Clinical features associated with COVID-19 outcome in multiple myeloma: first results from the International Myeloma Society data set

In the current severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2)¹  pandemic, known as COVID-19, cancer represents a major risk factor^2-4  for COVID-19–associated death. Cancer patients with COVID-19 represented 8.3% of deaths in the New York city area, and 7.6% of deaths from the Wuhan area in China.^5,6  An even higher fatality rate (20.3%) was observed in Italy.⁷  A recent study aimed at identifying risk factors for death in cancer patients developing COVID-19 interestingly (but not surprisingly) reported age over 65 years as a risk factor. In this study, treatment with checkpoint inhibitors was a risk factor, but not ongoing chemotherapy.⁵ 

Multiple myeloma (MM) is a hematological cancer involving plasma cells, mostly within the bone marrow.⁸  Apart from the specific cancer-related symptoms, most patients display immunosuppression⁹  involving both the B- and T-cell compartments. Infections are common disease complications,¹⁰  and unfortunately remain a major cause of death. Furthermore, corticosteroids, and especially dexamethasone, are used as treatment throughout the disease course, usually at high doses.^11,12  This MM therapy may increase the immunosuppression observed in patients with MM, though low doses seem to improve mortality in hospitalized patients. MM usually affects the elderly population, a more vulnerable group of patients due to immunosenescence together with other comorbidities. In addition, younger MM patients are usually treated with high-dose chemotherapy followed by autologous stem cell transplant,¹³  with high infection susceptibility during the 3-month period following transplant.¹⁴  For all of these reasons, MM could theoretically represent a high risk factor for poor outcomes with COVID-19.¹⁵ 

In this international study, we have collected data and investigated the risk and outcome of COVID-19 infection in MM patients globally, both to evaluate the death rate and to identify potential risk factors that could be modified to improve patient outcomes during the current pandemic and in the future. With this in mind, we have predominantly focused our analysis on patients requiring hospitalization for COVID-19 infection.

The international MM COVID-19 data set created by the International Myeloma Society (IMS) has retrospectively collected data for 650 patients with a plasma cell disorder from 10 different countries and multiple centers. All patients in the study had confirmed positive SARS-CoV-2 tests, according to the protocols in their respective countries. A questionnaire created by IMS was shared with participant institutes/investigators, and all required information was collected by the participating investigators. Data cleaning, preprocessing, and quality control were completed before the data analysis. COVID-19 outcome is defined as recovery from the virus and discharge from hospital or death due to COVID-19. Those patients who required an ongoing treatment at the time of data collection, unknown COVID-19 outcomes, and not hospitalized, were excluded from the cohort for statistical analysis. The IMS COVID-19 data set is reviewed by the New England Institutional Review Board based on federal regulation 45 CFR 46 and associated guidance; it is determined to be human subjects research exempt and approved for a waiver of authorization. All data captured are deidentified and comply with Health Insurance Portability and Accountability Act (HIPAA) Safe Harbor regulations.

All statistical analyses were performed using R (www.r-project.org). Descriptive statistics for demographic information and clinical variables are reported. Parametric 2-group comparison was used for age, univariate logistic regression was used to evaluate the association between COVID-19 outcome and variables, and odds ratios (ORs) with 95% confidence intervals (CIs) were estimated. Multivariate analysis was performed using only variables that were associated with the outcome on univariate analysis.

Overall, 650 patients with a plasma cell disorder and COVID-19 infection are included in this study, with the majority of patients being from Spain (28.62%), France (28.46%), the United States (19.38%), and the United Kingdom (14.77%) (Table 1; Figure 1A; supplemental Table 1 [available on the Blood Web site]). Median age was 69 years (range, 34-92 years), and 58.5% of patients were male (Table 2). The vast majority of patients (95.5%) had MM, whereas 29 patients (4.5%) had another plasma cell disorder. The MM immunoglobulin subtype included 55% immunoglobulin G (IgG), 21% IgA, and 20% light chain (Table 2). Patients were equally distributed between International Staging System (ISS) stages 1 to 3, with 32.1% of patients having high-risk cytogenetics and 26.5% having renal dysfunction. Fifty-four percent of patients were receiving first-line therapy, whereas 23.5% of patients had 3 or more lines of therapy. Additional demographic data are presented in Table 1, Table 2, and supplemental Tables 1 and 2.

Thirty-three percent of patients died following COVID-19 diagnosis. The death rate increased from 4% for those who were outpatients to 31% for hospitalized patients not on ventilator support to 80% for patients on ventilator support (Table 1; supplemental Table 1). The variability in death rates across 4 major countries is shown in Figure 1B and Table 1. The death rate in patients with other plasma cell disorders was 31% (9 of 29).

We have further focused on analyzing the hospitalized patients, for whom the mortality rates ranged from 27% in Germany, Italy, Brazil, The Netherlands, Portugal, and Turkey to 57% in the United Kingdom (Table 2). Age was significantly associated with COVID-19 outcome (P < .001). The estimated probability of death for 40-, 60-, and 80-year old patients was 17.76%, 31.43%, and 49.3%, respectively (Figure 1C). Forty percent of hospitalized patients were female, and in contrast to prior reports, sex was not associated with outcome. Of note, the mean age for male patients (69 years) was significantly lower than female patients (71.5 years) (P = .01).

Of the patients with available data, those diagnosed with MM in 2019 or 2020 accounted for 35.6% of the cohort, and those with ≤1 line of therapy accounted for 54% of the cohort; 32.9% of patients were diagnosed on or before 2015. Neither time from diagnosis nor number of prior lines of treatment had any impact on outcome of COVID-19 infection. Immunoglobulin type distribution was similar to the general MM population, and isotypes were not associated with outcome. Univariate analysis identified ISS3 vs ISS1 (P = .04), high-risk disease [del 17p, t(4;14), amp 1q or t(14;16)] (P = .07), renal disease (P = .007), inadequate MM control (active disease or progressive disease [PD] vs complete response) (P = .01), and 1 or more comorbidities (P = .04) correlating with higher rates of death (Figure 1D).

Eighty-seven percent of patients who had a known treatment status were on active MM therapy at the time of COVID-19 diagnosis, and 89% patients had their therapy held during COVID-19 diagnosis and management (Table 2; supplemental Table 2). A history of prior transplant or transplant within a year of COVID-19 diagnosis did not impact outcome. In fact, patients with a history of stem cell transplant within a year of COVID-19 diagnosis had a lower death rate; however, this difference was cofounded by a 10-year age difference between transplant and nontransplant patients, and was not observed when adjusted for age. Similarly, we did not observe any significant difference in outcome from COVID-19 infection whether patients underwent transplant within 6 months or >6 months before their COVID-19 diagnosis. Approximately 86% of patients had prior exposure to proteasome inhibitors (PIs), 80% to immunomodulatory (IMiD) agents, and 30% to anti-CD38 antibody. In univariate analysis, prior PI, IMiD, or anti-CD38 treatment was not associated with outcome. Although univariate analysis showed that patients who were receiving IMiD treatment at the time of COVID-19 diagnosis had decreased mortality compared with patients not on IMiDs, multivariate analysis failed to identify IMiD or any of these features as being related with outcome (Table 3). We did not observe any significant correlation between active PI, IMiD, anti-CD38 monoclonal antibody, alkylating agents, steroids, or other treatments (venetoclax, 96-hour infusional regimens, bispecific T-cell engagers, belantamab, chimeric antigen receptor T cell, elotuzumab, histone deacetylase) and the COVID-19 outcome.

The treatments of COVID-19 were very heterogeneous, with the most frequent therapies including combination strategies (70%), antibiotics (14%), and hydroxychloroquine (10%). No therapies for COVID-19 appeared to be protective or associated with worse outcomes.

Of the aforementioned variables that were associated with an increased risk of death by univariate analysis, only age (OR = 1.04; 95% CI, 1.01-1.08), high-risk MM [del 17p, t(4;14), amp 1q or t(14;16)] (OR = 2.35; 95% CI, 1.20-4.66), renal disease (OR = 2.71; 95% CI, 1.23-6.08), and active or progressive MM (OR = 1.91; 95% CI, 0.96-3.81) remained as independent predictors of adverse outcome on multivariate analysis (Table 3).

The COVID-19 infection has affected patients globally, with high incidence in Europe and the Americas. The disease has involved patients of all age groups; however, heterogeneity in outcome of COVID-19 infection has been observed to be associated with comorbidities, racial differences, as well as individual characteristics such as smoking.^16-18  Of note, the presence of comorbidities has been extensively studied to identify patients at greater risk of infection and those with worse outcome. In this regard, our current study focuses on a single type of cancer, MM, to understand both impact and outcome of patients when they develop COVID-19 infection. As MM patients have hallmark immunosuppression, it is of great interest to understand the impact of both the disease and its treatment, that is, the immunosuppressive effects of high-dose therapy with autologous transplantation, as well as novel targeted therapies.

Here, we report data primarily from 4 countries (Spain, France, the United Kingdom, and the United States) having a high prevalence of COVID-19 infection and with the highest frequencies of COVID-19 in MM patients. Differing access to testing likely led to the majority of outpatients (55%) coming from the United States.

Interestingly, recent data from institutions in New York City showed that ∼19% of 127 patients with COVID-19 actually had MM precursor conditions (plasmacytoma, monoclonal gammopathy of undetermined significance, smoldering MM).¹⁹  Based on the fact that data collection was feasible predominantly in patients who were hospitalized, our study has focused on hospitalized patients and their outcomes, and is unable to provide insight into a question regarding susceptibility and outcome of COVID-19 in patients with MM precursor conditions.

According to Surveillance, Epidemiology, and End Results Program (SEER) data, 22% of the patients with myeloma had their diagnosis of MM in 2019 or 2020. In our cohort, 36% of the patients with COVID-19 infection were diagnosed with MM in 2019 or 20, suggesting higher susceptibility even in earlier stages of the disease. Even if we account for variability in diagnosis and selection of cases, there seems to be no evidence of increased hospitalizations in advanced multiply relapsed MM, as was initially postulated. However, differing access to COVID-19 testing or care between newly diagnosed and established patients at academic hospitals collaborating with the IMS may impact these results. Of note, a prior report noted that hypogammaglobulinemia (IgG < 700 mg/dL) was not associated with outcomes of COVID-19 infection, but severe hypogammaglobulinemia (IgG < 400 mg/dL) was associated with death.²⁰  The impact of disease stage and clinical immunoparesis is being evaluated in prospective studies. The sex distribution in this analysis was similar to the general incidence of MM, and thus suggests a similar susceptibility to COVID-19 infection between male and female patients with MM. A clear association between age and outcome was confirmed in these patients, as is true in other settings.

The differences reported here between various countries reflects, at least in part, differing diagnostic and management practices, as well as the resources available and used in management of COVID-19 at the height of the pandemic, as well as referral patterns. Health system differences between countries may influence the ability to seek or obtain SARS-CoV-2 testing and health care, including admission to hospital. Health care providers should always consider local COVID-19 prevalence and local guidelines, and recommendations made here should only be used as a reference. For example, the number of patients who received ventilator support differed from 7% to 31%, and there are also differences in outpatient management vs hospitalization. Nonetheless, hospitalized patients can be considered to have more severe COVID-19–related complications requiring more intensive support.

Our data suggest higher mortality in hospitalized patients with MM and COVID-19 infection than nonmyeloma patients. A recent study from Spain found a higher mortality rate in MM patients with COVID-19 (34%) compared with age- and sex-matched non-MM patients with COVID-19 (23%).²¹  A recent publication from France confirmed overall mortality in all hospitalized patients with COVID-19 to be 16%, which is significantly lower than mortality observed in hospitalized MM patients in France (39%).²²  Our data clearly suggest a lack of relationship between prior lines of therapy or prior type of therapy, or those receiving active MM therapy at the time of diagnosis with COVID-19 and outcome. Interestingly, neither past nor recent high-dose therapy had significant impact on outcome. These data indicate that postponing indicated MM therapies, including high-dose therapy, may not be necessary during the COVID-19 pandemic. Within the limitation of our sample size and retrospective nature of the study, there is no clear suggestion for the need to avoid any specific MM treatment. Importantly, patients with good MM control (complete response) had superior outcome compared with those with relapsed disease or partial response. A similar finding was observed in a study of 928 cancer patients with COVID-19, in which patients with active cancer (progressing vs remission) had an adjusted OR of 5.2 for 30-day mortality, but there was no association with recent noncytotoxic therapy nor recent cytotoxic systemic chemotherapy.²  As most patients receive dexamethasone as part of combination therapy, judging its independent impact on outcome was not possible, which is important because a recent report suggests superior survival for those COVID-19 patients given dexamethasone as part of their COVID-19 therapy.²³ 

Our multivariate model identifies age, high-risk or progressive MM, and presence of renal disease as indicators of poor outcome. MM therapy to achieve deep response may therefore also protect patients from adverse outcome from COVID-19 infection. Although little is known about the recovery of patients with MM from COVID-19 infection, Wang et al found that the median time to polymerase chain reaction (PCR) negativity was 43 days (range, 19-68 days) from initial positive PCR.²⁰  Interestingly, 96% of MM patients (22 of 23) developed antibodies to SARS-CoV-2 at a median of 32 days after initial diagnosis.

Based on the observations reported here, young patients with high-risk and/or active MM need to receive therapies to control their disease, which will also improve their outcome if infected with COVID-19. For elderly patients with a higher death rate from COVID-19, disease control is also beneficial, but may be achieved using regimens that decrease frequency of office visits (eg, oral drug) in order to avoid exposure to the virus. Importantly, continued therapy including steroids and high-dose therapy are not contraindicated, and in fact should be continued to achieve better MM control, which is associated with improved outcome even with COVID-19 infection.

In conclusion, this collaborative international effort provides the first large-scale analysis and initial IMS suggestions on the management of, and outcomes for, patients with MM during the current COVID-19 pandemic (Table 4). As the pandemic and data accumulation rapidly increase, we need prospective studies on treatment options and additional patient characteristics to further understand the variables associated with COVID-19–associated death in MM patients. The high mortality noted in MM patients highlights the critical importance of measures to prevent contracting COVID-19, such as social distancing and wearing masks, in patients with MM.

The authors thank all medical centers and health workers for their fight against the global pandemic.

This work was supported by the efforts of the International Myeloma Society and its members globally.

Explanation of novelty: This study investigated the characteristics and outcomes of MM patients with COVID-19 infection in 10 countries globally.

Contribution: A.C., M.K.S., K.C.A., J.S.-M., N.C.M., and H.A.-L. designed the research and analyzed the data; and all authors contributed to data collection and wrote the manuscript.

Conflict-of-interest disclosure: A.C. is consultant to/on the advisory boards for Janssen, Celgene, Novartis, Amgen, Bristol Myers Squibb, Karyopharm, Sanofi Genzyme, Seattle Genetics, Oncopeptides, Millennium/Takeda, Antengene, GlaxoSmithKline, and Secura Bio, and has received research funding from Janssen, Celgene, Novartis, Amgen, Pharmacyclics, Seattle Genetics, and Millennium/Takeda. J.M.-L. has received honoraria for participation in advisory boards from Novartis, Roche, Bristol Myers Squibb, Adaptive Biotech, Incyte, Amgen, and Janssen-Cilag. K.W. has received honoraria for participation in advisory boards from Adaptive Biotech, Amgen, Bristol Myers Squibb, Celgene, GlaxoSmithKline, Janssen, Karyopharm, Sanofi, and Takeda, and has received research funding from Amgen, Celgene, Sanofi, and Janssen. C.R. has received honoraria from Akcea and Celgene and research funding from Amgen. K.C.A. is a consultant for Bristol Myers Squibb, Millennium-Takeda, Janssen, Sanofi, Tolero, and Precision Biosciences, and is a scientific founder of OncoPep and C4 Therapeutics. P.M. has received honoraria for participation in advisory boards from Janssen, Celgene/Bristol Myers Squibb, Amgen, Sanofi, and AbbVie. M.-V.M. has received honoraria for lectures and participation in advisory boards from Janssen, Celgene–Bristol Myers Squibb, Amgen, Takeda, AbbVie, GlaxoSmithKline, Adaptive Biotech, Roche, Seattle Genetics, Pfizer, and Regeneron. J.S.-M. has received honoraria for lectures and participation in advisory boards from Amgen, Bristol Myers Squibb, Celgene, Janssen, Merck, Novartis, Takeda, Sanofi, and Roche. N.C.M. is a consultant for Bristol Myers Squibb, Janssen, OncoPep, Amgen, Karyopharm, Legend Biotech, AbbVie, Takeda, and GlaxoSmithKline, and is a scientific founder of OncoPep. The remaining authors declare no competing financial interests.

Correspondence: Hervé Avet-Loiseau, Centre de Recherche en Cancérologie de Toulouse, INSERM U1037, Toulouse, France; e-mail: avetloiseau.herve@iuct-oncopole.fr.