| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Ann Thorac Surg 2006;81:1958-1962
© 2006 The Society of Thoracic Surgeons
a Department of Surgery, University of Virginia School of Medicine, Charlottesville, Virginia
b Department of Health Evaluation Sciences, University of Virginia School of Medicine, Charlottesville, Virginia
c Department of Pathology, University of Virginia School of Medicine, Charlottesville, Virginia
Accepted for publication December 7, 2005.
* Address correspondence to Dr Jones, Thoracic and Cardiovascular Surgery, PO Box 800679, University of Virginia, Lee Street, 4th Floor, Suite 4823 Charlottesville, VA 22908-0679 (Email: djones{at}virginia.edu).
Presented at the Fifty-first Annual Meeting of the Southern Thoracic Surgical Association, Cancun, Mexico, Nov 2–4, 2004.
| General thoracic surgery:
The Annals of Thoracic Surgery CME Program is located online at http://cme.ctsnetjournals.org. To take the CME activity related to this article, you must have either an STS member or an individual non-member subscription to the journal.
|
 |
Abstract |
|---|
 Top Abstract IntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
METHODS: A retrospective evaluation of 119 consecutive patients from 1999 to 2004 with a pathologic diagnosis of T2N0 nonsmall cell lung cancer was performed. Patient follow-up was 97%. Overall survival and disease-free survival rates were calculated by the Kaplan-Meier method. Univariate analysis was performed using the log rank test and multivariate analysis by Cox's proportional hazard model. Data were significant if p < 0.05.
RESULTS: The 4-year overall survival and disease-free survival rates were 62% and 60%, respectively. The local and distant recurrence rates were 5% and 18%, respectively. Tumor size (p = 0.001), histologic grade (p = 0.002), the Eastern Cooperative Oncology Group performance status (p = 0.002), angioinvasion (p = 0.03), and visceral pleural involvement (p = 0.02) were predictors of overall survival by univariate analysis. Multivariate analysis demonstrated increasing tumor size (1.26 [95% confidence intervals 1.12, 1.64]) and histologic grade (4.05 [95% confidence intervals 1.38, 11.90]) to be significant independent predictors of a worse overall survival. The 4-year survival of patients without any of these variables was 89% compared with 56% if one or more of these factors were present (p = 0.03).
CONCLUSIONS: There is significant heterogeneity in the T2N0 class of nonsmall cell lung cancer. Risk stratification using specific histopathologic variables may help determine which patients will benefit most from adjuvant therapy.
Despite an enhanced appreciation of complete nodal dissection and the need for an R0 resection, the overall survival for pathologically staged T2N0M0 or stage IB nonsmall cell lung cancer (NSCLC) remains approximately 60% [1]. This modest result for node-negative NSCLC has been a source of frustration for thoracic surgeons and their patients. One explanation for the lower than expected survival of these patients is the heterogeneity in the T2 classification. The current T2 classification includes tumors that are greater than 3.0 cm, tumors that invade the visceral pleura regardless of size, tumors that involve the mainstem bronchus but are greater than or equal to 2 cm distal from the carina, and tumors that result in atelectasis or lobar collapse radiographically [1]. This T2 classification schema has remained unchanged for the past 30 years. Consideration of these criteria suggests that the T2 classification is based chiefly on tumor size and location, which is a simplistic approach at best, and an inaccurate one at worst.
Advances in tumor biology have resulted in the evaluation of numerous proteins that may provide a more accurate, or at least a complimentary, prognostication tool compared with the current TNM staging system [2]. Although clinical application of molecular signatures in NSCLC remains a laudable goal, which subset of biologic markers to use and universal agreement on the type and method of technique for measurement of such markers continue to limit widespread use of biomarkers to help stage or recommend treatment to patients.
Recent findings of the International Adjuvant Lung Cancer Trial suggest that adjuvant chemotherapy offers improved survival for R0 resected pathologic stages IB to IIIA [3]. Similarly, analysis of the phase III multi-institutional Cancer and Leukemia Group B 9633 trial suggests that adjuvant chemotherapy improves overall and disease-free survival in patients with resected T2N0M0 NSCLC [4]. What is unclear from these studies, and others [5], is whether all patients with pT2N0M0 disease should receive chemotherapy, and, if not, whether there are criteria that may identify those patients who are likely to receive the most benefit from such an approach.
Given the heterogeneity of the current T2 classification, the limitations with clinical applicability of biomarker(s), as well as the resurgence of interest in adjuvant chemotherapy, we sought to determine if specific histopathologic variables may help better define which patients with pT2N0M0 tumors had a worse prognosis, and thus may potentially serve as factors to consider in future clinical trial development and staging system revisions.
|
Material and Methods |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
Patient demographics, performance status, tumor location, type of resection, and overall and disease-free survivals were documented. The criteria for T2 classification were also noted. Specific histopathologic variables evaluated included tumor histology, tumor size, histologic grade, the presence of lymphatic invasion, angioinvasion, or visceral pleural involvement. Low histologic grade was defined as well-differentiated or mildly-differentiated lesions, whereas high-grade tumors had moderately-differentiated or poorly-differentiated lesions. Finally, the site of first recurrence was documented.
Follow-Up
Patient follow-up was complete in 97% of the study patients. A combination of medical records, tumor registry, and telephone interviews were used to document survival and the presence of any recurrence of their disease. Median follow-up was 25 months (range, 1 to 55 months).
Statistical Evaluation
Basic summary statistics including frequencies and percentiles were calculated. Kaplan-Meier survival estimates and 95% confidence intervals were obtained for overall survival (OS) and disease-free survival (DFS), both overall and univariately for each of several possible categorical predictors. Log rank tests were used to assess differences in survival across levels of these predictors. Similarly, univariate Cox proportional hazard modeling was used to examine continuous predictors. Multivariate Cox proportional hazards models yielded estimated hazard ratios and 95% confidence intervals used to examine relationships between survival and several predictors of interest. Furthermore, in the Cox proportional hazards context, a more flexible method of modeling a continuous predictor, the restricted cubic spline, was used as needed to model more complex relationships between survival and a particular predictor.
|
Results |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
As shown in Table 1, there were slightly more men than women, with the median age being 68 years. The majority of patients had an Eastern Cooperative Oncology Group performance status of 0, whereas 38% of patients had a higher performance status. Seventy-eight percent of patients had their tumor resected with a lobectomy, whereas only 7% of patients required a pneumonectomy. The in-hospital mortality rate was 4%, with the cause of death being cardiac (2), pneumonia (2), and pulmonary embolus (1).
|
|
|
|
5 cm) or high-grade tumor histology.
|
|
|
Comment |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
Despite advances in the imaging of lung cancer, as well as a better understanding of its tumor biology, there have been no significant changes to the staging system to reflect this progress. Although there remains significant hope for using immunohistochemical protein analysis or other molecular biology or gene-array analyses of tumors to prognosticate and guide therapy, published results for the majority of such biomarkers are frequently conflicting, and their use in clinical practice has been limited to a discouragingly few clinical trials. In contrast, basic histopathologic tumor variables are uniformly utilized and reported [8], although their ability to guide adjuvant therapy has never been rigorously evaluated. We believe that this is the first study that has evaluated all of these variables with respect to pathologic stage IB NSCLC only.
As shown in our data, the classification of a T2 lesion is largely based on either size or visceral pleural invasion criteria, with no patient being classified solely on the presence of atelectasis or lobar obstruction or the 2 cm carinal proximity rule. This suggests that the latter stated two T2 criteria could be omitted in future iterations of the staging system.
Given the 57% 5-year survival for pathologic staged T2N0M0 NSCLC [1], it is highly likely that there are unrecognized tumor biologic variables in the current staging classification that account for worse prognosis than would be expected for node-negative NSCLC. In an analysis of isolated stage IB patients, Padilla and colleagues [9] found that tumor size was the only predictive variable for a worse prognosis. The importance of tumor size in T2 lesions has also been emphasized by Carbone and colleagues [10] who have a shown 5-year survivals of 62% and 51% for T2NO lesions greater than 3.0 cm and less than or equal to 5.0 cm, and greater than 5.0 cm, respectively. Similarly several authors have identified visceral pleural invasion to be an independent negative prognosticator in stage IB patients as well as other stages [11–13].
Our study confirmed that both increasing tumor size (as a continuous variable) and histologic tumor grade were significant predictors of a decreased overall survival (Table 4). Moreover, tumor size and tumor angioinvasion were both predictors of a worse OS and DFS on multivariate analysis. These results suggest that the histopathologic analysis of tumor size, histologic grade, and angioinvasion may permit the identification of a suset of T2N0 patients that have a high likelihood of having recurrent disease develop.
An examination of these variables and their association with either locoregional or distant recurrence reveals that more than 50% of patients with distant recurrence had high-grade tumor histology and larger tumors ( 5.0 cm) (Table 5). In addition, the presence of visceral pleural invasion or tumor angioinvasion was also found in 36% and 27%, respectively, of patients with distant recurrence. Evidence of the impact of these variables on overall survival is demonstrated in Figure 1] in which those patients with smaller tumors sizes (< 5.0 cm), no angioinvasion, and a low-grade histology had a significantly superior 4-year survival of 89% compared with a 56% survival for patients with one or more of these variables. Similar significant results were seen with disease-free survivals (data not shown).
Identification of which patient is at high-risk for recurrence and cancer death after resection of their stage IB NSCLC is important. Historically, the use of adjuvant chemotherapy after an R0 resection for patients with any stage NSCLC has not been a standard practice as there has been no demonstrated survival benefit [14]. Similarly there has been no demonstrable survival benefit for adjuvant radiation therapy in pathologic T2N0 lesions [15]. The treatment paradigm of post-resection observation has recently been challenged as the results of four, prospective, randomized, phase III studies have demonstrated a significant survival benefit with adjuvant chemotherapy after resection [3–5, 16]. Focusing only on stage IB disease, the Cancer and Leukemia Group B 9633 demonstrated a 4-year overall survival of 71% and 59% with adjuvant doublet chemotherapy and observation, respectively [4]. In addition, the Cancer and Leukemia Group B 9633 does have biologic data for their study, that once published, may offer additional insight into which variables may be important in predicting a response to adjuvant chemotherapy. Our 4-year survival of 62% is nearly identical to those results, although our lung cancer-specific mortality is only 13% compared with 26% in their study. In contrast, the International Adjuvant Lung Cancer Trial study demonstrated a 4.1% absolute overall survival benefit with adjuvant chemotherapy at 5 years for stages IB to IIIA. Although not adequately powered to detect differences between stages, it is interesting to note that there was no survival benefit for stages IA (n = 183) and IB (n = 498) disease [3]. This suggests certain subsets, but not all, of patients with NSCLC may benefit from adjuvant chemotherapy, and it highlights the need for appropriate risk stratification criteria.
The limitations associated with our article include the fact that it is retrospective and it is from a single institution. Although these are valid criticisms, all patient data is entered into our general thoracic surgery database concurrent with their care, and by the very nature of this type of analysis, an initial retrospective analysis is required. In addition, the number of patients available for review is reasonably robust, although a larger cohort may have permitted the identification of other significant prognostic factors. Finally, our data do not permit any direct correlation of these histologic variables and any potential survival benefits to adjuvant chemotherapy.
In conclusion, our data confirms the heterogeneous nature of the current stage IB classification and suggests that all stage IB NSCLC is not the same. We demonstrate that subset analysis of this stage, based on uniformly recognized and reported tumor histopathologic features, may better identify patients with T2N0 lesions who may benefit from adjuvant therapy. Specifically, these include tumor size, histologic grade, and the presence of angioinvasion. These findings need confirmation by either other large-scale retrospective reviews or prospective analyses, and when done, these should be considered for incorporation into future clinical trials for adjuvant therapy and in revisions of the staging system.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
DR JONES: Thank you, Dr Cerfolio. When we put together the manuscript and began to look at what is coming out in the literature about maximum SUV values with respect to prognostication, I wish we would have had those maximum SUV numbers, but we do not. One problem is that probably 50% of the patients had their PET scan at University of Virginia and the other 50% had outside studies. So even with a retrospective analysis, I could probably only identify an SUV value on maybe 60 or so of these patients.
|
References |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentDiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  C.-P. Hsu, J.-Y. Hsia, G.-C. Chang, C.-Y. Chuang, S.-E. Shai, S.-S. Yang, M.-C. Lee, and P.-C. Kwan Surgical-pathologic factors affect long-term outcomes in stage IB (pT2 N0 M0) non-small cell lung cancer: a heterogeneous disease. J. Thorac. Cardiovasc. Surg., August 1, 2009; 138(2): 426 - 433. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S.-H. I. Ou and J. A. Zell Response Chest, June 1, 2009; 135(6): 1695 - 1696. [Full Text] [PDF]
|
|
|
|
|
|
S.-H. I. Ou, J. A. Zell, A. Ziogas, and H. Anton-Culver Prognostic Significance of the Non-Size-Based AJCC T2 Descriptors: Visceral Pleura Invasion, Hilar Atelectasis, or Obstructive Pneumonitis in Stage IB Non-small Cell Lung Cancer Is Dependent on Tumor Size Chest, March 1, 2008; 133(3): 662 - 669. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
