Clinical value of tumor-associated antigens and autoantibody panel combination detection in the early diagnostic of lung cancer

Abstract

BACKGROUND:

This study aimed to investigate the efficiency of combining tumor-associated antigens (TAAs) and autoantibodies in the diagnosis of lung cancer.

METHODS:

The serum levels of TAAs and seven autoantibodies (7-AABs) were detected from patients with lung cancer, benign lung disease and healthy controls. The performance of a new panel by combing TAAs and 7-AABs was evaluated for the early diagnosis of lung cancer.

RESULTS:

The positive rate of 7-AABs was higher than the single detection of antibody. The positive rate of the combined detection of 7-AABs in lung cancer group (30.2%) was significantly higher than that of healthy controls (16.8%), but had no statistical difference compared with that of benign lung disease group (20.8%). The positive rate of 7-AABs showed a tendency to increase in lung cancer patients with higher tumor-node-metastasis (TNM) stages. For the pathological subtype analysis, the positive rate of 7-AABs was higher in patients with squamous cell carcinoma and small cell lung cancer than that of adenocarcinoma. The levels of carcinoembryonic antigen (CEA) and cytokeratin 19 fragment 211 (CYFRA 21-1) were significantly higher than that of benign lung disease and healthy control groups. An optimal model was established (including 7-AABs, CEA and CYFRA21-1) to distinguish lung cancer from control groups. The performance of this model was superior than that of single markers, with a sensitivity of 52.26% and specificity of 77.46% in the training group. Further assessment was studied in another validation group, with a sensitivity of 44.02% and specificity of 83%.

CONCLUSIONS:

The diagnostic performance was enhanced by combining 7-AABs, CEA and CYFRA21-1, which has critical value for the screening and early detection of lung cancer.

Keywords

Lung cancer tumor-associated antigen tumor-associated autoantibodies CEA CYFRA 21-1

1. Introduction

Lung cancer is one of the most common malignant tumors, as well as the leading causes of cancer-related death worldwide [1, 2]. The average 5-year survival rate of lung cancer is approximately 17.4%, but this decreases dramatically as the cancer becomes more advanced. Early screening, detection and diagnosis are crucially important to reduce the mortality and improve the overall survival rate of patients with lung cancer [3, 4, 5].

For the early diagnosis of lung cancer, low-dose computed tomography (LDCT) screening is recommended in patients with high risk, demonstrating a sensitivity of 93.8%. The application of LDCT reduces a mortality of 20% in lung cancer patients [6]. However, LDCT cannot distinguish benign nodules from early malignant tumors accurately, which leads to high rate of false-positive results. The repeated radiation exposure and substantial costs also limit the widespread application of CT as a screening procedure in clinical practice [7, 8]. Histopathology is typically used to diagnose lung cancer, but it is invasive. Current serum diagnostic biomarkers for lung cancer focused on tumor-associated antigens (TAAs), such as carcinoembryonic antigen (CEA), carbohydrate antigen 125 (CA-125), cytokeratin 19 fragment 21-1 (CYFRA 21-1), squamous cell carcinoma related antigen (SCC-Ag) and Progastrin releasing peptide (ProGRP). These TAA markers showed an increased positive rate in patients with advanced stages, but are rarely used as early biomarkers because of their low sensitivity and specificity [9]. Recent studies reported that detection of serum tumor-associated autoantibodies (TAAbs), which are produced by host cells against TAAs, may be a promising cancer screening method [10]. TAAbs are more stable in peripheral blood than TAAs, and exhibit better sensitivity and specificity in the previous studies [11, 12]. For single antibody detection, the sensitivity was low and panels of different autoantibodies showed better diagnostic performance, with the sensitivity ranging from 30.0% to 94.8%, specificity ranging from 73% to 100.0%, and auras under ROC curve (AUCs) ranging from 0.630 to 0.982 [12]. The combination of multiple autoantibodies had potential efficacy used as diagnostic tools in tumors [13, 14]. Currently, a panel of seven autoantibodies (7-AABs), including p53, PGP9.5, SOX2, GAGE7, GBU4-5, MAGEA1 and CAGE are more prevalent and proved to be useful blood biomarkers for the diagnosis of early lung cancer or distinguishing benign nodules from malignant tumors [5, 12, 15, 16]. However, the diagnostic sensitivity of 7-AABs varied greatly in different studies, and whether the combination of 7-AABs with TAAs could improve the diagnosis of lung cancer still needs further investigated. Correspondingly, it’s essential to explore more effective combination of serum markers for the screening and early diagnosis of lung cancer.

This study aimed to investigate the clinical value of 7-AABs combined with TAAs, including CEA, CYFRA21-1, SCC-Ag and ProGRP for the identifying of lung cancer not only from healthy controls but also from benign lung diseases.

2. Materials and methods

2.1 Patients

This was a retrospective study and the subjects were recruited from Tongji hospital, Tongji Medical College, Huazhong University of Science and Technology, from June to October 2020. The lung cancer and benign lung disease groups were included according to the pathology results after surgery. In the training set, a total of 199 patients with lung cancer and 72 patients with benign lung diseases (lung benign disease and non-tumor benign disease) according to the pathology results were included, Tumor histology was classified according to the World Health Organization guidelines for histologic type of lung tumors. Tumor-node-metastasis (TNM) stage was determined according to the American Joint Committee on Cancer staging manual (seventh edition). In addition, 172 age and sex matched healthy controls (HCs) were also recruited and determined by interview, physical examination and with normal CT results. The exclusive criteria include HIV and HCV positive, and patients who had antineoplastic therapy, radiotherapy or chemotherapy before surgery or cancer diagnosis. To validate this study, the following subjects were recruited from Sino-French New City of Tongji hospital, including 258 lung cancer, 81 benign lung disease and 170 healthy controls. The study protocol was approved by the ethical committee of Tongji hospital, Tongji Medical College, Huazhong University of Science and Technology (TJ-IRB20210331). All of the subjects provided written informed consent.

2.2 Specimen collection and processing

Serum from 3–5 ml fasting blood was separated by centrifugation at 3000 rpm for 10 min within 4 hours of sample collection. Samples that could not be processed immediately were stored at $-$ 80 ${}^{\circ}$ C for future testing.

2.3 Quantitation of autoantibodies in serum samples

The serum concentrations of autoantibodies were quantitated by a commercial enzymelinked immunosorbent assay (ELISA)-based test kit (Hangzhou Cancer Probe Biotech Co., Ltd, Hangzhou, China). The ELISA kit was tested according to the manufacturer’s recommendations. In brief, the detection kit components were equilibrated to room temperature and diluted based on the instructions. We washed the antigen-coated wells with 200–300 $\mu$ L, 1 $\times$ PBS, for 1 minute. Then 50 $\mu$ L diluted serum samples, standards, and controls were added to the antigen-coated wells and incubated at room temperature for 1 hour. After washing the plate thrice with Microplate Washer using the standard procedure, 50 $\mu$ L of diluted secondary antibodies antihuman IgG HRP was added, and then incubated for 30 minutes. Thereafter, the plate was once again washed thrice as above, and then 100 $\mu$ L of substrate was added, followed by incubation for 15 minutes at room temperature. Subsequently, 50 $\mu$ L stop solution was added to each well and mixed thoroughly. The OD value at 450 nm was read using a spectrophotometer within 30 minutes. The seven autoantibodies’ positive reference values were as follows: p53 $\geqslant$ 13.1 U/mL, PGP9.5 $\geqslant$ 11.1 U/mL, SOX2 $\geqslant$ 10.3 U/mL, GAGE7 $\geqslant$ 14.4 U/mL, GBU4-5 $\geqslant$ 7.0 U/mL, MAGE A1 $\geqslant$ 11.9 U/mL, and CAGE $\geqslant$ 7.2 U/mL. If one of the seven autoantibodies is positive, it will be judged as positive. If all seven autoantibodies are negative, it will be judged as negative.

2.4 Statistical analysis

The data were analyzed using SPSS 22.0, Graphpad Prism 6.0 and MedCalc (Version 18.2.1). The results are presented as mean $\pm$ standard deviation (SD). Differences of antibody levels between two groups were analyzed using nonparametric Mann-Whitney U test. Positivity rates were evaluated by mean of a standard $\chi^{2}$ tests with the respective degrees of freedom. In the training group, binary logistic regression was performed to screen for variables to comprise a panel with the best diagnostic performance. ROC curve was drawn to analyze the diagnostic efficiency. The AUC was estimated with 95% confidence intervals (CIs) using predicted probability values to assess the discriminatory capacity of the multiple-variable model. ROCs for single markers and different subgroups were constructed and compared by MedCalc. A two-tailed $p<$ 0.05 was considered statistically significant.

3. Results

This study enrolled 443 participants for the training group and 509 participants for the validation group. The basic characteristics of the included population were shown in Table 1. A total of 199 pathologically confirmed lung cancer (including 162 adenocarcinoma (AD), 31 squamous cell carcinoma (SCC), 5 small cell lung cancer (SCLC) and 1 large cell lung cancer (LCLC)), 72 benign lung disease (including 6 lung benign disease and 66 non-tumor benign disease) and 172 healthy controls were included in the training group. The validation group comprised 258 lung cancer (including 222 AD, 27 SCC, 8 SCLC and 1 LCLC), 81 benign lung disease (17 lung benign disease and 64 non-tumor benign disease) and 170 healthy controls. According to the TNM stage, 65.3% and 75.6% of the patients with lung cancer were classified as stage I in the training and validation cohorts, respectively.

Table 1
Clinical characteristics of all study participants and positive rates of autoantibodies

	Training group (443)			Validation group ( $n=$ 509)
Characteristics	Lung cancer ( $n=$ 199)	Benign lung disease ( $n=$ 72)	Healthy controls ( $n=$ 172)	Lung cancer ( $n=$ 258)	Benign lung disease ( $n=$ 81)	Healthy controls ( $n=$ 170)
Age, median (years)	55.3	55.5	56.4	56.8	54.9	53.6
Gender
Male (%)	96 (48.2%)	44 (61.1%)	102 (59.3%)	127 (49.2%)	38 (46.9%)	96 (56.5%)
Female (%)	103 (51.8%)	28 (38.9%)	75 (40.7%)	131 (50.8%)	43 (53.1%)	74 (43.5%)
Pathologic type
Adenocarcinoma	162 (81.4%)	0		222 (86.0%)	0
Squamous cell carcinoma	31 (15.6%)	0		27 (10.5%)	0
Small cell lung cancer	5 (2.5%)	0		8 (3.1%)	0
Large cell lung cancer	1 (0.5%)	0		1 (0.4%)	0
Lung benign disease	0	6 (8.3%)		0	17 (21.0%)
Non-tumor benign disease	0	66 (91.7%)		0	64 (79%)
Clinical stage
I	130 (65.3%)			195 (75.6%)
II	16 (8.0%)			25 (9.7%)
III	26 (13.1%)			26 (10.1%)
IV	27 (13.6%)			12 (4.6%)
Tumor-associated autoantibodies
CAGE	7 (3.5%)	0	2 (1.2%)	7 (2.7%)	2 (2.5%)	0
GACE7	12 (6.0%)	2 (2.8%)	0	13 (5.3%)	0	1 (0.6%)
GBU4-5	22 (11.1%)	9 (12.5%)	16 (9.3%)	21 (8.1%)	5 (6.2%)	13 (7.6%)
MACE A1	6 (3.0%)	0	0	7 (2.7%)	1 (2.2%)	1 (0.6%)
PGP9.5	5 (2.5%)	2 (2.8%)	3 (1.7%)	6 (2.3%)	2 (2.5%)	3 (1.8%)
SOX2	12 (6.0%)	3 (4.2%)	8 (4.7%)	27 (10.5%)	8 (9.9%)	5 (2.9%)
p53	9 (4.5%)	2 (2.8%)	5 (2.9%)	7 (2.7%)	3 (3.7%)	2 (1.2%)
7-AABs	60 (30.2%)	15 (20.8%)	29 (16.8%)	60 (23.3%)	18 (22.2%)	25 (14.7%)

Data were presented as median or numbers (%).

Table 2

The positive rate of combined autoantibody detection in different groups of lung cancer patients

	Training group ( $n=$ 199)		$P$ value	Validation group ( $n=$ 258)		$P$ value	Whole group ( $n=$ 457)		$P$ value
Parameters	Cases	Positive rate (%)		Cases	Positive rate (%)		Cases	Positive rate (%)
Age range (year)
$<$ 60	100	32 (32.0%)	0.644	106	28 (26.4%)	0.662	206	60 (29.1%)	0.240
$\geqslant$ 60	99	28 (28.3%)		152	32 (18.4%)		251	60 (23.9%)
Gender
Male	96	31 (32.3%)	0.540	127	35 (27.6%)	0.140	223	66 (29.6%)	0.136
Female	103	29 (28.2%)		131	25 (19.1%)		234	54 (23.1%)
Clinical stage
I	130	28 (21.5%)	0.003	195	37 (19.0%)	0.007	325	68 (20.9%)	$<$ 0.001
II	16	7 (43.8%)		25	8 (32.0%)		41	15 (36.6%)
III	26	11 (42.3%)		26	8 (30.8%)		52	19 (36.5%)
IV	27	14 (51.9%)		12	7 (58.3%)		39	18 (46.2%)
Histology
Adenocarcinoma	162	40 (24.7%)	0.004	222	42 (18.9%)	$<$ 0.001	384	87 (22.7%)	$<$ 0.001
Squamous cell carcinoma	31	16 (51.6%)		27	14 (51.9%)		58	25 (43.1%)
Small cell lung cancer	5	3 (60.0%)		8	4 (50.0%)		13	7 (53.8%)

The serum concentrations of the 7-AABs (including p53, GAGE7, PGP9.5, CAGE, MAGEA1, SOX2, and GBU4-5) were determined by ELISA. By using the commercial assay cutoffs, positivity is defined as having an elevated AABs assay signal. The positive rate of 7-AABs was higher than the single detection of antibody ( $p<$ 0.001) (Table 1). In the training group, the positive rate of 7-AABs in lung cancer group (30.2%) showed no statistical significance compared with that of benign lung disease group (20.8%) ( $\chi^{2}=$ 0.501, $p=$ 0.479), but higher than that of healthy controls (16.8%) ( $\chi^{2}=$ 8.937, $p=$ 0.003). In the validation group, similar results were also observed. The positive rate of 7-AABs in lung cancer (23.3%) had no significant difference compared with that of benign lung disease groups (22.2%) ( $\chi^{2}=$ 0.037, $p=$ 0.848), but higher than that of healthy controls (14.7%) ( $\chi^{2}=$ 4.707, $p=$ 0.030). A detailed analysis of lung cancer patients was conducted and our results show that the positive rate of 7-AABs detection was not affected by age, gender in the training group, validation group and the whole control group. However, the positive rate of 7-AABs showed a tendency to increase in in lung cancer patients with higher tumor-node-metastasis (TNM) stages ( $p<$ 0.001). For the pathological subtype analysis, the positive rate of 7-AABs was higher in squamous cell carcinoma and small cell lung cancer than that of adenocarcinoma patients (Table 2).

Comparisons of single autoantibody biomarker levels in the lung cancer, benign lung disease and healthy control groups were conducted. In the training group, only serum levels of PGP9.5 were markedly higher in lung cancer patients compared with that of benign lung disease and healthy control groups (Fig. 1A and B). In the validation group, the concentrations of CAGE, MACE A1, PGP9.5 were significantly higher in lung cancer group (Fig. 1C). Both in the training and validation cohorts, the levels of CEA and CYFRA21-1 in lung cancer group were significantly increased, but not for SCC-Ag and ProGRP (Fig. 1B and D).

Figure 1.

Comparisons of each tumor-associate autoantibodies and antigens among lung cancer, benign lung disease and healthy controls in the training and validation groups. Data were shown as mean $\pm$ SD in the bar graphs. HCs: healthy controls. * $p<$ 0.05, ** $p<$ 0.01, *** $p<$ 0.001.

ROC analysis was conducted to estimate the performance of 7-AABs and traditional tumor biomarkers using the data of training group (Fig. 2A). The performance of 7-AABs seemed to be unsatisfied for discriminating lung cancer from benign lung disease (AUC 0.547, 95% CI: 0.485–0.607) or from healthy controls (AUC 0.566, 95% CI: 0.514–0.617). New diagnostic panels were established according to the logistic regression analysis. When benign lung disease used as control group, only CEA and CYFRA21-1 were significant variables for predicting the diagnosis of lung cancer. The probability of lung cancer was calculated based on the equation from the logistic regression model:

$\displaystyle P=1/[1+{\rm e}^{[-(-0.33+0.132*\text{CEA}+0.374*\text{CYFRA}21-1% )]}.$

We compared the diagnostic value of the model (AUC: 0.678, 95%CI: 0.606–0.721) and found that the comprehensive performance of the diagnostic model was better than CEA (AUC: 0.607, 95%CI: 0.546–0.666) and CYFRA21-1 (AUC: 0.638, 95%CI: 0.578–0.696) ( $p<$ 0.05). This model showed a sensitivity of 51.76% and specificity of 83.33% to differentiate lung cancer from benign lung disease (Table 3). Additionally, 7-AABs, CEA and CYFRA21-1 were included in the final model in distinguishing lung cancer patients from health controls. The probability of lung cancer was calculated from the following equation:

$\displaystyle P=1/[1+$ $\displaystyle{\rm e}^{[-(-1.114+0.604*7-\text{AABs}+0.276*\text{CEA}+0.141*% \text{CYFRA}21-1)]}.$

The performance of the diagnostic model (AUC: 0.681, 95%CI: 0.631-0.728) was superior than 7-AABs (AUC: 0.566, 95%CI: 0.514–0.617), CEA (AUC: 0.651, 95%CI: 0.546–0.666) and CYFRA21-1 (AUC: 0.592, 95%CI: 0.540–0.642) ( $p<$ 0.05). This model had a sensitivity of 44.72% and specificity of 84.88% to distinguish lung cancer from healthy controls (Table 3). Then benign lung disease and healthy controls were merged as a whole control group, a new diagnostic model was established as the following equation:

$\displaystyle P=1/[1+$ $\displaystyle{\rm e}^{[-(-1.479+0.503*7-\text{AABs}+0.226*\text{CEA}+0.199*% \text{CYFRA}21-1)]}.$

At last, this model exhibited an AUC of 0.686 (95%CI: 0.640–0.729) and was statistically better than the single markers, with a sensitivity of 52.26% and specificity of 77.46% (Table 3).

In the validation group, ROCs of the 7-AABs panel and TAAs were analyzed (Fig. 2B). Between lung cancer and benign lung disease groups, the diagnostic model was validated with an AUC of 0.657 (95%CI: 0.604–0.708), sensitivity of 56.03% and specificity of 69.14%. The diagnostic model for discriminating lung cancer from healthy controls also showed a better diagnostic efficiency with an AUC of 0.663 (95%CI: 0.616–0.708), sensitivity of 47.47% and specificity of 73.26%. In the whole control group, the performance of diagnostic model exhibited an AUC of 0.668 (95%CI: 0.626–0.709), with a sensitivity of 44.02% and specificity of 83.00% (Table 3).

Table 3

Comparison of different markers in distinguishing lung cancer from benign lung disease and healthy controls

		LC VS. Benign lung disease				LC VS. HCs				LC VS. Whole control group
Group	Variable	AUC	95% CI	Sensitivity	Specificity	AUC	95% CI	Sensitivity	Specificity	AUC	95% CI	Sensitivity	Specificity
Training group	AAB	0.547	0.485 to 0.607	30.15	79.17	0.566	0.514 to 0.617	30.15	83.14	0.560	0.513 to 0.607	30.15	81.97
	CEA	0.607	0.546 to 0.666	48.74	72.22	0.651	0.600 to 0.700	83.14	38.69	0.638	0.592 to 0.683	39.20	81.97
	CYFRA21-1	0.638	0.578 to 0.696	52.26	76.39	0.592	0.540 to 0.642	46.73	73.10	0.606	0.558 to 0.651	46.73	74.90
	ProGRP	0.516	0.454 to 0.577	31.16	77.78	0.527	0.474 to 0.579	38.69	75.44	0.514	0.466 to 0.562	70.35	19.34
	SCC-Ag	0.504	0.443 to 0.565	80.90	8.33	0.500	0.448 to 0.552	23.62	86.55	0.501	0.454 to 0.549	76.38	13.58
	Model	0.678	0.606 to 0.721	51.76	83.33	0.681	0.631 to 0.728	51.76	84.88	0.686	0.640 to 0.729	52.26	77.46
Validation group	AAB	0.506	0.451 to 0.560	23.35	77.78	0.532	0.483 to 0.580	23.35	80.23	0.523	0.479 to 0.568	23.35	81.42
	CEA	0.591	0.537 to 0.644	34.24	81.48	0.643	0.595 to 0.688	81.32	37.79	0.626	0.582 to 0.668	28.40	88.93
	CYFRA21-1	0.592	0.538 to 0.645	42.41	76.54	0.580	0.531 to 0.627	43.19	71.35	0.584	0.539 to 0.627	42.41	73.41
	ProGRP	0.524	0.470 to 0.579	37.74	70.37	0.515	0.466 to 0.563	36.96	74.27	0.518	0.473 to 0.562	36.96	73.02
	SCC-Ag	0.555	0.500 to 0.609	21.79	90.12	0.535	0.487 to 0.583	19.07	90.06	0.542	0.497 to 0.585	19.07	90.48
	Model	0.657	0.604 to 0.708	56.03	69.14	0.663	0.616 to 0.708	49.61	75.00	0.668	0.626 to 0.709	44.02	83.00

LC: lung cancer, HCs: healthy controls.

Figure 2.

Receiver operator characteristic (ROC) curves for each single tested biomarkers and diagnostic models in different cohorts. (A) Comparison of the diagnostic ability of the model and each biomarker in the training group. (B) Comparison of the diagnostic ability of the model and each biomarker in the validation group. LC: lung cancer, HCs: healthy controls.

4. Discussion

Lung cancer is one of the most common cancers as well as the leading cause of cancer-related death worldwide [17]. In the early stages, patients with lung cancer are asymptomatic or atypical, and most patients have local or distant metastasis at the time of diagnosis, leading to the low 5-year survival rate (16%) [18]. CT scanning could reduce the incidence of patients with late stage lung cancer, but low specificity of CT might expose patients to radiation and unnecessary further examination [6]. Various blood biomarkers have been reported for the diagnosis of lung cancer, such as circulating tumor cells, tumor DNA, microRNA, traditional TAAs and tumor-associated autoantibodies [18, 19, 20]. Autoantibodies are produced in the process of oncogenesis and can be detected before the presence of TAAs [21, 22]. Combined detection of autoantibodies have been reported to have potential efficacy as diagnostic and prognostic tools in lung cancer [12]. The sensitivity of a single autoantibody is very low due to the complexity and heterogeneity of lung cancer as well as the expression of TAAs [23]. Previous studies have reported that the performance of TAAbs combination detection have been evaluated, and the specificity was high but with low sensitivity [12, 16, 21, 22]. Therefore, to improve the clinical value in early screening and diagnosis of lung cancer in combination of TAAs and TAAbs still needs to be developed.

In the present study, a panel of seven autoantibodies (including p53, PGP9.5, SOX2, GAGE7, GBU4-5, MAGEA1 and CAGE) were assessed and appeared to more frequently present in the serum of lung cancer than that of healthy controls, but not for benign lung disease. Our results show that the positive rate of 7-AABs was higher in SCC and SCLC than AD [16], and had an increased tendency in patients with progressed TNM stages. However, some studies presented that no obvious differences of positive rates were observed in lung cancer patients with different TNM stages, which might be due to the different participants and needs to be investigated in larger populations [5, 24].

The traditional tumor biomarkers, such as CEA, CYFRA21-1, SCC-Ag and ProGRP have been widely applied as diagnostic and predictive markers of lung cancer [25]. CEA was reported to be highly associated with primary tissue, lymphatic metastasis, and distant metastasis [26]. CYFRA21-1 was expressed in respiratory epithelial cells and could be detected in NSCLC tissue [27]. The expression of SCC-Ag was correlated with differentiation degree of squamous cell carcinoma tissue but had no additional value when used in combination with CYFRA21-1 due to its low sensitivity [28]. ProGRP was a useful biomarker for the diagnostic and evaluation of SCLC [29]. In the current study, the levels of CEA and CYFRA21-1 were observed significantly increased in lung cancer group both in the training and validation cohorts, which suggested to be useful markers to establish a well optimized model with strong predictive value.

Further multiple logistic regression analysis was performed to select biomarkers for a mixed model. Previous studies showed that combination detection of 7-AABs and serum tumor markers could improve the diagnostic accuracy for lung cancer [30]. Unexpectedly, only CEA and CYFRA21-1 were included to yield a final model for distinguishing lung cancer from benign lung disease. When distinguishing lung cancer from healthy controls, 7-AABs, CEA and CYFRA21-1 were significant variables. Moreover, benign lung disease and healthy controls were used as a whole control group, the diagnostic performance was evaluated using a new model including 7-AABs, CEA and CYFRA21-1, with a sensitivity of 52.26, specificity of 77.46%. Further validation was also conducted among different groups and similar results were obtained in the validation group. Accordingly, 7-AABs could contribute the early diagnostic of lung cancer, especially when differentiating from healthy controls and combination of 7-AABs, CEA and CYFRA21-1 could improve the whole diagnostic performance. The performance of 7-AABs in the current study was not as superior as reported, which might be due to the heterogeneity of cohorts and detection platform within different studies [31, 32]. Therefore, larger numbers of samples to validate the efficiency of TAAs and autoantibody combination detection should be done and further non-invasive markers still need to be investigated for the early diagnostic of lung cancer, especially when differentiated from benign lung disease.

5. Conclusion

This study showed that the positive rate of 7-AABs was higher in lung cancer group than that of healthy controls. A new panel of detecting CEA, CYFRA21-1 and 7-AABs was established and could contribute the early detection of lung cancer in some extent. Biomarkers for the screening and early diagnostic of lung cancer still need to be explored.

Funding

This study was supported by the National Mega Project on Major Infectious Disease Prevention (2017 ZX10103005-007).

Author contribution

Hongyan Hou designed the project and wrote the original draft; Renren Ouyang, Shiji Wu, Ting Wang, Wei Wei, Min Huang collected the data; Bo Zhang, Botao Yin, Jin Huang and Minxia Zhang analyzed the data; Yun Wang and Feng Wang reviewed and edited the manuscript.

Ethics statement

The study protocol was approved by the ethical committee of Tongji hospital, Tongji Medical College, Huazhong University of Science and Technology.

Data availability statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.

Footnotes

Acknowledgments

This study was supported by the National Mega Project on Major Infectious Disease Prevention (2017 ZX10103005-007).

Conflict of interest

The authors have no conflicts of interest to declare.

References

Feng

R.M.

Zong

Y.N.

Cao

S.M.

and Xu

R.H.

, Current cancer situation in China: Good or bad news from the 2018 Global Cancer Statistics? Cancer Commun (Lond) 39 (2019), 22.

Miller

K.D.

Nogueira

Mariotto

A.B.

Rowland

J.H.

Yabroff

K.R.

Alfano

C.M.

Jemal

Kramer

J.L.

and Siegel

R.L.

, Cancer treatment and survivorship statistics, CA Cancer J Clin 69 (2019), 363–385.

Ren

Zhang

Jiang

Cai

Zhou

Liu

Zhang

Zhou

and Hirsch

F.R.

, Early detection of lung cancer by using an autoantibody panel in Chinese population, Oncoimmunology 7 (2018), e1384108.

Sullivan

F.M.

Farmer

Mair

F.S.

Treweek

Kendrick

Jackson

Robertson

Briggs

McCowan

Bedford

Young

Vedhara

Gallant

Littleford

Robertson

Sewell

Dorward

Sarvesvaran

and Schembri

, Detection in blood of autoantibodies to tumour antigens as a case-finding method in lung cancer using the EarlyCDT(R)-Lung Test (ECLS): Study protocol for a randomized controlled trial, BMC Cancer 17 (2017), 187.

Wang

Yan

Yuan

Slamon

Hou

Wang

and Wang

, Significance of tumor-associated autoantibodies in the early diagnosis of lung cancer, Clin Respir J 12 (2018), 2020–2028.

National Lung Screening Trial Research Team Aberle

D.R.

Adams

A.M.

Berg

C.D.

Black

W.C.

Clapp

J.D.

Fagerstrom

R.M.

Gareen

I.F.

Gatsonis

Marcus

P.M.

and Sicks

J.D.

, Reduced lung-cancer mortality with low-dose computed tomographic screening, N Engl J Med 365 (2011), 395–409.

Brenner

D.J.

and Hall

E.J.

, Computed tomography–an increasing source of radiation exposure, N Engl J Med 357 (2007), 2277–2284.

Swensen

S.J.

Jett

J.R.

Hartman

T.E.

Midthun

D.E.

Mandrekar

S.J.

Hillman

S.L.

Sykes

A.M.

Aughenbaugh

G.L.

Bungum

A.O.

and Allen

K.L.

, CT screening for lung cancer: Five-year prospective experience, Radiology 235 (2005), 259–265.

Nakamura

and Nishimura

, History, molecular features, and clinical importance of conventional serum biomarkers in lung cancer, Surg Today 47 (2017), 1037–1059.

10.

Gerber

H.P.

Sibener

L.V.

Lee

L.J.

and Gee

, Intracellular targets as source for cleaner targets for the treatment of solid tumors, Biochem Pharmacol 168 (2019), 275–284.

11.

Desmetz

Mange

Maudelonde

and Solassol

, Autoantibody signatures: Progress and perspectives for early cancer detection, J Cell Mol Med 15 (2011), 2013–2024.

12.

Yang

Ren

and Yin

, Autoantibodies as diagnostic biomarkers for lung cancer: A systematic review, Cell Death Discov 5 (2019), 126.

13.

Ushigome

Nabeya

Soda

Takiguchi

Kuwajima

Tagawa

Matsushita

Koike

Funahashi

and Shimada

, Multi-panel assay of serum autoantibodies in colorectal cancer, Int J Clin Oncol 23 (2018), 917–923.

14.

Xie

and Sun

, Clinical value of seven autoantibodies combined detection in the diagnosis of lung cancer, J Clin Lab Anal 34 (2020), e23349.

15.

Tang

Z.M.

Ling

Z.G.

Wang

C.M.

Y.B.

and Kong

J.L.

, Serum tumor-associated autoantibodies as diagnostic biomarkers for lung cancer: A systematic review and meta-analysis, PLoS One 12 (2017), e0182117.

16.

Huang

Luo

Sun

Wang

and Zhang

, The diagnostic efficiency of seven autoantibodies in lung cancer, Eur J Cancer Prev 29 (2020), 315–320.

17.

Jemal

Siegel

and Ward

, Cancer statistics, 2010, CA Cancer J Clin 60 (2010), 277–300.

18.

Hennessey

P.T.

Sanford

Choudhary

Mydlarz

W.W.

Brown

Adai

A.T.

Ochs

M.F.

Ahrendt

S.A.

Mambo

and Califano

J.A.

, Serum microRNA biomarkers for detection of non-small cell lung cancer, PLoS One 7 (2012), e32307.

19.

Hofman

Bonnetaud

Ilie

M.I.

Vielh

Vignaud

J.M.

Flejou

J.F.

Lantuejoul

Piaton

Mourad

Butori

Selva

Poudenx

Sibon

Kelhef

Venissac

Jais

J.P.

Mouroux

Molina

T.J.

and Hofman

, Preoperative circulating tumor cell detection using the isolation by size of epithelial tumor cell method for patients with lung cancer is a new prognostic biomarker, Clin Cancer Res 17 (2011), 827–835.

20.

Krebs

M.G.

Hou

J.M.

Sloane

Lancashire

Priest

Nonaka

Ward

T.H.

Backen

Clack

Hughes

Ranson

Blackhall

F.H.

and Dive

, Analysis of circulating tumor cells in patients with non-small cell lung cancer using epithelial marker-dependent and -independent approaches, J Thorac Oncol 7 (2012), 306–315.

21.

Chapman

C.J.

Murray

McElveen

J.E.

Sahin

Luxemburger

Tureci

Wiewrodt

Barnes

A.C.

and Robertson

J.F.

, Autoantibodies in lung cancer: Possibilities for early detection and subsequent cure, Thorax 63 (2008), 228–233.

22.

Tan

H.T.

Low

Lim

S.G.

and Chung

M.C.

, Serum autoantibodies as biomarkers for early cancer detection, FEBS J 276 (2009), 6880–6904.

23.

Zhang

J.Y.

Casiano

C.A.

Peng

X.X.

Koziol

J.A.

Chan

E.K.

and Tan

E.M.

, Enhancement of antibody detection in cancer using panel of recombinant tumor-associated antigens, Cancer Epidemiol Biomarkers Prev 12 (2003), 136–143.

24.

Huo

Guo

Gao

Liu

Zhang

and Qin

, Case study of an autoantibody panel for early detection of lung cancer and ground-glass nodules, J Cancer Res Clin Oncol 146 (2020), 3349–3357.

25.

Grunnet

and Sorensen

J.B.

, Carcinoembryonic antigen (CEA) as tumor marker in lung cancer, Lung Cancer 76 (2012), 138–143.

26.

Chen

Z.Q.

Huang

L.S.

and Zhu

, Assessment of seven clinical tumor markers in diagnosis of non-small-cell lung cancer, Dis Markers 2018 (2018), 9845123.

27.

Pujol

J.L.

Grenier

Daures

J.P.

Daver

Pujol

and Michel

F.B.

, Serum fragment of cytokeratin subunit 19 measured by CYFRA 21-1 immunoradiometric assay as a marker of lung cancer, Cancer Res 53 (1993), 61–66.

28.

Kagohashi

Satoh

Ishikawa

Ohtsuka

and Sekizawa

, A re-evaluation of squamous cell carcinoma antigen (SCC) as a serum marker for non-small cell lung cancer, Med Oncol 25 (2008), 187–189.

29.

Wojcik

and Kulpa

J.K.

, Pro-gastrin-releasing peptide (ProGRP) as a biomarker in small-cell lung cancer diagnosis, monitoring and evaluation of treatment response, Lung Cancer (Auckl) 8 (2017), 231–240.

30.

Zang

Jin

Wang

Lei

Che

Mao

Huang

Liu

Zheng

Zhou

Gao

Sun

and He

, Enhancement of diagnostic performance in lung cancers by combining CEA and CA125 with autoantibodies detection, Oncoimmunology 8 (2019), e1625689.

31.

Broodman

Lindemans

van Sten

Bischoff

and Luider

, Serum protein markers for the early detection of lung cancer: A focus on autoantibodies, J Proteome Res 16 (2017), 3–13.

32.

Dai

Tsay

J.C.

Yie

T.A.

Munger

J.S.

Pass

Rom

W.N.

Zhang

Tan

E.M.

and Zhang

J.Y.

, Autoantibodies against tumor-associated antigens in the early detection of lung cancer, Lung Cancer 99 (2016), 172–179.

Clinical value of tumor-associated antigens and autoantibody panel combination detection in the early diagnostic of lung cancer

Abstract

BACKGROUND:

METHODS:

RESULTS:

CONCLUSIONS:

Keywords

1. Introduction

2. Materials and methods

2.1 Patients

2.2 Specimen collection and processing

2.3 Quantitation of autoantibodies in serum samples

2.4 Statistical analysis

3. Results

Table 1 Clinical characteristics of all study participants and positive rates of autoantibodies

5. Conclusion

Funding

Author contribution

Ethics statement

Data availability statement

Footnotes

Acknowledgments

Conflict of interest

References

Table 1
Clinical characteristics of all study participants and positive rates of autoantibodies