Portal de Periódicos da CAPES

Low mortality from COVID ‐19 infection in patients with B‐cell lymphoma after bispecific CD20xCD3 therapy

2023; Wiley; Volume: 204; Issue: 1 Linguagem: Inglês

10.1111/bjh.19156

1365-2141

Emil Ramsø Kyvsgaard, Caroline Hasselbalch Riley, Michael Roost Clausen, Mads Harsløf, Line Dam Heftdal, Carsten Utoft Niemann, Kirsten Grønbæk, Martin Hutchings, Simon Husby,

COVID-19 Clinical Research Studies

Patients with haematological malignancies and haematopoietic cell transplantation recipients are at increased risk of morbidity and mortality after COVID-19 infection.1-5 Patients treated with chimeric antigen receptor T-cell (CAR-T) therapy for B-cell malignancies have shown poor outcomes after COVID-19, both in case reports and retrospective multicentre studies, resulting in reported COVID-19 attributable mortality rates between 41% and 50%.5-8 A proposed explanation for these poor outcomes is a combination of a cancer-related immunocompromised state and therapy-related side effects, such as B-cell depletion, hypogammaglobulinaemia and neutropenia caused by conditioning regimens for CAR-T-cell therapy. These side effects can also be observed in patients treated for B-cell non-Hodgkin lymphoma (B-NHL) with CD20xCD3 bispecific antibodies such as glofitamab, epcoritamab and mosunetuzumab.9-13 To inform the field of the impact of COVID-19 on patients with B-NHL who have received bispecific antibodies, we here describe the clinical course of COVID-19 in a cohort of 130 Danish patients. Using clinical data from electronic health charts, the national Danish database of microbiology (MiBA) and the national drug- and vaccination registry (SharedMedicineCard), we collected data on date of positive COVID-19 test, quantitative PCR-determined COVID-19 variant, symptoms, reactivations, resolution of disease (either clinical, confirmed by negative PCR test or death), vaccination status and treatment. Disease duration was defined as the time from diagnosis of COVID-19 to clinical resolution. Reactivation was defined as either having a negative PCR test or having been confirmed without symptoms of COVID-19 for more than 2 weeks with a subsequent recurrence of symptoms and a positive PCR test. A total of 130 patients who received bispecific antibodies (glofitamab or epcoritamab) as part of clinical trials from November 2019 to February 2023 in Denmark were assessed. This study entails all patients treated at Rigshospitalet and Vejle hospital with bispecific antibodies during the mentioned period. Of the 130 patients, 109 were alive after 8 March 2020. We identified 43 patients infected with COVID-19, 34 of whom were during CD20xCD3-specific treatment or after end-of-treatment (Table 1). A total of 26 patients (79%) were in complete remission at the time of the COVID-19 diagnosis; 5 patients (15%) had progressive disease, 2 patients (6%) had stable disease and 1 patient had no data on lymphoma status at the time of the COVID-19 diagnosis. Among those who were infected after active antibody treatment ceased, the mean time since the last dose was 469 days and the median time since the last dose was 299 days (range, 3–1380). With regards to COVID-19 symptom burden, using the WHO definition of COVID-19 severity, 2 patients had critical COVID-19, 2 had severe COVID-19 and 30 had non-severe COVID-19, of which 4 patients (12%) were asymptomatic during infection.14 The most common symptoms included fever (n = 15), cough (n = 12) and dyspnea (n = 6). There were 16 patients (48%) admitted to the hospital for the treatment of COVID-19. Of these, two patients required oxygen support. The reasons for hospitalization in our study included fever (n = 9) and suspicion of CRS (n = 7), hypoxia (n = 1) and exacerbation of asthma (n = 1). The median length of hospital stay from the initial infection was 1.5 days (range, 1–15). At admission, disease severity was non-severe in 14 patients and severe in 2 patients. Twelve patients (35%) received no specific treatment for COVID-19, while the most commonly prescribed treatments were sotrovimab in 13 patients (42%) and remdesivir in 9 patients (29%). A total of eight patients receiving only sotrovimab were alive after 6 months. Among those who received only anti-virals (n = 8), two patients had died 6 months after the initial infection; however, only one death was related to COVID-19. Among those who received both anti-virals and sotrovimab (n = 5), one patient died of causes related to COVID-19. Among the 11 patients who received no treatment, no one had died after 6 months. Anti-virals included paxlovid, remdesivir and molnupiravir. At 6 months after the COVID-19 infection, three patients had died, two before recovering from COVID-19. The two cases, where COVID-19 presumably was the cause of death, had a combination of multiple bacterial infections and the progression of malignant disease. The 6-month cumulative incidence of death related to COVID-19 was 6.4% (95% CI 0–14.9, Figure 1), while the cumulative incidence of death overall was 9.6% (95% CI 0–19.9). Of the 25 patients who were alive and had data available at 6-month follow-up after COVID-19 infection, 20 were in complete remission, while the rest (n = 5) had refractory/relapsed lymphoma. The cumulative incidence of death related to COVID-19 is thus much lower than the 41% mortality rate of patients diagnosed with COVID-19 after CAR-T-cell therapy in a study by the EBMT Infectious Diseases Working Party and the EHA Lymphoma Group.6 The infected patients in our study were mostly vaccinated (n = 41, 95%), and among those, we could confirm the variant, mainly those infected with the omicron COVID-19 variant (n = 19, 90%). Estimating variants by infection date (in accordance with the most common circulating variant in Denmark) yielded similar results with mainly omicron infections (n = 38, 88%, Note S1). This may explain the observed difference, as earlier studies have been made in an era where the alpha/beta/delta variants were more common. The cumulative incidence of death related to COVID-19 found in our study is in line with the 9.3% found by Jiménez et al. in a vaccinated cohort of patients with diverse haematological malignancies infected by primarily omicron variants (80%) and the 2% 30-day omicron mortality in CLL found by Niemann et al.15, 16 Our findings may be due to selection bias, as the patients included in our study are clinical trial participants and presumed to have fewer comorbidities and a better performance status compared to a general lymphoma population in a haematology–oncology department.1, 4, 6, 7 Conversely, the patients in our cohort all have all patients with refractory/relapsed B-NHL, have received multiple prior lines of anti-CD20 immunochemotherapy, and thus represent a vulnerable population. Patients in our cohort have less severe disease and lower mortality compared to a recent American study on 22 patients with COVID-19 treated with CD3xCD20 bispecific antibodies conducted by Nachar et al.17 The study by Nachar et al. found 21% of patients to have died from COVID-19. This could be due to a number of factors. While Nachar et al. only had complete data on 19 patients, our study had complete data on all patients, and we were able to cross-reference our patients with all COVID-19 tests conducted in Denmark during this period. This entails that our study is much more likely to identify mild and asymptomatic infections, which is probably underreported in the study by Nachar et al. Furthermore, Nachar et al. lack information on variants, which have a very large influence on mortality. The vaccination rates also differed, with the American cohort having a lower proportion of vaccinated patients (74% having received at least one dose), while in the Danish cohort 97% had received two or more doses. We expect that this makes us able to identify the most realistic COVID-19 mortality in a vaccinated population being infected with currently circulating variants and receiving BsAbs against B-cell lymphomas. Of the patients with COVID-19 during CD20xCD3-specific treatment or after end-of-treatment, 13 out of 34 had at least one reactivation. The mean number of COVID-19 reactivations was 1.2 (95% CI 0.44–2.03), ranging from 0 up to 9. The median time from the onset of reactivation was 25 days from the initial diagnosis. Eight patients were hospitalized as a direct result of their reactivation, and the 6-month cumulative incidence rate of death was 12.5% (95% CI 0–35.4) among these patients. An additional patient had a reactivation 20 days after the first infection and 4 days after the first treatment with bispecific antibodies. Only 9% (1 out of 11) of the patients who tested positive for SARS-CoV-2 before bispecific antibody treatment had a reactivation, compared to 38% of those testing positive after or during bispecific antibody treatment. In an adjusted multivariable regression for age and sex, only hospitalization during the first COVID-19 infection was significantly associated with reactivations (OR: 18.9, 95% CI 2.84–238, p = 0.007). Looking at both white blood cell count and absolute lymphocyte count at baseline, we were unable to find any effect on odds of reactivation. The median time to clinical resolution of COVID-19 was 7 days (IQR: 4–14). The median time to clinical resolution of 7 days was notably shorter when compared to the 20 days found in patients treated with CAR-T-cell therapy.6 To our knowledge, this study presents the most complete data on COVID-19 incidence and severity in patients with relapsed/refractory lymphoma who have received CD3xCD20 bispecific antibodies. The COVID-19 attributable mortality after bispecific CD20xCD3 antibody therapy, when compared to CAR T-cell therapy, was much lower and comparable to the mortality reported in other cohorts of less heavily treated haematological malignancies infected with omicron variants. However, these patients present with an increased number of COVID-19 reactivations. In conclusion, it seems relatively safe to prescribe bispecific CD20xCD3 antibodies for lymphoma for patients who have been vaccinated for COVID-19 with the currently circulating virus variants. Emil R. Kyvsgaard wrote the paper. Emil R. Kyvsgaard, Martin Hutchings and Simon Husby designed the study. Emil R. Kyvsgaard, Simon Husby and Mads Harsløf did the statistical analysis. Caroline Riley, Michael Roost Clausen, Martin Hutchings and Emil R. Kyvsgaard collected data. Line Dam Heftdal contributed data on disease variants and contributed to the analysis. All authors contributed to the interpretation of the data. This study was conducted with funding from ‘Rigshospitalets Research Foundation’, ‘Aase og Ejnar Danielsens Foundation’ and ‘Frk Amalie Jørgensens Memorial Scholarship’. Kirsten Grønbæk received research support from Janssen and is on the advisory board of Nanexa and GSK. Martin Hutchings has a consulting or advisory role at Takeda, Roche and Genmab and has received research funding from Celgene, Genmab, Roche, Takeda and Novartis. Carsten Niemann has received research funding and/or consultancy fees from AstraZeneca, Janssen, AbbVie, Beigene, Genmab, CSL Behring, Octapharma, Takeda and Novo Nordisk Foundation. The laws on handling personal information were obeyed in accordance with the Danish Scientific Ethical Committees Act (Komitéloven) § 20, stk 1. nr. 4. The study has been approved by Datatilsynet under the Capital Region umbrella application (RH-2020-561), and the demands of Datatilsynet concerning documentation of data security have been followed. Consent has been obtained where applicable, else exemption from obtainment of informed consent from these patients has been obtained from the Danish ethical council through the Research Biobank and Clinical Database of patients with Lymphoproliferative Malignancies (LM). No material from other sources has been reproduced. The clinical trial has been registered as DALYCA in pactiuus with the trial number RH-2020-561. Note S1. Table S1. Table S2. Table S3. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.