Martínez-López et al. Blood Cancer Journal (2020)10:103 https://doi.org/10.1038/s41408-020-00372-5 Blood Cancer Journal ARTICLE Open Access Multiple myeloma and SARS-CoV-2 infection: clinical characteristics and prognostic factors of inpatient mortality Joaquín Martínez-López 1, María-Victoria Mateos 2, Cristina Encinas3, Anna Sureda4, José Ángel Hernández-Rivas5, Ana Lopez de la Guía6, Diego Conde7, Isabel Krsnik8, Elena Prieto9, Rosalía Riaza Grau10, Mercedes Gironella 11, María Jesús Blanchard12, Nerea Caminos13, Carlos Fernández de Larrea14, María Alicia Senin15, Fernando Escalante16, José Enrique de la Puerta 17, Eugenio Giménez18, Pilar Martínez-Barranco19, Juan José Mateos20, Luis Felipe Casado21, Joan Bladé14, Juan José Lahuerta 1, Javier de la Cruz22 and Jesús San-Miguel 23 Abstract There is limited information on the characteristics, prognostic factors, and outcomes of patients with multiple myeloma (MM) hospitalized with COVID-19. This retrospective case series investigated 167 patients reported from 73 hospitals within the Spanish Myeloma Collaborative Group network in March and April, 2020. Outcomes were compared with 167 randomly selected, contemporary, age-/sex-matched noncancer patients with COVID-19 admitted at six participating hospitals. Among MM and noncancer patients, median age was 71 years, and 57% of patients were male; 75 and 77% of patients, respectively, had at least one comorbidity. COVID-19 clinical severity was 1234567890():,; 1234567890():,; 1234567890():,; 1234567890():,; moderate–severe in 77 and 89% of patients and critical in 8 and 4%, respectively. Supplemental oxygen was required by 47 and 55% of MM and noncancer patients, respectively, and 21%/9% vs 8%/6% required noninvasive/invasive ventilation. Inpatient mortality was 34 and 23% in MM and noncancer patients, respectively. Among MM patients, inpatient mortality was 41% in males, 42% in patients aged >65 years, 49% in patients with active/progressive MM at hospitalization, and 59% in patients with comorbid renal disease at hospitalization, which were independent prognostic factors on adjusted multivariate analysis. This case series demonstrates the increased risk and identifies predictors of inpatient mortality among MM patients hospitalized with COVID-19. Introduction CoV-2), was detected in Wuhan, China1. This rapidly In December 2019, a cluster of patients with pneumonia evolved into a pandemic of the heterogeneous clinical linked to a previously unknown coronavirus, now named disease named coronavirus disease 2019 (COVID-19) that severe acute respiratory syndrome coronavirus 2 (SARS- has challenged healthcare systems and led to thousands of deaths worldwide. The clinical presentation is highly variable, ranging from asymptomatic cases to severe Correspondence: María-Victoria Mateos (mvmateos@usal.es) respiratory infections. It is associated with a pro- 1 Hospital Universitario 12 de Octubre, i+12, Complutense University, CNIO, inflammatory status, leading to cytokine release syn- CIBERONC number CB16/12/00369, Madrid, Spain 2 drome and hypercoagulation, with multi-organ damage Hospital Universitario de Salamanca, Instituto de Investigación Biomédica de Salamanca (IBSAL), Centro de Investigación del Cancer (IBMCC-USAL, CSIC), and fatal outcome2. It was soon recognized that various CIBER-ONC number CB16/12/00233, Salamanca, Spain factors, including increased age (>65 years, and particu- Full list of author information is available at the end of the article larly >75 years), male sex, and presence of comorbidities These authors contributed equally: Joaquín Martínez-López, María- Victoria Mateos (cardiovascular comorbidity, obesity, hypertension, and These authors jointly supervised this work: Javier de la Cruz, Jesús San-Miguel © The Author(s) 2020 Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/. Blood Cancer Journal Martínez-López et al. Blood Cancer Journal (2020)10:103 Page 2 of 11 diabetes) contributed substantially to the severity of noncancer cohort of COVID-19 patients, and identify COVID-19. Investigations of novel strategies for the preadmission prognostic factors of inpatient mortality. prevention and treatment of COVID-19 (including anti- malarials, antiviral drugs, antibiotics, anti-interleukin-6/-1 Methods therapies, and anticoagulants), together with appropriate This multicenter case-series analysis was performed at a intensive supportive care, have aimed to improve patient national level by the Spanish Myeloma Collaborative outcome3,4. There are cumulative data indicating that Group (Programa Español de Tratamientos en Hemato- patients with cancer may be at increased risk for more logía14,20/Grupo Español de Mieloma8). The study was severe COVID-19 and associated complications, including approved by the Institutional Review Board (IRB) of those receiving or not receiving treatment within the Hospital Universitario 12 de Octubre (n 20/208). This month prior to infection5,6, although other recent results case-series study includes MM patients who were suggest mortality may be primarily associated with age, admitted to a participating hospital with a SARS-CoV-2 male sex, and comorbidities7. polymerase chain reaction (PCR)-positive test between Multiple myeloma (MM) is a malignant proliferation of March 1 and April 30, 2020. Additionally, a series of clonal B lymphoid cells at the last stage of maturation patients with no cancer history, matched by age and sex, (plasma cells) that are responsible for the secretion of were randomly selected from cohorts of COVID-19 immunoglobulins (Igs). Impairment of Ig production, patients admitted during the same time period at six with the coexistence of a monoclonal component (M- participating hospitals. component) plus immunosuppression of normal Igs, leads Clinical specimens were obtained by nasopharyngeal to severely impaired humoral immunity. Additionally, swab collection in accordance with Spanish disease con- MM patients display dysfunctional cellular and innate trol and prevention guidelines. Samples were processed at immunity; this complex failure of immunosurveillance local microbiology laboratories, and SARS-CoV-2 one- mechanisms makes MM patients highly susceptible to step real-time reverse transcriptase (rRT)-PCR diagnostic viral and bacterial infections8. Some MM treatments may assay was performed. also positively or negatively impact the immune system. A modified version of a questionnaire proposed by the The most common side-effect of MM treatment is the International Myeloma Society was used for data extrac- induction of cytopenia, which is clearly associated with tion. A common set of demographic, clinical, laboratory, increased susceptibility to bacterial and viral infections. treatment, and outcome variables were collected for the Chemotherapy, particularly high-dose melphalan, results MM and noncancer cohorts of hospitalized COVID-19 in these side effects. Corticosteroids have well-established patients. Specific disease status and treatment data were immunosuppressive effects. Proteasome inhibitors induce obtained for MM patients. Data extraction was performed T-cell dysfunction and are associated with an increased locally by hematology departments in participating hos- risk for varicella-zoster virus reactivation9. Immunomo- pitals and centrally processed by the PETHEMA/GEM dulatory drugs may have a protective effect by increasing data-monitoring unit. Hospital outcome status was natural killer (NK) and T-cell function but are also defined as discharge from the hospital or receiving associated with cytopenia and an increased risk for venous ongoing care (survivor group) or death (non-survivor thromboembolism10. CD38 monoclonal antibodies group). reduce NK and immunosuppressive regulatory T cells, and are associated with higher incidences of viral and Statistical analysis bacterial infections8,11,12. Demographic, clinical, laboratory, and therapy covari- To date, there is very limited information regarding ates and hospital outcomes were summarized using characteristics and outcomes of MM patients with descriptive statistics. Continuous variables were reported COVID-19, with only a single case report13 and a small as the median and interquartile range (IQR). Generic UK series of 75 heterogeneous patients that did not report COVID-19 patient characteristics and specific MM fea- a non-MM reference group of patients with COVID-1914. tures were analyzed according to the two hospital out- Several international and national guidelines have been come categories. Crude and adjusted odd ratios (OR) with reported15–19, providing guidance to physicians on the 95% confidence intervals (CI) were estimated with logistic management of patients with MM during the COVID-19 regression analysis for a predefined set of well-established pandemic; however, these recommendations are based on prognostic factors. Preadmission characteristics, MM consensus and general disease concepts, and lack specific status, and comorbidities constituted the reference model data derived from infected patients. To our knowledge, and were used to adjust the association of relevant MM this is the first large case-series study to describe com- features with mortality. To limit model overfitting, the prehensively the clinical characteristics of COVID-19 in number of events per candidate variable was restricted to hospitalized MM patients, compare outcomes with a >10. Calibration and discrimination of the resulting Blood Cancer Journal Martínez-López et al. Blood Cancer Journal (2020)10:103 Page 3 of 11 models were assessed using the Hosmer–Lemeshow with 108 (65%) reported from two provinces with high goodness-of-fit test and the c-statistic21. We assessed population excess mortality. internal validity with a bootstrapping procedure that used The characteristics and clinical features of the 167 MM 500 replicate samples to gain insight into the stability of patients and the 167 noncancer patients are described in the reference model22. Analyses were generated using Table 1. Concerning COVID-19 clinical features, 77 and SAS/STAT software, Version 9.4, SAS Institute Inc. 89% of MM and noncancer patients, respectively, pre- sented with moderate–severe disease, and 8 and 4% had a Results critical disease. The proportion of patients requiring Between March 1 and April 30, 2020, 216 patients with supplemental oxygen was 47 and 55% in MM and non- MM and SARS-Cov-2 infection detected by RT-PCR were cancer patients, respectively. Numerically, more MM reported from 73 healthcare centers. Thirty patients were patients required noninvasive and invasive ventilation excluded because of a diagnosis of a plasma cell disorder (21% and 9%, respectively) compared to the noncancer other than MM; 19 patients were excluded because they group (8% and 6%, respectively). At hospital admission, did not require hospital admission (Fig. 1). As of May 20, MM patients had numerically lower median levels of 2020, one patient remained hospitalized and receiving neutrophils, lymphocytes, and platelets, and numerically ongoing care. The temporal and geographical distribution higher concentrations of D-dimers and ferritin compared of reported cases matched the distribution of the epi- with noncancer patients (Table 1). Treatments adminis- demic in Spain; 114 (68%) were confirmed in the first tered to the two groups of patients are summarized in 3 weeks of the lockdown period (March 15 to April 4). Table 1; 31% of MM patients received triple-combination Eligible cases were notified from 29 of 50 provinces (58%), therapy including (hydroxy)chloroquine, azithromycin, PETHEMA/GEM-affiliated centers reporting multiple myeloma patients with COVID-19 between Reporting March 1 and April 30, 2020 centers n = 73 From 17/17 (100%) regions and 38/50 (76%) provinces in Spain Centers excluded as reporting 0 cases n = 11 (15%) All reported cases in Spain Reported n = 216 Patients excluded, due to not cases From 62 centers in 15/17 (88%) having a diagnosis of multiple regions and 32/50 (64%) provinces myeloma n = 30 (14%) • Amyloidosis (n = 10) • Monoclonal gammopathy of undetermined significance All reported cases in patients with (n = 10) multiple myeloma • Smoldering multiple myeloma n = 186 (n = 10) From 59 centers in 14/17 (82%) regions and 31/50 (62%) provinces Patients excluded due to being PCR-negative for COVID-19 n=0 All reported cases in patients with multiple myeloma and PCR-positive for COVID-19 n = 186 Patients excluded due to being treated as outpatients for COVID-19 n = 19 (10%) All cases in hospitalized patients with multiple myeloma and Cases in PCR-positive for COVID-19 analysis n = 167 (90%) From 57 centers in 14/17 (82%) regions and 29/50 (58%) provinces Fig. 1 Flow chart of the patients included in the study. Flow diagram of patients with multiple myeloma admitted to hospital with COVID-19 who were included in the present analysis. Blood Cancer Journal Martínez-López et al. Blood Cancer Journal (2020)10:103 Page 4 of 11 Table 1 Preadmission features, COVID-19 clinical, Table 1 continued laboratory, and therapy features, and COVID-19 outcome in multiple myeloma (MM) and noncancer patients MM patients Noncancer hospitalized with COVID-19 diagnosis. (n = 167), patients (n = N (%) 167), N (%) MM patients Noncancer (n = 167), patients (n = D-dimers, μg/mL 0.861 0.685 N (%) 167), N (%) (0.52–1.207) (0.422–1.383) Ferritin, ng/mL 995 (486–2070) 761 (401–1528) Preadmission features Age-/sex- matched COVID-19 therapy Age, median (IQR), years 71 (62–78) 71 (63–78) (Hydroxy)chloroquine 148 (89) 157 (94) Age groups Azithromycin 91 (54) 115 (69) <65 years 55 (33) 55 (33) Antiretrovirals 103 (62) 103 (62) 65–74 years 55 (33) 55 (33) Antiviral combination therapy ≥75 years 57 (34) 57 (34) None 11 (7) 5 (3) Sex (Hydroxy)chloroquine only 16 (10) 14 (8) Male 95 (57) 95 (57) (Hydroxy)chloroquine and 47 (28) 31 (19) retroviral Female 72 (43) 72 (43) (Hydroxy)chloroquine and 34 (20) 42 (25) Comorbidities azithromycin At least one 126 (75) 129 (77) (Hydroxy)chloroquine, 51 (31) 70 (42) None 41 (25) 38 (23) azithromycin, and retroviral Cardiac disease 35 (21) 32 (19) Other combination 8 (5) 5 (3) Pulmonary disease 23 (14) 34 (20) Steroids 83 (50) 62 (37) Diabetes 28 (17) 36 (22) Anti-interleukin-6 receptor 22 (13) 27 (16) Renal disease 32 (19) 13 (8) antibody therapy Hypertension 67 (40) 79 (47) Steroids and anti-interleukin-6 21 (13) 16 (10) COVID-19 clinical features receptor antibody therapy Clinical severitya Heparin 109 (65) 117 (70) Mild 26 (15) 11 (7) COVID-19 outcome (hospital outcome) Moderate–severe 128 (77) 149 (89) Receiving ongoing care 1 (<1) 2 (1) Critical 13 (8) 6 (4) Hospital discharge 110 (66) 127 (76) Respiratory supporta Hospital death 56 (34) 38 (23) No supplemental oxygen 39 (23) 51 (31) COVID-19 coronavirus disease 2019, IQR interquartile range. Supplemental oxygen 78 (47) 91 (55) a Data missing for one patient in the noncancer group. Noninvasive ventilation 35 (21) 14 (8) Invasive ventilation 15 (9) 10 (6) and retroviral compared with 42% of the noncancer COVID-19 laboratory group. Regarding outcomes, inpatient mortality was 34% measures at hospital in the MM group and 23% in the noncancer group. admission, median (IQR) Absolute neutrophil count, ×109/ 2.78 (1.7–4.7) 4.8 (3.54–6.3) MM- and COVID-19-related characteristics according to L hospital outcome Lymphocyte count, ×109/L 0.515 0.99 (0.7–1.42) Preadmission characteristics of MM patients according (0.24–0.815) to hospital outcome are presented in Table 2; 56 patients (34%) died during hospitalization (non-survivor group), Platelets, ×109/L 129 (78–176) 215 (116–285) 110 (66%) were discharged, and 1 patient remained hos- pitalized receiving ongoing care (survivor group). Mor- tality was 41% in males and 24% in females. Mortality was Blood Cancer Journal Martínez-López et al. Blood Cancer Journal (2020)10:103 Page 5 of 11 Table 2 Characteristics of multiple myeloma (MM) Table 2 continued patients with COVID-19, according to hospital outcome. Patients with hospital outcome, N (%) Patients with hospital outcome, N (%) All (n = 167) Patients Patients All (n = 167) Patients Patients discharged who died discharged who died (n = 111)a (n = 56) (n = 111)a (n = 56) Year of diagnosis of MM Age, median (IQR), years 71 (62–78) 70 (58–77) 75 (68–79) (time since diagnosis) Age groups, years, n (%) % shown as the proportion 2020 (0–3 months) 25 (15) 13 (52) 12 (48) of “all” 2019 (4–18 months) 30 (18) 24 (80) 6 (20) ≤49 6 (4) 6 (100) 0 ≤2018 (≥18 months) 112 (67) 74 (66) 38 (34) 50–59 29 (17) 24 (83) 5 (17) Prior stem cell 60–69 41 (24) 25 (61) 16 (39) transplantationd 70–79 60 (36) 37 (62) 23 (38) Yes 51 (31) 42 (82) 9 (18) ≥80 31 (19) 19 (61) 12 (39) No 113 (69) 67 (59) 46 (41) Sex Prior lines of treatment Males 95 (57) 56 (59) 39 (41) for MMe Females 72 (43) 55 (76) 17 (24) 1 82 (51) 58 (71) 24 (29) MM features % shown as the proportion 2 39 (24) 23 (59) 16 (41) of “all” >3 41 (25) 27 (66) 14 (34) Monoclonal component Prior therapy class Immunoglobulin G 83 (50) 61 (73) 22 (27) Proteasome inhibitor 138 (83) 92 (67) 46 (33) Immunoglobulin A 42 (25) 25 (60) 17 (40) No proteasome inhibitor 29 (17) 19 (66) 10 (34) Immunoglobulin D 2 (1) 2 (100) 0 (0) Immunomodulatory drug 119 (71) 80 (67) 39 (33) Immunoglobulin M 3 (2) 2 (67) 1 (33) No 48 (29) 31 (65) 17 (35) Non-secretory disease 3 (2) 2 (67) 1 (33) immunomodulatory drug Light chain only disease 34 (21) 19 (56) 15 (44) Monoclonal antibody 38 (23) 25 (66) 13 (34) b No monoclonal antibody 129 (77) 86 (67) 43 (33) International staging system disease stage I 47 (28) 34 (72) 13 (28) MM status and comorbidities at % shown the proportion hospital admission for COVID-19 of “all” II 61 (37) 42 (69) 19 (31) MM status III 57 (35) 35 (61) 22 (39) Active disease or 43 (26) 22 (51) 21 (49) Cytogenetic classification by c progression FISH Partial response 72 (43) 52 (72) 20 (28) High risk 36 (24) 23 (64) 13 (36) Complete response 52 (31) 37 (71) 15 (29) Other/standard risk 114 (76) 75 (66) 39 (34) Comorbidities Bone disease at diagnosis of MM At least one 126 (75) 79 (63) 47 (37) Yes 129 (77) 86 (67) 43 (33) None 41 (25) 32 (78) 9 (22) No 38 (23) 25 (66) 13 (34) Cardiac disease 35 (21) 22 (63) 13 (37) Renal disease at diagnosis No cardiac disease 132 (79) 89 (67) 43 (33) of MM Pulmonary disease 23 (14) 14 (61) 9 (39) Yes 45 (27) 22 (49) 23 (51) No pulmonary disease 144 (86) 97 (67) 47 (33) No 122 (73) 89 (73) 33 (27) Diabetes 28 (17) 18 (64) 10 (36) Blood Cancer Journal Martínez-López et al. Blood Cancer Journal (2020)10:103 Page 6 of 11 Table 2 continued disease, the inpatient mortality rate was 49% compared to 28 and 29% for patients in partial or complete response, Patients with hospital outcome, N (%) respectively (Table 2). Patients with at least one comor- bidity had an inpatient mortality rate of 37% vs 22% in All (n = 167) Patients Patients those without comorbidities. The mortality rate was discharged who died numerically higher in MM patients with vs without car- (n = 111)a (n = 56) diac (37% vs 33%) or pulmonary (39% vs 33%) comor- No diabetes 139 (83) 93 (67) 46 (33) bidities or hypertension (42% vs 28%); the presence of renal disease was associated with the numerically highest Renal disease 32 (19) 13 (41) 19 (59) mortality rate (59%). Immunoparesis did not affect the No renal disease 135 (81) 9 (73) 37 (27) mortality rate. Hypertension 67 (40) 39 (58) 28 (42) Table 3 summarizes clinical and laboratory features at No hypertension 100 (60) 71 (72) 28 (28) admission according to hospital outcome. Patients with f higher values (above the median) for the neutrophil count, Immunoparesis 104 (69) 67 (65) 37 (31) D-dimers, and ferritin had higher mortality rates (43%, No immunoparesis 46 (31) 30 (65) 16 (35) 40%, and 32%, respectively) than those in patients with COVID-19 coronavirus disease 2019, FISH fluorescence in situ hybridization, IQR lower values (below the median; 25%, 22%, and 22%, interquartile range. a respectively). The opposite effect was observed for lym- Includes one patient who remains hospitalized receiving ongoing care. b Data missing for two patients who died in the hospital. phocyte and platelet counts, with lower values associated c Data missing for 17 patients, 13 who were discharged and 4 who died in the with a higher mortality rate (40% vs 28 and 41% vs 25%, hospital. d Data missing for three patients, two who were discharged and one who died in respectively). the hospital. The association with inpatient mortality was estimated e Data missing for five patients, three who were discharged and two who died in the hospital. for nine prognostic factors assessed before admission f Data missing for 16 patients, 13 who were discharged and 3 who died in (Table 4). In an unadjusted analysis, seven factors were hospital. associated with increased odds of death: male sex, age >65 years, renal disease and hypertension as comorbidities, 16% in patients aged <65 years compared to 42% in those diagnosis of MM in 2020, renal disease at diagnosis, and aged >65 years, including 40% in patients aged 65–75 active/progressive disease at the time of COVID-19 hos- years and 44% in those aged >75 years. None of the female pitalization (Table 4). IgG M-component and prior ASCT patients aged <65 years died. were associated with lower inpatient mortality. In an Concerning MM features, inpatient mortality was 27 adjusted analysis, male sex, age >65 years, active/pro- and 28% in patients with an IgG M-component and stage gressive disease, and renal disease as comorbidity were I disease at diagnosis, respectively. Cytogenetic abnorm- independent prognostic factors of inpatient mortality. The alities and the presence of bone disease did not impact combination of this set of independent factors constituted inpatient mortality. However, in patients with renal a stable prognostic model of inpatient mortality; the impairment at MM diagnosis inpatient mortality was 51% bootstrapping procedure corrected the discriminatory vs 27% in patients with normal renal function. Prior ability of this reference model in terms of c-statistic from treatment with immunomodulatory drugs, proteasome 0.79 to 0.765. No other candidate variable improved the inhibitors, or monoclonal antibodies did not impact prognostic performance of the reference model in terms inpatient mortality (Table 2). of calibration and discrimination (Table 4). With regards to the year of MM diagnosis, 12 of 25 patients (48%) diagnosed between January and April 2020 Discussion (during the emerging pandemic) did not survive COVID- This study provides a comprehensive description of the 19. To understand this high mortality rate, we analyzed characteristics and outcomes of MM patients hospitalized the characteristics of the cohort. Of note, 10 (83%) of with confirmed SARS-CoV-2 infection in Spain. Mortality the 12 non-survivors were male, 11 (92%) had a M- was 50% higher in MM patients (34%) vs noncancer component other than IgG, 9 (75%) had international patients (23%). The main predictors of inpatient mortality staging system stage III disease, 7 (58%) presented with for MM were male sex, age >65 years, renal disease, and renal impairment, and 7 (58%) had active disease. By active/progressive disease. contrast, 82% of patients who had received autologous This is a highly representative series for gaining insights stem cell transplantation (ASCT) survived COVID-19, into the impact of COVID-19 in the MM population; all while 41% of the non-ASCT group died (Table 2). Spanish hospitals were put under the governance of the Regarding MM status and comorbidities at the time of Ministry of Health, and the structure of PETHEMA/GEM COVID-19 admission, in patients with active/progressive includes most centers treating MM patients within Blood Cancer Journal Martínez-López et al. Blood Cancer Journal (2020)10:103 Page 7 of 11 Table 3 Patient characteristics at hospital admission for Table 3 continued COVID-19, according to hospital outcome. Patients with hospital outcome, no. (%) Patients with hospital outcome, no. (%) All Patients Patients All Patients Patients (n = 167) discharged who died (n = 167) discharged who died (n = 111)a (n = 56) (n = 111)a (n = 56) Antiviral combination therapy Clinical features during admission % shown as the proportion of None 11 (6) 6 (55) 5 (45) “all” (Hydroxy)chloroquine only 16 (10) 11 (69) 5 (31) Clinical severity (Hydroxy)chloroquine and 47 (28) 30 (64) 17 (36) Mild 26 (16) 24 (92) 2 (8) retrovirals Moderate–severe 128 (77) 85 (66) 43 (34) (Hydroxy)chloroquine and 34 (20) 27 (79) 7 (21) Critical 13 (8) 2 (15) 11 (85) azithromycin Respiratory support (Hydroxy)chloroquine, 51 (31) 33 (65) 18 (35) No supplemental oxygen 39 (23) 36 (92) 3 (8) azithromycin, and retrovirals Supplemental oxygen 78 (47) 56 (72) 22 (28) Other combination therapy 8 (5) 4 (50) 4 (50) Noninvasive ventilation 35 (21) 13 (37) 22 (63) Steroids 83 (50) 51 (61) 32 (39) Invasive ventilation 15 (9) 6 (40) 9 (60) No steroids 84 (50) 60 (71) 24 (29) Laboratory measures at hospital % shown as the proportion of Anti-interleukin-6 receptor 22 (13) 15 (68) 7 (32) admission “all” antibody therapy Absolute neutrophil countb No anti-interleukin-6 145 (87) 96 (66) 49 (34) receptor antibody therapy Above median 80 (50) 46 (57) 34 (43) Steroids and anti- 21 (13) 14 (67) 7 (33) Below median 79 (50) 59 (75) 20 (25) interleukin-6 receptor Lymphocyte countc antibody therapyf Above median 79 (50) 57 (72) 22 (28) No steroids and anti- 143 (87) 95 (66) 48 (34) Below median 78 (50) 47 (60) 31 (40) interleukin-6 receptor Plateletsd antibody therapyf Above median 77 (51) 58 (75) 19 (25) Heparin 109 (65) 77 (71) 32 (29) Below median 75 (49) 44 (59) 31 (41) No heparin 58 (35) 34 (59) 24 (41) D-dimers COVID-19 coronavirus disease 2019, IQR interquartile range. a Includes one patient who remains hospitalized receiving ongoing care. Above mediane 67 (51) 40 (60) 27 (40) b Data missing for eight patients, six who were discharged and two who died in hospital. Below medianf 64 (49) 50 (78) 14 (22) c Data missing for 11 patients, 8 who were discharged and 3 who died in g hospital. Ferritin d Data missing for 16 patients, 10 who were discharged and 6 who died in hospital. Above median 41 (50) 28 (68) 13 (32) e Data missing for 36 patients, 21 who were discharged and 15 who died in the Below median 41 (50) 32 (78) 9 (22) hospital. f Data missing for 85 patients, 51 who were discharged and 34 who died in the COVID-19 therapy % shown as the proportion of hospital. g Data missing for three patients, two who were discharged and one who died in “all” the hospital. (Hydroxy)chloroquine 148 (89) 101 (68) 47 (32) No (hydroxy)chloroquine 19 (11) 10 (53) 9 (47) Azithromycin 91 (54) 62 (68) 29 (32) Spanish territory. The selected time-frame corresponded to peak COVID-19 incidence and population excess No azithromycin 76 (46) 49 (64) 27 (36) mortality in Spain. Additionally, to gain a more precise Antiretrovirals 103 (62) 65 (63) 38 (37) picture of the differential features of these patients, we No antiretrovirals 64 (38) 46 (72) 18 (28) obtained clinical and laboratory characteristics and hos- pital outcomes from a concomitant series of SARS-CoV- 2-infected noncancer patients. To assure uniformity, the Blood Cancer Journal Martínez-López et al. Blood Cancer Journal (2020)10:103 Page 8 of 11 Table 4 Prognostic factors of inpatient mortality in multiple myeloma (MM) patients hospitalized with COVID-19. Prognostic factorsa N Inpatient Unadjusted analysisb Adjusted analysisc GoF c-statistic mortality, no. (%) P value (95% CI) Odds ratio P value Odds ratio P value (95% CI) (95% CI) All patients 167 56 (34) Age > 65 years 112 47 (42) 3.7 (1.6–8.3) 0.002 3. (1.4–8.4) 0.006 NA NA Males 95 39 (41) 2.3 (1.1–4.4) 0.018 3.8 (1.7–8.4) 0.001 NA NA MM comorbidities and status at hospital admission for COVID-19 Renal disease 32 19 (59) 3.9 (1.7–8.7) <0.001 4.6 (1.9–11.3) <0.001 NA NA Hypertension 67 28 (42) 1.8 (0.96–3.5) 0.065 1.7 (0.8–3.5) 0.18 NA NA Active disease or progression 43 21 (49) 2.4 (1.2–4.9) 0.015 2.7 (1.2–6.0) 0.017 NA NA Reference model: 0.7 0.79 (0.72–0.86) MM features at diagnosis and treatment Monoclonal component, 83 22 (27) 0.5 (0.3–1.0) 0.05 0.6 (0.3–1.3) 0.18 0.6 0.80 (0.73–0.87) immunoglobulin G Renal disease at diagnosis 45 23 (51) 2.8 (1.4–5.7) 0.004 1.3 (0.4–3.7) 0.6 0.8 0.79 (0.72–0.86) Diagnosis in 2020 (time since 25 12 (48) 2.0 (0.9–4.9) 0.10 2.7 (0.7–5.8) 0.19 0.5 0.79 (0.72–0.86) diagnosis ≤ 3 months) Prior stem cell transplantation 51 9 (17) 0.3 (0.1–0.7) 0.004 0.6 (0.2–1.7) 0.4 0.4 0.79 (0.71–0.86) CI confidence interval, COVID-19 coronavirus disease 2019, GoF goodness-of-fit, NA not applicable. a Predefined set of well-established prognostic factors assessed before admission; all variables were dichotomized according to standard categories. b Crude odds ratio and 95% confidence interval. c Each logistic model included age, sex, myeloma status, comorbidities (hypertension, renal disease) at diagnosis of COVID-19 (reference model), and one variable from the “Multiple myeloma at diagnosis and treatment” set. Calibration and discrimination of the models were assessed with the Hosmer–Lemeshow goodness-of-fit test (GoF P value) and the c-statistic. study focused solely on patients requiring hospital study bias with the QIPS tool for prognostic factors studies admission—we did not include MM patients who were and concluded that the reported results are unlikely to be treated as outpatients for COVID-19 and either was PCR- distorted, spurious, or biased23. The independent prog- positive or just had COVID-19-compatible symptoma- nostic factors that we identified were measurements tology since, in such a population, obtaining a complete available prior to hospital admission. A prognostic model case history could not be guaranteed. including these factors could be developed to risk-stratify Participating centers covered the whole geographical MM patients and to implement preventive strategies in the territory of Spain, although half of the cases were reported event of a healthcare crisis. from the Madrid region, where a higher number of As noted, inpatient mortality was higher in MM vs COVID-19 cases were confirmed at the national level. A noncancer patients. This occurred despite the similarity large number of participating hospitals (n = 73) ensured a between groups in the clinical severity of COVID-19. representative mix of cases. The central data-monitoring Rates of comorbidities appeared comparable between unit worked closely with local sites to confirm data and on groups, except for renal disease, which has emerged as a patient follow-up. Participating centers were long-term critical feature in the outcome of SARS-CoV-2-infected contributors to the PETHEMA/GEM research program; MM patients. Interestingly, the presence of lymphopenia local data extractors at each site implemented common and higher levels of D-dimers and ferritin, laboratory data element definitions. findings associated with higher mortality rates in several The selection of prognostic factors for estimating adjus- global COVID-19 patient series2,4,24,25, were more com- ted associations with inpatient mortality rate was based on mon in MM vs noncancer patients. Nevertheless, the clinical credibility and restricted in number to limit model mortality rate observed for MM patients is similar to that overfitting. The assessment of internal validity confirmed reported in patients with other cancers (28.6%) and the stability of the model, and that factor estimates were hematologic diseases (33%), as well as immunocompro- realistic for similar future patients. We assessed the risk of mised patients (27.8%)5,7,26,27. Blood Cancer Journal Martínez-López et al. Blood Cancer Journal (2020)10:103 Page 9 of 11 The higher inpatient mortality rate in MM vs noncancer In conclusion, MM patients hospitalized with SARS- patients can be explained through patient-related factors CoV-2 infection have a higher inpatient mortality rate and MM features, besides the biology of the disease itself. than noncancer patients. This had been suspected since Regarding patient-related factors, age >65 years and male the beginning of the COVID-19 pandemic, and conse- sex emerged as two of the most important risk factors quently national and international societies have pub- predicting inpatient mortality. Notably, and in contrast to lished or made general recommendations about optimal noncancer COVID-19 populations, in which 75 years has management of MM patients. Our study assessed the been reported as the cutoff predicting peak risk of death, impact of prior MM treatments based on immunomo- the mortality rate in MM did not differ above 65 years of dulatory drugs, proteasome inhibitors, or monoclonal age, suggesting that there are other factors intrinsic to antibodies, and none of these influenced in the inpatient MM that influence the outcome. Regarding MM features, mortality rate. Although we cannot answer the question renal insufficiency is a major hallmark of MM (one of the of whether to treat or not—and how—during the pan- four myeloma-defining events); in the present series, its demic, we can stress the need for appropriate disease presence at hospital admission for COVID-19 emerged as control in all patients, to minimize hospital visits, parti- the most important factor for survival (OR = 3.8). Renal cularly in the most vulnerable MM populations such as impairment at diagnosis was also an adverse factor, but male patients, those aged >65 years (who are already at this significance was lost in the adjusted multivariate high risk, even at this relatively young age), and those with analysis. However, the strength of association with inpa- renal impairment, which is the most critical factor. tient mortality of renal insufficiency at admission Accordingly, MM patients requiring treatment due to increased, and the independent prognostic value was active disease should be tested for COVID-19 by PCR and, retained in the multivariate analysis. Indeed, kidney dis- if negative, we should proceed to optimize treatment in ease as a prognostic factor predicting increased inpatient the most protective environment possible. In patients mortality rate in a general patient population has been with the non-active disease, MM treatment could be previously described in a single-center study in China28. postponed for 1–2 months; however, if patients are By contrast, hypertension as a comorbidity was not a receiving maintenance therapy, treatment should con- significant factor in the adjusted model. Other MM fea- tinue. Further research is required to identify the most tures such as high-risk cytogenetics and immunoparesis effective treatments for COVID-19, as well as vaccination did not influence the outcome; the latter is of interest strategies that should be employed as early as possible in since humoral immunodeficiency, which is common in the most vulnerable patients. Future studies should vali- MM, would be considered a potential factor for elevated date criteria for stratifying MM patients who are at greater risk. Additionally, we observed that IgG isotype and stage risk in a healthcare crisis such as the COVID-19 pan- I disease were associated with lower mortality. demic and who could benefit from reducing their risk- Regarding MM status at hospital admission for COVID- exposure through a more stringent preventive strategy. 19, as would be expected, the presence of uncontrolled disease had a detrimental effect on survival (OR = 2.6) and Acknowledgements was one of the independent factors predicting outcome. We would like to thank Roberto Maldonado from the PETHEMA Foundation for This would help explain the significantly higher mortality data monitoring and technical support, and other members of the Grupo Español de Mieloma (GEM) participating in the study. Following are the rate observed in patients diagnosed at the time of pandemic hospital and corresponding investigators: Hospital Universitario 12 de Octubre, (January–April 2020), since in most their disease was not Mario Rodriguez; Hospital Universitario 12 de Octubre, Nerea Castro; CA de adequately controlled; it is well known that newly diagnosed Palencia, José María Alonso Alonso; Hospital Universitario de Guadalajara, Dunia de Miguel Llorente; Hospital Universitario la Princesa, Adrián Alegre; patients have an elevated risk of mortality due to infectious Hospital Clínico San Carlos, Belen Iñigo; Complejo Asistencial de Avila, complications in the first 3 months of treatment. The Abelardo Bárez García; Hospital Universitario Mútua de Terrassa, Josep M. Martí benefit of disease control (complete/partial response) in the Tutusaus; Complejo Hospitalario Universitario A Coruña, Ana María Vale López; Complejo Hospitalario Universitario de Albacete, Ángela Ibáñez García; context of COVID-19 would likewise explain the favorable Complejo Hospitalario de Ourense, José Angel Méndez Sánchez; Complejo outcome observed in patients who had previously received Hospitalario de Pontevedra, Ana María Dios Loureiro; Hospital de Txagorritxu, ASCT since most of them were in response. Ernesto Pérez Persona; Hospital Universitario Doctor Peset, Javier de la Rubia; Hospital Universitari Joan XXIII de Tarragona, María Lourdes Escoda Teigell; ICO The non-participation of MM patients with unconfirmed Girona, Yolanda González Montes; MD Anderson Cancer Center, Rebeca COVID-19 or who were not hospitalized is a limitation in Iglesias del Barrio; Complejo Hospitalario de Navarra, Mª Ángeles Goñi Herranz; terms of the broader applicability of the study results. Hospital del Sureste, Rosalía Bustelos; Complejo Hospitalario de Cáceres, Carmen Cabrera Silva; Hospital Universitario la Fe, Mario Arnao Herraiz; Further studies could be conducted to implement our Complejo Hospitalario Universitario Santiago, Marta Sonia González Pérez; analysis protocol in a less restricted population of patients. Hospital Virgen de los Lirios, Alcoy, Carmen Amorós Pérez; Hospital del Mar, Another limitation is that our study is not able to identify Eugenia Abella Monreal; Hospital San Pedro, María Josefa Nájera Irazu; Hospital Sant Joan de Déu, Cristina Motlló; Hospital Virgen de la Concha, Roberto optimal management and treatment of newly diagnosed Hernández; Hospital Universitario Miguel Servet, Ana Cristina Godoy Molías; MM patients with COVID-19. Hospital Universitario Marqués de Valdecilla, Noemi Fernandez Escalada; Blood Cancer Journal Martínez-López et al. Blood Cancer Journal (2020)10:103 Page 10 of 11 Hospital General Universitario de Alicante, Natalia Gutiérrez Pastor; Hospital del E.P.P. has received honoraria for advisory boards from Amgen and for lectures Bierzo, Erik de Cabo López; Hospital Universitario de Canarias, Miguel T. from BMS/Celgene and Janssen. C.F.d.L. has received honoraria for advisory Hernández; Hospital Universitario Quirón Madrid, Carmen Martínez Chamorro; boards from BMS/Celgene, Amgen, Takeda, and Janssen. L.F.C. reports Hospital San Jorge, Ilda Maria Murillo; Hospital Universitario HM Sanchinarro, honoraria for lectures from and membership on advisory boards with Celgene, Jaime Pérez de Oteyza; Hospital General de Ciudad Real, Maria Del Carmen Janssen, Roche, Novartis, Bristol–Myers Squibb, Amgen, Takeda, Pfizer, Incyte, Calle Primo; Hospital Santa Bárbara de Soria, Carlos Aguilar; Hospital and AbbVie. The remaining authors declare that they have no conflict of Universitario Virgen del Rocío, Marta Reinoso Segura; Hospital Universitario de interest. Fuenlabrada, Pilar Bravo; Hospital Clínico de Valladolid, Alfonso García de Coca; Hospital Universitario de Cabueñes, María Esther González García; Hospital Clínico Universitario Lozano Blesa, Ana Isabel Pimentel; Hospital de Jaén, María Publisher’s note Magdalena Anguita-Arance; Hospital Universitario Infanta Elena, Daniel Naya Springer Nature remains neutral with regard to jurisdictional claims in Errea; Hospital del Tajo, Elena Ruiz Sainz; Hospital de Vinalopó, Beatriz Sánchez published maps and institutional affiliations. Quiñones; Hospital Universitario Virgen de Valme, Eduardo Ríos; Hospital Universitario Virgen de la Arrixaca, Angela Carrillo Molina; Hospital Universitario Reina Sofía, Miguel Ángel Álvarez; Hospital Universitario Lucus Augusti, Received: 31 July 2020 Revised: 6 August 2020 Accepted: 6 August 2020 Esperanza Lavilla; Hospital Son Llatzer, Joan Bargay Lleonart; Hospital Universitario Virgen de la Victoria, Ricarda García Sánchez; Hospital Universitario central de Asturias, Ana Pilar González. We thank Steve Hill, PhD, of FireKite, an Ashfield company, part of UDG Healthcare plc, for editorial assistance in the preparation of this paper for submission, which was funded References by PETHEMA/GEM. 1. World Health Organization. Novel Coronavirus–China. World Health Organi- zation. https://www.who.int/china/news/detail/09-01-2020-who-statement- regarding-cluster-of-pneumonia-cases-in-wuhan-china (WHO, 2020). Author details 2. Zhou, F. et al. Clinical course and risk factors for mortality of adult inpatients 1 Hospital Universitario 12 de Octubre, i+12, Complutense University, CNIO, with COVID-19 in Wuhan, China: a retrospective cohort study. Lancet 395, CIBERONC number CB16/12/00369, Madrid, Spain. 2Hospital Universitario de 1054–1062 (2020). Salamanca, Instituto de Investigación Biomédica de Salamanca (IBSAL), Centro 3. Wu, C. et al. Risk factors associated with acute respiratory distress syndrome de Investigación del Cancer (IBMCC-USAL, CSIC), CIBER-ONC number CB16/12/ and death in patients with coronavirus disease 2019 pneumonia in Wuhan, 00233, Salamanca, Spain. 3Hospital Universitario Gregorio Marañón, Instituto de China. JAMA Intern. Med. 180, 934–943 (2020). Investigación Sanitaria Gregorio Marañón (IiSGM), Madrid, Spain. 4Hospital 4. Guan, W.-J. et al. Clinical characteristics of coronavirus disease 2019 in China. N. Durán i Reynals—ICO, L´Hospitalet, Spain. 5Hospital Universitario Infanta Engl. J. Med. 382, 1708–1720 (2020). Leonor, Madrid, Spain. 6Hospital Universitario La Paz, Madrid, Spain. 7Hospital 5. Zhang, L. et al. Clinical characteristics of COVID-19-infected cancer patients: a Universitario Príncipe de Asturias, Madrid, Spain. 8Hospital Universitario Puerta retrospective case study in three hospitals within Wuhan, China. Ann. Oncol. de Hierro, Madrid, Spain. 9Hospital Universitario Fundación Jiménez Diaz, 31, 894–901 (2020). Madrid, Spain. 10Hospital Severo Ochoa, Madrid, Spain. 11Hopital Universitario 6. Liang, W. et al. Cancer patients in SARS-CoV-2 infection: a nationwide analysis Vall d´Hebron, Barcelona, Spain. 12Hospital Universitario Ramon y Cajal, Madrid, in China. Lancet Oncol. 21, 335–337 (2020). Spain. 13Hospital Universitario de Donostia, Donostia, Spain. 14Hospital Clínic 7. Lee, L. Y. W. et al. COVID-19 mortality in patients with cancer on che- IDIBAPS, Barcelona, Spain. 15Institut Català d’Oncologia i Institut Josep Carreras, motherapy or other anticancer treatments: a prospective cohort study. Lancet Badalona, Spain. 16Hospital de León, León, Spain. 17Hospital de Galdakao, 395, 1919–1926 (2020). Vizcaya, Spain. 18Hospital Infanta Sofía, Madrid, Spain. 19Hospital de Alcorcón, 8. Blimark, C. et al. Multiple myeloma and infections: a population-based study Madrid, Spain. 20Hospital de Cruces, Bilbao, Spain. 21Hospital Virgen de la Salud, on 9253 multiple myeloma patients. Haematologica 100, 107–113 (2015). Toledo, Spain. 22Hospital Universitario 12 de Octubre, Instituto de Investigación 9. Basler, M., Lauer, C., Beck, U. & Groettrup, M. The proteasome inhibitor bor- Sanitaria i + 12, RedSAMID RD16/0022/0011, Madrid, Spain. 23Clinica tezomib enhances the susceptibility to viral infection. J. Immunol. 183, Universidad de Navarra, Centro de Investigación Medica Aplicada (CIMA), 6145–6150 (2009). Instituto de Investigación Sanitaria de Navarra (IDISNA), CIBER-ONC number 10. Chen, M. et al. Immunomodulatory drugs and the risk of serious infection in CB16/12/00369, Pamplona, Spain multiple myeloma: systematic review and meta-analysis of randomized and observational studies. Ann. Hematol. 97, 925–944 (2018). Author contributions 11. Feng, X. et al. Targeting CD38 suppresses induction and function of T reg- J.M.-L., M.-V.M., J.d.l.C., and J.S.-M. had full access to all of the data in the study ulatory cells to mitigate immunosuppression in multiple myeloma. Clin. Cancer and take responsibility for the integrity of the data and the accuracy of the Res. 23, 4290–4300 (2017). data analysis. Concept and design: J.M.-L., M.-V.M., J.d.l.C., and J.S.-M. 12. Nahi, H. et al. Infectious complications and NK cell depletion following dar- Acquisition, analysis, or interpretation of the data: all authors. Statistical atumumab treatment of multiple myeloma. PLoS ONE 14, e0211927 (2019). analysis: J.d.l.C. Drafting of the paper: J.M.-L., M.-V.M., J.d.l.C., and J.S.-M. Critical 13. Zhang, X. et al. First case of COVID-19 in a patient with multiple myeloma revision of the paper for important intellectual content: all authors. successfully treated with tocilizumab. Blood Adv. 4, 1307–1310 (2020). 14. Cook, G. et al. Real-world assessment of the clinical impact of symptomatic Conflict of interest infection with severe acute respiratory syndrome coronavirus (COVID-19 dis- M.-V.M. has received honoraria for lectures and participation in advisory boards ease) in patients with multiple myeloma receiving systemic anti-cancer from Janssen, Celgene-BMS, Amgen, Takeda, Abbvie, GSK, Adaptive, Roche, therapy. Br. J. Haematol. https://doi.org/10.1111/bjh.16874 (2020). Seatle Genetics, Pfizer, and Regeneron. J.M.-L. has received honoraria for 15. Terpos, E. et al. Management of patients with multiple myeloma in the era of participation in advisory boards from Novartis, Roche, BMS, Adaptive, Incyte, COVID-19 pandemic: a consensus paper from the European Myeloma Net- Amgen, and Janssen-Cilag. J.S.-M. has received honoraria for lectures and work (EMN). Leukemia. https://doi.org/10.1038/s41375-41020-40876-z (2020). advisory boards from Amgen, Bristol–Myers Squibb, Celgene, Janssen, Merck, 16. Di Ciaccio, P. et al. Australian and New Zealand consensus statement on the Novartis, Takeda, Sanofi, and Roche. J.B. has received honoraria for lectures and management of lymphoma, chronic lymphocytic leukaemia and myeloma advisory boards from Janssen, Celgene, Amgen, Takeda, and Oncopeptides. during the COVID-19 pandemic. Intern. Med. J. https://doi.org/10.1111/ J.J.L. has received honoraria for lectures and advisory boards from Janssen, imj.14859 (2020). Celgene, Amgen, and Takeda. A.L. has received honoraria for advisory boards 17. Hungria, V. et al. Managing patients with multiple myeloma during the from Celgene, Amgen, and Janssen. P.M.-B. has received honoraria for advisory COVID-19 pandemic: recommendations from an expert panel—ABHH boards from Amgen. A.S. has received honoraria for advisory boards from Monoclonal Gammopathies Committe. Hematol. Transfus Cell Ther. https://doi. Takeda, BMS, MSD, Sanofi, Roche, Novartis, Janssen, and Sandoz, and for a org/10.1016/j.htct.2020.1005.1001 (2020). consultancy from Takeda, BMS, Novartis, Celgene, Janssen, Gilead, and Sanofi, 18. Mian, H. et al. Caring for older adults with multiple myeloma during the and has received honoraria as a member of a Speakers Bureau for Takeda. COVID-19 pandemic: perspective from the international forum for optimizing F.E. has received honoraria for advisory boards from Celgene and Amgen. care of older adults with myeloma. J. Geriatr. Oncol. 11, 764–768 (2020). Blood Cancer Journal Martínez-López et al. Blood Cancer Journal (2020)10:103 Page 11 of 11 19. Al Saleh, A. S., Sher, T. & Gertz, M. A. Multiple myeloma in the time of COVID- 24. Bhatraju, P. K. et al. Covid-19 in critically Ill patients in the Seattle region—case 19. Acta Haematol. https://doi.org/10.1159/000507690 (2020). series. N. Engl. J. Med. 382, 2012–2022 (2020). 20. Barragan, E. et al. Prognostic value of FLT3 mutations in patients with acute 25. Grasselli, G., Pesenti, A. & Cecconi, M. Critical care utilization for the COVID-19 promyelocytic leukemia treated with all-trans retinoic acid and anthracycline outbreak in Lombardy, Italy: early experience and forecast during an emer- monochemotherapy. Haematologica 96, 1470–1477 (2011). gency response. J. Am. Med. Assoc. https://doi.org/10.1001/jama.2020.4031 21. Moons, K. G. et al. Transparent reporting of a multivariable prediction model (2020). for individual prognosis or diagnosis (TRIPOD): explanation and elaboration. 26. Martín-Moro, F. et al. Survival study of hospitalised patients with concurrent Ann. Intern. Med. 162, W1–W73 (2015). COVID-19 and haematological malignancies. Br. J. Haematol. https://doi.org/ 22. Sauerbrei, W., Taube, S. E., McShane, L. M., Cavenagh, M. M. & Altman, D. G. 10.1111/bjh.16801 (2020). Reporting recommendations for tumor marker prognostic studies (REMARK): 27. Fernández-Ruiz, M. et al. COVID-19 in solid organ transplant recipients: a an abridged explanation and elaboration. J. Natl Cancer Inst. 110, 803–811 single-center case series from Spain. Am. J. Transplant. https://doi.org/10.1111/ (2018). ajt.15929 (2020). 23. Hayden, J. A., van der Windt, D. A., Cartwright, J. L., Côté, P. & Bombardier, C. 28. Cheng, Y. et al. Kidney disease is associated with in-hospital death of patients Assessing bias in studies of prognostic factors. Ann. Intern. Med. 158, 280–286 with COVID-19. Kidney Int. 97, 829–838 (2020). (2013). Blood Cancer Journal