Increased expression of sialyl Lewisx antigen correlates with poor survival in patients with colorectal carcinoma: clinicopathological and immunohistochemical study.

We have previously shown that sialyl Lewisx antigen (sLex) (NeuAc alpha 2-3Gal beta 1-4(Fuc alpha 1-3)GlcNAC-R) has an important functional role in defining the invasion and metastasis of human colorectal carcinoma. The results were derived from the clinical specimens obtained at surgery or experimental metastasis of human colon carcinoma variant expressing different levels of sLex in nude mice. In the present study, we immunohistochemically examined 132 human colorectal carcinomas for the expression of sLex to investigate whether this antigen expression could serve as a prognostic parameter. The tumors were divided into two groups: sLex positive and sLex negative. The incidence of sLex positive was correlated with the depth of tumor invasion, the presence of the lymph node metastasis, lymphatic invasion, and the disease stage. The difference was statistically significant (P = 0.0026; P = 0.0002; P = 0.003; P = 0.0013; respectively). Based on the data on 114 patients who underwent curative resections, incidence of the disease recurrence was assessed. The sLex-positive patients had higher incidence of recurrence in distant organs, especially in the liver, than that of the sLex-negative patients. The 5-year disease free survival rates of sLex-positive and -negative patients were 57.7 and 89.1%, respectively (P = 0.0002). The difference of 5-year overall survival rates between the two were also significant (sLex positive, 58.3%; sLex negative, 93.0%: P < 0.0001). By Cox multivariate analysis, sLex expression levels remained the best discriminant of disease-free survival (P = 0.035) and overall survival (P = 0.0081). These results suggest that increased expression of sLex is correlated with the extent of malignancy and high incidence of recurrence and consequently with survival of colorectal carcinoma patients. Thus sLex may prove to be a potent marker of recurrence in colorectal carcinoma patients.

Expression of carbohydrate antigen sialyl Le(a): a new functional prognostic factor in gastric cancer.

PURPOSE: The prognostic value of the altered expression of carbohydrate antigens sialyl Le(a) (sLe(a)) and sialyl Le(x) (sLe(x)), which have been implicated as functional ligands in heterotypic-cell-adhesion systems in the multistep process of tumor metastasis, were evaluated. PATIENTS AND METHODS: The level of expression of sLe(a) and sLe(x) antigens was examined immunohistochemically in paraffin-embedded tumor samples from 137 patients who underwent resection for gastric cancer. Correlation between the antigens' expression, various established clinicopathologic factors, and prognosis were studied by univariate and multivariate analysis. RESULTS: Tumors that were positive for the sLe(a) antigen were significantly more likely to be large (P = .035), to be localized at the proximal third of the stomach (P = .018), to have an infiltrate appearance (P = .013), to have an invasive mode both in depth of invasion (P = .028) and in lymphatic invasion (P = .002), and to be classified as late stage (P = .011) than those that were negative for sLe(a), whereas the sLe(x) antigen status was not correlated with any clinicopathologic factors. The overall survival of patients with an sLe(a)-antigen-positive tumor was significantly poorer than that of those with an sLe(a)-antigen-negative tumor (P = .0001). Survival within each pathologic stage differed also (stage I, P = .030; stage II, P = .046; stage III, P = .026, respectively). A Cox regression analysis with multiple covariates showed that positive sLe(a) antigen status was an independent prognostic factor for a worse outcome in patients with gastric cancer. According to the mode of recurrence, increased sLe(a) antigen expression significantly affected both peritoneal dissemination and liver metastasis. CONCLUSION: Increased expression of the sLe(a) antigen may serve as a potent prognostic indicator for recurrence in patients with gastric cancer. Careful follow-up and intensive therapy are required for patients with an sLe(a)-antigen-positive gastric cancer.

Expression of N-acetylglucosaminyltransferase V in colorectal cancer correlates with metastasis and poor prognosis.

N-Acetylglucosaminyltransferase V (GnT-V) is an enzyme that catalyzes beta 1-6 branching of N-acetylglucosamine on asparagine-linked oligosaccharides of cell proteins. Metastatic potential of various cancer cells has been shown to correlate with increase of GnT-V activity and concomitant beta 1-6 branching of N-acetylglucosamine. However, protein expression of GnT-V in human cancer tissue and its clinical significance have not yet been demonstrated. To clarify the possible relationship between metastasis and GnT-V in human colorectal cancer, protein expression of GnT-V was studied using surgically resected specimens. We established a monoclonal antibody against GnT-V and performed immunohistochemical analysis of 103 human colorectal cancer cases. Of 103 cases, 26 cases (25.2 %) showed specific expression of GnT-V in colorectal cancer tissues. The expression of GnT-V was significantly correlated with distant metastasis (P < 0.05, chi2 test). Overall 5-year survival rate was 52.8% for GnT-V-positive patients and 81.7% for GnT-V-negative patients (P < 0.01, Log-rank test). We showed direct evidence for the relationship between GnT-V and metastasis in human colorectal cancer. Screening of GnT-V expression in colorectal cancer may provide useful information for prognosis of postoperative patients.

Phosphodiesterase type III inhibitor, cilostazol, inhibits colon cancer cell motility.

Metastasis of cancer cells is initiated by the cellular migration into extracellular matrix and surrounding vessels. We previously showed that elevation of cAMP levels in cancer cells suppressed trans-cellular migration in vitro. Drugs that can elevate cAMP levels in cancer cells effectively may be applied to prevent metastasis in cancer patients. Cilostazol, an oral anti-platelet drug, is a specific cAMP phosphodiesterase type III inhibitor and has been clinically used to treat thrombosis patients. In chemotaxis assay, cellular migration of human colon cancer cells, DLD- 1, was induced by 10 microg/ml of soluble fibronectin or 10% of fetal bovine serum (FBS). Treatment with cilostazol (50 microM) suppressed 92.3% or 84.6% of the migration in control cells, respectively. When DLD-1 cells were stimulated by soluble fibronectin in phagokinetic assay, migration assessed by the area of gold particle phagocytosis track was induced and cilostazol also decreased 67.3% of the cellular migration in control cells. Furthermore, in the trans-cellular migration assay, cilostazol suppressed cancer cell invasion induced by FBS. Thus, cilostazol can suppress colon cancer cell motility and might be effective as an anti-metastasis drug for cancer patients.

Myoepithelioma of the lung: report of two cases and review of the literature.

Myoepithelial tumors occur mainly in the salivary glands, the sweat glands or the breast, but uncommonly in the lung. Herein, we describe two cases of myoepithelioma of the lung. Both patients were 58-year-old men, in whom the tumors were located in the right-upper bronchus and in the left-upper bronchus, respectively, with endobronchial growth pattern. Surgery was performed, but metastasis occurred into the forearm and hip muscles in the former case, and into the liver in the latter. Histologically, the tumor in the former was a spindle-plasmacytoid type, and that in the latter was a plasmacytoid type in part with squamous differentiation. Based on histochemical, immunohistochemical and ultrastructural analyses, both were compatible with myoepithelioma. The clinicopathological uniqueness of this neoplasm is discussed, together with a review of reports of this disease in the literature.

Application of sentinel node biopsy to gastric cancer surgery.

BACKGROUND: Sentinel node (SN) biopsy has been tried in the management of a variety of cancers with the hope that it would eliminate many unnecessary lymph node dissections, resulting in less morbidity. This important technique, however, has not been tried in gastric cancer surgery. The feasibility of SN biopsy and its accuracy in predicting the lymph node status in patients with gastric cancer were examined in the current study. PATIENTS AND METHODS: SN biopsy was performed in patients with T1 (n = 44) or T2 (n = 30) gastric cancers (ie, immediately after laparotomy, indocyanine green was injected around the primary tumor, and the green-stained nodes [SNs: 2.6 +/- 1.7 nodes per patient] were removed). Then, gastrectomy with extended lymphadenectomy was performed. The unstained nodes (non-SNs: 39 +/- 18 nodes per patient) were obtained from the resected specimens. Both SNs and non-SNs were subjected to histologic examination with hematoxylin-eosin. RESULTS: SNs could be identified in 73 of 74 patients (success rate, 99%). Of these 73 patients, 10 had lymph node metastases in SNs or non-SNs, or both; 6 in both SNs and non-SNs; 3 in SNs alone; and 1 in non-SNs alone. The sensitivity of the SN status in the diagnosis of the lymph node status of the patient was 90% (9/10) and specificity was 100% (63/63). Sensitivity was 100% in the T1 group (n = 44) and 88% in the T2 group (n = 29). CONCLUSIONS: SN biopsy using indocyanine green can be performed with a high success rate, and the SN status can predict the lymph node status with a high degree of accuracy, especially in patients with T1 gastric cancer.

Practical grouping of positive lymph nodes in pancreatic head cancer treated by an extended pancreatectomy.

BACKGROUND: Until recently long-term survival has not been expected when at least one positive node was detected at any site in pancreatic head cancer treated by conventional pancreatectomy. However, even when nodal involvement is seen, there has been an increasing number of long-term survivors after extended pancreatectomy in which a wide range of lymphatic and connective tissues were cleared. Thus the purpose of the present study was to establish a practical rational grouping of positive lymph nodes in pancreatic head cancer treated by extended pancreatectomy. METHODS: In 81 patients who tolerated extended pancreatectomy for cancer of the pancreatic head, a mean of 56 +/- 23 (range, 28 to 89) lymph nodes in each patient were examined under a microscope to determine the presence or absence of cancer. They were classified anatomically into 14 lymph node groups, and the incidence, distribution, and number of positive nodes were examined. A simplified grouping was made on the basis of the histologic findings and was checked against long-term survival rates. RESULTS: Nodal involvement was detected in 59 (73%) of 81 patients, and positive nodes were more commonly observed in the posterior pancreaticoduodenal (PPD), superior mesenteric (SM), and anterior pancreaticoduodenal (APD) groups than in the 11 other groups (p < 0.05). The PPD, APD, and SM groups offered the sole sites of nodal involvement with incidence levels of 23%, 17%, and 6%, respectively, whereas none of the 11 other groups did. Thus patients were classified into four groups: (a), negative in all 14 lymph node groups (n = 22); (b), positive but limited to the PPD/APD groups (n = 14); (c), also positive in the SM group, but negative in the 11 other groups (n = 13); and (d), also positive in at least one of the 11 other groups (n = 32). This classification was associated well with the 5-year survival rate: 59% in group (a), 53% in group (b), 15% in group (c), and 0% in group (d) [p < 0.05; group (b) versus group (c)]. Also this grouping associated well with the total number of positive nodes (p < 0.05). The 5-year survival rate in patients with one to three positive nodes was 47% and was more than 6% in patients with four to seven positive nodes (p < 0.05). CONCLUSIONS: In the clinicopathologic staging of the lymphatic spread from carcinoma of the pancreatic head, the PPD and APD groups were considered the first stations of lymphatic metastasis, whereas the 12 other groups-including the SM group-were categorized as second or more distant stations.

Pancreatic invasion as the prognostic indicator of duodenal adenocarcinoma treated by pancreatoduodenectomy plus extended lymphadenectomy.

UNLABELLED: Pancreatoduodenectomy has become the standard procedure in resection of the duodenal adenocarcinoma, and some adjuvant therapies can be added to obtain further improvement in postoperative outcome. However, for patient selection, it is necessary to have a predictive indicator showing, if possible before laparotomy, which instances are noncurable by surgery alone or need adjuvant therapies. METHODS: A retrospective analysis was made for 24 consecutive patients whose duodenal adenocarcinoma were treated by pancreatoduodenectomy plus a wide range of lymphadenectomies without any adjuvant therapies at Osaka Medical Center for Cancer and Cardiovascular Diseases. Patient survival rates were related to macroscopic and microscopic findings and to findings obtained by preoperative imaging techniques. RESULTS: The overall survival rate was 69% at 3 years and 57% at 5 years; locoregional recurrence was the primary cause of death. Although the 5-year survival rate was 44% in patients with nodal involvement and 76% in those without, this difference did not reach statistical significance (P = .079). Instead, invasion into the pancreatic parenchyma at a macroscopic level was the most significant prognostic factor; the 5-year survival rate was 78% in the 16 patients without and 16% in the 8 patients with pancreatic invasion (P = .0047). Invasion into the pancreas correlated well with the angiographic findings; the 5-year survival rate was 25% in patients whose angiograms delineated the pancreatic invasion and 83% in patients whose angiograms did not (P = .0084). CONCLUSION: When duodenal adenocarcinoma was treated by pancreatoduodenectomy plus a wide range of lymphadenectomy, pancreatic invasion at a macroscopic level was most associated with patient survival. Pancreatic invasion was well delineated by the preoperative angiogram, which would be helpful in patient selection.

Does pleural lavage cytology before thoracic closure predict both patient's prognosis and site of cancer recurrence after resection of esophageal cancer?

BACKGROUND: Operative manipulation occasionally exfoliates and spreads cancer cells in the surgical field, and it is a matter of concern whether the exfoliated cancer cells actually affect the patient's prognosis and sites of cancer recurrence. METHODS: In 240 patients with esophageal cancers, lavage cytology (LC) of the right pleural cavity was performed before and after esophageal resection combined with regional lymphadenectomy. The cytologic results were compared with the pathologic factors associated with cancer extension, postoperative survival, and cause of surgical failure. RESULTS: Only 3 patients (1.3%) were LC positive before resection. Of the 237 LC-negative patients, LC was also negative after resection in 215 patients (90.7%) (LC-/-), but LC became positive after resection in 22 patients (9.3%) (LC-/+). The 3-year survival rate was 0% in the LC-/+ group versus 65% in the LC-/- group, and the median survival rates were 10.9 months and 25.0 months, respectively (P <.0001). Multivariate analysis revealed that LC-/+ was an independent prognostic factor (P =.0331), along with nodal involvement and depth of cancer invasion. However, there were no significant differences in the sites of cancer recurrence between the 2 groups. Only 1 patient was found to develop the first recurrence in the pleural cavity. The LC-/+ group had a higher incidence of bulky lymph-node metastasis (P =.0009). CONCLUSIONS: Pleural LC after resection of esophageal cancer seems to be a prognostic indicator of overall recurrence, but not necessarily in the pleural cavity. Patients with a positive LC after resection may benefit most by effective systemic adjuvant chemotherapy.

The prognostic significance of surgical margin in liver resection of patients with hepatocellular carcinoma.

OBJECTIVE: To evaluate the prognostic significance of surgical margin in liver resection of patients with hepatocellular carcinoma. DESIGN: Retrospective study. SETTING: The Center for Adult Diseases, Osaka, Japan, between 1980 and 1989. PATIENTS: One hundred eighty-five patients with hepatocellular carcinoma who underwent liver resection with complete extirpation. MAIN OUTCOME MEASURE: Cumulative survival rate. RESULTS: The patients were divided into a wide surgical margin group, in which the lesion was excised with a margin of 1.0 cm or more, and a narrow surgical margin group, in which the margin was less than 1.0 cm. No significant differences could be detected in survival rates for 3 years or longer. Mean +/- SE tumor sizes were 3.4 +/- 0.4 cm and 4.4 +/- 0.3 cm, respectively, in the wide and narrow surgical margin groups. The patients were divided into three groups according to tumor size: group 1, 2.0 cm or less in diameter; group 2, greater than 2.0 cm but 5.0 cm or less in diameter; and group 3, greater than 5.0 cm in diameter. In groups 2 and 3, no significant differences in survival rates were found between the wide and narrow surgical margin groups. In group 1, the survival rate was significantly higher in the wide surgical margin group than in the narrow surgical margin group (P < .05). CONCLUSIONS: Small hepatocellular carcinomas of 2.0 cm or less in diameter should be resected with an adequate surgical margin. However, surgical margin was not a significant factor in the resection of hepatocellular carcinomas larger than 2.0 cm.

Liver perfusion chemotherapy via both the hepatic artery and portal vein to prevent hepatic metastasis after extended pancreatectomy for adenocarcinoma of the pancreas.

Since hepatic metastasis is a common cause of treatment failure after curative pancreatectomy for adenocarcinoma of the pancreas, we developed a new method of postoperative hepatic perfusion chemotherapy via both the hepatic artery and portal vein. The present study was conducted to determine if this method decreases the hepatic recurrence and improves the survival rate. Following extended pancreatectomy with wide lymphatic and connective tissue clearance for pancreatic cancer, one catheter was placed in the hepatic artery and one in the portal vein. Immediately after surgery, 5-fluorouracil (125 mg/d) was continuously infused via these two routes simultaneously for 28 to 35 days. There were no treatment-related complications in the 20 patients who survived surgery. The 3-year survival rate was 54%, and the cumulative rate of death from hepatic metastasis was 8%. These figures were significantly better than those of our historical control groups. We conclude that this method should be evaluated in a prospective, randomized controlled study.

Adjuvant therapies in extended pancreatectomy for ductal adenocarcinoma of the pancreas.

BACKGROUND/AIMS: The poor prognosis after pancreatic cancer resection is attributed to a high incidence of locoregional recurrence and hepatic metastasis. The present study was carried out in order to determine whether adjuvant therapies improve patient survival. METHODOLOGY: Preoperative irradiation and postoperative liver perfusion chemotherapy have been added to the extended pancreatectomy procedure, and a retrospective analysis of 111 patients was performed. RESULTS: Preoperative irradiation (n = 17) significantly decreased the incidence of locoregional recurrence but increased the incidence of hepatic metastasis, with no improvement in the 5-year survival rate (from 27% to 23%). In contrast, postoperative liver perfusion chemotherapy (n = 27) succeeded in improving the 5-year survival rate to 39% by significantly decreasing the incidence of hepatic metastasis. CONCLUSION: These data suggest that the patients' survival rate may be increased when these two types of adjuvant therapy are combined with extended pancreatectomy.

Regional chemotherapy to prevent hepatic metastasis after resection of pancreatic cancer.

BACKGROUND/AIMS: Liver metastasis is one of the major causes of cancer death after resection of pancreatic cancer. To deal with this problem, we developed a liver perfusion chemotherapy via the hepatic artery and portal vein. The present paper is designed to introduce the technique of this method and report on the promising results. METHODOLOGY: Between 1987 and 1995, 98 consecutive patients with adenocarcinoma of the pancreas received an extended pancreatectomy, in which a wide range of lymphatic and connective tissue clearance was performed in addition to the conventional pancreatectomy, at Osaka Medical Center for Cancer and Cardiovascular Diseases. All patients were followed-up to determine whether and where cancer recurrence developed. RESULTS: Excluding 4 patients who died of postoperative complications, our liver perfusion chemotherapy was performed on 27 patients. The other 67 patients did not receive this treatment. In the group that underwent liver perfusion chemotherapy, no patients experienced complications such as leucocytopenia (< 3000/mm3), thrombocytopenia (< 50,000/mm3), or liver disfunction. The cumulative survival rate differed remarkably between the two groups: 92% vs 62% at one year (p < 0.05); 51% vs 35% at three years (p < 0.05); and 41% vs 25% at five years (ns). CONCLUSIONS: When this method was performed after an extended pancreatectomy for adenocarcinoma of the pancreas, our preliminary report (Am J Surg. 1994; 168:361-364) indicated that it was useful not only for preventing hepatic metastasis, but also for improving patients' survival rate. Since then, further positive and supportive results have been obtained in an additional investigation.

Treatments for second malignancies after gastrectomy for stomach cancer.

BACKGROUND/AIMS: Metachronous primary cancers after stomach cancer were reported as a cause of late death. It is necessary to treat these second malignancies to improve the result of surgical treatment for gastric cancer. MATERIAL AND METHODS: Among 3,570 patients who underwent gastrectomy during 1961-1986, 160 second primary malignancies were observed as of December of 1991. Liver (27 patients) and colorectal (18 patients) cancers were the largest in number. RESULTS: By the end of 1992, for 32 liver tumors small resection, transcatheter arterial embolization and percutaneous ethanol injection were carried out and 3-year survival of resected cases (6 patients; 80.0%) was better than that of nonresected cases (26; 3.8%) (p < 0.05). In 19 colorectal cancers, surgical resection (13/19) and combined locoregional chemotherapy (2/19) were actively performed, and 5-year survival of resected cases (13; 55.4%) was better than that of nonresected cases (6; 0%) (p < 0.05). CONCLUSION: With an intensive long-term follow-up after gastrectomy for cancer, early diagnosis and effective therapy for second primary malignancies is improved.

Palliative surgical treatment for recurrent and non-resectable hepatocellular carcinoma.

Of 275 patients who had a radical hepatic resection for HCC, 143 (52%) experienced recurrences. Of these 143 patients, the liver was the site of first recurrence in 130 patients (91%). The first form of therapy for recurrent patients was transarterial chemoembolization (TACE) in 82 (74%), hepatic resection in 15 (14%) and percutaneous ethanol injection in 6 (5%) of the 111 patients who received regional therapy. Post-recurrent survival rates of TACE and hepatic resection were 37%, 77% at 3 years and 17%, 77% at 5 years, respectively. Wrapping therapy was performed in seven patients, two of whom had received repeat TACE for intrahepatic recurrence; the remaining 5 were unresectable cases. After wrapping therapy, the area of previous collateral feeders was be supplied by the hepatic artery. The high level of AFP decreased dramatically. This procedure resulted in a complete response in 2 patients, a partial response in 3 and no change in 2. Overall survival was 4-54 months, and the median survival was 18 months. Two patients are still alive, 18 and 20 months after the procedure.

Necrosis of hepatocellular carcinoma caused by spontaneously arising arterial thrombus.

This paper describes a 65-year-old Japanese man with hepatocellular carcinoma (HCC) in whom the alpha-fetoprotein level decreased remarkably without any treatment. Plain computed tomography disclosed a low-density area in the left lateral segment. Liver scintigraphy revealed a filling defect with 99mTc-Sn colloid and increased uptake of 67Ga-citrate. The latter was smaller in area than the former. This indicates that non-necrotic HCC was still present at this time. There was no hypervascular lesion in the hepatic angiogram obtained 22 days after liver scintigraphy. The tumor was resected by partial hepatectomy 24 days after hepatic angiography. The histological section showed almost complete necrotization of the tumor, and the necrotic change consisted of old and recent necrosis. An arterial thrombus was formed in non-tumor liver tissue. It was presumed that coagulative necrosis was produced by interruption of the blood supply due to the spontaneous formation of an arterial thrombus.

A new method of intra-arterial regional chemotherapy with more selective drug delivery for locally advanced pancreatic cancer.

BACKGROUND/AIMS: In order to deliver the anti-cancer drugs more selectively into the cancer tissues, we have developed a new method of intra-arterial regional chemotherapy for locally unresectable cancer of the exocrine pancreas. MATERIALS AND METHODS: This method involved placing the catheters selectively into the splenic artery and/or into the gastroduodenal artery during laparotomy. Postoperatively, via the catheter, we infused 50-100 mg of Methotrexate mixed with 10 micrograms of Angiotensin-II with an intent of increasing the blood flow in the tumor tissue but decreasing that to the non-tumor tissues. Simultaneously, a bolus iv-infusion of 5-fluorouracil (5-Fu, 500 mg) was performed. One day after each chemotherapy, citrovorum factor (Leucovorin, 30mg) was given per orally, For 15 patients with locally non-resectable pancreatic cancer, this treatment was repeated weekly or biweekly at our out-patient clinic. RESULTS: As a result, the toxicity was so slight that all patients could tolerate this treatment as long as the catheter was patent (11 +/- 8 postoperative months). The survival period was 16 +/- 9 months (range: 5-36 months; median: 14 months), and one-, two- and three- year survival rates were 60%, 23% and 11%, respectively. Patients could take care of themselves within 12 +/- 9 months, and either complete (50%) or partial (50%) pain-relief was obtained among the 12 patients with severe pain. Only one patient experienced local tumor regression during the chemotherapy, and the incidence of liver metastasis was as low as 13%. CONCLUSIONS: Comparing with the previously reported data in the traditional chemo- and/or radio-therapies, we consider that our method of intra-arterial chemotherapy is quite useful not only for the prolongation of patient's survival but also for improving the quality of life. Thus, this new treatment seems worthy of entering into the prospective randomized study.

Fine needle aspiration cytology of malignant schwannoma metastatic to the breast.

A case of malignant schwannoma metastatic to the breast is reported. The tumor was found as a solitary breast lump in a 43-year-old woman and was clinically considered to be a primary tumor. Fine needle aspiration (FNA) smears of the mass were interpreted cytologically as a malignant neurogenic tumor, compatible with a malignant schwannoma. The patient did not have multiple neurofibromatoses. Eight months previously, an excision of a solitary nodule located in the left axilla had been performed in another hospital. Comparison of the histologic section from that excision with the FNA confirmed that the breast tumor was a metastatic malignant schwannoma, and the patient was treated accordingly.

Survival and postoperative complications after extended surgery for non-small-cell lung cancer. A retrospective study.

OBJECTIVE: We evaluated survival after extended surgery in patients with non-small-cell lung cancer and the effects of induction therapy on results and complications. SUBJECTS AND METHODS: Between April 1987 and March 1998, 127 patients with pathological T3 (pT3) or T4 (pT4) non-small-cell lung cancer underwent extended surgery combined with resection of neighboring organs. Of these, 35 received induction therapy. In the remaining 92, surgery preceded other therapy. Long-term results and postoperative respiratory complications were analyzed and compared between the patients with and without induction therapy. RESULTS: Overall 5-year survival after extended surgery was 37%. Five-year survival rates in the pT3 was 41% and that in the pT4 group 28% (p = 0.030). Five-year survival rate in the pN0-1 was 46% and that in the pN2-3 group 26% (p = 0.003). No significant difference was observed in survival curves between patients with and without induction therapy. Induction therapy responders showed better survival than nonresponders. To prevent postoperative fatal complications due to bronchopleural fistula, we prophylactically covered the bronchial stump using autologous tissue in 31 induction therapy patients, and no mortality due to complications was seen in this group. CONCLUSION: Long-term survival after extended surgery was observed in pT3 and pT4 patients, especially among those with a pN0-1 status. Induction therapy responders may be considered good candidates for extended surgery because of the favorable prognosis in contrast to that for nonresponders.

Transmediastinal approach to exploring the lung contralateral to the thoracotomy site.

OBJECTIVE: The approach to contralateral lung through the mediastinum is assumed useful in managing oddly distributed bilateral lung tumors. SUBJECTS AND METHODS: To remove a tumor located in the contralateral lung, a transmediastinal approach from the thoracotomy site to the contralateral lung was used in 6 patients having oddly distributed bilateral lung tumors, 1 of which was located in the contralateral lung close to the anterior or posterior mediastinum. RESULTS: All cases were treated successfully. One patient required an additional small incision on the contralateral anterior chest wall to insert an endoscopic stapler without intraoperative postural change. The postoperative course was uneventful and, to date, no local recurrence has been seen at the resected margin of the contralateral lung. CONCLUSION: This novel approach is useful, offering the advantages of reduced invasiveness and pain, shorter surgical duration, and favorable cosmetic results for patients with a tumor close to the mediastinum in the contralateral lung.