Tumor biomarkers predict clinical outcome of COVID-19 patients.

Dear Editor,

A recent article in Journal of Infection by Fu and colleagues have summarized the clinical characteristics of coronavirus disease 2019 (COVID-19) in China, and described that those with medical comorbidities tend to have more severe clinical symptoms and higher case-fatality rate, according to data of 43 studies involving 3600 patients. Of the data from China, 81% cases were mild, 14% were severe, and 5% were critical, and the case-fatality rate was 2.3% in all cases and 49.0% in critical cases. Older age and comorbidities, such as cardiovascular disease, confer a higher risk for severe disease, and young and otherwise healthy patients are also at risk for complications. ARDS (Acute respiratory distress syndrome) and respiratory failure, sepsis, acute cardiac injury and heart failure were the most common critical complications during exacerbation of COVID-19. Several laboratory outcomes indicated the severity and the clinical outcome of COVID-19 patients. Previous studies reported that tumor biomarkers, such as carcino embryonic antigen (CEA), cytokeratin 19 fragment (CYFRA21-1), neuron-specific enolase (NSE), squamous cell carcinoma antigen (SCCA) and Pro-Gastrin Releasing Peptide (ProGRP), were elevated in the patients with benign lung disorders, such as pneumonia and pulmonary fibrosis. , , We would like to share our findings that the role of tumor markers related lung cancer in COVID-19 patients as predictive indicators for clinical outcome.

A total of 129 patients diagnosed as COVID-19, with 20 moderate (15.50%), 73 severe (56.59%) and 36 critical severe cases (27.91%) on admission, were included in this study. In addition, a total of 80 age-and gender- matched health individuals were enrolled as controls. The patients self-reported medical history of comorbidities were recorded on admission and were classified as hypertension, cardiovascular disease, diabetes (type 2), chronic obstructive pulmonary disease (COPD) and others. Of 129 cases, 114 cases (88.37%) were discharged from hospital for their recovery from COVID-19, and 15 cases (11.63%) were deceased during the treatment, shown in Supplementary Table 1. For the characteristics of patients, we observed that the mean age of patients was significantly different among the subgroup of severity, and the mean age of patients with critical severe was significant higher than those who with severe or moderate. The distribution of patients with diabetes, chronic kidney disease and others comorbidities have significant differences among the sub-groups of disease severity. Most deceased cases (14/15) were with the critical severe COVID-19 and one with severe COVID-19. Patients who deceased have significant higher ration of comorbidities of chronic kidney disease (p=0.001), shown in Table 1 .

Table 1

Clinical characteristics of the patients according to disease severity and the clinical outcomes

Clinical characteristics	Severity of Disease			P-value	Clinical outcome		P-value
	Moderate	Severe	Critical severe		Discharged	Deceased
No	20	73	36		114	15
Age, Mean(SD),y	61.90(12.89)	66.59(11.49)	70.92(12.32)	0.026	66.40±12.24	72.13±11.86	0.090
Gender, n(%)
Male	11(14.10)	41(52.56)	26(33.33)	0.235	67(85.90)	11(14.10)	0.278
Female	9(17.65)	32(62.75)	10(19.61)		47(92.16)	4(7.84)
Hypertension, n(%)
Yes	5(9.09)	31(56.36)	19(34.55)	0.131	47(85.45)	8(14.55)	0.373
No	15(20.27)	42(56.76)	17(22.97)		67(90.54)	7(9.46)
Cardiovascular disease, n(%)
Yes	4(17.39)	9(39.13)	10(43.48)	0.135	19(82.61)	4(17.39)	0.341
No	16(15.09)	64(60.38)	26(24.53)		95(89.62)	11(10.38)
Diabetes, type 2, n(%)
Yes	5 (17.86)	10 (35.71)	13 (46.43)	0.026	23(82.14)	5(17.86)	0.245
No	15(14.85)	63(62.38)	23(22.77)		91(90.09)	10(9.90)
COPD, n(%)
Yes	0(0)	9(81.82)	2(18.18)	0.163	11(100)	0(0)	0.208
No	20(16.95)	64(54.24)	34(28.81)		103(87.29)	15(12.71)
Chronic kidney disease, n(%)
Yes	1(11.11)	2(22.22)	6(66.67)	0.025	5(55.56)	4(44.44)	0.001
No	19(15.83)	71(59.17)	30(25.00)		109(90.83)	11(9.17)
Others, n(%)
Yes	1(2.78)	21(58.33)	14(38.89)	0.025	30(83.33)	6(16.67)	0.267
No	19(20.43)	52(55.91)	22(23.66)		84 (90.32)	9(9.68)
Clinical outcome, n(%)
Discharge	20(17.54)	72(63.16)	22(19.30)	0.000
Die	0(0)	1(6.67)	14(93.33)
Tumor biomarkers, IQR
CEA (ng/mL)	2.39(1.20,3.82)	3.48(2.26,4.95)	5.03(3.09,8.58)	0.000	3.39(2.14,5.05)	5.43(3.55,11.42)	0.003
CYFRA21-1(ng/mL)	2.24(1.79,3.03)	3.30(2.43,4.09)	5.06(2.78,9.83)	0.000	3.14(2.35,4.05)	9.72(7.32,12.08)	0.000
NSE (ng/mL)	12.76(11.85,15.11)	13.01(10.67,16.28)	12.56(10.74,18.28)	0.970	12.40(10.86, 15.34)	15.92(12.77,28.41)	0.022
SCC (ng/mL)	0.99(0.64,1.68)	1.03(0.73,1.39)	2.62(1.18,4.10)	0.000	1.07(0.75,1.59)	3.59(2.75,8.77)	0.000
proGRP(pg/mL)	42.13(36.56,46.41)	44.06(35.67,57.15)	56.27(32.55,88.89)	0.219	44.61(36.38,61.35)	34.60 (16.57,102.55)	0.476

The plasma concentration of all the five biomarkers were significantly elevated in cases than those in controls (pall<0.01). In addition, the significant differences of plasma level of CEA, CYFRA21-1 and SCCA were observed among the subgroups of severity of disease and clinical outcome (Table 1) and plasma level of CEA, CYFRA21-1, SCCA were significantly increased with the advance serverity of disease. Whereas, there were no significant differences of NSE and proGRP contration amonge the different severity subgroups, shown in Supplementary Figure 1.

To further analyze prognostic role of tumor biomarker in COVID-19 patients, a logistic regression was applied to measure the associations of tumor biomarkers level to risk of death. Crude OR, OR adjusted for age and gender (model 1), and OR adjusted for molel1 plus comordities (model 2) were used to assess the relative risk, respectively. The results revealed that increased level of CEA (OR=1.13, 95%CI:1.03-1.23, p=0.010; adjust model 1 OR=1.12, 95%CI: 1.02-1.23, p=0.016; adjust model 2 OR=1.12, 95%CI: 1.01-1.26, p=0.029), CYFRA21-1 (OR=1.73, 95%CI:1.35-2.21, p=0.000; adjust model 1 OR=1.673, 95%CI: 1.34-2.36, p=0.000; adjust model 2 OR=1.73, 95%CI: 1.32-2.28, p=0.000), NSE (OR=1.09, 95%CI: 1.02-1.17, p=0.016; adjust model 1 OR=1.11, 95%CI: 1.04-1.19, p=0.003; adjust model 2 OR=1.15, 95%CI: 1.05-1.27, p=0.004) and SCCA (OR=1.28, 95%CI: 1.11-1.48, p=0.016; adjust model 1 OR=1.22, 95%CI: 1.06-1.41, p= 0.007; adjust model 2 OR=1.24, 95%CI: 1.07-1.45, p= 0.006) were associated with increased risk of death, respectively, shown in Table 2 .

Table 2

The relative risk of tumor biomarkers to death

Patients	CEA		CYFRA21-1		NSE		SCCA		proGRP
	OR(95%CI)	P value	OR(95%CI)	P value	OR(95%CI)	P value	OR(95%CI)	P value	OR(95%CI)	P value
Discharged	reference		reference		reference		reference		reference
Deceased	1.13(1.03,1.23)	0.010	1.73(1.35,2.21)	0.000	1.09(1.02,1.17)	0.016	1.28(1.11,1.48)	0.001	1.01(0.99,1.02)	0.417
Model1	1.12(1.02,1.23)	0.016	1.73(1.34,2.36)	0.000	1.11(1.04,1.19)	0.003	1.27(1.09,1.48)	0.002	1.00(0.99,1.02)	0.832
Model2	1.12(1.01,1.26)	0.029	1.73(1.32,2.28)	0.000	1.15(1.05,1.27)	0.004	1.24(1.07,1.45)	0.006	1.00(0.98,1.01)	0.601

Model1, adjusted for age and gender; Model2, adjusted for model1 plus comorbidities.

The ROC curve revealed that SCCA (AUC: 0.937, p=0.000, cut-off: 2.57 ng/ml), CYFRA21-1(AUC: 0.882, p=0.000, cut-off: 7.29 ng/ml) and CEA (AUC: 0.737, p=0.003, cut-off: 8.55 ng/ml) could predicte the clinical outcome of COVID-19 patients, shown in Figure 2. The correlation of biomarkers dynamics and patient outcome was also evaluated with 22 discharged and 11 deceased patients, and the result revealed that the increased concentration of CYFRA21-1, SCCA and NSE were the risk of death (Supplementary Table 2), indicating that dynamic monitor for the three biomarkers could predict the clinical outcome of COVID-19 patients.

This study revealed that age, diabetes, chronic kidney disease and other diseases were associated with the severity of COVID-19 patients. In which, chronic kidney disease was also regarded as a risk of death of COVID-19 patients, which was consistent the result of publised data. Acutally, the most common cause of death in COVID-19 patients is viral pneumonia leading to inflammatory response results in the progression to multi-organ failure. Therefore, those patients have history of diabetes, chronic kidney disease were more susceptiable to develop multi-organ failure and lead to death. Tumor biomarkers related lung cancer that CEA, CYFRA21-1, NSE, SCCA, ProGRP were previously reported to be elevated in the pneumonia patients or benign lung diseases. , In this study, we observed that all five tumor biomarkers were significantly increased in the plasma of COVID-19 patients than those in health controls, that CEA, CYFRA21-1 and SCCA were significantly different among the subgroups of severity of disease and clinical outcome, and that CEA, CYFRA21-1, SCCA could predicte the clinical outcome of COVID-19 patients. This study firstly reported the role of tumor biomarkers in COVID-19 patients.

In short, we concluded that the concentrations of tumor biomarkers of CEA, CYFRA21-1, NSE, SCCA, ProGRP were elevated in COVID-19 patients, and that CEA, CYFRA21-1, SCCA could predicte the clinical outcome of COVID-19 patients.

Declaration of Competing Interest

The authors declare no competing interests