|Year : 2019 | Volume
| Issue : 2 | Page : 298-304
Prospective and prognostic factors for hepatic metastasis of gastric carcinoma: A retrospective analysis
Jin Cheng Song, Xiao Lei Ding, Yang Zhang, Xian Zhang, Xiu Hua Sun
Department of Oncology, The Second Affiliated Hospital of Dalian Medical University, Zhongshan, Dalian, Liaoning Province, China
|Date of Web Publication||1-Apr-2019|
Dr. Xiao Lei Ding
Department of Oncology, The Second Affiliated Hospital of Dalian Medical University, Zhongshan, Road No. 467, Shahekou District, Dalian 116027, Liaoning Province
Source of Support: None, Conflict of Interest: None
Aims: The aim of the study was to prospectively explore the prognostic factor for gastric cancer with liver metastasis (GCLM), since no prognostic factor was reported to be consistently significant across studies.
Patients and Methods: One hundred and five patients with GCLM treated at our center between January 1, 2010, and March 31, 2016, were included and their clinical data were retrospectively analyzed. The univariate analyses were first applied for identify the potential independent prognostic and predictive factors for liver metastasis. These factors were further evaluated with Cox proportional-hazard regression model testing. Finally, survival curves were estimated.
Results: The Eastern Cooperative Oncology Group (ECOG) score, number of other distant metastases, levels of cancer antigen (CA), and carcinoembryonic antigen (CEA) were independent prognostic factors (adjusted relative risk [RR]: 1.362–2.887; P = 0.000–0.027). The survival of patients who received radical gastrectomy would be associated with the ECOG score, staging (T stage and N stage), CA 19-9, and CEA levels (RR: 2.169–3.787; P = 0.000–0.027). Patients with following indicators 1 month postoperatively were prone to liver metastasis after radical gastrectomy (median, 6.9–12.03 months; P = 0.007–0.042): Venous/lymphatic invasion, pathological Stage IV (especially combined with T4 stage), intestinal Lauren type, and combined elevation of CEA and CA 19-9 levels.
Conclusions: The therapy design for patients with GCLM should consider the general conditions and personal clinicopathological characters of patients. After balancing the benefit and risk factors, multidisciplinary treatment and individual treatment should be developed based on evidence-based medicine model for each patient.
Keywords: Gastric cancer, liver metastasis, prognostic factor, retrospective study
|How to cite this article:|
Song JC, Ding XL, Zhang Y, Zhang X, Sun XH. Prospective and prognostic factors for hepatic metastasis of gastric carcinoma: A retrospective analysis. J Can Res Ther 2019;15:298-304
|How to cite this URL:|
Song JC, Ding XL, Zhang Y, Zhang X, Sun XH. Prospective and prognostic factors for hepatic metastasis of gastric carcinoma: A retrospective analysis. J Can Res Ther [serial online] 2019 [cited 2019 May 21];15:298-304. Available from: http://www.cancerjournal.net/text.asp?2019/15/2/298/255100
| > Introduction|| |
Gastric cancer (GC) is the fourth most common cancer in the world, which accounted for approximately 8.6% of all new cancers. The mortality rate of GC has been second only to that of lung cancer. In 2016, its estimated 26,370 people were diagnosed as GC and 10,730 people were predicted to eventually die of this disease in the United States. However, more than half of patients with GC have been diagnosed at an advanced stage, losing the opportunity to receive radical surgery. Although surgical treatment could be curative, the recurrence rate remained higher for these patients.
The liver metastasis has been most common in cases with advanced GC. Liver metastasis could be observed in 4%–14% of patients with GC at the time of diagnosis, as well as half of those patients at the time of death. The prognosis of GC with liver metastasis (GCLM) was very poor, with a 5-year survival rate of <10%., Thus, several clinical studies have tried to explore the prognostic and predictive factors associated with GCLM. Unfortunately, extensive prospective data remained to be lack and no prognostic factor was reported to be consistently significant across studies., For this reason, this study aimed to estimate the risk factors for hepatic metastasis after curative surgical treatment and to investigate prognostic factors in patients with GCLM. The results may be conductive to the rational selection and accurate implementation of treatment after surgery, as well as the planning of follow-up programs. The early diagnosis of recurrence and the prognosis of GC may be generally improved.
| > Patients and Methods|| |
One hundred and five patients with GCLM who were treated at the Department of Oncology, the Second Affiliated Hospital of Dalian Medical University, between January 1, 2010, and March 31, 2016, were included in our study. Their clinical data were collected and analyzed retrospectively. Fifty-seven patients (the ratio of male: female: 3.38:1) received radical gastrectomy. This study was reviewed and approved by the Ethics Committee of Dalian Medical University, and informed consent was obtained from patients or their family members. Liver metastasis was confirmed by imaging examinations (including ultrasound, computed tomography, or magnetic resonance imaging). Patients who died of cardio-cerebrovascular or other nontumor-related diseases, as well as those who were lost to follow up, were excluded from the study. Follow-up was conducted through the consultation of the hospital's clinical database or telephone conversations with patients or their family members. The follow-up period was extended until death or March 31, 2017. Survival time (in months) was calculated from the date of liver metastasis diagnosis to the end of follow-up. Disease stage at the time of initial liver diagnosis was defined with the pathological tumor, nodes, and metastasis (TNM) system for surgical cases and the clinical TNM system for nonsurgical cases, according to the seventh edition of the American Joint Committee on Cancer Staging Manual.
The following demographic and clinical variables were evaluated at the time of liver metastasis diagnosis: age (<70 or ≥70 years), sex (male or female), performance status according to Eastern Cooperative Oncology Group (ECOG) criteria (<3 or ≥3 years), and radical gastrostomy (receiving or not). The following clinicopathological variables were evaluated by experienced pathologists: histological type (diffuse [poorly differentiated adenocarcinoma, signet-ring cell, and mucinous carcinomas] or intestinal [papillary, well-differentiated, and moderately differentiated adenocarcinomas] Lauren type), pathological stage (I/II, III, or IV), T stage (T1/T2, T3, or T4), and N stage (N0/N1 or N2/N3). Staging variables were also monitored when patients receiving primary gastric resection. No follow-up data on Lauren type were available for patients with advanced disease who did not receive gastrectomy. Data on the timing of liver metastasis detection (synchronous [at the time of primary tumor diagnosis or within 1–6 months of primary gastric resection] or metachronous), distribution of liver metastasis (unilobar or bilobar), number of liver metastases (solitary or multiple), maximum diameter of liver metastases (<30 or ≥30 mm), number of other distant metastases, treatment mode of liver metastasis (best supportive care [BSC], systematic chemotherapy, or systematic chemotherapy + local treatment/palliative operation), and the levels of carcinoembryonic antigen (EA), cancer antigen (CA) 19-9, and CA 72-4 (the cutoff value for defining positivity: 5 ng/ml for CEA, 37 U/ml for CA 19-9, and 7 U/ml for CA 72-4) were also collected.
Statistical analyses were performed with SPSS for Windows version 19.0 (SPSS Inc., Chicago, IL, USA). Univariate analyses were conducted with the t-test and analysis of variance for continuous variables. Then, the potential predictive factors for each liver metastasis pattern were identified. Survival curves were estimated with the Kaplan–Meier method and compared with the log-rank test. Cox proportional hazard regression models were applied to estimate hazard ratios models. P < 0.05 was considered to be statistically significant.
| > Results|| |
There were 79 (75.2%) male and 26 (24.8%) female patients (the ratio of male:female: 3.03:1) included in this study. At the time of liver metastasis diagnosis, the median age was 63 (ranged 41–84) years. Seven patients were alive at the end of follow-up. The mean survival time after liver metastasis diagnosis was 13.1 (95% confidence interval [CI]: 10.402–15.796) months. The median survival time was 9.2 (95% CI: 7.192–11.208) months. The 1-, 2-, and 3-year survival rates were 31.4%, 13.1%, and 6.9%, respectively. With the results of single-variable Kaplan–Meier analyses, the significant prognostic factors were explored including sex, ECOG score, distribution of liver metastasis, maximum diameter of liver metastasis, number of other distant metastases, hepatic metastasis treatment mode, and elevated CEA level [P ≤ 0.001 to P = 0.038; [Table 1]. With Cox regression analysis, the results showed that the ECOG score (relative risk [RR]: 2.887; 95% CI: 1.827–4.561; P = 0.000), elevated CEA level (RR: 2.037; 95% CI: 1.327–3.126; P = 0.001), and number of other distant metastases (RR: 1.362; 95% CI: 1.036–1.791; P = 0.027) were independent prognostic factors [Table 2].
|Table 1: Clinicopathological data and univariate data on survival after liver metastasis diagnosis|
Click here to view
|Table 2: Results of Cox regression analysis of survival after liver metastasis diagnosis|
Click here to view
Patients receiving radical gastrectomy
Fifty-seven patients (the ratio of male:female: 3.38:1) received radical gastrectomy. The median age at the time of radical gastrectomy was 60 (ranged 38–83) years. Five patients were alive at the end of follow-up. The mean survival time was 15.3 (95% CI: 11.040–19.592) months and the median survival time was 11.3 (95% CI: 9.417–13.183) months. The 1-, 2-, and 3-year survival rates were 36.8%, 17.4%, and 8.7%, respectively. With the results of single-variable Kaplan–Meier analyses, significant prognostic factors were explored, including T stage, N stage, hepatic metastasis treatment mode, ECOG score, number of other distant metastases, elevated CEA level, and elevated CA 19-9 level [P ≤ 0.001 to P = 0.046; [Table 3]. The results of Cox regression analysis showed that T stage levels were independent prognostic factors [Table 4].
|Table 3: Univariate data on survival after liver metastasis diagnosis in patients who underwent radical gastrectomy|
Click here to view
|Table 4: Survival after liver metastasis diagnosis in patients who underwent radical gastrectomy (based on Cox regression)|
Click here to view
Potential predictive factors for liver metastasis
The mean and median times from surgery to liver metastasis diagnosis were 14.3 and 10.0 months, respectively. With the univariate analysis, the significant predictive factors were venous/lymphatic invasion, pathological stage, T stage, Lauren type, and elevated CEA level combined with elevated CA 19-9 level at 1 month postoperatively [P = 0.007–0.042; [Table 5].
| > Discussion|| |
The prognosis of GC was typically poor since the recurrence and diagnosis at advanced stages. The recurrence in liver was commonly observed in approximately 21.8%–49.5% of GC patients. The incidence of liver metastasis was up to 13.5%–30.0% after gastrectomy. Even in early GC, the rate of liver metastasis after curative gastrectomy was up to 2%. The reported 1-, 3-, and 5-year survival rates of GCLM were 51.3%–79.2%, 20.0%–33.3%, and 2.6%–10%, respectively, with a median survival time of 5–34 months.,, The survival rates calculated in this study have also fallen within these ranges. The outcome of GC would be worse with unsuitable hepatic metastasis resection, with 1-, 3-, and 5-year survival rates of 21.1%, 1.1%, and 0%, respectively. These survival rates were slightly lower than those obtained in this study, which may be largely resulted from recent improvements in multidisciplinary treatment.
The prognostic factors for GCLM have been investigated in many studies. Turanli reported that the location and number of liver metastases, presence of ascites, surgical margins, and CEA and CA 19-9 levels were independent prognostic factors. Other factors, such as the stage of primary GC, timing of hepatectomy, extent of lymph node metastasis, and therapeutic strategy, have also been associated with survival. In our study, survival after diagnosis of liver metastasis was associated with the ECOG score, number of other distant metastases, and elevated CEA level, regardless of whether patients received radical gastrectomy. A higher ECOG score was always accompanied with a greater number of other distant metastases. At this time point, the general conditions of these patients were very poor, thus unable to tolerate the systemic chemotherapy and local therapy, such as hepatic artery infusion (HAI), transarterial chemoembolization (TACE), or radiofrequency ablation (RA). Previous studies have reported better outcomes were observed for patients receiving these treatments. Thus, the poor conditions of patients in our study also contributed to the shortened survival time. In our study, the survival time was 11.5 months for patients receiving systemic chemotherapy + local treatment/palliative operation, which was much longer than those receiving BSC or systemic chemotherapy alone. Most liver metastases from GC were multiple, bilobar and combined with peritoneal or extensive lymph node metastasis,, while local treatment was only suitable for only a few patients with single, isolated lesions measuring <5 cm located in one hemi-liver. The survival seemed to be better for liver metastases' resection in selected patients. A meta-analysis with systematic review identified that some patients with liver-confined performance status and disease and amenable of surgical resection were ideal candidates for liver resection; however, the rate was <1% of GC patients. Approximately 900 patients from 23 studies were pooled, the reported median OS were 22 months and 5-year OS rate were 24% after hepatic resection of GC liver metastases. Hepatectomy was not contemplated according to current guidelines in patients with Stage IV GC for the lack of data from prospective trials. Thus, most patients with GCLM would have no opportunity to receive local treatment/palliative operation, leading to a poor outcome. Performance status was strongly associated with prognosis for patients with GCLM. Therefore, appropriate treatment should be selected to avoid overtreatment and decrease the treatment refusal for patients with poor performance status. For those showing good performance, a stronger chemotherapy regimen and/or local treatment (HAI, TACE, and RA) or palliative operation could be selected.
The metastasis of GC to the liver was primarily developed through hematogenous dissemination, lymphatic dissemination, and serous invasion of the primary tumor. Hematogenous metastasis has been the most generally accepted theory for the development of liver metastasis. Deeper invasion has been associated with a higher incidence of hematogenous or lymphatic metastasis, and the reported 1-year survival rates in T1, T2, T3, and T4 cases were 50%, 52%, 27%, and 0%, respectively. The risk of lymph node metastasis in GC with submucosal invasion was four times higher than that of in mucosal GC. Among patients with no peritoneal metastasis, the prognosis was worse for patients with N2/N3 stages, compared with those with N0/N1 stages. In our study, the better outcome was also observed in patients with T1/T2 and N0/N1 stages. Patients with advanced pathological stage GC, especially with deeper invasion, should receive adjunct therapy as early as possible to reduce the risk of liver metastasis and prolong the survival time.
In prospective research, the reported independent risk factors for liver metastasis were submucosal invasion, vascular involvement, tumor differentiation, elevated level of tumor marker (preoperative CEA and alpha-fetoprotein), extranodal invasion, and GC size.,, Liver metastasis after radical surgical treatment was prone to be observed in patients with high preoperative CEA levels, lymph node involvement, and Lauren intestinal histotype. The median interval between gastric resection and the diagnosis of hepatic metastasis was 10–13.1 months. In our study, this interval was 10.0 months, and the occurrence of metastasis was approximately 3 months earlier in patients with the intestinal type of GC than in those with the diffuse type, which was related to the clinicopathological features. The intestinal type accounted for a larger proportion of cases in patients aged >60 years and was characterized by a high incidence of venous or lymphatic metastasis., Both of which may explain its greater tendency of liver metastasis.
Tumor markers would be increased several months before the imaging abnormalities. Several studies have documented the significantly more frequent CEA positivity in patients with liver metastasis, as well as the increased CA 19-9 positivity in patients with lymph node, peritoneal, and serous involvement. For CEA-positive patients, there were larger tumors, deeper serosal invasion, and more frequent lymphatic and vascular involvement than that of in CEA-negative patients., The measurement of serum CEA levels was useful for the prediction of progression and prognosis. Univariate and multivariate analyses have shown that CEA levels <50 ng/ml were significantly associated with poor overall survival. In our study, CEA levels were found to be higher in patients with advanced GC and more nonhepatic distant metastases, higher EOCG scores, and larger liver metastasis lesions. These findings indicated a heavy tumor burden and thus a poor prognosis. Marrelli reported that the combination of three markers (CEA, CA 19-9, and CA 72-4) was highly sensitive for predicting liver metastasis. We also demonstrated that patients with higher CEA and CA 19-9 levels 1 month postoperatively were more associated with liver metastasis. In addition, the elevated CEA and CA 19-9 levels at the time of liver metastasis diagnosis were significantly associated with worsened survival. Therefore, the levels of these markers should be followed closely after surgery, especially 1 month postoperatively, in patients with GC. The elevation of both levels always indicated a poor clinical outcome.
| > Conclusions|| |
In this study, the performance status of patients with GCLM was strongly associated with prognosis. For patients in poor conditions, appropriate treatment should be selected for avoiding overtreatment and decreasing the treatment refusal. For patients with advanced GC, especially with deeper invasion, adjunct therapy should be applied as early as possible to reduce the risk of liver metastasis and prolong the survival. The pathological stage and the levels of CEA and CA 19-9 were not only associated with prognosis but also related to liver metastasis. Thus, all these indicators should be closely followed postoperatively. The intestinal type of GC was more prone to liver metastasis, which was also associated with the clinicopathological features. The development of therapeutic strategies for patients with GCLM must consider their general conditions and individual clinicopathological characteristics for balancing benefits and risks. Multidisciplinary and individual treatments should be considered, according to the evidence-based medicine models in each case.
There were also several limitations in this study including its retrospective design and small sample size. Despite these limitations, an awareness of prospective and prognostic factors would be useful for selecting appropriate and rational treatment. It would be greatly beneficial to improve the prognosis of patients with GC.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
| > References|| |
Jemal A, Bray F, Center MM, Ferlay J, Ward E, Forman D, et al.
Global cancer statistics. CA Cancer J Clin 2011;61:69-90.
Siegel RL, Miller KD, Jemal A. Cancer statistics, 2016. CA Cancer J Clin 2016;66:7-30.
Shen YH, Chen S, Zhang WT, Ji Y, Yu L, Sun HC, et al.
Clinical analysis of gastroenteropancreatic neuroendocrine tumor with liver metastasis, compared with primary hepatic neuroendocrine tumor. J Cancer Res Ther 2014;10:276-80.
Kerkar SP, Kemp CD, Avital I. Liver resections in metastatic gastric cancer. HPB (Oxford) 2010;12:589-96.
Mastoraki A, Benetou C, Mastoraki S, Papanikolaou IS, Danias N, Smyrniotis V, et al.
The role of surgery in the therapeutic approach of gastric cancer liver metastases. Indian J Gastroenterol 2016;35:331-6.
Kakeji Y, Morita M, Maehara Y. Strategies for treating liver metastasis from gastric cancer. Surg Today 2010;40:287-94.
Martella L, Bertozzi S, Londero AP, Steffan A, De Paoli P, Bertola G, et al.
Surgery for liver metastases from gastric cancer: A meta-analysis of observational studies. Medicine (Baltimore) 2015;94:e1113.
Hu X. Risk factors and prognosis of liver metastasis from gastric cancer. Zhonghua Wei Chang Wai Ke Za Zhi 2014;17:108-11.
Jerraya H, Saidani A, Khalfallah M, Bouasker I, Nouira R, Dziri C, et al.
Management of liver metastases from gastric carcinoma: Where is the evidence? Tunis Med 2013;91:1-5.
Khazaei S, Rezaeian S, Soheylizad M, Khazaei S, Biderafsh A. Global incidence and mortality rates of stomach cancer and the human development index: An ecological study. Asian Pac J Cancer Prev 2016;17:1701-4.
Tiberio GA, Roviello F, Donini A, de Manzoni G; Italian Research Group for Gastric Cancer. Hepatic metastases from gastric cancer: A surgical perspective. World J Gastroenterol 2015;21:11489-92.
Liu J, Chen L. Current status and progress in gastric cancer with liver metastasis. Chin Med J (Engl) 2011;124:445-56.
Ishida M, Morita S, Saka M, Fukagawa T, Taniguchi H, Katai H, et al.
Metachronous liver metastasis from early gastric cancer. J Gastrointest Surg 2012;16:837-41.
Ikeguchi M, Saito H, Tatebe S, Wakatsuki T. Outcome of treatment of liver metastasis after curative surgery for gastric cancer. Am Surg 2011;77:1274-6.
Liu J, Li JH, Zhai RJ, Wei B, Shao MZ, Chen L, et al.
Predictive factors improving survival after gastric and hepatic surgical treatment in gastric cancer patients with synchronous liver metastases. Chin Med J (Engl) 2012;125:165-71.
Li J, Cui J, Shen W, Xi H, Liu S, Chen L, et al.
Comparison of clinical efficacy of different treatment methods for synchronous liver metastasis from gastric cancer. Zhonghua Wei Chang Wai Ke Za Zhi 2014;17:117-20.
Turanli S. The value of resection of primary tumor in gastric cancer patients with liver metastasis. Indian J Surg 2010;72:200-5.
Kim HR, Cheon SH, Lee KH, Ahn JR, Jeung HC, Lee SS, et al.
Efficacy and feasibility of radiofrequency ablation for liver metastases from gastric adenocarcinoma. Int J Hyperthermia 2010;26:305-15.
Tao F, Lv J, Wang W, Jin K. Clinical modalities for management of gastric cancer hepatic metastasis. Int J Clin Exp Med 2015;8:19850-8.
Chen J, Tang Z, Dong X, Gao S, Fang H, Wu D, et al.
Radiofrequency ablation for liver metastasis from gastric cancer. Eur J Surg Oncol 2013;39:701-6.
Petrelli F, Coinu A, Cabiddu M, Ghilardi M, Borgonovo K, Lonati V, et al.
Hepatic resection for gastric cancer liver metastases: A systematic review and meta-analysis. J Surg Oncol 2015;111:1021-7.
Li F, Zhang R, Liang H, Liu H, Quan J. The pattern and risk factors of recurrence of proximal gastric cancer after curative resection. J Surg Oncol 2013;107:130-5.
Ueda K, Iwahashi M, Nakamori M, Nakamura M, Naka T, Ishida K, et al.
Analysis of the prognostic factors and evaluation of surgical treatment for synchronous liver metastases from gastric cancer. Langenbecks Arch Surg 2009;394:647-53.
Hanaoka N, Tanabe S, Mikami T, Okayasu I, Saigenji K. Mixed-histologic-type submucosal invasive gastric cancer as a risk factor for lymph node metastasis: Feasibility of endoscopic submucosal dissection. Endoscopy 2009;41:427-32.
Tiberio GA, Coniglio A, Marchet A, Marrelli D, Giacopuzzi S, Baiocchi L, et al.
Metachronous hepatic metastases from gastric carcinoma: A multicentric survey. Eur J Surg Oncol 2009;35:486-91.
Wang W, Zhang R, Zhang H, Deng J, Wang X, Xue Q, et al.
Risk factors in metachronous liver metastasis from gastric cancer. Zhonghua Wei Chang Wai Ke Za Zhi 2014;17:121-3.
Ohno S, Fujii T, Ueda S, Nakamoto T, Kinugasa S, Yoshimura H, et al.
Predictive factors and timing for liver recurrence after curative resection of gastric carcinoma. Am J Surg 2003;185:258-63.
Marrelli D, Roviello F, De Stefano A, Fotia G, Giliberto C, Garosi L, et al.
Risk factors for liver metastases after curative surgical procedures for gastric cancer: A prospective study of 208 patients treated with surgical resection. J Am Coll Surg 2004;198:51-8.
Garancini M, Uggeri F, Degrate L, Nespoli L, Gianotti L, Nespoli A, et al.
Surgical treatment of liver metastases of gastric cancer: Is local treatment in a systemic disease worthwhile? HPB (Oxford) 2012;14:209-15.
Qiu MZ, Cai MY, Zhang DS, Wang ZQ, Wang DS, Li YH, et al.
Clinicopathological characteristics and prognostic analysis of lauren classification in gastric adenocarcinoma in China. J Transl Med 2013;11:58.
Nesi G, Basili G, Girardi LR, Manetti A, Biliotti G, Barchielli A, et al.
Pathological predictors of lymph node involvement in submucosal gastric carcinoma: A retrospective analysis of long-term outcome. In Vivo
Shimada H, Noie T, Ohashi M, Oba K, Takahashi Y. Clinical significance of serum tumor markers for gastric cancer: A systematic review of literature by the Task Force of the Japanese Gastric Cancer Association. Gastric Cancer 2014;17:26-33.
Ikeda Y, Mori M, Kajiyama K, Kamakura T, Maehara Y, Haraguchi Y, et al.
Indicative value of carcinoembryonic antigen (CEA) for liver recurrence following curative resection of stage II and III gastric cancer. Hepatogastroenterology 1996;43:1281-7.
Ucar E, Semerci E, Ustun H, Yetim T, Huzmeli C, Gullu M, et al.
Prognostic value of preoperative CEA, CA 19-9, CA 72-4, and AFP levels in gastric cancer. Adv Ther 2008;25:1075-84.
Teker F, Canbaz F, Kemal Y, Yucel I. A case of gastric cancer with liver metastases had a complete response with cisplatin and capecitabine as third-line chemotherapy. J Cancer Res Ther 2015;11:1026.
[Table 1], [Table 2], [Table 3], [Table 4], [Table 5]