*Contributed equally
The early diagnosis of lung cancer is closely associated with the decline of mortality. A panel consisting of seven lung cancer-related autoantibodies (7-AABs) has been shown to be a reliable and specific indicator for the early detection of lung cancer, with a specificity of ~90% and a positive predictive value of ~85%. However, its low sensitivity and negative predictive value limit its wide application. To improve its diagnostic value, the diagnostic efficiencies of 7-AABs in combination with non-specific tumor markers were retrospectively investigated for the detection of early-stage lung cancer. A total of 217 patients with small lung nodules who presented with ground-glass opacity or solid nodules as well as 30 healthy controls were studied. The concentrations of 7-AABs and heat shock protein 90a (HSP90a) were assessed using ELISA. Automated flow fluorescence immune analysis was used for the assessment of CEA, CYFRA21-1, CA199 and CA125 levels. The results showed that 7-AABs + HSP90a possessed a remarkably improved diagnostic efficiency for patients with small pulmonary nodules or for patients with lung nodules of different types, which suggested that 7-AABs in combination with HSP90a could have a high clinical value for the improvement of the diagnostic efficiency of early-stage lung cancer.
Lung cancer is currently the leading cause of cancer-related mortality worldwide (
Recently, lung cancer-related autoantibodies (AABs) have been reported as biomarkers with high specificity, good stability and non-invasion have been identified and have become a focus of research in the early diagnosis of high-risk lung cancer cohorts (
A previous study reported that heat shock protein 90a (HSP90a), which maintains the stability of various protein molecules, is closely related to the occurrence and development of malignant tumors (
In the present study, the level of each lung cancer-related AAB in patients with malignant lung nodules (MLNs) or benign lung nodules (BLNs), as well as HCs, was investigated. The sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV) and relative risk (RR) of the 7-AABs with combinations of the 7-AABs and other non-specific tumor markers, such as HSP90a, CEA, CYFRA21-1, carbohydrate antigen 199 (CA199) or carbohydrate antigen 125 (CA125) were assessed. Furthermore, the diagnostic efficiency of the aforementioned markers in patients with different sizes, types and stages of lung nodules was evaluated.
A total of 217 individuals who were in good health, with lung nodules (ground-glass opacity or solid nodule) firstly diagnosed using high-resolution CT (HRCT) scans during routine examinations were enrolled as patients. Patients were recruited from October 2020 to August 2021 in The First Affiliated Hospital of Wannan Medical College (Wuhu, China). There were 78 males and 139 females, and the average age was 54.5±1.3 (mean ± standard deviation) years. The exclusion criteria for patients were as follows: Chronic obstructive pulmonary disease (including chronic bronchitis, pneumonectomies and pulmonary heart disease), pulmonary fibrosis, or a history of treated or untreated pulmonary malignancy. A total of 30 individuals who were in good health without underlying lung diseases, nor other major diseases screened for in the routine examinations were selected as healthy controls. The present study was approved by the Ethics Review Committee of the First Affiliated Hospital of Wannan Medical College (approval no. 202076; Wuhu, China) and the experimental procedures were in agreement with The Declaration of Helsinki. Written informed consent was obtained from all participants.
The serum concentrations of 7-AABs were quantified using the Seven Autoantibodies Detection Kit (ELISA) (cat. no. 20210106. CancerProbe Biotechnology Co., Ltd.). The concentration of HSP90a in the blood was assessed using the Human Heat Shock Protein HSP90-α ELISA kit (cat. no. HUEB0886; Assay Genie Co., Ltd.). CEA, CYFRA21-1, CA199 and CA125 were assessed using a Multiple Tumor Markers Detection Kit (cat. no. LP120280; Shanghai Tellgen Life Science Co., Ltd.) and were quantified using a Tesmi F4000 automated flow fluorescence immune analyzer (Shanghai Tellgen Life Science Co., Ltd.).
Tissue samples from the patients with lung nodules were immersed and fixed using 4% formaldehyde at room temperature (~20˚C) for 24 h, embedded in paraffin, and then sectioned at 3-5 µm thickness. Sections were baked at 70˚C for 1 h and dewaxed using xylene, before rehydration in a 100, 95, 85 and 75% ethanol series for 5 min at a time. Then, the sections were transferred into distilled water for 5 min. Next, the sections were stained using H&E at room temperature for 5 min, followed by dehydration with absolute ethyl alcohol at room temperature for 5 min and sealed using neutral gum (cat. no. N116470, Shanghai Aladdin Biochemical Technology Co., Ltd.) for assessment. Following the World Health Organization Classification of Tumours (5th Edition)-Thoracic Tumours (
Data were statistically analyzed using SPSS version 22.0 (IBM Corp.). After the Shapiro-Wilk test was performed, comparison of the levels of each AAB (namely, p53, PGP9.5, SOX2, GAGE7, GBU4-5, MAGEA1 and CAGE) and HSP90a among the three groups was performed using the Kruskal-Wallis test followed by Dunn's test for non-parametric data. Comparison of the diagnostic efficiency of 7-AABs for discrimination between MLNs and BLNs in different groups was performed using the χ2 test. Clinical performance was presented in terms of sensitivity (the percentage of true positives) and specificity (the percentage of true negatives). PPV (the probability of MLNs given a positive test result), NPV (the probability of BLNs given a negative test result) and RR (the proportion of cases with a positive outcome in the two groups) were also calculated. McNemar's test was used for comparison of the sensitivity between 7-AABs + HSP90a and other groups (such as 7-AABs, 7-AABs + CEA, 7-AABs + CYFRA21-1, 7-AABs + CA199 and 7-AABs + CA125). For comparison of the NPV between 7-AABs + HSP90a and other groups, the marginal regression model based on the generalized estimation equation was used as follows: Model, g[P(D=1|Z, X=0)]=αN + βNZ, where g represented the contiguous function, Z represented the indicator variable and X represented the diagnostic result. Null hypothesis (H0) β=0 was used to check the difference between tests. When g took the natural logarithmic function, eβ represented an estimate of the relative NPV. The diagnostic efficiencies of 7-AABs, 7-AABs + HSP90a, 7-AABs + CEA, 7-AABs + CYFRA21-1, 7-AABs + CA199 and 7-AABs + CA125 were assessed using receiver operating characteristic (ROC) curves. All tests were two-tailed, and P<0.05 was considered to indicate a statistically significant difference. The data were presented as mean ± standard deviation, number of subjects (n) and percentage (%). Three replicate experiments were performed for each measurement.
A total of 217 patients with lung nodules and 30 HCs were enrolled in the present study.
Of the 217 enrolled patients, 68 (31.3%) and 149 (68.7%) patients, with lung nodules ≤8 mm and 8-20 mm, respectively, were evaluated using HRCT. Moreover, 58 (26.7%), 97 (44.7%) and 62 (28.6%) patients with pure ground-glass opacity (pGGO), mix ground-glass opacity (mGGO) and solid nodule (SN), respectively, were also evaluated using HRCT (
The levels of 7-AABs, namely p53, PGP9.5, SOX2, GAGE7, GBU4-5, MAGEA1 and CAGE in serum samples and plasma HSP90a levels in 217 patients and 30 HCs were assessed using ELISA. The levels of p53, PGP9.5, SOX2, GAGE7, MAGEA1 and HSP90a in the MLN group were significantly elevated compared with those in the HC group (
The diagnostic values of 7-AABs alone or in combination with non-specific tumor markers for patients with lung nodules were evaluated. It was found that although the 7-AABs demonstrated a high specificity (77.6%; 95% CI, 64.4-87.1%) and PPV (84.0%; 95% CI, 73.8-90.8%) in patients with lung nodules, a low sensitivity (42.8%; 95% CI, 35.0-50.9%) and NPV (33.1%; 95% CI, 25.4-41.7%) were also demonstrated. Moreover, 7-AABs reflected a 1.3-fold increase in RR of MLNs, which demonstrated a good diagnostic efficiency (AUC=0.612, P=0.012). However, statistical analysis indicated that 7-AABs + HSP90a exhibited an elevated sensitivity (87.4%; 95% CI, 81.0-92.0%; P<0.0001) and NPV (66.1%; 95% CI, 52.5-77.6%; P<0.0001) compared with 7-AABs alone or in combination with other non-specific tumor markers, such as CEA, CYFRA21-1, CA199 or CA125. Furthermore, a positive result for 7-AABs + HSP90a reflected a 2.6-fold increase in RR for MLNs, which was significantly higher compared with 7-AABs alone or other combinations (
The diagnostic values of 7-AABs alone or in combination with non-specific tumor markers for patients with pulmonary nodules of 8-20 mm or <8 mm was assessed. The results demonstrated a better diagnostic value of 7-AABs for discrimination between MLNs and BLNs in the 8-20 mm group (χ2=5.4, P=0.02) compared with the <8 mm group (χ2=2.9, P=0.09). Furthermore, 7-AABs + HSP90a demonstrated an elevated sensitivity (all P<0.0001) and NPV (all P<0.0001) compared with 7-AABs alone or in combination with other non-specific tumor markers, such as CEA, CYFRA21-1, CA199 and CA125, in both the 8-20 mm and <8 mm groups. A positive result for 7-AABs + HSP90a indicated a 2.2-fold and 4.0-fold increase in the RR of MLNs in the 8-20 mm and <8 mm groups, respectively (
The diagnostic values of 7-AABs alone or in combination with non-specific tumor markers for patients with pGGO, mGGO or SN were evaluated. Statistical analysis indicated that a better diagnostic value of 7-AABs for discrimination between MLNs and BLNs was demonstrated in the pGGO group (χ2=3.9, P=0.048) compared with the mGGO (χ2=1.2, P=0.27) and SN (χ2=2.1, P=0.15) groups. It was noteworthy that 7-AABs + HSP90a exhibited an increased sensitivity and NPV compared with 7-AABs alone or in combination with other non-specific tumor markers, such as CEA, CYFRA21-1, CA199 and CA125, in the three groups. A positive result for 7-AABs + HSP90a indicated a 3.3-fold, 1.5-fold and 5.5-fold increase in the RR for MLNs in the pGGO, mGGO and SN groups, respectively (
Early-stage NSCLC has a notably better prognosis compared with advanced-stage NSCLC, and can usually be treated radically with relatively benign outcomes (
In the present study, 217 patients with different types of small pulmonary nodules (≤20 mm), such as pGGO, mGGO and SN, were assessed. The results indicated that p53, PGP9.5, SOX2, GAGE7 and MAGEA1 protein expression levels were significantly increased in MLNs compared with BLNs and HCs. The specificity, sensitivity, PPV and NPV of 7-AABs for discrimination between MLNs and BLNs were 77.6, 42.8, 84.0 and 33.1%, respectively, which was similar those values previously reported (
However, due to the limitation in the number of patients with SCLC, larger pulmonary nodules (>20 mm) or advanced progression (such as, stage III-IV lung cancer), the diagnostic values of 7-AABs, alone or in combination with non-specific tumor markers, were not evaluated in these patients in the present study. Therefore, these aspects require further study.
In conclusion, the findings of the present study demonstrated that 7-AABs + HSP90a had good clinical value for the diagnosis of early-stage lung cancer, including patients with different types of small nodules, which suggested its value for further application in clinical practice.
Not applicable.
The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.
QC, SZ, WH and GF conceived and designed the study. QC, SZ, GG, WZ, ZZ, GF and NJ performed the data analysis and drafted the manuscript. QC, SZ and WH performed data collection. GG, WZ, GF, NJ and WH analyzed the results. WH and GF edited the manuscript and confirm the authenticity of all the raw data. All authors have read and approved the final manuscript.
The present study was approved by the Ethics Review Committee of The First Affiliated Hospital of Wannan Medical College (approval no. 202076; Wuhu, China), and the experimental procedures were in agreement with The Declaration of Helsinki. Written informed consent was obtained from all participants.
Not applicable.
The authors declare that they have no competing interests.
Pathological examination of H&E staining was used to confirm patients with (A) adenocarcinoma
High-resolution computed tomography was used to evaluate patients with (A) pGGO (indicated by the red arrow), (B) mGGO (indicated by the red arrow) or (C) SN (indicated by the red arrow). pGGO, pure ground-glass opacity; mGGO, mix ground-glass opacity; SN, solid nodule.
Levels of each autoantibody and HSP90a in the MLN, BLN and HC groups. *P<0.05, **P<0.01 and ****P<0.0001. After the normality test was performed, Kruskal-Wallis test followed by Dunn's test was used to analyze the data. MLN, malignant lung nodules; BLN, benign lung nodules; HC, healthy control; p53, Tumor protein 53; MAGEA1, melanoma antigen 1; PGP9.5, protein gene product 9.5; SOX2, sex-determining region Y-box 2; GAGE7, cancer/testis G antigen 7; GBU4-5, ATP-dependent RNA helicase 4-5; CAGE, cancer-associated antigen; HSP90a, heat shock protein 90a; ns, not significant.
Diagnostic efficiency of 7-AABs alone or in combination with non-specific tumor markers. (A) Diagnostic efficiency was assessed using the ROC curve for patients with lung nodules. The results of 7-AABs, 7-AABs + CEA, 7-AABs + CYFRA21-1 and 7-AABs + CA199 were very similar to the ROC analysis using the overlapped data. (B) Diagnostic efficiency was assessed using the ROC curve for patients with lung nodules of different sizes. The results of 7-AABs, 7-AABs + CEA, 7-AABs + CYFRA21-1, 7-AABs + CA199 and 7-AABs + CA125 were similar to the ROC analysis using the overlapped data in the <8 mm group. The results of 7-AABs, 7-AABs + CEA, 7-AABs + CYFRA21-1 and 7-AABs + CA199 were very similar to the ROC analysis using the overlapped data in the 8-20 mm group. (C) Diagnostic efficiency was assessed using the ROC curve for patients with lung nodules of different types. The results of 7-AABs, 7-AABs + CEA, 7-AABs + CYFRA21-1, 7-AABs + CA199 and 7-AABs + CA125 were similar to the ROC analysis using the overlapped data in the pGGO and mGGO groups. The results of 7-AABs alone or in combination with non-specific tumor markers were identical to the ROC analysis using the overlapped data in the SN group. 7-AABs, seven lung cancer-related autoantibodies; ROC, receiver operating characteristic; CEA, carcinoembryonic antigen; CYFRA21-1, cytokeratin fragment antigen 21-1; CA199, carbohydrate antigen 199; CA125, carbohydrate antigen 125; pGGO, pure ground-glass opacity; mGGO, mix ground-glass opacity; SN, solid nodule; HSP90a, heat shock protein 90a.
Clinical characteristics of patients with lung nodules and HCs.
Clinical characteristic | Lung nodule | MLN | BLN | HC |
---|---|---|---|---|
Number of patients | 217 | 159 | 58 | 30 |
Age, years | 54.5±11.3 | 54.5±11.7 | 54.5±10.1 | 53.2±9.1 |
Sex (M/F), n | 78/139 | 50/109 | 28/30 | 11/19 |
NSCLC, n | 158 | 158 | - | - |
Adenocarcinoma |
66 (41.8) | 66 (41.8) | - | - |
Invasive adenocarcinoma, n (%) | 91 (57.6) | 91 (57.6) | - | - |
Squamous cell carcinoma, n (%) | 1 (0.6) | 1 (0.6) | - | - |
SCLC, n | 1 | 1 | - | - |
Stage, n (%) | ||||
0 | 66 (41.5) | 66 (41.5) | - | - |
I | 91 (57.2) | 91 (57.2) | - | - |
II | 1 (0.6) | 1 (0.6) | - | - |
III | 1 (0.6) | 1 (0.6) | - | - |
Size, n (%) | 217 | 159 | 58 | |
<8 mm | 68 (31.3) | 42 (26.4) | 26 (44.8) | - |
8-20 mm | 149 (68.7) | 117 (73.6) | 32 (55.2) | - |
Type, n (%) | 217 | 159 | 58 | |
Pure GGO | 58 (26.7) | 45 (28.3) | 13 (22.4) | - |
Mix GGO | 97 (44.7) | 85 (53.5) | 12 (20.7) | - |
Solid nodule | 62 (28.6) | 29 (18.2) | 33 (56.9) | - |
HC, healthy control; MLN, malignant lung nodule; BLN, benign lung nodule; NSCLC, non-small cell lung cancer; SCLC, small cell lung cancer; M, male; F, female; GGO, ground-glass opacity.
Diagnostic values of 7-AABs alone or in combination with non-specific tumor markers for patients with lung nodules.
Diagnostic value | 7-AABs | 7-AABs + HSP90a | 7-AABs + CEA | 7-AABs + CYFRA21-1 | 7-AABs + CA199 | 7-AABs + CA125 |
---|---|---|---|---|---|---|
Specificity, % | 77.6 | 67.2 | 70.7 | 72.4 | 75.9 | 77.6 |
Sensitivity, % | 42.8 | 87.4 |
45.9 | 49.1 | 44.0 | 47.8 |
PPV, % | 84.0 | 88.0 | 81.1 | 83.0 | 83.3 | 85.4 |
NPV, % | 33.1 | 66.1 |
32.3 | 34.1 | 33.1 | 35.2 |
RR (95% CI) | 1.3 (1.1-1.5) | 2.6 (1.8-3.7) | 1.2 (1.0-1.4) | 1.3 (1.1-1.5) | 1.2 (1.1-1.5) | 1.3 (1.1-1.5) |
aP<0.0001 compared with other groups, namely 7-AABs, 7-AABs + CEA, 7-AABs + CYFRA21-1, 7-AABs + CA199 or 7-AABs + CA125, respectively. McNemar's test was used for the comparison of the sensitivity between 7-AABs + HSP90a and other groups. The marginal regression model based on the generalized estimation equation was used for the comparison of the NPV between 7-AABs + HSP90a and other groups. 7-AABs, seven lung cancer-related autoantibodies; CEA, carcinoembryonic antigen; CYFRA21-1, cytokeratin fragment antigen 21-1; CA199, carbohydrate antigen 199; CA125, carbohydrate antigen 125; HSP90a, heat shock protein 90a; PPV, positive predictive value; NPV, negative predictive value; RR, relative risk; CI, confidence interval.
Diagnostic values of 7-AABs alone or in combination with non-specific tumor markers for patients with lung nodules of different sizes.
A, tumor size <8 mm | ||||||
---|---|---|---|---|---|---|
Diagnostic value | 7-AABs | 7-AABs + HSP90a | 7-AABs + CEA | 7-AABs + CYFRA21-1 | 7-AABs + CA199 | 7-AABs + CA125 |
Specificity, % | 73.1 | 61.5 | 73.1 | 69.2 | 73.1 | 73.1 |
Sensitivity, % | 47.6 | 90.5 |
50.0 | 52.4 | 47.6 | 47.6 |
PPV, % | 74.1 | 79.2 | 75.0 | 73.3 | 74.1 | 74.1 |
NPV, % | 46.3 | 80.0 |
47.5 | 47.4 | 46.3 | 46.3 |
RR (95% CI) | 1.4 (1.0-1.4) | 4.0 (1.6-9.6) | 1.4 (1.0-2.1) | 1.4 (1.0-2.0) | 1.4 (1.0-2.0) | 1.4 (1.0-2.0) |
B, tumor size 8-20 mm | ||||||
Diagnostic value | 7-AABs | 7-AABs + HSP90a | 7-AABs + CEA | 7-AABs + CYFRA21-1 | 7-AABs + CA199 | 7-AABs + CA125 |
Specificity, % | 81.3 | 71.9 | 68.8 | 75.0 | 78.1 | 81.3 |
Sensitivity, % | 41.0 | 86.3 |
44.4 | 47.9 | 42.7 | 47.9 |
PPV, % | 88.9 | 91.8 | 83.9 | 87.5 | 87.7 | 90.3 |
NPV, % | 27.4 | 60.0 |
25.3 | 28.2 | 27.2 | 29.9 |
RR (95% CI) | 1.2 (1.0-1.4) | 2.2 (1.5-3.3) | 1.1 (1.0-1.3) | 1.2(1.0-1.4) | 1.2 (1.0-1.4) | 1.3 (1.1-1.5) |
aP<0.0001 compared with other groups, namely 7-AABs, 7-AABs + CEA, 7-AABs + CYFRA21-1, 7-AABs + CA199 or 7-AABs + CA125, respectively. McNemar's test was used for the comparison of the sensitivity between 7-AABs + HSP90a and other groups. The marginal regression model based on the generalized estimation equation was used for the comparison of the NPV between 7-AABs + HSP90a and other groups. 7-AABs, seven lung cancer-related autoantibodies; CEA, carcinoembryonic antigen; CYFRA21-1, cytokeratin fragment antigen 21-1; CA199, carbohydrate antigen 199; CA125, carbohydrate antigen 125; HSP90a, heat shock protein 90a; PPV, positive predictive value; NPV, negative predictive value; RR, relative risk.
Diagnostic values 7-AABs alone or in combination with non-specific tumor markers for patients with lung nodules of different types.
A, pure ground-glass opacity type lung nodules | ||||||
---|---|---|---|---|---|---|
Diagnostic value | 7-AABs | 7-AABs + HSP90a | 7-AABs + CEA | 7-AABs + CYFRA21-1 | 7-AABs + CA199 | 7-AABs + CA125 |
Specificity, % | 84.6 | 76.9 | 84.6 | 84.6 | 84.6 | 84.6 |
Sensitivity, % | 51.1 | 91.1 |
51.1 | 57.8 | 51.1 | 53.3 |
PPV, % | 92.0 | 93.2 | 92.0 | 92.9 | 92.0 | 92.3 |
NPV, % | 33.3 | 71.4 |
33.3 | 36.7 | 36.7 | 36.7 |
RR (95% CI) | 1.4 (1.1-1.8) | 3.3 (1.4-7.5) | 1.4 (1.1-1.8) | 1.5 (1.1-2.0) | 1.4 (1.1-1.8) | 1.4 (1.1-1.9) |
B, mix ground-glass opacity type lung nodules | ||||||
Diagnostic value | 7-AABs | 7-AABs + HSP90a | 7-AABs + CEA | 7-AABs + CYFRA21-1 | 7-AABs + CA199 | 7-AABs + CA125 |
Specificity, % | 83.3 | 66.7 | 75.0 | 75.0 | 83.3 | 83.3 |
Sensitivity, % | 37.6 | 84.7 |
41.2 | 42.4 | 38.8 | 43.5 |
PPV, % | 94.1 | 94.7 | 92.1 | 92.3 | 94.3 | 94.9 |
NPV, % | 15.9 | 38.1 |
15.5 | 15.5 | 16.1 | 16.1 |
RR (95% CI) | 1.1 (1.0-1.3) | 1.5 (1.1-2.1) | 1.1 (1.0-1.3) | 1.1 (1.0-1.3) | 1.1 (1.0-1.3) | 1.1 (1.0-1.3) |
C, solid type lung nodules | ||||||
Diagnostic value | 7-AABs | 7-AABs + HSP90a | 7-AABs + CEA | 7-AABs + CYFRA21-1 | 7-AABs + CA199 | 7-AABs + CA125 |
Specificity, % | 72.7 | 63.6 | 63.6 | 66.7 | 69.7 | 72.7 |
Sensitivity, % | 44.8 | 89.7 |
51.7 | 55.2 | 48.3 | 51.7 |
PPV, % | 59.1 | 68.4 | 55.6 | 59.3 | 58.3 | 62.5 |
NPV, % | 60.0 | 87.5 |
60.0 | 62.9 | 60.5 | 63.2 |
RR (95% CI) | 1.5 (0.9-2.5) | 5.5 (1.9-16.1) | 1.4 (0.8-2.4) | 1.6 (0.9-2.7) | 1.5 (0.9-2.5) | 1.7 (1.0-2.9) |
aP<0.0001 and
bP<0.001 compared with other groups, namely 7-AABs, 7-AABs + CEA, 7-AABs + CYFRA21-1, 7-AABs + CA199 or 7-AABs + CA125, respectively. McNemar's test was used for the comparison of the sensitivity between 7-AABs + HSP90a and other groups. The marginal regression model based on the generalized estimation equation was used for the comparison of the NPV between 7-AABs + HSP90a and other groups. 7-AABs, seven lung cancer-related autoantibodies; CEA, carcinoembryonic antigen; CYFRA21-1, cytokeratin fragment antigen 21-1; CA199, carbohydrate antigen 199; CA125, carbohydrate antigen 125; HSP90a, heat shock protein 90a; PPV, positive predictive value; NPV, negative predictive value; RR, relative risk.