Long and persistent COVID-19 in patients with hematologic malignancies: from bench to bedside

Abstract

Purpose of review:

Cancer patients, especially those with hematologic malignancies, are at increased risk for COVID-19 related complications and mortality. We describe the incidence, clinical characteristics, risk factors, and outcomes of persistent COVID-19 infection in patients with hematologic malignancies.

Recent findings:

The syndrome of persistent COVID-19 in patients with hematologic malignancies manifests as a chronic protracted illness marked by waxing and waning or progressive respiratory symptoms and prolonged viral shedding. Immunosuppressed patients with lymphoid malignancies may serve as partially immune reservoirs for the generation of immune-evasive viral escape mutants.

Summary:

Persistent COVID-19 infection is a unique concern in patients with hematologic malignancies. While vaccination against SARS-CoV-2 has reduced the overall burden of COVID-19 in patients with hematologic cancers, whether vaccination or other novel treatments for COVID-19 prevent or alleviate this syndrome remains to be determined.

Introduction

Patients with hematologic malignancies are at increased risk of complications due to severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection, with elderly, immunocompromised, and co-morbid patients at the highest risk of poor outcomes.^1–5 In the early stages of the pandemic, reports from several countries showed a higher incidence of severe disease and mortality in patients with hematologic malignancies.^5–10 In a large metanalysis of 38 studies that included 3,377 hematologic malignancy patients with acute COVID-19, the pooled risk of death among hospitalized adult patients with hematologic malignancies was 39%, and approached 50% in patients 60 years or older.¹¹ Although the availability of vaccines has dramatically reduced the overall risk of COVID-19 related complications and mortality, vaccinated patients with a hematologic malignancy, especially those on active treatment, remain at a higher risk for hospitalization (relative risk [RR] 3.13) and death (RR 1.66) when compared to matched vaccinated controls without cancer.^12,13

Immune dysregulation has been strongly linked to COVID-19 morbidity and mortality, beginning with early reports demonstrating a heightened inflammatory response in patients with severe disease.¹⁴ This, coupled with randomized controlled data showing a benefit to dexamethasone treatment in patients with severe COVID-19 infection, suggested that COVID-19 pathology might be driven largely by an overzealous immune response.¹⁵ At the same time, the finding that both autoantibodies against type I interferon as well as mutations in types I and III interferon genes were associated with severe COVID-19 infection suggested that immunosuppression might also be a risk factor for poor outcomes.^16,17 Specifically, in patients with hematologic malignancies, delayed neutralizing antibody production and lymphopenias were associated with severe disease, while decreased CD8+ T-cell counts were predictive of mortality.^18–20

Unfortunately, patients who recover from acute COVID-19 infection often experience a distinct set of lingering symptoms known collectively as “long COVID.” Results from a recent international online survey found that symptoms of long COVID-19 persisted for longer than six months in 65.2% of the participants.²¹ Up to one quarter of participants with post-COVID-19 sequelae may have an initial asymptomatic infection, suggesting that this syndrome does not merely represent residual symptoms in patients with severe disease.²² The most frequent and disabling symptoms of long COVID-19 included fatigue, post-exertional malaise, and cognitive dysfunction, and many patients exhibit features of postural tachycardia syndrome and Myalgic Encephalomyelitis, all of which have been observed after other viral infections.^21,23 Furthermore, previously COVID-infected patients can develop defects involving multiple organ systems via mechanisms that remain incompletely understood. For example, a correlation between COVID-19 and a heightened risk of new onset diabetes and coagulopathy have now been shown in multiple studies, including among those with initial mild infection.^24–26 How the presence of hematologic malignancy affects the risk of long COVID remains unknown.

Unlike patients without cancer, however, immunosuppressed patients including those with hematologic malignancies can develop a distinct chronic viral syndrome marked by persistent PCR positivity, migrating radiographic findings, chronic respiratory and systemic symptoms, evidence of intra host viral evolution, and subdued or aberrant immune responses.^27–31 In addition to the direct toxicity of prolonged infection, these patients also suffer from delays in the initiation or resumption of cancer-directed therapy that can ultimately lead to inferior cancer-specific outcomes.^31–33 Given that this syndrome is primarily seen in immunocompromised patients, evidence-based strategies to either prevent or manage persistent infection are lacking.

In this review, we discuss the current understanding of the incidence, clinical characteristics, risk factors, and outcomes of persistent COVID-19 infection in patients with hematologic malignancy. We further elaborate on the immunologic drivers of SARS-CoV-2 persistence based on the complex interactions between the virus and host immune state. Additionally, we discuss the fundamental implications of actively replicating viral reservoirs in a host environment that could promote immune escape from vaccines and therapeutics. Finally, we share our perspective on how to approach management of patients who develop this unique syndrome.

Clinical experience of persistent COVID-19 in patients with hematologic malignancy

The syndrome of persistent COVID-19 in patients with lymphoid malignancies manifests as a chronic protracted illness marked by viral RNA detection with concomitant respiratory and systemic symptoms that often require hospital readmission (Figure 1). In one of the earliest reports of its kind, Yasuda et al. reported a case of a patient with follicular lymphoma undergoing rituximab maintenance therapy with persistent COVID-19 pneumonia at two months and positive SARS-CoV-2 reverse transcription polymerase chain reaction (RT-PCR) at day 46 and a follow-up report published one year later found that the patient had waxing and waning clinical manifestations for 10 months, resolution of lung lesions at 12 months, and never developed anti-SARS-CoV-2 antibodies.^34,35

Figure 1.

Risk factors and symptoms associated with acute, persistent, and long COVID-19.

Multiple case series followed by two large cohort studies reporting on persistent SARS-CoV-2 infection in patients with hematologic malignancies have been subsequently published.^{28,31,36–38} In Table 1 we review the clinical presentation, duration of chronic infection, virologic assessment, and clinical predictors of COVID-19 persistence and outcomes in these critical studies. Hueso et al. were the first to report protracted COVID-19 infection in 17 consecutive patients with B-cell lymphopenia including 15 patients with hematologic malignancies.³⁶ In their study, 15 patients had received anti-CD20 monoclonal antibody treatment within the past 2 years with an interval between last rituximab infusion and COVID-19 symptom onset of 4 months. Patients suffered from COVID-19 symptoms for a median of 56 days (range 7–83 days). Subsequently, Betrains et al. reported on 5 lymphoma patients with severe B-cell depletion who suffered from protracted clinical manifestations and decreased neutralizing antibody titer formation.²⁸ This was one of the first studies to demonstrate reduced viral clearance in patients with hematologic malignancies as nasopharyngeal SARS-CoV-2 RT-PCR tests remained positive for a median of 74 days (range 44–109 days) following disease onset. Further, the authors showed a clear association between waxing and waning symptoms from COVID-19, nasopharyngeal viral load as measured by semi-quantitative SARS-CoV-2 RT-PCR, and the absence of neutralizing antibodies. Interestingly, although SARS-CoV-2 IgG antibodies were absent at follow-up in all patients, no relapse was observed after the four surviving patients’ median follow-up of 117 days (range 88–148 days).

Table 1:

Epidemiological, clinical and immunological characteristics of patients with hematologic malignancies and persistent COVID-19 infection

Manuscript	Patient characteristics	Clinical Course	Immunologic Assessment	Predictors of outcome	Outcome
Yasuda et al35	19 pts 100% Lymphoma 16/19 with anti-CD20 included in most recent therapy	Viral persistence: median 65 days Symptoms: up to 10 months Radiographic findings: up to 12m	Failure to develop antibodies in 11/11 evaluated pts	n/a	13 pts recovered 5 pts died 1 pt with ongoing COVID-19
Hueso et al36	17 pts 15 with heme malignancy, 1 with MS, 1 with CVID 15 pts with anti-CD20 in past 2y	Symptoms: 56 days (7–83d) 10 pts required supplemental O2 2 pts required intubation	SARS-CoV-2-specific T-cell responses in all pts No SARS-CoV-2-specific antibodies in any pts	Decrease in blood SARS-CoV-2 RNA with clinical improvement	16 recovered 1 died
Poncelet et al38	3 pts All undergoing chemotherapy for aggressive lymphoma 2/3 receiving anti-CD20 therapy	All 3 pts with multiple hospital admissions 2 required ICU 1 required mechanical ventilation Viral persistence: up to 43 days	n/a	n/a	3 readmitted 2 recovered 1 died
Betrains et al28	5 pts All with lymphoma All receiving anti-CD20 therapy Median time since anti-CD20 treatment: 8 months (4–15m)	Viral persistence: 74 days (44–109d) 2 pts required supplemental O2	Elevated inflammatory markers in all pts Absent anti-SARS-CoV-2 IgG in all pts	SARS-CoV-2 qRT-PCR associated with symptoms	4 recovered 1 died of pulmonary aspergillosis
Duléry et al37	111 pts All with lymphoma 94 with B-NHL 63 (57%) with anti-CD20 in past 12 months	Nasopharyngeal viral persistence: up to 143 days Peripheral blood RNA: positive in 10/13 pts, median 35d (4–123d) Symptoms: 14 days (1–235d) Symptoms in pts with prolonged symptoms: 58 days (31–235d) Among 32 hospitalized pts: -15% (n=5) no O2 support -30% (n=11) low-dose O2 -12% (n=2) high-flow O2 -42% (n=14) mechanical ventilation	2/19 pts with prolonged stay had +SARS-CoV-2 antibodies	Predictors of prolonged stay: -anti-CD20 therapy -age ≥70 years -relapsed/refractory lymphoma	24 (22%) died within 30d Among 87 remaining pts: -55 (63%) discharged -31 (35%) remain hospitalized -1 pt readmitted -69% alive at 6m
Lee et al31	382 pts All with hematologic malignancy 214 with lymphoid malignancy	PCR+ ≥30 days in 13.9% (n=51) Viral persistence: 59 days (26–344) Hospitalization: 27% (n=102) Severe COVID-19: 83% (n=85) O2 support: 72% (n=73) ICU-level care: 31% (n=32) Mechanical ventilation: 22% (n=22)	Of patients with 2+ hospital admissions for COVID-19: - 14/19 with anti-CD20 therapy - 11/12 with CD19+ cells <50/mcL	Predictors of persistent PCR positivity: -lymphopenia -anti-CD20 within 1y -cellular therapy within 1 year Predictors of death: -Cardiovascular disease -active cancer therapy -CAR-T cell therapy	Of 102 hospitalized pts: 57 (56%) recovered 26 (25%) died 19 (18%) recovered and re-admitted

Abbreviations: BAL, bronchoalveolar lavage; B-NHL; B-cell non-Hodgkin lymphoma; CAR-T, chimeric antigen receptor; CVID, common variable immunodeficiency; COVID-19, coronavirus disease 2019; d; days; HR, hazard ratio; ICU, intensive care unit; IgG, immunoglobulin G; LOS, length of stay; m; months; mcL, microliter; MS, multiple sclerosis; O2, oxygen; pts, patients; RT-PCR, reverse transcriptase polymerase chain reaction; SARS-CoV-2, severe acute respiratory syndrome coronavirus 2

In a retrospective multicenter study from France, Duléry et al. examined prolonged length of stay in 111 lymphoma patients hospitalized for COVID-19.³⁷ They found that 32 patients (29%) suffered from persistent COVID-19 symptoms lasting over 30 days requiring an extended hospital stay while the median duration of hospitalization was 14 days (range 1–235 days), and a 69% survival rate at 6-months. On multivariate analysis, recent administration of anti-CD20 therapy was associated with prolonged length of stay (HR 2.26, p-value <0.001) and higher risk of death (HR 2.17, p-value = 0.039). Viral persistence via nasal swabs or bronchoalveolar lavage SARS-CoV-2 RT-PCR was observed up to 143 days. Of the 19 patients with prolonged hospital stay in whom sufficient data was available, only two demonstrated seroconversion of antibodies targeting COVID-19.

Most recently and in the largest cohort of persistent symptomatic COVID-19 infection in patients with lymphoid malignancies, Lee et al. described the clinical course and predictors of persistent COVID-19 in 382 patients with hematologic cancer.³¹ 73 of 102 (72%) hospitalized patients in their cohort required supplemental oxygen, 31% required intensive care unit-level care, and 22% required mechanical ventilation. Case fatality rate during the index hospitalization was 22%, and 31% (n=32) experienced an extended length of stay over 21 days while 19 of the remaining 77 patients (25%) who were discharged following the index hospitalization were readmitted with recrudescent respiratory symptoms. Multivariate analysis revealed cardiovascular disease, active cancer treatment, and chimeric antigen receptor T-cell therapy, but not anti-CD20 therapy, to be independently associated with mortality. The authors also found a high rate of viral persistence as 13.9% of patients demonstrated persistent SARS-CoV-2 RT-PR positivity after ≥30 days. Lymphopenia, receiving anti-CD20 antibodies, and cellular therapy were independent risk factors for viral persistence. This study was the first to demonstrate that distinct clinical, laboratory and immunologic features were associated with acute and chronic COVID-19-driven pathology.

Immunologic drivers of viral persistence, variants, and outcomes

Patients with hematologic malignancies offer a unique opportunity to dissect the contributions of specific immune cell types to the unique clinical manifestations of SARS-CoV-2 infection, given that these patients often suffer from deficits in specific arms of the host immune response, due either to disease or therapy-dependent suppression of immune cell types. For example, while increased CD8+ T-cell activity has been linked to severe disease in immune competent patients, acute mortality was highest in hematologic malignancy patients with low CD8+ T-cell counts, either due to bone marrow dysfunction or cytotoxic chemotherapy.^19,39 As noted above, loss of humoral immune responses, mostly observed among those receiving B-cell-targeted therapies, appears to be the dominant signature associated with prolonged or recurrent COVID-19 infection.³¹ The immune signatures associated with either acute mortality or chronic infection are notably distinct from the aforementioned “long COVID” syndrome, in whom persistent circulating inflammatory factors appear to play an important role.⁴⁰

A significant public health concern regarding chronic persistent SARS-CoV-2 infection is the threat of intra-host evolution. Accelerated viral evolution has been described in immunosuppressed patients with persistent SARS-CoV-2 infection.^29,41–43 Intra-host viral evolution and the emergence of mutations found in variants of concern have been observed amongst patients with hematologic malignancies suffering from chronic SARS-CoV-2 infection.^27,44–48 In these studies, viral persistence resulted from ongoing infection by the same strain rather than re-infection or co-infection with additional viral strains, and active treatment of hematologic malignancy and the resultant suppression of cellular immunity that occurs was the strongest predictor of viral entropy.³¹ Together, these findings have led to growing concern that immunosuppressed patients with lymphoid malignancies may serve as reservoirs for the generation of viral escape mutants with immune evading mechanisms. The sudden emergence of the Omicron variant in South Africa and Botswana in November 2021 with the closest ancestral strain isolated in mid-2020 has further fueled this speculation as a possible source of Omicron origin through intra-host evolution in an immunosuppressed patient.^49,50 This phenomenon has been demonstrated in several reports in immunosuppressed patients, including those with blood cancers and patients with poorly controlled HIV.^{27,29,44,51,52}

Although a few case reports of successful viral clearance with passive immunotherapy have been published, it is unclear if the current treatment strategies to prevent or treat COVID-19 infection reduce the risk of intra-host evolution in this patient population.^28,35,36,53 There is limited experience with recently authorized antivirals in treating chronic infection, and enthusiasm for chronic use of these agents is tempered by the possibility of developing therapy-related mutations, as has been reported for one of the currently available antiviral drugs, molnupiravir.^54–57 This risk is expected to be amplified in immunosuppressed patients with uncontrolled viral replication and hence caution is recommended when considering this agent in patients with underlying immune dysfunction. The role of combined treatment with antivirals and effective monoclonal antibodies has not been studied yet. Still, with the substantial COVID-related morbidity, there may be a benefit to early treatment with both agents in patients at risk for viral persistence.

SARS-CoV-2 vaccine efficacy amongst patients with hematologic malignancies

While widespread vaccination efforts have reduced the overall risk of COVID-19 associated morbidity and mortality for much of the global population, cancer patients, especially those with hematologic malignancies, have a lower likelihood of achieving target immune responses following vaccination against SARS-CoV-2. Preliminary reports found that certain patients with hematologic malignancies had reduced humoral response and failed to produce high titers of anti-SARS-CoV-2 antibodies in response to vaccination.^58–60 A recent meta-analysis found a heterogeneous and markedly blunted serologic response among participants with hematologic malignancies with a significantly lower seroconversion rate after complete immunization compared to those with solid tumors (65% vs 94%; p-value < 0.0001).¹³

Patients on anti-CD20 therapy have consistently shown poor response to mRNA vaccines and the time interval between last treatment dose and vaccination is the strongest predictor of serological response.^61–67 A retrospective study from Israel found a higher incidence of COVID-19 infection (RR 1.60, 95% CI 1.12–2.37), symptomatic disease (RR 1.72, CI 1.05–2.85), COVID -19-related hospital admissions (RR 3.13, CI 1.68–7.08), severe COVID-19 (RR 2.27, CI 1.18–5.19), and COVID-19-related death (RR 1.66, CI 0.72–4.47) among 32,516 vaccinated patients with hematologic malignancies compared with vaccinated matched controls.¹² When the analysis was restricted to patients on active treatment for hematologic malignancies the risk of all studied outcomes was increased. Population-level data from the Center for Disease Control and Prevention (CDC) have shown that vaccinated person who are immunosuppressed are at elevated risk for severe outcomes from COVID-19 (adjusted OR 1.91, 95% CI 1.37–266) and that the effectiveness of mRNA vaccines against COVID-19-related hospitalization is lower among immunocompromised adults including patients with hematologic malignancies (77%, 95 CI 74%–80%) than among non-immunocompromised controls (90%, 95% CI 89%–91%).^68,69

Nevertheless, vaccination against SARS-CoV-2 has clearly reduced COVID-19-related complications amongst patients with hematologic cancers and guideline recommendations strongly recommend vaccination in this population.^70–72 This is likely due to T-cell memory responses generated to vaccination in up to 50–75% of serological non-responders.^63,67,73 T-cell immunity following vaccination has been demonstrated in both hematologic malignancy patients undergoing B-cell-directed therapies as well as in patients receiving B-cell depleting therapies for rheumatologic disorders.^65,74,75 Biological data demonstrating that T-cell responses largely protect against COVID-19 severity while being insufficient to achieve rapid COVID-19 viral clearance in the absence of B-cell function are consistent with these observations.

A recently published observational cohort study investigating the immunogenicity and reactogenicity of SARS-CoV-2 vaccines in hematologic and solid organ cancer patients found that two doses of an mRNA vaccine (either BNT162b2/Pfizer-BioNTech or mRNA-1273/Moderna) were associated with higher protective immune responses compared with one dose of the adenoviral vector vaccine Ad26.COV2.S/Janssen.⁷⁶ Additional mRNA vaccine doses may improve serological response in a subset of patients and forms the basis for the current CDC recommendation for a three-dose mRNA primary vaccine series followed by a boosters as recommended by the Advisory Committee on Immunization Practices ( https://www.cdc.gov/vaccines/covid-19/clinical-considerations/interim-considerations-us.html#immunocompromised ).⁷² Heterologous boosting may have advantages and is encouraged especially in those who initially received the vectored vaccine.⁷⁷ The impact of vaccination on post-acute sequalae in hematologic malignancy patients is not known.

Clinical approach to patients with hematologic malignancy and persistent COVID-19

Consensus is lacking on the optimal management approach of persistent COVID-19 in patients with hematologic malignancies and underlying immune dysregulation, in large part because randomized controlled trials of therapeutic agents did not enroll these patients. In the absence of high-quality clinical data, we offer our suggestions for both prevention and treatment of COVID-19 infection in patients with hematologic malignancies:

• In uninfected patients, vaccination against COVID-19 infection remains the most well-supported strategy to prevent COVID-19 infection and/or reduce disease severity, even in patients who do not mount an adequate humoral response. We emphasize that in immunocompromised patients, a total of four doses is indicated and associated with the highest rates of seroconversion.^78,79 See https://www.cdc.gov/vaccines/covid-19/clinical-considerations/interim-considerations-us.html#immunocompromised for the interim vaccination schedule in high-risk immunocompromised patients. In patients who are unable to mount a humoral immune response to vaccination, the combination of tixagevimab and cilgavimab received emergency use authorization (EUA) for pre-exposure prophylaxis in December 2021, and subsequent recommendations on revised higher doses was made for the Omicron subvariants. As the only drug now approved for pre-exposure prevention, this combination Mab should be routinely offered to high-risk patients as long as neutralization activity against circulating SARS CoV-2 variants is retained. The recommended interval for tixagevimab and cilgavimab after vaccination is 2 weeks. No waiting period is advised for vaccination after tixagevimab and cilgavimab. Patients with acute infection who are treated with a monoclonal antibody should receive tixagevimab and cilgavimab as soon as they are clinically recovered. For new and emerging variants, the empiric use of monoclonal antibody should be based on the invitro neutralization activity.

• In patients with newly diagnosed COVID-19 infection, the ability of approved or EUA agents to prevent development of chronic or long COVID-19 remains unknown. However, given data showing that monoclonal antibodies and anti-virals including remdesivir and nirmatrelvir/ritonavir reduce the risk of hospitalization in mild to moderate COVID illness and accelerate viral clearance,. we recommend these agents during acute infection with a preference for either nirmatrelvir/ritonavir or an authorized monoclonal antibody for treatment [ fda.gov].^80–82 A three-day infusion of remdesivir is an alternate option in non-hospitalized patients who are at high risk for COVID-19 progression but requires access to an infusion center on consecutive days.⁵⁶ Molupiravir is the least preferred agent due to concerns about low barriers to resistance. The role of combination anti-viral therapy has not been studied at this time.

• For patients with severe disease, steroids and/or tocilizumab can be used as clinically indicated but should not be used indefinitely as long-term immunosuppression can potentially facilitate chronic infection. Remdesivir can be considered in patients hospitalized with COVID-19 who do not require mechanical ventilation or extracorporeal membrane oxygenation because some data suggest it may reduce time to recovery and risk of mechanical ventilation.⁸³ Barcitinib can be used in combination with remdesivir if steroids are not ideal.

• Unfortunately, there are no currently approved therapies for patients who develop chronic COVID-19 infection, as all approved or EUA agents are indicated only for use in newly symptomatic patients. We eagerly await studies to determine whether either anti-viral therapy or passive immunization with recombinant monoclonal antibodies or high-titer convalescent plasma can promote viral clearance in these patients. Given the ongoing observed lung inflammation, patients with chronic COVID-19 are often on long-term steroid therapy. We would again recommend judicious use of steroids with tapering when clinically feasible given the potential negative consequences of long-term immunosuppression, which may potentially underlie the increased risk of invasive fungal infections reported in patients with severe COVID-19 infection.^84,85

Conclusions

Patients with hematologic malignancies are at unique risk for persistent COVID-19, a syndrome of progressive and relapsing respiratory symptoms with prolonged viral shedding. The long-term clinical sequelae of chronic SARS-CoV-2 infection in patients with lymphoid malignancies have wide-ranging implications that extend beyond individual patient outcomes. While the availability of vaccines has dramatically reduced the overall risk of COVID-19 related complications and mortality, the optimal clinical approach to persistent COVID-19 in patients with hematologic malignancies is uncertain and further studies are urgently needed on various fronts.

Key Points:

• Patients with hematologic malignancies, especially those lacking B-cell function, are at increased risk for persistent COVID-19 infection

• Persistent COVID-19 in patients with hematologic malignancies manifests as a chronic protracted illness marked by a progressive or relapsing respiratory syndrome with persistent viral RNA detection

• Optimal management of persistent COVID-19 in patients with hematologic malignancies is uncertain, though vaccination against SARS-CoV-2 has dramatically reduced the overall risk of COVID-19 related complications and mortality