Objectives:
The long-term effects of severe acute respiratory syndrome coronavirus 2 (coronavirus disease 2019 [COVID-19]) infection in patients with cancer are unknown. We examined 1-year mortality and prevalence of long COVID in patients with and without cancer after initial hospitalization for acute COVID-19 infection.
Methods:
We previously studied 585 patients hospitalized from March to May 2020 with acute COVID-19 infection at Weill Cornell Medicine (117 patients with cancer and 468 age, sex, and comorbidity-matched non-cancer controls). Of the 456 patients who were discharged, we followed 359 patients (75 cancer and 284 non-cancer controls) for COVID-related symptoms and death, at 3, 6, and 12 months after initial symptom onset. Pearson χ2 and Fisher exact tests were used to determine associations between cancer, postdischarge mortality, and long COVID symptoms. Multivariable Cox proportional hazards models adjusting for potential confounders were used to quantify the risk of death between patients with and without cancer.
Results:
The cancer cohort had higher mortality after hospitalization (23% vs 5%, P < 0.001), a hazard ratio of 4.7 (95% CI: 2.34-9.46) for all-cause mortality, after adjusting for smoking and oxygen requirement. Long COVID symptoms were observed in 33% of patients regardless of cancer status. Constitutional, respiratory, and cardiac complaints were the most prevalent symptoms in the first 6 months, whereas respiratory and neurological complaints (eg, “brain fog” and memory deficits) were most prevalent at 12 months.
Conclusions:
Patients with cancer have higher mortality after hospitalization for acute severe acute respiratory syndrome coronavirus 2 infections. The risk of death was highest in the first 3 months after discharge. About one-third of all patients experienced long COVID.
Key Words: COVID, Long COVID, SARS-CoV-19, cancer, risk factors
BACKGROUND/RATIONALE
In the United States, there are currently more than 16.9 million people with a history of cancer and more than 1.9 million people will be diagnosed with cancer in 2022.1 Patients with cancer have an estimated 5% to 15% prevalence of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) (coronavirus disease [COVID-19]) infection and COVID-19 mortality as high as 25%.2 We reported that patients with cancer hospitalized from March to May 2020 in New York City due to COVID-19 did not have higher rates of mortality, intensive care unit transfer, or intubation compared with age, sex, and comorbidity-matched controls without cancer during their initial hospitalization.3 This observation was echoed by other research groups, which reported that other patient-level factors such as age and sex were independently associated with increased risk of intensive care unit admission and death, but cancer status and cytotoxic therapy were not associated with increased mortality.4 As patients survive their initial COVID-19 infection, new questions regarding posthospitalization mortality and morbidity have emerged, particularly as it relates to “long COVID”.
Patients with cancer may be at particular risk for long COVID symptoms given their compromised immune systems and weakened physiological reserve. Dysregulated functions of natural killers, helper T cells, and macrophages have been implicated in immune evasion5,6 and may leave patients with cancer vulnerable to infection, which is further compounded by cytotoxic treatments.7 A “long COVID” syndrome, as defined by the World Health Organization, encompasses a range of symptoms experienced at least 3 months after COVID-19 symptom onset, lasting for 2 or more months, and which cannot be explained by an alternative diagnosis.8 Long COVID is highly variable, but usual symptoms include myalgia, fatigue, respiratory symptoms like dyspnea, cough, and chest pain, as well as neurological symptoms, gastrointestinal dysfunction, and skin manifestations.9,10 Long COVID is a significant health burden, affecting 3 to 5 million individuals,11 and can occur in up to 70% of patients who were hospitalized during their initial infection.10,12,13 This study sought to document the prevalence and characteristics of long COVID and postdischarge mortality among patients with cancer who survived their initial acute, severe COVID-19 infection.
METHODS
This study uses a cohort formulated to examine COVID-19 in patients with cancer and compare them to age, sex, and comorbidity-matched non-cancer controls.3 There were 1935 patients hospitalized at either New York/Presbyterian-Weill Cornell Medicine or Lower Manhattan Hospital with COVID between March 1, 2020 and May 15, 2020, from which a cohort of cancer and non-cancer matched controls were selected for further analysis. We examined 117 patients with active malignancy and 468 patients without cancer who were matched (1:4) for age, sex, and the number of comorbidities.3 Comorbidities used for matching included obesity, diabetes, hypertension, chronic obstructive pulmonary disease, asthma, end-stage renal disease, cirrhosis, coronary artery disease, heart failure, and human immunodeficiency virus.3 In this report, we followed those patients who survived their initial hospitalization for mortality, COVID-related death, and long COVID symptoms at 3, 6, and 12 months after the onset of initial symptoms. The follow-up data were abstracted after allowing a range of +/− 14 days from the date of symptom onset. Loss to follow-up was defined by an absence of electronic medical record (EMR) chart data between initial hospital discharge and 12 months postsymptom follow-up. This study was approved by the Weill Cornell Medicine Institutional Review Board, which waived informed consent.
Primary Outcome: Mortality
The primary outcome was postdischarge all-cause mortality. Survival at each time point was identified and confirmed by a combination of in-person visits, EMR data, telephone calls with patient and/or surrogate medical decision makers, or documentation of in-person presentation for laboratory or imaging studies.14 Dates of death were obtained from the EMR or from the patient’s report. The cause of death was recorded when indicated in the patient's EMR or could be inferred.
Secondary Outcome: Long Coronavirus Disease
Long COVID symptoms at each time interval were collected from data abstracted from the EMR and met the WHO case definition for long COVID.8 To allow for meaningful comparison across different studies, symptoms were grouped and analyzed together based on the impacted organ system. For example, “shortness of breath” and “pleuritic chest pain” are considered distinct long COVID symptoms, but for purposes of analysis, these were grouped into “respiratory symptoms” (Supplemental Table 3, Supplemental Digital Content 1, http://links.lww.com/AJCO/A464). Any symptom that could be attributed to cancer or other cause was not considered a long COVID symptom, as per the WHO definition.
Statistical Analyses
We first compared all patients from the initial COVID-19 hospitalization to those who survived hospitalization, and again at the end of the 1-year follow-up period, on sex, age, and number of comorbidities to determine whether the balance was maintained despite loss due to death and/or drop-out. Balance was assessed again for patients that had any data on symptoms during the year versus those who did not. Patients with and without cancer who were lost to follow-up were censored from the final analyses. All tests of associations were completed using Pearson χ2 and Fisher exact tests. Categorical variables are presented as counts and percentages; we reported medians and interquartile ranges for continuous variables. Significance was set at P <0.05. All analyses were conducted in R 4.1.0.15 We examined survival using a Kaplan-Meier plot with a log-ranked test to determine whether there is a difference in risk of death over 1 year. A multivariate Cox proportional hazards model adjusting for smoking and oxygen requirement at emergency room presentation was used to determine the magnitude of the difference in risk of death between the two cohorts. We then examined associations between active cancer and the prevalence of any long COVID-19 symptoms.
Within the cancer cohort, additional subanalyses compared morbidity and mortality outcomes for patients with solid versus hematologic malignancies and outcomes for patients who received cytotoxic therapy within 90 days of presentation versus those who did not.
RESULTS
Our initial cohort included 117 patients with active malignancy ranging from stage 1 to metastatic disease; 42 had hematologic malignancy (6 acute leukemia, 19 chronic leukemia, 7 lymphomas, and 10 multiple myelomas) and 75 had solid malignancy (15 breasts, 19 gastrointestinal, 22 genitourinary, 6 gynecologic, and 1 thoracic). Of the 585 patients in our initial cohort who were hospitalized with acute COVID-19 infection from March 1 to May 15, 2020, 456 patients (77.9%) survived their initial acute infection and were discharged from the hospital (88 with cancer and 368 matched non-cancer controls). Of these patients, 97 (21.2%) were lost to follow-up, leaving a final study population of 359 patients (75 with cancer and 284 controls) (Table 1). Among patients with complete follow-up, the median follow-up time is 360 days. The final cohort was profiled and remained balanced on the matched characteristics (Supplemental Table 1, Supplemental Digital Content 1, http://links.lww.com/AJCO/A464). An analysis comparing patients who were lost-to-follow-up to those who were not also found no differences in the matched characteristics (Supplemental Table 2, Supplemental Digital Content 1, http://links.lww.com/AJCO/A464). An equal proportion of patients with and without cancer were retained for follow-up.
TABLE 1.
Characteristics of Patients With Cancer and a Matched, Non-cancer Cohort, Who Survived Their Initial SARS-CoV-2 Acute Infection That Resulted in Hospitalization at New York-Presbyterian Hospital From March 1, 2020 to May 15, 2020
| Characteristic | Overall, N = 359 * ; n (%) | Cancer patients, N = 75 * ; n (%) | Control patients, N = 284 * ; n (%) | P |
|---|---|---|---|---|
| Age | 70 (61, 77) | 71 (61, 83) | 70 (60, 77) | 0.2 |
| Sex | 174 (48) | 34 (45) | 140 (49) | 0.5 |
| Race | 0.5 | |||
| White | 153 (43) | 38 (51) | 115 (40) | — |
| Black | 53 (15) | 12 (16) | 41 (14) | — |
| Asian | 59 (16) | 10 (13) | 49 (17) | — |
| Other | 66 (18) | 10 (13) | 56 (20) | — |
| Unreported | 28 (7.8) | 5 (6.7) | 23 (8.1) | — |
| Admitted from | ||||
| Home | 311 (87) | 61 (81) | 250 (88) | — |
| Rehabilitation/nursing home | 27 (7.5) | 9 (12) | 18 (6.3) | 0.3 |
| Other hospital | 6 (1.7) | 1 (1.3) | 5 (1.8) | — |
| Other/undomiciled | 15 (4.2) | 4 (5.3) | 11 (3.9) | — |
| Smoking history | 0.4 | |||
| Never smoked | 259 (72) | 50 (67) | 209 (74) | — |
| Former smoker | 87 (24) | 23 (31) | 64 (23) | — |
| Current smoker | 13 (3.6) | 2 (2.7) | 11 (3.9) | — |
| Comorbidities | ||||
| Obesity (BMI ≥30 kg/m2 [≥28 kg/m2 for Asians]) | 89 (25) | 17 (23) | 72 (25) | 0.6 |
| Diabetes | 107 (30) | 24 (32) | 83 (29) | 0.6 |
| Hypertension | 202 (56) | 40 (53) | 162 (57) | 0.6 |
| COPD | 21 (5.8) | 5 (6.7) | 16 (5.6) | 0.8 |
| Asthma | 23 (6.4) | 4 (5.3) | 19 (6.7) | 0.8 |
| ESRD | 21 (5.8) | 3 (4.0) | 18 (6.3) | 0.5 |
| Cirrhosis | 3 (0.8) | 1 (1.3) | 2 (0.7) | 0.5 |
| CAD | 54 (15) | 14 (19) | 40 (14) | 0.3 |
| Heart failure | 27 (7.5) | 9 (12) | 18 (6.3) | 0.1 |
| HIV | 3 (0.8) | 0 | 3 (1.1) | >0.9 |
| Total count of home medications (excluding over-the-counter medications) <5 | 188 (52) | 26 (35) | 162 (57) | <0.001 |
| Use of immunosuppressive medication | ||||
| None | 317 (88) | 60 (80) | 257 (90) | 0.012 |
| Prednisone <20 mg/d | 16 (4.5) | 5 (6.7) | 11 (3.9) | 0.3 |
| Prednisone ≥20 mg/d | 5 (1.4) | 1 (1.3) | 4 (1.4) | >0.9 |
| TNF α_inhibitor | 0 | 0 | 0 | — |
| Other monoclonal antibodies | 3 (0.8) | 1 (1.3) | 2 (0.7) | 0.5 |
| Tacrolimus | 12 (3.3) | 2 (2.7) | 10 (3.5) | >0.9 |
| Mycophenolate MMF | 12 (3.3) | 2 (2.7) | 10 (3.5) | >0.9 |
| Methotrexate | 1 (0.3) | 0 | 1 (0.4) | >0.9 |
| Other | 12 (3.3) | 9 (12) | 3 (1.1) | <0.001 |
Median (interquartile range).
BMI indicates body mass index; CAD, coronary artery disease; COPD, chronic obstructive pulmonary disease; ESRD, end-stage renal disease; HIV, human immunodeficiency virus; MMF, mycophenolate mofetil; SARS-CoV-2, severe acute respiratory syndrome coronavirus 2; TNF, tumor necrosis factor.
Overall, 32 patients (8.9%) died after discharge from their initial COVID-19 hospitalization. Patients with cancer had higher 1-year mortality compared with matched non-cancer controls (23% vs 5.3%, P < 0.001). The highest mortality was among patients in the first 3 months after discharge; mortality in the cancer cohort was 12% (95% CI: 6.0-22.2), compared with 2.8% (95% CI: 1.3-5.7) in the non-cancer cohort, P = 0.003. In the cancer cohort, mortality due to COVID-19 was 5.3% (95% CI: 1.7-14) compared with 1.1% in non-cancer patients (95% CI: 0.27-3.3, P 0.037). From 3 to 6 months, the overall mortality fell to 0.8% (95% CI: 0.22-2.6): 4.0% (95% CI: 1.0-12) in the cancer cohort compared with 0% in the non-cancer group (P = 0.009). From 6 to 12 months, 5 (6.7%) patients with cancer died versus 7 (2.5%) patients without cancer (Table 2).
TABLE 2.
Mortality After Discharge From Initial Admission for COVID-19 Hospitalization, at 3, 6, and 12 Months
| Characteristic | Overall, N = 359; n (%) | 95% CI | Cancer patients, N = 75; n (%) | 95% CI | Control patients, N = 284; n (%) | 95% CI | P * |
|---|---|---|---|---|---|---|---|
| Deceased (within 1 y) | 32 (8.9) | 6.3, 12 | 17 (23) | 14, 34 | 15 (5.3) | 3.1, 8.7 | <0.001 |
| Deaths (discharge to 3 mo) | 17 (4.7) | 2.9, 7.6 | 9 (12) | 6.0, 22 | 8 (2.8) | 1.3, 5.7 | 0.003 |
| Deaths from (3 to 6 mo) | 3 (0.8) | 0.22, 2.6 | 3 (4.0) | 1.0, 12 | 0 | 0, 1.7 | 0.009 |
| Deaths (6 to 12 mo) | 12 (3.3) | 1.8, 5.9 | 5 (6.7) | 2.5, 16 | 7 (2.5) | 1.1, 5.2 | 0.14 |
| Deaths due to COVID (discharge to 3 mo) | 7 (1.9) | 0.86, 4.2 | 4 (5.3) | 1.7, 14 | 3 (1.1) | 0.27, 3.3 | 0.037 |
| Deaths due to COVID (3 to 6 mo) | 0 | 0 | 0 | 0 | 0 | 0 | — |
| Deaths due to COVID (6 to 12 mo) | 0 | 0 | 0 | 0 | 0 | 0 | — |
Pearson χ2 test; Fisher exact test.
COVID indicates coronavirus disease.
Of the 32 total deaths, 7 deaths were attributable to COVID-19, 4 (5.3%, 95% CI: 1.7-14) in the cancer cohort and 3 (1.1%, 95% CI: 0.27-3.3%) in the non-cancer control cohort, P = 0.037. No deaths were observed from COVID-19–related causes in either cohort after 3 months from discharge (Table 2). Among the 17 patients who died after discharge in the cancer cohort, 4 patients (23.5%) died from complications of COVID-19, 8 patients (47%) died due to their malignancy, 1 (5.8%) from sepsis unrelated to COVID-19, and 4 patients (23.5%) from unknown causes. The time-to-event analysis revealed lower survival probability in the cancer cohort versus controls, P < 0.001 (Fig. 1). The cancer cohort had 4.7 times (95% CI: 2.34, 9.46) the hazard of all-cause mortality compared with the control cohort after adjusting for smoking and oxygen requirement within 3 hours of emergency room presentation.
FIGURE 1.
Overall Survival probability after hospital discharge from acute severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection. Patients without cancer (control patients) were matched 4:1 to patients with cancer (cancer patients) for multiple comorbidities. All patients were admitted during the initial SARS-CoV-2 outbreak in New York. After discharge, patients were followed for survival.
Among the 75 patients with cancer, 26 had a hematologic malignancy (acute leukemia, chronic leukemia, lymphoma, or myeloma), and 49 had a solid tumor diagnosis (Supplemental Table 3, Supplemental Digital Content 1, http://links.lww.com/AJCO/A464). There was no difference in all-cause mortality between patients with solid tumor or hematologic malignancies (24% mortality in solid tumor patients at 1 year compared with 19% for hematologic malignancies, P = 0.6). We also did not observe a difference in mortality between patients who received cytotoxic therapy within 90 days of admission and those who did not (29% of patients who died within a year received cytotoxic therapy whereas 20% did not, P = 0.4).
TABLE 3.
Long COVID Symptoms at 3, 6, and 12 Months After Initial Hospitalization for COVID-19
| Characteristic | Overall, N = 252; n (%) | 95% CI* | Cancer patient cohort, N = 52; n (%) | 95% CI* | Non-cancer control cohort, N = 200; n (%) | 95% CI* | P † |
|---|---|---|---|---|---|---|---|
| Any long COVID | 83 (33) | 27, 39 | 15 (29) | 18, 43 | 68 (34) | 28, 41 | 0.5 |
| Long COVID discharge to 3 mo (excluding 49 died/LFU) | 71 (35) | 29, 42 | 14 (28) | 17, 43 | 57 (37) | 30, 45 | 0.2 |
| Long COVID 3 to 6 mo (excluding 77 died/LFU) | 43 (25) | 19, 32 | 6 (15) | 6.1, 30 | 37 (28) | 20, 36 | 0.091 |
| Long COVID 6 to 12 mo (excluding 78 died/LFU) | 28 (16) | 11, 23 | 3 (8.3) | 2.2, 24 | 25 (18) | 12, 26 | 0.2 |
Wilson CI.
Fisher exact test; Pearson χ2 test.
COVID-19 indicates coronavirus disease 2019; LFU, lost to follow-up.
Long Coronavirus Disease
Overall, 33% of all patients reported long COVID symptoms (Supplemental Table 4, Supplemental Digital Content 1, http://links.lww.com/AJCO/A464) in the 12-month follow-up period. The prevalence of any type of long COVID symptom at 3 months was 28% for patients with cancer (95% CI: 17-43) and 37% for the control patients (95% CI: 30-45, P = 0.2). At 6 months, 15% (95% CI: 6.1-30) and 28% (95% CI: 20-36, P = 0.091) of cancer and control cohorts, respectively, reported symptoms, and at 12 months, 8.3% (95% CI: 2.2-24) of patients with cancer and 18% (95% CI: 12-26, P = 0.2) of the non-cancer control cohort reported symptoms. At each time point, a lower proportion of cancer patients reported each symptom compared with control patients (Table 3). Supplemental Table 5 (Supplemental Digital Content 1, http://links.lww.com/AJCO/A464) also provides the long COVID rates at 3, 6, and 12 months, along with proportions of patients who were deceased and lost to follow-up at each time point.
Specific symptoms reported and the organ system, to which they were categorized are found in Table 4. For the most common complaints (respiratory, constitutional, and neurological), fewer patients with cancer reported symptoms compared with their non-cancer counterparts. For respiratory symptoms, the rates reported in the cancer cohort at 3, 6, and 12 months after initial symptom onset were 16%, 2.4%, and 2.8%, respectively, compared with rates of 21%, 15%, and 8% reported in the non-cancer cohort at the same time points. Constitutional and neurological symptoms were reported with a consistent trend of lower rates for patients with cancer at all time points. All patients reported fewer symptoms after 3 months (Table 4).
TABLE 4.
Long COVID Symptoms Over Time in Patients With and Without Cancer
| 3 mo | 6 mo | 12 mo | |||||||
|---|---|---|---|---|---|---|---|---|---|
| Characteristic | Cancer patients, N = 50; n (%) | Control patients, N = 153; n (%) | P * | Cancer patients, N = 41; n (%) | Control patients, N = 134; n (%) | P * | Cancer patients, N = 36; n (%) | Control patients, N = 138; n (%) | P * |
| Any long COVID | 14 (28) | 57 (37) | 0.2 | 6 (15) | 37 (28) | 0.091 | 3 (8.3) | 25 (18) | 0.2 |
| Respiratory | 8 (16) | 32 (21) | 0.4 | 1 (2.4) | 20 (15) | 0.029 | 1 (2.8) | 11 (8.0) | 0.5 |
| Constitutional | 8 (16) | 34 (22) | 0.3 | 5 (12) | 16 (12) | >0.9 | 1 (2.8) | 6 (4.3) | >0.9 |
| Psychiatric | 3 (6.0) | 7 (4.6) | 0.7 | 1 (2.4) | 4 (3.0) | >0.9 | 1 (2.8) | 4 (2.9) | >0.9 |
| Neurological | 3 (6.0) | 11 (7.2) | 1 | 0 | 9 (6.7) | 0.12 | 0 | 9 (6.5) | 0.2 |
| Musculoskeletal | 0 | 2 (1.3) | >0.9 | 0 | 0 | NA | 0 | 1 (0.7) | >0.9 |
| Cardiac | 8 (16) | 33 (22) | 0.4 | 0 | 1 (0.7) | >0.9 | 0 | 0 | NA |
| Integumentary | 0 | 2 (1.3) | >0.9 | 0 | 1 (0.7) | >0.9 | 1 (2.8) | 0 | 0.2 |
| HEENT | 0 | 0 | NA | 0 | 0 | NA | 1 (2.8) | 0 | 0.2 |
| GI | 1 (2.0) | 0 | 0.2 | 0 | 1 (0.7) | >0.9 | 0 | 0 | NA |
The most frequent symptom and any symptom with over 10% incidence are highlighted.
Pearson χ2 test; Fisher exact test.
COVID indicates coronavirus disease; GI, gastrointestinal; HEENT, Head, Eyes, Ears, Nose, Throat; NA, not applicable.
The most frequently reported symptoms at 3 months were respiratory (eg, persistent shortness of breath, cough, and/or wheezing, which were reported in 16% of patients with cancer vs 21% of patients without cancer, P = 0.4) and constitutional (eg, fatigue and weakness, which were reported in 16% of patients with cancer vs in 22% of patients without cancer, P = 0.3). Respiratory symptoms persisted as the most common long COVID symptom in the non-cancer cohort throughout the year (15% of long COVID symptoms at 6 months, and 8.0% at 12 months. In the cancer cohort, the most prevalent long COVID symptom at 6 months was constitutional (12%).
The neurological symptoms most often documented were “brain fog” and memory deficits, though these resolved in all patients with cancer by 6 months. Many patients reported symptoms affecting multiple organ systems, often with a background of persistent fatigue and malaise.
DISCUSSION
COVID-19 remains a lethal virus for patients with cancer. Even after surviving the initial acute infection, we found that 4.7% of all patients in our study died within 3 months, including 12% of patients in the cancer cohort, and 2.8% of the matched control cohort. Within 1 year of discharge from their initial acute COVID-19 infection that resulted in hospitalization, 8.9% of the total population died within 1 year of discharge, 23% from the cancer cohort, and 5.3% from the matched non-cancer control cohort. Patients with cancer had higher mortality than patients without cancer and were nearly 5 times more likely to die within 1 year of discharge than their matched non-cancer controls. The most pronounced difference in mortality is in the first 3 months after symptom onset.
The high mortality of COVID-19, even after surviving the initial acute infection, is supported by other studies. In a study of over 70,000 veterans affairs patients (among whom cancer status was not specified) with mild or asymptomatic SARS-CoV-2 infection, there was a 1.59-fold increased risk of death in the patients who survived their initial COVID-19 infection compared with control patients.16 Patients hospitalized with COVID-19 also had a 1.51-fold increased risk of death at 6 months compared with influenza patients, highlighting the severity of this viral illness. However, we recognize that our relatively small cohort retained for our study could magnify the relative risk of long COVID symptoms and mortality compared with what was observed in other studies with larger populations. Our mortality findings are similar to another hospitalized cancer patient cohort from China that observed an 8% all-cause mortality rate in cancer patients and 0.4% all-cause mortality in patients without cancer at 1 year.17 However, in the Chinese cohort, they observed that after the initial hospitalization, 12-month mortality from COVID-19 seems to be similar between cancer and non-cancer patients.17
Despite observed differences in mortality between patients with and without cancer, we did not observe a significant difference in the prevalence of “long COVID” symptoms between the groups in the first 12 months after infection. Irrespective of cancer status, approximately one-third of patients who were hospitalized with acute SARS-CoV-2 infection subsequently developed long COVID. The underlying pathophysiology of long COVID remains unclear and may include ongoing viral replication driving a persistent immune inflammatory response7,18,19 autoimmunity, or overly exuberant host factor immune response to the initial or persistent SARS-CoV-2 infection.20 Persistence of the virus in the host triggers continuous recruitment of activated T cells, monocytes, and neutrophils, which in turn stimulate inflammatory cytokine release; in acute, severe COVID-19 infection, multiple hyper-inflammatory and pro-apoptotic pathways are triggered coupled with inhibition or delay in the induction of type I interferon (IFN) by infected cells.9 It is unknown whether these immunologic changes persist long COVID. But, there is evidence of persistent expression of cytokines such as type I (IFNβ) and type III (IFNλ1) IFN, IL-6, and others in patients with long COVID symptoms suggestive of persistent viral antigen-triggering persistent immune activation.18,21 These immune changes may explain why there seems to be a slightly lower prevalence of the long COVID symptoms in some patients with cancer: if the components of the immune system that drive the release of the cytokines like tumor necrosis factor-alpha (such as macrophages and monocytes22) and IL-6 are depleted by cancer or cancer-directed therapies, patients with cancer may have some protection against long COVID symptoms. Specifically, cytotoxic cancer therapies often induce myelosuppression, depleting populations of neutrophils, T cells, and monocytes23 that may drive the release of cytokines implicated in severe COVID-19 infection and the development of long COVID.
Immune suppression or general immune dysfunction in patients with cancer6,7 may dampen the exuberant immune activity theorized to drive long COVID symptoms. This may explain some of the lower observed rates of symptoms in cancer patients. Notably, we did not observe significant differences in symptoms by treatment type (eg, radiation only vs cytotoxic therapy), and we did not compare outcomes of different immunotherapies such as B-cell–directed therapies or checkpoint inhibitors. Patients receiving immunotherapy or who are immunosuppressed secondary to a solid organ transplant may be at risk of having a more severe acute infection,24 although the impact of long COVID remains unknown.
CONCLUSIONS
This is the first study to our knowledge to report the long-term mortality and morbidity after severe COVID-19 infection in patients with cancer as compared with a non-cancer–matched control cohort in a Western population. Patients with cancer are significantly more likely to die within 1 year of discharge from hospitalization with acute COVID-19 than matched non-cancer counterparts with the most significant risk of death in the first 3 months after symptom onset. With the COVID-19 pandemic extending into its third year, the impact of this infection and its variants on patients with cancer remains highly relevant. The significantly higher mortality in the first 3 months after hospitalization for acute infection in patients with cancer underscores the continued need to minimize infectious risk for patients with cancer. Further, about one-third of all patients with severe COVID-19 infection in our study reported long COVID symptoms regardless of their cancer status. Overall, our results identify significant long-term morbidity and mortality risk of this disease and the specific vulnerability of cancer patients to mortality in the first 3 months after symptom onset.
Limitations
Nearly 20% of patients who survived their initial SARS-CoV-2 hospitalization were lost to follow-up after hospital discharge. This may have biased symptom reporting, as patients feeling well may not have presented for follow-up visits. The presence of active malignancy, itself an impactful comorbidity, introduced an imbalance between the two cohorts. However, the consistency of matching for age, sex, and number of nonmalignant comorbidities between the patients who survived and remained actively followed as compared with those who were lost to follow-up supports our findings as a representative sample. We were in addition unable to specify the cause of death for several patients because of the decentralized cause of death documentation. Further, it is possible that different SARS-CoV-2 strains may have a differential impact on mortality after recovery from acute infection or persistent long COVID symptoms. Our cohorts were enrolled contemporaneously and likely represented exposure to the same SARS-CoV-2 strain. Finally, the patient’s reinfection status, repeat testing, rapid test results, vaccination status, and booster status were not available for the initial analysis and were not accounted for in the subsequent follow-up of these patients after their initial hospitalization. This information likely impacted the persistence and severity of long COVID symptoms but was not analyzed as part of this study due to inconsistent reporting of this information and the introduction of vaccines late in the follow-up period. Tixagebimab/cilgavimab was not available for postexposure prophylaxis during the study period, thus its impact on long COVID or posthospitalization mortality is not available. It was also impossible to account for behaviors such as avoiding crowded spaces, travel, and personal protective equipment usage that may have impacted the subject’s risk for reinfection.
Supplementary Material
Footnotes
This study received support from NewYork-Presbyterian Hospital (NYPH) and Weill Cornell Medical College (WCMC), including the Clinical and Translational Science Center (CTSC) (UL1 TR000457) and Joint Clinical Trials Office (JCTO).
The authors declare no conflicts of interest.
Supplemental Digital Content is available for this article. Direct URL citations are provided in the HTML and PDF versions of this article on the journal's website, www.amjclinicaloncology.com.
Contributor Information
Olivia Fankuchen, Email: orf9006@nyp.org.
Jennifer Lau, Email: jel4013@med.cornell.edu.
Mangala Rajan, Email: mar2834@med.cornell.edu.
Brandon Swed, Email: brs2023@med.cornell.edu.
Peter Martin, Email: pem9019@med.cornell.edu.
Manuel Hidalgo, Email: mah4006@med.cornell.edu.
Samuel Yamshon, Email: sjy9001@med.cornell.edu.
Laura Pinheiro, Email: lcp2003@med.cornell.edu.
Manish A. Shah, Email: mas9313@med.cornell.edu.
REFERENCES
- 1. America Cancer Society. Cancer treatment and survivorship facts and figures, 2022-2024. Atlanta: American Cancer Society; 2022. [Google Scholar]
- 2. Fillmore NR, La J, Szalat RE, et al. Prevalence and outcome of COVID-19 infection in cancer patients: a National Veterans Affairs Study. J Natl Canc Inst. 2021;113:691–698. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Brar G, Pinheiro L, Shusterman M, et al. COVID-19 severity and outcomes in patients with cancer: a matched cohort study. J Clin Oncol. 2020;38:3914–3924. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Lee LY, Cazier JB, Angelis V, et al. COVID-19 mortality in patients with cancer on chemotherapy or other anticancer treatments: a prospective cohort study. Lancet. 2020;395:1919–1926. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Biswas SK. Metabolic reprogramming of immune cells in cancer progression. Immunity. 2015;43:435–449. [DOI] [PubMed] [Google Scholar]
- 6. Finn OJ. Immuno-oncology: understanding the function and dysfunction of the immune system in cancer. Ann Oncol. 2012;23(suppl 8):viii6–viii9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Merad M, Blish CA, Sallusto F, et al. The immunology and immunopathology of COVID-19. Science. 2022;375:1122–1127. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. World Health Organization. A clinical case definition of post COVID-19 condition by a Delphi consensus. World Health Organization; 2021. https://www.who.int/publications/i/item/WHO-2019-nCoV-Post_COVID-19_condition-Clinical_case_definition-2021.1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Brodin P. Immune determinants of COVID-19 disease presentation and severity. Nat Med. 2021;27:28–33. [DOI] [PubMed] [Google Scholar]
- 10. Carfi A, Bernabei R, Landi F, et al. Persistent symptoms in patients after acute COVID-19. JAMA. 2020;324:603–605. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Tenforde MW, Devine O, Reese HE, et al. Point prevalence estimates of activity-limiting long-term symptoms among US Adults >=1 month after reported SARS-CoV-2 Infection, November 1, 2021. J Infect Dis. 2023;227:855–863. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Martimbianco ALC, Pacheco RL, Bagattini AM, et al. Frequency, signs and symptoms, and criteria adopted for long COVID-19: a systematic review. Int J Clin Pract. 2021;75:e14357. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Thompson EJ, Williams DM, Walker AJ, et al. Long COVID burden and risk factors in 10 UK longitudinal studies and electronic health records. Nat Commun. 2022;13:3528. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Kingery JR, Safford MM, Martin P, et al. Health status, persistent symptoms, and effort intolerance one year after acute COVID-19 infection. J Gen Intern Med. 2022;37:1218–1225. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Team RCR. A language and environment for statistical computing. Vienna, Austria: R Foundation for Statistical Computing; 2020. https://www.r-project.org/ [Google Scholar]
- 16. Al-Aly Z, Xie Y, Bowe B. High-dimensional characterization of post-acute sequelae of COVID-19. Nature. 2021;594:259–264. [DOI] [PubMed] [Google Scholar]
- 17. Chai C, Feng X, Lu M, et al. One-year mortality and consequences of COVID-19 in cancer patients: a cohort study. IUBMB Life. 2021;73:1244–1256. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Groff D, Sun A, Ssentongo AE, et al. Short-term and long-term rates of postacute sequelae of SARS-CoV-2 infection: a systematic review. JAMA Netw Open. 2021;4:e2128568. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Nalbandian A, Sehgal K, Gupta A, et al. Post-acute COVID-19 syndrome. Nat Med. 2021;24:601–615. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Phetsouphanh C, Darley DR, Wilson DB, et al. Immunological dysfunction persists for 8 montsh following intial mild-to-moderate SARS-CoV-2 infection. Nat Immunol. 2022;23:210–216. [DOI] [PubMed] [Google Scholar]
- 21. Mehandru S, Merad M. Pathological sequelae of long-haul COVID. Nat Immunol. 2022;23:194–202. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Locksley RM, Killeen N, Lenardo MJ. The TNF and TNF receptor superfamilies: integrating mammalian biology. Cell. 2001;104:487–501. [DOI] [PubMed] [Google Scholar]
- 23. Qiao J, Liu Z, Fu YX. Adapting conventional cancer treatment for immunotherapy. J Mol Med (Berl). 2016;94:489–495. [DOI] [PubMed] [Google Scholar]
- 24. Fung M, Babik JM. COVID-19 in immunocompromised hosts: what we know so far. Clin Infect Dis. 2021;72:340–350. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.