Chest ACCP Member Benefits
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Guest Access | Sign In via User Name/Password
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF) Free
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Article Archive
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via ISI Web of Science (123)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Brundage, M. D.
Right arrow Articles by Mackillop, W. J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Brundage, M. D.
Right arrow Articles by Mackillop, W. J.
(Chest. 2002;122:1037-1057.)
© 2002 American College of Chest Physicians

Prognostic Factors in Non-small Cell Lung Cancer*

A Decade of Progress

Michael D. Brundage, MD, MSc; Diane Davies, MSc and William J. Mackillop, MB, ChB

* From the Department of Oncology, Radiation Oncology Research Unit, Queen’s University, Kingston, ON, Canada.

Correspondence to: Michael D. Brundage, MD, MSc, Radiation Oncology Research Unit, Apps Level 4, Kingston General Hospital, Kingston, ON, Canada K7L 2V7; e-mail: michael.brundage{at}krcc.on.ca


   Abstract
TOP
 Abstract
Introduction
 Materials and Methods
 Results
 Breadth of Prognostic Factors...
 Integration and Clinical...
 References
 
Study objectives: To provide a systematic overview of the literature investigating patient and tumor factors that are predictive of survival for patients with non-small cell lung cancer (NSCLC), and to analyze patterns in the design of these studies in order to highlight problematic aspects of their design and to advocate for appropriate directions of future studies.

Design: A systematic search of the MEDLINE database and a synthesis of the identified literature.

Measurements and results: The database search (January 1990 to July 2001) was carried out combining the MeSH terms prognosis and carcinoma, nonsmall cell lung. Eight hundred eighty-seven articles met the search criteria. These studies identified 169 prognostic factors relating either to the tumor or the host. One hundred seventy-six studies reported multivariate analyses. Concerning 153 studies reporting a multivariate analysis of prognostic factors in patients with early-stage NSCLC, the median number of patients enrolled per study was 120 (range, 31 to 1,281 patients). The median number of factors reported to be significant in univariate analyses was 4 (range, 2 to 14 factors). The median number of factors reported to be significant in multivariate analyses per study was 2 (range, 0 to 6 factors). The median number of studies examining each prognostic factor was 1 (range, 1 to 105 studies). Only 6% of studies addressed clinical outcomes other than patient survival.

Conclusions: While the breadth of prognostic factors studied in the literature is extensive, the scope of factors evaluated in individual studies is inappropriately narrow. Individual studies are typically statistically underpowered and are remarkably heterogeneous with regard to their conclusions. Larger studies with clinically relevant modeling are required to address the usefulness of newly available prognostic factors in defining the management of patients with NSCLC.

Key Words: carcinoma • multivariate analysis • non-small cell lung cancer • prognosis


   Introduction
TOP
 Abstract
 Introduction
Materials and Methods
 Results
 Breadth of Prognostic Factors...
 Integration and Clinical...
 References
 
Non-small cell lung cancer (NSCLC) is best conceptualized as a group of heterogeneous clinical entities that share molecular and cellular origins but that have different clinical behaviors, and hence, different prognoses. Determining the prognosis for an individual patient with NSCLC is difficult, in part because of the marked clinical heterogeneity of patients with the disease. Over 90% of patients present with one or more of a constellation of possible presenting signs or symptoms, and the clinical course of the disease can evolve along a seemingly infinite combination of branching pathways. This variation in clinical presentation and potential progression, are, in turn, due to the multiple potential manifestations of the primary tumor, of involved metastatic sites, and of paraneoplastic syndromes.

Despite the heterogeneity of the clinical manifestations of lung cancer, the prognosis for a population of patients with lung cancer is remarkably predictable. The overall mortality rates for lung cancer in North America over the last 15 years have remained unchanged.1 A population-based study2 of > 12,000 patients with unresected NSCLC who were registered in seven Ontario regional cancer centers (from 1982 to 1991) demonstrated no significant differences in patient survival rates either between centers or over time. The predictability of population survival outcomes, however, is of limited usefulness to clinicians due to the marked heterogeneity of the patients comprising the overall population. Prognostic factors are, thus, used to divide the population into subgroups in order to realize the benefits of prognostic stratification,3 including improved medical decision making,4 improved personal decision making beyond the treatment decision, more appropriate research design and analysis, and more appropriate health policy development.5

A substantial amount of clinical and basic science research has focused on the prognostic factors for patients with lung cancer. Early investigations focused on clinical characteristics of the tumor and of the patient, such as extent of the disease and weight loss, respectively.6 A number of clinical laboratory tests, such as serum lactate dehydrogenase (LDH) level, were subsequently identified as being relevant,7 followed most recently by investigation of a plethora of new factors arising out of an increased understanding of the cellular and molecular biology of lung cancer.6 8 9

The literature has grown rapidly and has identified > 150 prognostic factors pertaining to the tumor, to the patient, or to the environment. Given the extent and heterogeneity of the literature, many review articles addressing prognosis in lung cancer patients have attempted to identify clinically important and/or promising new prognostic factors in patients with lung cancer (for example, Buccheri and Ferrigno6 ). International consensus workshops aimed at achieving similar goals have been conducted.7

The main purpose of this study was to evaluate the last decade of literature pertaining to prognostic factors for patients from an epidemiologic perspective. In doing so, our goals were to provide an overview of the breadth of prognostic factors described to date, to analyze patterns in the design of this literature, to highlight problematic aspects of the studies, and to advocate for appropriate directions of future research.


   Materials and Methods
TOP
 Abstract
 Introduction
 Materials and Methods
Results
 Breadth of Prognostic Factors...
 Integration and Clinical...
 References
 
We performed a systematic overview of the last decade of literature investigating patient and tumor factors that were predictive of patient survival rates for patients with NSCLC. A MEDLINE database search (January 1990 to December 1999) was carried out combining the MeSH terms prognosis and carcinoma, nonsmall cell lung. At the request of an external reviewer, the eligibility period was extended to June 2001. Indexed articles were reviewed and sorted by the targeted patient cohort (ie, clinically staged, surgically resected, locally advanced, or advanced disease). Additional recorded parameters from each study included the publication year, clinical end points of the analysis, prognostic factors that were analyzed, and sample size. For articles reporting one or more multivariate analyses, we recorded the factors found to be significantly predictive of the stated clinical outcome in univariate and in multivariate modeling. Descriptive statistics were to summarize patterns of publication of prognostic factor studies.


   Results
TOP
 Abstract
 Introduction
 Materials and Methods
 Results
Breadth of Prognostic Factors...
 Integration and Clinical...
 References
 
Overall, 887 articles met the search criteria. These studies identified 169 prognostic factors relating either to the tumor or the host. Figure 1 shows the pattern of this indexed research and illustrates that the literature regarding prognostic issues in NSCLC seems to have grown almost exponentially.



View larger version (28K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 1.. Indexed literature on lung cancer prognosis from 1990 to 2000. The left-hand ordinate indicates the number of indexed entries by year, according to study type. The right-hand ordinate indicates the cumulative number of all entries. The index criteria are found in the text.

 
The recognized heterogeneity among lung cancer patients has led to the definition of major clinical subgroups that are seen as being clinically dissimilar from one another, and these groups are generally defined on the basis of tumor stage. NSCLC is generally staged using the TNM staging system,10 which has been consistently shown to be the most powerful prognostic tool for predicting the survival rates of NSCLC patients overall, as illustrated by Table 1 . Given the overarching importance of disease stage in NSCLC, we have stratified our search results according to three common presenting clinical scenarios (ie, advanced disease, locally advanced disease, and potentially resectable disease). Our focus was on factors that were predictive of patient survival, given the magnitude of the literature addressing this end point.


View this table:
[in this window]
[in a new window]

  		Table 1.. The Prognostic Importance of Stage in NSCLC

 

   Breadth of Prognostic Factors Identified in the Literature
TOP
 Abstract
 Introduction
 Materials and Methods
 Results
 Breadth of Prognostic Factors...
Integration and Clinical...
 References
 
Advanced Disease
Tumor, host, and environment factors pertaining to the prognosis of patients with advanced NSCLC are listed in Table 2 . As shown in the table, factors essential to decision making are the extent of disease, weight loss, and performance status, as these are the most predictive indicators of median patient survival time after undergoing systemic chemotherapy. For patients without substantial systemic manifestations of illness, chemotherapy is known to improve median survival time when compared to the best supportive care alone.11


View this table:
[in this window]
[in a new window]

  		Table 2.. Prognostic Factors in Patients With Advanced NSCLC*

 
Additional factors that may be used to refine the accuracy of prognosis also are listed in Table 2 . These include markers relating to the extent of clinically detectable disease, hematologic or biochemical markers associated with disease extent, and the "intangible" factor inherent in the opinion of the physician (Table 2) . In a large study of 2,531 patients who were enrolled in a variety of clinical trials, Albain and colleagues12 identified good performance status, female sex, and age <= 70 years as the most important factors that were predictive of favorable survival rates overall. The study also identified the following three prognostic subgroups for patients receiving cisplatin chemotherapy that had significantly distinct survival expectations: based on performance status, age, and hemoglobin and serum LDH levels. Other studies employing secondary analysis of clinical trial information have reached similar conclusions,13 as have retrospective studies of patients who were not enrolled in clinical trial protocols.14 15

New and promising prognostic factors also are listed in Table 2 . In contrast to the scenario of resected disease, only a few studies have investigated the impact of molecular markers in the setting of advanced disease, and most of these do so only in the context of evaluating all patients from whom tissue is available. Thus, determining the most appropriate way to integrate knowledge regarding these factors with more readily available tumor and patient-based indicators requires further study.

An increasing amount of research has addressed the use of patient-reported parameters, such as quality-of-life scores and/or anxiety and depression measures, as prognostic factors.16 17 18 19 20 These pretreatment assessments may reflect the impact of disease and, hence, may be seen as providing information that is complementary to performance status and weight loss assessments. However, the repeated demonstration that these assessments provide independent prognostic information also may reflect patients’ inherent characteristics or degree of emotional support that may independently predict better disease outcomes, possibly through psychophysiologic mechanisms. Further research is required to clarify the nature of the observed association between favorable scores on these assessments and superior clinical outcomes.

Locally Advanced Disease
The majority of patients with locally advanced disease will have significant symptoms or other general manifestations of illness such as weight loss or poor performance status. Accordingly, most studies of prognostic factors consider locally advanced and metastatic NSCLC under the general rubric of "advanced" disease. The distinction between the two entities, however, is important for the consideration of particular subgroups of patients, given the treatment decision-making implications of refined prognostic information.

Patients without systemic manifestations of illness, who generally are defined as those patients with no substantial weight loss and high performance status, have been shown in a number of clinical trials to have higher survival rates when they receive induction chemotherapy followed by radiotherapy, or concurrent chemoradiotherapy, compared to radiotherapy alone.11 The same subgroup of patients has been shown to experience higher survival rates when treated with continuous hyperfractionated and accelerated radiotherapy compared to conventional fractionation, and when treated with higher doses of conventional radiation compared to lower doses.21 The role of surgery in relation to induction chemotherapy and radiotherapy is currently being investigated, as is the role of combination chemoradiotherapy in more symptomatic patients.1

A second notable subgroup of patients is that having cT3 N0M0 disease located in the superior pulmonary sulcus (ie, Pancoast tumor). A number of reviews highlights the role of prognostic factors such as neurologic involvement and vertebral body involvement in this particular clinical setting.22 23 24 25

Clinically Resectable Disease
Few studies systematically addressed the prognosis of patients with early-stage disease solely on the basis of clinically available information (including clinical stage). Since surgery is considered to be the standard management of patients who are medically fit for thoracotomy,26 clinical decision making in this scenario generally focuses on patient-related factors that estimate the patient’s likelihood of surviving a pulmonary resection (eg, predictors of postoperative pulmonary function), tumor-related factors that estimate the likelihood of complete resection (eg, cN2 or cT4 disease), or prognostic considerations with respect to competing surgical options (eg, wedge resection vs lobectomy).26

One notable exception to the trend away from assessing prognosis based on clinical features alone is the work of Feinstein and Wells,27 who have proposed, developed, and validated a "clinical-severity" staging system. They have pointed out that the anatomic and morphologic descriptors of lung cancer are classifiable according to widely accepted criteria, but the clinical manifestations of disease, while also widely accepted to have important prognostic value, are not systematically classified in a standard taxonomy. Through a process of sequential sequestration of prognostic strata and subsequent consolidation of clinical indicators, five clinical-severity stages have been defined. The stages are based on the presence of, and severity of, clinical manifestations and functional effects of the tumor and/or patient comorbidity. Feinstein and Wells27 have demonstrated that this system is more robust and reliable than indicators such as patients’ performance status or weight loss applied in isolation.

A special case of prognosis in the setting of clinically resectable disease occurs for patients who have disease that is amenable to resection but who are inoperable due to medical reasons. Primary radiotherapy with curative intent is generally recommended for this scenario, and although no modern randomized trials have directly compared surgery to radiotherapy, radiotherapy is generally believed to afford results that are inferior to surgical resection, both in terms of local disease control and overall survival rates.1 Nonetheless, substantial cure rates have been reported in patients receiving primary radiotherapy, and prognostic factors predicting patient survival have been studied retrospectively. Wigren and colleagues,28 for example, have reported on a multivariate model derived from a study of > 500 patients demonstrating that tumor size was the most powerful predictor of survival. Other independent predictors included tumor stage and patient symptoms, performance status, and hemoglobin level.28 Choi and colleagues29 have reported a recent review of additional factors relevant to patients with lung cancer receiving radiation therapy with curative intent.

Surgically Resected NSCLC
Perhaps somewhat surprisingly, the vast majority of studies addressing the prognosis of patients with NSCLC do so in patients who have already undergone surgical resection of the primary disease, who constitute a minority of NSCLC patients,10 rather than in patients who are treatment-naïve. Patients are, thus, selected in these studies as those who are well enough to undergo successful resection, who have disease that is amenable to complete (or, in some cases, attempted) resection, and who have been staged by pathologic TNM criteria.

The rationale for the study of prognostic factors in this scenario is that even with complete resection, recurrence rates are substantial (20 to 85%, depending on tumor stage10 ), thus making the determination of prognosis clinically relevant. Moreover, the strategy allows sufficient access to fresh or archived pathologic specimens to allow multiple biochemical or molecular evaluations of tumor factors. Table 3 lists prognostic factors that have been identified as independently predictive of patient survival rates in cases of resected NSCLC.


View this table:
[in this window]
[in a new window]

  		Table 3.. Prognostic Factors in Patients With Surgically Resected NSCLC*

 
Tumor-Related Factors: As discussed earlier, tumor stage is the most important prognostic factor identified for patients with resected NSCLC. Revisions to the TNM system were made in 1997 in order to provide greater specificity for identifying patient subgroups, recognizing the prognostic importance of pTI (in comparison to pTII), of the amenability of the disease to definitive surgical resection within stage III, and of the presence of intrapulmonary ipsilateral satellite tumor metastases.10

An important circumstance of prognostic importance is that of incomplete resection, either with gross disease remaining or with positive microscopic resection margins, in which postoperative therapy (radiotherapy or chemoradiotherapy) is usually recommended.1 The majority of studies confirm that microscopically involved margins are a strong negative predictor of survival, even when additional therapy is provided, suggesting that the biological characteristics of the tumor that are associated with microscopic residual disease also are associated with the systemic spread of disease.

As shown in Table 3 , additional tumor-related prognostic factors include other elements of the anatomic extent of disease, and conventional histologic parameters, that are clinically useful in estimating prognosis. For example, the primary T classification has been shown to have independent significance within stage I, and tumor size and volume have been shown to provide additional information beyond the T category. With regard to histologic parameters, the prognostic significance of tumor cell type (eg, large cell undifferentiated, adenocarcinoma, or squamous cell) has been studied extensively. Many of these studies have concluded that adenocarcinoma has an independent negative impact on survival prognosis (Table 3) , whereas other studies of comparable design have not shown cell type to have independent prognostic value.30 31 Bronchoalveolar cancer32 and carcinoid tumors33 34 constitute notable exceptions, as each is considered to be a distinct clinical entity with natural histories that differ from the more common NSCLC tumor types.

Many other tumor factors have been shown in at least one study to have independent prognostic significance, but these factors are not typically assessed in routine clinical practice. A detailed discussion of each factor is beyond the scope of this review. Table 3 provides appropriate references, and excellent reviews are available.6 35 36 These factors are briefly summarized as (1) histologic features, (2) clinical chemistry and serum tumor markers, (3) markers of tumor proliferation, (4) markers of cellular adhesion, and (5) other molecular biological markers, including regulators of cellular growth (eg, ras oncogene or protein, retinoblastoma, epidermal growth factor receptor, erb-b2, motility-related protein-1, and hepatocyte growth factor), regulators of the metastatic cascade (eg, tissue polypeptide antigen [TPA], cyclin D-1, and cathepsin), regulators of apoptosis (p53 and bcl-2), and others. The potential clinical application of these factors is discussed below.

Host-Related Factors: Many studies of prognostic factors have addressed patient characteristics as predictors of survival after complete surgical resection. In general, these factors have been found to be less powerful predictors of outcome, particularly in stage I disease, than in the advanced-disease setting. Thus, these factors are not generally considered to be important for clinical decision making in this scenario. One promising molecular factor that is related to the host is the patient’s CYPIA-1 gene polymorphism (CYPIA1) status. This gene is responsible for the metabolic activation of benzopyrene found in cigarette smoke, and high susceptibility to smoking-related lung cancer has been associated with CYP1A-1. In addition, the susceptible genotype has been found to be associated with higher disease recurrence rates and lower survival rates in multivariate analyses.37

Environment-Related Factors: A full description of therapeutic options and their efficacy is beyond our intended scope in this article. Medical decision making must consider at least two conditional prognoses (eg, the prognosis with one treatment and the prognosis with an alternative treatment), hence, the prognostic significance of a therapeutic intervention in the individual case is clearly of paramount importance in oncology. More complicated, however, is the interaction of a specific treatment effect with other known, or potential, prognostic indicators, as will be discussed below.


   Integration and Clinical Application of Prognostic Factors
TOP
 Abstract
 Introduction
 Materials and Methods
 Results
 Breadth of Prognostic Factors...
 Integration and Clinical...
References
 
In the context of a very large amount of literature reporting putative prognostic factors in lung cancer patients, many authors have highlighted methodological issues in the interpretation and integration of existing studies.6 9 In this section, we address the integration and clinical application of research knowledge pertaining to prognostic factors by discussing seven main observations and their implications.

First, existing reports of prognostic factors comprise a body of literature that is markedly heterogeneous for a number of criteria, as has been noted by others.6 The interstudy variations include the following: variation in the study populations; variation in the diagnostic and staging criteria used to identify the study patients; variation in the type of statistical analyses applied in the studies; variation in the prognostic factors included in the analyses; variation in the methods used to define, measure, or classify the factors; variation in the treatments received by the study patients; and variation in the appropriate statistical correction for analysis of post-treatment factors. These variations result in the following potential problems: patient selection bias, leading to spurious results; low statistical power, leading to false-negative results; differences in laboratory techniques used to determine the presence of specific factors, leading to conflicting results between studies; limited opportunity to integrate studies, leading to the failure of meta-analytic approaches; and inability to define the true influence of prognostic factors due to uncontrolled treatment effects or inappropriate patient selection criteria, among others.6

Second, the quality of published reports is variable. While fully described studies can be seen to differ from one another, some studies are described in insufficient detail to determine the study methods and to identify important variations,38 and this missing information further complicates the interpretation of the literature. This problem is germane to studies of molecular markers, in which differences in laboratory techniques have been shown to have a marked impact on the study results,39 and is equally relevant to assessing patient-based factors, in which valid methods for measuring and evaluating psychosocial and behavioral factors and related outcomes are required.

Third, the breadth of prognostic factors studied in the literature is extensive, and growing, whereas the scope of prognostic factors evaluated in individual studies is often inappropriately narrow. Table 4 , 4B , 4C , 4D , 4E summarizes the reports of prognostic factors in patients with resected NSCLC that have been published since 1990, were reported in English, were electronically indexed (as noted earlier), and used at least one multivariate analysis method. We chose this time frame in order to limit the scope of our summary to relatively recent publications, thereby limiting heterogeneity due to variation in methods over time. Table 4 displays the factors examined in each study, but that were not found to be independently significant, as open circles, and it displays those factors statistically significantly associated with survival in the final analytical model as closed circles.


View this table:
[in this window]
[in a new window]

  		Table 4.. Multivariate Studies of Prognostic Factors for Early-Stage NSCLC*

 

View this table:
[in this window]
[in a new window]

  		Table 4B.. Continued

 

View this table:
[in this window]
[in a new window]

  		Table 4C.. Continued

 

View this table:
[in this window]
[in a new window]

  		Table 4D.. Continued

 

View this table:
[in this window]
[in a new window]

  		Table 4E.. Continued

 
Table 4 shows that most individual studies addressed only a few prognostic factors, typically controlling for patient stage and readily available patient and tumor characteristics such as age, sex, cell type, and/or tumor grade. More specifically, the median number of prognostic factors, per study, that were significantly associated with patient survival rate in univariate analyses was four (range, 1 to 12 factors). The median number of factors that were independently predictive of patient survival in multivariate models was two (range, zero to six factors). The median number of studies investigating a specific prognostic factor was 1 (range, 1 to 105 studies). No substantive changes in these patterns are evident over the time period evaluated. For example, comparing the 76 multivariate studies published from 1990 to 1998 to the 77 multivariate studies published from 1999 to 2001, there were no statistically significant differences in the mean number of variables considered per study (7.77 vs 8.07, respectively; t = -0.61; p = 0.54), or in the number of independent significant variables identified (2.41 vs 2.65, respectively; t = -1.4; p = 0.17).

Fourth, with the exception of a few predictive factors, the literature demonstrates conflicting evidence of the prognostic power of each factor. For a small number of factors, studies are largely consistent in their findings. For example, in the great majority of studies, at least one representation of prognostic TNM stage has been demonstrated as being significantly related to prognosis, the exceptions typically being studies that focus on a stage-specific cohort (for example, pathologic stage I patients). In contrast, three factors (ie, tumor cell type, patient sex, and patient age) have consistently been found not to be significantly associated with survival outcomes, the exceptions being the few studies in which these factors were of weak predictive power. Moreover, it is possible that some of the significant associations demonstrated for these factors might represent type I statistical errors, given the number of studies investigating these patient attributes (Table 4) .

For most of the factors studied in more than one multivariate model, however, the literature is remarkably variable in the conclusions reached about the prognostic value of individual factors. Furthermore, for many factors, the strength of the independent association of that factor with survival outcomes is also quite variable. There are many potential reasons for this observed heterogeneity. Some studies are clearly statistically underpowered, given that the median number of patients enrolled per study was only 120 (range, 31 to 1,281 patients). In addition, variation in the case mix of the study populations, variation in the other explanatory variables included in each analysis, and variation in the methods used to define and quantify prognostic factors will be reflected in the heterogeneity of the apparent strength of association between the prognostic factor and the relevant outcome. Interstudy variation in technical factors has been illustrated for some molecular markers,40 as has the need to elucidate interstudy variations more fully.38 These interstudy differences have resulted in some controversy regarding the clinical value of many prognostic factors, and our review illustrates that this controversy exists for many factors. Excluding anatomic stage, and those factors examined in fewer than three studies, the factors found to be significantly associated with survival most often were the following: markers of angiogenesis (13 of 16 studies; 81%), p21 status (4 of 5 studies; 80%), status of the serum assay for detection of the cytokeratin 19 fragment (4 of 5 studies; 80%), status of the argyrophilic nucleolar organizer region (3 of 4 studies; 75%), p185 status (3 of 4 studies; 75%), Ki-67 status (4 of 6 studies; 67%), vascular endothelial growth factor status (7 of 11 studies; 64%), vessel invasion (11 of 21 studies; 52%), and p53 status (16 of 38 studies; 42%). We point out that these proportions are intended to illustrate variation in study findings and should not be taken as evidence for or against the validity of these factors, since other factors (as discussed earlier) may contribute to our observations. An appropriate assessment of each factor would require careful review in its own right (although the heterogeneity that we have observed in study designs clearly would compromise the application of meta-analytic techniques). Some meta-analyses of well-studied factors are now available (eg, for p53 status41 42 and tumor ploidy43 ) and have recognized this limitation of study heterogeneity.42 Given these complexities, a thorough analysis of each factor is beyond our intended scope.

Fifth, few studies attempt to integrate newly described factors with significant factors identified in earlier studies. A related finding is that only a few reports represent prospectively designed efforts to validate the predictive models identified in earlier patient cohorts (for example, those studies validating the TMN staging system44 45 46 47 48 ). A strategy of combining available databases may be useful for integrating the findings of existing individual studies of nonanatomic prognostic factors.

Sixth, the multidimensional nature of prognosis is underinvestigated, the focus of the majority of research being on predictors of survival. Further investigation of factors that are predictive of survival is clearly required, given the limitations in the existing studies that has been outlined thus far. Of concern, however, is that the prognostic factors predicting the many other clinical end points that are of potential interest to patients and their physicians are investigated only to a limited extent, or not at all. Research investigating the information needs of patients with lung cancer has shown that patients are interested in a broad spectrum of end points other than survival, such as the chances of symptom relief, chances of response to therapy, likely impact on quality of life, likely patterns of disease recurrence, and so on.49 Prognosis regarding these outcomes may be pertinent to treatment decision making or may be important to the patient for purposes of personal decision making, for understanding the nature of the illness, or for other reasons.49 Some studies, for example, have examined the impact of lung cancer treatment on patterns of disease recurrence or on symptom relief, and fewer still have examined what factors predict these important clinical end points. Since it has been demonstrated that many cancer patients in general, and many lung cancer patients specifically,49 50 wish to participate in treatment decision making, enhanced understanding of these predictors will be necessary to improve the quality of information that clinicians are able to provide to individual patients.

Seventh, comparatively little research has focused on patients at time points beyond their initial presentation. For example, patients presenting with recurrent metastatic disease (following treatment with curative intent) are not generally distinguished in the literature from those whose initial presentation is with stage IV disease. Although some studies have considered the prognosis of patients with recurrent disease, prognostic factors relevant to the internal frame of reference for a given patient (for example, time since initial diagnosis or extent of initial disease) rarely have been elucidated.

We conclude that more comprehensive research is required to fully realize the benefits of prognosis in lung cancer patients. New studies are needed to better evaluate how factors that are already established to have prognostic value are best combined in clinically useful stratification models, and to validate these models. At the level of the specific patient, more accurate predictors of outcomes that are germane to the individual patient will enhance medical and personal decision making. At the level of patient populations, further research is necessary to establish the risks and benefits of competing treatment approaches as they apply to patient strata that have distinct prognoses. For example, ongoing randomized clinical trials that have been designed to determine the risks and benefits of adjuvant chemotherapy after complete surgical resection are also evaluating whether the k-ras status of the tumor is associated with the efficacy of chemotherapy. At the level of the disease, a better integration of prognostic factors in individual studies may reveal unanticipated associations that, in turn, may lead to new insights into opportunities for therapeutic intervention.


   Acknowledgements
 
The authors gratefully acknowledge the helpful comments of the three anonymous reviewers, and the excellent research assistance of Ms. Leanne Findlay in updating and collating the database and tables.


   Footnotes
 
Abbreviations: CYP1A1 = CYP1A1 gene polymorphism; LDH = lactate dehydrogenase; NSCLC = non-small cell lung cancer; TPA = tissue polypeptide antigen

Supported in part by a grant from the National Cancer Institute of Canada.

Received for publication May 31, 2001. Accepted for publication January 25, 2002.


   References
TOP
 Abstract
 Introduction
 Materials and Methods
 Results
 Breadth of Prognostic Factors...
 Integration and Clinical...
 References
 

  1. Ginsberg, RJ, Vokes, EE, Raben, A (1999) Cancer of the lung. DeVita, VT Hellman, S Rosenberg, SA eds. Cancer: principles and practice of oncology. ,849-950 Lippincott-Raven Philadelphia, PA.
  2. Mackillop, WJ, Zhou, Y, Dixon, P, et al Variation in the management and outcome of non-small cell lung cancer in Ontario. Radiother Oncol 1994;32,106-115[CrossRef][ISI][Medline]
  3. Feinstein, AR Clinical biostatistics XIV: the purposes of prognostic stratification. Clin Pharmacol Ther 1972;13,285-297[ISI][Medline]
  4. Brundage, MD, Feldman-Stewart, D, Cosby, R, et al Phase I study of a decision aid for patients with locally advanced non-small cell lung cancer. J Clin Oncol 2001;19,1326-1335[Abstract/Free Full Text]
  5. Mackillop, WJ The importance of prognosis in cancer medicine. Gospodarowicz, MK Henson, DE Hutter, RBPet al eds. Prognostic factors in cancer. 2001 Wiley & Sons New York, NY.
  6. Buccheri, G, Ferrigno, D Prognostic factors in lung cancer: tables and comments. Eur Respir J 1994;7,1350-1364[Abstract]
  7. Feld, R, Borges, M, Giner, V, et al Prognostic factors in non-small cell lung cancer. Lung Cancer 1994;11(suppl),S19-S23
  8. Smit, EF, Groen, HJ, Splinter, TA, et al New prognostic factors in resectable non-small cell lung cancer. Thorax 1996;51,638-646[ISI][Medline]
  9. Mountain, CF New prognostic factors in lung cancer: biologic prophets of cancer cell aggression. Chest 1995;108,246-254[Medline]
  10. Mountain, CF Revisions in the International System for Staging Lung Cancer. Chest 1997;111,1710-1717[CrossRef][ISI][Medline]
  11. Stewart, LA, Pignon, JP Chemotherapy in non-small cell lung cancer: a meta-analysis using updated data on individual patients from 52 randomised clinical trials. BMJ 1995;311,899-909[Abstract/Free Full Text]
  12. Albain, KS, Crowley, JJ, LeBlanc, M, et al Survival determinants in extensive-stage non-small-cell lung cancer: the Southwest Oncology Group experience. J Clin Oncol 1991;9,1618-1626[Abstract]
  13. Paesmans, M, Sculier, JP, Libert, P, et al Prognostic factors for survival in advanced non-small-cell lung cancer: univariate and multivariate analyses including recursive partitioning and amalgamation algorithms in 1,052 patients: the European Lung Cancer Working Party. J Clin Oncol 1995;13,1221-1230[Abstract]
  14. Hespanhol, V, Queiroga, H, Magalhaes, A, et al Survival predictors in advanced non-small cell lung cancer. Lung Cancer 1995;13,253-267[CrossRef][ISI][Medline]
  15. Takigawa, N, Segawa, Y, Okahara, M, et al Prognostic factors for patients with advanced non-small cell lung cancer: univariate and multivariate analyses including recursive partitioning and amalgamation. Lung Cancer 1996;15,67-77[CrossRef][ISI][Medline]
  16. Ganz, PA, Lee, JJ, Siau, J Quality of life assessment: an independent prognostic variable for survival in lung cancer. Cancer 1991;67,3131-3135[CrossRef][ISI][Medline]
  17. Coates, A, Porzsolt, F, Osoba, D Quality of life in oncology practice: prognostic value of EORTC QLQ-C30 scores in patients with advanced malignancy. Eur J Cancer 1997;33,1025-1030[CrossRef][ISI][Medline]
  18. Herndon, JE, Fleishman, S, Kornblith, AB, et al Is quality of life predictive of the survival of patients with advanced nonsmall cell lung carcinoma? Cancer 1999;85,333-340[CrossRef][ISI][Medline]
  19. Montazeri, A, Milroy, R, Hole, D, et al Quality of life in lung cancer patients: as an important prognostic factor. Lung Cancer 2001;31,233-240[CrossRef][ISI][Medline]
  20. Langendijk, H, Aaronson, NK, de Jong, JM, et al The prognostic impact of quality of life assessed with the EORTC QLQ-C30 in inoperable non-small cell lung carcinoma treated with radiotherapy. Radiother Oncol 2000;55,19-25[CrossRef][ISI][Medline]
  21. Emami, B, Perez, CA Lung. Perez, CA eds. Radiation oncology. 1993 Lippincott Philadelphia, PA.
  22. Anderson, TM, Moy, PM, Holmes, EC Factors affecting survival in superior sulcus tumors. J Clin Oncol 1986;4,1598-1603[Abstract/Free Full Text]
  23. Beyer, DC, Weisenburger, T Superior sulcus tumors. Am J Clin Oncol 1986;9,156-161[ISI][Medline]
  24. Detterbeck, FC Pancoast (superior sulcus) tumors. Ann Thorac Surg 1997;63,1810-1818[Abstract/Free Full Text]
  25. Shaw, RR Pancoast’s tumor. Ann Thorac Surg 1984;37,343-345[Abstract]
  26. Sabiston, DCJ, Spencer, FC Surgery of the chest. 1995 WB Saunders Philadelphia, PA.
  27. Feinstein, AR, Wells, CK A clinical-severity staging system for patients with lung cancer. Medicine (Baltimore) 1990;69,1-33[Medline]
  28. Wigren, T, Oksanen, H, Kellokumpu-Lehtinen, P A practical prognostic index for inoperable non-small-cell lung cancer. J Cancer Res Clin Oncol 1997;123,259-266[ISI][Medline]
  29. Choi, N, Baumann, M, Flentjie, M, et al Predictive factors in radiotherapy for non-small cell lung cancer: present status. Lung Cancer 2001;31,43-56[CrossRef][ISI][Medline]
  30. Mori, M, Kohli, A, Baker, SP, et al Laminin and cathepsin B as prognostic factors in stage I non-small cell lung cancer: are they useful? Mod Pathol 1997;10,572-577[ISI][Medline]
  31. Fontanini, G, Lucchi, M, Vignati, S, et al Angiogenesis as a prognostic indicator of survival in non-small-cell lung carcinoma: a prospective study. J Natl Cancer Inst 1997;89,881-886[Abstract/Free Full Text]
  32. Clayton, F The spectrum and significance of bronchioloalveolar carcinomas. Pathol Annu 1988;23,361-394[Medline]
  33. Perkins, P, Kemp, BL, Putnam, JBJ, et al Pretreatment characteristics of carcinoid tumors of the lung which predict aggressive behavior. Am J Clin Oncol 1997;20,285-288[CrossRef][ISI][Medline]
  34. Kayser, K, Kayser, C, Rahn, W, et al Carcinoid tumors of the lung: immuno- and ligandohistochemistry, analysis of integrated optical density, syntactic structure analysis, clinical data, and prognosis of patients treated surgically. J Surg Oncol 1996;63,99-106[CrossRef][ISI][Medline]
  35. D’Amico, TA, Massey, M, Herndon, JE, et al A biologic risk model for stage I lung cancer: immunohistochemical analysis of 408 patients with the use of ten molecular markers. Thorac Cardiovasc Surg 1999;117,736-743
  36. Johnson, BE Biologic and molecular prognostic factors: impact on treatment of patients with non-small cell lung cancer. Chest 1995;107,287S-290S[Medline]
  37. Goto, I, Yoneda, S, Yamamoto, M, et al Prognostic significance of germ line polymorphisms of the CYP1A1 and glutathione S-transferase genes in patients with non-small cell lung cancer. Cancer Res 1996;56,3725-3730[Abstract/Free Full Text]
  38. Watine, J Further comments on "A practical prognostic index for inoperable non-small-cell lung cancer": a clinical biologist’s point of view. J Cancer Res Clin Oncol 1998;124,581-583[CrossRef][ISI][Medline]
  39. Jacobson, DR ras mutations in lung cancer. Brambilla, C Brambilla, E eds. Lung tumors: fundamental biology and clinical management. 1999 Marcel Dekker New York, NY.
  40. Michalides, RJ Deregulation of the cell cycle in lung cancer. Brambilla, C Brambilla, E eds. Lung tumors: fundamental biology and clinical management. 1999 Marcel Dekker New York, NY.
  41. Huncharek, M, Kupelnick, B, Geschwind, JF, et al Prognostic significance of p53 mutations in non-small cell lung cancer: a meta-analysis of 829 cases from eight published studies. Cancer Lett 2000;153,219-226[CrossRef][ISI][Medline]
  42. Mitsudomi, T, Hamajima, N, Ogawa, M, et al Prognostic significance of p53 alterations in patients with non-small cell lung cancer: a meta-analysis. Clin Cancer Res 2000;6,4055-4063[Abstract/Free Full Text]
  43. Choma, D, Daures, JP, Quantin, X, et al Aneuploidy and prognosis of non-small-cell lung cancer: a meta-analysis of published data. Br J Cancer 2001;85,14-22[CrossRef][ISI][Medline]
  44. van Rens, MT, de la Riviere, AB, Elbers, HR, et al Prognostic assessment of 2,361 patients who underwent pulmonary resection for non-small cell lung cancer, stage I, II, and IIIA. Chest 2000;117,374-379[Medline]
  45. Saito, Y, Yamakawa, Y, Kiriyama, M, et al Evaluation of new TNM lung cancer classification. Jpn Thorac Cardiovasc Surg 2000;48,499-505[Medline]
  46. Bulzebruck, H, Bopp, R, Drings, P, et al New aspects in the staging of lung cancer: prospective validation of the International Union Against Cancer TNM classification. Cancer 1992;70,1102-1110[CrossRef][ISI][Medline]
  47. Buccheri, G, Ferrigno, D Prognostic value of stage grouping and TNM descriptors in lung cancer. Chest 2000;117,1247-1255[Medline]
  48. Jassem, J, Skokowski, J, Dziadziuszko, R, et al Results of surgical treatment of non-small cell lung cancer: validation of the new postoperative pathologic TNM classification. Thorac Cardiovasc Surg 2000;119,1141-1146
  49. Davidson, JR, Brundage, MD, Feldman-Stewart, D Lung cancer treatment decisions: patients’ desires for participation and information. Psychooncology 1999;8,511-520[CrossRef][Medline]
  50. Brundage, MD, Davidson, JR, Mackillop, W, et al Using a treatment-tradeoff methods to elicit preferences for the treatment of locally advanced non-small cell lung cancer. Med Decis Making 1998;18,256-267[Abstract/Free Full Text]
  51. Buccheri, G, Ferrigno, D Prognostic value of the tissue polypeptide antigen in lung cancer. Chest 1992;101,1287-1292[Medline]
  52. Palomares, MR, Sayre, JW, Shekar, KC, et al Gender influence on weight-loss pattern and survival of nonsmall cell lung carcinoma patients. Cancer 1996;78,2119-2126[CrossRef][ISI][Medline]
  53. Thorogood, J, Bulman, AS, Collins, T, et al The use of discriminant analysis to guide palliative treatment for lung cancer patients. Clin Oncol (R Coll Radiol) 1992;4,22-26
  54. Takai, E, Yano, T, Iguchi, H, et al Tumor-induced hypercalcemia and parathyroid hormone-related protein in lung carcinoma. Cancer 1996;78,1384-1387[CrossRef][ISI][Medline]
  55. Buccheri, G, Ferrigno, D, Tamburini, M, Karnofsky,, ECOG. performance status scoring in lung cancer: a prospective, longitudinal study of 536 patients from a single institution. Eur J Cancer 1996;32A,1135-1141[CrossRef][ISI][Medline]
  56. Bonomi, P, Gale, M, Rowland, K, et al Pre-treatment prognostic factors in stage III non-small cell lung cancer patients receiving combined modality treatment. Int J Radiat Oncol Biol Phys 1991;20,247-252[ISI][Medline]
  57. Sugiura, S, Ando, Y, Minami, H, et al Prognostic value of pleural effusion in patients with non-small cell lung cancer. Clin Cancer Res 1997;3,47-50[Abstract]
  58. Komaki, R, Pajak, TF, Byhardt, RW, et al Analysis of early and late deaths on RTOG non-small cell carcinoma of the lung trials: comparison with CALGB 8433. Lung Cancer 1993;10,189-197[CrossRef][ISI][Medline]
  59. Ando, M, Ando, Y, Sugiura, S, et al Prognostic factors for short-term survival in patients with stage IV non-small cell lung cancer. Jpn J Cancer Res 1999;90,249-253[CrossRef][ISI]
  60. Wigren, T Confirmation of a prognostic index for patients with inoperable non-small cell lung cancer. Radiother Oncol 1997;44,9-15[CrossRef][ISI][Medline]
  61. Kreisman, H, Lisbona, A, Olson, L, et al Effect of radiologic stage III substage on nonsurgical therapy of non-small cell lung cancer. Cancer 1993;72,1588-1596[CrossRef][ISI][Medline]
  62. Thomas, M, Rube, C, Semik, M, et al Trimodality therapy in stage III non-small cell lung cancer: prediction of recurrence by assessment of p185neu. Eur Respir J 1999;13,424-429[Abstract]
  63. Naito, T, Satoh, H, Ishikawa, H, et al Pleural effusion as a significant prognostic factor in non-small cell lung cancer. Anticancer Res 1997;17,4743-4746[ISI][Medline]
  64. Hilsenbeck, SG, Raub, WAJ, Sridhar, KS Prognostic factors in lung cancer based on multivariate analysis. Am J Clin Oncol 1993;16,301-309[ISI][Medline]
  65. Buccheri, G, Ferrigno, D, Ginardi, C, et al Haemostatic abnormalities in lung cancer: prognostic implications. Eur J Cancer 1997;33,50-55[CrossRef][ISI][Medline]
  66. Jacoulet, P, Depierre, A, Moro, D, et al Long-term survivors of small-cell lung cancer (SCLC): a French multicenter study; Groupe d’Oncologie de Langue Francaise. Ann Oncol 1997;8,1009-1014[Abstract/Free Full Text]
  67. Taguchi, O, Gabazza, EC, Yasui, H, et al Prognostic significance of plasma D-dimer levels in patients with lung cancer. Thorax 1997;52,563-565[Abstract]
  68. Wojtukiewicz, MZ, Zacharski, LR, Moritz, TE, et al Prognostic significance of blood coagulation tests in carcinoma of the lung and colon. Blood Coagul Fibrinolysis 1992;3,429-437[ISI][Medline]
  69. Pedersen, LM, Milman, N Prevalence and prognostic significance of proteinuria in patients with lung cancer. Acta Oncol 1996;35,691-695[ISI][Medline]
  70. Buccheri, G Depressive reactions to lung cancer are common and often followed by a poor outcome. Eur Respir J 1998;11,173-178[Abstract/Free Full Text]
  71. Miyamoto, H, Harada, M, Isobe, H, et al Prognostic value of nuclear DNA content and expression of the ras oncogene product in lung cancer. Cancer Res 1991;51,6346-6350[ISI][Medline]
  72. Pence, JC, Kerns, BJ, Dodge, RK, et al Prognostic significance of the proliferation index in surgically resected non-small-cell lung cancer. Arch Surg 1993;128,1382-1390[Abstract]
  73. Pifarre, A, Rosell, R, Monzo, M, et al Prognostic value of replication errors on chromosomes 2p and 3p in non-small-cell lung cancer. Br J Cancer 1997;75,184-189[ISI][Medline]
  74. Kawasaki, M, Nakanishi, Y, Kuwano, K, et al The utility of p53 immunostaining of transbronchial biopsy specimens of lung cancer: p53 overexpression predicts poor prognosis and chemoresistance in advanced non-small cell lung cancer. Clin Cancer Res 1997;3,1195-1200[Abstract]
  75. Volm, M, Drings, P, Wodrich, W Prognostic significance of the expression of c-fos, c-jun and c-erbB-1 oncogene products in human squamous cell lung carcinomas. J Cancer Res Clin Oncol 1993;119,507-510[CrossRef][ISI][Medline]
  76. Buccheri, G, Ferrigno, D, Vola, F Carcinoembryonic antigen (CEA), tissue polypeptide antigen (TPA) and other prognostic indicators in squamous cell lung cancer. Lung Cancer 1993;10,21-33[CrossRef][ISI][Medline]
  77. Carles, J, Rosell, R, Ariza, A, et al Neuroendocrine differentiation as a prognostic factor in non-small cell lung cancer. Lung Cancer 1993;10,209-219[CrossRef][ISI][Medline]
  78. Diez, M, Torres, A, Ortega, L, et al Value of serum neuron-specific enolase in nonsmall cell lung cancer. Oncology (Huntingt) 1993;50,127-131[ISI][Medline]
  79. Takekawa, H, Takaoka, K, Tsukamoto, E, et al Thallium-201 single photon emission computed tomography as an indicator of prognosis for patients with lung carcinoma. Cancer 1997;80,198-203[CrossRef][ISI][Medline]
  80. Bulzebruck, H, Drings, P, Kayser, K, et al Classification of lung cancer: first experiences with the new TNM classification (4th edition). Eur Respir J 1991;4,1197-1206[Abstract]
  81. Mountain, CF, Dresler, CM Regional lymph node classification for lung cancer staging. Chest 1997;111,1718-1723[CrossRef][ISI][Medline]
  82. Lima, C, Matsui, K, Kitagawa, M, et al Analysis of prognostic factors in patients with resected peripheral T1 adenocarcinoma of the lung. Pathol Res Pract 1993;189,1149-1153[ISI][Medline]
  83. Shimizu, J, Watanabe, Y, Oda, M, et al Results of surgical treatment of stage I lung cancer. Nippon Geka Gakkai Zasshi 1993;94,505-510[Medline]
  84. Harpole, DHJ, Herndon, JE, Wolfe, WG, et al A prognostic model of recurrence and death in stage I non-small cell lung cancer utilizing presentation, histopathology, and oncoprotein expression. Cancer Res 1995;55,51-56[Abstract/Free Full Text]
  85. Martini, N, Bains, MS, Burt, ME, et al Incidence of local recurrence and second primary tumors in resected stage I lung cancer. Thorac Cardiovasc Surg 1995;109,120-129
  86. Sagawa, M, Sakurada, A, Fujimura, S, et al Five-year survivors with resected pN2 nonsmall cell lung carcinoma. Cancer 1999;85,864-868[CrossRef][ISI][Medline]
  87. Vansteenkiste, JF, De Leyn, PR, Deneffe, GJ, et al Survival and prognostic factors in resected N2 non-small cell lung cancer: a study of 140 cases: Leuven Lung Cancer Group. Ann Thorac Surg 1997;63,1441-1450[Abstract/Free Full Text]
  88. Lacasse, Y, Bucher, HC, Wong, E, et al "Incomplete resection" in non-small cell lung cancer: need for a new definition; Canadian Lung Oncology Group. Ann Thorac Surg 1998;65,220-226[Abstract/Free Full Text]
  89. Chandrachud, LM, Pendleton, N, Chisholm, DM, et al Relationship between vascularity, age and survival in non-small-cell lung cancer. Br J Cancer 1997;76,1367-1375[ISI][Medline]
  90. Jefferson, MF, Pendleton, N, Faragher, EB, et al "Tumour volume" as a predictor of survival after resection of non-small-cell lung cancer (NSCLC). Br J Cancer 1996;74,456-459[ISI][Medline]
  91. Bourke, W, Milstein, D, Giura, R, et al Lung cancer in young adults. Chest 1992;102,1723-1729[Medline]
  92. Massard, G, Moog, R, Wihlm, JM, et al Bronchogenic cancer in the elderly: operative risk and long-term prognosis. Thorac Cardiovasc Surg 1996;44,40-45[ISI][Medline]
  93. Conill, C, Astudillo, J, Verger, E Prognostic significance of metastases to mediastinal lymph node levels in resected non-small cell lung carcinoma. Cancer 1993;72,1199-1202[CrossRef][ISI][Medline]
  94. Kondo, H, Asamura, H, Suemasu, K, et al Prognostic significance of pleural lavage cytology immediately after thoracotomy in patients with lung cancer. Thorac Cardiovasc Surg 1993;106,1092-1097
  95. Buhr, J, Berghauser, KH, Gonner, S, et al The prognostic significance of tumor cell detection in intraoperative pleural lavage and lung tissue cultures for patients with lung cancer. Thorac Cardiovasc Surg 1997;113,683-690
  96. Fukuse, T, Hirata, T, Tanaka, F, et al Prognosis of ipsilateral intrapulmonary metastases in resected nonsmall cell lung cancer. Eur J Cardiothorac Surg 1997;12,218-223[Abstract]
  97. Ichinose, Y, Hara, N, Ohta, M, et al Is T factor of the TNM staging system a predominant prognostic factor in pathologic stage I non-small-cell lung cancer? A multivariate prognostic factor analysis of 151 patients. Thorac Cardiovasc Surg 1993;106,90-94
  98. Sobue, T, Suzuki, T, Fujimoto, I, et al Prognostic factors for surgically treated lung adenocarcinoma patients, with special reference to smoking habit. Jpn J Cancer Res 1991;82,33-39[CrossRef][ISI]
  99. Wertzel, H, Siebert, H, Lange, W, et al Results after surgery in stage-I bronchogenic carcinoma. Thorac Cardiovasc Surg 1998;46,365-369[ISI][Medline]
  100. Diez, M, Torres, A, Maestro, ML, et al Prediction of survival and recurrence by serum and cytosolic levels of CEA, CA125 and SCC antigens in resectable non-small-cell lung cancer. Br J Cancer 1996;73,1248-1254[ISI][Medline]
  101. Gail, MH, Eagan, RT, Feld, R, et al Prognostic factors in patients with resected stage I non-small cell lung cancer: a report from the Lung Cancer Study Group. Cancer 1984;54,1802-1813[CrossRef][ISI][Medline]
  102. Muller, T, Marshall, RJ, Cooper, EH, et al The role of serum tumour markers to aid the selection of lung cancer patients for surgery and the assessment of prognosis. Eur J Cancer Clin Oncol 1985;21,1461-1466[CrossRef][ISI][Medline]
  103. Clee, MD, Hockings, NF, Johnston, RN Bronchial carcinoma: factors influencing postoperative survival. Br J Dis Chest 1984;78,225-235[ISI][Medline]
  104. Macchiarini, P, Fontanini, G, Hardin, MJ, et al Blood vessel invasion by tumor cells predicts recurrence in completely resected T1 N0 M0 non-small-cell lung cancer. Thorac Cardiovasc Surg 1993;106,80-89
  105. Brechot, JM, Chevret, S, Charpentier, MC, et al Blood vessel and lymphatic vessel invasion in resected nonsmall cell lung carcinoma: correlation with TNM stage and disease free and overall survival. Cancer 1996;78,2111-2118[CrossRef][ISI][Medline]
  106. Kessler, R, Gasser, B, Massard, G, et al Blood vessel invasion is a major prognostic factor in resected non-small cell lung cancer. Ann Thorac Surg 1996;62,1489-1493