Clinical evaluation of seven tumour markers in lung cancer diagnosis: can any combination improve the results?

Our plain-English summary of Clinical evaluation of seven tumour markers in lung cancer diagnosis: can any combination improve the results?, by M. Plebani, D. Basso, F. Navaglia, et al (British Journal of Cancer, 1995).

The study

• Authors: M. Plebani, D. Basso, F. Navaglia, M. De Paoli, A. Tommasini, A. Cipriani
• Journal: British Journal of Cancer
• Published: 1995;72(1):170-3
• Cited by: 71 articles, as counted by NIH iCite
• Read it: PubMed · Publisher · Free full text

Abstract

In this study we compared the diagnostic utility of: (1) neuron-specific enolase (NSE); (2) squamous cell carcinoma antigen (SCC); (3) carcinoembryonic antigen (CEA); and (4) cytokeratin markers (CYFRA 21-1, TPA, TPM, TPS) in patients with small-cell lung cancer (SCLC) (21 cases) and non-small-cell lung cancer (94 cases). For comparison we also studied 66 patients with benign lung diseases and nine with pleural mesothelioma. NSE levels in SCLC patients were significantly higher than those in all the other groups studied. No significant variations were found among the SCC levels in all groups. CEA levels in patients with adenocarcinoma were significantly higher than those in all other groups studied. CYFRA 21-1 serum levels significantly increased in patients with squamous cell carcinoma and mesothelioma, while TPA, TPS and TPM increased in patients with lung cancer irrespective of the histological type. In patients with SCLC, high levels of all markers except SCC were found when the disease was extensive. In patients with non-SCLC, the highest levels of all tumour markers were usually found in those with advanced disease, although CYFRA 21-1 gave a sensitivity of 44% when a specificity of 95% was fixed in stage I non-SCLC patients. An analysis of receiver operating characteristic curves revealed that the highest diagnostic accuracies in distinguishing benign from malignant lung diseases were achieved with TPM (81%), CYFRA 21-1 (72%), CEA (78%) or TPA (78%) when using cut-off values of 46 Ul-1, 3.0 micrograms l-1, 2.0 micrograms l-1 and 75 Ul-1 respectively. When all patients were considered, the combined evaluation of more than one marker did not significantly improve the results obtained with TPM alone. However, taking into consideration the fact that CYFRA 21-1 is the most sensitive index of early lung tumours and that its combined determination with TPM did not worsen the overall sensitivity and specificity of the latter, the combined use of these two markers may be suggested as a useful took for the diagnosis of lung tumours.

In plain English

This study from the University of Padova, Italy, measured seven blood (serum) tumour markers to see how well they told lung cancer apart from benign lung disease. It did not study asbestos exposure. Serum was taken at diagnosis from 190 people: 66 with benign lung diseases, 115 with lung cancer and nine men with pleural mesothelioma. The cytokeratin markers TPA, TPS and TPM were significantly higher in lung cancer than in benign disease. In the nine mesothelioma patients, CYFRA 21-1 was significantly higher than in the benign, squamous and small-cell groups, and average TPA and TPM levels were the highest of any group. Cut-offs were then set so that 95% of benign cases tested negative. At those cut-offs, CYFRA 21-1 detected 52% of cancers and 44% of stage I non-small-cell cancers. A separate analysis used each marker's best-balanced cut-off. In it, TPM had the highest accuracy, 81%, at a lower cut-off of 46 units per litre (79% sensitivity, 84% specificity). Adding CYFRA 21-1 to TPM did not significantly change TPM's results. The authors concluded that TPM and CYFRA 21-1 were the most sensitive and specific markers and recommended using them together.

We wrote this plain-English summary from the full article with the help of AI, and a second AI model checked each sentence against the article. It is not the authors' wording; the abstract and the article are their own account.

What the study did and found

• Design: Clinical diagnostic study comparing serum tumour marker levels, measured once at diagnosis, across patient groups, with analysis of variance and ROC curves
• Who was studied: 190 subjects from the University of Padova, Italy (dates of enrolment not given). 66 had benign lung disease (42 males, 24 females, age 19-70). 115 had histologically confirmed lung cancer (103 males, 12 females, age 45-74): 54 squamous cell carcinoma, 40 adenocarcinoma and 21 small-cell lung cancer. 9 males had pleural mesothelioma (age 50-65).
• Exposure: No asbestos or fiber exposure was studied. The study measured seven serum tumour markers: CYFRA 21-1, TPA, TPS and TPM (cytokeratin markers), NSE, SCC and CEA.
• Main results:
  • Patients with lung cancer of every histological type had significantly higher TPA, TPS and TPM than patients with benign lung disease.
  • In the pleural mesothelioma group (9 patients), CYFRA 21-1 levels were significantly higher than in the benign lung disease, squamous cell carcinoma and small-cell lung cancer groups.
  • Mean TPA in the mesothelioma group was 876 U per litre (standard error 491) and mean TPM was 422 U per litre (standard error 129). Both were marked as p less than 0.001 compared with all other groups. The benign lung disease group had means of 68 and 29 U per litre.
  • NSE levels in patients with small-cell lung cancer were significantly higher than in all other groups.
  • CEA levels were significantly higher in patients with adenocarcinoma than in those with benign lung disease, squamous cell carcinoma and small-cell lung cancer.
  • SCC levels did not differ significantly among the groups.
  • With specificity fixed at 95%, sensitivity for lung cancer was 52% for CYFRA 21-1, 47% for TPM (cut-off 100 U per litre) and 43% for TPA. CEA was 35%, TPS 25%, NSE 23% and SCC 15%.
  • In stage I non-small-cell lung cancer, CYFRA 21-1 was the most sensitive marker, at 44% (with specificity fixed at 95%).
  • In small-cell lung cancer, limited disease (10 patients) was distinguished from benign lung disease in only a very small percentage of cases.
  • In the ROC analysis, which used a different cut-off from the 95% specificity analysis, TPM had the highest diagnostic accuracy. It reached 81% (sensitivity 79%, specificity 84%) at a cut-off of 46 U per litre. TPA and CEA each reached 78%. CYFRA 21-1 reached 72% (sensitivity 64%, specificity 86%).
  • Combining CYFRA 21-1 with TPM (79% sensitivity, 79% specificity) did not significantly change the results of TPM alone. Three-marker combinations raised sensitivity to as much as 94%, but specificity never exceeded 45%.
• The authors' conclusion: The authors concluded that TPM and CYFRA 21-1 were the most sensitive and specific indices of lung cancer and recommended using them together, because CYFRA 21-1 was found in 44% of early tumours while TPM was found mainly in advanced tumours.
• Limitations the authors state:
  • The authors noted that a main drawback of serum tumour markers is that the highest levels are usually found when the disease is at an advanced stage.

Topics

The National Library of Medicine indexes this study under subjects our site covers: Lung cancer.

Medical terms in this study, explained in our Lexicon: Lung Cancer, Squamous Cell Carcinoma, Carcinoembryonic Antigen (CEA), Pleural Mesothelioma, Adenocarcinoma, Mesothelioma.

Related studies

• Evaluation of seven tumour markers in pleural fluid for the diagnosis of malignant effusions — M. Miédougé, P. Rouzaud, et al. (1999)
• Biomarker assessments in asbestos-exposed workers as indicators for selective prevention of mesothelioma or bronchogenic carcinoma: rationale and practical implementations — E. Pluygers, P. Baldewyns, et al. (1991)
• Evaluation of pleural CYFRA 21-1 and carcinoembryonic antigen in the diagnosis of malignant pleural effusions — G. Salama, M. Miédougé, et al. (1998)
• Calretinin, thrombomodulin, CEA, and CD15: a useful combination of immunohistochemical markers for differentiating pleural epithelial mesothelioma from peripheral pulmonary adenocarcinoma — C.E. Comin, L. Novelli, et al. (2001)
• Combination of MiR-103a-3p and mesothelin improves the biomarker performance of malignant mesothelioma diagnosis — D.G. Weber, S. Casjens, et al. (2014)
• Combined circulating epigenetic markers to improve mesothelin performance in the diagnosis of malignant mesothelioma — L. Santarelli, S. Staffolani, et al. (2015)

Browse more research in British Journal of Cancer, or return to medical research.