Late stage (III and IV) non-small cell cancer of the lung: results of surgical resection at Inova Fairfax Hospital.

Abstract: One hundred forty-two patients underwent surgery and related treatment for advanced stage (III, IV) non-small cell cancer of the lung. One hundred seventeen patients underwent up-front surgery, with a hospital mortality rate of 1.7% (2/117). Kaplan-Meier 5-year survival in this group was 31% ([+ or -] 5). Twenty-five patients underwent neoadjuvant therapy followed by surgical resection, with respective rates of hospital mortality, complete pathologic response, and major pathologic response of 0%, 16%, and 64%. Kaplan-Meier 5-year survival in this latter group was 34% ([+ or -] 11). Of the 16 patients undergoing neoadjuvant therapy who had complete pathologic response or significant downstaging from stage III disease, Kaplan-Meier 5-year survival was 61% ([+ or -] 15). Three clinical observations of interest emerged regarding survival. First, in those patients with postresection FE[V.sub.1] < 1.0 L, hospital mortality rate was 20%, and there were no 5-year survivors (P < 0.0001). Second, where neoadjuvant therapy was associated with complete pathologic response or significant downstaging of disease, there was a trend for improved survival in the downstaged group, but it did not reach statistical significance (P = 0.14). Third, adjuvant therapy was associated with improved 5-year survival (P = 0.03), particularly for combination chemotherapy and radiotherapy (P = 0.02).

Key Words: lung cancer, non-small cell, late stage, neoadjuvant therapy, adjuvant therapy

**********

Carcinoma of the lung is the single leading cause of death from cancer in men and women, responsible for greater than one quarter of all such deaths. (1) Many studies (2-4) attest to the benefit of surgical resection of non-small cell carcinoma of the lung (NSCCL), when diagnosed in advance of metastases to mediastinal lymphatics (stage III-A) or even more advanced (stage III-B and IV) disease. Until quite recently, with few exceptions, (5-8) surgical treatment for advanced disease was considered only palliative and the end results dismal. (3,6,9) More recently, we (10) and others (11-14) have reported surgical series of advanced stage NSCCL with significant improvements over historic reports for both hospital mortality and long-term survival.

Between January 1, 1991, and October 31, 2003, 666 patients underwent complete surgical resection for NSCCL at Inova Fairfax Hospital. Of that group, 24% (163/666) had advanced late stage (III, IV) disease. Although prior studies have reported much poorer long-term survival, (3,6,9) our experience in this group (10) was so encouraging that we decided to examine in more detail our patients with advanced late stage disease.

Materials and Methods

Of the original 163 patients who had advanced-stage disease and good-to-excellent performance status, 21 patients were eliminated for this review. Nine patients were eliminated because the differentiation of N2 from N1 lymph node might reasonably be questioned; level 10 was classified as an N1 lymph node. (15) Six patients with neuroendocrine histology were eliminated to remove any doubts as to whether survival benefit accrued from possible inadvertent inclusion of carcinoid-type tumors. Four patients were eliminated because a definitive histology was not possible--two of whom underwent needle biopsy elsewhere. Finally, two patients were eliminated when a previous staging classification of stage III (T3N0) was carried over inadvertently as stage III in the newer lung cancer classification. (16)

All of the remaining 142 patients underwent complete surgical resection of stage III or IV NSCCL, as detailed by clinical and pathologic staging according to the New International Staging System for Lung Cancer, (16,17) as accepted by the American Joint Committee on Cancer (18) (Table 1 and Fig. 1). Complete resection was defined by Watanabe criteria, as "(any) procedure in which all accessible lymph nodes were removed, even if the most distal node was found to be involved." (6) Final staging was based on total histologic review of resected lung, lymph nodes, and tissues.

No single regimen of preoperative (neoadjuvant) chemotherapy was followed. Multiple oncology groups used a variety of cis/carboplatin regimens, with varying number of cycles and supplemental agents. Neoadjuvant radiation ranged from 4,500 to 6,300 Gray (Gy), most commonly 4,500 to 5,000 Gy. For postoperative (adjuvant) care, the style and degree of chemotherapy and radiotherapy also varied considerably.

Patient survival was examined by age, sex, cell type, TNM classification, type of therapy, and extents of tumor and resection. Hospital mortality rate was also calculated. Information with regard to lymph node staging was obtained from tissue submitted for pathology review, whether from resected pulmonary parenchyma and nodes obtained at thoracotomy, from prethoracotomy surgical staging procedures, or both.

Follow-up in this cohort was 100%, and the censor date for those patients still alive was October 31, 2003. Survival was calculated according to the method of Kaplan-Meier, (19) and the log-rank test was used to compare two or more survival curves. (20) Statistical significance was defined as P [less than or equal to] 0.05. Also, simple alive/dead proportions were calculated, irrespective of staging, therapy or cause of death.

Results

Population characteristics

Of the 142 patients, 56% were male (80/142) and 44% (62/142) were female. Patient ages ranged from 32 to 82 years and did not differ significantly by gender. Preoperative estimates of postoperative 1-second forced expiratory volumes (FE[V.sub.1]) in these patients were characterized as <1.0 L, [greater than or equal to]1.0 L, and unknown. The corresponding proportions of patients falling into these categories were 7% (10/142), 80% (113/142), and 13% (19/142). Tumors were classified according to histology, stage, and TNM subgroups. With regard to cell type, 58% (83/142) of these patients had squamous cell tumors, whereas 37% (53/142) had adenocarcinomas and 4% (6/142) had large cell undifferentiated carcinomas.

With regard to final stage of disease, 16 patients were downstaged by neoadjuvant therapy to stage 0 or complete pathologic response (CPR) (4 patients), I-A (3 patients), I-B (4 patients), and II-B (5 patients). Seventy-one patients were classified as stage III-A and 40 patients as stage III-B. Fifteen patients were stage IV, either resected of multiple cancers in one lobe with concomitant nodal metastases or of separate cancers in different lobes at the same operation. Finally, 49 of 68 patients with N2 disease harbored single-level mediastinal metastases, whereas 12 patients harbored multiple-level metastases and 7 harbored indeterminate metastases.

[FIGURE 1 OMITTED]

Diagnostics and therapeutics

All patients in this series underwent complete surgical resection according to Watanabe criteria. (6) All patients treated neoadjuvantly completed their preoperative therapy before undergoing resection. Follow-up in these patients was 100%, encompassing all groups and subgroups. Eighty-two percent (117/142) of these patients underwent up-front surgical resection, 31% of these (36/117) with mediastinoscopy/mediastinotomy. Also in this up-front surgical resection group, 26% (30/117) underwent positron emission tomography (PET) and 100% underwent CT.

Of the total group, 18% (25/142) underwent preoperative (neoadjuvant) therapy. Sixteen patients received chemotherapy and 9 patients received chemotherapy plus radiotherapy before undergoing surgical resection. All 25 patients underwent preoperative CT scanning. Of the 16 patients pretreated with chemotherapy, downstaging occurred in 63% (10/16). Thirteen of the 16 tested positive either for hypermetabolic mediastinum on PET (10 patients) and/or with positive nodal pathology at mediastinoscopy/mediastinotomy (6 patients). Of the remaining 3 patients, 1 underwent negative mediastinoscopy, emerging stage III-A (T3N1) after neoadjuvant therapy and subsequent resection. The final two patients were treated on the basis of CT alone, emerging after therapy with III-B and I-B disease. The patient with stage 1-B disease died 2 months after surgery.

In 10 instances in which mediastinoscopy/mediastinotomy was negative or not performed in the face of hypermetabolic (positive) mediastinum on PET, PET was repeated before surgery but after chemotherapy in four patients. Two of the four repeat PET scans were negative and two PET scans remained positive. The two latter patients were resected of I-B and II-B disease and, at postoperative follow-up, both patients remained alive, the former at 13 months and the latter at 16 months. Of the remaining 6 patients, 3 were resected of stage III disease. Two of these remained alive at 60 and 30 months, respectively, and one died at 15 months after surgery. The final three were resected of stage 0, 0 and II-B disease. Two remained alive at 49 and 43 months, respectively, and one died at 2 months after surgery. In the above 6 patients, no granulomatous changes or any other explanation for the mediastinally positive PET scans were found.

Of the 9 patients treated neoadjuvantly with chemotherapy plus radiotherapy, downstaging occurred in 67% (6/9). Eight patients tested positive with either hypermetabolic mediastinum on PET scan (6 patients) and/or with nodal pathology at mediastinoscopy/mediastinotomy (4 patients). The remaining patient was resected of stage III (T2N2) disease. In all four instances in which mediastinoscopy/mediastinotomy was negative or not performed in the face of hypermetabolic (positive) mediastinum on PET, PET was repeated before surgery but after chemotherapy and radiotherapy. All four repeat PET scans were negative. The mediastinal nodes were negative in all four at surgery, with three patients staged II-B and one staged I-B.

Lobectomy was the most commonly performed lung resection technique, involving 96 patients, of whom 7 underwent bilobectomy and 1 sleeve lobectomy. Pneumonectomy was performed in 29 patients (13 right, 16 left). Wedge resection was performed in 17 patients. Resection of a portion of the chest wall or mediastinum was also performed in 15 and 4 patients, respectively.

Survival

Hospital mortality rate was 1.4% (2/142). Both deaths occurred after performance of right pneumonectomy, and both occurred in the group of 10 patients whose preoperatively calculated postoperative FE[V.sub.1] was less than 1.0 L. There were no statistically significant differences in 5-year survival patterns among patients examined by extent of resection: pneumonectomy versus lobectomy versus wedge resection (P = 0.49). The single most significant factor in early and late patient death was predicted, postoperative FE[V.sub.1] (P < 0.0001, Table 2).

No statistically significant differences in 5-year survival were found between the 117 patients undergoing up-front surgical resection and the 25 patients treated neoadjuvantly (P = 0.73, Table 3-A). No statistically significant survival differences were found between neoadjuvant chemotherapy and neoadjuvant chemotherapy plus radiotherapy (P = 0.19, Table 3-B). The lack of a statistically significant difference here may be due to small sample size and insufficient accrual survival time in the neoadjuvant chemotherapy plus radiotherapy group to permit a 5-year comparison. Finally, no statistically significant differences in survival were found between neoadjuvant chemotherapy plus surgery and up-front surgery (P = 0.32, Table 3-C1). Again, the limited sample of patients treated neoadjuvantly with chemotherapy plus surgery may be an issue here.

Sixteen patients were treated with preoperative (neoadjuvant) chemotherapy before undergoing surgical resection. Six of these patients had N2 disease that was not pathologically proven, either before or after surgery. At follow-up, 1 patient with stage 0 disease died at 2 months, 1 patient with stage I-B disease died at 2 months, 1 patient with stage 0 was alive at 49 months, 1 patient with stage I-B disease was alive at 13 months, and 2 patients with stage II-B disease were alive at 16 and 43 months, respectively. To examine whether the inclusion of these six patients might create an unintended bias toward longer survival, we eliminated them from the neoadjuvant chemotherapy plus surgery group. We then compared the remaining 10 patients with the 117 patients in the up-front surgery group (Table 3-C2) and found no statistically significant differences in survival (P = 0.41). Again, the limited sample of patients treated neoadjuvantly with chemotherapy plus surgery may be an issue here.

Table 4 depicts the survival experience of our stage III/IV patients. Of the 25 patients who were treated neoadjuvantly, 64% (16/25) were postoperatively downstaged to one degree or another. Five-year survival in the downstaged group was 61% versus 25% in the group with N2 disease, although this difference was not statistically significant (P = 0.14). Although patients with single-level N2 disease appeared to have 5-year survival rates superior to those with multilevel disease (34% versus 16%), there was no statistically significant difference (P = 0.19).

Finally, Table 5 depicts the survival experience of patients by postoperative (adjuvant) therapy. Patients treated adjuvantly with chemotherapy with or without radiation showed the best survival (P = 0.02).

Discussion

This review of our survival experience in stages III and IV NSCCL has several limitations. First, the review is retrospective. Second, although we are able to distinguish single-level N2 disease from multilevel N2 disease, (12,14) we did not differentiate between complete and incomplete resection on the basis of Mountain's lymphatic criteria. (21) Third, not all N2 disease was pathologically proven before surgery. Fourth, different and varying regimens of chemotherapy and radio-therapy were used in the neoadjuvant and adjuvant groups. Nevertheless, the 5-year survival achieved with an aggressive surgical strategy was 31%, at minimum.

Although the latter are Kaplan-Meier estimates, not actuarial survivals as recently reported by Machtay, (11) our survivals with surgery up front would seem at first glance to equal that achieved at the University of Pennsylvania (11) with radiotherapy to 45 to 54 Gy and concurrent chemotherapy. Clearly, such is not the case, as it has long been known that extended survivals and cures may be achieved when operating on and resecting occult, mediastinal (N2) disease. (3,5,7,8) We also have continued Martini's (22) lead into neoadjuvant application of chemotherapy plus or minus radiotherapy before operating for advanced disease. Just as the University of Pennsylvania (11) celebrates improved 5-year survival to 48% with major pathologic response, we also note improvement to 61% in 5-year survival in that group downstaged by neoadjuvant therapy (refer to Table 4, "downstaged"). As fully 64% of our neoadjuvantly treated patients are downstaged significantly, application of neoadjuvant therapy to aggressive surgical resection is increasingly seen to bear improved results. (11,22)

Compared with the results of Machtay et al, (11) our results using single-agent, neoadjuvant chemotherapy improved CPR by 46% (19% versus 13%) and overall major pathologic response by 50% (63% versus 42%), and our hospital mortality rate of 0% in this group compares favorably with their "early death rate of 9%." Even if the six patients not histologically proven before surgery to harbor stage III disease were deleted from the neoadjuvant chemotherapy group, the remaining 10 patients had respective hospital mortality rates, CPR, major pathologic response, and 5-year survival rates of 0%, 10%, 40%, and 39%. In view of these results and those of Memorial Sloan-Kettering, (23) in which "the addition of (preoperative) radiotherapy did not significantly increase survival," we continue to favor simple but aggressive neoadjuvant chemotherapy alone for patients with bulky mediastinal disease whose clinical performance status is good to excellent.

Given the benefits and limitations of neoadjuvant therapy, we do not subscribe to a forced choice of either neoadjuvant or adjuvant therapy. If patients are deemed fit enough, we will aggressively pursue a triple regimen (neoadjuvant therapy, surgical resection, and adjuvant therapy), as adjuvant therapy has its own potential advantages. With minimal residue of disease, and those cells preponderantly in their growth phase, postoperative remnants are maximally susceptible to chemo/radiotherapy. (24) This approach seems particularly reasonable, given our results with adjuvant therapy (Table 5) and recent developments favoring adjuvant chemotherapy for lung cancer as a new standard of care. (25,26) Furthermore, if one adopts a neoadjuvant program of chemotherapy alone, then postoperative radiotherapy may be given in higher doses and in an uninterrupted fashion, if required.

Adjuvant therapy may be delivered as chemotherapy, radiotherapy, or combination chemoradiation therapy (sequential or concurrent). Impressive results have been achieved recently with adjuvant cisplatin-based chemotherapy. (25) Almost two decades ago, the Lung Cancer Study Group (27) demonstrated the ability of adjuvant radiotherapy to minimize local recurrence, at the cost of increased morbidity and non-cancer-related mortality rates. Also noted in that same study was a trend toward increased survival in stage III-A N2-positive patients. Subsequently, the Eastern Cooperative Oncology Group demonstrated that adjuvant, combination chemoradiation therapy (50.4 Gy) did not improve survival benefit over radiotherapy alone. (28) However, in our experience (Table 5), adjuvant radiotherapy, supplemented with sensitizing chemotherapy, was associated with the best 5-year survival.

Are we making progress in the treatment of NSCCL? In a 1992 review, (3) the University of Toronto reported that N2-positive patients surviving negative mediastinoscopy and complete resection achieved a 5-year survival of 35%, (29) somewhat similar to our results. However, when operative mortality, incomplete resection, or inability to resect were included in their data analysis, 5-year survival dropped to 24% when mediastinoscopy was negative and 9% when mediastinoscopy was positive. (30) Thus, in our experience, where we have included all hospital deaths and have achieved survivals ranging from 31 to 61%, we do seem to be making some progress. When examining our results against the experiences of Toronto (3) and the University of Pennsylvania, (11) two points stand out in patient selection. First, treatment must be based on extent of disease. Second, treatment must also be based on clinical performance, as our results in patients with poor predicted postoperative FE[V.sub.1] well document. It bears repeating that there were no long-term survivors when preoperatively predicted postoperative FE[V.sub.1] was less than 1.0.

Conclusions

Our experience suggests that aggressive therapy for advanced stage (III, IV) NSCCL, including surgical resection, appears worthwhile in patients with reasonable pulmonary function and good-to-excellent performance status. (7,10) First, the appearance of advanced disease may be misleading, as Watanabe (6) documented false-positive CT scans for N2 disease in 39% of patients ultimately resected with stage I or II disease. Second, up-front surgery alone may yield a 5-year survival rate equal to 31%. Third, skillfully combining surgery with other neoadjuvant and/or adjuvant therapies may further improve 5-year survival to 45 to 61%.

We currently favor neoadjuvant, multiagent chemotherapy and combination adjuvant chemotherapy plus radiotherapy, when combination therapy is desired. With only a 1% treatment-related mortality rate associated with adjuvant chemotherapy, (25) benefit/risk analysis for stage III patients satisfactorily recovering from surgery favors treatment. Although future experience will more clearly define what combination of treatments are best, at present we have seen significant gains in 5-year survival with very low early mortality rates.

References

1. Ries LAG, Hawkey BF, Miller BA, et al. Cancer statistics review 1973-1988. Natl Cancer Inst NIH 1991. pp 1791-1789.

2. Williams DE, Pairolero PC, Davis CS, et al. Survival of patients surgically treated for stage I lung cancer. J Thorac Cardiovasc Surg 1981;82:70-76.

3. Miller JD, Gorenstein LA, Patterson G. Staging: The key to rational management of lung cancer. Ann Thorac Surg 1992;53:170-178.

4. Kiernan PD, Sheridan M, Hostetter AL, et al. Stage I non-small cell cancer of the lung: Results of surgical resection at Fairfax Hospital. Va Med Q 1993;120:146-149.

5. Martini N, Flehinger BJ. The role of surgery in N2 lung cancer. SCNA 1987;67:49-1037.

6. Watanabe Y, Shimizu J, Oda M, et al. Aggressive surgical intervention in N2 non-small cell cancer of the lung. Ann Thorac Surg 1991;51:253-261.

7. Kiernan PD, Sheridan M, Byrne WD, et al. Stage II and III-A non-small cell cancer of the lung: Results of surgical resection at INOVA Fairfax Hospital. Va Med Q 1994;121:172-178.

8. Daly BDT, Mueller JD, Faling LJ, et al. N2 lung cancer: Outcome in patients with false negative computed tomographic scans of the chest. J Thorac Cardiovasc Surg 1993;105:904-911.

9. Naruke T, Goya T, Tsuchiya R, et al. The importance of surgery to non-small cell carcinoma of the lung with mediastinal lymph node metastasis. Ann Thorac Surg 1988;46:603-610.

10. Kiernan PD, Graling PR, Hettrick VL, et al. A pragmatic and successful approach to the treatment of non-small cell carcinoma of the lung. AORN 2004;80:840-862.

11. Machtay M, Lee JH, Stevenson JP, et al. Two commonly used neoadjuvant chemo and radiation therapy regimens for locally advanced stage III non-small cell lung carcinoma: Long-term results and associations with pathologic response. J Thorac Cardiovasc Surg 2004;127:108-113.

12. Inoue M, Sawabata N, Takeda S, et al. Results of surgical intervention for P-stage III-A (N2) non-small cell lung cancer: Acceptable prognosis predicted by complete resection in patients with single N2 disease with primary tumor in the upper lobe. J Thorac Cardiovasc Surg 2004;127:1100-1106.

13. Okada M, Nishio W, Sakamoto T, et al. Evolution of surgical outcomes for non-small cell lung cancer: time trends in 1,465 consecutive patients undergoing complete resection. Ann Thorac Surg 2004;77:1926-1931.

14. Keller SM, Vangel MG, Wagner H, et al. Prologed survival in patients with resected non-small cell lung cancer and single-level N2 disease. J Thorac Cardiovasc Surg 2004;128:130-137.

15. Tisi GM, Friedman PJ, Peters RM, et al. Clinical staging of primary lung cancer. Am Rev Respir Dis 1983;127:659-670.

16. Mountain CF. Revisions in the international system for staging lung cancer. Chest 1997;111:1710-1717.

17. Mountain CF, Dresler CM. Regional lymph node classification for lung cancer staging. Chest 1997;111:1718-1723.

18. American Joint Committee on Cancer Staging Manual. Philadelphia, Lippincott-Raven Publishing, 1997, 5th ed, pp 37-127.

19. Cox DR, and Oakes D. Analysis of Survival Data. London, Chapman and Hall, 1984.

20. Peto R, Pike MC. Conservatism of the approximate sigma (O-E) 2-E in the log rank test for survival data in tumor-incidence data. Biometrics 1973;29:579-584.

21. Mountain CF. The biologic operability of stage III non-small cell lung cancer. Ann Thorac Surg 1985;40:60-64.

22. Martini N, Kris MG, Gralla RJ, et al. The effects of preoperative chemotherapy on the resectability of non-small cell lung cancer with mediastinal lymph node metastasis (N2M0). Ann Thorac Surg 1988;45:370-379.

23. Martin J, Ginsberg RJ, Venkatraman ES, et al. Long-term results of combined-modality therapy in resectable non-small cell lung cancer. J Clin Oncol 2002;20:1989-1995.

24. Simpson-Herren L, Sanford A, Holmquist J. Effects of surgery in the cell kinetics of residual tumor. Cancer Treat Res 1976;60:1744-1760.

25. International Adjuvant Lung Cancer Trial Collaborative Group. Cisplatin-based adjuvant chemotherapy in patients with completely resected non-small cell lung cancer. N Engl J Med 2004;350:351-360.

26. Blum RH. Adjuvant chemotherapy for lung cancer: A new standard of care. N Engl J Med 2004;350:404-405.

27. Weisenburger T. Effects of postoperative mediastinal radiation on completely resected stage II and stage III epidermoid cancer of the lung. N Engl J Med 1986;315:1377-1381.

28. Keller S, Adak S, Wagner H, et al. A randomized trial of postoperative adjuvant therapy in patients with completely resected stage II or stage IIIA non-small cell lung cancer. N Engl J Med 2000;343:1217-1222.

29. Pearson FG. Lung cancer: The past twenty-five years. Chest 1986;89:200S-205S.

30. Pearson FG, Delarue NC, Ilves, TRJ, et al. Significance of positive superior mediastinal nodes. identified at mediastinoscopy in patients with resectable cancer of the lung. J Thorac Cardiovasc Surg 1982;83:1-11.

Believe those who seek the truth. Doubt those who find it.
--Andre Gide

Paul D. Kiernan, MD, Michael J. Sheridan, SCD, James Lamberti, MD, Thomas LoRusso, MD, Vivian Hetrick, RN, Betty Vaughan, RN, and Paula Graling, RN

From the Inova Health System, Inova Fairfax Hospital, Sections of Thoracic Surgery, Epidemiology and Biostatistics, Pulmonary Medicine, and Thoracic Operating Room Nursing, Annandale, Virginia.

Reprint requests to Dr. Kiernan, Section of Thoracic Surgery, Inova Fairfax Hospital, c/o Suite 301, 3301 Woodburn Road, Annandale, VA 22003. E-mail: [e-mail]

Accepted May 10, 2005.

RELATED ARTICLE: Key Points

* One hundred seventeen patients underwent up-front surgery with respective hospital mortality rates and 5-year survival of 1.7% and 31%.

* Twenty-five patients, most with bulky mediastinal disease, underwent neoadjuvant therapy followed by surgical resection with respective rates of hospital mortality, complete pathologic response, major pathologic response, and 5-year survival of 0%, 16%, 64%, and 34%.

* Of the 16 patients undergoing neoadjuvant therapy who had complete pathologic response or significant downstaging of disease, 5-year survival was 61%.

* Among patients with an estimated postoperative FE[V.sub.1] < 1.0 L, there were no 5-year survivors (P < 0.0001).

* Adjuvant therapy was associated with improved 5-year survival (P = 0.03), particularly for combination chemotherapy and radiotherapy (P = 0.02).

Table 1. International (TNM) staging system for lung cancer (a)

TNM descriptors

Primary tumor (T)
  0, No evidence of primary tumor
  1, [less than or equal to] 3 cm in greatest dimension; not in main
  bronchus
  2, > 3 cm in greatest dimension; involves main bronchus > 2 cm distal
  to carina; invades visceral pleura
  3, of any size that directly invades chest wall, parietal or
  mediastinal pleura, pericardium, diaphragm, or tumor in the main
  bronchus < 2 cm distal to the carina without involvement of the
  carina
  4, of any size that invades "vital mediastinum" such as heart, great
  vessels, trachea, esophagus, vertebral body, carina, tumor with
  malignant pleural or pericardial effusion, or with satellite tumor
  nodule(s) within ipsilateral primary tumor lobe of the lung
Lymph nodes (N)
  0, No metastasis
  1, Metastasis to ipsilateral hilar and/or intrapulmonary nodes
  2, Metastasis to ipsilateral mediastinal and/or subcarinal nodes
  3, Metastasis to contralateral mediastinal or scalene or
  supraclavicular nodes
Distant metastasis (M)
  0, No distant metastasis
  1, Distant metastasis, including nodule(s) in ipsilateral nonprimary
  tumor lobe(s) of the lung

              Stage groupings

I-A        T1N0M0      III-A      T3N1M0
I-B        T2N0N0                 Any N2M0
II-A       T1N1M0      III-B      Any T4M0
II-B       T2N1M0                 Any N3M0
           T3N0M0      IV         Any M1

(a) TNM, tumor, node, metastasis.
Adapted from Mountain CF. (16)

Table 2. Advanced stage non-small cell carcinoma of the lung survival by
predicted postoperative 1-second forced expiratory volume status. Inova
Fairfax Hospital January 1, 1991, to October 31, 2003 (a)

                                                 Kaplan-Meier survival
Predicted postoperative               No.            estimates (SE)
FE[V.sub.1]                       alive/dead    1 yr     2 yr     3 yr

< 1.0 L                              0/10       22 (14)  22 (14)  *
[greater than or equal to] 1.0 L    54/59       82 (4)   82 (4)   46 (5)

                                  Kaplan-Meier survival
Predicted postoperative               estimates (SE)
FE[V.sub.1]                        4 yr    5 yr      P

< 1.0 L                           *       *       <0.0001
[greater than or equal to] 1.0 L  38 (6)  36 (6)

*Insufficient accrual time for comparison.
(a) SE, standard error of Kaplan-Meier survival estimate.

Table 3. Advanced stage non-small cell carcinoma of the lung survival by
type of therapy, with and without neoadjuvant therapy. Inova Fairfax
Hospital, January 1, 1991, to October 31, 2003 (a)

                                        Kaplan-Meier survival
                                           estimates (SE)
Therapy               No. alive/dead    1 yr    2 yr     3 yr

A. Surgery up front       45/72        77 (4)  53 (5)   41 (5)
vs
Neoadj Rx + surgery       12/13        92 (5)  58 (11)  34 (11)
B. Type neoadj Rx
  Chemo only               9/7         88 (8)  64 (13)  45 (14)
  vs
  Chemo + radiation        3/6        100 (0)  47 (19)  *
C. Therapy
C1. Chemo + surgery        9/7         88 (8)  64 (13)  45 (14)
  vs
C2. Surgery up front      45/72        77 (4)  53 (5)   41 (5)
D. Chemo + surgery         5/5        100 (0)  78 (14)  39 (17)
vs
Surgery up front          45/72        77 (4)  53 (5)   41 (5)

                          Kaplan-Meier survival
                              estimates (SE)
Therapy                  4 yr       5 yr       P

A. Surgery up front     33 (5)     31 (5)
vs                                            0.73
Neoadj Rx + surgery     34 (11)    34 (11)
B. Type neoadj Rx
  Chemo only            45 (14)    45 (14)
  vs                                          0.19
  Chemo + radiation     *          *
C. Therapy
C1. Chemo + surgery     45 (14)    45 (14)
  vs                                          0.32
C2. Surgery up front    33 (5)     31 (5)
D. Chemo + surgery      39 (17)    39 (17)
vs                                            0.41
Surgery up front        33 (5)     31 (5)

*Insufficient accrual time for comparison.
(a) Neoadj Rx, neoadjuvant therapy.

Table 4. Advanced stage non-small cell carcinoma of the lung survival by
final pathology stage. Inova Fairfax Hospital, January 1, 1991, to
October 31, 2003 (a)

                                           Kaplan-Meier survival
Final stage (No.)                              estimates (SE)
(N = 142)               No. alive/dead    1 yr      2 yr     3 yr

CPR (a) (4)                   3/1        75 (22)   75 (22)  75 (22)
I-A (b) (3)                   2/1       100 (0)   100 (0)   50 (35)
I-B (c) (4)                   1/3        75 (22)  *         *
II-B (b) (5)                  4/1       100 (0)    75 (22)  75 (22)
III-A (71)                   26/45       80 (5)    55 (6)   44 (6)
III-B (40)                   17/23       74 (7)    52 (9)   36 (9)
IV (15)                       4/11       86 (9)    50 (13)  29 (12)

                                           Kaplan-Meier survival
                                               estimates (SE)
Disease category (No.)  No. alive/dead    1 yr      2 yr     3 yr

Downstaged (b) (16)          11/5        88 (9)    61 (15)  61 (15)
N2 positive (68)             22/46       81 (5)    51 (6)   38 (6)

                         Kaplan-Meier survival
Final stage (No.)            estimates (SE)
(N = 142)                4 yr     5 yr      P

CPR (a) (4)             75 (22)  *
I-A (b) (3)             50 (35)  50 (35)
I-B (c) (4)             *        *
II-B (b) (5)            *        *         0.48
III-A (71)              33 (6)   31 (6)
III-B (40)              36 (9)   36 (9)
IV (15)                 21 (11)  21 (11)

                         Kaplan-Meier survival
                             estimates (SE)
Disease category (No.)   4 yr     5 yr      P

Downstaged (b) (16)     61 (15)  61 (15)   0.14
N2 positive (68)        28 (6)   25 (6)

*Insufficient accrual time for comparison.
(a) CPR, Complete pathologic response.
(b) Downstaged by neoadjuvant therapy.

Table 5. Advanced stage non-small cell carcinoma of the lung survival by
postoperative (adjuvant) therapy. Inova Fairfax Hospital, January 1,
1991, to October 31, 2003

                                             Kaplan-Meier survival
Type postoperative therapy                       estimates (SE)
(No.) (N = 142)             No. alive/dead   1 yr     2 yr     3 yr

Chemo + radiation (39)           21/18      95 (4)   67 (8)   55 (9)
Chemo only (21)                  11/10      84 (8)   50 (12)  39 (13)
Radiation only (23)               6/17      72 (10)  50 (11)  39 (11)
None (59)                        19/40      71 (6)   47 (7)   31 (7)

                             Kaplan-Meier survival
Type postoperative therapy       estimates (SE)
(No.) (N = 142)              4 yr     5 yr      P

Chemo + radiation (39)      51 (9)   51 (9)
Chemo only (21)             39 (13)  39 (13)  0.03*
Radiation only (23)         20 (10)  20 (10)
None (59)                   26 (7)   21 (7)

*For trend in postoperative therapy survival. For chemotherapy with
radiation versus other postoperative therapies, P = 0.02.