Cancer Immunol Immunother (1992) 35:175 - 180
ancer mmunolggy mmunotherapy @ Springer-Verlag 1992
Glycosidically bound sialic acid levels as a predictive marker of postoperative adjuvant therapy in gastric cancer Kyoji Ogoshi, Yasumasa Kondoh, Tomoo Tajima, and Toshio Mitomi Department of Surgery, Tokai University School of Medicine, Kanagawa, Japan Received 8 January 1992/Accepted 21 February 1992
Summary. A group of 293 gastric cancer patients were examined to see if the preoperative value of glycosidically bound sialic acid is a predictor of prognosis and effectiveness of postoperative adjuvant therapy. All patients had gastrectomies and were histologically confirmed to have primary adenocarcinoma of the stomach. Some patients then received either postoperative adjuvant chemotherapy or immunochemotherapy. Patients with sialic acid levels less than 74.5 mg/dl survived significantly longer than those with sialic acid levels of 74.5 mg/dl or of 85.3 mg/dl and over. No significant differences in survival were found among patients treated by gastrectomy alone, gastrectomy plus chemotherapy and gastrectomy plus immunochemotherapy. However, patients with abnormally elevated levels of sialic acid survived significantly longer when they were treated with immunochemotherapy after gastrectomy than those treated by gastrectomy alone or with chemotherapy after gastrectomy. By using C o x ' s multivariate regression model, p T N M stages, postoperative adjuvant therapy (chemotherapy and immunochemotherapy) and preoperative serum levels of sialic acid were examined as prognostic variables. Postoperative therapy was a significant prognostic variable in patients with abnormally elevated levels of sialic acid. The preoperative serum level of sialic acid is a promising predictive marker of the response to postoperative adjuvant immunochemotherapy. Key words: Gastric c a n c e r - Sialic a c i d - P r o g n o s i s - P S K
Introduction Postoperative adjuvant immunochemotherapy and chemotherapy were reported to be effective in gastric cancer in
Offprint requests to: K. Ogoshi, Department of Surgery, School of Medicine, Tokai University,Bohseidai Isehara, Kanagawa, 259-11 Japan
Japan [9, 29] and in colon cancer in USA [18]. Mitomycin C, fluoropyrimidines, the protein-bound polysaccharide termed PSK and levamisole were used as chemotherapeutic or immunomodulating agents. But the reasons for the effectiveness of postoperative adjuvant therapy have not yet been sufficiently evaluated. Winzler reported significant elevations in proteinbound sialic acid (N-acetylneuraminic acid) in cancer patients [32]. Subsequently many investigators have verified increases in serum lipid-bound or lipid-associated sialic acid levels in the majority of patients with a variety of cancers [4, 11, 15, 19, 20]. In this investigation, we studied the clinical utility of the preoperative serum sialic acid status of gastric cancer patients to determine whether their preoperative values could act as possible predictors of the prognosis and the effectiveness of postoperative adjuvant therapy.
Materials and methods The study was conducted from March 1979 to December 1989 at the Department of Surgery of Tokai University.Preoperative serum levels of sialic acid were examined in 293 resected gastric cancer patients with histologically confirmed primary adenocarcinoma. Their ages ranged from 21 to 87 years (median: 58 years), and 73.4% were men. They were selected for chemotherapy or immunochemotherapyby cancer stage on the basis of gross findings accordingto the Japanese Research Society for gastric cancer [ l 2]. The use of chemotherapeuticregimens was left to the judgment of doctors. With the exclusion of 57 cases, in which tumor invaded the mucosa without lymph node metastasis, 236 of 293 patients were analyzed for survival and effectiveness of postoperative adjuvant therapy. The patients who refused postoperative therapy themselves or those whose doctors did not indicate postoperative adjuvant therapy underwent gastrectomy alone and were used as controls. Of the 236 patients, 94 underwent total gastrectomy and 142 had subtotal gastrectomy (128 distal gastrectomy, 14 proximal gastrectomy). Twenty-twopatients underwent gastrectomywith incompleteremoval of perigastric lymph nodes within 3 cm of the edge of the primary tumor (R0 dissection of regional lymph nodes), while 20 had R1 dissection (completeremoval of perigastric lymph nodes within 3 cm of the edge of the primary tumor), and 194 had R2 dissection (complete removal of perigastric lymph nodes and the lymph nodes along the left gastric,
176
Table 1. Values/scores of variables used in the Cox regression model Prognostic factor
Values/scores
pTNM stage a
1.
IA; 2, IB; 3, II; 4, IIIA; 5, IIIB; 6, IV
Therapy
1. 2.
Gastrectomy and chemotherapy Gastrectomy and chemotherapy plus PSK b
Sialic acid before operation
Serum sialic acid level of each patient
Based on histological finding according to the 1987 Unio Intemationalis Contra Cancrum (UICC) [30] b Protein-bound polysaccharide K a
common hepatic, splenic, and celiac arteries). All resected specimens were subjected to a detailed pathological examination, and the patients were then staged according to the 1987 Unio Internationalis Contra Cancrum (UICC) on pTNM staging [30]. For 42 patients, the sole therapy was total or subtotal gastrectomy; 98 cases received gastrectomy plus postoperative adjuvant chemotherapy, which consisted of intravenous administration of mitomycin C (Kyowa Hakko Kogyo Co. Ltd., Tokyo, Japan) (20 mg intraoperatively and 10 mg on postoperative day 1). This was followed on postoperative day 14 with the oral administration of fluoropyrimidines, such as Nl-(2'-tetrahydrofuryl)-5-fluorouracil (Futraful) (Taiho Pharmaceutical Co. Ltd., Tokyo, Japan) (600 mg/day) or 5-fluorouracil (Kyowa Hakko Kogyo Co., Ltd., Tokyo, Japan) (150 mg/day). Of these patients, 16 received mitomycin C alone, 39 fluoropyrimidines alone, and 43 mitomycin C plus fluoropyrimidines. A total of 96 patients received postoperative adjuvant chemotherapy plus a nonspecific immunopotentiator, PSK (Kureha Chemical Industry Co. Ltd., Tokyo, Japan) (3.0 g/day), starting on postoperative day 14 after gastrectomy. Chemotherapeutic regimens of immunochemotherapy were as follows: 6 received mitomycin C alone, 49 fluoropyrimidines alone, and 41 mitomycin C plus fluoropyrimidines. Out of 98 patients, 78 (79.6%) received chemotherapy, and 73 out of 96 patients (76.1%) who received immunochemotherapy had oral administration of fluoropyrimidines over a period of 3 months. Out of 96 patients, 80 (83.3%) received immunochemotherapy and were given PSK orally over a period of 3 months.
Survival was assessed from the day of surgery until death or the most recent update. Follow-up information for all patients was obtained by direct patient contact or by telephone contact with the patients or their families in December 1991. The percentages of patients who have been followed-up until death or over 5 years and 3 years were 65.3% and 89.8% respectively. Survival curves were calculated using the Kaplan-Meier product-limit estimate [14] and differences in survival distribution were assessed by the generalized Wilcoxon test and the log-rank test [7, 25]. Operative deaths only were excluded. Multivariate analyses utilizing Cox's model were performed with variables such as pTNM stages, postoperative adjuvant therapy and preoperative serum levels of sialic acid [ 1, 3, 13] in patients with chemotherapy and immunochemotherapy. The values/scores of variables employed in the Cox regression model are shown in Table 1. Blood samples were collected by venipuncture, allowed to clot at room temperature, and then centrifuged at 3000 rpm. The serum was then stored at -70 ° C for subsequent analysis of serum levels of sialic acid. The sialic acid level was determined by using a modified resorcinol reagent kit (Kyokuto sialic acid test, Kyokuto Pharmaceutical Co. Ltd., Tokyo, Japan) and has been described in detail elsewhere [8]. As the mean and standard deviation of sialic acid values for 61 healthy subjects were 63.7_+ 10.8 mg/dl (37.5-91.5 mg/dl), patients were divided into three groups according to the preoperative serum levels of sialic acid: less than 74.5 mg/dl (1 standard deviation), 74.5 mg/dl and over to less than 85.3 mg/dl (2 standard deviations) and 85.3 mg/dl and over. In this study the upper limit of normal sialic acid was defined as 85.3 mg/dl [22].
Results T a b l e 2 s h o w s t h e s e r u m l e v e l s o f s i a l i c a c i d a c c o r d i n g to the histopathological grading and stage grouping. Serum l e v e l s o f s i a l i c a c i d i n c r e a s e d i n r e l a t i o n to t u m o r p r o g r e s sion. H o w e v e r , t h e r e w a s a n o c o r r e l a t i o n b e t w e e n c e l l u l a r differentiation and sialic acid level. F i g u r e 1 s h o w s t h e s u r v i v a l c u r v e s o f 2 9 3 p a t i e n t s acc o r d i n g to t h e s e r u m l e v e l s o f p r e o p e r a t i v e s i a l i c a c i d : l e s s
Table 2. Serum levels of sialic acid (mg/dl) according to the histopathological grading and stage grouping Stage
Total
Histopathological grading a Well differentiated
Moderately differentiated
Poorly differentiatedb
IA+IB
67.2_+ 16.3 (n = 154)
67.0-+ 13.6 (n = 57)
67.8 ___12.6 (n = 27)
67.1 -+ 19.1 (n = 67) {66.8+ 19.2 (n = 39)}
II
69.6_+16.5 ( n = 41)
72.2+_16.7 ( n = 9)
64.7 -+_11.4 (n = 12)
70.5 -+ 18.8 (n = 18) {62.8_+ 8.9 (n = 10)}
IIIA+IIIB
74.9_+ 18.3 (n = 45)*~
73.5 _+21.1 (n = 12)
80.5 _+15.3 (n =
72.7 _+18.6 (n = 22) {77.9_+ 6 . 4 ( n = 4)}
IV
8 7 . 9 + _ 2 1 . 9 (n = 5 3 ) *2
98.1 +_20.1 (n = 7) *3
88.7 _+ 19.5 (n = 15) *5
85.7 _+24.0 (n = 29) *6 {81.0_+27.9 (n = 1 0 ) } *7
Total
72.4-+ 19.3 (n = 293)
71.0 _+ 17.6 (n = 85)
74.1 -+ 17.3 (n = 64)
72.4 -+21.2 (n = 136) {69.1 -+ 19.2 (n = 63)}
a Excluded, 8 patients with mucinous adenocarcinoma. Their mean serum level of sialic acid was 74.3 _+ 19.6 mg/dl b { }, Signet ring cell formation in poorly differentiated adenocarcinoma • 1 vs stages IA+IB (P = 0.0037) • 2 vs stages IA+IB (P
ancer mmunolggy mmunotherapy @ Springer-Verlag 1992
Glycosidically bound sialic acid levels as a predictive marker of postoperative adjuvant therapy in gastric cancer Kyoji Ogoshi, Yasumasa Kondoh, Tomoo Tajima, and Toshio Mitomi Department of Surgery, Tokai University School of Medicine, Kanagawa, Japan Received 8 January 1992/Accepted 21 February 1992
Summary. A group of 293 gastric cancer patients were examined to see if the preoperative value of glycosidically bound sialic acid is a predictor of prognosis and effectiveness of postoperative adjuvant therapy. All patients had gastrectomies and were histologically confirmed to have primary adenocarcinoma of the stomach. Some patients then received either postoperative adjuvant chemotherapy or immunochemotherapy. Patients with sialic acid levels less than 74.5 mg/dl survived significantly longer than those with sialic acid levels of 74.5 mg/dl or of 85.3 mg/dl and over. No significant differences in survival were found among patients treated by gastrectomy alone, gastrectomy plus chemotherapy and gastrectomy plus immunochemotherapy. However, patients with abnormally elevated levels of sialic acid survived significantly longer when they were treated with immunochemotherapy after gastrectomy than those treated by gastrectomy alone or with chemotherapy after gastrectomy. By using C o x ' s multivariate regression model, p T N M stages, postoperative adjuvant therapy (chemotherapy and immunochemotherapy) and preoperative serum levels of sialic acid were examined as prognostic variables. Postoperative therapy was a significant prognostic variable in patients with abnormally elevated levels of sialic acid. The preoperative serum level of sialic acid is a promising predictive marker of the response to postoperative adjuvant immunochemotherapy. Key words: Gastric c a n c e r - Sialic a c i d - P r o g n o s i s - P S K
Introduction Postoperative adjuvant immunochemotherapy and chemotherapy were reported to be effective in gastric cancer in
Offprint requests to: K. Ogoshi, Department of Surgery, School of Medicine, Tokai University,Bohseidai Isehara, Kanagawa, 259-11 Japan
Japan [9, 29] and in colon cancer in USA [18]. Mitomycin C, fluoropyrimidines, the protein-bound polysaccharide termed PSK and levamisole were used as chemotherapeutic or immunomodulating agents. But the reasons for the effectiveness of postoperative adjuvant therapy have not yet been sufficiently evaluated. Winzler reported significant elevations in proteinbound sialic acid (N-acetylneuraminic acid) in cancer patients [32]. Subsequently many investigators have verified increases in serum lipid-bound or lipid-associated sialic acid levels in the majority of patients with a variety of cancers [4, 11, 15, 19, 20]. In this investigation, we studied the clinical utility of the preoperative serum sialic acid status of gastric cancer patients to determine whether their preoperative values could act as possible predictors of the prognosis and the effectiveness of postoperative adjuvant therapy.
Materials and methods The study was conducted from March 1979 to December 1989 at the Department of Surgery of Tokai University.Preoperative serum levels of sialic acid were examined in 293 resected gastric cancer patients with histologically confirmed primary adenocarcinoma. Their ages ranged from 21 to 87 years (median: 58 years), and 73.4% were men. They were selected for chemotherapy or immunochemotherapyby cancer stage on the basis of gross findings accordingto the Japanese Research Society for gastric cancer [ l 2]. The use of chemotherapeuticregimens was left to the judgment of doctors. With the exclusion of 57 cases, in which tumor invaded the mucosa without lymph node metastasis, 236 of 293 patients were analyzed for survival and effectiveness of postoperative adjuvant therapy. The patients who refused postoperative therapy themselves or those whose doctors did not indicate postoperative adjuvant therapy underwent gastrectomy alone and were used as controls. Of the 236 patients, 94 underwent total gastrectomy and 142 had subtotal gastrectomy (128 distal gastrectomy, 14 proximal gastrectomy). Twenty-twopatients underwent gastrectomywith incompleteremoval of perigastric lymph nodes within 3 cm of the edge of the primary tumor (R0 dissection of regional lymph nodes), while 20 had R1 dissection (completeremoval of perigastric lymph nodes within 3 cm of the edge of the primary tumor), and 194 had R2 dissection (complete removal of perigastric lymph nodes and the lymph nodes along the left gastric,
176
Table 1. Values/scores of variables used in the Cox regression model Prognostic factor
Values/scores
pTNM stage a
1.
IA; 2, IB; 3, II; 4, IIIA; 5, IIIB; 6, IV
Therapy
1. 2.
Gastrectomy and chemotherapy Gastrectomy and chemotherapy plus PSK b
Sialic acid before operation
Serum sialic acid level of each patient
Based on histological finding according to the 1987 Unio Intemationalis Contra Cancrum (UICC) [30] b Protein-bound polysaccharide K a
common hepatic, splenic, and celiac arteries). All resected specimens were subjected to a detailed pathological examination, and the patients were then staged according to the 1987 Unio Internationalis Contra Cancrum (UICC) on pTNM staging [30]. For 42 patients, the sole therapy was total or subtotal gastrectomy; 98 cases received gastrectomy plus postoperative adjuvant chemotherapy, which consisted of intravenous administration of mitomycin C (Kyowa Hakko Kogyo Co. Ltd., Tokyo, Japan) (20 mg intraoperatively and 10 mg on postoperative day 1). This was followed on postoperative day 14 with the oral administration of fluoropyrimidines, such as Nl-(2'-tetrahydrofuryl)-5-fluorouracil (Futraful) (Taiho Pharmaceutical Co. Ltd., Tokyo, Japan) (600 mg/day) or 5-fluorouracil (Kyowa Hakko Kogyo Co., Ltd., Tokyo, Japan) (150 mg/day). Of these patients, 16 received mitomycin C alone, 39 fluoropyrimidines alone, and 43 mitomycin C plus fluoropyrimidines. A total of 96 patients received postoperative adjuvant chemotherapy plus a nonspecific immunopotentiator, PSK (Kureha Chemical Industry Co. Ltd., Tokyo, Japan) (3.0 g/day), starting on postoperative day 14 after gastrectomy. Chemotherapeutic regimens of immunochemotherapy were as follows: 6 received mitomycin C alone, 49 fluoropyrimidines alone, and 41 mitomycin C plus fluoropyrimidines. Out of 98 patients, 78 (79.6%) received chemotherapy, and 73 out of 96 patients (76.1%) who received immunochemotherapy had oral administration of fluoropyrimidines over a period of 3 months. Out of 96 patients, 80 (83.3%) received immunochemotherapy and were given PSK orally over a period of 3 months.
Survival was assessed from the day of surgery until death or the most recent update. Follow-up information for all patients was obtained by direct patient contact or by telephone contact with the patients or their families in December 1991. The percentages of patients who have been followed-up until death or over 5 years and 3 years were 65.3% and 89.8% respectively. Survival curves were calculated using the Kaplan-Meier product-limit estimate [14] and differences in survival distribution were assessed by the generalized Wilcoxon test and the log-rank test [7, 25]. Operative deaths only were excluded. Multivariate analyses utilizing Cox's model were performed with variables such as pTNM stages, postoperative adjuvant therapy and preoperative serum levels of sialic acid [ 1, 3, 13] in patients with chemotherapy and immunochemotherapy. The values/scores of variables employed in the Cox regression model are shown in Table 1. Blood samples were collected by venipuncture, allowed to clot at room temperature, and then centrifuged at 3000 rpm. The serum was then stored at -70 ° C for subsequent analysis of serum levels of sialic acid. The sialic acid level was determined by using a modified resorcinol reagent kit (Kyokuto sialic acid test, Kyokuto Pharmaceutical Co. Ltd., Tokyo, Japan) and has been described in detail elsewhere [8]. As the mean and standard deviation of sialic acid values for 61 healthy subjects were 63.7_+ 10.8 mg/dl (37.5-91.5 mg/dl), patients were divided into three groups according to the preoperative serum levels of sialic acid: less than 74.5 mg/dl (1 standard deviation), 74.5 mg/dl and over to less than 85.3 mg/dl (2 standard deviations) and 85.3 mg/dl and over. In this study the upper limit of normal sialic acid was defined as 85.3 mg/dl [22].
Results T a b l e 2 s h o w s t h e s e r u m l e v e l s o f s i a l i c a c i d a c c o r d i n g to the histopathological grading and stage grouping. Serum l e v e l s o f s i a l i c a c i d i n c r e a s e d i n r e l a t i o n to t u m o r p r o g r e s sion. H o w e v e r , t h e r e w a s a n o c o r r e l a t i o n b e t w e e n c e l l u l a r differentiation and sialic acid level. F i g u r e 1 s h o w s t h e s u r v i v a l c u r v e s o f 2 9 3 p a t i e n t s acc o r d i n g to t h e s e r u m l e v e l s o f p r e o p e r a t i v e s i a l i c a c i d : l e s s
Table 2. Serum levels of sialic acid (mg/dl) according to the histopathological grading and stage grouping Stage
Total
Histopathological grading a Well differentiated
Moderately differentiated
Poorly differentiatedb
IA+IB
67.2_+ 16.3 (n = 154)
67.0-+ 13.6 (n = 57)
67.8 ___12.6 (n = 27)
67.1 -+ 19.1 (n = 67) {66.8+ 19.2 (n = 39)}
II
69.6_+16.5 ( n = 41)
72.2+_16.7 ( n = 9)
64.7 -+_11.4 (n = 12)
70.5 -+ 18.8 (n = 18) {62.8_+ 8.9 (n = 10)}
IIIA+IIIB
74.9_+ 18.3 (n = 45)*~
73.5 _+21.1 (n = 12)
80.5 _+15.3 (n =
72.7 _+18.6 (n = 22) {77.9_+ 6 . 4 ( n = 4)}
IV
8 7 . 9 + _ 2 1 . 9 (n = 5 3 ) *2
98.1 +_20.1 (n = 7) *3
88.7 _+ 19.5 (n = 15) *5
85.7 _+24.0 (n = 29) *6 {81.0_+27.9 (n = 1 0 ) } *7
Total
72.4-+ 19.3 (n = 293)
71.0 _+ 17.6 (n = 85)
74.1 -+ 17.3 (n = 64)
72.4 -+21.2 (n = 136) {69.1 -+ 19.2 (n = 63)}
a Excluded, 8 patients with mucinous adenocarcinoma. Their mean serum level of sialic acid was 74.3 _+ 19.6 mg/dl b { }, Signet ring cell formation in poorly differentiated adenocarcinoma • 1 vs stages IA+IB (P = 0.0037) • 2 vs stages IA+IB (P
