|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
© 2003 American Society for Clinical Oncology Interaction of Splenectomy and Perioperative Blood Transfusions on Prognosis of Patients With Proximal Gastric and Gastroesophageal Junction Cancer
From the Departments of Surgery, Epidemiology and Biostatistics, and Pathology, Memorial Sloan-Kettering Cancer Center, New York, NY. Address reprint requests to Daniel Coit, MD, FACS, Gastric and Mixed Tumor Service, Department of Surgery, Memorial Sloan-Kettering Cancer Center, 1275 York Ave, New York, NY 10021; e-mail: coitd{at}mskcc.org.
Purpose: To assess the interaction of splenectomy and perioperative allogeneic blood transfusions on the prognosis of patients undergoing a potentially curative resection of proximal gastric and gastroesophageal junction (GEJ) cancer, because reports from the transplantation literature demonstrated that the immunosuppressive effects of transfusions are dependent on the presence of an intact spleen. Patients and Methods: Between July 1, 1985, and July 30, 2001, 240 patients underwent complete resection (R0) of a proximal gastric or GEJ cancer (Siewert type II or III). Clinical and pathologic factors were collected in a prospective database. The survival data were modeled using the method of Kaplan and Meier and analyzed by the log-rank test and Cox regression. Results: The median follow-up of the patients was 25 months (40 months for survivors). The median relapse-free survival was 30 months, and the median disease-specific survival was 45 months. Univariate analysis suggested an interaction of splenectomy and perioperative transfusion in their effect on relapse-free survival. Patients who received a perioperative transfusion but did not undergo splenectomy demonstrated the worst prognosis on multivariate analysis independent of other prognostic factors. In patients who underwent splenectomy, perioperative transfusion had no effect on relapse-free survival on multivariate analysis. Conclusion: Our study suggests an interaction of blood transfusion and splenectomy in their effect on survival paralleling the findings in the transplantation literature. The adverse effect of allogeneic blood transfusion on prognosis in patients with gastric cancer seems to be associated with the presence of an intact spleen and is abrogated by its absence.
IN 2002, approximately 21,600 people were diagnosed and 12,400 people died from gastric cancer in the United States.1 Complete surgical resection remains the only potentially curative modality for gastric adenocarcinoma. In addition to several well-defined pathologic criteria,2 treatment-related factors, such as extent of lymphadenectomy, splenectomy, and perioperative allogeneic blood transfusion, have been associated with outcome of patients with gastric cancer. Extended lymphadenectomy, splenectomy, or a combination of both might theoretically improve prognosis by achieving better lymph node clearance. However, neither of these factors was associated with an improved outcome in randomized trials specifically addressing this issue.3–5 It has been shown that splenectomy and perioperative blood transfusions are associated with a worse outcome, and an immunosuppressive effect has been postulated to explain these findings.6–12 Data from the transplantation literature demonstrate that the immunosuppressive effect of blood transfusion seems to depend on the presence of an intact spleen, suggesting that splenectomy and perioperative transfusions may have important interactions on the immune competence of patients.13–16 This effect, to our knowledge, has not been evaluated in patients with gastric cancer. Splenectomy is associated with an increased incidence of postoperative infectious complications, which may have further implications for prognosis.4,17,18 Although most authors recommend splenic preservation in the surgical treatment of gastric cancer, splenectomy is still considered for proximal gastric and gastroesophageal junction (GEJ) cancers (type II and III), because the incidence of lymph node metastases in the splenic hilum is thought to be higher in these tumors.19–21 Splenectomy is also performed for local tumor invasion and in case of accidental injury during operation. The aim of this study was to examine the interaction of splenectomy and perioperative blood transfusions in a homogenous group of patients with proximal gastric and type II and III GEJ cancer undergoing potentially curative resection.
Patients and Treatment A total of 1,256 patients underwent R0 (no residual gross or microscopic tumor) resection of gastric adenocarcinoma at Memorial Sloan-Kettering Cancer Center between July 1, 1985, and July 30, 2001, and were entered into a prospective database. From this patient cohort, we identified 240 patients with proximal gastric or GEJ cancer (Siewert type II or III) who underwent resection through a transabdominal or left thoraco-abdominal approach. Tumor position was determined both at operation and pathologically. Proximal gastric cancer was defined as a cancer in the proximal one third of the stomach; GEJ cancer was classified according to previously published guidelines.22 Patient demographics, tumor characteristics, treatment related factors, and postoperative course were recorded and analyzed. Perioperative transfusion was defined as allogeneic blood transfusion during operation or the first 2 postoperative days. Transfusion was performed at the discretion of the treating surgeon and anesthesiologist. Hemodynamically significant intraoperative blood loss, or a hemoglobin level of 8 to 10 g/dL, were used as thresholds for transfusion, depending on the patient’s comorbidities. Infectious complications included sepsis, anastomotic leak, intra-abdominal abscess, pneumonia, catheter sepsis, wound, and urinary tract infections. Noninfectious complications included: renal failure, postoperative hemorrhage, pulmonary embolism, atelectasis, cardiac complications, pneumothorax, pleural effusion, deep vein thrombosis, and urinary retention. Tumor stage and grade were classified according to the fifth edition of the tumor-node-metastasis system classification of the International Union Against Cancer and the American Joint Committee on Cancer.23
Statistical Analysis
Patient Characteristics The characteristics of the 240 patients included in this study are displayed in Table 1  . The median age of the patients was 64 years (range, 21 to 89 years). Fifty patients received preoperative radiotherapy, chemotherapy, or both, and 54 patients received postoperative radiotherapy, chemotherapy, or both, either as standard of care or as part of different clinical trials. Indications for splenectomy in 75 patients (31%) included: extent of tumor in 51 patients, bleeding in 14 patients, and ischemia of the spleen in one patient. The basis for splenectomy could not be determined in nine patients. The median intraoperative blood loss in patients receiving perioperative blood transfusions was 900 mL (range, 100 to 6,300 mL) compared with 500 mL (range, 50 to 2,100 mL) in patients not receiving transfusions (P < .0001). Transfusion was also associated with more advanced tumors; 62% of the patients receiving blood transfusions had either T3 or T4 tumors versus 50% of patients without transfusions (P = .03).
Follow-Up and Univariate Characterization of Prognostic Factors The median follow-up for all patients was 25 months, with a range of 3 to 193 months (median of 40 months for survivors; range, 3 to 193 months). The estimated median relapse-free survival was 30 months; the estimated median disease-specific survival was 45 months. At the time of last follow-up, 92 patients (38%) had no evidence of disease, 11 patients (5%) were alive with disease, 111 patients (46%) had died of disease, and 26 patients (11%) had died of unrelated causes. Data regarding the location of recurrence were available for 109 patients. With regard to initial site, distant recurrence occurred in 59 patients, local recurrence (anastomosis, lymph nodes, gastric bed) occurred in 66 patients, and peritoneal recurrence occurred in 17 patients, with 33 patients experiencing recurrence at two sites. Table 2 depicts the univariate analysis of the different factors on relapse-free and disease-specific survival. Perioperative blood transfusion and infectious complications, as well as several pathologic criteria, such as T and N stage, vascular and perineural invasion, and poor differentiation, were associated with a poor outcome.
The interaction between perioperative blood transfusions and splenectomy on relapse-free survival is shown in Fig 1 . In patients who underwent splenectomy, perioperative blood transfusion had no prognostic relevance (estimated median survival, 26 v 22 months; P = not significant, log-rank test). However, in patients who did not undergo splenectomy, perioperative blood transfusion was associated with a significantly worse prognosis (estimated median survival, 20 v 66 months; P < .01, log-rank test). The interaction of splenectomy and perioperative transfusion had a similar association with disease-specific survival: perioperative blood transfusion was significantly associated with a worse prognosis only in patients who did not undergo splenectomy (Fig 2).
Multivariate Characterization of Prognostic Factors The first attempt at building a multivariate model for relapse-free survival did not use any interaction terms. Included were all factors that had prognostic potential as suggested by the univariate analysis (P < .1). The final model defined N stage, perineural invasion, and infectious complications as independent risk factors. Because univariate analysis confirmed the a priori hypothesis of an interaction between splenectomy and perioperative transfusion, we then included an interaction term that combined splenectomy and perioperative blood transfusion as a single variable in the multivariate model for relapse-free survival. After stepwise elimination, the model identified T and N stage, perineural invasion, and the interaction term splenectomy/perioperative transfusion as independent risk factors (Table 3 ). This model reveals that splenectomy and perioperative blood transfusions should be considered together when assessing their impact on relapse-free survival. Patients who did not receive a perioperative blood transfusion were not at an increased risk for a decreased relapse-free survival. Of those patients who did receive a blood transfusion, an increased risk for a decreased relapse-free survival was only seen if no splenectomy was performed (hazard ratio, 1.7; 95% CI, 1.1 to 2.7). Patients who received a blood transfusion and underwent a splenectomy, however, were not at an increased risk for a decreased relapse-free survival.
Splenectomy is associated with an increased incidence of infectious complications (data not shown), and these complications are associated with a worse prognosis on univariate analysis (Table 2  ). On multivariate analysis including the interaction term splenectomy/transfusion, however, infectious complications were not an independent predictor of survival. Infectious complications, therefore, did not eliminate the beneficial effect of splenectomy in patients requiring perioperative blood transfusions. Separate multivariate analyses were performed for local, distant, and peritoneal recurrences. N stage and perineural invasion were predictors for local and distant recurrence. T and N stage were predictors for peritoneal recurrence. The interaction term splenectomy/perioperative transfusion did not reach significance in these analyses.
Table 4
This study investigated the interaction between perioperative allogeneic blood transfusions and splenectomy on the prognosis of patients undergoing resection of proximal gastric or GEJ cancer. Previous studies have demonstrated that Siewert type II and III GEJ tumors show a pattern of lymphatic spread similar to that of proximal gastric cancer, which is the rationale for combining these tumors in this study.26,27 The most striking finding of this study is that perioperative allogeneic blood transfusion and splenectomy show an interaction in their effect on survival. Patients who received a perioperative blood transfusion showed a shorter relapse-free survival only if the spleen was preserved. Allogeneic blood transfusions are known to be immunosuppressive, leading to improved survival of transplanted organs.28,29 Evidence for an interaction between splenectomy and blood transfusion was first demonstrated in animal organ transplantation experiments. Immunosuppressive effects of transfusion were abrogated by splenectomy in these experiments: animals that underwent splenectomy and blood transfusion showed an increased rate of rejection of the transplanted organs compared with animals that only received blood transfusions before transplantation.13–16 Animal experiments have also demonstrated that allogeneic blood transfusions enhance tumor growth. Of special interest in context of this study is that this effect of an enhanced tumor growth after allogeneic blood transfusions can be passively transferred to other animals through spleen cells of transfused animals. Tumor-bearing animals that received spleen cells from allogeneically transfused animals showed significantly increased tumor growth compared with animals that had received spleen cells from syngeneically transfused animals.30 These experiments show that allogeneic blood transfusions actively induce an immunosuppressive factor in the spleen, most likely suppressor T lymphocytes, leading to an increased tumor growth. Animal experiments have also demonstrated that the timing of splenectomy and blood transfusion is important. If the animals received transfusions up to 2 weeks after splenectomy, the interaction between transfusion and splenectomy could no longer be demonstrated.31 This finding implies that other immunocompetent sites can be stimulated to produce suppressor cell activity after splenectomy has been performed and is the reason why we limited our analysis to transfusions given in the perioperative period. The interaction of splenectomy and blood transfusions on survival has never been investigated in patients with gastric cancer. Splenectomy and blood transfusions have been evaluated only as individual factors in these patients. The effect of splenectomy on prognosis in patients with gastric cancer remains controversial. Splenectomy might facilitate a more complete lymphadenectomy by thorough clearance of the lymph nodes in the splenic hilum. It is, however, possible to perform a sufficient lymphadenectomy in the splenic hilum without splenectomy. Numerous retrospective as well as prospective randomized trials, however, have not demonstrated a prognostic benefit for splenectomy or extended lymphadenectomy.5 Some retrospective studies even demonstrated a worse survival with splenectomy.3,4,6–8,17,32 Our results concur with those of published studies suggesting that splenectomy does not improve survival by a superior lymphadenectomy. This finding is important, as previous authors have recommended splenectomy in patients with proximal gastric or GEJ cancer to address the increased likelihood of lymph node metastases in the splenic hilum.19,20 The immunologic effects of splenectomy that may influence survival in patients undergoing resection for gastric cancer are not clearly defined. Splenectomy is thought to be associated with impaired phagocytic activity, decreased antibody response, altered levels of immunoglobulins, and altered T-cell function.6 Other studies, however, have shown that the spleen is also an essential organ for immunosuppressive activity, because the spleen must be present to permit the development of tolerance to certain antigens.6 Some reports claim a decreased survival or impaired immunologic function after splenectomy, whereas other reports suggest that splenectomy might be beneficial.7,8,33,34 Blood transfusions have been associated with decreased survival in patients with a variety of malignancies, including gastric cancer.9–12 Decreased helper/suppressor T-cell ratios, decreased natural-killer cell activity, decreased macrophage antigen presentation, suppression of lymphocyte blastogenesis, and decreased delayed-type hypersensitivity have been observed after transfusion, demonstrating a relative immunosuppression.28 The mechanism of this transfusion effect is not well defined. Different hypotheses, such as overload of the reticuloendothelial system, alterations in interleukin-2 and prostaglandin metabolism, and immunosuppression by clonal deletion or generation of active suppressor factors have been proposed. Active suppressor factors include anti-idiotypic antibodies or suppressor T lymphocytes that are generated after transfusion.28 The latter effect is of special interest with respect to the results of this study, as splenectomy precludes the development of suppressor T-lymphocytes, explaining a possible interaction between the effects of splenectomy and blood transfusion on the immune system.35 To our knowledge, this is the first study demonstrating a possible interaction of splenectomy and blood transfusion on disease recurrence of patients with resected gastric cancer. Although the statistical analysis of the data indicated that this observation is valid and that there is only a minimal likelihood that our results are merely caused by chance, it cannot assess a causal explanation in an observational setting. However, the observations are consistent with the hypothesis specified before the data analysis. Because we included an interaction term only to test such a hypothesis, it is unlikely that our results are merely an accidental statistical association. Therefore, immunologic effects must be strongly considered as an explanation for the results of this study. Our results may also contribute to the understanding of the transfusion effect and demonstrate the importance of the spleen in this context. The next question would be whether differences in tumor recurrences after curative resection could be explained by altered immune function. It is well known that after potentially curative resection, disseminated tumor cells can be detected in several organ systems, including blood and bone marrow in patients with gastric cancer.36–38 It could be hypothesized that constant immunologic surveillance of these cells is of major importance in these patients. This theory is supported by clinical data demonstrating that a prognostic influence of blood transfusion was found only in patients with gastric cancer in whom disseminated tumor cells could be detected in bone marrow samples.39 The specific effect of splenectomy was not evaluated in that study. On the basis of our data, we hypothesize that splenectomy should be viewed with a different perspective in patients undergoing potentially curative resection of gastric cancer. Our data do not support the notion of a routine splenectomy in patients with proximal gastric or GEJ junction cancer to improve prognosis by a more extended lymphadenectomy. Our data strongly suggest that the adverse effect of allogeneic blood transfusion in the perioperative period on prognosis is associated with the presence of an intact spleen and is abrogated by its absence, which points toward an immunologic mechanism. One potential way of testing our hypothesis would be to perform a randomized trial comparing splenectomy versus no splenectomy in patients who are receiving a blood transfusion during gastric resection. Additionally, ongoing randomized trials regarding the value of splenectomy in gastric cancer patients could be stratified for perioperative blood transfusion to further investigate this hypothesis.40
The authors indicated no potential conflicts of interest.
Supported by a grant from the Gelb Foundation, New York, NY.
1. Jemal A, Thomas A, Murray T, et al: Cancer Statistics. CA Cancer J Clin 52:23–47, 2002 2. Karpeh M, Leon L, Klimstra D, et al: Lymph node staging in gastric cancer: Is location more important than number? Ann Surg 232:362–371, 2000[CrossRef][Medline] 3. Cuschieri A, Weeden S, Fielding J, et al: Patient survival after D1 and D2 resections for gastric cancer: Long-term results of the MRC randomized surgical trial. Br J Cancer 79:1522–1530, 1999[CrossRef][Medline]
4. Bonenkamp JJ, Hermans J, Sasako M, et al: Extended lymph-node dissection for gastric cancer. N Engl J Med 340:908–914, 1999 5. Csendes A, Burdiles P, Rojas J, et al: A prospective randomized study comparing D2 total gastrectomy versus D2 total gastrectomy plus splenectomy in 187 patients with gastric carcinoma. Surgery 131:401–407, 2002[CrossRef][Medline] 6. Llende M, Santiago-Delpin EA, Lavergne J: Immunobiological consequences of splenectomy: A review. J Surg Res 40:85–94, 1986[CrossRef][Medline]
7. Griffith JP, Sue-Ling HM, Dixon MI, et al: Preservation of the spleen improves survival after radical surgery for gastric cancer. Gut 36:684–690, 1995 8. Wanebo HJ, Kennedy BJ, Winchester DP, et al: Role of splenectomy in gastric cancer surgery: Adverse effect of elective splenectomy on long-term survival. J Am Coll Surg 185:177–184, 1997[CrossRef][Medline] 9. Maeta M, Shimizu N, Oka A, et al: Perioperative allogeneic blood transfusion exacerbates surgical stress-induced postoperative immunosuppression and has a negative effect on prognosis in patients with gastric cancer. J Surg Oncol 55:149–153, 1994[Medline] 10. Hyung WJ, Noh SH, Shin DW, et al: Adverse effects of perioperative transfusion on patients with stage III and IV gastric cancer. Ann Surg Oncol 9:5–12, 2002[CrossRef][Medline] 11. Fong Y, Karpeh M, Mayer K, et al: Association of perioperative transfusions with poor outcome in resection of gastric adenocarcinoma. Am J Surg 167:256–260, 1994[CrossRef][Medline] 12. Dhar DK, Kubota H, Tachibana M, et al: A tailored perioperative blood transfusion might avoid undue recurrences in gastric carcinoma patients. Dig Dis Sci 45:1737–1742, 2000[CrossRef][Medline] 13. Marquet RL, Heineman E, Tank B, et al: Abrogation of the beneficial blood transfusion effect in dogs by splenectomy. World J Surg 8:408–413, 1984[CrossRef][Medline] 14. Shelby J, Wakley E, Corry RJ: Suppressor cell induction in donor-specific transfused mouse heart recipients. Surgery 96:296–301, 1984[Medline] 15. Shelby J, Wakley E, Corry RJ: Splenectomy abrogates the improved graft survival achieved by donor-specific transfusion. Transplant Proc 17:1083–1086, 1985 16. Yamaguchi Y, Goto M, Makino Y, et al: Prolonged survival of rat hepatic allografts pretreated with single donor-specific blood transfusion: The distribution of donor cells expressing class I major histocompatibility complex antigens in the recipient. J Surg Res 61:23–29, 1996[CrossRef][Medline]
17. Brady MS, Rogatko A, Dent L, et al: Effect of splenectomy on morbidity and survival following curative gastrectomy for carcinoma. Arch Surg 126:359–364, 1991 18. Cuschieri A, Fayers P, Fielding J, et al: Postoperative morbidity and mortality after D1 and D2 resections for gastric cancer: Preliminary results of the MRC randomised controlled surgical trial. Lancet 347:995–999, 1996[CrossRef][Medline] 19. Monig SP, Collet PH, Baldus SE, et al: Splenectomy in proximal gastric cancer: Frequency of lymph node metastasis to the splenic hilus. J Surg Oncol 76:89–92, 2001[CrossRef][Medline] 20. Kitamura K, Nishida S, Yamamoto K, et al: Lymph node metastasis in gastric cancer in the upper third of the stomach: Surgical treatment on the basis of the anatomical distribution of positive node. Hepatogastroenterology 45:281–285, 1998[Medline] 21. Sakaguchi T, Sawada H, Yamada Y, et al: Indication for splenectomy for gastric carcinoma involving the proximal part of the stomach. Hepatogastroenterology 48:603–605, 2001[Medline] 22. Siewert JR, Stein HJ: Classification of adenocarcinoma of the oesophagogastric junction. Br J Surg 85:1457–1459, 1998[CrossRef][Medline] 23. Sobin LH, Wittekind C: UICC: TNM Classification of Malignant Tumors. London, United Kingdom, John Wiley & Sons, 1997 24. Kaplan EL, Meier P: Nonparametric estimation from incomplete observations. J Am Stat Assoc 53:457–481, 1958[CrossRef] 25. Cox DR: Regression models and life-tables. J R Stat Soc (B) 34:187–220, 1972 26. Siewert JR, Feith M, Werner M, et al: Adenocarcinoma of the esophagogastric junction. Ann Surg 232:353–361, 2000[CrossRef][Medline] 27. Dresner SM, Lamb PJ, Bennett MK, et al: The pattern of metastatic lymph node dissemination from adenocarcinoma of the esophagogastric junction. Surgery 129:103–109, 2001[CrossRef][Medline] 28. Brunson ME, Alexander JW: Mechanisms of transfusion-induced immunosuppression. Transfusion 30:651–658, 1990[CrossRef][Medline] 29. Opelz G, Sengar DPS, Mickey MR, et al: Effect of blood transfusions on subsequent kidney transplants. Transplant Proc 5:253–259, 1973[Medline]
30. Blajchman MA, Bardossy L, Carmen R, et al: Allogeneic blood transfusion-induced enhancement of tumor growth: Two animal models showing amelioration by leukodepletion and passive transfer using spleen cells. Blood 81:1880–1882, 1993 31. Banowsky LH, Saunders P, Riehl B, et al: Effect of pretransplant blood transfusions and splenectomy on renal allograft survival in the Lewis rat. J Urol 130:996–998, 1983[Medline] 32. Martin RC, Jaques DP, Brennan MF, et al: Extended local resection for advanced gastric cancer: Increased survival versus increased morbidity. Ann Surg 236:159–165, 2002[CrossRef][Medline] 33. Saji S, Sakamoto J, Teramukai S, et al: Impact of splenectomy and immunochemotherapy on survival following gastrectomy for carcinoma: Covariate interaction with immunosuppressive acidic protein, a serum marker for the host immune system—Tumor Marker Committee for the Study Group of Immunochemotherapy with PSK for Gastric Cancer. Surg Today 29:504–510, 1999[CrossRef][Medline] 34. Okuno K, Tanaka A, Shigeoka H, et al: Suppression of T-cell function in gastric cancer patients after total gastrectomy with splenectomy: Implications of splenic autotransplantation. Gastric Cancer 2:20–25, 1999[CrossRef][Medline]
35. Sy MS, Miller SD, Kowach HB, et al: A splenic requirement for the generation of suppressor T cells. J Immunol 119:2095–2099, 1977
36. Heiss MM, Simon EH, Beyer BCM, et al: Minimal residual disease in gastric cancer: Evidence of an independent prognostic relevance of urokinase receptor expression by disseminated tumor cells in the bone marrow. J Clin Oncol 20:2005–2016, 2002 37. Heiss MM, Allgayer H, Gruetzner KU, et al: Individual development and uPA-receptor expression of disseminated tumour cells in bone marrow: A reference to early systemic disease in solid cancer. Nat Med 1:1035–1039, 1995[CrossRef][Medline] 38. Kienle P, Koch M: Minimal residual disease in gastrointestinal cancer. Semin Surg Oncol 20:282–293, 2001[CrossRef][Medline] 39. Heiss MM, Allgayer H, Gruetzner KU, et al: Prognostic influence of blood transfusion on minimal residual disease in resected gastric cancer patients. Anticancer Res 17:2657–2662, 1997[Medline]
40. Sano T, Yamamoto S, Sasako M: Randomized controlled trial to evaluate splenectomy in total gastrectomy for proximal gastric carcinoma: Japan clinical oncology group study JCOG 0110-MF. Jpn J Clin Oncol 32:363–364, 2002 Submitted December 26, 2002; accepted September 27, 2003.
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||
|
|||||||||||
|
|
Copyright © 2003 by the American Society of Clinical Oncology, Online ISSN: 1527-7755. Print ISSN: 0732-183X
|
TOP
ABSTRACT
INTRODUCTION
5% (P 
2 test. The time of follow-up was calculated from the date of first operation. Survival was estimated according to the Kaplan-Meier method and compared using the log-rank test.
