Society of Clinical Oncology Clinical Outcomes of B-cell NHL in Latin America Manuscript writing: All authors Gladys Patricia Agreda-Vasquez Final approval of manuscript: All authors Honoraria: Janssen, Roche Accountable for all aspects of the work: All authors Consulting or Advisory Role: Asofarma Gerardo Machnicki AUTHORS’ DISCLOSURES OF POTENTIAL CONFLICTS OF Employment: Janssen INTEREST Stock and Other Ownership Interests: Janssen The following represents disclosure information provided by authors of Paula Barreyro this manuscript. All relationships are considered compensated unless Employment: Janssen-Cilag otherwise noted. Relationships are self-held unless noted. I = Immediate Stock and Other Ownership Interests: Janssen-Cilag Family Member, Inst = My Institution. Relationships may not relate to the subject matter of this manuscript. For more information about ASCO’s Damila Trufelli conflict of interest policy, please refer to www.asco.org/rwc or ascopubs. Employment: Janssen Oncology org/go/authors/author-center. Leadership: Janssen Oncology Open Payments is a public database containing information reported by Stock and Other Ownership Interests: Janssen Oncology companies about payments made to US-licensed physicians (Open Honoraria: Janssen Oncology Payments). Pamella Villanova Miguel Pavlovsky Employment: Janssen Honoraria: AbbVie/Pharmacyclics, Roche, Novartis, Janssen, Stock and Other Ownership Interests: Janssen AstraZeneca Honoraria: Janssen Consulting or Advisory Role: AbbVie/Pharmacyclics, Roche, Novartis, Travel, Accommodations, Expenses: Janssen Janssen Speakers’ Bureau: AbbVie/Pharmacyclics, Roche, Novartis, Janssen No other potential conflicts of interest were reported. Travel, Accommodations, Expenses: Roche, Sanofi, AbbVie, Raffo REFERENCES 1. Curado MP, de Souza DL: Cancer burden in Latin America and the Caribbean. Ann Glob Health 80:370-377, 2014 2. Diumenjo MC, Abriata G, Forman D, et al: The burden of non-Hodgkin lymphoma in Central and South America. Cancer Epidemiol 44:S168-S177, 2016 (suppl 1) 3. Ortega V, Verastegui E, Flores G, et al: Non-Hodgkin’s lymphomas in Mexico. A clinicopathological and molecular analysis. Leuk Lymphoma 31:575-582, 1998 4. Muller AM, Ihorst G, Mertelsmann R, et al: Epidemiology of non-Hodgkin’s lymphoma (NHL): Trends, geographic distribution, and etiology. Ann Hematol 84:1-12, 2005 5. Bray F, Colombet M, Mery L, et al: Cancer Incidence in Five Continents, Vol. XI. IARC Scientific Publication No. 166. Lyon: International Agency for Research on Cancer. https://publications.iarc.fr/597 6. Chiattone C, Gomez-Almaguer D, Pavlovsky C, et al: Results from Hemato-Oncology Latin America Observational Registry (HOLA) providing real world outcomes for the treatment of patients with CLL. Blood 128:5578, 2016 7. Tietsche De Moraes Hungria V, Chiattone C, Pavlovsky M, et al: Epidemiology of hematologic malignancies in real-world settings: Findings from the Hemato- Oncology Latin America Observational Registry study. JCO Glob Oncol 5:1-19, 2019 8. de Moraes Hungria VT, Martınez-Baños DM, Peñafiel CR, et al: Multiple myeloma treatment patterns and clinical outcomes in the Latin America Haemato- Oncology (HOLA) Observational Study, 2008–2016. Br J Haematol 188:383-393, 2020 9. Tan D, Horning SJ, Hoppe RT, et al: Improvements in observed and relative survival in follicular grade 1-2 lymphoma during 4 decades: The Stanford University experience. Blood 122:981-987, 2013 10. Provencio M, Sabin P, Gomez-Codina J, et al: Impact of treatment in long-term survival patients with follicular lymphoma: A Spanish Lymphoma Oncology Group Registry. PLoS One 12:e0177204, 2017 11. Schieber M, Gordon LI, Karmali R: Current overview and treatment of mantle cell lymphoma. F1000Res 7:F1000 Faculty Rev-1136, 2018 ´ ´ ´ ´ 12. Cordova-Ramırez AC, Sanchez-Valledor LF, Colon-Otero G, et al: Mantle cell lymphoma may have a different clinical course in Mexican Mestizos: Real-world data from a single center. Rev Invest Clin 73(2):94-9, 2021 13. Raderer M, Kiesewetter B, Ferreri AJM: Clinicopathologic characteristics and treatment of marginal zone lymphoma of mucosa-associated lymphoid tissue (MALT lymphoma). CA Cancer J Clin 66:152-171, 2016 14. Nastoupil LJ, Sinha R, Byrtek M, et al: The use and effectiveness of rituximab maintenance in patients with follicular lymphoma diagnosed between 2004 and 2007 in the United States. Cancer 120:1830-1837, 2014 15. Huang H-H, Wen Y-C, Chen H-M, et al: Rituximab maintenance improves overall survival in follicular lymphoma: A retrospective nationwide real-world analysis from Taiwan Cancer Registry Database. Cancer Med 7:3582-3591, 2018 16. Belada D, Prochazka V, Janikova A, et al: The influence of maintenance therapy of rituximab on the survival of elderly patients with follicular lymphoma. A retrospective analysis from the database of the Czech Lymphoma Study Group. Leuk Res 73:29-38, 2018 17. Goss PE, Lee BL, Badovinac-Crnjevic T, et al: Planning cancer control in Latin America and the Caribbean. Lancet Oncol 14:391-436, 2013 18. Tilly H, Gomes da Silva M, Vitolo U, et al: Diffuse large B-cell lymphoma (DLBCL): ESMO clinical practice guidelines for diagnosis, treatment and follow-up. Ann Oncol 26:v116-v125, 2015 (suppl 5) nn n JCO Global Oncology 9 Pavlovsky et al APPENDIX 1.0 0.8 0.6 0.4 Censored 0.2 DLBCL FL MCL 08 20 40 600 100 Time (months) No. at risk: DLBCL 147 44 24 9 1 1 FL 66 37 19 10 4 0 MCL 36 14 10 3 0 FIG A1. OS of patients by NHL subtypes from second relapse. Kaplan-Meier survival curves depicting the OS of patients with DLBCL (n = 147), FL (n = 66), and MCL (n = 36) from second relapse; the results are described when the number of patients at risk at the beginning of the observation period was ≥ 10. In total, 152 patients with DLBCL, 70 patients with FL, 38 patients with MCL, and 3 patients with MALT lymphoma received treatment for second relapse. For DLBCL, the median OS from second relapse was 1.9 years (95% CI, 0.43 to n.e.), whereas the median OS from second relapse for FL or MCL was 5.0 years (95% CI, 1.9 to n.e.) or 4.4 years (95% CI, 0.8 to n.e.), respectively. Outcome analysis was not conducted for MALT lymphoma because of small patient sample size. DLBCL, diffuse large B-cell lymphoma; FL, follicular lymphoma; MALT, mucosa-associated lymphoid tissue; MCL, mantle-cell lymphoma; n.e., not estimable; NHL, non-Hodgkin lymphoma; OS, overall survival. 10 © 2022 by American Society of Clinical Oncology OS (probability) Clinical Outcomes of B-cell NHL in Latin America 1.0 0.8 0.6 0.4 Censored 0.2 DLBCL FL MCL 020 40 60 80 Time (months) No. at risk: DLBCL 145 27 12 3 0 FL 64 29 12 5 0 MCL 35 8 2 0 FIG A2. PFS of patients by NHL subtypes from second relapse. Kaplan-Meier survival curves depicting the PFS of patients with DLBCL (n = 145), FL (n = 64), and MCL (n = 35) from second relapse; the results are described when the number of patients at risk at the beginning of the observation period was ≥ 10. The median PFS from second relapse was 1.3 years (95% CI, 0.4 to 6.2) in DLBCL, 4.3 years (95% CI, 1.8 to n.e.) in FL, and 1.5 years (95% CI, 0.8 to 4.4) in MCL. DLBCL, diffuse large B-cell lymphoma; FL, follicular lymphoma; MCL, mantle-cell lymphoma; n.e., not estimable; NHL, non-Hodgkin lymphoma; PFS, progression-free survival. JCO Global Oncology 11 PFS (probability)

Journal

JCO: Global Oncology – Wolters Kluwer Health

Published: Apr 29, 2022

There are no references for this article.