Received: 24 May 2021 | Revised: 6 April 2022 | Accepted: 20 April 2022 DOI: 10.1002/rth2.12736 ORIGINAL ARTICLE Enoxaparin versus dalteparin or tinzaparin in patients with cancer and venous thromboembolism: The RIETECAT study Javier Trujillo-­Santos PhD1 | Dominique Farge-­Bancel PhD2 | José María Pedrajas PhD3 | Covadonga Gómez-­Cuervo PhD4 | Aitor Ballaz PhD5 | Andrei Braester MD6 | 7 8 9 Isabelle Mahé PhD | Aurora Villalobos PhD | José Antonio Porras PhD | Manuel Monreal MD, PhD10 | The RIETE Investigators 1 Department of Internal Medicine, Hospital General Universitario Santa Lucía, Universidad Católica de Murcia, Murcia, Spain 2 Unité de Médecine Interne, Maladies Auto-­immunes et Pathologie Vasculaire (UF 04), IRSL, EA-­3518, Université de Paris, Paris, France 3 Department of Internal Medicine, Hospital Clínico San Carlos, Madrid, Spain 4 Department of Internal Medicine, Hospital Universitario 12 de Octubre, Madrid, Spain 5 Department of Pneumonology, Hospital de Galdakao, Vizcaya, Spain 6 Department of Haematology, Azrieli Faculty of Medicine, Bar-­Ilan University, Safed, Israel 7 Department of Internal Medicine, Hôpital Louis Mourier, Colombes (APHP), University Paris, France 8 Department of Internal Medicine, Hospital Regional Universitario de Málaga, Málaga, Spain 9 Department of Internal Medicine, Hospital Universitario Joan XXIII de Tarragona, Tarragona, Spain 10 Department of Internal Medicine, Hospital Germans Trias i Pujol, Badalona, Universidad Católica de Murcia, Universidad Autónoma de Barcelona, Badalona, Spain Correspondence Manuel Monreal, Department of Internal Abstract Medicine, Hospital Universitari Germans Background: Venous thromboembolism (VTE) is a frequent complication in patients Trias i Pujol, Badalona, Universidad Católica de Murcia, Spain. with cancer and a leading cause of morbidity and death. Email: manuelmonrealbosch@gmail.com Objectives: The objective of the RIETECAT study was to compare the long-­term ef- Funding information fectiveness and safety of enoxaparin versus dalteparin or tinzaparin for the secondary Sanofi prevention of VTE in adults with active cancer. Handling Editor: Dr Suzanne Cannegieter Methods: We used the data from the multicenter, multinational RIETE registry to com- pare the rates of VTE recurrences, major bleeding, or death over 6 months in patients with active cancer and acute VTE using full doses of enoxaparin versus dalteparin or tinzaparin, and a multivariable Cox proportional hazard model was used to analyze the primary end point. Results: From January 2009 to June 2018, 4451 patients with active cancer received full doses of the study drugs: enoxaparin, 3526 patients; and dalteparin or tinzapa- rin, 925 (754 + 171) patients. There was limited difference in VTE recurrences (2.0% vs 2.5%) and mortality rate (19% vs 17%) between the enoxaparin and dalteparin or A full list of the RIETE investigators is given in the Appendix A1. This is an open access article under the terms of the Creative Commons Attribution-­NonCommercial-­NoDerivs License, which permits use and distribution in any medium, provided the original work is properly cited, the use is non-­commercial and no modifications or adaptations are made. © 2022 The Authors. Research and Practice in Thrombosis and Haemostasis published by Wiley Periodicals LLC on behalf of International Society on Thrombosis and Haemostasis (ISTH). Res Pract Thromb Haemost. 2022;6:e12736.  wileyonlinelibrary.com/journal/rth2 | 1 of 10 https://doi.org/10.1002/rth2.12736 2 of 10 | TRUJILLO-­SANTOS et al. tinzaparin subgroups. However, there was a slight numerical increase in major bleeding (3.1% vs 1.9%). Propensity score matching confirmed that there were no differences in the risk for VTE recurrences (adjusted hazard ratio [aHR], 0.81; 95% confidence interval [CI], 0.48-­1.38), major bleeding (aHR, 1.40; 95% CI, 0.80-­2.46), or death (aHR, 1.07; 95% CI, 0.88-­1.30) between subgroups. Conclusions: In RIETECAT, in patients with cancer and VTE receiving full-­dose enoxa- parin or dalteparin or tinzaparin, no statistically significant differences were observed regarding effectiveness and safety outcomes over a 6-­month period. KEYWORDS cancer, cohort, dalteparin, enoxaparin, LMWH, recurrences, tinzaparin, venous thromboembolism Essentials • Patients with cancer and venous thromboembolism (VTE) should receive long-­term anticoagulant therapy. • RIETECAT compared the effectiveness and safety of enoxaparin versus dalteparin or tinzaparin. • There was no statistical difference between treatments in recurrent VTE. • There was no statistical difference between treatments in bleeding complications or death. 1 | I NTRO D U C TI O N largest database of patients with VTE. The RIETECAT study is an analysis of patients with active cancer from RIETE presenting with Venous thromboembolism (VTE) is a frequent complication in pa- VTE and receiving initial and long-­term therapy with full doses of tients with cancer and a leading cause of death, morbidity, delays in enoxaparin versus dalteparin or tinzaparin. The information on care, and increased costs.1-­6 Current practice guidelines recommend, LMWH regimen (once daily or twice daily) was introduced in RIETE on the basis of data from randomized trials, that patients with can- on January 1, 2009. Thus, the study included patients from January cer and VTE receive long-­term therapy with low-­molecular-­weight 2009 to June 2018. Full doses were defined as (i) enoxaparin: 1 mg/ heparin (LMWH) over vitamin K antagonists (VKAs) or direct oral kg ± 20% twice daily or 1.5 mg/kg ± 20% once daily for initial and anticoagulants (DOACs).7-­10 At the time of this study, in Europe, only long-­term therapy; (ii) dalteparin: 200 IU/kg ± 20% once daily during dalteparin and tinzaparin have this specific indication for patients the first month, and then 150 IU/kg ± 20% once daily; and (iii) tinza- with cancer mentioned in their label. Enoxaparin, another LMWH, parin: 175 IU/kg ± 20% once daily for initial and long-­term therapy. is also used in patients with cancer and VTE, although at the time it Patients were assessed for up to 6 months following the index date was not approved for this indication. or until the first individual occurrence of each clinical outcome (VTE The efficacy and safety of enoxaparin versus VKAs for the long-­ recurrences, major bleeding, nonmajor bleeding of clinical signifi- term therapy of VTE in patients with cancer was demonstrated in cance, death) or due to loss during the 6-­month period. The baseline two randomized controlled trials.11,12 In the real-­world setting, how- characteristics and treatment exposures were reported to describe ever, the effectiveness and safety of enoxaparin compared with the population included in RIETECAT. The rates of VTE recurrences, dalteparin or tinzaparin in patients with cancer remains unexplored. bleeding, and death were compared over a period of 6 months after Thus, we used the data in the RIETE (Registro Informatizado de la the index VTE. Enfermedad TromboEmbólica) registry, to compare the effective- ness and safety of enoxaparin versus dalteparin or tinzaparin in pa- tients with active cancer and acute VTE over a 6-­month period. 2.2 | Patient selection Consecutive patients were included who had an acute episode of 2 | M E TH O D S symptomatic, objectively confirmed VTE between January 1, 2009, and June 30, 2018, and fulfilled the following criteria: age ≥18 years; 2.1 | Study design active cancer (defined by a histological or cytological confirmation of malignancy and at least one of the following features: cancer diagno- RIETE (clinicaltrials.gov identifier: NCT02832245) is an ongoing pro- sis within 6 months before VTE, metastatic disease or hematologi- spective, multicenter, multinational registry of consecutive patients cal malignancy not in complete remission, or treatment for cancer 13 with objectively confirmed VTE. To date, RIETE is the world’s during the previous 6 months); start of treatment with full-­dose TRUJILLO-­SANTOS et al. | 3 of 10 enoxaparin, dalteparin, or tinzaparin within the first 48 hours after hazard ratio (HR) was lower than 1.5. It was assumed that the HR was VTE diagnosis. constant over the study period. Cox proportional hazards models To be enrolled in the registry, patients needed to have had an were used to analyze the data. Secondary effectiveness outcomes acute deep vein thrombosis (DVT), pulmonary embolism (PE), or were the individual components of VTE recurrences, that is, symp- both, confirmed by objective tests (ie, contrast venography or com- tomatic, objectively confirmed DVT recurrences and PE recurrences pression ultrasonography for suspected DVT, helical computed (fatal and nonfatal). The secondary objective of RIETECAT was to tomography scan, ventilation-­perfusion lung scintigraphy, or pulmo- compare the safety of enoxaparin versus dalteparin or tinzaparin in nary angiography for suspected PE). adult patients with active cancer over a 6-­month period. Secondary Patients were excluded from the RIETECAT study if they safety outcomes were major bleeding events (fatal or nonfatal), all-­ had: (i) prior VTE <12 months before the index event; (ii) started cause death, fatal PE, fatal bleeding, and nonmajor but clinically rel- LMWH therapy >48 hours after the index VTE; (iii) started enox- evant (NMCR) bleeding. Only bleeding events that occurred within aparin, dalteparin, or tinzaparin but switched to other drugs be- 24 hours after the last dose of the study drug (enoxaparin, daltepa- fore day 90; or (iv) started enoxaparin, dalteparin, or tinzaparin rin, or tinzaparin) were considered. Major bleeding was defined as at full doses, but then switched to other doses before day 90, in bleeding events that were overt and required a transfusion of ≥2 the absence of VTE recurrences or bleeding events. Patients who units of blood; were retroperitoneal, spinal, intracranial, intrathecal, started on enoxaparin, dalteparin, or tinzaparin at full doses but intrapericardial, or intraocular; or were fatal. NMCR bleeding were then switched to nonfull doses or other anticoagulants before day those overt bleeds not meeting criteria for major bleeding but re- 90 in the absence of VTE recurrences or bleeding events were quiring medical assistance. Fatal bleeding was defined as any death excluded from the main study but were included in a sensitivity occurring within 10 days of a major bleeding episode, in the absence analysis. of an alternative cause of death. Fatal PE was defined as any death occurring within 10 days of a PE episode (either the index event or recurrent PE) in the absence of an alternative cause of death. 2.3 | Data elements Variables routinely collected in RIETE included baseline character- 2.5 | Statistical analyses istics (sex, age, body weight); initial VTE presentation (proximal-­, bilateral-­ or upper-­limb DVT); systolic blood pressure levels; heart Patient characteristics at the time of initial VTE diagnosis were rate; blood oxygen saturation levels (in patients with PE); cancer described and compared across the two exposure categories. characteristics (location, time since cancer diagnosis, presence of Differences across subgroups were assessed using a t test for metastases, and oncologic therapy); additional risk factors for VTE continuous variables, and a chi-­s quare test for categorical vari- (recent surgery within 2 months before index VTE, recent immobil- ables. For each treatment exposure category, the proportion of ity with bathroom privileges for >4 days within 2 months, estrogen patients with primary or secondary outcomes at 6 months was use, pregnancy or postpartum, and personal history of VTE); comor- obtained. Multivariable analyses were performed. A multivariable bidities and blood tests at baseline (including anemia, leukocyte and Cox proportional hazard model was used to analyze the primary platelet count, and creatinine clearance [CrCl] levels); and concomi- end point. Crude and adjusted HRs, and corresponding 95% CIs tant therapies, including antiplatelet drugs and corticosteroids. of the primary and secondary outcomes were estimated using Cox proportional hazards models. Treatment exposure served as a time-­d ependent variable. Models were adjusted for the follow- 2.4 | Study objectives and outcomes ing covariates: sex, age, body weight, initial presentation of VTE (unstable PE, stable PE, or DVT), location of cancer (according to The primary objective of RIETECAT was to assess the noninferior- Khorana score), metastases, treatment for cancer, recent immobil- ity of enoxaparin versus dalteparin or tinzaparin in preventing VTE ity, recent surgery, chronic heart or lung disease, recent (<30 days recurrences in adult patients with active cancer over a 6-­month prior) major bleeding, concomitant therapy with antiplatelets and/ period. VTE recurrences were defined as composite outcomes of or corticosteroids, anemia, leukocyte count, platelet count, CrCl symptomatic, objectively confirmed DVT and fatal or nonfatal PE. levels, size of hospital (>500 beds, 250-­5 00 beds, or <250 beds), A noninferiority margin of 1.5 was determined based on a review of and year of diagnosis. the literature and inputs from clinical experts. With an overall sam- In addition, an analysis was conducted on the primary end ple size of 4451 subjects, and considering the expected VTE recur- point, using propensity score matching to adjust for differences rence rate in the reference group was 2.5% at 6 months, the power in the baseline characteristics of patients exposed to enoxaparin to show noninferiority with the margin of 1.5 was 80%, using a one-­ versus dalteparin or tinzaparin (same covariates as those used in sided test with a 0.03 significance level. Noninferiority of enoxa- the Cox model of the primary analysis). As no adjustments for mul- parin versus dalteparin or tinzaparin on VTE incidence would be tiplicity were made for secondary evaluation variables, all tests demonstrated if the upper limit of 95% confidence interval (CI) of the were exploratory. Statistical analyses were conducted using IBM 4 of 10 | TRUJILLO-­SANTOS et al. SPSS Statistics for Windows version 25.0 (IBM Corp., Armonk, NY, ≥4 days (16%) or to receive corticosteroids (19%) or antiplatelet drugs USA). (16%) concomitantly than patients receiving dalteparin or tinzaparin Similar analyses were conducted for the sensitivity analysis. (12%, 13%, and 16%, respectively) but less likely to have had prior VTE Patient characteristics at baseline were compared across the (7.5% compared with 9.6% in the dalteparin or tinzaparin subgroup). two exposure categories; for each treatment exposure category, Comorbidities did not differ between the two subgroups re- the proportion of patients with primary or secondary outcomes garding chronic heart failure, chronic lung disease, gastrointesti- at 6 months was obtained, and a multivariable Cox proportional nal diseases, or current anemia. In each subgroup, most patients hazard model was used to analyze the primary end point. Crude (97.9% vs 98.6%) had no recent history of major bleeding (within and adjusted HRs, and corresponding 95% CIs of the primary and 1 month before index VTE). Comorbidities that might influence secondary outcomes were estimated using Cox proportional haz- coagulation were reported, and there were no significant differ- ards models. Treatment exposure served as a time-­d ependent ences between the two treatment subgroups regarding the inci- variable. Models were adjusted for the same covariates as those dence of patients with liver cirrhosis (29 [0.82%] vs 8 [0.86%]), listed above. chronic liver disease (41 [1.2%] vs 11 [1.2%]), chronic thrombocy- topenia (5 [0.14%] vs 2 [0.22%]), or antiphospholipid syndrome (5 [0.14%] vs 1 [0.11%]). 3 | R E S U LT S From March 2001 to June 2018, 84 918 patients with VTE were 3.2 | Cancer characteristics recruited in RIETE. Of these, 21 234 had active cancer. The use of LMWH for the long-­term therapy of VTE in patients with active can- The majority (≈92%) of patients in both subgroups had solid cancers. cer progressively increased from 40% to 80%, the use of VKAs de- There were no differences in cancer sites, except for breast can- creased, and the use of DOACs increased from zero to 10% (Figure cer, which was slightly less likely in patients receiving enoxaparin. S1A). Among all LMWHs, enoxaparin accounted for the largest The median time elapsed from cancer diagnosis to VTE was slightly proportion (Figure S1B). Over 50% of patients in Europe received shorter in patients on enoxaparin (4 vs 5 months; Table 2). LMWH as long-­term therapy for VTE (Table S1). Its use in other countries also reached a high proportion, except in Brazil, where VKAs accounted for 93% of use. 3.3 | Treatment characteristics In total, 8167 patients with active cancer and VTE were diag- nosed from January 2009 to June 2018. For the current study, For initial therapy, the median duration was 8 days in patients we excluded 334 patients (4.1%) who received long-­term ther- on enoxaparin and 27 days in those on dalteparin or tinzaparin, apy with other drugs, 408 (5.0%) with no information on LMWH until the regimen or doses changed (Table 3). Mean daily doses doses or regimen, 664 (8.1%) who started with LMWH beyond were 195 ± 16 IU/kg/d in patients on twice-­daily enoxaparin, the first 48 hours, and 870 (10.6%) who switched from LMWH 148 ± 15 IU/kg/d in those on once-­daily enoxaparin, 176 ± 22 IU/ to other drugs (mostly VKAs) within the first 3 months (Figure 1). kg/d in those on once-­daily dalteparin, and 177 ± 14 IU/kg/d in Thus, there were 5891 patients: 4704 (80%) received enoxaparin, those on once-­daily tinzaparin. Overall, 34% of patients on enoxa- 257 (4.4%) received dalteparin, and 930 (16%) received tinzapa- parin and 21% of patients on dalteparin or tinzaparin had changed rin. Among these, we excluded 1440 patients (18%) who did not their daily dose during the 6-­m onth period (Table 3). Most of them receive the recommended doses (±20%) or regimen of LMWH experienced a one-­t ime dose decrease (patients were on initial (Table S2). Thus, the main study analysis included 4451 patients: therapy and switched to long-­term therapy). Moreover, 78% of pa- 3526 treated with enoxaparin, 171 treated with dalteparin, and tients receiving enoxaparin (and all patients on dalteparin or tin- 754 treated with tinzaparin. zaparin) maintained the initial regimen (once daily alone or twice daily alone). In patients whose regimen was modified, the majority was a switch from twice-­daily to once-­daily injections, while very 3.1 | Baseline characteristics few had a switch from once-­daily to twice-­daily injections or expe- rienced multiple changes. Patients receiving enoxaparin were more likely to initially present with PE (58%) than those on dalteparin or tinzaparin (44%) (Table 1). Among patients presenting with PE, those receiving enoxaparin were 3.4 | Clinical outcomes more likely to have tachycardia (36%) than those on dalteparin or tin- zaparin (30%). Among patients initially presenting with DVT, those In the 6 months following the acute VTE, of a total of 4451 patients, on enoxaparin were more likely to have proximal lower-­limb DVT 93 patients presented with VTE recurrences (PE recurrences, 47; (80%) than those on dalteparin or tinzaparin (62%). Patients receiv- DVT recurrences, 47), 129 had major bleeding (intracranial, 21), 111 ing enoxaparin were also more likely to have been immobilized for had NMCR bleeding, and 823 died (fatal PE, 39; fatal bleeding, 20). TRUJILLO-­SANTOS et al. | 5 of 10 F I G U R E 1 Flowchart of patients. DVT, deep vein thrombosis; LMWH, low-­ molecular-­weight heparin; PE, pulmonary embolism; VTE, venous thromboembolism The rate of recurrent VTE was 2.0% (n = 70/3526) in patients recurrences (aHR, 0.83; 95% CI, 0.41-­1.70; P = .62), PE recurrences receiving enoxaparin and 2.5% (n = 23/925) in the dalteparin or (aHR, 0.79; 95% CI, 0.41-­1.56; P = .51), nonmajor bleeding of clini- tinzaparin subgroup (Table 4). Similarly, there was limited differ- cal significance (aHR, 0.88; 95% CI, 0.55-­1.41; P = 0.60) or all-­cause ence in DVT recurrences between the enoxaparin and dalteparin or death (aHR, 0.97; 95% CI, 0.81-­1.16; P = .77). The rate of major tinzaparin subgroups (1.0% vs 1.2%) or PE recurrences between the bleeding was nonsignificantly higher in patients treated with enoxa- enoxaparin and dalteparin or tinzaparin subgroups (0.99% vs 1.3%) parin (aHR, 1.52; 95% CI, 0.90-­2.58; P = 0.12; Table 5). (Table 4). The rate of major bleeding was numerically higher in the Results of the propensity score matching involved 1662 pa- enoxaparin subgroup compared with the dalteparin or tinzaparin tients on enoxaparin and 903 patients on dalteparin or tinzaparin. subgroup (3.1% vs 1.9%). There were no differences in the rates of The matched analysis revealed no significant differences in the risk NMCR bleeding (2.5% vs 2.6%) or all-­cause death (19% vs 17%) be- for DVT recurrences (aHR, 0.90; 95% CI, 0.43-­1.90; P=0.78), PE tween the two subgroups. Similarly, there was no difference in the recurrences (aHR, 0.73; 95% CI, 0.35-­1.55; P = 0.42), major bleed- rates of the composite of fatal PE or fatal bleeding (1.4% vs 1.2%) be- ing (aHR, 1.40; 95% CI, 0.80-­2 .46; P = 0.24), nonmajor bleeding tween the treatment subgroups (Table 4). The rates of major bleed- of clinical significance (aHR, 0.88; 95% CI, 0.53-­1.49; P = .64), or ing in specific sites are given in Table S3. death (aHR, 1.07; 95% CI, 0.88-­1.30; P = .48) between subgroups. 3.5 | Multivariable analysis 3.6 | Sensitivity analysis On the adjusted analyses, there were no differences between treat- The sensitivity analysis had a total of 5128 (n = 4099 enoxaparin; ment subgroups in terms of VTE recurrences with enoxaparin meet- n = 1029 dalteparin/tinzaparin) patients, which included 677 of the ing the prespecified criterion for noninferiority of 1.5 (adjusted HR 1440 excluded patients who started on full-­dose LMWH but were [aHR], 0.82; 95% CI, 0.50-­1.34; P = .008 for noninferiority), DVT then moved to another anticoagulant treatment or to a nonfull dose 6 of 10 | TRUJILLO-­SANTOS et al. TA B L E 1 Patient demographics, clinical characteristics, TA B L E 2 Cancer characteristics and underlying conditions at baseline in patients receiving recommended doses of LMWH Dalteparin or Enoxaparin tinzaparin Dalteparin or Patients, N 3526 925 Enoxaparin tinzaparin Location, n (%) Patients, N 3526 925 Lung 639 (18) 151 (16) Demographics Colorectal 478 (14) 134 (15) Male sex, n (%) 1670 (47) 448 (48) Breast 433 (12) 140 (15) Age (mean years ± SD) 68 ± 13 67 ± 13 Prostate 291 (8.3) 62 (6.7) Body weight (mean kg ±SD) 73 ± 14 74 ± 14 Hematological 277 (7.9) 70 (7.6) a Race/ethnicity , n (%) Bladder 183 (5.2) 49 (5.3) White 1315 (37) 292 (32) Pancreas 178 (5.0) 50 (5.4) Latino 39 (1.1) 5 (0.5) Brain 147 (4.2) 29 (3.1) Asian 19 (0.5) 3 (0.3) Gastric 130 (3.7) 24 (2.6) Arab 4 (0.1) 4 (0.4) Uterine 129 (3.7) 43 (4.6) Mixed/other 15 (0.4) 6 (0.6) Ovary 120 (3.4) 41 (4.4) Not provided 2134 (61) 615 (66) Kidney 76 (2.2) 19 (2.1) Initial VTE presentation, n (%) Oropharynx/larynx 62 (1.8) 18 (1.9) Carcinoma of unknown 57 (1.6) 14 (1.5) PE 2058 (58) 411 (44) origin SBP levels <100 mm Hg 188 (9.1) 32 (7.8) Esophagus 39 (1.1) 10 (1.1) Heart rate >100 bpm 727 (36) 122 (30) Others 287 (8.1) 71 (7.7) Sat O2 levels <90% 298 (29) 43 (24) Time since cancer diagnosis Isolated DVT 1468 (42) 514 (56) Mean months ±SD 20 ± 42 23 ± 45 Proximal 1178 (80) 320 (62) Cancer stage, n (%) Bilateral lower limb 69 (4.7) 25 (4.9) With metastases 1845 (52) 504 (54) Upper limb 309 (21) 198 (38) Current cancer therapy, n (%) Comorbidities, n (%) Radiotherapy 496 (14) 107 (12) Chronic heart disease 185 (5.2) 50 (5.4) Chemotherapy 1540 (44) 525 (57) Chronic lung disease 443 (13) 99 (11) Hormonal 410 (12) 112 (12) Chronic liver disease 100 (2.8) 22 (2.4) Other 40 (1.1) 9 (0.97) Atrial fibrillation 115 (3.3) 32 (3.5) None 1040 (29.0) 172 (19) Recent (<30 days) major 75 (2.1) 13 (1.4) Abbreviations: SD, standard deviation. bleeding Anemia 1442 (41) 351 (38) 3 before day 90, and the 4451 patients from the primary analysis. Platelet count <100,000/mm 174 (4.9) 44 (4.8) The 677 patients consisted of 279 patients on enoxaparin and 21 CrCl levels ˂30 mL/min 142 (4.0) 29 (3.1) patients on dalteparin or tinzaparin who were prescribed recom- Additional risk factors for VTE mended doses within the first 48 hours after VTE diagnosis but were Postoperative 486 (14%) 117 (13%) then transferred to an alternative treatment, and 294 patients on Recent immobility ≥4 days 579 (16%) 114 (12%) enoxaparin and 83 patients on dalteparin or tinzaparin who received Prior VTE 266 (7.5%) 89 (9.6%) recommended doses within the first 48 hours after VTE diagnosis Concomitant medications, n (%) but were then switched to nonfull doses. The 6-­month outcomes Corticosteroids 613 (19) 135 (16) observed with this larger cohort were similar to those observed in the primary analysis. The rate of recurrent VTE was 2.1% in the Antiplatelets 540 (16) 113 (13) enoxaparin group and 2.4% in the dalteparin or tinzaparin subgroup, Abbreviations: bpm, beats per minute; CrCl, creatinine clearance; and the rate of major bleeding was 2.8% in the enoxaparin subgroup DVT, deep vein thrombosis; LMWH, low-­molecular-­weight heparin; and 1.7% in the dalteparin or tinzaparin subgroup. The results from PE, pulmonary embolism; SBP, systolic blood pressure; SD, standard deviation; VTE, venous thromboembolism. the adjusted analyses were also similar to those from the primary a Information on race/ethnicity was incorporated in RIETE on April analysis and similar between the subgroups in terms of VTE recur- 2014. rences (aHR, 0.90; 95% CI, 0.57-­1.41), DVT recurrences (aHR, 0.90; TRUJILLO-­SANTOS et al. | 7 of 10 TA B L E 3 Treatment characteristics Dalteparin or Enoxaparin tinzaparin Patients, N 3526 925 Switching LMWH doses Yes, n (%) 1215 (34) 197 (21) Number of days until switch Mean days ±SD 17 ± 27 28 ± 28 Median days (min-­max) 8 (1-­178) 27 (1-­150) Changes of daily doses, n (%) Increase in doses 225 (6.4) 46 (5.0) Decrease in doses 1095 (31) 172 (19) Both increase and decrease 105 (3.0) 21 (2.3) Number of doses switch 1 1027 (29) 173 (19) 2 134 (3.8) 18 (1.9) ≥3 54 (1.5) 6 (0.65) Regimen, n (%) Once daily 891 (25) 925 (100) Twice daily 1856 (53) 0 Regimen modification Twice daily to once daily 687 (20) 0 Once daily to twice daily 42 (1.2) 0 Multiple dosing changes 50 (1.4) 0 Different drugs 0 49 (5.3) Abbreviations: LMWH, low-­molecular-­weight heparin; SD, standard deviation. 95% CI, 0.46-­1.76), PE recurrences (aHR, 0.92; 95% CI, 0.49-­1.73), the contrary, patients receiving dalteparin or tinzaparin had more nonmajor bleeding of clinical significance (aHR, 0.86; 95% CI, 0.56-­ consistent treatment regimens following their respective labels. 1.33) or all-­cause death (aHR, 1.06; 95% CI, 0.89-­1.27). The rate of There were unbalanced patient numbers (with a ratio of 4:1 for major bleeding was slightly higher in patients treated with enoxapa- enoxaparin versus dalteparin or tinzaparin) and unbalanced charac- rin (aHR, 1.57; 95% CI, 0.93-­2.66; Tables S4-­S8). teristics at baseline (in the enoxaparin subgroup there was a higher proportion of patients initially presenting with PE, and more prox- imal DVT). Moreover, patients on enoxaparin were more likely to 4 | DISCUSSION have been immobilized or to be using corticosteroids or antiplate- let drugs concomitantly. Following multivariable analysis, the risks To our knowledge, this is the largest study comparing the effec- for VTE recurrences, NMCR bleeding, and death between the two tiveness and safety of different LMWHs in a real-­world cohort of treatment subgroups (enoxaparin vs the other LMWHs) were com- patients with cancer and VTE. In our cohort, 25% of patients on parable. There was no statistically significant difference between enoxaparin, 12% on dalteparin, and 14% on tinzaparin were not pre- the subgroups for major bleeding, although there was a tendency to scribed full doses, and thus were excluded from the study. Our find- numerically more major bleeds in the enoxaparin subgroup. While an ings reveal that in real-­life clinical practice, most physicians followed increased risk of major bleeding cannot be excluded based on these label instructions of dalteparin or tinzaparin to treat their patients, study results, there were differences in baseline characteristics be- but enoxaparin (which was lacking guidance for the long-­term ther- tween the subgroups that may explain this finding. The variability of apy of VTE in patients with active cancer) was used variably. This is the treatment regimens in the enoxaparin subgroup compared with important since LMWHs accounted for >50% of clinical use as long-­ the dalteparin or tinzaparin subgroup should also be taken into con- term therapy in our study,14 and around 50% of these patients were sideration, and further analysis regarding regimen variance between prescribed enoxaparin. We also found different treatment regimens subgroups may be required. Importantly, the rate of fatal bleeding adopted by clinicians (once-­daily injection, twice-­daily injections, or risks was comparable between the enoxaparin and dalteparin or switch from initial twice daily to once daily as long-­term therapy). On tinzaparin subgroups. 8 of 10 | TRUJILLO-­SANTOS et al. TA B L E 4 Six-­month outcomes While the proportion of patients with hematological malignan- Dalteparin or cies in RIETECAT was small, this is reflective of the real-­life inci- Enoxaparin tinzaparin dence and is also similar to the proportion included across clinical n = 3526 n = 925 trials with LMWHs. In the Cox proportional model, cancer location Main outcome, n (%) (type) was considered as a covariate for data adjustment, and VTE Recurrent VTEa 70 (2.0) 23 (2.5) recurrence risk did not differ between patients treated with enoxa- Secondary outcomes parin and those treated with dalteparin or tinzaparin. However, due Recurrent DVTb 36 (1.0) 11 (1.2) to the small number of patients in each cancer subgroup, it was not Recurrent PE c 35 (0.99) 12 (1.3) feasible to perform any comparison of effectiveness and safety for Safety outcomes each cancer type. Major bleeding c 111 (3.1) 18 (1.9) The efficacy and safety of enoxaparin versus VKAs for the long-­ term therapy of VTE in patients with cancer have been studied in Recurrent VTE or major 181 (5.1) 41 (4.4) bleeding two randomized controlled trials, which found nonsignificant dif- Nonmajor bleeding 87 (2.5) 24 (2.6) ferences between subgroups.11,12 However, ours is the first study consistently comparing enoxaparin with the other two LMWHs as All-­c ause death 666 (19) 157 (17) a subgroup. The incidence rate of VTE recurrences, major bleeding, Cause of death or death in our cohort was lower than in previous studies.15,16 This Pulmonary embolism 33 (0.94) 6 (0.65) is consistent with recent reports that found a progressive improve- Bleeding 15 (0.43) 5 (0.54) ment in the outcomes of VTE patients, particularly in those with Fatal PE or fatal bleeding 48 (1.4) 11 (1.2) cancer.17,18 Sudden, unexpected death 3 (0.09) 3 (0.32) The present study has potential limitations. First, given the Abbreviations: DVT, deep vein thrombosis; PE, pulmonary embolism; noninterventional nature of the study, comparability of patients VTE, venous thromboembolism. treated with enoxaparin versus dalteparin or tinzaparin may be a Composite effectiveness outcome = symptomatic DVT and fatal or difficult to achieve. For this reason, we used a multivariable Cox nonfatal PE. b model adjusted by relevant covariables and propensity score Symptomatic. c matching to minimize the confounding bias (although these could Fatal and nonfatal. TA B L E 5 Time-­dependent Cox Cox proportional hazards proportional hazards for the crude and Propensity adjusted association between outcomes Crude analysis Adjusted analysis score and treatment Recurrent DVTa 0.88 (0.45-­1.73) 0.83 (0.41-­1.70) 0.90 (0.43-­1.90) Recurrent PEb 0.78 (0.41-­1.51) 0.79 (0.41-­1.56) 0.73 (0.35-­1.55) Recurrent VTEc 0.82 (0.51-­1.31) 0.82 (0.50-­1.34) 0.81 (0.48-­1.38) Major bleedingb 1.67 (1.02-­2.76) 1.522 (0.90-­2.58) 1.40 (0.80-­2.46) Non-­major bleeding 0.97 (0.62-­1.53) 0.88 (0.55-­1.41) 0.89 (0.53-­1.49) All-­c ause death 1.14 (0.96-­1.36) 0.97 (0.81-­1.16) 1.07 (0.88-­1.30) Fatal PE 1.46 (0.61-­3.47) 1.05 (0.43-­2.55) 1.00 (0.37-­2.71) Fatal bleeding 0.81 (0.30-­2.24) 0.57 (0.20-­1.64) 0.44 (0.12-­1.65) Note: Results expressed as hazard ratio (95% confidence interval). Abbreviations: DVT, deep vein thrombosis; PE, pulmonary embolism; VTE, venous thromboembolism. a Symptomatic. b Fatal and nonfatal. c Composite effectiveness outcome = symptomatic DVT and fatal or nonfatal PE. TRUJILLO-­SANTOS et al. | 9 of 10 not adjust for unmeasured confounding). Second, data from reg- House, Third Avenue, Globe Park, Marlow, Buckinghamshire, SL7 istries are susceptible to selection bias if a nonrepresentative 1EY, UK, and was funded by Sanofi. sample of patients is selected for analysis. However, the RIETE registry captures a broad range of consecutive patients with R E L AT I O N S H I P D I S C LO S U R E symptomatic VTE from multiple medical centers, countries, and This RIETECAT study was sponsored and funded by Sanofi, but treatment settings, making it less likely that the study cohort is Sanofi did not have any access to the data at any time, nor did they made up of a skewed population. The primary analysis excluded have any influence or implication on the preparation of the manu- patients who switched from full-­d ose treatment with enoxaparin, script. The manuscript was written, reviewed, and approved by the dalteparin, or tinzaparin to a non–­f ull-­d ose or to another antico- authors. None of the authors perceived any fee or other forms of agulant treatment in the first 90 days. Exclusion of patients based compensation for the study or the manuscript. on information obtained during follow-­up can introduce bias.19 A switch in dose or treatment during follow-­up to avoid bleeding ORCID complications (eg, following admission to hospital for an invasive Andrei Braester https://orcid.org/0000-0002-1567-7562 diagnostic test, surgical intervention, or other complications) can Manuel Monreal https://orcid.org/0000-0002-0494-0767 be indicative of a different bleeding risk at baseline. However, these patients were included in the sensitivity analysis and the T WITTER results from the 6-­m onth outcomes and adjusted analyses were Aitor Ballaz @AitorBallaz similar to those observed in the primary analysis. This multina- Isabelle Mahé @isabellemahe1 tional study provided real-­world data on the 6 months compar- Manuel Monreal @mmonrealriete ative effectiveness of enoxaparin versus dalteparin or tinzaparin for secondary VTE prevention in patients with active cancer by REFERENCES leveraging existing data. The main strength of our observation is 1. Khorana AA. Venous thromboembolism and prognosis in cancer. that the population-­b ased sample we used describes the effects Thromb Res. 2010;125:490-­493. 2. Monreal M, Falga C, Valdes M, et al. Fatal pulmonary embo- of initial therapy for VTE in “real-­world” clinical care, as opposed lism and fatal bleeding in cancer patients with venous thrombo- to a protocol-­driven randomized trial, and enhances the generaliz- embolism: findings from the RIETE registry. J Thromb Haemost. ability of our findings. 2006;4:1950-­1956. To conclude, in RIETECAT, patients with active cancer and acute 3. Trujillo-­Santos J, Nieto J, Tiberio G, et al. Predicting recurrences or major bleeding in cancer patients with venous thrombo- symptomatic VTE receiving full-­dose enoxaparin had a nonsignifi- embolism. Findings from the RIETE registry. Thromb Haemost. cantly lower risk for VTE recurrences, a nonsignificantly higher risk 2008;100:435-­439. for major bleeding, and similar risks for NMCR bleeding or death 4. Chee CE, Ashrani AA, Marks RS, et al. Predictors of venous throm- over 6 months compared with patients treated with full-­dose dalte- boembolism recurrence and bleeding among active cancer patients: a population-­based cohort study. Blood. 2014;123:3972-­3978. parin or tinzaparin. 5. Farge D, Trujillo-­Santos J, Debourdeau P, et al. Fatal events in can- cer patients receiving anticoagulant therapy for venous thrombo- AU T H O R C O N T R I B U T I O N S embolism. Medicine (Baltimore). 2015;94(32):e1235. JT-­S participated in the design of the study, recruited patients, per- 6. Fuentes HE, Tafur AJ, Caprini JA, et al. Prediction of early mortality formed the statistical analyses, and revised the final version of the in patients with cancer-­associated thrombosis in the RIETE data- base. Int Angiol. 2019;38:173-­184. manuscript. DF-­B, JMP, CG-­G , AB, ABr, IM, AV, and JAP participated 7. Kearon C, Akl EA, Ornelas J, et al. Antithrombotic therapy for in the design of the study, recruited patients, and helped to write the VTE disease: chest guideline and expert panel report. Chest. manuscript. MM participated in the design of the study, recruited 2016;149:315-­352. patients, wrote the manuscript, and obtained funds. 8. Watson HG, Keeling DM, Laffan M, Tait RC, Makris M, British Committee for Standards in Haematology. Guideline on as- pects of cancer-­related venous thrombosis. Br J Haematol. AC K N OW L E D G M E N T S 2015;170:640-­6 48. We express our gratitude to Sanofi Spain and LEO PHARMA for sup- 9. Farge D, Bounameaux H, Brenner B, et al. International clinical porting this registry with an unrestricted educational grant. We also practice guidelines including guidance for direct oral anticoagulants in the treatment and prophylaxis of venous thromboembolism in thank the RIETE Registry Coordinating Center, S&H Medical Science patients with cancer. Lancet Oncol. 2016;17(10):e452-­e 466. Service, for their quality control data, logistic, and administrative 10. Lyman GH, Bohlke K, Khorana AA, et al. Venous thromboembo- support; and Prof. Salvador Ortiz, Universidad Autónoma Madrid, lism prophylaxis and treatment in patients with cancer. American and Silvia Galindo, both statistical advisors in S&H Medical Science Society of Clinical Oncology Clinical Practice Guideline update 2014. J Clin Oncol. 2015;33(6):654-­656. Service, for the statistical analysis of the data presented in this 11. Meyer G, Marjanovic Z, Valcke J, et al. Comparison of low-­molecular-­ paper. The manuscript was written, reviewed, and approved by the weight heparin and warfarin for the secondary prevention of ve- authors. Editorial support was provided by Jane Juif and Ann-­Marie nous thromboembolism in patients with cancer: a randomized Shaw from Lucid Group Communications Ltd, First Floor, Jubilee controlled study. Arch Intern Med. 2002;162:1729-­1735. 10 of 10 | TRUJILLO-­SANTOS et al. 12. Deitcher SR, Kessler CM, Merliet G, et al. Secondary prevention 18. Morillo R, Jiménez D, Aibar MA, et al. DVT management and out- of venous thromboembolic events in patients with active cancer: come trends, 2001 to 2014. Chest. 2016;150:374-­383. enoxaparin alone versus initial enoxaparin followed by warfarin for 19. Hernán MA, Sauer BC, Hernández-­Díaz S, Platt R, Shrier I. a 180-­day period. Clin Appl Thromb Hemost. 2006;12:389-­396. Specifying a target trial prevents immortal time bias and other 13. Bikdeli B, Jimenez D, Hawkins M, et al. Rationale, design and meth- self-­inflicted injuries in observational analyses. J Clin Epidemiol. odology of the computerized registry of patients with venous 2016;79:70-­75. thromboembolism (RIETE). Thromb Haemost. 2018;118:214-­224. 14. Mahé I, Sterpu R, Bertoletti L, et al. Long-­term anticoagulant ther- apy of patients with venous thrombo-­embolism. What are the prac- S U P P O R T I N G I N FO R M AT I O N tices? PLoS One. 2015;10(6):e0128741. Additional supporting information may be found in the online 15. Lee AYY, Levine MN, Baker RI, et al. Low-­molecular-­weight heparin version of the article at the publisher’s website. versus a coumarin for the prevention of recurrent venous thrombo- embolism in patients with cancer. N Engl J Med. 2003;349:146-­153. 16. Lee AYY, Bauersachs R, Janas MS, et al. CATCH: a randomised clin- ical trial comparing long-­term tinzaparin versus warfarin for treat- How to cite this article: Trujillo-­Santos J, Farge-­Bancel D, ment of acute venous thromboembolism in cancer patients. BMC Pedrajas JM, et al; The RIETE Investigators. Enoxaparin Cancer. 2013;13:284. versus dalteparin or tinzaparin in patients with cancer and 17. Jiménez D, de Miguel-­Díez J, Guijarro R, et al. Trends in the man- agement and outcomes of acute pulmonary embolism. Analysis venous thromboembolism: The RIETECAT study. Res Pract from the RIETE registry. J Am Coll Cardiol. 2016;67:162-­170. Thromb Haemost. 2022;6:e12736. doi:10.1002/rth2.12736 APPENDIX Coordinator of the RIETE Registry: Manuel Monreal. A Latorre, J Lima, P Llamas, JL Lobo, L López-­Jiménez, P López-­ RIETE Steering Committee Members: Paolo Prandoni, Benjamin Miguel, JJ López-­Núñez, R López-­Reyes, JB López-­Sáez, A Lorenzo, Brenner, and Dominique Farge-­Bancel. O Madridano, A Maestre, PJ Marchena, F Martín-­Martos, D RIETE National Coordinators: Raquel Barba (Spain), Pierpaolo Martínez-­Urbistondo, C Mella, MI Mercado, J Moisés, M Monreal, Di Micco (Italy), Laurent Bertoletti (France), Sebastian Schellong MV Morales, A Muñoz-­Blanco, N Muñoz-­Rivas, MS Navas, JA Nieto, (Germany), Inna Tzoran (Israel), Abilio Reis (Portugal), Marijan E Nofuentes-­Pérez, MJ Núñez-­Fernández, B Obispo, M Olid, MC Bosevski (R. Macedonia), Henri Bounameaux (Switzerland), Radovan Olivares, JL Orcastegui, J Osorio, S Otalora, R Otero, D Paredes, P Malý (Czech Republic), Peter Verhamme (Belgium), Joseph A. Caprini Parra, JM Pedrajas, G Pellejero, JA Porras, J Portillo, F Rivera-­Civico, (USA), Hanh My Bui (Vietnam). DA Rodríguez-­Chiaradía, C Rodríguez-­Matute, J Rogado, V Rosa, P RIETE Registry Coordinating Center: S & H Medical Science Ruiz-­Artacho, N Ruiz-­Giménez, J Ruiz-­Ruiz, P Ruiz-­Sada, G Salgueiro, Service. R Sánchez-­Martínez, JF Sánchez-­Muñoz-­Torrero, T Sancho, S Soler, B Suárez-­Rodríguez, JM Suriñach, R Tirado, MI Torres, C Tolosa, J Members of the RIETE Group Trujillo-­Santos, F Uresandi, B Valero, R Valle, JF Varona, G Vidal, A SPAIN: MD Adarraga, J Aibar, MA Aibar, C Amado, JI Arcelus, A Villalobos, P Villares, C Zamora, BELGIUM: M Engelen, T Vanassche, Asuero, A Ballaz, R Barba, C Barbagelata, M Barrón, B Barrón-­ P Verhamme, CZECH REPUBLIC: J Hirmerova, R Malý, FRANCE: N Andrés, A Blanco-­Molina, E Botella, AM Camon, I Casado, J Castro, Ait Abdallah, L Bertoletti, A Bura-­Riviere, J Catella, F Couturaud, M Castro, L Chasco, J Criado, C de Ancos, J del Toro, P Demelo-­ B Crichi, P Debourdeau, O Espitia, N Falvo, D Farge-­Bancel, H Rodríguez, AM Díaz-­Brasero, JA Díaz-­Peromingo, MV Di Campli, A Helfer, K Lacut, R Le Mao, I Mahé, F Moustafa, G Poenou, I Quere, Dubois-­Silva, JC Escribano, F Espósito, C Falgá, AI Farfán-­Sedano, GERMANY: S Schellong, ISRAEL: A Braester, B Brenner, I Tzoran, C Fernández-­C apitán, JL Fernández-­Reyes, MA Fidalgo, K Flores, IRAN: R Nikandish, ITALY: F Bilora, B Brandolin, M Ciammaichella, P C Font, L Font, I Francisco, C Gabara, F Galeano-­Valle, MA García, Di Micco, E Imbalzano, R Maida, F Pace, R Pesavento, P Prandoni, R F García-­Bragado, M García de Herreros, R García de la Garza, C Quintavalla, A Rocci, C Siniscalchi, A Tufano, A Visonà, B Zalunardo, García-­Díaz, R García-­Hernáez, A García-­Raso, A Gil-­Díaz, M LATVIA: J Birzulis, A Skride, A Zaicenko, PORTUGAL: S Fonseca, Giménez-­Suau, C Gómez-­Cuervo, E Grau, L Guirado, J Gutiérrez, F Martins, J Meireles, REPUBLIC OF MACEDONIA: M Bosevski, L Hernández-­Blasco, E Hernando, L Jara-­Palomares, MJ Jaras, D SWITZERLAND: H Bounameaux, L Mazzolai, USA: CI Ochoa-­Chaar, Jiménez, R Jiménez, C Jiménez-­Alfaro, MD Joya, S Lainez-­Justo, I Weinberg, VIETNAM: HM Bui.