Summary
The indication for resection of gastric/esophagogastric-junction liver metastasis (GELM) has yet to be established. This study aimed to investigate prognostic factors in patients undergoing GELM resection. From 2001 to 2015, 31 consecutive patients underwent resection for GELM; and factors for poor prognosis were evaluated. Of the 31 patients, 23 (74.2%) developed multiple liver metastases. The histology of gastric cancer was intestinal-type adenocarcinoma in 21 patients (67.7%). Median overall survival (OS) was 3.2 years. The 1-, 3-, and 5-year OS rates were 92.8%, 56.2%, and 42.2%, respectively. The 1-, 3-, and 5-year recurrence-free survival (RFS) rates were 58.5%, 31.3%, and 31.3%, respectively. Multivariate analysis indicated that intestinal-type adenocarcinoma was associated with a significantly lower risk of OS (hazard ratio [HR], 0.26; p =0.022) and RFS (HR, 0.25; p = 0.008). In multiple logistic regression analysis, intestinal-type adenocarcinoma (odds ratio, 0.14; p = 0.012) reduced incidence of extra-hepatic recurrence after GELM resection. In conclusion, GELM resection in patients with intestinal-type histology is preferable because intestinal-type adenocarcinoma is associated with better prognosis and a lower incidence of extra-hepatic recurrence than diffuse/other-type adenocarcinoma.
Keywords: Gastric liver metastasis, gastric/esophagogastric-junction liver metastasis, the intestinal-type adenocarcinoma, liver resection
Introduction
The prognosis of gastric cancer has improved over the last two decades, but it remains the third highest cause of cancer-related death worldwide (1-3). Surgical resection of the stomach is the mainstay of management for resectable gastric cancer, but cumulative recurrence rates still remain high; 79% within 2 years of operation (1). Liver is one of the major organs that develop gastric cancer metastases, with an incidence of 4-34% (2).
Although chemotherapy is regarded as the standard treatment for gastric/esophagogastric-junction liver metastases (GELMs), several retrospective studies have reported favorable prognosis for liver resection concerning GELM (3-7). These studies demonstrated the following risk factors for poor prognosis: number and maximum size of liver metastases; R1/R2 resection; synchronous metastases; primary tumor stage pT4; and the presence of other distal metastases. However, the study periods were mainly limited before the year 2000. The appropriate indication criteria for GELM resection are still debatable because effective chemotherapies for gastric cancer were introduced in the early 2000s. Additionally, factors for poor prognosis regarding liver resection for GELM are not well established, as compared with those for colorectal liver metastases.
The aim of this study was to investigate prognostic factors for GELM by evaluating patients who underwent liver resection for GELM.
Materials and Methods
Indication for liver resection for GELMs
Liver resection was indicated for three or fewer GELMs without metastases at other sites, based on previous reports (6). In patients with four or more GELMs, preoperative chemotherapy was performed. In cases where no extra-hepatic gastric metastases occurred after chemotherapy, liver resection was indicated. Simultaneous resection of the stomach and GELM was performed for synchronous GELMs, when they were easily removed using limited non-anatomic liver resection. The final surgical procedures were planned to resect all GELMs to secure negative histologic margins.
Definition of histology
Histopathological classification of gastric cancer was classified into three groups, intestinal type, diffuse type, and other type, based on the criteria of Japanese Classification of Gastric Cancer third edition. Intestinal-type adenocarcinoma was defined as a tumor with glandular architecture, resembling colonic carcinoma, whereas diffuse-type adenocarcinoma was defined as a tumor composed of solitary or small clusters of cells, and lacking glandular structures. Gastric cancer with uncommon variant was classified as other type (8).
Patients
Between January 2001 and December 2015, 31 consecutive patients underwent liver resection for GELM at the University of Tokyo Hospital. The clinical records of these patients were retrospectively reviewed from a prospectively maintained database. Patient characteristics are summarized in Table 1. All operations were performed after obtaining informed consent from each patient, and all aspects of the procedures were conducted according to the principles expressed in the Declaration of Helsinki. In the preparation of this study, all efforts have been made to protect patient privacy and anonymity. The study was approved by the institutional review board at the University of Tokyo (2158-5).
Table 1. Patient characteristics.
| Variables | Value |
|---|---|
| Number of patients | 31 |
| Patient factor | |
| Age, years [range] | 73 [47-84] |
| Sex, n (%) | |
| Male | 27 (87.1) |
| Female | 4 (12.9) |
| ASA score, n (%) | |
| 1 | 13 (41.9) |
| 2 | 18 (58.1) |
| 3 | 0 (0.0) |
| BMI, kg/m2 [range] | 21.6 [14.6-30.7] |
| AFP, U/mL [range] | 9.0 [1.0-32.5×105] |
| CEA, ng/mL [range] | 9.0 [1.0-5370] |
| CA19-9, U/mL [range] | 100 [69-100] |
| Primary lesion factors | |
| Location, n (%) | |
| Esophagogastric junction | 4 (12.9) |
| Upper | 5 (16.1) |
| Middle | 11 (35.5) |
| Lower | 11 (35.5) |
| Maximum size, cm [range] | 4.0 [0.4-19] |
| Histology, n (%) | |
| Diffuse/other-type adenocarcinoma | 10 (32.3) |
| Intestinal-type adenocarcinoma | 21 (77.7) |
| T classification, n (%) | |
| T1 | 9 (29.0) |
| T2 | 3 (9.7) |
| T3 | 12 (38.7) |
| T4 | 7 (22.6) |
| N classification, n (%) | |
| N0 | 15 (48.4) |
| N1 | 3 (9.7) |
| N2 | 9 (29.0) |
| N3 | 4 (12.9) |
| Liver metastasis factor | |
| Timing of liver metastases, n (%) | |
| Synchronous | 13 (41.9) |
| Metachronous | 18 (58.1) |
| Tumor number, n (%) | |
| 1 | 23 (74.2) |
| 2-3 | 5 (16.1) |
| ≥ 4 | 3 (9.7) |
| Maximum tumor size, cm [range] | 3.1 [0.8-22] |
| Tumor distribution, n (%) | |
| Unilobular | 23 (74.2) |
| Bilobular | 8 (25.8) |
| Preoperative chemotherapy, n (%) | 13 (41.9) |
| Regimen of preoperative chemotherapy, n (%) | |
| S-1 and CDDP | 6 (19.4) |
| CPT-11 and CDDP | 2 (6.5) |
| CPT-11 and MMC | 1 (3.2) |
| S-1 and CDDP and Tmab | 2 (6.5) |
| S-1 and Oxaliplatin | 1 (3.2) |
| CDDP and 5-FU | 1 (3.2) |
| wPTX and Tmab | 1 (3.2) |
Abbreviations: ASA, American society of anesthesiologists; BMI, body mass index; AFP, α-fetoprotein; CEA, carcinoembryonic antigen; CA19-9, carbohydrate antigen 19-9; S-1, Tegafur gimestat otastat potassium; CDDP, Cisplatin; CPT-11, Irinotecan; MMC, Mitomycin C; Tmab, Trastuzumab; 5-FU, 5-Fluorouracil; wPTX, weekly Paclitaxel.
Preoperative evaluation
The surgical procedure was planned with reference to tumor location, size, and the results of the volumetric analysis. All patients underwent ultrasonography, plain and contrast-enhanced computed tomography (CT), and magnetic resonance imaging (MRI) for the staging of GELM; they underwent chest X-ray, chest CT, gastroscopy, and, if necessary, positron emission tomography-CT for the surveillance of extra-hepatic metastases. Intraoperative tumor surveillance was performed using visual inspection, manual palpation, and intraoperative ultrasonography, and the final surgical procedures were planned to resect all GELMs and secure negative histologic margins.
Surgical procedures
Liver resection was indicated under criteria based on preoperative liver function parameters, such as the presence/absence of uncontrolled ascites, serum bilirubin level, and indocyanine green retention rate at 15 min (9,10). Non-anatomical limited resection was principally performed to preserve as much liver parenchyma as possible. A major anatomical hepatectomy was performed when liver metastases were adherent to or invading major hepatic vessels and/or were identified in the hemi-liver. After retrieving surgical specimens, the distance between tumors and the cut surface were measured, and the shortest distance from multiple tumors was defined as a surgical margin. When one of the surgical margins was positive, the tumor was defined as having a positive surgical margin. Major hepatectomy was defined as the resection of ≥ 3 contiguous segments, according to Couinauds classification (11).
Statistical analysis
Categorical variables are expressed as numbers (%). Continuous variables are expressed as the median and range. The TNM classification and stage were determined according to the International Union Against Cancer (version 7), when gastrectomy or synchronous gastric and liver resection were performed.
Survival curves were constructed using the Kaplan- Meier method and compared using the log-rank test. Overall survival (OS) was calculated from the day of liver resection in patients undergoing upfront resection, or the initiation of neoadjuvant chemotherapy in patients undergoing neoadjuvant chemotherapy. Loss to follow-up and death without recurrence were censored for the recurrence-free survival (RFS) analysis.
Factors with a p value < 0.05 using the Cox proportional-hazards model were considered as potential risk factors and were further analyzed using a multivariate Cox model. Factors with a p value < 0.05 using logistic regression in univariate analysis were considered as potential predictors and were further analyzed in a multiple logistic regression analysis. Hazard ratios (HR), odds ratio (OR), and 95% confidence interval (CI) were calculated for each factor. The cutoff level for estimated blood loss in our study was set at 1,000 mL, based on previous reports (12). Tumor markers were categorized by institutional upper limits: carcinoembryonic antigen (≥ 5 vs. < 5), carbohydrate antigen 19-9 (≥ 37 vs. < 37), and α-fetoprotein (≥ 9 vs. < 9). Other continuous variables were categorized using the median value. A p value < 0.05 was considered to indicate statistical significance.
Statistical analysis was performed using JMP software (version 11.0.6; SAS Institute Inc., Cary, NC, USA).
Results
Patient characteristics
The median maximum GELM size was 3.1 (range, 0.8-22.0) cm. Histological outcomes of gastric cancer were intestinal type in 21 patients (67.7%) and diffuse type/other type in 10 patients (32.3%). Liver resection for synchronous and metachronous metastases were performed in 13 (41.9%) patients and 18 (58.1%) patients, respectively. Before liver resection for GELM, 45.2% of patients (n = 14) underwent chemotherapy with regimens that mainly included S-1 and/or cisplatin, including three patients (9.7%) who were treated with neoadjuvant chemotherapy.
Intraoperative, postoperative outcomes
Intraoperative and postoperative outcomes are summarized in Table 2. Major hepatectomy was performed in six patients (19.3%). The morbidity rate was 12.9% (n = 4) including no Clavien-Dindo III-V complications. Resection rates of R0, R1, and R2 were 71.0 % (n = 22), 22.5% (n = 7), and 6.5% (n = 2), respectively. Postoperative chemotherapy was prescribed in 15 patients (48.4%). All histopathological findings of GELM were consistent with those of a primary tumor.
Table 2. Intraoperative and postoperative outcomes.
| Variables | Value |
|---|---|
| Number of patients | 31 |
| Intraoperative outcomes | |
| Operative time, min [range] | 358 [146-724] |
| Estimated blood loss, mL [range] | 690 [20-3270] |
| Blood transfusion, n (%) | 10 (32.3) |
| Major hepatectomy, n (%) | 6 (19.3) |
| Postoperative outcomes | |
| Morbidity rate | 16.1% |
| Clavien-Dindo classification, n (%) | |
| ≥ IIIA | 0 (0.0) |
| I-II | 4 (12.9)* |
| Length of hospital stay, days [range] | 14 [5-49] |
| R1 and R2 resection, n (%) | 9 (29.0) |
| Postoperative chemotherapy, n (%) | 15 (48.4) |
*Cholangitis in two patients (6.5%), congestive heart failure in one patient (3.2%), and ileus in one patient (3.2%).
Overall survival and recurrence-free survival
The median follow-up period was 3.3 (range, 0.3-8.4) years. The 1-, 3-, and 5-year OS rates were 92.8%, 56.2%, and 42.2%, respectively. The median OS was 3.2 years (Figure 1A). The 1-, 3-, and 5-year RFS rates were 58.5%, 31.3%, and 31.3%, respectively. The median RFS was 1.4 years (Figure 1B). Recurrence after liver resection for GELM occurred in 18 (58.1%) patients; this included the liver in eight (25.8%), the lung in two (6.5%), the bone in two (6.5%), the lymph nodes in one (3.2%), and the peritoneum in one (3.2%), including multiple site recurrence in four (22.2%).
Figure 1.
(A) Overall survival in patients with gastric/ esophagogastric-junction liver metastasis. (B) Recurrence-free survival in patients with gastric/esophagogastric-junction liver metastasis.
Risk factors for OS and RFS
Intestinal-type adenocarcinoma, major hepatectomy, and R1/R2 resection were found to be significantly associated with OS (Table 3). Of these factors, intestinal-type adenocarcinoma was associated with a significantly lower risk of OS (HR, 0.24; 95% CI, 0.07-0.81; p = 0.022). In contrast, R1/R2 resection (HR, 5.31; 95% CI, 1.40-20.5; p = 0.015) was an independent risk factor for OS. Primary gastric location (esophagogastric junction and upper stomach) and intestinal-type adenocarcinoma were found to be significantly associated with RFS (Table 4). Of the two factors, intestinal-type adenocarcinoma was associated with a lower risk of RFS (HR,0.34; 95% CI, 0.09-0.72; p = 0.008).
Table 3. Univariate and multivariate analysis of overall survival.
| Variables | Univariate Analysis |
Multivariate Analysis |
||||
|---|---|---|---|---|---|---|
| HR | 95% CI | p value | HR | 95% CI | p value | |
| Patient factors | ||||||
| Sex, | ||||||
| Male / Female | 1.11 | 0.20-20.6 | 0.918 | |||
| Age | ||||||
| ≥ 70 years / ≤ 69 years | 1.21 | 0.36-3.86 | 0.756 | |||
| BMI | ||||||
| ≥ 25.0 kg/m2 / ≤ 24.9 kg/m2 | 1.23 | 0.55-2.75 | 0.684 | |||
| ASA | ||||||
| ≥ 2 / ≤ 1 | 0.60 | 0.21-1.45 | 0.183 | |||
| AFP | ||||||
| ≥ 9.0 IU/mL / ≤ 8.9 IU/mL | 0.41 | 0.06-1.80 | 0.251 | |||
| CEA | ||||||
| ≥ 5.0 IU/mL / ≤ 4.9 IU/mL | 0.88 | 0.19-2.75 | 0.787 | |||
| CA19-9 | ||||||
| ≥ 37.0 IU/mL / ≤ 36.9 IU/mL | 0.91 | 0.19-3.40 | 0.902 | |||
| Preoperative chemotherapy | 0.91 | 0.31-2.48 | 0.924 | |||
| Primary cancer-related factors | ||||||
| Tumor location | ||||||
| EGJ, Upper / Middle, Lower | 1.78 | 0.55-4.84 | 0.314 | |||
| Maximum primary tumor size | ||||||
| ≥ 5 cm / ≤ 4.9 cm | 1.72 | 0.59-5.52 | 0.312 | |||
| T classification | ||||||
| ≥ 3 / ≤ 2 | 1.17 | 0.73-2.94 | 0.240 | |||
| N classification | ||||||
| ≥ 1 / ≤ 0 | 1.35 | 0.50-3.94 | 0.552 | |||
| Histological type | ||||||
| Intestinal / Diffuse and other | 0.26 | 0.08-0.85 | 0.027 | 0.24 | 0.07-0.81 | 0.022 |
| Liver metastases-related factors | ||||||
| Timing of liver metastases | ||||||
| Metachronous / Synchronous | 0.68 | 0.63-4.94 | 0.284 | |||
| Tumor number | ||||||
| Multiple / Single | 1.06 | 0.33-2.86 | 0.914 | |||
| Tumor distribution | ||||||
| Bilobular / Unilobular | 1.34 | 0.42-3.68 | 0.586 | |||
| Maximum tumor size | ||||||
| ≥ 5 cm / ≤ 4.9 cm | 0.66 | 0.15-2.09 | 0.518 | |||
| Operative procedures | ||||||
| Synchronous hepatectomy | 1.65 | 0.54-4.62 | 0.354 | |||
| Major hepatectomy | 3.43 | 1.09-44.7 | 0.044 | |||
| Operating time | ||||||
| ≥ 360 min / ≤ 359 min | 1.57 | 0.57-4.45 | 0.371 | |||
| Estimated blood loss | ||||||
| ≥ 1000 mL / ≤ 999 mL | 1.26 | 0.44-3.46 | 0.648 | |||
| Resection | ||||||
| ≥ R1 / R0 | 4.86 | 1.32-17.9 | 0.018 | 5.31 | 1.40-20.5 | 0.015 |
| Postoperative factors | ||||||
| Clavien-Dindo classification | ||||||
| ≥ I / ≤ 0 | 0.81 | 0.59-2.00 | 0.786 | |||
| Duration of hospital stay | ||||||
| ≥ 14 days / ≤ 13 days | 1.34 | 0.48-3.85 | 0.567 | |||
| Postoperative chemotherapy | 1.12 | 0.41-3.09 | 0.815 | |||
Abbreviations: BMI, body mass index; ASA, American society of anesthesiologists; AFP, α-fetoprotein; CEA, carcinoembryonic antigen; CA19- 9, carbohydrate antigen 19-9; EGJ, esophagogastric junction; Por, undifferentiated adenocarcinoma; HR, hazard ratios; CI, confidence intervals.
Table 4. Univariate and multivariate analysis of recurrence-free survival.
| Variables | Univariate Analysis |
Multivariate Analysis |
||||
|---|---|---|---|---|---|---|
| HR | 95% CI | p value | HR | 95% CI | p value | |
| Patient factors | ||||||
| Sex | ||||||
| Male / Female | 1.08 | 0.32-6.72 | 0.938 | |||
| Age | ||||||
| ≥ 70 years / ≤ 69 years | 1.02 | 0.41-2.59 | 0.949 | |||
| BMI | ||||||
| ≥ 25.0 kg/m2 / ≤ 24.9 kg/m2 | 0.73 | 0.38-1.30 | 0.307 | |||
| ASA | ||||||
| ≥ 2 / ≤ 1 | 0.91 | 0.28-2.20 | 0.460 | |||
| AFP | ||||||
| ≥ 9.0 IU/mL / ≤ 8.9 IU/mL | 1.68 | 0.54-5.09 | 0.356 | |||
| CEA | ||||||
| ≥ 5.0 IU/mL / ≤ 4.9 IU/mL | 4.88 | 0.53-4.85 | 0.447 | |||
| CA19-9 | ||||||
| ≥ 37.0 IU/mL / ≤ 36.9 IU/mL | 2.65 | 0.89-8.16 | 0.077 | |||
| Preoperative chemotherapy | 0.89 | 0.44-2.99 | 0.821 | |||
| Primary cancer-related factors | ||||||
| Tumor location | ||||||
| EGJ, Upper/ Middle, Lower | 3.03 | 1.08-8.12 | 0.039 | |||
| Maximum primary tumor size | ||||||
| ≥ 5 cm / ≤ 4.9 cm | 1.93 | 0.71-4.86 | 0.183 | |||
| T classification | ||||||
| ≥ 3 / ≤ 2 | 1.28 | 0.51-3.45 | 0.596 | |||
| N classification | ||||||
| ≥ 1 / ≤ 0 | 1.03 | 0.41-2.62 | 0.934 | |||
| Histological type | ||||||
| Intestinal /Diffuse and other | 0.25 | 0.10-0.68 | 0.006 | 0.34 | 0.09-0.72 | 0.008 |
| Liver metastases-related factors | ||||||
| Timing of liver metastases | ||||||
| Metachronous / Synchronous | 0.88 | 0.24-2.25 | 0.786 | |||
| Tumor number | ||||||
| Multiple / Single | 1.61 | 0.59-4.02 | 0.329 | |||
| Tumor distribution | ||||||
| Bilobular / Unilobular | 1.24 | 0.43-3.16 | 0.661 | |||
| Maximum tumor size | ||||||
| ≥ 5 cm / ≤ 4.9 cm | 1.93 | 0.71-4.86 | 0.183 | |||
| Portal vein thrombosis | 2.18 | 0.60-6.30 | 0.210 | |||
| Operative procedures | ||||||
| Synchronous hepatectomy | 0.73 | 0.17-1.83 | 0.650 | |||
| Major hepatectomy | 3.16 | 0.77-9.98 | 0.093 | |||
| Operating time | ||||||
| ≥ 360 min / ≤ 359 min | 1.23 | 0.48-3.17 | 0.654 | |||
| Estimated blood loss | ||||||
| ≥ 1000 mL / ≤ 999 mL | 1.05 | 0.64-2.71 | 0.922 | |||
| Blood transfusion | 0.92 | 0.32-2.39 | 0.881 | |||
| Resection | ||||||
| ≥ R1 / R0 | 1.82 | 0.62-4.82 | 0.256 | |||
| Postoperative factors | ||||||
| Clavien-Dindo classification | ||||||
| ≥ I / ≤ 0 | 1.16 | 0.56-2.14 | 0.646 | |||
| Duration of hospital stay | ||||||
| ≥ 14 days / ≤ 13 days | 0.86 | 0.31-2.26 | 0.776 | |||
| Postoperative chemotherapy | 0.79 | 0.29-1.98 | 0.631 | |||
Abbreviations: BMI, body mass index; ASA, American society of anesthesiologists; AFP, α-fetoprotein; CEA, carcinoembryonic antigen; CA19- 9, carbohydrate antigen 19-9; EGJ, esophagogastric junction; Por, undifferentiated adenocarcinoma; HR, hazard ratios; CI, confidence intervals.
Factors predicting extra-hepatic recurrence
Intestinal-type adenocarcinoma and a maximum tumor size ≥ 5 cm were found to be significantly associated with extra-hepatic recurrence after liver resection for GELM (Table 5). Subsequent multiple logistic regression analysis revealed that intestinal-type adenocarcinoma (OR, 0.13; 95% CI, 0.02-0.66; p = 0.012) was associated with a lower incidence of extra-hepatic recurrence. Extra-hepatic recurrence rates after liver resection were significantly lower in patients with intestinal-type adenocarcinoma than in those with diffuse-type adenocarcinoma/other type (23.8% [5/21] vs. 60.0% [6/10]; p = 0.002).
Table 5. Univariate and multivariate analysis of extra-hepatic recurrence.
| Variables | Univariate Analysis |
Multivariate Analysis |
||||
|---|---|---|---|---|---|---|
| HR | 95% CI | p value | HR | 95% CI | p value | |
| Patient factors | ||||||
| Sex | ||||||
| Male / Female | 0.56 | 0.02-5.13 | 0.630 | |||
| Age | ||||||
| ≥ 70 years / ≤ 69 years | 0.57 | 0.11-2.52 | 0.463 | |||
| BMI | ||||||
| ≥ 25.0 kg/m2 / ≤ 24.9 kg/m2 | 0.82 | 0.16-3.26 | 0.363 | |||
| ASA | ||||||
| ≥ 2 / ≤ 1 | 1.03 | 0.52-2.82 | 0.623 | |||
| AFP | ||||||
| ≥ 9.0 IU/mL / ≤ 8.9 IU/mL | 3.00 | 0.46-26.7 | 0.256 | |||
| CEA | ||||||
| ≥ 5.0 IU/mL / ≤ 4.9 IU/mL | 0.53 | 0.09-2.64 | 0.449 | |||
| CA19-9 | ||||||
| ≥ 37.0 IU/mL / ≤ 36.9 IU/mL | 1.50 | 0.31-7.66 | 0.610 | |||
| Preoperative chemotherapy | 2.66 | 0.57-15.0 | 0.213 | |||
| Primary cancer-related factors | ||||||
| Tumor location | ||||||
| EGJ, Upper/ Middle, Lower | 0.30 | 0.05-1.48 | 0.140 | |||
| Maximum primary tumor size | ||||||
| ≥ 5 cm / ≤ 4.9 cm | 1.67 | 0.26-2.46 | 0.253 | |||
| T classification | ||||||
| ≥ 3 / ≤ 2 | 1.54 | 0.33-7.11 | 0.568 | |||
| N classification | ||||||
| ≥ 1 / ≤ 0 | 0.47 | 0.09-2.06 | 0.318 | |||
| Histological type | ||||||
| Intestinal /Diffuse and other | 0.14 | 0.02-0.67 | 0.006 | 0.13 | 0.02-0.66 | 0.012 |
| Liver metastases-related factors | ||||||
| Timing of liver metastases | ||||||
| Metachronous / Synchronous | 3.03 | 0.69-14.5 | 0.141 | |||
| Tumor number | ||||||
| Multiple / Single | 1.07 | 0.20-6.27 | 0.943 | |||
| Tumor distribution | ||||||
| Bilobular / Unilobular | 2.30 | 0.42-18.1 | 0.345 | |||
| Maximum tumor size | ||||||
| ≥ 5 cm / ≤ 4.9 cm | 1.08 | 1.02-5.06 | 0.049 | |||
| Operative procedures | ||||||
| Synchronous hepatectomy | ||||||
| Major hepatectomy | 1.77 | 0.27-11.4 | 0.531 | |||
| Operating time | ||||||
| ≥ 360 min / ≤ 359 min | 1.08 | 0.20-4.14 | 0.919 | |||
| Estimated blood loss | ||||||
| ≥ 1000 mL / ≤ 999 mL | 1.23 | 0.13-4.07 | 0.803 | |||
| Blood transfusion | 1.55 | 0.32-8.91 | 0.589 | |||
| Resection | ||||||
| ≥ R1 / R0 | 2.67 | 0.54-14.0 | 0.221 | |||
| Postoperative factors | ||||||
| Clavien-Dindo classification | ||||||
| ≥ I / ≤ 0 | 0.72 | 0.20-3.08 | 0.654 | |||
| Duration of hospital stay | ||||||
| ≥14 days / ≤13 days | 1.94 | 0.43-9.61 | 0.387 | |||
| Postoperative chemotherapy | 2.40 | 0.55-11.1 | 0.241 | |||
Abbreviations: BMI, body mass index; ASA, American society of anesthesiologists; AFP, α-fetoprotein; CEA, carcinoembryonic antigen; CA19- 9, carbohydrate antigen 19-9; EGJ, esophagogastric junction; Por, undifferentiated adenocarcinoma; OR, odds ratio; CI, confidence intervals.
OS and RFS in patients with intestinal-type adenocarcinoma
OS (p = 0.014) and RFS (p = 0.013) differed significantly between patients with intestinal-type adenocarcinoma and patients with diffuse-type adenocarcinoma/other type (Figure 2). The 1-, 3-, and 5-year OS rates in patients with intestinal-type adenocarcinoma were 100.0%, 72.9%, and 52.0%, respectively, whereas diffuse-type adenocarcinoma/ other type were 88.8%, 19.4%, and 19.4%, respectively. The 1-, 3-, and 5-year RFS rates in patients with intestinal-type adenocarcinoma were 74.5%, 48.0%, and 48.0%, respectively, while those in patients with diffuse-type adenocarcinoma/other type were 30.0%, 0.0%, and 0.0%, respectively. Recurrence after liver resection occurred in nine of 10 (90%) patients with diffuse-type adenocarcinoma/other type, including in the liver in three, in the lymph nodes in one, in the peritoneum in one, in the bone in one, and in multiple sites in three.
Figure 2.
(A) Overall survival in patients with intestinal-type adenocarcinoma and with diffuse-type adenocarcinoma/other type in GELM. (B) Recurrence-free survival in patients with intestinal-type adenocarcinoma and with diffuse-type adenocarcinoma/other type in GELM. GELM, gastric/esophagogastric-junction liver metastasis.
Summary of outcomes for GELM resection in previous study
Table 6 shows Indication, long-term outcomes, and prognostic factors of patients undergoing GELM resection in previous studies that included > 30 patients in recent years.
Table 6. Surgical indication, long-term outcomes, and prognostic factors reported in the previous and present studies.
| Author | Year | n | Surgical indication | Median OS (month) |
5-year RFS rate (%) |
5-year OS rate (%) |
Prognostic factors for poor survival |
|---|---|---|---|---|---|---|---|
| Shildberg et al. | 2012 | 31 | Without other distal metastasis | NS | NS | 13 | Multiple liver metastases R1/R2 resection Synchronous |
| Takemura et al. | 2012 | 64 | Three or fewer GELMs (More liver metastases at the surgeon's discretion) Only R0 resection |
34 | 27 | 37 | ≥ 5 cm in size pT4 of primary tumor |
| Wang et al. | 2012 | 30 | Without other distal metastasis preoperatively Synchronous GELMs Only R0 resection |
11 | NS | 16.7 | Peritoneal dissemination Multiple distal metastases |
| Kinoshita et al. | 2014 | 256† | Three or fewer GELMs (More liver metastases at the surgeon's discretion) Only R0 resection |
31 | 30.1 | 31.1 | pT4 of primary tumor ≥ 5 cm in size ≥ 3 GELMs |
| Tiberio et al. | 2014 | 53† | Without other distal metastasis | 34 | NS | 31.5 | ≥ 6 cm in size D2 dissection |
| Wang et al. | 2014 | 39 | Without other distal metastasis | 14 | 7.7 | 10.3 | Lymph node metastasis Multiple distal metastases |
| Guner et al. | 2015 | 68† | Not stated in detail (case by case) | 24 | 26.0 | 30.0 | ≥ 3 cm in size |
| Liu et al. | 2015 | 35 | Without other distal metastasis | 33 | NS | 14.3 | Lymphovascular invasion Multiple liver metastasis |
| Oki et al. | 2015 | 94† | Without other distal metastasis | 34 | 27.7 | 42.3 | ≥ 3 cm in size Multiple GELMs ≥ N2 of primary tumor |
| Present study | 2018 | 31 | Three or fewer GELMs Controllable after chemotherapy Only R0 resection |
38 | 31.3 | 42.2 | Diffuse/other type R1/R2 resection |
Abbreviations: OS, overall survivals; RFS, recurrence-free survivals; GELM, gastric/esophagogastric-junction liver metastasis; NS, not stated. †, Multicenter cohort study.
Discussion
Our study demonstrated that the 5-year OS and RFS rates in selected patients who underwent liver resection for GELM were 42.2% and 31.3%, respectively. Intestinal-type adenocarcinoma was associated with a lower risk for both OS and RFS, and with a lower incidence for extra-liver recurrence after liver resection for GELM.
The 5-year OS and RFS rates in our study are similar to those reported in previous studies. The OS and RFS rates reportedly ranged from 9.3% to 42.1% and from 8.6% to 27.7%, respectively (3-7,12-18). The median OS time was 38.0 months for selected patients in our study, which was also comparable to previous studies, where it ranged from 11 to 36 months (3,6,15,16,18,19). In contrast, according to a recent phase III clinical trial for GELM (20), the median OS was 9.5-14.1 months without liver resection. However, to address appropriate selection criteria for GELM resection, factors for poor prognosis concerning GELM resection and predictors for extra-hepatic recurrence after liver resection should be investigated; this is because gastric cancer develops peritoneal dissemination and lymph node metastases more frequently than colorectal cancer (21). In the present study, intestinal-type adenocarcinoma reduced a risk for OS and RFS. Additionally, intestinal-type adenocarcinoma was associated with a lower incidence of extra-hepatic recurrence. This finding is reasonable because diffuse-type histology is associated with infiltrative growth and peritoneal dissemination (22). Actually, the peritoneal dissemination rate is reported to be higher in patients with diffuse-type adenocarcinoma than those with intestinal-type adenocarcinoma (31% vs. 6%) (23). In our study, extra-hepatic recurrence rates after liver resection were significantly higher in patients with diffuse-type adenocarcinoma/other type than in those with intestinal-type adenocarcinoma (60% [6/10] vs. 23.8% [5/21]; p = 0.002). Accordingly, GELM resection is preferable for patients with intestinal-type adenocarcinoma. It would be reasonable to limit solitary GELM and/or to use a mandatory neoadjuvant chemotherapy strategy for non-intestinal-type histology, instead of upfront liver resection, although the effect of perioperative chemotherapy for GELM remains unclear. Additionally, the use of a strong adjuvant chemotherapy regimen can be recommended for GELM with diffuse-type adenocarcinoma/other type.
According to previous studies that included > 30 patients in the past 5 years, multiple liver metastases, R1/R2 resection, synchronous metastases, maximum size of liver metastases, pT4 of primary tumor, and other distant metastases were reported to be risk factors for OS (3,5,6,16,18,19). Unlike the previously reported covariates, the diffuse-type adenocarcinoma in our study is one of the prognostic factors for OS and RFS, and it is a predictor of extra-liver metastasis development. This is most likely because the previous series included advanced-stage patients with GELM and other organ metastases, and patients with four or more GELMs. These factors tempered the influence of the diffuse/ other-type adenocarcinoma of primary gastric cancer in the analysis. In contrast, our indication criteria are more restrictive than the previous series, namely three or fewer GELMs without any distant metastases; in addition, the diffuse-type carcinoma was found to be a factor for poor prognosis.
The present study had several limitations. Its retrospective nature and the small number of patients enrolled may weaken the reliability of the statistical analyses. Genomic expressions including α-fetoprotein and human epidermal growth factor receptor-related 2 (HER2) were not evaluated in the study. Further investigations with a large number of patients in a well-designed multicenter study are needed to evaluate appropriate patient selection criteria for GELM resection.
In conclusion, intestinal-type adenocarcinoma was associated with a lower risk for OS and RFS; it was also associated with a lower incidence of extra-hepatic recurrence, under the GELM resection criteria involving three or fewer tumors without distant metastases. Therefore, GELM resection is preferable for patients with intestinal-type histology. A strict indication such as solitary GELM and/or the use of mandatory neoadjuvant chemotherapy, and the use of a strong adjuvant chemotherapy regimen, can be recommended for GELM with diffuse-type adenocarcinoma/other type.
References
- 1. D'Angelica M, Gonen M, Brennan MF, Turnbull AD, Bains M, Karpeh MS. Patterns of initial recurrence in completely resected gastric adenocarcinoma. Ann Surg. 2004; 240:808-816. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Douglass HO, Jr, Hundahl SA, Macdonald JS, Khatri VP. Gastric cancer: D2 dissection or low Maruyama Index-based surgery - a debate. Surg Oncol Clin N Am. 2007; 16:133-155. [DOI] [PubMed] [Google Scholar]
- 3. Oki E, Tokunaga S, Emi Y, et al. Surgical treatment of liver metastasis of gastric cancer: a retrospective multicenter cohort study (KSCC1302). Gastric Cancer. 2016; 19:968-976. [DOI] [PubMed] [Google Scholar]
- 4. Oguro S, Imamura H, Yoshimoto J, Ishizaki Y, Kawasaki S. Liver metastases from gastric cancer represent systemic disease in comparison with those from colorectal cancer. J Hepatobiliary Pancreat Sci. 2016; 23:324-332. [DOI] [PubMed] [Google Scholar]
- 5. Liu Q, Bi J-J, Tian Y-T, Feng Q, Zheng Z-X, Wang Z. Outcome after simultaneous resection of gastric primary tumour and synchronous liver metastases: survival analysis of a single-center experience in China. Asian Pac J Cancer Prev. 2015; 16:1665-1669. [DOI] [PubMed] [Google Scholar]
- 6. Kinoshita T, Kinoshita T, Saiura A, Esaki M, Sakamoto H, Yamanaka T. Multicentre analysis of long-term outcome after surgical resection for gastric cancer liver metastases. Br J Surg. 2015; 102:102-107. [DOI] [PubMed] [Google Scholar]
- 7. Guner A, Son T, Cho I, Kwon IG, An JY, Kim HI, Cheong JH, Noh SH, Hyung WJ. Liver-directed treatments for liver metastasis from gastric adenocarcinoma: comparison between liver resection and radiofrequency ablation. Gastric Cancer. 2016; 19:951-960. [DOI] [PubMed] [Google Scholar]
- 8. Japanese Gastric Cancer Association. Japanese classification of gastric carcinoma: 3rd English edition. Gastric Cancer. 2011; 14:101-112. [DOI] [PubMed] [Google Scholar]
- 9. Makuuchi M, Kosuge T, Takayama T, Yamazaki S, Kakazu T, Miyagawa S, Kawasaki S. Surgery for small liver cancers. Semin Surg Oncol. 1993; 9:298-304. [DOI] [PubMed] [Google Scholar]
- 10. Kubota K, Makuuchi M, Kusaka K, Kobayashi T, Miki K, Hasegawa K, Harihara Y, Takayama T. Measurement of liver volume and hepatic functional reserve as a guide to decision-making in resectional surgery for hepatic tumors. Hepatology. 1997; 26:1176-1181. [DOI] [PubMed] [Google Scholar]
- 11. Mise Y, Satou S, Shindoh J, Conrad C, Aoki T, Hasegawa K, Sugawara Y, Kokudo N. Three-dimensional volumetry in 107 normal livers reveals clinically relevant inter-segment variation in size. HPB (Oxford). 2014; 16:439-447. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Dittmar Y, Altendorf-Hofmann A, Rauchfuss F, Gotz M, Scheuerlein H, Jandt K, Settmacher U. Resection of liver metastases is beneficial in patients with gastric cancer: report on 15 cases and review of literature. Gastric Cancer. 2012; 15:131-136. [DOI] [PubMed] [Google Scholar]
- 13. Garancini M, Uggeri F, Degrate L, Nespoli L, Gianotti L, Nespoli A, Uggeri F, Romano F. Surgical treatment of liver metastases of gastric cancer: is local treatment in a systemic disease worthwhile? HPB (Oxford). 2012; 14:209-215. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Schildberg CW, Croner R, Merkel S, Schellerer V, Muller V, Yedibela S, Hohenberger W, Peros G, Perrakis A. Outcome of operative therapy of hepatic metastatic stomach carcinoma: a retrospective analysis. World J Surg. 2012; 36:872-878. [DOI] [PubMed] [Google Scholar]
- 15. Takemura N, Saiura A, Koga R, Arita J, Yoshioka R, Ono Y, Hiki N, Sano T, Yamamoto J, Kokudo N, Yamaguchi T. Long-term outcomes after surgical resection for gastric cancer liver metastasis: an analysis of 64 macroscopically complete resections. Langenbecks Arch Surg. 2012; 397:951-957. [DOI] [PubMed] [Google Scholar]
- 16. Qiu JL, Deng MG, Li W, Zou RH, Li BK, Zheng Y, Lao XM, Zhou K, Yuan YF. Hepatic resection for synchronous hepatic metastasis from gastric cancer. Eur J Surg Oncol. 2013; 39:694-700. [DOI] [PubMed] [Google Scholar]
- 17. Komeda K, Hayashi M, Kubo S, et al. High survival in patients operated for small isolated liver metastases from gastric cancer: a multi-institutional study. World J Surg. 2014; 38:2692-2697. [DOI] [PubMed] [Google Scholar]
- 18. Wang W, Liang H, Zhang H, Wang X, Xue Q, Zhang R. Prognostic significance of radical surgical treatment for gastric cancer patients with synchronous liver metastases. Med Oncol. 2014; 31: [DOI] [PubMed] [Google Scholar]
- 19. Petrelli F, Coinu A, Cabiddu M, Ghilardi M, Borgonovo K, Lonati V, Barni S. Hepatic resection for gastric cancer liver metastases: A systematic review and meta-analysis. J Surg Oncol. 2015; 111:1021-1027. [DOI] [PubMed] [Google Scholar]
- 20. Bang YJ, Van Cutsem E, Feyereislova A, et al. Trastuzumab in combination with chemotherapy versus chemotherapy alone for treatment of HER2-positive advanced gastric or gastro-oesophageal junction cancer (ToGA): a phase 3, open-label, randomised controlled trial. Lancet. 2010; 376:687-697. [DOI] [PubMed] [Google Scholar]
- 21. Cheon SH, Rha SY, Jeung HC, Im CK, Kim SH, Kim HR, Ahn JB, Roh JK, Noh SH, Chung HC. Survival benefit of combined curative resection of the stomach (D2 resection) and liver in gastric cancer patients with liver metastases. Ann Oncol. 2008; 19:1146-1153. [DOI] [PubMed] [Google Scholar]
- 22. Kwon KJ, Shim KN, Song EM, Choi JY, Kim SE, Jung HK, Jung SA. Clinicopathological characteristics and prognosis of signet ring cell carcinoma of the stomach. Gastric Cancer. 2014; 17:43-53. [DOI] [PubMed] [Google Scholar]
- 23. Adachi Y, Yasuda K, Inomata M, Sato K, Shiraishi N, Kitano S. Pathology and prognosis of gastric carcinoma: well versus poorly differentiated type. Cancer. 2000; 89:1418-1424. [PubMed] [Google Scholar]