Lymphatic Vessel Invasion As a Prognostic Factor in Patients With Primary Resected Adenocarcinomas of the Esophagogastric Junction
http://www.100md.com
《临床肿瘤学》
the Departments of Surgery and Pathology, Klinikum rechts der Isar, Technical University of Munich, Munich, Germany
ABSTRACT
PATIENTS AND METHODS: We prospectively evaluated 459 patients undergoing primary surgical resection for tumors of the esophagogastric junction at our institution between 1992 and 2000 (180 adenocarcinomas of the distal esophagus, AEG I; 140 carcinomas of the cardia, AEG II; and 139 subcardial gastric cancers, AEG III). Median follow-up was 36.8 months. The prevalence of LVI was evaluated by two independent pathologists. Univariate and multivariate analysis of prognostic factors was performed.
RESULTS: The total rate of LVI was 49.9%, with a significant difference between AEG I (38.9%) and AEGII/III (57.0%, P = .0002). Univariate analysis showed a significant correlation between LVI and T category (P < .0001), N category (P < .0001), and resection status (R [residual tumor] category; P < .0001). This was shown for the group of all AEG tumors, as well as for the subgroups AEG I and AEG II/III. On multivariate analysis, LVI was identified as a significant and independent prognostic factor (P = .050) in the population of all patients and in patients with AEG II/III, but not in the subgroup with AEG I.
CONCLUSION: These data demonstrate the prognostic significance of LVI in patients with AEG tumors, with marked differences between the subgroups AEG I versus AEG II/III. The lower prevalence and lack of prognostic significance of LVI in AEG I might be explained by inflammation involved in the pathogenesis of this entity.
INTRODUCTION
Three entities are (according to internationally accepted criteria) subsumed under the term adenocarcinoma of the esophagogastric junction (AEG): adenocarcinoma of the distal esophagus (AEG I), true carcinoma of the cardia (AEG II), and subcardial gastric carcinoma (AEG III14,15; Fig 1).
The objective of this study was to clarify the relation between well-defined histopathologic prognostic factors and the presence of LVI in patients with AEG tumors, and to determine the prognostic significance of LVI.
PATIENTS AND METHODS
Surgical Approach
The primary goal of the surgical approach was complete resection of the primary tumor and its lymphatic drainage. An individualized surgical strategy was applied as previously described16-18:
Patients with AEG I were treated with transhiatal or transthoracic esophagectomy with proximal gastric resection and lymphadenectomy in the lower posterior mediastinum, as well as the upper abdominal compartment. After esophagectomy, gastrointestinal continuity was restored by gastric pull up, and in a few cases with prior gastric surgery, by colon interposition.
Patients with AEG II/III received an extended total gastrectomy with transhiatal resection of the distal esophagus and en bloc removal of the lymphatic drainage in the lower posterior mediastinum, along the celiac axis and superior border of the pancreas (D2 lymphadenectomy16,17). After gastrectomy, reconstruction was made with an esophagojejunostomy Roux-en-Y.
The different surgical strategy between AEG type I tumors and AEG type II/III tumor is basis for the subgroup analysis (AEG I v AEG II/III).
Tumors treated in a neoadjuvant setting with chemotherapy followed by surgical resection16,17 were excluded from this study, as only primary resected cases were analyzed.
No routine postoperative (adjuvant) therapy was applied. All patients were examined preoperatively according to published criteria19 and were considered clinically fit for surgery.
Histopathologic Examination
For fixation of the specimens, a 4% formaldehyde solution was used. Incubation time was 24 hours. The complete tumor was cut in slices of 0.5 cm. The resection margins were marked with ink. All lymph nodes were dissected and analyzed. The tissue was paraffin embedded. Serial sections of 5-μm thickness were cut and stained with hematoxylin-eosin and van Gieson.
All specimens were analyzed by two independent, experienced pathologists. The tumors were classified according to the TNM system of classification of the International Union Against Cancer (UICC),20 including R (residual tumor) category and grading (G). For AEG I tumors, the TNM system classification for esophageal tumors was used. AEG II/III tumors were classified using the TNM classification of gastric tumors, because the UICC provides no separate classification for adenocarcinomas of the esophagogastric junction.20
LVI was defined as tumor cell spread through the lymphatic vessels (ie, carcinoma cells floating within the endothelial-lined space). LVI was separated from venous invasion on the basis of the features of smooth muscle and elastic fibers, and the presence or absence of valvulae as defined by Shirouzu et al.1
Minor (LVI1) and extensive (LVI2) LVI was distinguished as shown in Figure 2. Fewer than three sites of LVI per low-power field (LPF) were regarded as "minor LVI," whereas more than three sites of LVI per LPF were classified as "extensive LVI."
In all of the 459 patients of the study population, there was a complete agreement between the two pathologists regarding the presence or absence of LVI on blinded and independent review of the slides; in 19 of 459 patients, there was difference in the estimation of the severity of LVI, but this was resolved unanimously on subsequent review of the slides, with both pathologists present. In five additional patients, there was a disagreement on the presence (n = 2) or degree (n = 3) of LVI, which could not be resolved. These five patients were excluded from the subsequent statistical analysis and are not part of the reported patient population (N = 459).
Follow-Up
Follow-up was performed by the oncologic outpatient clinic and by personal contact with the patient, their general practitioner, and/or their relatives by letter and/or by phone call. The final follow-up date was December 31, 2002, to reach at least 1 year of follow-up time in all patients. The median follow-up was 36.8 months (range, 12 to 120 months). To date, follow-up is complete for 95.8% (440 of 459 patients).
Statistical Analysis
Survival rates were calculated according to the Kaplan-Meier method, including 95% CIs, and tested with the log-rank test.21 Postoperative 30-day and hospital mortality were not excluded from this calculation.
Differences in proportion of patients were analyzed with a {chi}2 test. All tests were two sided, and a P value of less than .05 was considered significant.
Death of a patient was chosen as an indisputable and objective end point for the multivariate analysis of prognostic factors. To reach more independent and accurate results, we performed the Cox multiple regression analysis.22 Tested variables included "depth of tumor infiltration" (four categories: PT0/1 v PT2 v PT3 v PT4), absence or presence of lymph node metastases (two categories: PN0 v PN+), residual tumor category after resection (two categories: R0 v R+), and absence or presence of LVI (2 categories: LVI– v LVI+).
All analyses were performed using the statistical package SPSS for Windows (release 11.0; SPSS Inc, Chicago, IL).
RESULTS
In the subgroup of patients with primary resected AEG I tumors, LVI was demonstrated in 70 (38.9%) of 180 patients. The prevalence of LVI in the AEG II/III subgroup was a markedly higher 57.0% (159 of 279)—a statistically significant difference (P = .0002, {chi}2 test). The results of the histopathologic work-up and the subclassification into minor and extensive LVI are presented in Table 3.
LVI correlated with PT, PN, and R categories, with a higher prevalence in more advanced cases (higher PT category, PN+, R+ cases). Results of the prevalence of LVI and its correlation with PT, PN, and R are given in Table 4.
A substantial amount of LVI was found in early cancers (PT1) and in cases without nodal involvement (negative PN status). LVI was found in 11.6% of the patients with early (PT1) AEG I tumors and in 13.5% of the patients with AEG II/III tumors. The node-negative cases showed LVI in 12.5% (AEG I) and 24.7% (AEG II/III).
The results of the survival analysis are shown as Kaplan-Meier curves (Fig. 3, 4, and 5). On univariate analysis, the presence and extent of LVI was a significant prognostic factor for the entire patient population, as well as for the subgroups AEG I and AEG II/III.
Multivariate analysis (Cox regression) revealed prognostic significance and independence as determinant of survival in the group of all AEG tumors (P = .050; Table 5) and in the subgroup of AEG II/III tumors (P = .048; Table 6). In contrast to this finding, LVI lost its prognostic significance in the subgroup of patients with AEG I tumors.
DISCUSSION
LVI has previously been shown to be a strong and independent prognostic factor in esophageal squamous cell cancers7-10 and gastric cancers,5,6 but its role in distal esophageal adenocarcinomas and adenocarcinomas of the esophagogastric junction has not been proven so far. LVI is included in the TNM system as "L" category, but its mention in the histopathology report is optional.
LVI is thought to precede or occur coincidently with lymph node metastasis. This is reflected by the correlation between LVI and N and T categories in our study. Our analysis clearly shows that LVI is associated with more advanced disease, as patients with more advanced tumors have a higher prevalence of LVI.
Nevertheless, patients with early cancers (T1) and cancers without nodal involvement (N0) showed a substantial amount of LVI. This reflects the aggressiveness of the AEG tumors. Patients with early local tumor growth often have lymphatic tumor spread, additionally to locally limited disease.
Multivariate analysis confirmed that the impact of LVI is independent from other prognostic markers in the group of all patients with AEG. In the analysis of the subgroups, LVI lost its prognostic significance in patients with AEG I: LVI is not independent from other strong prognostic factors (T, N, and R). By contrast, in the subgroup of patients with AEG II/III, LVI remains a strong and independent prognostic factor. Rather surprisingly, macroscopically and microscopically complete tumor resection (R0) and the T category lost prognostic significance in the subgroup of patients with AEG II/III tumors when LVI was included in the multivariate model. Thus, LVI seems to be the leading and independent prognostic factor (next to frank lymphatic metastases) in AEG II/III tumors, but not in AEG I tumors.
One possible explanation for the discrepancies between AEG I and AEG II/III is the markedly different pathogenetic mechanisms of these tumor entities.23,24 Development of AEG I follows a metaplasia–intraepithelial neoplasia–carcinoma sequence, which is promoted by the chronic damaging effect of gastroesophageal reflux.24 One major feature of this disease process is chronic inflammation (Fig 6). This might result in stepwise degradation of lymphatic vessels, and thus, a lesser degree of LVI in AEG I tumors. In contrast, chronic inflammation and reflux has so far not been shown to play a role in AEG II/III tumors.
These results support a distinction of esophageal carcinomas according to the tumor type: esophageal adenocarcinomas and squamous cell cancers differ in respect to tumor biology, etiology, and pathogenesis.18 Our results show that LVI is of minor importance and prognostic impact in distal esophageal adenocarcinomas (AEG I), in sharp contrast to esophageal squamous cell cancers, where it is a strong and independent prognostic factor, as previously shown.7 These results support the distinction of these markedly different entities in the TNM staging system.
In summary, our data stress the importance of a systematic histopathologic work-up and the inclusion of LVI in the histopathology report. The results of this study support the inclusion of LVI into the TNM staging system of the UICC.20 The thus far optional L category should gainobligatory character, especially in AEG II/III tumors, in which it is a strong and independent prognostic factor.
Authors' Disclosures of Potential Conflicts of Interest
NOTES
Supported by Deutsche Krebshilfe (grant No. 70-2789-Si3).
Authors' disclosures of potential conflicts of interest are found at the end of this article.
REFERENCES
1. Shirouzu K, Isomoto H, Morodomi, et al: Carcinomatous lymphatic permeation: Prognostic significance in patients with rectal carcinoma—A long term prospective study. Cancer 75:4-10, 1995
2. Minsky BD, Mies C, Rich TA, et al: Lymphatic vessel invasion is an independent prognostic factor for survival in colorectal cancer. Int J Radiat Oncol Biol Phys 17:311-318, 1989
3. Zamcheck N, Doos WG, Prudente R, et al: Prognostic factors in colon carcinoma: Correlation of serum carcinoembryonic antigen level and tumor histopathology. Hum Pathol 6:31-45, 1975
4. Baur M, Schlappack O, Jaskez R, et al: [Differential prognosis of patients with breast cancer with pleuritis carcinomatosa and with pulmonary lymphangiosis carcinomatosa.] Wien Klin Wochenschr 104:714-716, 1992
5. Gabbert HE, Meier S, Gerharz CD, et al: Incidence and prognostic significance of vascular invasion in 529 gastric-cancer patients. Int J Cancer 49:203-207, 1991
6. Gabbert HE, Meier S, Gerharz CD, et al: Tumor-cell dissociation at the invasion front: A new prognostic parameter in gastric cancer patients. Int J Cancer 50:202-207, 1992
7. Brücher BL, Stein HJ, Werner M, et al: Lymphatic vessel invasion is an independent prognostic factor in patients with a primary resected tumor with esophageal squamous cell carcinoma. Cancer 92:2228-2233, 2001
8. Torres CM, Wang HH, Turner JR, et al: Pathologic prognostic factors in oesophageal squamous cell carcinoma: A follow –up study of 74 patients with or without preoperative chemoradiation therapy. Mod Pathol 12:961-968, 1999
9. Watanabe M, Kuwano H, Araki K, et al: Prognostic factors in patients with submucosal carcinoma of the oesophagus. Br J Cancer 83:609-613, 2000
10. Sarbia M, Bittinger F, Porschen R, et al: Tumor vascularization and prognosis in squamous cell carcinoma of the esophagus. Anticancer Res 16:2117-2121, 1996
11. Brechot JM, Chevret S, Charpentier MC, et al: Blood vessel and lymphatic vessel invasion in resected nonsmall cell lung carcinoma: Correlation with TNM stage and disease free and overall survival. Cancer 78:2111-2118, 1996
12. Ota S, Muro H: [The prognostic significance of lymphatic vessel and blood vessel invasion in peripherally located adenocarcinoma of the lung] Nippon Rinsho 60:276-280, 2002 (suppl 5)
13. Ikehara M, Oshita F, Kameda Y, et al: Expression of survivin correlated with vessel invasion is a marker of poor prognosis in small adenocarcinoma of the lung. Oncol Rep 9:835-838, 2002
14. Siewert JR, Hlscher AH, Becker K, et al: [Cardia cancer: Attempt at a therapeutically relevant classification]. Chirurg 58:25-34, 1987
15. Siewert JR, Stein HJ: Classification of carcinoma of the oesophagogastric junction. Br J Surg 85:1457-1459, 1998
16. Stein HJ, Feith M, Siewert JR: Cancer of the esophagogastric junction. Surg Oncol 9:35-41, 2000
17. Stein HJ, Feith M, Siewert JR: Individualized surgical strategies for cancer of the esophagogastric junction. Ann Chir Gynaecol 89:191-198, 2000
18. Siewert JR, Feith M, Werner M, et al: Adenocarcinoma of the esophagogastric junction. Results of surgical therapy based on anatomical/topographic classification in 1,002 consecutive patients. Ann Surg 232:353-361, 2000
19. Bartels H, Stein HJ, Siewert JR: Preoperative risk analysis and postoperative mortality of oesophagectomy for resectable oesophageal cancer. Br J Surg 85:840-844, 1998
20. Sobin LH, Wittekind Ch: UICC: TNM Classification of Malignant Tumors (ed 6). New York, NY, Wiley-Liss, 2002
21. Cox DR: Regression models and life tables. J R Stat Soc 34:1987-2001, 1972
22. Kaplan EL, Meier P: Nonparametric estimation from incomplete observations. J Am Stat Assoc 75:457-487, 1958
23. Stein HJ, von Rahden BHA, Hfler H, et al: [Carcinoma of the oesophagogastric junction and Barrett esophagus: An almost clear oncologic model] Chirurg 74:703-708, 2003
24. von Rahden BHA, Stein HJ, Siewert JR: Barrett's esophagus and Barrett's carcinoma. Curr Oncol Rep 5:203-209, 2003(Burkhard H.A. von Rahden,)
ABSTRACT
PATIENTS AND METHODS: We prospectively evaluated 459 patients undergoing primary surgical resection for tumors of the esophagogastric junction at our institution between 1992 and 2000 (180 adenocarcinomas of the distal esophagus, AEG I; 140 carcinomas of the cardia, AEG II; and 139 subcardial gastric cancers, AEG III). Median follow-up was 36.8 months. The prevalence of LVI was evaluated by two independent pathologists. Univariate and multivariate analysis of prognostic factors was performed.
RESULTS: The total rate of LVI was 49.9%, with a significant difference between AEG I (38.9%) and AEGII/III (57.0%, P = .0002). Univariate analysis showed a significant correlation between LVI and T category (P < .0001), N category (P < .0001), and resection status (R [residual tumor] category; P < .0001). This was shown for the group of all AEG tumors, as well as for the subgroups AEG I and AEG II/III. On multivariate analysis, LVI was identified as a significant and independent prognostic factor (P = .050) in the population of all patients and in patients with AEG II/III, but not in the subgroup with AEG I.
CONCLUSION: These data demonstrate the prognostic significance of LVI in patients with AEG tumors, with marked differences between the subgroups AEG I versus AEG II/III. The lower prevalence and lack of prognostic significance of LVI in AEG I might be explained by inflammation involved in the pathogenesis of this entity.
INTRODUCTION
Three entities are (according to internationally accepted criteria) subsumed under the term adenocarcinoma of the esophagogastric junction (AEG): adenocarcinoma of the distal esophagus (AEG I), true carcinoma of the cardia (AEG II), and subcardial gastric carcinoma (AEG III14,15; Fig 1).
The objective of this study was to clarify the relation between well-defined histopathologic prognostic factors and the presence of LVI in patients with AEG tumors, and to determine the prognostic significance of LVI.
PATIENTS AND METHODS
Surgical Approach
The primary goal of the surgical approach was complete resection of the primary tumor and its lymphatic drainage. An individualized surgical strategy was applied as previously described16-18:
Patients with AEG I were treated with transhiatal or transthoracic esophagectomy with proximal gastric resection and lymphadenectomy in the lower posterior mediastinum, as well as the upper abdominal compartment. After esophagectomy, gastrointestinal continuity was restored by gastric pull up, and in a few cases with prior gastric surgery, by colon interposition.
Patients with AEG II/III received an extended total gastrectomy with transhiatal resection of the distal esophagus and en bloc removal of the lymphatic drainage in the lower posterior mediastinum, along the celiac axis and superior border of the pancreas (D2 lymphadenectomy16,17). After gastrectomy, reconstruction was made with an esophagojejunostomy Roux-en-Y.
The different surgical strategy between AEG type I tumors and AEG type II/III tumor is basis for the subgroup analysis (AEG I v AEG II/III).
Tumors treated in a neoadjuvant setting with chemotherapy followed by surgical resection16,17 were excluded from this study, as only primary resected cases were analyzed.
No routine postoperative (adjuvant) therapy was applied. All patients were examined preoperatively according to published criteria19 and were considered clinically fit for surgery.
Histopathologic Examination
For fixation of the specimens, a 4% formaldehyde solution was used. Incubation time was 24 hours. The complete tumor was cut in slices of 0.5 cm. The resection margins were marked with ink. All lymph nodes were dissected and analyzed. The tissue was paraffin embedded. Serial sections of 5-μm thickness were cut and stained with hematoxylin-eosin and van Gieson.
All specimens were analyzed by two independent, experienced pathologists. The tumors were classified according to the TNM system of classification of the International Union Against Cancer (UICC),20 including R (residual tumor) category and grading (G). For AEG I tumors, the TNM system classification for esophageal tumors was used. AEG II/III tumors were classified using the TNM classification of gastric tumors, because the UICC provides no separate classification for adenocarcinomas of the esophagogastric junction.20
LVI was defined as tumor cell spread through the lymphatic vessels (ie, carcinoma cells floating within the endothelial-lined space). LVI was separated from venous invasion on the basis of the features of smooth muscle and elastic fibers, and the presence or absence of valvulae as defined by Shirouzu et al.1
Minor (LVI1) and extensive (LVI2) LVI was distinguished as shown in Figure 2. Fewer than three sites of LVI per low-power field (LPF) were regarded as "minor LVI," whereas more than three sites of LVI per LPF were classified as "extensive LVI."
In all of the 459 patients of the study population, there was a complete agreement between the two pathologists regarding the presence or absence of LVI on blinded and independent review of the slides; in 19 of 459 patients, there was difference in the estimation of the severity of LVI, but this was resolved unanimously on subsequent review of the slides, with both pathologists present. In five additional patients, there was a disagreement on the presence (n = 2) or degree (n = 3) of LVI, which could not be resolved. These five patients were excluded from the subsequent statistical analysis and are not part of the reported patient population (N = 459).
Follow-Up
Follow-up was performed by the oncologic outpatient clinic and by personal contact with the patient, their general practitioner, and/or their relatives by letter and/or by phone call. The final follow-up date was December 31, 2002, to reach at least 1 year of follow-up time in all patients. The median follow-up was 36.8 months (range, 12 to 120 months). To date, follow-up is complete for 95.8% (440 of 459 patients).
Statistical Analysis
Survival rates were calculated according to the Kaplan-Meier method, including 95% CIs, and tested with the log-rank test.21 Postoperative 30-day and hospital mortality were not excluded from this calculation.
Differences in proportion of patients were analyzed with a {chi}2 test. All tests were two sided, and a P value of less than .05 was considered significant.
Death of a patient was chosen as an indisputable and objective end point for the multivariate analysis of prognostic factors. To reach more independent and accurate results, we performed the Cox multiple regression analysis.22 Tested variables included "depth of tumor infiltration" (four categories: PT0/1 v PT2 v PT3 v PT4), absence or presence of lymph node metastases (two categories: PN0 v PN+), residual tumor category after resection (two categories: R0 v R+), and absence or presence of LVI (2 categories: LVI– v LVI+).
All analyses were performed using the statistical package SPSS for Windows (release 11.0; SPSS Inc, Chicago, IL).
RESULTS
In the subgroup of patients with primary resected AEG I tumors, LVI was demonstrated in 70 (38.9%) of 180 patients. The prevalence of LVI in the AEG II/III subgroup was a markedly higher 57.0% (159 of 279)—a statistically significant difference (P = .0002, {chi}2 test). The results of the histopathologic work-up and the subclassification into minor and extensive LVI are presented in Table 3.
LVI correlated with PT, PN, and R categories, with a higher prevalence in more advanced cases (higher PT category, PN+, R+ cases). Results of the prevalence of LVI and its correlation with PT, PN, and R are given in Table 4.
A substantial amount of LVI was found in early cancers (PT1) and in cases without nodal involvement (negative PN status). LVI was found in 11.6% of the patients with early (PT1) AEG I tumors and in 13.5% of the patients with AEG II/III tumors. The node-negative cases showed LVI in 12.5% (AEG I) and 24.7% (AEG II/III).
The results of the survival analysis are shown as Kaplan-Meier curves (Fig. 3, 4, and 5). On univariate analysis, the presence and extent of LVI was a significant prognostic factor for the entire patient population, as well as for the subgroups AEG I and AEG II/III.
Multivariate analysis (Cox regression) revealed prognostic significance and independence as determinant of survival in the group of all AEG tumors (P = .050; Table 5) and in the subgroup of AEG II/III tumors (P = .048; Table 6). In contrast to this finding, LVI lost its prognostic significance in the subgroup of patients with AEG I tumors.
DISCUSSION
LVI has previously been shown to be a strong and independent prognostic factor in esophageal squamous cell cancers7-10 and gastric cancers,5,6 but its role in distal esophageal adenocarcinomas and adenocarcinomas of the esophagogastric junction has not been proven so far. LVI is included in the TNM system as "L" category, but its mention in the histopathology report is optional.
LVI is thought to precede or occur coincidently with lymph node metastasis. This is reflected by the correlation between LVI and N and T categories in our study. Our analysis clearly shows that LVI is associated with more advanced disease, as patients with more advanced tumors have a higher prevalence of LVI.
Nevertheless, patients with early cancers (T1) and cancers without nodal involvement (N0) showed a substantial amount of LVI. This reflects the aggressiveness of the AEG tumors. Patients with early local tumor growth often have lymphatic tumor spread, additionally to locally limited disease.
Multivariate analysis confirmed that the impact of LVI is independent from other prognostic markers in the group of all patients with AEG. In the analysis of the subgroups, LVI lost its prognostic significance in patients with AEG I: LVI is not independent from other strong prognostic factors (T, N, and R). By contrast, in the subgroup of patients with AEG II/III, LVI remains a strong and independent prognostic factor. Rather surprisingly, macroscopically and microscopically complete tumor resection (R0) and the T category lost prognostic significance in the subgroup of patients with AEG II/III tumors when LVI was included in the multivariate model. Thus, LVI seems to be the leading and independent prognostic factor (next to frank lymphatic metastases) in AEG II/III tumors, but not in AEG I tumors.
One possible explanation for the discrepancies between AEG I and AEG II/III is the markedly different pathogenetic mechanisms of these tumor entities.23,24 Development of AEG I follows a metaplasia–intraepithelial neoplasia–carcinoma sequence, which is promoted by the chronic damaging effect of gastroesophageal reflux.24 One major feature of this disease process is chronic inflammation (Fig 6). This might result in stepwise degradation of lymphatic vessels, and thus, a lesser degree of LVI in AEG I tumors. In contrast, chronic inflammation and reflux has so far not been shown to play a role in AEG II/III tumors.
These results support a distinction of esophageal carcinomas according to the tumor type: esophageal adenocarcinomas and squamous cell cancers differ in respect to tumor biology, etiology, and pathogenesis.18 Our results show that LVI is of minor importance and prognostic impact in distal esophageal adenocarcinomas (AEG I), in sharp contrast to esophageal squamous cell cancers, where it is a strong and independent prognostic factor, as previously shown.7 These results support the distinction of these markedly different entities in the TNM staging system.
In summary, our data stress the importance of a systematic histopathologic work-up and the inclusion of LVI in the histopathology report. The results of this study support the inclusion of LVI into the TNM staging system of the UICC.20 The thus far optional L category should gainobligatory character, especially in AEG II/III tumors, in which it is a strong and independent prognostic factor.
Authors' Disclosures of Potential Conflicts of Interest
NOTES
Supported by Deutsche Krebshilfe (grant No. 70-2789-Si3).
Authors' disclosures of potential conflicts of interest are found at the end of this article.
REFERENCES
1. Shirouzu K, Isomoto H, Morodomi, et al: Carcinomatous lymphatic permeation: Prognostic significance in patients with rectal carcinoma—A long term prospective study. Cancer 75:4-10, 1995
2. Minsky BD, Mies C, Rich TA, et al: Lymphatic vessel invasion is an independent prognostic factor for survival in colorectal cancer. Int J Radiat Oncol Biol Phys 17:311-318, 1989
3. Zamcheck N, Doos WG, Prudente R, et al: Prognostic factors in colon carcinoma: Correlation of serum carcinoembryonic antigen level and tumor histopathology. Hum Pathol 6:31-45, 1975
4. Baur M, Schlappack O, Jaskez R, et al: [Differential prognosis of patients with breast cancer with pleuritis carcinomatosa and with pulmonary lymphangiosis carcinomatosa.] Wien Klin Wochenschr 104:714-716, 1992
5. Gabbert HE, Meier S, Gerharz CD, et al: Incidence and prognostic significance of vascular invasion in 529 gastric-cancer patients. Int J Cancer 49:203-207, 1991
6. Gabbert HE, Meier S, Gerharz CD, et al: Tumor-cell dissociation at the invasion front: A new prognostic parameter in gastric cancer patients. Int J Cancer 50:202-207, 1992
7. Brücher BL, Stein HJ, Werner M, et al: Lymphatic vessel invasion is an independent prognostic factor in patients with a primary resected tumor with esophageal squamous cell carcinoma. Cancer 92:2228-2233, 2001
8. Torres CM, Wang HH, Turner JR, et al: Pathologic prognostic factors in oesophageal squamous cell carcinoma: A follow –up study of 74 patients with or without preoperative chemoradiation therapy. Mod Pathol 12:961-968, 1999
9. Watanabe M, Kuwano H, Araki K, et al: Prognostic factors in patients with submucosal carcinoma of the oesophagus. Br J Cancer 83:609-613, 2000
10. Sarbia M, Bittinger F, Porschen R, et al: Tumor vascularization and prognosis in squamous cell carcinoma of the esophagus. Anticancer Res 16:2117-2121, 1996
11. Brechot JM, Chevret S, Charpentier MC, et al: Blood vessel and lymphatic vessel invasion in resected nonsmall cell lung carcinoma: Correlation with TNM stage and disease free and overall survival. Cancer 78:2111-2118, 1996
12. Ota S, Muro H: [The prognostic significance of lymphatic vessel and blood vessel invasion in peripherally located adenocarcinoma of the lung] Nippon Rinsho 60:276-280, 2002 (suppl 5)
13. Ikehara M, Oshita F, Kameda Y, et al: Expression of survivin correlated with vessel invasion is a marker of poor prognosis in small adenocarcinoma of the lung. Oncol Rep 9:835-838, 2002
14. Siewert JR, Hlscher AH, Becker K, et al: [Cardia cancer: Attempt at a therapeutically relevant classification]. Chirurg 58:25-34, 1987
15. Siewert JR, Stein HJ: Classification of carcinoma of the oesophagogastric junction. Br J Surg 85:1457-1459, 1998
16. Stein HJ, Feith M, Siewert JR: Cancer of the esophagogastric junction. Surg Oncol 9:35-41, 2000
17. Stein HJ, Feith M, Siewert JR: Individualized surgical strategies for cancer of the esophagogastric junction. Ann Chir Gynaecol 89:191-198, 2000
18. Siewert JR, Feith M, Werner M, et al: Adenocarcinoma of the esophagogastric junction. Results of surgical therapy based on anatomical/topographic classification in 1,002 consecutive patients. Ann Surg 232:353-361, 2000
19. Bartels H, Stein HJ, Siewert JR: Preoperative risk analysis and postoperative mortality of oesophagectomy for resectable oesophageal cancer. Br J Surg 85:840-844, 1998
20. Sobin LH, Wittekind Ch: UICC: TNM Classification of Malignant Tumors (ed 6). New York, NY, Wiley-Liss, 2002
21. Cox DR: Regression models and life tables. J R Stat Soc 34:1987-2001, 1972
22. Kaplan EL, Meier P: Nonparametric estimation from incomplete observations. J Am Stat Assoc 75:457-487, 1958
23. Stein HJ, von Rahden BHA, Hfler H, et al: [Carcinoma of the oesophagogastric junction and Barrett esophagus: An almost clear oncologic model] Chirurg 74:703-708, 2003
24. von Rahden BHA, Stein HJ, Siewert JR: Barrett's esophagus and Barrett's carcinoma. Curr Oncol Rep 5:203-209, 2003(Burkhard H.A. von Rahden,)