- Email: [email protected]
Incidence of gastric cancer in Sri Lanka: analysis of the cancer registry data and comparison with other South Asian populations
Poster Session, Sunday 29 January 2017 was required in only 1 case in D2 group. The postoperative mortality was 0% in both groups. Conclusion: There is no significant difference in the complications between D2+CME group and traditional D2 group. Therefore, we ensure that this trial is safe and thus ongoing. No conflict of interest. 544 ORAL Sarcopenia outperforms the Charlson Comorbidity Index in risk prediction in patients undergoing pancreatic resections D. Wagner1 , K. Marsoner1 , A. Tomberger2 , H. Cerwenka1 , G. Werkgartner1 , H.J. Mischinger1 , P. Kornprat1 . 1 Landeskrankenhaus Univ. Klinikum Graz, Department of Surgery- Division for General Surgery, Graz, Austria; 2 Medical University of Graz, Department of Surgery- Division for General Surgery, Graz, Austria Introduction: Sarcopenia is a known predictor in patients undergoing major pancreatic surgeries. We sought to combine sarcopenia with established risk predictors to improve their prognostic capacity for postoperative outcome and morbidity. Methods: As established parameters to predict preoperative mortality risk for patients, the ASA classification and the Charlson Comorbidity Index (CCI) were used. The Hounsfield Units Average Calculation (HUAC) was measured to define sarcopenia in 424 patients undergoing pancreatic resections for malignancies. Patients in the lowest sex-adjusted quartile for HUAC were defined as having sarcopenia (muscle wasting). Multivariable Cox regression analysis was utilized to identify preoperative risk factors associated with postoperative morbidity. Results: Median patient age was 63 years (19−87), 47.9% patients were male, and half the cohort had multiple comorbidities (Charlson Comorbidity Index [CCI] >6, 63.2%), 30-day mortality was 10.3% and 126 (29.7%). Median HUAC was 19.78 HU (IQR: 15.94–23.54) with 145 patients (34.2%) having sarcopenia. Preoperative frailty defined by sarcopenia was associated with an increased risk for postoperative complications (OR 1.55, 95% CI 0.98–2.45, p = 0.014), and a higher 30-day mortality (HR 5.17, 95% CI 1.57–16.69, p = 0.004). With an AUC of 0.85 HUAC showed the highest predictability for 30-day mortality (95% CI 0.78–0.91, p = 0.0001). Patients with CCI 6 and sarcopenia defined by the HUAC had a 9.78 higher risk of dying in the immediate postoperative phase (HR 9.78, 95% CI 2.98−32, p = 0.0001). Conclusion: Sarcopenia predicts postoperative mortality and complications best and it should be incorporated to conventional risk scores to identify high risk patients. No conflict of interest.
Poster Session (Sunday 29 January 2017) Gastrointestinal Malignancies − Upper GI 596 POSTER Adjvant HIPEC in gastric cancer patients with high risk of peritoneal carcinomatosis A. Aladashvili1 , R. Croner2 , I. Pantsulaia3 . 1 NCC, GI cancer, Tbilisi, Georgia; 2 FAU, General Surgery, Erlangen, Germany; 3 TSMU, Microbiology and Immunology, Tbilisi, Georgia Background: The peritoneum is one of the most common site of recurrence in gastric cancer. Median survival for PC is only about fore months, if untreated and the benefit of palliative systemic chemotherapy is limited. Only a quarter of patients are eligible for curative treatment, consisting of CRS/HIPEC. These clinical problems underline the need for effective adjuvant therapy in high-risk patients to minimize the risk of outgrowth of peritoneal micro metastases. Adjuvant HIPEC seems to be suitable for this purpose. Without the need for CRS, adjuvant HIPEC can be performed with a low complication rate and short hospital stay. The aim of this study is to determine the effectiveness of adjuvant HIPEC in preventing the development of PC in patients with gastric cancer at high risk of peritoneal recurrence. Patients and Methods: This study was performed in the Georgian NCC, starting in February 2014. Eligible for inclusion are patients who underwent curative resection for T4, cM0 stage gastric cancer. After resection of the primary tumour, 46 patients will be randomized to adjuvant HIPEC comparing with routine systemic chemotherapy only in the control arm. Different cytostatic agents were used for 60−90 min at 42−43ºC. Postoperatively in both arms of the study in patients without evidence of disease based on routine follow-up using CT, CEA and CA 19-9. Results: Morbidity and mortality were 32.81% and 3.12%, respectively, with three cases (4.68%) of peritoneal recurrence, from total number 64
Abstracts
S73
patients 62% were male and 48% female. Mean age was 57.6 years, range 31 to 73 years. In the beginning of treatment, the KPS was over 80% for all patients. Median follow-up was 23 months, ranging from 7 to 52 months. 2 patients were diagnosed with a pancreatic fistulae through the identification of an abnormal discharge in the closed suction drain placed during surgery, and confirmed by a fluid amylase examination, no additional treatment was necessary. 2 patients had an intraabdominal abscess that required re-laparotomy. Conclusions: Adjuvant HIPEC is assumed to reduce the expected 42% absolute risk of PC in patients with T4 GC to a risk of 14%. This reduction is likely to translate into a prolonged overall survival. In light of our experience and supported by literature data, we can affirm that HIPEC has a potential role in the prevention of gastric carcinomatosis. Certainly further studies are required on a larger scale to validate this new but promising approach. No conflict of interest. 597 POSTER Inhibitory effect of (−)-epigallocatechin-3-gallate and bleomycin on human pancreatic cancer MIA Paca-2 cell growth S. Bimonte1 , M. Leongito1 , A. Barbieri2 , F. Izzo1 . 1 National Cancer Institute Fondazione G. Pascale, Division of Abdominal Surgical Oncology- Hepatobiliary Unit, Naples, Italy; 2 National Cancer Institute Fondazione G. Pascale, SSD Animal Sperimentation, Naples, Italy Background: Human pancreatic cancer is currently one of the deadliest cancers with high mortality rate. It has been previously shown that (−)-epigallocatechin-3-gallate (EGCG), the most abundant catechin found in green tea, has shown suppressive effects on human pancreatic cancer cells. Bleomycin (BLM), an anti-cancer chemotherapeutic drug that induces DNA damage, has antitumor effects by induction of apoptosis in several cancer cell lines and also in pancreatic cancer cells. The present study investigated for the first time, the inhibitory effect of EGCG and BLM on pancreatic cancer cell growth. Methods: Using the pancreatic cancer cell lines MIA PaCa-2 cells the efficacy and synergism of EGCG and BLM were evaluated by in vitro tests. Inhibition of cell proliferation was determined by MTT assay. Mitochondrial depolarization was performed with JC-1 probe. Viability and apoptosis were determined by Flow Cytometry with annexin V, propidium iodide staining and DNA fragmentation assay. Results: Cell proliferation assay revealed significant additive inhibitory effects with combination of EGCG and BLM at 72 h in a dose dependent manner. The combination of EGCG and BLM induced cell cycle S-phase arrest and mitochondrial depolarization. Viability, apoptosis and DNA fragmentation assay indicated that the combination of EGCG and bleomycin potentiated apoptosis. Conclusions: Our results indicate that EGCG and BLM have additive antiproliferative effects in vitro by induction of apoptosis of MIA PaCa-2 cells. This combination could represent a new strategy with potential advantages for treatment of pancreatic cancer. To date, this is the first report published of the inhibitory effect of EGCG and BLM on human pancreatic cancer MIA Paca-2 cell growth. No conflict of interest. 598 POSTER Incidence of gastric cancer in Sri Lanka: analysis of the cancer registry data and comparison with other South Asian populations D. Wickramasinghe1 , N. Wickramasinghe2 , N. Samarasekera1 . 1 Faculty of Medicine- University of Colombo, Department of Surgery, Coombo 8, Sri Lanka; 2 Ministry of health, Ministry of Health, Colombo, Sri Lanka Background: This study aims to report the incidence of gastric carcinoma (GCa) in Sri Lanka (SL) and to compare these findings with other cancer registry data of the region and with migrant populations. Materials and Methods: We compared the data published by the National Cancer Control Program of Sri Lanka over the last 2 decades with data from the National Cancer Registry Programme of the Indian Council of Medical Research and Karachi cancer registry. SEERstat was used to analyse the Surveillance, Epidemiology, and End Results database to analyse data on Indian migrant population. Results: Gastric CA was the 10th most common cancer in males. The incidence of Gastric CA rises with age in both sexes, with a peak in the 70−74 year age group. There was a disproportionately higher number of GCa in the tamil population (Chi square test, p = 0.0022). The commonest type of Gastric CA in Sri Lanka was Adenomas and Adenocarcinomas, NOS (n = 175, 61.6%), followed by Cystic/mucinous/serous neoplasms second. (n = 83, 7.0%). India, Pakistan and Sri Lanka had comparable Age Adjusted Incidence (AAI) and age distribution of Gastric CA. All migrant populations had lower
S74
Abstracts
Poster Session, Sunday 29 January 2017
incidence of Gastric CA than original population or population in their present country. Cigarette smoking is more prevalent in Sri Lankan males than females. Conclusions: The incidence of Gastric CA and its distribution among age groups in Sri Lanka was comparable to other countries of the region. Persons of Tamil ethnicity have a higher risk of developing Gastric CA. Migrant populations had a lower incidence of Gastric CA than native populations. No conflict of interest. 599 POSTER Types and patterns of gastrointestinal (GI) polyps encountered in oesophago-gastro-duodenoscopy at a tertiary care center 1
1
2
1
D. Wickramasinghe , I. Almeida , C. De Silva , N. Samarasekera . 1 Faculty of Medicine- University of Colombo, Department of Surgery, Colombo 8, Sri Lanka; 2 Faculty of Medicine- University of Colombo, Department of Pathology, Colombo 8, Sri Lanka Background: We present the types and patterns of upper GI polyps from a single tertiary care center. Materials and Methods: Endoscopy and pathology reports of a single unit from 2006 to 2016 were analyzed retrospectively. Spearman’s correlation coefficient and Chi square were used to identify correlations. Results: There were a total of 74 patients (M:F 35:39) with polyps encountered in 2834 oesophago-gastro-duodenoscopies (OGD) (Incidence − 2.6%). There were single polyps in a majority (n = 64, 86.5%). A total of 79 polyps were seen and most were seen in the stomach (n = 42, 53.2%) followed by oesophagus (n = 21, 26.6%) and duodenum (n = 16, 20.3%). The histology in a majority was normal (n = 33, 41.8%) followed by fundic gland polyps (n = 13, 16.5%) and hyperplastic polyps (n = 10, 12.7%). The majority of these polyps were benign (n = 71, 89.9%). The mean ages of patients with benign and malignant polyps were 53.1 years (SD 15.3) and 57.0 years (SD 24.7), respectively. There was no statistically significant correlation with malignancy and age (Mann–Whitney test, P = 0.45) or between gender and malignancy (Fisher’s exact test, p = 1). Conclusions: A wide range of pathological types of upper GI polyps were encountered. The incidence of polyps in our sample is lower than the values reported in the USA. The incidence of multiple polyps was also lower (25% vs. 18%). However an equal gender distribution was observed similar to previous reports. There was no statistically significant association between age or gender and malignant change in the polyps. No conflict of interest. 600 POSTER Approaches for optimal choice in treatment of gastric cancer with liver metastases F. Djuraev1 , N. Atakhanova1 . 1 Tashkent Medical Academy, Department of Oncology, Tashkent, Uzbekistan Background: Gastric cancer (GC) is still one of the leading cause of cancer deaths with the incidence more than 700,000 registered new cases annually. Despite of the fact that incidence of the gastric cancer had slightly decreased for the past decade, most of the patient at the moment of initial diagnosing already have metastatic lesions, especially in developing countries. In this retrospective trial we tried to determine the optimal choice for treatment of GC with liver metastases. Materials and Methods: The trial include analysis of treatment of 74 patients with gastric cancer with liver metastases, who undergone treatment between 2004 and 2016. All patients had the same stage of disease, T4aN2M1, In addition metastatic lesion affected 1 or 2 segments of one liver lobe, number of metastatic nodes varied 1−5. The sizes of metastatic nodes: in 31 (41.9%) cases − 0.7−2.0 cm, and in 43 (58.1%) cases − 2.0−3.0 cm. According to the treatment option patients were divided into 3 groups: 1. Systemic chemotherapy − 29 (39.2%) patients 2. Hepatic arterial chemo infusion − 22 (29.7%) patients 3. Surgery + adjuvant chemo DCF − in 23 (31.1%) cases The schemes of chemotherapy regimen in 1st and 2nd group were DCF (Docetaxel + Cisplatin + Fluorouracil). However in 2nd group drugs were administered intra-hepatic artery, via transfemoral artery route. A catheter was inserted into the celiac trunk by interventional radiological techniques. In 3rd group all patients underwent surgery: D2 gastrectomy + resection of affected liver segments + adjuvant chemotherapy with DCF scheme. Results: Evaluation of the results and optimal choice for treatment was based on next rates: 1 year survival rate, overal survival rate. In 1st group of patients overall survival rate constituted 8.2+0.3 months, none of these
patients had 1 year survival. In 2nd group 1 year survival rate made up 4.5% that is only 1 patient, overal survival 11.3+0.4 months (P1−2 < 0.05). In 3rd group 1 year survival rate − 78.3%, and this is the only group where we observed 3 year survival rate in 17.8% of cases. Conclusion: Cases of resectable GC with liver metastases require combined treatment including D2 gastrectomy with liver resection + adjuvant chemotherapy, as only this treatment option improves 1 year survival rate up to 78.3% and 3 year survival to 17.8%. However in cases of unresectable GC we offer hepatic arterial chemo infusion as it authentically improves overall survival from 8.2+0.3 to 11.3+0.4 months. No conflict of interest. 601 POSTER Serum tumor markers carcinoma antigen (CA) 72−4 versus carcinoembryonic antigen (CEA) in patients with early stage (IA−IB) gastric cancer F. Lumachi1 , P. Ubiali2 , R. Tozzoli3 , A. Del Conte4 , S.M.M. Basso2 . 1 University of Padova, School of Medicine- Department of SurgeryOncology & Gastroenterology, Padova, Italy; 2 S. Maria degli Angeli Hospital, Department of Surgery- General Surgery, 33170 Pordenone, Italy; 3 S. Maria degli Angeli Hospital, Department of Laboratory MedicineClinical Pathology Laboratory, 33170 Pordenone, Italy; 4 S. Maria degli Angeli Hospital, Clinical Oncology, 33170 Pordenone, Italy Background: In the European Union, gastric cancer (GC) is the seventh most common cancer. Its prevalence is relatively low compared to the eastern Countries, accounting for approximately 14 cases per 100,000 per year. A number of serum tumor markers (TM) are commonly measured in patients with GC, both for diagnostic and prognostic purposes, including carcinoembryonic antigen (CEA), and carcinoma antigen (CA) 125, CA 19-9, CA-242 and CA 72-4. The aim of this study was to evaluate the usefulness of CA 72-4 versus CEA in patients with early stage GC. Material and Methods: We retrospectively reviewed the medical charts of 36 selected patients (median age 61 years, range 35−79 years) who underwent curative surgical resection for IA and IB AJCC/UICC stage GC. There were 26 (72.2%) men and 10 (27.8) women. Exclusion criteria were a history of other malignancy, preoperative neoadjuvant chemotherapy, alcohol abuse or smoking, or the use of drugs potentially interfering with the serum levels of TMs, as well as those with liver or kidney failure. Controls were 37 sex- and age matched patients in whom gastroscopy excluded the presence of GC or peptic ulcer. Written informed consent was obtained from all the participants. TMs assay was obtained with commercially available kits, according to manufacturer’s instructions. CEA was measured by automated homogeneous chemiluminescent (CLIA) immunoassay (luminescent oxygen channeling immunoassay-LOCI, that uses a dyoxetan derivative as luminescence substrate), whilst CA 72-4 was measured by a solid phase enzyme-linked immunosorbent assay (ELISA) based on the sandwich principle. The cut-off levels were 5 ng/ml and 4 U/mL for CEA and CA 72-4, respectively. Results: The results are reported in the Table. The sensitivity of CA 72-4 was significantly higher than that of CEA (30.6% vs. 11.1%, c2 = 4.31, p = 0.04) at the same good specificity (94.6%) and PPV (84.6% vs. 66.7%), but the accuracy did not differ (63.0% vs. 53.4%, c2 = 1.38, p = 0.24). Using the two TMs together, both the sensitivity (47.2%, c2 = 2.10, p = 0.14) and the accuracy (72.6%, c2 = 1.54, p = 0.21) did not improve significantly.
Sensitivity Specificity Positive predictive value (PPV) Negative predictive value (NPV) False positive rate (a) False negative rate (b) Likelihood ratio positive Likelihood ratio negative Odds ratio (OR)
CEA (95% CI)
CA 72-4 (95% CI)
CEA+CA72-4 together (95% CI)
11.1% (0.84–21.38) 94.6% (87.3–99.9) 66.7% (28.9–99.9) 52.2% (40.3–64.2) 5.41% 88.89% 2.06 0.94 21.2 (0.37–12.81)
30.6% (15.5–45.6) 94.6% (87.3–99.9) 84.6% (65.0–99.9) 58.3% (45.9–70.8) 5.41% 69.44% 5.65 0.73 7.7 (1.56–37.82)
47.2% (30−9–63.5) 97.3% (92.1–99.9) 94.4% (83.9–99.9) 65.4% (52.9–78.0) 2.70% 52.78% 17.47 0.54 32.21 (3.97–260.94)
Conclusions: CA 72−2 is more sensitive than CEA in detecting patients with GC, at the same specificity and similar PPV. However, simultaneous measurement of both TMs is not useful. No conflict of interest. 602 POSTER The epidemiological study of upper gastrointestinal cancer screening in rural areas in Sichuan, China Z. Song1 . 1 Sichuan Cancer Hospital, Urology, Chengdu, China Background: Upper gastrointestinal cancers are leading causes of cancer mortality in China. We conducted an upper gastrointestinal cancer
Poster Session (Sunday 29 January 2017) Gastrointestinal Malignancies − Upper GI 596 POSTER Adjvant HIPEC in gastric cancer patients with high risk of peritoneal carcinomatosis A. Aladashvili1 , R. Croner2 , I. Pantsulaia3 . 1 NCC, GI cancer, Tbilisi, Georgia; 2 FAU, General Surgery, Erlangen, Germany; 3 TSMU, Microbiology and Immunology, Tbilisi, Georgia Background: The peritoneum is one of the most common site of recurrence in gastric cancer. Median survival for PC is only about fore months, if untreated and the benefit of palliative systemic chemotherapy is limited. Only a quarter of patients are eligible for curative treatment, consisting of CRS/HIPEC. These clinical problems underline the need for effective adjuvant therapy in high-risk patients to minimize the risk of outgrowth of peritoneal micro metastases. Adjuvant HIPEC seems to be suitable for this purpose. Without the need for CRS, adjuvant HIPEC can be performed with a low complication rate and short hospital stay. The aim of this study is to determine the effectiveness of adjuvant HIPEC in preventing the development of PC in patients with gastric cancer at high risk of peritoneal recurrence. Patients and Methods: This study was performed in the Georgian NCC, starting in February 2014. Eligible for inclusion are patients who underwent curative resection for T4, cM0 stage gastric cancer. After resection of the primary tumour, 46 patients will be randomized to adjuvant HIPEC comparing with routine systemic chemotherapy only in the control arm. Different cytostatic agents were used for 60−90 min at 42−43ºC. Postoperatively in both arms of the study in patients without evidence of disease based on routine follow-up using CT, CEA and CA 19-9. Results: Morbidity and mortality were 32.81% and 3.12%, respectively, with three cases (4.68%) of peritoneal recurrence, from total number 64
Abstracts
S73
patients 62% were male and 48% female. Mean age was 57.6 years, range 31 to 73 years. In the beginning of treatment, the KPS was over 80% for all patients. Median follow-up was 23 months, ranging from 7 to 52 months. 2 patients were diagnosed with a pancreatic fistulae through the identification of an abnormal discharge in the closed suction drain placed during surgery, and confirmed by a fluid amylase examination, no additional treatment was necessary. 2 patients had an intraabdominal abscess that required re-laparotomy. Conclusions: Adjuvant HIPEC is assumed to reduce the expected 42% absolute risk of PC in patients with T4 GC to a risk of 14%. This reduction is likely to translate into a prolonged overall survival. In light of our experience and supported by literature data, we can affirm that HIPEC has a potential role in the prevention of gastric carcinomatosis. Certainly further studies are required on a larger scale to validate this new but promising approach. No conflict of interest. 597 POSTER Inhibitory effect of (−)-epigallocatechin-3-gallate and bleomycin on human pancreatic cancer MIA Paca-2 cell growth S. Bimonte1 , M. Leongito1 , A. Barbieri2 , F. Izzo1 . 1 National Cancer Institute Fondazione G. Pascale, Division of Abdominal Surgical Oncology- Hepatobiliary Unit, Naples, Italy; 2 National Cancer Institute Fondazione G. Pascale, SSD Animal Sperimentation, Naples, Italy Background: Human pancreatic cancer is currently one of the deadliest cancers with high mortality rate. It has been previously shown that (−)-epigallocatechin-3-gallate (EGCG), the most abundant catechin found in green tea, has shown suppressive effects on human pancreatic cancer cells. Bleomycin (BLM), an anti-cancer chemotherapeutic drug that induces DNA damage, has antitumor effects by induction of apoptosis in several cancer cell lines and also in pancreatic cancer cells. The present study investigated for the first time, the inhibitory effect of EGCG and BLM on pancreatic cancer cell growth. Methods: Using the pancreatic cancer cell lines MIA PaCa-2 cells the efficacy and synergism of EGCG and BLM were evaluated by in vitro tests. Inhibition of cell proliferation was determined by MTT assay. Mitochondrial depolarization was performed with JC-1 probe. Viability and apoptosis were determined by Flow Cytometry with annexin V, propidium iodide staining and DNA fragmentation assay. Results: Cell proliferation assay revealed significant additive inhibitory effects with combination of EGCG and BLM at 72 h in a dose dependent manner. The combination of EGCG and BLM induced cell cycle S-phase arrest and mitochondrial depolarization. Viability, apoptosis and DNA fragmentation assay indicated that the combination of EGCG and bleomycin potentiated apoptosis. Conclusions: Our results indicate that EGCG and BLM have additive antiproliferative effects in vitro by induction of apoptosis of MIA PaCa-2 cells. This combination could represent a new strategy with potential advantages for treatment of pancreatic cancer. To date, this is the first report published of the inhibitory effect of EGCG and BLM on human pancreatic cancer MIA Paca-2 cell growth. No conflict of interest. 598 POSTER Incidence of gastric cancer in Sri Lanka: analysis of the cancer registry data and comparison with other South Asian populations D. Wickramasinghe1 , N. Wickramasinghe2 , N. Samarasekera1 . 1 Faculty of Medicine- University of Colombo, Department of Surgery, Coombo 8, Sri Lanka; 2 Ministry of health, Ministry of Health, Colombo, Sri Lanka Background: This study aims to report the incidence of gastric carcinoma (GCa) in Sri Lanka (SL) and to compare these findings with other cancer registry data of the region and with migrant populations. Materials and Methods: We compared the data published by the National Cancer Control Program of Sri Lanka over the last 2 decades with data from the National Cancer Registry Programme of the Indian Council of Medical Research and Karachi cancer registry. SEERstat was used to analyse the Surveillance, Epidemiology, and End Results database to analyse data on Indian migrant population. Results: Gastric CA was the 10th most common cancer in males. The incidence of Gastric CA rises with age in both sexes, with a peak in the 70−74 year age group. There was a disproportionately higher number of GCa in the tamil population (Chi square test, p = 0.0022). The commonest type of Gastric CA in Sri Lanka was Adenomas and Adenocarcinomas, NOS (n = 175, 61.6%), followed by Cystic/mucinous/serous neoplasms second. (n = 83, 7.0%). India, Pakistan and Sri Lanka had comparable Age Adjusted Incidence (AAI) and age distribution of Gastric CA. All migrant populations had lower
S74
Abstracts
Poster Session, Sunday 29 January 2017
incidence of Gastric CA than original population or population in their present country. Cigarette smoking is more prevalent in Sri Lankan males than females. Conclusions: The incidence of Gastric CA and its distribution among age groups in Sri Lanka was comparable to other countries of the region. Persons of Tamil ethnicity have a higher risk of developing Gastric CA. Migrant populations had a lower incidence of Gastric CA than native populations. No conflict of interest. 599 POSTER Types and patterns of gastrointestinal (GI) polyps encountered in oesophago-gastro-duodenoscopy at a tertiary care center 1
1
2
1
D. Wickramasinghe , I. Almeida , C. De Silva , N. Samarasekera . 1 Faculty of Medicine- University of Colombo, Department of Surgery, Colombo 8, Sri Lanka; 2 Faculty of Medicine- University of Colombo, Department of Pathology, Colombo 8, Sri Lanka Background: We present the types and patterns of upper GI polyps from a single tertiary care center. Materials and Methods: Endoscopy and pathology reports of a single unit from 2006 to 2016 were analyzed retrospectively. Spearman’s correlation coefficient and Chi square were used to identify correlations. Results: There were a total of 74 patients (M:F 35:39) with polyps encountered in 2834 oesophago-gastro-duodenoscopies (OGD) (Incidence − 2.6%). There were single polyps in a majority (n = 64, 86.5%). A total of 79 polyps were seen and most were seen in the stomach (n = 42, 53.2%) followed by oesophagus (n = 21, 26.6%) and duodenum (n = 16, 20.3%). The histology in a majority was normal (n = 33, 41.8%) followed by fundic gland polyps (n = 13, 16.5%) and hyperplastic polyps (n = 10, 12.7%). The majority of these polyps were benign (n = 71, 89.9%). The mean ages of patients with benign and malignant polyps were 53.1 years (SD 15.3) and 57.0 years (SD 24.7), respectively. There was no statistically significant correlation with malignancy and age (Mann–Whitney test, P = 0.45) or between gender and malignancy (Fisher’s exact test, p = 1). Conclusions: A wide range of pathological types of upper GI polyps were encountered. The incidence of polyps in our sample is lower than the values reported in the USA. The incidence of multiple polyps was also lower (25% vs. 18%). However an equal gender distribution was observed similar to previous reports. There was no statistically significant association between age or gender and malignant change in the polyps. No conflict of interest. 600 POSTER Approaches for optimal choice in treatment of gastric cancer with liver metastases F. Djuraev1 , N. Atakhanova1 . 1 Tashkent Medical Academy, Department of Oncology, Tashkent, Uzbekistan Background: Gastric cancer (GC) is still one of the leading cause of cancer deaths with the incidence more than 700,000 registered new cases annually. Despite of the fact that incidence of the gastric cancer had slightly decreased for the past decade, most of the patient at the moment of initial diagnosing already have metastatic lesions, especially in developing countries. In this retrospective trial we tried to determine the optimal choice for treatment of GC with liver metastases. Materials and Methods: The trial include analysis of treatment of 74 patients with gastric cancer with liver metastases, who undergone treatment between 2004 and 2016. All patients had the same stage of disease, T4aN2M1, In addition metastatic lesion affected 1 or 2 segments of one liver lobe, number of metastatic nodes varied 1−5. The sizes of metastatic nodes: in 31 (41.9%) cases − 0.7−2.0 cm, and in 43 (58.1%) cases − 2.0−3.0 cm. According to the treatment option patients were divided into 3 groups: 1. Systemic chemotherapy − 29 (39.2%) patients 2. Hepatic arterial chemo infusion − 22 (29.7%) patients 3. Surgery + adjuvant chemo DCF − in 23 (31.1%) cases The schemes of chemotherapy regimen in 1st and 2nd group were DCF (Docetaxel + Cisplatin + Fluorouracil). However in 2nd group drugs were administered intra-hepatic artery, via transfemoral artery route. A catheter was inserted into the celiac trunk by interventional radiological techniques. In 3rd group all patients underwent surgery: D2 gastrectomy + resection of affected liver segments + adjuvant chemotherapy with DCF scheme. Results: Evaluation of the results and optimal choice for treatment was based on next rates: 1 year survival rate, overal survival rate. In 1st group of patients overall survival rate constituted 8.2+0.3 months, none of these
patients had 1 year survival. In 2nd group 1 year survival rate made up 4.5% that is only 1 patient, overal survival 11.3+0.4 months (P1−2 < 0.05). In 3rd group 1 year survival rate − 78.3%, and this is the only group where we observed 3 year survival rate in 17.8% of cases. Conclusion: Cases of resectable GC with liver metastases require combined treatment including D2 gastrectomy with liver resection + adjuvant chemotherapy, as only this treatment option improves 1 year survival rate up to 78.3% and 3 year survival to 17.8%. However in cases of unresectable GC we offer hepatic arterial chemo infusion as it authentically improves overall survival from 8.2+0.3 to 11.3+0.4 months. No conflict of interest. 601 POSTER Serum tumor markers carcinoma antigen (CA) 72−4 versus carcinoembryonic antigen (CEA) in patients with early stage (IA−IB) gastric cancer F. Lumachi1 , P. Ubiali2 , R. Tozzoli3 , A. Del Conte4 , S.M.M. Basso2 . 1 University of Padova, School of Medicine- Department of SurgeryOncology & Gastroenterology, Padova, Italy; 2 S. Maria degli Angeli Hospital, Department of Surgery- General Surgery, 33170 Pordenone, Italy; 3 S. Maria degli Angeli Hospital, Department of Laboratory MedicineClinical Pathology Laboratory, 33170 Pordenone, Italy; 4 S. Maria degli Angeli Hospital, Clinical Oncology, 33170 Pordenone, Italy Background: In the European Union, gastric cancer (GC) is the seventh most common cancer. Its prevalence is relatively low compared to the eastern Countries, accounting for approximately 14 cases per 100,000 per year. A number of serum tumor markers (TM) are commonly measured in patients with GC, both for diagnostic and prognostic purposes, including carcinoembryonic antigen (CEA), and carcinoma antigen (CA) 125, CA 19-9, CA-242 and CA 72-4. The aim of this study was to evaluate the usefulness of CA 72-4 versus CEA in patients with early stage GC. Material and Methods: We retrospectively reviewed the medical charts of 36 selected patients (median age 61 years, range 35−79 years) who underwent curative surgical resection for IA and IB AJCC/UICC stage GC. There were 26 (72.2%) men and 10 (27.8) women. Exclusion criteria were a history of other malignancy, preoperative neoadjuvant chemotherapy, alcohol abuse or smoking, or the use of drugs potentially interfering with the serum levels of TMs, as well as those with liver or kidney failure. Controls were 37 sex- and age matched patients in whom gastroscopy excluded the presence of GC or peptic ulcer. Written informed consent was obtained from all the participants. TMs assay was obtained with commercially available kits, according to manufacturer’s instructions. CEA was measured by automated homogeneous chemiluminescent (CLIA) immunoassay (luminescent oxygen channeling immunoassay-LOCI, that uses a dyoxetan derivative as luminescence substrate), whilst CA 72-4 was measured by a solid phase enzyme-linked immunosorbent assay (ELISA) based on the sandwich principle. The cut-off levels were 5 ng/ml and 4 U/mL for CEA and CA 72-4, respectively. Results: The results are reported in the Table. The sensitivity of CA 72-4 was significantly higher than that of CEA (30.6% vs. 11.1%, c2 = 4.31, p = 0.04) at the same good specificity (94.6%) and PPV (84.6% vs. 66.7%), but the accuracy did not differ (63.0% vs. 53.4%, c2 = 1.38, p = 0.24). Using the two TMs together, both the sensitivity (47.2%, c2 = 2.10, p = 0.14) and the accuracy (72.6%, c2 = 1.54, p = 0.21) did not improve significantly.
Sensitivity Specificity Positive predictive value (PPV) Negative predictive value (NPV) False positive rate (a) False negative rate (b) Likelihood ratio positive Likelihood ratio negative Odds ratio (OR)
CEA (95% CI)
CA 72-4 (95% CI)
CEA+CA72-4 together (95% CI)
11.1% (0.84–21.38) 94.6% (87.3–99.9) 66.7% (28.9–99.9) 52.2% (40.3–64.2) 5.41% 88.89% 2.06 0.94 21.2 (0.37–12.81)
30.6% (15.5–45.6) 94.6% (87.3–99.9) 84.6% (65.0–99.9) 58.3% (45.9–70.8) 5.41% 69.44% 5.65 0.73 7.7 (1.56–37.82)
47.2% (30−9–63.5) 97.3% (92.1–99.9) 94.4% (83.9–99.9) 65.4% (52.9–78.0) 2.70% 52.78% 17.47 0.54 32.21 (3.97–260.94)
Conclusions: CA 72−2 is more sensitive than CEA in detecting patients with GC, at the same specificity and similar PPV. However, simultaneous measurement of both TMs is not useful. No conflict of interest. 602 POSTER The epidemiological study of upper gastrointestinal cancer screening in rural areas in Sichuan, China Z. Song1 . 1 Sichuan Cancer Hospital, Urology, Chengdu, China Background: Upper gastrointestinal cancers are leading causes of cancer mortality in China. We conducted an upper gastrointestinal cancer