Lymphocyte May Be a Reference Index of the Outcome of Cancer Patients in COVID-19 Infection

Background: The novel coronavirus (COVID-19)–infected pneumonia is an international concern as it spreads through human populations and across national and international borders. Methods: In this retrospective study, we consecutively included all cancer cases who had been identied as having a nucleic acid-conrmed COVID-19 infection from two designated hospitals in Wuhan, China. Non-cancer patients were also enrolled for comparison. The clinical data were gathered from the medical recordsfrom Jan 14 to March 12. Results: Among the 117 cancer patients infected with COVID19, the median age was 63 years and 48.7% were male. Male, hematologic cancer, dyspnea on admission, and anti-cancer therapy signicantly increased the risk of death. The amounts of cytokines and immune cells were correlated with the outcomeofcancer patients infected with COVIP-19. However, high level of TNF-a, IL-2R, IL-6, IL-8 did not increase the risk of death in non-cancer patients. Moreover, IL-2R and IL-6 markedly decreased in cancer patients recovered from COVID-19. Conclusions: Cancer patients with COVID-19 were associated with high mortality (23.9%).The amounts of cytokines and lymphocytes could be utilized as the reference index in predicting the survival outcome of cancer patients with COVID-19.


Background
In December 2019, several pneumonia cases of unknown aetiology were identi ed in Wuhan, China [1].
The pathogen has been identi ed as a new enveloped RNA betacoronavirus known as 2019 novel coronavirus disease . Until June 28, 2020, more than 10 million laboratory-con rmed cases have been recorded globally.
Coronaviruses mainly cause respiratory tract infections, such as Middle East respiratory syndrome coronavirus (MERS-CoV) and severe acute respiratory syndrome coronavirus (SARS-CoV) [2], which have resulted in more than 10000 cases in the past twenty years, with death rates of 37% for MERS-CoV and 10% for SARS-CoV [3,4]. In contrast to MERS-CoV and SARS-CoV, COVID-19 have caused more deaths by multiple organ failure rather than respiratory failure, due to the widespread distribution of ACE2 (angiotensin converting enzyme 2)-the receptor for COVID-19-in various organs [5]. The main conditions predisposing to infection were malignancy and immune-suppressive therapy, such as chemotherapy or surgery [6]. Thus, cancer patients might be more susceptible to COVID-19 and have a poor prognosis.
Given the rapid spread of COVID-19, we aimed to describe the epidemiological, clinical, and laboratory parameters, treatment, and outcomes of cancer patients con rmed to have COVID-19 infection, and to nd the factor predicting the outcome of cancer patients infected with COVID-19.

Patients
In this retrospective, two-center study, we reviewed the medical records for all patients with laboratorycon rmed COVID-19 infection and cancer history who were hospitalized in Tongji hospital of Tongji Medical College, Huazhong University of Science and Technology and Renmin Hospital of Wuhan University. The admission data of these patients was from Jan 14, 2020 to March 12, 2020. Following the WHO interim guidance, COVID-19 was diagnosed by real-time quantitative reverse transcriptasepolymerase chain reaction (RT-PCR) assay or high-throughput sequencing of pharyngeal swab specimens. Nucleic acid con rmation of COVID-19 was conducted at the Chinese Center for Disease Prevention and Control until January 23, 2020, and subsequently in the two designated hospital in Wuhan, China. Only nucleic acid-con rmed patients were incorporated into the analysis.
We extracted the clinical symptom, comorbidity, and laboratory ndings from electronic medical records. Patients were followed up to April 20, 2020. If additional information or clari cation was required, data were obtained directly from attending doctors or other health care providers. For each patient, the symptoms during their hospitalization were documented in the present study, including fever, cough, expectoration, fatigue, diarrhea, etc. Fever was de ned as axillary temperature more than 37.5 °C. All laboratory assessments were carried out depending on our institutional guidelines and clinical conditions of the patients. Laboratory testing included whole blood count, blood chemistry, coagulation test, lymphocyte subpopulation, and measures of procalcitonin, C-reactive protein (CRP), and serum cytokines.
The primary composite end-point was death. The study was granted approval by the ethical committee of our institute with a waiver of written informed consent for emerging infectious diseases.

Statistical Analysis
Continuous variables were shown as median and compared by the Mann-Whitney U test; categorical variables were summarized as number (%) and compared by χ2 test or Fisher's exact test between different groups. Scatter plots were drawn to describe lymphocyte subpopulation and plasma cytokine concentrations. The Kaplan-Meier method was utilized to measure the correlation of cytokines and lymphocytes with survival, and the Cox proportion hazard analysis was utilized to clarify potential signi cant differences in outcome. A two-sided p value of less than 0.05 was considered signi cant for all applied statistical test. All the analyses were calculated with the use of R language, version 3.6.3 (http://www.r-project.org/).

Demographic and clinical characteristics
In the 117 cancer patients with laboratory-con rmed COVID-19 infection who had been hospitalized at the two designated hospital as of March 12, 2020, 56 had received anti-cancer treatment, including surgery (16.2%), chemotherapy (19.7%), or radiotherapy (4.3%) as the most recent treatment in the last year and the other patients were cancer survivors in routine follow-up. The demographic and clinical parameters of the 117 patients are listed in Table 1. The median age was 63 years (interquartile range, 56 to 70). A total of 48.7% were male. Lung cancer was the most common type (25.6%); less common were digestive system cancer (23.9%), breast cancer (15.4%), thyroid cancer (8.5%), urinary system cancer (7.7%), gynecological tumor (6.8%), hematologic malignancies (6.8%), sarcoma (2.6%), head and neck cancer (0.9%) and cancer of unknown primary site (1.7%). Among the overall population, 48.7% had more than one coexisting illness, including diabetes (15.4%), hypertension (30.8%), cardiovascular disease (9.4%) and chronic obstructive pulmonary disease (COPD) (14.5%). Fever was present in 71.8% of the patients on admission. Less common symptom was cough (65.0%) and myalgia (50.4%); sputum production (39.3%), and diarrhea (14.5%) was uncommon.  Annotation: Lymphocytopenia was de ned as a lymphocyte count of less than 1000 per cubic millimeter. Thrombocytopenia was de ned as a platelet count of less than 100,000 per cubic millimeter. APTT indicates activated partial thromboplastin time. AST indicates aspartate aminotransferase Compared to alive patients, dead patients had increased levels of TNF-α, IL-2R, IL-6, IL-8, and IL-10 tested on admission ( Fig. 1 and Supplemental Fig. 1). However, high level of TNF-α, IL-2R, IL-6, IL-8 did not raise the risk of death in non-cancer patients. Importantly, the levels of TNF-α, IL-2R, IL-6, IL-8, and IL-10 were signi cantly correlated with survival time of cancer patients with COVIP-19. These odds were further proved by logistic regression model after adjusting for other risk factors, including age, gender, cancer type, comorbidities, symptom and anti-cancer treatment (Table 3). In addition, T lymphocytes, B lymphocytes, T helper (Th) cells, and T suppressor (Ts) cells were diminished in cancer patients with a primary composite end-point event of death (Fig. 2). Cancer patients with more B cells and Th cells had longer survival than those with less B cells and Th cells (p = 0.007, p = 0.035). We further examined if the levels of cytokines were changed when cancer patients recovered from COVID-19. Figure 3 showed that IL-2R and IL-6 signi cantly decreased in the patients with better illness. These results strongly suggesting

Discussion
In this study, we exhibited clinical data of 117 cancer patients and paired non-cancer patients with nuclei acid-con rmed COVID-19 infections. Similar to previous studies, fever and cough were the main symptoms and digestive symptoms were infrequent, which indicates a difference in viral tropism as compared to MERS-CoV, SARS-CoV, and seasonal in uenza [7,8] Our study demonstrated that cancer patients had more prominent laboratory abnormalities than noncancer patients, accompanied with cancer patients seriously ill as compared to patients without history of cancer [1,9,10]. 28 of the 117 cases progressed to severe pneumonia as of March 30, 2020, and died during hospitalization. Additionally, we showed that male, hematologic malignancies and dyspnea had been related to increased risk of death; patients who underwent anti-cancer therapy in the past year, had worse survival outcomes from COVID-19, alerting as a timely reminder to doctors that more attention should be paid to cancer patients, especially in male patients or those with speci c cancer type.
Lymphocytopenia was common and associated with death, a nding that was in agreement with previous observations [11]. Many cytokines also regulate the immune system and play a role in virus immunity. TNFα secretion promotes T lymphocyte proliferation and survival, whereas IL-2 levels re ect the T lymphocyte activation status toward a Th1 immune response [12]. IL-6 is involved in the differentiation of B lymphocytes into Ig-secreting cells and participates in lymphocyte and monocyte differentiation [13]. IL-8 is a chemotactic factor that attracts T lymphocytes, neutrophils, basophils [14].
Previous studies have proved that elevated concentration of serum proin ammatory cytokines was related to pulmonary in ammation and respiratory failure in SARS patients [15]. MERS-CoV infection was also found to induce increased levels of proin ammatory cytokines [16]. In this study, cancer patients with higher concentration of TNF-α, IL-2R, IL-6 and IL-10 had higher risk of death. In addition, we analyzed the alteration of major immune cell types (CD4 + , CD8 + T cells, B cells and NK cells) in 117 cancer patients and showed that lymphocytes were diminished in cancer patients died from COVID-19. This result suggests that COVID-19 might mainly act on immune cells, especially lymphocytes. Virus spreads through the respiratory and digestive mucosa and infects target cells, leads to a cytokine storm in severity illness cancer patients, generates the immune responses, and causes alteration in leukocytes and lymphocytes. Patients with severe illness developed acute respiratory distress syndrome and required mechanical ventilation. The decrease in the counts of lymphocytes suggests that the virus consumes the immune cells and blocks the immune function [17]. Lymphocyte de ciency might accelerate to exacerbations of patients [11]. In our study, IL-2R and IL-6 signi cantly decreased in the patients recovered from COVID-19. These results strongly suggesting that the number of cytokines and lymphocytes could be used as the prognostic factor in predicting the outcome of cancer patients with COVID-19. Further studies that characterize the immune response in cancer patients with COVID-19 are required.
This study has a few notable limitations. First, our study was undertaken in two designated hospital in Wuhan. Large-scale multicenter clinical trials are needed to clarify our ndings. Second, data collection was clinically driven and not systematic. Third, we might leave out a series of asymptomatic patients or patients who had mild symptoms and were treated at home. Thus, our cohort represented the more severe population of cancer patients with COVID-19.

Conclusions
COVID-19 has spread rapidly since December 2019, and has been demonstrated to have a severe impact on human health. Moreover, compared to non-cancer population, COVID-19 infection caused higher proportion of multiple organ dysfunction syndrome in cancer patients and was linked to higher mortality (23.9%). Special attention should be given to these patients.

Declarations
Ethics approval and consent to participate The study isapproved by the ethical committee of Tongji hospital, Tongji Medical College, Huazhong University of Science and Technology with a waiver of written informed consent for emerging infectious diseases.

Consent for publication
Not applicable.