Lung cancer is the leading cause of cancer death and exhibits high incidence rates among men and women worldwide. It is estimated that 1.8 million new lung cancer cases occurred in 2012, accounting for approximately 13% of total cancer diagnoses. Lung cancers are classified into 2 main types: small cell lung cancer (SCLC) and non-small cell lung cancer (NSCLC). The latter represents ∼85% of patients and contains 3 major histological subtypes, including adenocarcinoma, squamous cell carcinoma, and large cell carcinoma, in which surgery is the only treatment to achieve a possible cure.[2–4] SCLC contributes to the remaining ∼15% of lung cancer cases. It differs from NSCLC on aspects of neuroendocrine differentiation and rapid growth rate and early metastasis in lymph nodes or distant organs.[5,6] The majority of SCLC patients cannot be detected at early stage and lost the opportunity of surgical therapy. However, SCLC is highly sensitive to radiotherapy and chemotherapy, early diagnosis followed by appropriate combined treatment may produce a favorable outcome. As previously reported, around 20% of SCLC patients can achieve long-term survival with aggressive therapy when diagnosed with limited-stage disease, versus just 5% when diagnosed at advanced stage.[8,9] Thus, it is of great importance in the differential diagnosis of lung cancer histological subtype at initial presentation to choose appropriate therapeutic intervention, which can help seize the chance to improve survival rates.
Serum analysis of tumor markers, which is considered as an effective and non-invasive detection, has crucial clinical significances in the histologic differentiation of lung cancer.[10,11] Currently, neuron-specific enolase (NSE) is the most frequently used marker of SCLC in clinical laboratory in China.[12,13] However, it is not ideal because of low sensitivity and determination easily affected by hemolysis in samples. Carcinoembryonic antigen (CEA) has been extensively studied in NSCLC but it lacks lung cancer specificity with abnormal levels being found in other malignancies. Progastrin-releasing peptide (ProGRP) is a promising tumor marker for small cell neuroendocrine carcinoma.[11,16–18] It is the precursor of the neuropeptide gastrin-releasing peptide (GRP) whose instability in the serum precluded its use in a conventional setting. In China, ProGRP is routinely measured by enzyme-linked immunosorbent assay (ELISA) in the majority of clinical laboratories. The Elecsys ProGRP assay (Roche Diagnostics GmbH, Penzberg, Germany) is a new inspection method called ECLIA designed to quantitatively detect ProGRP levels in both human serum and plasma, which was adopted in the present study of serum tumor markers.
The aims of this research were (i) to evaluate the role of ProGRP in this new detection system for the differential diagnosis of lung cancer, specially SCLC; (ii) to compare ProGRP with the other 2 biomarkers (CEA and NSE) used in the patients with benign pulmonary diseases and lung cancer, to assess their diagnostic performances individually or in combination, and analyze their utility in the histological diagnosis of lung cancer (small cell vs squamous vs adenocarcinoma).
Participants and methods
A total of 183 consecutive patients diagnosed with lung cancer under the care of Sun Yat-sen Memorial Hospital and Sun Yat-sen University Cancer Center (Guangzhou, Guangdong Province, China) were enrolled in the present study, including 66 patients with SCLC, 73 with adenocarcinoma, and 44 with squamous cell carcinoma. All patients were investigated initially by physical examination, chest x-ray, fiberoptic bronchoscopy, computed tomography (CT) scan of chest and brain, CT and ultrasonography of the upper abdomen, and bone scan. Lung cancer histologic types were classified according to the 1999 World Health Organization recommendations. All patients were staged according to international guideline TNM classification.
In addition, 45 patients with non-malignant pulmonary diseases were recruited, including pneumonia (n = 25), chronic obstructive pulmonary disease (n = 11), bronchial asthma (n = 3), pulmonary emphysema (n = 2), a solitary benign nodule (n = 2), and pulmonary atelectasis (n = 2). Furthermore, 50 healthy controls were chosen randomly to donate blood for research if they had no lung diseases or other health problems in the past. The experimental protocol was approved by the ethics committee of Sun Yat-Sen Memorial Hospital and Sun Yat-Sen University Cancer Center in China. All participants provided informed consent.
Serum tumor marker assays
Serum samples were collected before the start of treatment and stored at −80°C until analysis was performed. The serum levels of ProGRP, NSE, and CEA were measured with the Modular Analytics E170 (Elecsys; Roche Diagnostics) using the ECLIA technique. The reference value of each parameter was NSE ≤ 16.3 ng/mL, CEA ≤ 5 ng/mL, and ProGRP ≤ 66 pg/mL. As for SCLC, the serum levels of NSE and ProGRP above the threshold levels were considered to be positive, whereas CEA below threshold level was regarded as positive. There were 4 combined detection modes (CDMs) studied in this research, including CDM1 (ProGRP + NSE), CDM2 (ProGRP + CEA), CDM3 (NSE + CEA), and CDM4 (ProGRP + NSE + CEA). The test was defined as positive if one of the markers was positive in combined detection, and negative if all of the markers were negative.
Due to the asymmetric distribution of the biomarker values, the data were presented as medians and interquartile range. Associations between categorical variables were analyzed by Fisher's exact test or chi-square test. Non-parametric tests were used for comparisons between numeric variables. The Mann–Whitney U test and Kruskal–Wallis test were performed to compare biomarker distributions across 2 and >2 subgroups of patients. Sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), positive likelihood ratio (+LR), and negative likelihood ratio (−LR) were calculated using standard formulas.
Receiver operating characteristic (ROC) curves were constructed to compare the predictive ability of each biochemical marker and their CDMs to discriminate among the patients with lung cancers of different histological groups, benign pulmonary diseases and healthy individuals. The corresponding area under the curve (AUC) value and 95% confidence interval (CI) were calculated. P < .05 was considered statistically significant. All statistical analyses were performed using SPSS 17.0 (SPSS, Chicago, IL) and GraphPad Prism 7.0 (GraphPad Inc., La Jolla, CA).
Clinical characteristics of all participants
The clinical characteristics of all patients and healthy controls were listed in Table 1. There were no significant differences in sex or age among these 5 groups (healthy vs benign vs small cell vs adenocarcinoma vs squamous). All patients with malignant lung tumor were staged with the TNM classification system.
Differential diagnostic values of ProGRP, NSE, and CEA serum levels in SCLC
Compared with the patients with benign lung diseases and NSCLC, the serum level of NSE was significantly higher in the patients with SCLC (Fig. 1). Similarly, ProGRP discriminated clearly among the groups concerning SCLC versus adenocarcinoma, SCLC versus squamous cell carcinoma, as well as SCLC versus benign lung diseases. Notably, extremely the high ProGRP level was found in patients with SCLC. Whereas CEA showed the higher level in the patients with adenocarcinoma compared with the other groups.
When analyzing the diagnostic performance of the tumor markers in patients with SCLC versus those with benign pulmonary diseases (Table 2 and Fig. 2A), the sensitivity, specificity, PPV, NPV, and +LR of ProGRP were 71.2%, 91.1%, 92.1%, 68.3%, and 8.0, respectively, which were obviously higher compared with those of NSE. ROC analyses showed that the highest AUC of each individual biomarker was found for ProGRP (0.815, 95% CI: 0.737–0.893), followed by NSE (0.713, 95% CI: 0.617–0.809). Combinations of NSE with ProGRP (CDM1) increased sensitivity and NPV, further improved the diagnostic power of NSE alone and yielded an AUC of 0.849 (95% CI: 0.779–0.919).
In addition, the sensitivity, specificity, PPV, NPV, +LR, and −LR of ProGRP for distinguishing SCLC from squamous cell carcinoma were 71.2%, 93.2%, 94.0%, 68.9%, 10.5, and 0.3, respectively (Table 2). All of these evaluation indexes were better than those of NSE. Furthermore, ProGRP also indicated better diagnostic utility in terms of higher AUC than NSE and CEA (Fig. 2B, 0.859 vs 0.669 vs 0.534, P < .05). Besides, combination of ProGRP with NSE (CDM1) also gave a comparatively ideal predictive value in the discrimination between SCLC and squamous cell carcinoma and yielded an AUC of 0.894 (95% CI: 0.833–0.955), but it did not significantly differ from ProGRP (P = .15).
In discriminating SCLC from adenocarcinoma, serum ProGRP offered a sensitivity, specificity, PPV, NPV, +LR, and −LR of 71.2%, 93.1%, 90.4%, 78.2%, 10.5, and 0.3, respectively (Table 2). Interestingly, CEA exhibited a higher sensitivity of 74.2%. Of note, the sensitivity and NPV increased markedly reaching 100% when these 2 biomarkers were assessed together. ROC analysis (Fig. 2C) showed that ProGRP was most efficiently used to identify the SCLC patients with an AUC of 0.835 (95% CI: 0.760–0.910), followed by CEA (AUC 0.693, 95% CI: 0.606–0.780) and NSE (AUC 0.670, 95% CI: 0.576–0.765). The combination of ProGRP and CEA (CDM2) further increased the diagnostic capacity indicated by a high AUC of 0.928 (95% CI: 0.886–0.0.970), which was close to the AUC concerning the combinations of ProGRP with NSE and CEA (CDM4 0.929, 95% CI: 0.888–0.970, P = .75). To further analyze the diagnostic capacity of ProGRP and CEA in SCLC versus adenocarcinoma, we divided the patients with SCLC and adenocarcinoma into 4 groups according to serum levels of these 2 markers (Table 3). Intriguingly, none of the patients with SCLC were in group B (ProGRP ≤ 66 pg/mL, CEA > 5 ng/mL), that is, in this group 100% of the patients were diagnosed with adenocarcinoma. Besides, the patients with SCLC accounted for 96.8% and 85% in groups C (ProGRP > 66 pg/mL, CEA ≤ 5 ng/mL) and D (ProGRP > 66 pg/mL, CEA > 5 ng/mL), respectively.
In this study, serum ProGRP was analyzed by ECLIA technique and we evaluated the role of ProGRP in this new detection system for the differential diagnosis of benign pulmonary diseases and lung cancer histological subtypes (especially SCLC vs squamous vs adenocarcinoma), as well as its use in combination with CEA and NSE.
Our study revealed that ProGRP serum level was significantly higher in SCLC than in benign pulmonary diseases and NSCLC (adenocarcinoma and squamous). In the ECLIA detection system, ProGRP showed the sensitivity of 71.2% in the patients with SCLC, which was similar to those determinated by ELISA in other groups (65%–80%).[19,23–26] Likewise, Tang et al assessed the diagnostic value of serum ProGRP detected by ELISA in the diagnosis of SCLC using meta-analysis and proved that ProGRP had the pooled sensitivity of 72% (95% CI: 70%–75%) and pooled specificity of 93% (95% CI: 92%–94%), which was comparable to the values of ProGRP analyzed by ECLIA in our study. With the combination of ProGRP and NSE, the diagnostic sensitivity of SCLC further increased, reaching 81.8%.
Comparing with NSE and CEA, ProGRP exhibited the strongest sensitivity–specificity relationship in discriminating SCLC from benign lung diseases. Its sensitivity, specificity, PPV, and +LR were up to 71.2%, 91.1%, 92.1%, and 8.0, respectively. Whereas the corresponding indexes of NSE were 56.1%, 73.3%, 75.5%, and 2.1, respectively. However, there was no significant difference in AUC–ROC between ProGRP and NSE (0.815 vs 0.713). Moreover, both of them were found to be more superior to CEA for detecting SCLC and benign lung diseases. Notablely, the combination of ProGRP and NSE could significantly enhance the diagnostic performance of NSE alone, while this improvement over ProGRP alone was not obvious. Shibayama et al found that NSE showed a higher specificity and PPV than ProGRP in the discrimination of SCLC and benign lung diseases (specificity: 100% vs 93.2%; PPV: 100% vs 91.4%), but our findings reflected a discrepancy: ProGRP possessed a better specificity and PPV. One of the probable reasons was that in our study serum samples were not separated from the clot within 1 hour after sampling strictly according to formulary requirement and might lead to a false increase of NSE.[14,16] Since NSE may leak from platelets and erythrocytes, collecting a suitable sample seems not easy. Alternatively, ProGRP might be a promising tumor marker.
The histological diagnosis of lung cancer is of great importance for the initiation of treatment and prognostic implications. However, the diagnosis is not always easy, especially for the patients with advanced tumors. Thus, how to improve the histological diagnosis using serum tumor markers is still a challenging issue worthy of special consideration. Our study paid more attention to the roles of serum ProGRP, NSE, CEA, and their combinations in the discrimination between SCLC and NSCLC (especially squamous cell carcinoma and adenocarcinoma).
Against the squamous cell carcinoma of lung cancer, the diagnostic sensitivity of serum ProGRP and NSE for SCLC were, respectively, estimated to be 93.0% and 40.9% at a set specificity of 75.0%. Our results demonstrated that the tumor markers with the highest discriminatory capacity between SCLC and squamous cell carcinoma were ProGRP, followed by NSE (AUC–ROC 0.859 vs 0.669, P = .008), which further confirmed the superiority of ProGRP to NSE for diagnosis of SCLC and was in line with earlier reports on ProGRP analyzed by ELISA.[25,26,28–30] However, the AUC–ROC of ProGRP was not significantly increased by the addition of NSE, no significant differences were found in the comparison. In view of the cost savings for patients, ProGRP still might be an optimal choice for discriminating SCLC from squamous cell carcinoma, and NSE might add additional value to ProGRP.
In the discrimination of SCLC and adenocarcinoma, serum ProGRP showed a much higher sensitivity compared with NSE at a specificity of 76.0% (93.2% vs 41.1%). Serum level of CEA was significantly lower in the SCLC patients compared with the adenocarcinoma cases. Our present results further validated previous findings.[18,23,26] Moreover, the combination of ProGRP and CEA improved the diagnostic performance in comparison with ProGRP alone (AUC–ROC 0.928 vs 0.835, P = .04). In the present study, we found an interesting phenomenon: 100% of the patients were diagnosed with adenocarcinoma in the group of ProGRP ≤ 66 pg/mL and CEA > 5 ng/mL when analyzing the cases with SCLC and adenocarcinoma. That is to say, when ProGRP was less than 66 pg/mL, CEA was of significant value in discriminating SCLC from adenocarcinoma. If CEA was less than 5 ng/mL, the patient was considered SCLC with a great possibility. On the contrary, the patient was more likely to be identified as adenocarcinoma. In addition, the combined mode of ProGRP + NSE + CEA provided almost equivalent AUC compared with ProGRP + CEA (0.929 vs 0.928), which suggested that the former mode was not appropriate as a routine clinical protocol and might lead to unnecessary expenditure for patients.
In summary, our study provided valid information about the 3 biomarkers with respect to their diagnostic values for discriminating SCLC from benign pulmonary diseases and NSCLC (adenocarcinoma and squamous cell carcinoma). Among these studied markers, ProGRP was the most efficient in identifying SCLC. Moreover, ProGRP and NSE were demonstrated to have approximately equivalent diagnostic performance in discriminating SCLC from benign lung diseases. For the patients with squamous cell carcinoma, we recommended the use of ProGRP; however, for those with adenocarcinoma, ProGRP, and CEA were preferred.
We thank all the participants who consented to participate in this study. Especially, we sincerely thank Xiaohui Li who helped us collect the samples.
LSH and HYY wrote the paper. LSH, LQZS, YX and DHC carried out the experiments and interpreted the data. XHL and XLC collected the samples. CHD and XHL participated in study design and supervised the study. All authors approved the final version of the paper.
This work was supported by the Guangdong Natural Science Foundation (grant number S2013010014007, 2014A030313070) and Guangdong Province Science & Technology Project Plan & Social Development Foundation (grant number 2010A030400006). The funders did not participate in data collection and analysis, article writing or submission.
Institutional review board statement
The study was developed under the principles of Declaration of Helsinki and approved by the Ethics Committee of Sun Yat-sen Memorial Hospital and Sun Yat-sen University Cancer Center.
Conflicts of interest
The authors declare that they have no conflicts of interest.
1. Torre LA, Bray F, Siegel RL, et al. Global cancer statistics, 2012. CA Cancer J Clin
2. Pikor LA, Ramnarine VR, Lam S, et al. Genetic alterations defining NSCLC subtypes and their therapeutic implications. Lung Cancer
3. Schneider BJ. Non-small cell lung cancer
staging: proposed revisions to the TNM system. Cancer Imaging
4. Gotfrit J, Zhang T, Zanon-Heacock S, et al. Patients with advanced non-small cell lung cancer
requiring inpatient medical oncology consultation: characteristics, referral patterns, and outcomes. Clin Lung Cancer
5. Jett JR, Schild SE, Kesler KA, et al. Treatment of small cell lung cancer
6. Semenova EA, Nagel R, Berns A. Origins, genetic landscape, and emerging therapies of small cell lung cancer
. Genes Dev
7. Lampaki S, Zarogoulidis P, Lagoudi K, et al. Small cell lung cancer
: current and future strategies. Oncomedicine
8. Chan BA, Coward JI. Chemotherapy advances in small-cell lung cancer
. J Thorac Dis
2013; 5 (suppl 5):S565–S578.
9. Kalemkerian GP. Chemotherapy for small-cell lung cancer
. Lancet Oncol
10. Lou E, Johnson M, Sima C, et al. Serum biomarkers for assessing histology and outcomes in patients with metastatic lung cancer
. Cancer Biomark
11. Holdenrieder S, Pawel von J, Dankelmann E, et al. Nucleosomes, ProGRP, NSE, CYFRA 21-1, and CEA in monitoring first-line chemotherapy of small cell lung cancer
. Clin Cancer Res
12. de Herder WW. Biochemistry of neuroendocrine tumours. Best Pract Res Clin Endocrinol Metab
13. Eriksson B, Oberg K, Stridsberg M. Tumor markers in neuroendocrine tumors. Digestion
2000; 62 (suppl 1):33–38.
14. Ramont L, Thoannes H, Volondat A, et al. Effects of hemolysis and storage condition on neuron-specific enolase
(NSE) in cerebrospinal fluid and serum: implications in clinical practice. Clin Chem Lab Med
15. Schneider J, Philipp M, Velcovsky HG, et al. Pro-gastrin-releasing peptide (ProGRP), neuron specific enolase (NSE), carcinoembryonic antigen
(CEA) and cytokeratin 19-fragments (CYFRA 21-1) in patients with lung cancer
in comparison to other lung diseases. Anticancer Res
16. Korse CM, Taal BG, Vincent A, et al. Choice of tumour markers in patients with neuroendocrine tumours is dependent on the histological grade. A marker study of chromogranin A, neuron specific enolase, progastrin-releasing peptide
and cytokeratin fragments. Eur J Cancer
17. Molina R, Filella X, Augé JM. ProGRP: a new biomarker for small cell lung cancer
. Clin Biochem
18. Yang DW, Zhang Y, Hong QY, et al. Role of a serum-based biomarker panel in the early diagnosis of lung cancer
for a cohort of high-risk patients. Cancer
2015; 121 (suppl 17):3113–3121.
19. Lamy PJ, Grenier J, Kramar A, et al. Pro-gastrin-releasing peptide, neuron specific enolase and chromogranin A as serum markers of small cell lung cancer
. Lung Cancer
20. Korse CM, Holdenrieder S, Zhi XY, et al. Multicenter evaluation of a new progastrin-releasing peptide
(ProGRP) immunoassay across Europe and China. Clin Chim Acta
21. Travis WD, Colby TV, Corrin B, et al. Histological Classification of Lung and Pleural Tumours. 1999; Berlin, Heidelberg:Springer, 21–24.
22. Mountain CF. Revisions in the international system for staging lung cancer
23. Molina R, Auge JM, Filella X, et al. Pro-gastrin-releasing peptide (proGRP) in patients with benign and malignant diseases: comparison with CEA, SCC, CYFRA 21-1 and NSE in patients with lung cancer
. Anticancer Res
24. Nisman B, Biran H, Ramu N, et al. The diagnostic and prognostic value of ProGRP in lung cancer
. Anticancer Res
25. Shibayama T, Ueoka H, Nishii K, et al. Complementary roles of pro-gastrin-releasing peptide (ProGRP) and neuron specific enolase (NSE) in diagnosis and prognosis of small-cell lung cancer
(SCLC). Lung Cancer
26. Molina R, Augé JM, Bosch X, et al. Usefulness of serum tumor markers, including progastrin-releasing peptide
, in patients with lung cancer
: correlation with histology. Tumour Biol
27. Tang JH, Zhang XL, Zhang ZH, et al. Diagnostic value of tumor marker pro-gastrin-releasing peptide in patients with small cell lung cancer
: a systematic review. Chin Med J
28. Molina R, Marrades RM, Augé JM, et al. Assessment of a combined panel of six serum tumor markers for lung cancer
. Am J Respir Crit Care Med
29. Huang Z, Xu D, Zhang F, et al. Pro-gastrin-releasing peptide and neuron-specific enolase
: useful predictors of response to chemotherapy and survival in patients with small cell lung cancer
. Clin Transl Oncol
30. Niho S, Nishiwaki Y, Goto K, et al. Significance of serum pro-gastrin-releasing peptide as a predictor of relapse of small cell lung cancer
: comparative evaluation with neuron-specific enolase
and carcinoembryonic antigen
. Lung Cancer