Currently, no further therapy in addition to surgery is recommended in completely resected NSCLC stage I patients. However, the 5-year survival rate at this stage has been reported to be approximately 60%, i.e. 40% of patients had a lower survival rate. The aim of the study was to identify those patients at increased risk by using the tumor markers CYFRA21-1 and CEA as prognostic factors. One hundred and fifty-three stage I NSCLC patients, who were treated exclusively by surgery between 1996 and 1998, entered this retrospective study. It was shown, by multivariate analysis, that elevated CYFRA 21-1 (>3.3ng/ml) and CEA (>9.8ng/ml) levels were associated with a worse outcome in 21.3% and 13.1% of the patients under study, respectively. The corresponding 3-year survival rates were found to be 60.2% for increased CYFRA 21-1 levels (p=0.029) and approximately 40% for increased CEA levels (p=0.022), compared to a rate of 78.4% and 79.0% in case of normal marker levels, respectively. The relative risk (95% confidence interval) was found to be 2.156 (1.08-4.29) for elevated CYFRA 21-1 and 2.707 (1.15-6.36) for elevated CEA. The detection rate for the identification of patients with worse outcome increased when a combination of both markers was used. Thereby, it was possible to identify 32% of patients where one or both markers were elevated. The 3-year survival rate was 55.7% in this group compared to that of 82.5% in those patients where both markers were in the normal range (p=0.0014). In order to consider the degree of marker elevation that is thought to reflect tumor burden, we introduced a tumor marker index (TMI) corresponding to the geometric mean of normalized CYFRA21-1 and CEA levels (marker value divided by diagnostic cut-off). Thereby, we were able to identify 3 groups of patients at different risk levels: the first group (22.7%) had a 3-year survival rate of 96.7%, the second group (42.6%) had one of 77.2% and the third group (34.7%) had one of only 55.7%. In conclusion, elevated CYFRA 21-1 and CEA levels were able to identify a group of curatively operated NSCLC patients who were at high risk of early death. Those patients may benefit from more aggressive treatment approaches. The group of patients with a 3-year survival rate of 96.7% probably does not need further treatment.

To evaluate the possible role of cysteine proteases and serine proteases, as well as their respective inhibitors and receptors, as new prognostic factors in NSCLC, we examined, for the first time, 10 biological parameters related to three proteolytic systems within a homogeneous collective of 147 cases of NSCLC. Activities (cath B(AT), cath B(A7.5)) and protein levels of cath B(C), cath L(C), uPA, PAI-1, uPAR [measured by three different assays uPAR (ADI), uPAR (HD13), uPAR (IIIF10)] and TF were measured in homogenates of lung tumour tissue and corresponding non-malignant lung parenchyma. Total cath B activity (cath B(AT)) and enzymatic activity of the fraction of cath B, which is stable and active at pH 7.5 (cath B(A7.5)), were determined by a fluorogenic assay using synthetic substrate Z-Arg-Arg-AMC. The concentrations of cath B(C), cath L(C), uPA, PAI-1, uPAR and TF were determined by ELISAs. uPAR was determined using three different ELISA formats. The median levels of cath B(AT) (5.1-fold), cath B(A7.5) (2.5-fold), cath B(C), (8.5-fold), cath L(C) (6.6-fold), uPA (6.5-fold), PAI-1 (4.2-fold), uPAR (ADI) (2.2-fold), uPAR (HD13) (4.0-fold) and uPAR (IIIF10) (2.6-fold) were higher in tumour tissue compared to the lung parenchyma. Cath B(AT), cath B(A7.5) and cath B(C) in primary tumours correlated with lymph node metastases. Regarding histologies, the concentration of PAI-1 seems to be associated with the histological cell types of NSCLC. We found the highest values of PAI-1 in large cell carcinoma > SCC, AC > carcinoid and lowest values in metastases of primary tumours of other organs. Only PAI-1 was significantly increased in poorly-differentiated cells (G3) compared to well- and moderately- differentiated cells (G1/G2). PAI-1 significantly correlated with cath B(AT) and cath B(A7.5) with uPAR (ADI), uPAR (HD13), uPAR (IIIF10) with uPA, and only weakly with TF, but not with cath B(C) and cath L(C). Significant correlations with overall survival in the total population of NSCLC patients were observed in univariate analysis for cath B(AT), cath B(C), PAI-1, uPAR (ADI), uPAR (HD13), and uPAR (IIIF10). Cath L(C) was not significantly associated with poor prognosis. Regarding the histological tumour type, only in patients with squamous cell carcinomas did cath B(A7.5) and PAI-1 remain significant prognostic factors. In multivariate survival analysis only two proteolytic factors, PAI-1 and uPAR (III101F), stayed significant. In conclusion, among 10 biological parameters evaluated within the same cohort of patients, only PAI-1, uPAR (ADI), uPAR (HD13), uPAR (IIIF10), cath B(AT) and cath B(C) are prognostic factors for overall survival of NSCLC patients. Moreover, PAI-1 and uPAR (IIIF10) add independent prognostic information with regard to established clinical and histomorphological factors in NSCLC.

Besides established factors, we assessed the prognostic impact of tumor markers CYFRA 21-1, CEA and NSE on survival probability in a series of 515 NSCLC patients who were treated by surgery. In addition, we studied the prognostic significance of CYFRA 21-1 and CEA with reference to the individual postoperative (p-) stages. It was found that p-stages correlated inversely with survival probability. Complete resection had a favourable impact on prognosis (R-0 versus R-1/R-2: p<0.0001). Among patient characteristics, we found that age (>70 years) was an unfavourable prognostic factor (p=0.0014). Performance status (PS, ECOG) also pointed to an adverse prognosis for PS 1-2 versus PS 0 (p=0.014), whereas gender was not prognostically important. Elevated LDH levels (>240 U/l), although non-specific, had adverse prognostic significance (p=0.0034). Patients with CYFRA 21-1 levels above 3.3 ng/ml had a median survival of 24.2 months compared to 55.0 months for patients with marker levels below that cut-off point (p<0.0001). CEA levels above 9.8 ng/ml were also associated with shorter survival. Median survival was 19.2 months versus 49.9 months in patients with CEA levels below 9.8 ng/ml (p<0.0001). NSE was of no prognostic value. However, increased NSE levels (>14.5 ng/ml) in a subgroup of patients with advanced stages IIIa up to IV were found to be associated with shortened survival (p=0.024). In p- stage 1, 3- year survival was shorter for patients with abnormal CYFRA 21-1 levels compared to patients with normal marker expression (60.2% versus 78.4%, p=0.015). CEA was also of prognostic value-in this stage (p=0.013). In p-stage II, patients with normal CYFRA 21-1 and CEA levels tended to have a better outcome (p=0.064 and 0.09, respectively). Combined use of both markers showed that stage I patients in whom both markers were normal had a 3-year survival rate of 82.5% compared to those where at least one marker was abnormal (3-year survival: 55.7%, p=0.0014). It is concluded that tumor markers CYFRA 21-1 and CEA proved to be important adjuncts to the staging system and may help to better assess prognosis. A subgroup of p-stage I patients at high risk was identified by elevated marker levels. These patients may benefit from adjuvant chemotherapy.

Abstract Secretory type 2 cystatins, like cystatins C, E/M and F, are thought to be involved in many pathobiological processes, including vascular amyloidosis, rheumatoid arthritis, Alzheimers disease, osteoporosis, viral and bacterial infections, inflammatory disorders and tumour invasion and metastasis. In order to define the levels of cystatins C, E/M, and F in pleural effusions and to investigate whether these cystatins correlate with diagnostic parameters of pleural and lung diseases, we determined their concentrations in 160 pleural effusions. The median concentration of cystatin C in pleural effusions was 1437 ug/l (95.8 nM), ranging between 18-3967 ug/l. Cystatin C did neither correlate with malignant nor with benign diseases. The concentration of cystatin E/M was significantly higher in effusions of primary pleural tumours (mesotheliomas) compared to secondary pleural tumours and benign diseases. Furthermore, there was a significant correlation between the concentration of cystatin E/M of mesotheliomas and the pleural fluid tumour cell count and of cystatin C. The median values of cystatin F were significantly increased in parapneumonic/ empyema thoracis pleural effusions and tuberculous pleurisy compared to malignant pleural effusions, respectively. The concentration of cystatin F in benign effusions correlated significantly with diagnostic parameters and inflammation (total protein; lactate dehydrogenase; C-reactive protein). Finally, only in the group of parapneumonic/empyema thoracis was there a significant correlation between cystatin F and the neutrophil count. In conclusion, pleural effusions of different origin contain high levels of cystatin C, perhaps constituting the major part of an inhibitor reservoir. The level of cystatin E/M appears to be significantly associated with primary pleural tumours and cystatin F correlates with inflammatory processes of lung disorders.

The meeting was held in Heidelberg in July, 2001 and provided the basis for a European cooperation in molecular research focused on translational research in lung cancer. During this meeting, molecular and immunological analysis was discussed as the basis for the development of successful future cancer treatment strategies.R. Rosell, M. Monzó and M. Taron reported on new biotechnological and molecular approaches in a European framework for lung cancer management.Novel methods for diagnosis and management of cancer are being developed today and the most rapidly evolving field is oncogenomics. Oncogenic transformation confers resistance to chemotherapy through a variety of mechanisms including suppression of apoptosis, increased drug metabolism and modification of target proteins. Extensive characterization of genetic alterations in lung cancer can provide rapidly assessable repertoires that can be used for clinical stratification in phase II–III studies. For the sake of simplicity, these repertoires can be classified in five categories: allelic imbalance, aberrant promoter methylation, gene RNA overexpression, microtubule alterations and polymorphisms. Interest at the meeting was centred on ERCC1 mRNA levels as a direct mechanism of cisplatin resistance that could be exploited in customised chemotherapy. European collaboration in molecular research will rely on two simple principles: serial measurements of serum/plasma tumour DNA burden at baseline and after three and six cycles of chemotherapy and very importantly, the isolation of RNA, not only from paraffin-embedded tissue, but also from serum and plasma. Some ongoing Spanish studies are looking at serum RNA. We have successfully met the first challenge of sending blood samples for RNA extraction from geographically distant institutions. Now, we are performing RNA assessment of potential key markers, such as ERCC1, SXR (steroid xenobiotic receptor), ribonucleotide reductase M2 subunit and α- and β-tubulin, which can provide some clues to the prediction of chemotherapy response in individual patients.J. Niklinski, L. Chyczewski, J. Laudanski and W. Niklinska presented the state of advancement of the ongoing projects in Bialystok, Poland.They first described the organisation of the multidisciplinary team, comprising surgeons, molecular biologists, pathologists, pneumonologists and oncologists. Dr Niklinski then presented the method of sample collection, basically from patients treated by surgery. The samples consisted of tissue specimens, serum and nucleated cells (lymphocytes) from the blood; the collection is performed in conditions allowing analyses at both DNA and RNA levels. He then presented a short review of their evaluation of critical genes involved in lung cancer; he also provided information about their different possibilities to analyse p53 consisting of immunohistochemistry, ELISA for p53 antibodies, indirect and direct sequencing of the gene and finally a yeast functional assay. Dr Niklinski commented that some approaches have a significant number of limitations; presently it seems that direct sequencing is the most adequate way to study p53, not only the region comprised between exons 5 and 8, but also exons 4 and 11, in which a significant number of mutations can be determined. A promising new approach seems to be the development of a functional yeast assay, since some preliminary reports mention a 70% sensitivity; however, more information and complete follow-up data are needed, especially in relation to prognosis. Dr Niklinski then spoke about the alteration, both genetic and epigenetic, of the p16 gene. Point mutations appear to be rather rare events, but LOH and methylation of the promoter region (determined by MS-PCR) are rather common in established lung cancer and during its premalignant stages. They should be taken into consideration for the multicenter evaluation. He then talked about ras evaluation, emphasising that its clinical significance is not clear today. Sensitive approaches based on the enriched PCR technique are available and by using this sensitive technique, it would be worth a multiple evaluation of the gene. Finally, Dr Niklinski gave a short presentation about molecular evaluations in premalignant conditions, in order to obtain a clear definition of premalignant changes prone to evolve to malignancy. It seems that laser microdissection techniques are the most correct to study molecular alterations in premalignant changes of the lung.H. Lahm and J.R. Fischer presented their program on immunological and molecular markers for the assessment of genetic and epigenetic alterations in lung cancer.Alterations of the DNA by mutations, deletions or epigenetic modification, as well as dysregulation of gene expression determine the biological behavior and the response of neoplastic lung cells significantly. The Immunology-Molecular Biology Laboratory (IML) at the Thoraxklinik Heidelberg gGmbH undertakes efforts to determine these effects both in the periphery and at the tumour site. Plasma/serum DNA serves as a source to detect mutations (K-RAS, β-tubulin, p53) and hypermethylated promoter regions (RARβ, p16INK4a, DAPK, MGMT, FHIT, RASSF1A). In addition, the immunological status of the patients will be assessed by determination of cytokine secretion (IL-2, IL-6, IL-10, TNFα) by peripheral blood lymphocytes via ELISA. Earlier investigations of this laboratory have established the predominant prognostic significance of impaired cytokine secretion for survival of patients with SCLC and NSCLC. Several lines of evidence suggest that this cytokine suppression is caused by immunosuppressive factors secreted by tumour cells. Furthermore, investigation of tumour material will include an analysis of mutations, the status of hypermethylation, a quantitative assessment of gene expression (IL-10, COX-2, ERCC1, TGFβ1, telomerase) by means of a real-time RT-PCR analysis. A careful comparison with the results obtained in the periphery will be performed. All results will be stored in a central database, which will enable a correlation with any relevant parameters of the patients (therapy, survival). A particular interest will be given to the combination of several factors with respect to the patients’ situation. The work is ultimately focussed on obtaining an individual pattern of molecular and immunological markers, which may serve as the basis to propose a customised and improved therapy for each patient.N. van Zandwijk, L. van't Veer, P. Baas and W.J. Mooi proposed a program for analysis of prognostic factors for non-small cell lung cancer by micro-array techniques in a joint project.For more than 30 years, lung cancers for therapeutic purposes have been subdivided into two groups: small cell lung cancer (SCLC) and non-small cell lung cancer (NSCLC). The majority (±80%) of lung cancers are of the NSCLC group, which comprises carcinomas with squamous, adeno- and some without a specific differentiation (large cell carcinomas). Although there is evidence that squamous cell lung cancers have a somewhat different metastatic pattern when compared with those lacking squamous differentiation, the histological subdivisions within the heterogenous group of non-small cell lung cancers bear no major prognostic significance. Taking also into account the inability to predict a successful outcome of surgery—only ≈50% of patients, who currently undergo surgery for NSCLC with curative intent will be alive after 5 years—there is an urgent need for prognostic factors that can reliably predict long term survival and/or the emergence of metastatic disease.

BACKGROUND. Tumor cells require specific proteolytic enzymes for invasion and metastasis, including lysosomal peptidases-cathepsins. Cathepsin B is a lysosomal cysteine peptidase, which appears to play a major role in invasion and metastasis of human turners. In this study, the authors focused on the possible role of cathepsin B in lymphogenic metastasis by investigating the enzyme localization and its activity in lung tumors and corresponding tumor-infiltrated lymph nodes.METHODS. Cathepsin B activity was determined in lung turners, lung parenchyma, and tumor cell-infiltrated and noninfiltrated regional lymph nodes of the same patient. The authors investigated 35 cancer patients suffering from nonsmall cell lung carcinoma. Cathepsin B throughout activity was measured by cleavage of the fluorogenic substrate Z-Arg-Arg-AMC at pH 6.0.RESULTS. The median specific cathepsin B activity was highest in tumors, followed by the infiltrated lymph nodes, noninfiltrated lymph nodes, and lung parenchyma. The authors showed a significant 1.8-fold increase in cathepsin B activity in turner-infiltrated lymph nodes compared with noninfiltrated regional lymph nodes and a 4.5-fold increase in lung tumor tissue compared with lung parenchyma. High cathepsin B activity, both in tumors and tumor cell-infiltrated lymph nodes, indicated poor prognosis for overall survival. Immunohistochemical analysis showed the presence of cathepsin B in histiocytes and tumor cells but not in lymphocytes of lymph node tissue.CONCLUSIONS. The authors' findings on higher cathepsin B levels in tumor cell-infiltrated lymph nodes show that increased level of cathepsin B activity is characteristic of the invasive tumor cell phenotype. This corroborates the hypothesis, that tumor cell associated cathepsin B may play a role in lymphogenic metastasis. The authors' results support the use of lymph node associated cathepsin B as a prognostic factor for survival of patients with lung carcinoma. Cancer 2000;89: 2282-91. (C) 2000 American Cancer Society.

In order to evaluate the role of cysteine peptidase cathepsin H (Cath H) in human lung cancer its protein levels were determined in 148 pairs of lung tumour tissue and adjacent non-tumourous lung parenchyma using the enzyme-linked immunosorbent assay technique. Additionally, Cath H levels were determined in sera of 171 patients with malignant tumours, 34 patients with benign lung diseases and 47 healthy controls. The median level of Cath H in tumour tissue was 0.64 times that in the corresponding lung parenchyma. Relating tumour levels with histological type we found higher Cath H levels in small-cell and adenocarcinomas and lower levels in squamous cell carcinoma, large-cell carcinoma and secondary tumours. A significant difference in Cath H level between lung tumour tissue and non-tumourous lung parenchyma was associated with the group of cigarette smokers (156 vs 263 ng mg–1protein, P< 0.001). For this group of patients Cath H tumour levels correlated with the survival rate, while for the entire patient population this was not the case. Smokers with high tumour levels of Cath H experienced poor survival. Cath H was significantly higher in sera of patients with malignant and benign lung diseases than in control sera (P< 0.001). The increase was significant for all histological types, being the highest in small-cell and squamous cell carcinomas. Our study reveals that in lung tumours there is different behaviour of Cath H compared with other cysteine peptidases, e.g. cathepsin B and cathepsin L. Variations between tissue and serum levels of Cath H indicate either reduced expression or enhanced secretion of this enzyme in lung tumours. © 2000 Cancer Research Campaign

Background: Cystoscopy is still considered to be the ‘Gold Standard’ in the diagnosis or urinary bladder cancer. However, this invasive technique is not free from subjective influences, especially in cases of superficial tumors of low malignancy. As is the case with cytological detection methods, cystoscopy suffers from a diagnostic gap in the early detection of urothelial carcinomas. Molecular genetic methods, on the other hand, are objective with regard to evaluation of early changes before any macroscopic tumor detection. Cytogenetic analyses of transitional cell carcinomas show structural changes of chromosomes 1, 3, 5, and 8 along with numerical aberrations of chromosomes 7, 8, 9, 10, and the Y chromosome. The detection of these changes in in vitro studies proved to be difficult for the need to establish tumor cell lines. Consequentially, fluorescence in situ hybridization (FISH) could be used as an alternative method to detect cytogenetic changes. Patients and Methods: Since 1995 we investigated 95 patients suffering from urinary bladder carcinoma in a prospective study. FISH was carried out on cells from spontaneous urine samples and bladder washes in parallel with conventional cytology. For chromosomes 7, 8, 9, and 12 centromere probes were used. Results: Comparing results on spontaneous urine samples and on bladder washes, the sensitivity of detecting tumor cells obtained by FISH was 27/32% (pTa), 72.6/92.8% (T1G1–2), and 73.3/86.6% (≤ pT1G3 and > pT1), respectively. In cases with papillomas and cystitis as well as in a control group, FISH yielded no false-positive results. Conclusion: Our data show that FISH is an objective, appropriate technique for diagnosis and follow-up of transitional cell carcinoma and represents a valuable complementation to the conventional cytological investigations.

The introduction of new regimens in the chemotherapy of inoperable non-small cell lung cancer (NSCLC) patients provides a useful extension of survival probability that may now justify the application of tumor markers for the disease monitoring. In a prospective study of 48 consecutive NSCLC patients with TNM stages IIIB/IV we compared changes in the serum levels of the cytokeratin 19 fragment CYFRA 21-1 with the clinical evaluations of response to therapy. CYFRA 21-1 levels were measured using the enzyme immunoassay of Boehringer, Mannheim (Germany). Clinical response to therapy was evaluated according to standard criteria of the WHO. For the assessment of response to therapy by changes in the marker levels the difference between two consecutive levels must exceed 30%. This value is based on the formula: Difference = 2 square root of 2 x CV (CV: inter-assay coefficient of variation of the marker test). CYFRA 21-1 was found to be elevated in 29/48 (60.4%) patients prior to therapy and in 10/48 (20.8%) patients at tumor progression. 91 evaluations have been recorded in these 39 patients. The overall concordance between changes in the marker levels and the clinical assessment was 59.3%. The decrease of CYFRA 21-1 levels at remission was rather low resulting in a concordance of only 42.9%, i.e. marker assays cannot replace the clinical restaging by imaging modalities. In contrast, changes in the marker levels at progression did exceed the required 30% in the majority of cases (64.7%). Most of discordant results (40.7%) could be explained by insufficient decrease or increase of CYFRA 21-1 levels or by extended lead-time. The most striking result was the detection of progressive disease by rising marker levels. Except one case, there was no false-positive elevation of CYFRA 21-1 levels. It is concluded that the detection of progressive disease by rising CYFRA 21-1 levels may avoid continuation of ineffective treatment.

We established a continuous semi-microassay, and for large-scale studies both a stopped and a continuous microtiter plate assay for the fluorometric determination of cathepsin L and cathepsin S activities in body fluids, tissues or cell extracts in the presence of cathepsin B. For the detection of enzymatic activities we used the synthetic substrate Z-Phe-Arg-AMC, and for discrimination between cathepsin L, S and cathepsin B the specific inhibitor CA-074 for blocking interfering cathepsin B activities was applied. Furthermore, we took advantage of the stability of cathepsin S at pH 7.5 for further differentiation between cathepsin L and cathepsin S activities. The kinetic assays were characterized in terms of imprecision, analytical sensitivity, accuracy and substrate concentration. The within-run coefficients of variation were found to be 4.9%-7.2% for the continuous semi-microassay, 10.3%-11.7% for the stopped, and 4.5%-11.8% for the continuous microtiter plate assay. The between-days coefficients of variation for the continuous semi-microassay were 8.1%-8.9%, while for the stopped and continuous microtiter plate assays the coefficients were 11.2%-13.5% and 5.8%-12.2%, respectively. Compared to the continuous semi-microassay, the stopped and the continuous microtiter plate assays showed 3-fold and 11-fold higher sensitivity, respectively. Comparison between the continuous enzyme activity assays at substrate concentrations of 40 microM and 200 microM demonstrated a significant correlation of r = 0.97 and r = 0.99, respectively. The newly developed microtiter plate assay will allow efficient, sensitive and high precision determination of cathepsin L and cathepsin S activities in large-scale studies of cysteine-cathepsin dependent diseases.

In order to determine the value of fluorescence in situ hybridization (FISH) in the diagnosis and follow-up of bladder cancer interphase cytogenetics was performed on cells from urine and bladder washings. 50 ml of urine or bladder washings were collected. FISH was carried out using centromere probes for chromosomes 7, 8, 9 and 12 according to standard protocols. In each case 100 cell nuclei were analysed. Fifty-four samples from urine and 67 samples from bladder washing were analysed by FISH in comparison with results obtained by conventional cytology. Sensitivity of detection of tumor cells by FISH was 68.5% in urine and 63% in bladder washings regardless of tumor stage and grade. Sensitivity obtained by conventional cytology was 50% in urine and 77.3% in bladder washings. FISH on cells from urine samples is an effective complement to the standard urine cytology. Using centromere probes this approach is characterized by high specificity and sensitivity in tumors with T-category higher than pTa and grade higher than G1.

Background: In order to identify the clinical and biochemical important prognostic factors in patients with inoperable non-small-cell lung cancer (NSCLC) receiving chemotherapy or radiochemotherapy, univariate and multivariate regression analyses were performed. Furthermore, we formulated a score for survival that allowed to divide patients into three prognostic groups Material and Methods: 229 previously untreated patients who had received chemotherapy or combined radiochemotherapy at the Thorax-klinik Heidelberg-Rohrbach between January 1987 and December 1990 were eligible for the analysis. We prospectively evaluated data on 30 pretreatment factors (7 clinical and 23 laboratory variables) and analyzed them by univariate and multivariate methods Results: The estimated overall median survival time was 161 days. The univariate analyses revealed that 20 parameters were statistically significant for survival. Cox's multivariate regression analysis indicated that female sex (p = 0.012), a sodium level > 137.5 mmol/l (p < 0.001) and a low stage of disease (TNM stages I, II and IIIA) (p < 0.001) contributed independently to survival. Based on these three independent predictive factors, a prognostic factor risk index was created for each patient to divide the patients into three prognostic groups (good, average, poor). The median survival time within these groups was 409 days vs. 160 days vs. 91 days (p < 0.001). Conclusions: Our findings confirm the important role of well-known independent prognostic factors for survival in NSCLC patients, such as TNM and sex. For sodium there are no comparable analyses at present, and therefore further investigation of this factor is needed. The proposed prognostic index appears to be a feasible tool to develop new treatment strategies in NSCLC.