References
1. Rosell R: Personal Communication
2. Roth J: Personal Communication
3. Burkes R: Personal Communication
4. Martini N: Personal Communication
5. Kumar P, et al. Trimodality treatment of surgically staged IIIa (N2) non-small cell lung cancer (NSCLC): Updated analysis of cancer and leukemia group B (CALGB) protocol 8935. Proceedings of the 8th World Conference on Lung Cancer. Lung Cancer (Suppl) 1997;18:(Abs. 271).
6. Albain K: Personal Communication
7. Darwish S, et al. Long-term follow-up in 46 patients (pts) with IIIa (clinical N2) non-small cell lung cancer (NSCLC) after neoadjuvant chemotherapy (CT). Proceedings of ASCO. 1997;16: 466.
8. Weitberg A, et al. Eleven year follow-up of trimodality therapy for stage IIIa non-small cell carcinoma of the lung. Proceedings of ASCO l997, l6: 487
9. Bonomi P and Faber P. Personal Communication.

____________________________________________________________________________

DOES SURGERY PROLONG SURVIVAL IN NON-SMALL CELL LUNG CANCER?
F.A. LEDERLE MD.
Dept of Medicine, VA Medical Center and University of Minnesota, Minneapolis, MN.

Surgical treatment for locally advanced non-small cell lung cancer (NSCLC) is based on the assumption that surgery is known to be effective for localized disease. The purpose of this discussion is to examine that assumption. Surgery is associated with improved survival in localized NSCLC, but whether this represents cause-and-effect or is the result of pre-existing differences between patients who do and do not receive surgery (i.e., the result of length, lead-time, and selection bias) remains unclear. Because of the complexity of the decision to perform surgery and the variable natural history of the disease, comparable groups of operated and unoperated patients can probably only be obtained through randomization (1). To assess the effectiveness of surgery for localized NSCLC, data from pertinent randomized trials are reviewed.

In a small trial comparing surgery with radiation therapy in 58 patients with localized lung cancer (2), survival was lower at 1 year and higher at 4 years in the surgery group, with neither difference achieving statistical significance and only 12 patients surviving to 2 years. A p<0.05 was incorrectly reported for patients with squamous cell carcinoma; the correct value is p=0.10, thus no statistically significant differences were observed.

A recent randomized comparison of lobectomy with limited (wedge or segmental) resection in 276 patients with T1N0 NSCLC provided an opportunity to look for a "dose-response" effect from surgery (3). Lobectomy significantly reduced local recurrence, but the reduction in overall mortality was not statistically significant.

Because a beneficial effect of surgery detected in locally advanced disease might reasonably be extrapolated to localized disease, studies in locally advanced disease are pertinent to this discussion. In one small study, 75 patients with N2 NSCLC who all received chemotherapy were randomized to also receive surgery or radiation therapy (4). There was no significant difference in the one-year or median survival between the two groups. In another study (SWOG-9019), 100 patients with selected Stage III NSCLC were randomized to receive chemoradiotherapy with or without surgery. The study closed in 1995 due to low recruitment, and no results have been reported to date.

Randomized trials of screening for lung cancer provide indirect evidence regarding the effectiveness of surgery. Five trials compared intensive lung cancer screening with less intensive screening in a total of 45,000 subjects and found no improvement in lung cancer mortality (5). Lung cancers detected in these trials were treated promptly by surgical resection whenever possible, thus no advantage was seen with early surgery compared with surgery performed later, often after the disease became symptomatic. The recent suggestion that the excess cancers detected by screening reflect a failure of randomization to produce comparable groups is possible but of low probability (about 1 in 800) (6). The accompanying suggestion that screening should be considered effective because of the improved survival of cancers detected by screening is, however, untenable because it a) effectively excludes from analysis all randomized patients who were not diagnosed with lung cancer, and b) invites length, lead-time, and overdiagnosis bias (7).

In addition to these completed trials, several other relevant studies are in progress. Two ongoing intergroup studies in the United States and Europe are randomizing N2 NSCLC patients to surgery vs. radiotherapy after induction chemoradiotherapy (8). Also, the US National Cancer Institute's Prostate, Lung, Colorectal and Ovarian Cancer (PLCO) Trial is randomizing 148,000 people aged 55-74 to receive usual care vs. annual screening (including a chest x-ray to detect lung cancer) for four years, followed by at least 10 years of follow-up (9). The study began in 1993, so final results are more than a decade away but results for lung cancer should be available sooner because of the high expected death rate.

Thus a number of randomized trials that could have demonstrated or implied a beneficial effect from surgery in localized NSCLC have failed to do so. While this does not prove surgery to be ineffective, it does challenge the standard argument that surgery is so clearly beneficial for localized NSCLC that a randomized trial would be unethical. It is also difficult to reconcile the two current beliefs that surgery is effective for NSCLC but screening is not. The trials currently in progress could provide convincing evidence that surgery is beneficial, but past experience suggests that negative results from these trials may not be persuasive. Whether surgery is beneficial for localized NSCLC can probably only be determined by a randomized trial. If such a study were undertaken, issues to be resolved include whether the alternative treatment arm should be radiation therapy, chemotherapy, or best supportive care, and whether enrollment should initially be limited to patients at increased operative risk. For decades, the main barrier to conducting this study has been the lack of support from expert physicians. Perhaps the renewed interest in evidence-based medicine will help to lower this barrier.

References
1. Lederle FA, Niewoehner DE. Lung cancer surgery: a critical review of the evidence.Arch Intern Med: 1994;154:2397-400.
2. Morrison R, Deeley TJ, Cleland WP. The treatment of carcinoma of the bronchus: a clinical trial to compare surgery and supervoltage radiotherapy. Lancet 1963;1:683-4.
3. Rubinstein LV, Ginsberg RJ. Lobectomy versus resection in T1N0 lung cancer (letter). Ann Thorac Surg 1996;62:1249.
4. Inculet A, Scott C, Dar R, et al. Phase III study comparing chemotherapy and radiation therapy with preoperative chemotherapy and surgical resection in patients with N2 NSCLC (RTOG 89-01). Lung Cancer 1997;18 (Suppl 2): abstract.
5. Richert-Boe KE, Humphrey LL. Screening for cancers of the lung and colon. Arch Intern Med 1992;152:2398-404.
6. Strauss GM, Gleason RE, Sugarbaker DJ. Screening for lung cancer: another look; a different view. Chest 1997;111:754-68.
7. Parkin DM, Piani P. Lung cancer screening (letter). Chest 1994;106:977.
8. Albain KS. Induction chemotherapy with/without radiation followed by surgery in Stage III non-small-cell lung cancer. Oncology 1997;11(Suppl. 9):51-7.
9. Prorok P. The National Cancer Institute multi-screening trial. Canad J Oncol, 1994; 4 (Suppl. l): 98- 101.

____________________________________________________________________________

PHASE II TRIAL OF HIGH-DOSE INDUCTION CHEMOTHERAPY FOLLOWED BY SURGICAL RESECTION AND RADIATION THERAPY IN MARGINALLY RESECTABLE STAGE IIIA NSCLC
M .GOLDBERG, R.J. SCHILDER, B. ROGERS, M.M. MILLENSON, C. DRESLER, S. JOHNSON, J.GALLO, D. KILPATRICK, C. YEUNG, K. HILLER, C.J. LANGER
Fox Chase Cancer Center, Philadelphia, Pennsylvania, 19111, USA

Most patients with regionally advanced NSCLC present with disease not amenable to initial curative resection. Of stage IIIA N2 patients only 10-20% present with "favorable" disease, defined as a normal mediastinum on CT of the chest or a single positive mediastinal node found at mediastinoscopy or thoracotomy. This group has a 40-50% two year survival and a 20-30% five year survival following complete resection, with or without adjuvant therapy. The remainder with stage IIIA disease have bulky mediastinal tumor visible on x-ray or extra-nodal spread. Currently they are not good candidates for initial surgical resection and traditionally have been treated with radical thoracic radiation alone, which though effective in achieving local tumor shrinkage and palliation of symptoms, produces a five year survival of 5-7%.

The poor survival rate of patients with stage IIIA bulky N2 disease is due to failure of radiotherapy to provide long-term local control and to the rapid development of metastases. Therefore, attention has focused on combined modality approaches with cytoreductive "induction" therapy administered prior to definitive locoregional treatment. The goals of "induction" therapy have been to convert unresectable to resectable disease, to eliminate distant micrometastases and to improve local control. Multiple phase II trials employing various drug regimens with or without radiation have demonstrated encouraging resectability and disease free survival rates.

Several groups have evaluated neoadjuvant chemotherapy for NSCLC. Memorial Sloan Kettering Cancer Center reported excellent results with mitomycin C, vindesine or vinblastine and cisplatin. 89 of 136 patients with bulky stage IIIA disease were able to undergo resection after chemotherapy with a 14% pathologic complete response rate (PCR). However, the Lung Cancer Study Group found no PCR after neoadjuvant cisplatin, mitomycin, and vinblastine in 31 patients. A Toronto phase II trial found a PCR rate of 7.7% in patients with cisplatin, mitomycin and vindesine and a median survival of 18.6 months.

Recently new agents have attempted to decrease toxicity and to improve response rate and survival. When carboplatin and cisplatin were each combined with etoposide, the carboplatin-based regimen was less toxic and produced an equivalent median survival. Paclitaxel has demonstrated the best response rate (21%) of all single agents in ten years of ECOG trials. The dose limiting toxicity is myelosuppression which is easily overcome by hematopoietic growth factors allowing doses up to 250 mg/m2. The non-hematologic toxicities which define the maximum tolerated dose of carboplatin are minimal up to approximately 2000 mg/m2 when supported by autologous bone marrow transplantation.

Recently, there have been reports of hematologic rescue from high dose chemotherapy using peripheral blood stem cells (PBSC). The technique offers numerous advantages over autologous bone marrow transplantation (ABMT)(1). Based upon data using high-dose carboplatin and paclitaxel (Schilder PASCO, 1996) we designed a phase II trial in patients with stage IIIA NSCLC. The objective was to determine the PCR rate and the survival rate of patients with locally advanced stage IIIA NSCLC treated with neoadjuvant high dose chemotherapy, surgical resection and postoperative radiation. All patients initially received etoposide (2g/m2) and G-CSF to mobilize and collect PBSC.

Two cycles, 28 days apart, of carboplatin (AUC=12 in 9 pts; AUC = 16 in 1 pt.) and paclitaxel (250mg/m2) were administered with G-CSF and PBSC support. After reevaluation, patients underwent radiotherapy (60Gy) or thoracotomy followed by radiotherapy (44Gy-60Gy). Three males and seven females entered the study after mediastinoscopy. The age range was 43 to 66 years with a median of 58.5 years. PS was 0 in 8 patients and 1 in two patients. Histology was adenocarcinoma in 7, squamous cell carcinoma in 1, broncheoalveolar carcinoma in 1 and large cell carcinoma in 1. Eight lesions originated in the RUL, one in the RML and one in the RLL. Based on CT scan the overall response rate was 40% (1CR, 3PR, 2MR, 2SD, 2PD). Eight of ten patients underwent resection with four right pneumonectomies, three RU lobectomies and one wedge resection of the RUL. Six patients had complete resections, two patients had incomplete resections and two patients had no resection. Of eight patients resected, four were downstaged by induction therapy, one was upstaged and three remained unchanged. The median dose of postoperative radiotherapy was 5400cGy. Follow-up in all patients has been from 25-122 weeks and a median of 70 weeks. To date, the median overall survival is 15.5 months and the median time to progression with recurrence is 9.6 months.

Currently three patients are NED (26, 46, 95 weeks), three are AWD (78, 100, 107 weeks) and three are DOD (43, 61, 122 weeks). Of those that are NED, all were adenocarcinoma, two were downstaged and one had remained stable after induction therapy. Causes of death in patients undergoing resection included pneumonectomy space recurrence, brain metastases, and pneumonectomy space infection. Neutropenic fever occurred in 2 pts during initial mobilization cycle only. The median units of RBC and/or platelets transfused were 0. There were no significant non-hematologic toxicities. High dose induction chemotherapy with stem cell rescue is feasible and safe with acceptable response rates. There appears to be a trend towards long-term benefit both in patients that had complete resections and in patients that had incomplete resections or no operative intervention.

References
1. Kessinger A. Reestablishing Hematopoiesis after Dose-Intensive Therapy with Peripheral Stem Cells. In: Armitage JO, Antman KH (eds): High Dose Cancer Therapy: Pharmacology, Hematopoietins, Stem Cells.
2. Baltimore,MD: Williams and Wilkins, 1992 : 182-194.

____________________________________________________________________________

SURGERY FOR P-STAGE-III, NON-SMALL CELL LUNG CANCER (NSCLC)
H. WADA, F. TANAKA, R. MIYAHARA, T. FUKUSE, S. HITOMI
Department of Thoracic Surgery, Chest Disease Research Institute, Kyoto University Kawaharacho, E-Mail address: wada @chest.kyoto-u.ac.jp (Hiromi Wada)

Key words: Non-small cell lung cancer (NSCLC), UFT, Adjuvant Chemotherapy, Biochemical modulation therapy (BCM), Cisplatin (CDDP)

1. Survival after surgery for p-stage III, non-small cell lung cancer (NSCLC)
The postoperative prognosis of NSCLC remains poor, and 5-year survival rates for all patients having undergone surgery for NSCLC have been reported to be only 20-35%; in p-stage IIIa and IIIb diseases, 5-year survival rates were only 15.1 - 44.2% and 0.0 - 8.2%, respectively [1,2]. Recently, in a retrospective analysis on the time trends of postoperative survival from 1976 through 1990 at Kyoto University, we have reported that the postoperative survival for p-stage IIIa has improved remarkably in the later years (5-year survival rates: 21.7% in the earlier period, 29.6% in the middle period, 49.5% in the later period, respectively). The improvement was caused by increase in the rates of complete tumor resection, decrease in the rates of operation-related death, and decrease in the ratio of pT3N2M0 disease with poor prognosis. In p-stage IIIb disease, 5-year survival rate remained to be only 23.1% even in the later period [3].

2. Postoperative adjuvant therapy for p-stage III, non-small cell lung cancer Although postoperative adjuvant therapy had been introduced to improve the prognosis, it had been concluded that radiation therapy or intensive chemotherapy was not effective as postoperative adjuvant therapy (Third IASLC Workshop on Therapy of NSCLC, Bruges, 1993). However, in a showment8], except for some. or NSCLCnt chemotherapyont. group (p = 0.019). The results of the subset-analysis showed that a better prognosis was obtained for p-stage III disease in the UFT group (5-year survival rate: 54.6%) as compared with the Cont. group (5-year survival rate: 16.7%, p = 0.014) [4]. To realize more favorite prognosis in p-stage IIIa, pN2 disease, we have introduced biochemical modulation chemotherapy (BCM) using UFT combined with CDDP as postoperative adjuvant therapy. The regimen is as follows: CDDP 2mg/m2/day, continuous intravenous infusion for 4 weeks UFT 300 or 400mg/day, oral administration for more than 1 year. Among 20 patients who received BCM after surgery for p-stage IIIa, pN2 NSCLC, no patient have died with 2 years after surgery of maximum follow-up interval.

3. Preoperative induction therapy for p-stage III, non-small cell lung cancer. Although it is proved that postoperative administration of UFT improve the prognosis of completely resected p-stage IIIa disease, postoperative prognosis of pT3N2M0 or p-stage IIIb disease remained to be poor. To improve these prognosis, many regimens have been tried as preoperative induction therapy, but the efficacy has not established yet.

We have recently introduced BCM using 5-FU combined with CDDP for advanced NSCLC shown as follows:

CDDP 6mg/m2/day, intravenous infusion day 1 - 5
5-FU 150mg/m2/day, intravenous infusion day 1 - 7
UFT 300 or 400mg/day , oral administration day 1 - 7

In early phase II study of BCM with or without concurrent radiation for c-stage IIIb or IV NSCLC, BCM without radiation proved to be effective without any severe side effects (response rate : 50.0%) [5].

Following the early phase II study, we have introduced the BCM regimen without radiation as preoperative induction therapy for c-stage IIIa, N2 or IIIb NSCLC. Thirteen patients (10 males and 3 females, average age: 58.5 years, 2 with c-stage IIIa, N2, 8 with c-stage IIIb, T4, and 2 with c-stage IIIb, N3 disease) received BCM followed by operation. Partial response (PR) was achieved in 6 patients (response rate: 46.2%); no progressive disease (PD) was seen. Severe side effect caused by BCM or postoperative complication was not seen in any case. One-year survival rate is 77.8%, and the mean survival time is 815 days. BCM may be an effective and safe way as preoperative induction therapy for advanced NSCLC, although long-term follow-up of the prognosis is needed to establish the efficacy.

References
1. Mountain CF. Chest 1986; 89: 225s-33s.
2. Naruke T, et al. J Thorac Cardiovasc Surg, 1988; 96: 440-7
3. Wada H, et al. J Thorac Cardiovasc Surg 112:349-355, 1996
4. Wada H, et al. J Clin Oncol 14:1048-1054, 1996
5. F.Tanaka, et al. Lung cancer 18s: 33, 1997

____________________________________________________________________________

LONG-TERM FOLLOW-UP IN 46 PATIENTS WITH IIIA (CLINICAL N2) NON-SMALL CELL LUNG CANCER AFTER INDUCTION CHEMOTHERAPY- THE IMPACT OF ANGIOGENESIS ON METASTASES AND SURVIVAL
S. DARWISH*, V. MINOTTI*, G. FONTANINI°, V. DE ANGELIS*, M. BETTI*, A. SIDONI, L. CRINÒ*, M. TONATO*
* Div. Medical Oncology, Perugia; ° Pathology, Pisa, Pathology, Perugia – Italy

Lung cancer is the most common cause of cancer-related death in the United States and in developed countries among both men and women, with an estimated 187,000 new cases and 179,000 deaths in 1997 in the USA. Prognosis for locally advanced non-small cell lung cancer (NSCLC) patients is poor. Treatment with surgery or radiation therapy is unsatisfactory and 5-years survival rates of less than 10% are reported for these patients. According to the staging system introduced in 1986 and revised in 1997 locally advanced disease is divided into stage IIIA and IIIB. Whereas stage IIIB is considered categorically unresectable, stage IIIA is potentially resectable but rarely curable. In an attempt to increase the complete resection and survival rates in stage IIIA disease, several pilot trials have explored multiple therapeutic modalities including induction chemotherapy.

From June 1988 to July 1991, 46 patients with locally advanced unresectable NSCLC stage III A (clinical N2) received an induction chemotherapy (I CT) with cisplatin and etoposide (PE) regimen, 39 male, 7 females, median age 58 years (range 45-70), median PS 90. The main cell-type was squamous cell carcinoma 74%, followed by 22% adenocarcinoma and 4% anaplastic large-cell carcinoma. Fourty-five pts were evaluable for response. Thirty-Seven of 45 pts (82%) had a major response after 2-3 cycles PE: 3 (7%) complete response and 34 (75%) partial response. Eight pts were considered non surgical-candidates. Thirty-five underwent surgical exploration and 33 pts (73%) were resected. Twenty-eight of them achieved a complete resection ( 62% of all pts / 80% of surgically explored pts. In 4 pts (2 CR, 2 PR) no tumour was found in surgical specimen. Of the 28 completely resected pts, 11 had recurrences: 5 locoregional and 6 distant. To date 38 pts have died, among them there are all unresectable pts. The median survival of the entire 46 pts was 24 months with a 6 years survival of 19%. Of 9 pts who entered the study 7 or more years ago, 7 survived for at least 7 years, one relapsed in the brain and 6 (13%) remain disease- free and enjoy a normal functional status.

We tried to verify the impact of angiogenesis on the metastases and survival in completely resected pts. Tumour angiogenesis was assessed in 22 of 28 completely resected pts counting microvessels (capillaries and venules) in the most active area of neovascularization by immunohistochemical staining of vascular endothelium for factor VIII- related antigene and grading their density scale from 1 to 4+, using light microscopy. Individual microvessels were counted on 200 X field and each count was expressed as the highest number of microvessels indentified within this area.

Microvessel-count correlated significantly with metastatic disease: "median±SD": 19(±9.6) and 10 (±6.5) for metastatic and non metastatic disease respectively, p= 0,02.

Blood vessel invasion (BVI) was positive in 5 pts (4 metastatic pts, 1 non metastatic pt) and negative in 17 pts (7 metastatic pts, 10 non metastatic pts). Microvessel count correlated significantly with BVI (median±SD): 22 ± 9.3 and 12.1 ± 8.3 in pts with positive BVI and negative BVI respectively, p= 0.03.

Microvessel-count in 8 alive pts was not significantly different with respect to 14 dead pts (median ± SD): 11.2 ± 7 and 17.5 ± 10.6, was respectively (p= N.S.).

Conclusion
- A reasonable number (13%) of completely resected patients, after response to induction chemotherapy still remains alive for 6 years and enjoys a normal functional status.
- In univariate analysis, the likelihood of metastases and blood vessel invasion became greater as the vessel count increased .
- Survival was not correlated significantly with micovessel- count

Bibliography
1. Mountain CF. A new International staging system for lung cancer. Chest 1986; 89(S):225S-33S.
2. Mountain CF. Revision in the Interntional System for Staging Lung cancer. Chest 1997; 111:1710-17.
3. Mountain CF: Biological, physiologic and technical determinants in surgical therapy for lung cancer. Straus M. (ed.): lung cancer, clinical diagnosis and treatment. New York, Grune and Stratton, 185-198, 1977.
4. Martini N, Kris MG, Flehinger BJ, Gralla RJ, Bains MS, Burt ME, Heelan R, McCormack PM, Pisters KMW, Rigas JR, Rusch VW, Ginsberg RJ: Preoperative chemotherapy for stage IIIa (N2) lung cancer: the Sloan Kettering experience with 136 patients. Ann Thorac Surg 1993. 55: 1365-1374.
5. Martini N. Flehinger BJ: The role of surgery in N2 lung cancer. Surg. Clin North Am 67:1037-1049. 1987.
6. Gralla RJ. Preoperative and adjuvant chemotherapy in non-small cell lung cancer. Sem in Oncol. vol. 15 ,N6 suppl. 7 ; 1988.
7. Darwish S, Minotti V, Crino L, Rossetti R, Maranzano E, Checcaglini F, Fiaschini P, Mercati U, Penza O, Vitali R, Davis S, Latini P, Tonato M: Neoadjuvant cisplatin and etoposide for stage IIIA (clinical N2) non-small cell lung cancer. Am J Clin Oncol 1994, 17: 64-67.
8. Weidner N, Semple JP, Welch WR, Folkman J:Tumor angiogenesis and metastasis- Correlation in invasive breast carcinoma.N Engl J Med 1991, 324: 1-8.
9. Fantanini G, Bigini D, Vignati S, BasoloF, Mussi A, et al: Microvessel count predicts metastatic disease and survival in non-small cell lung cancer.J Pathol, 1995, 177: 57-63.
10. Macchiarini P, Fontanini G, Hardin MJ, Squartini F, Angeletti CA: Rapporti tra neovascolarizzazione e metastasi del cancro polmonare non a piccole cellule. Lancet 1992, 340: 145-46.

____________________________________________________________________________

LONG-TERM RESULTS OF A SURGICALLY BASED MULTIMODALITY TREATMENT FOR LUNG CANCER
V. BELTRAMI, R. SACCO and E. MASCITELLI
Clinica Chirurgica Generale dell'Università "D'Annunzio", Chieti

Introduction
Our general case material for this study derives from a series of nearly 4.000 patients with bronchogenic carcinoma seen at our university clinic up to December 1997. A surgically-based multimodality treatment was applied when possible and the long-term results were studied in patients operated on 10, 5 and 3 years previously. Our impression is that the results could improve in the next years, due to the increasing use of induction therapy prior to lung resection.

Satistical analysis
We surgically treated 1507 patients with a lung cancer, selected from a total of 3.994 cases we studied. In the general series, the age ranged from 30 to 85 years, most cases belonging to the 61-70 years group; females were only 9 %. 2.299 cases were considered inoperable and 188 refused surgery; the percentage of exploratory procedures was 21 % before 1985, 7 % from 1986 to 1990 and below 3 % after 1991. We resected 1.232 cases , 1.169 NSCLC and 63 SCLC; 326 received some form of neoadjuvant treatment. Staging procedures were TC with biopsy , transbronchial needle aspiration and mediastinoscopy or VATS. Radiation (30 Gy in daily fractions of 2) was used in 46 patients with some chest-wall involvement; chemotherapy (cisplatin, epirubicin and VP-16 in two cycles with a three-week interval) was preferred in 172, for centrally located tumors and in N2 cases. Restaging was used to select eligible patients: a resection became possible in 49 % of III A and in 40 % of III B cases.

In the general series, 765 classic lobectomies and bilobectomies, 46 sleeve lobectomies and 84 segmental or sub-segmental resections were performed: minor resections by VATS are not included. Pneumonectomies were 337: in 19 some plastic procedure was applied to the carina and 147 were performed with an intrapericardial approach to the pulmonary artery and veins; an atrium suture was necessary in 8 patients. A pericardiectomy was also performed 21 times, a chest wall pleurectomy 72 times, 19 patients underwent resection of more than one rib and 10 showed involvement of the diaphragm.

61,60 % of the resected patients had a histological diagnosis of squamous carcinoma, 28,90 % of adenocarcinoma; large cell anaplastic tumors were observed in 4,20 % of cases and SCLC in 5,30 %. Tumor stages were 198 T1N0, 355 T2N0, 217 T3N0; the remaining stage II patients were 104. IIIA patients with positive N were 197 and IIIB 44; in 47 stage IV patients a previous or a sinchronous metastatectomy was performed. All operated SCLC were localized. The general mortality rate was around 5,5 % .

Patients submitted to previous induction treatment, cases of SCLC, patients who had surgery extended to the chest wall, the pericardium or the diaphragm, as well as all cases with positive nodes were eligible for an adjuvant Radiotherapy and/or Chemotherapy.

Long-term survival
Only 21 patients were lost to follow-up. A high percentage (around 55 %) of late deaths was related to the development of metastases; local recurrences were not common (20 %). A true second primary lung cancer (SPLC) was observed 23 times and 15 of such patients were successfully re-operated.

Long-term survival was calculated for all patients operated on up to 1987 (10 years) and 1992 (5 years): patients who had received induction treatment were also evaluated for possible 3-years survival and resected cases with a SCLC were considered separately. More than 72 % of the patients resected for a T1N0 NSCLC fared well after 5 years, while this figure was 46 % after 10 years. Survival after resection for T2N0 disease was 51 and 24 % , respectively. After 5 years, 31 % of T3N0 patients were alive, but after 10 years only 17 % of them survived. Patients with N2 disease proved to have a poor prognosis, but those having received previous induction treatment had a better chance: 45 % of them were alive after 3 years and 28 % after 5 years. Patients operated on for SCLC proved also to have some better chance of survival after induction and adjuvant Chemotherapy : 32 % of our cases who received such a multimodal treatment were alive at 5 years.

____________________________________________________________________________

IS THERE OPERABLE STAGE III NSCLC? THE ROLE OF PET SCANNING
K. VIRIK1, J. DUSSEK2, M. MALSEY3, D. YIP1, P. HARPER1
Departments of Medical Oncology 1, Cardiothoracic Surgery 2 & Clinical PET, Centre 3 Guy’s and St.Thomas’ Hospital Trust, London, U.K.

The optimum management of patients with NSCLC is dependant upon accurate staging. Clinical Stage I & II patients have a 40% 5 year survival with standard surgical resection.1,2 However, up to 70% of patients present with advanced - stage III & IV - disease. Despite the higher stage, accurate staging within this group is no less significant in terms of its influence on survival. The 5 year survival rates for locoregionally advanced disease may be up to 30%3, whereas in those with distant metastatic disease, 5 year survivals are anecdotal.

Critical to optimising appropriate therapy in locally advanced disease is the accurate assessment of operability. In general, N3 disease (contralateral mediastinal lymph node metastases) is considered inoperable. Standard non-invasive staging investigations to elucidate nodal involvement include CT and MRI. The imaging result is defined by tissue morphology and hence subject to limitations. Normal sized regional lymph nodes can be histologically involved with disease and enlarged nodes can be histologically benign.4

Positron emission tomography (PET) scanning is a non invasive technique of assessing cellular metabolic activity.

Malignant cells differ from normal mammalian cells in that they utilise anaerobic glycolysis in the production of energy. Since the energy (ATP) production is the same for both neoplastic and normal cells, the metabolic rate is necessarily higher in cancer cells in order to "compensate" for the relative inefficiency of anaerobic glycolysis.5

18F (fluoro-2-deoxy-D-glucose) is the positron emitting radioisotope most commonly used to measure metabolic activity in lung cancer.

A review of the published literature together with our hospital’s experience, suggests that PET scanning is "the way forward" in terms of staging modalities with a real potential of leading to improved patient outcome.

PET imaging modified the therapeutic outcome in more than 20% of patients in a prospective study comparing the accuracy of whole body FDG PET and conventional imaging. It correctly changed the M stage in 10% of patients.6 Sensitivity and specificity for the diagnosis of N2 disease were 83% and 94% for PET compared to 63% and 73% respectively for CT in a prospective evaluation of 99 patients.7 A retrospective study with 32 NSCLC patients found PET to be 80% sensitive, 100% specific and 87.5% accurate in staging thoracic lymph nodes cf CT scanning which was 50% sensitive, 75% specific and 59.4% accurate. The authors opinion was that PET appeared reliable in separating N1/N2 disease from N3.8

A recent prospective study with 47 patients utilised lymph node mapping correlated with extensive nodal sampling at thoracotomy, mediastinoscopy or both as the standard of reference in evaluating diagnostic accuracy of nodal staging between PET and CT. PET was found to correctly designate the N stage in 96% of cases and CT in 79% of cases.9 The authors correlated the amount of FDG uptake in a lesion with the amount in the brain in contrast to most other reports wherein FDG uptake was compared with uptake in the mediastinal blood pool. Their superior results suggest that refining PET parameters may well result in greater accuracy of assessment of nodal involvement.

Of intrigue, is the concept of image fusion of PET and CT allowing for the minimisation of the limitations inherent individually. This area appears ripe for further evaluation.

References
1. Ihde DC: Chemotherapy of lung cancer. N Eng J Med 327:1434-1441,1992.
2. Ihde DC, Minna JD: Non-small cell lung cancer, Biology, diagnosis and staging. Curr Probl Cancer 15:61-104, 1991.
3. Martini N, Flehinger BJ: The role of surgery in N2 lung cancer. Surg Clin North Am 67:1037-1049, 1987.
4. McLoud TC et al: Bronchogenic carcinoma: analysis of staging in the mediastinum with CT by correlative lymph node mapping and sampling. Radiology 182:319-323, 1992.
5. Hughes JMB: 18F-Fluorodeoxyglucose PET scans in lung cancer.Thorax 51(S2):S16-S22, 1996.
6. Bury T et al: Staging of non-small-cell cancer by whole-body fluorine-18 deoxyglucose positron emission tomography. Eur J Nuc Med 23:204-206, 1996.
7. Valk PE et al: Staging non-small cell lung cancer by whole-body positron emission tomographic imaging. Ann Thorac Surg 60:1573-1582, 1995.
8. Guhlmann A et al: Lymph node staging in non-small cell lung cancer: evaluation by (18F) FDG positron emission tomography (PET). Thorax 52: 438-441, 1997.
9. Steinert HC et al: Non- small cell lung cancer: Nodal staging with FDG PET versus CT with correlative lymph node mapping and sampling. Radiology 202:441-446, 1997.

____________________________________________________________________________

SUCCESSFUL TREATMENT OF LOCALLY ADVANCED NSCLC; WHERE WE HAVE BEEN, WHERE WE ARE GOING AND WHY WE WON’T GET THERE
J.C. RUCKDESCHEL, M.D., H. WAGNER, JR., M.D. and L.A. ROBINSON, M.D.
H. Lee Moffitt Cancer Center and Research Institute, Tampa, Florida, USA

Locally advanced NSCLC has been a therapeutic conundrum for several decades. Modality oriented trials of surgery alone or radiation alone have produced just enough long term survivors (~5%) to keep up the faintest hopes for both the patient and the physician.

Most therapeutic approaches have lacked intellectual rigor. The 1970s and 1980s saw several trials of post-operative chemotherapy or radiation. Despite reducing local recurrences, the routine application of post-operative radiation had no impact on survival. Post-operative cisplatin-based chemotherapy had a marginal (~4%) impact on survival but it took a meta-analysis to demonstrate the effect.

There were, however, virtually no trials of post-radiation chemotherapy despite the fact that the "evidence" for initial cytoreduction with radiation is at least as compelling as the "evidence" for initial cytoreduction by chemotherapy. The hypothesis that delivering the chemotherapy 6-8 weeks earlier makes a substantive impact on micrometastatic disease flies in the face of everything we know about lung cancer biology. Few patients with locally advanced NSCLC are cured by local control only, but it is equally true that few are cured without local control. Yet it is primarily local control that we have improved upon. Even the Intergroup repeat of the CALGB trial has failed to offer compelling evidence for a lasting "systemic" effect. The bid radiation arm at 3 years is now no different than the chemo-radiation arm mimicking the results of the earlier post-operative adjuvant trials. The current chemotherapeutic regimens for NSCLC appear superior to earlier regimens and there are several trials under way combining them sequentially with radiation (all prior to the radiation). Progress will, however, be only incremental.

Pre-operative, or neo-adjuvant therapy with chemotherapy alone or combined chemo-radiation has become nearly standard therapy in many North American cities for patients with potentially resectable NSCLC. The evidence for this approach, however, is far from finalized. Virtually all of the work in this area is in the form of phase II trials with little or no information on staging. The handful of ?positive,? randomized trials appear to be confined to patients with minimal N2 disease. We have been seduced by the extremely high response rates (?70%) and the remarkable survival statistics for the few patients who have complete histologic clearance of their tumors with the pre-operative therapy. There are, to date, no compelling trials of pre-operative therapy for more than minimal N2 disease.

It is the lack of an adequate staging system and our unwillingness to use even that system which will condemn us to at least another decade of inconclusive studies. The recent "modification" of the International Staging System corrected a few minor issues in the surgical approach to NSCLC but ignored the deficits in stage III disease. At a minimum the staging system needs to recognize the following categories of stage III disease.

We firmly believe that bi- or tri-modality therapy is superior to uni-modal therapy for all except stage IA and IV disease. We do not, however, have the data to make this view compelling.

____________________________________________________________________________

COMPARABLE FACTS OF IMPORTANCE FOR THE TREATMENT OF PATIENTS WITH NON-SCLC AS WELL AS PATIENTS WITH SCLC
K. KARRER and G. ANTYPAS
BUON-ISC-LCSG, Institute for Cancer Research, University Vienna, Austria and METAXA Cancer Hospital, Piraeus, Greece

One of the controversial questions of the therapeutic options for the locally advanced lung cancer is the indication for surgery for cure at the TNM-stage IIIA.

The therapeutic limitations of surgery are well recognized, and future progress to enhance the therapeutic effectiveness of surgery is necessary to increase the proportion of curable or at least long term surviving patients (1,2,3).

Our first hypothesis (4) seems still valid in view of the bulk of results of different chemotherapies used with increasing efficacy.

Adjuvant chemotherapy should be able to add chances for a long term survival from surgery for cure, not replacing the still generally accepted method of choice for the treatment of any malignant solid tumor, by complete surgical resection of the primary tumor, together with it's lymph drainage area (radical surgery with curative intent), with the stage of the tumor as the major determinant of the cure rate.

It seems constructively important, that at the most recent meeting at Geneva in May 1997, the UICC-TNM-committee accepted the subdivision of pTN2-patients to stage IIIB, if clinically detectable tumor involvement or bulky N2 disease is determined, whereas patients with only patho-histologically detectable just beginning tumorcell infiltration in one of the N2-lymphnodes belong to TNM-stage III A. The inaccuracy of clinical staging of the potentially surgical candidates does not only interfere in the treatment of SCLC, but also for non-SCLC, in particular for the local advanced stage-III patients. More so are the histologic subtypes and heterogeneity sources of controversy, as well as the presence of a varying percentage of non-small tumor cell types in association with SCLC (5).

The attempts to convert marginally resectable and non-resectable lesions into resectable ones, by the use of the so called neo-adjuvant (initial) chemotherapy, only further confuse the issue as to the proper role of surgical resection with this diseases. Last not least, the lack of sufficient numbers of potential surgical candidates has prevented the conduction of satisfactory randomized clinical trials of medical therapy alone versus the addition of surgical resection to a satisfactory pre- or postoperative medical regimen to resolve the question. This latter circumstance has been the major stumbling block for many in the acceptance of any role of surgical resection in the management of SCLC as well as for stage III of non-SCLC (6).

In consequence, from resulting relatively high numbers (125) of long term survivors (over 30 months p.o.) from 335 patients receiving surgery for cure and postoperative adjuvant chemotherapy within the ISC-studies I/II and V/VI, in the newly designed ongoing enlarged multinational cooperative BUON-ISC-LCSG-studies unselected operable patients would become randomized to receive pre- or postoperative chemotherapy to try to demonstrate whether surgery or chemotherapy is more potent to increase the chances of long term survival with SCLC. The stage-related operability would be defined by the best available diagnostic methods to exclude cTN2-tumors. It is emphasized, to subdivide patients with completely resected N2-tumors at pTN2R0-stages in group A as the very beginning of tumor involvement, which can become detected only pathohistologically from several sites of the operation specimens. And group B with several lynphonodes with larger tumor involvement, which can cause similar metastases as the primary tumor.

References
1. Shields TW, Matthews M: Adjuvant Therapy of Small Cell Bronchial Carcinoma. Cancer Treat.Rev. 1984;11:331-333.
2 . Meyer JA at al: Adverse prognostic effects of N2 disease in treated small cell carcinoma of the lung. J Thorac Cardiovasc Surg 1 984;88:495-501.
3. Ginsberg RJ, Karrer K: Surgery in small cell lung cancer: a consensus report. Lung Cancer 1989;5:139.
4. Denk W., Karrer K. - Modellversuch einer Rezidivprophylaxe des Karzinoms. Wien Klin Wochenschr 1955; 67:986.
5. Fraire AE et al: Lung Cancer heterogeneity: prognostic implications Cancer 1981; 47: 1863-66.
6. Shields TW and Karrer Karl: Surgery for Small Cell Lung Cancer, chapter 7 in LUNG CANCER 2.e. editors Toth, Cox, Hong, in print 1997 by Blackwell Science, Inc., Malden, MA, USA.
7. Karrer K, Shields TW at al: The importance of surgical and multimodality treatment for small cell bronchial carcinoma. J Thorac Cardiovasc Surg 1989;97: 168-176.
8. Schamanek A, Karrer K et al: Importance of surgery in the multimodality treatment for small cell lung cancer (SCLC). Radiol Oncol 1994; 28 :351-8.
9. Karrer K et al: Surgery for cure followed by chemotherapy in small cell carcinoma of the lung. Acta Oncol 1995;34:899-906.
10. Karrer K et al: Results of surgery in small cell lung cancer. P 369-374 in Balkan Congress of Oncology, Athens/Greece July 3-7, 1996. Editor George Antypas, by Monduzzi Editore S.p.A. Bologna, Italy 1996.