Search

362 Background: The increased efficacy of multimodality treatment allows for the consideration of conversion therapy with curative intent for initially unresectable esophageal squamous cell carcinoma (ESCC) after initial treatment. Conversion therapy has been commonly defined as treatment aiming at a curable intention after tumors initially deemed technically or oncologically unresectable or only marginally resectable respond to chemotherapy. In our institute, conversion therapy is considered for patients with para-aortic lymph node-positive advanced ESCC who have no other distant metastasis, if the para-aortic lymph node metastases have disappeared after systemic chemotherapy. Methods: There were 33 patients of unresectable esophageal cancer with positive para-aortic lymph node metastases from 2011 to 2021. Nine of these patients showed disappearance of para-aortic lymph node metastases and underwent conversion therapy, seven patients underwent esophagectomy and two patients underwent chemoradiation therapy (CRT). Characteristics, short-term and long-term results of these nine cases are studied. Results: All of the nine patients who had para-aortic lymph node metastases were treated with induction DCF (Docetaxel + Cisplatin + 5-FU) therapy as initial treatment. After disappearance of para-aortic lymph node metastasis, minimally invasive or open esophagectomy was performed in seven patients and CRT was performed in two patients. In operation group, four patients (57%) had Clavien-Dindo grade 2 or 3 complications, but no perioperative death was observed. In CRT group, there was no toxicity over CTCAE grade 3. Four of nine patients (44%) had recurrence, within mediastinum lymph node, abdominal lymph node or adrenal gland. 3-year DFS was 56% and 3-year OS was 67%. Conclusions: Conversion therapy is safe and feasible in cases that responded to systemic chemotherapy. Some cases showed good prognosis. Long-term survival is expected with conversion therapy. In our study, there was few cases that could led to conversion therapy. It is expected that more aggressive initial chemotherapy will be used in the future for patients with positive para-aortic lymph node metastases with a view to conversion therapy.

TPS474 Background: Supraclavicular lymph nodes (SCLN) have been regarded as regional nodes of thoracic esophageal cancer in Japan and resected along with cervical paraesophageal lymph nodes as a cervical part of the standard three-field dissection, while the involvement of SCLN is regarded as distant metastasis in Western countries. However, there is no clear evidence that supports or rejects prophylactic SCLN dissection. As postoperative SCLN metastasis can often be managed by either surgery or chemoradiotherapy at the time of relapse, routine prophylactic SCLN dissection may not be essential. Here we conduct the clinical trial to confirm the non-inferiority of esophagectomy with D2 lymphadenectomy except for SCLN to standard D2 lymphadenectomy in terms of overall survival (OS) for patients with resectable upper or middle thoracic esophageal cancer. Methods: Eligibility criteria include the followings; upper or middle thoracic esophageal squamous cell carcinoma, adenosquamous carcinoma, or basaloid cell carcinoma with clinical stage I, II, III, or IVA (except for T4) based on the 8th UICC-TNM, age 18 to 80, ECOG performance status 0 or 1, and adequate organ function. Neoadjuvant chemotherapy (cisplatin plus 5- fluorouracil or docetaxel plus cisplatin plus 5-fluorouracil) is allowed, but preoperative chemoradiotherapy is not allowed. Patients are randomly assigned to the following two arms using the minimization method with a random component to balance institution, clinical stage (Stage I (T1N0M0) versus Stage I (T1N1M0)/II/III/IVA), age (76 or older versus 75 or younger), and tumor location (upper thoracic versus middle thoracic). Patients undergo either esophagectomy with standard D2 dissection (Arm A) or esophagectomy with D2 lymphadenectomy except for SCLN (Arm B). The primary endpoint is OS in all randomized patients. The secondary endpoints are relapse-free survival, the incidence of perioperative and delayed complications, the incidence of SCLN recurrence, the incidence of salvage cervical treatment, the proportion of synchronous cervical and abdominal procedure, and operation time and the number of operating surgeons. A sample size of 456 patients would be required to observe 147 events to demonstrate the non-inferiority of Arm B in terms of OS, with a one-sided alpha of 5%, power of 70%, an accrual period of 5 years, a follow-up period of 3 years, and a non-inferiority margin of 6% at 3-year OS, assuming the expected 3-year OS of Arm A and Arm B are 80% and 81%, respectively. The planned sample size was set at 480 patients (240 patients per arm). This trial was registered at the Japan Registry of Clinical Trials (jRCT) as study number jRCT1030220248 and started in August 2022. Clinical trial information: jRCT1030220248 .

290 Background: NIVO + chemo and NIVO + IPI demonstrated superior overall survival (OS) vs chemo in CheckMate 648 (NCT03143153), leading to approvals in the US, EU, Japan, and other countries. We report longer follow-up results. Methods: Adults with previously untreated, unresectable advanced, recurrent, or metastatic ESCC were randomized to NIVO (240 mg Q2W) + chemo (fluorouracil + cisplatin Q4W), NIVO (3 mg/kg Q2W) + IPI (1 mg/kg Q6W) or chemo. Primary endpoints were OS and progression-free survival (PFS) per blinded independent central review (BICR). Hierarchical testing was done first in patients (pts) with tumor cell programmed death ligand 1 (PD-L1) ≥ 1%, then in all randomized pts. Results: 970 pts were randomized to NIVO + chemo, NIVO + IPI, or chemo. With 29-mo minimum follow-up, NIVO + chemo and NIVO + IPI continued to show improvement in OS vs chemo, including higher 24-mo OS rates, in pts with tumor cell PD-L1 ≥ 1% and all randomized pts. Responses were more durable and a larger proportion of responders had a duration of response (DOR) ≥ 24 mo with NIVO + chemo and NIVO + IPI vs chemo in pts with tumor cell PD-L1 ≥ 1% (22%, 36%, 13%, respectively) and all randomized pts (21%, 29%, 13%). Additional efficacy data by PD-L1 status will be presented. Any-grade treatment-related adverse events (TRAEs) occurred in 96% (grade 3/4, 49%) of pts with NIVO + chemo, 80% (33%) with NIVO + IPI, and 90% (36%) with chemo. Any-grade TRAEs leading to discontinuation occurred in 35% of pts with NIVO + chemo, 19% with NIVO + IPI, and 21% with chemo. Treatment-related deaths occurred in 2% of pts in each arm. Conclusions: NIVO + chemo and NIVO + IPI continued to demonstrate clinically meaningful survival benefit vs chemo, durable objective responses, and acceptable safety profiles with longer follow-up. This further supports each regimen as a new 1L treatment option for advanced ESCC. Clinical trial information: NCT03143153 . [Table: see text]

4035 Background: NIVO + chemo and NIVO + IPI demonstrated significant overall survival (OS) benefit vs chemo in previously untreated patients (pts) with advanced ESCC in the phase 3 CheckMate 648 study. We report expanded results from the primary analysis with 13-month (mo) minimum follow-up. Methods: Pts with previously untreated, unresectable advanced, recurrent, or metastatic ESCC were randomized to NIVO (240 mg Q2W) + chemo (fluorouracil + cisplatin Q4W), NIVO (3 mg/kg Q2W) + IPI (1 mg/kg Q6W), or chemo. Primary endpoints were OS and progression-free survival (PFS) per blinded independent central review (BICR) in pts with tumor cell programmed death ligand 1 (PD-L1) ≥ 1%. Secondary endpoints planned for hierarchical testing included OS, PFS, and objective response rate (ORR) per BICR in all randomized pts. Duration of response (DOR) per BICR and PFS2 per investigator (time from randomization to progression after subsequent systemic therapy, initiation of second subsequent systemic therapy, or death) were exploratory endpoints. Results: Among all pts randomized to NIVO + chemo (n = 321), NIVO + IPI (n = 325), or chemo (n = 324), PFS2 favored NIVO + chemo (HR 0.64, 95% CI 0.54–0.77) and NIVO + IPI (HR 0.74, 95% CI 0.62–0.88) vs chemo. ORR (95% CI) was 47% (42–53), 28% (23–33), and 27% (22–32), respectively. More responders with NIVO + chemo or NIVO + IPI vs chemo had prolonged DOR (≥12 mo; 39%, 48%, and 23%, respectively). Efficacy data by tumor cell PD-L1 and PD-L1 combined positive score will be presented. Grade 3/4 treatment-related adverse events with potential immunologic etiology (select TRAEs) occurred in ≤ 6% of pts with NIVO + chemo and NIVO + IPI, and non-endocrine select TRAEs resolved in 57%–95% of pts across organ categories (Table). Conclusions: NIVO + chemo and NIVO + IPI demonstrated favorable PFS2 and a higher proportion of pts with prolonged DOR vs chemo, as well as acceptable safety profiles. These results provide further support for each regimen as a new potential 1L standard of care in advanced ESCC. Clinical trial information: NCT03143153. [Table: see text]

4038 Background: Neoadjuvant chemotherapy followed by surgery (NAC-S) is the standard therapy for locally advanced esophageal squamous cell cancer (ESCC) in Japan. The aim of this phase II trial was to assess the efficacy and safety of the addition of adjuvant S-1 after R0 resection in patients treated with NAC-S. Methods: Key eligibility criteria were as follows: ESCC of clinical stage IB-III (without T4 disease); aged 20 to 75 years; ECOG performance status 0 or 1; and performed neoadjuvant chemotherapy (5-FU + cisplatin). All patients registered before surgery. Patients received adjuvant therapy with 4 cycles of S-1 (80 mg/m2/day) that is administered orally for 4 weeks of a 6-weeks cycle. The primary endpoint was three-year relapse free survival (RFS). Results: A total of 52 patients were enrolled between January 2016 and January 2019. Two patients (one with small cell carcinoma and the other with synchronous malignancy) were excluded from analysis. Five patients were diagnosed as R1 or R2. Seven patients did not receive adjuvant S-1 due to adverse events of surgery in 5 patients, refusal to adjuvant S-1 in a patient, and forgot to start in a patient. Thirty-eight patients received adjuvant S-1, with 32 patients completing 4 cycles. The median relative dose intensity of adjuvant S-1 was 85.8%. Median follow-up time among survivors after surgery was 4.5 years (range 0.2-5.6). Three-year RFS in intention to treat population was 72.3% (90% confidence interval [CI] 59.9-81.5), suggesting that the primary endpoint was met, and 3-year overall survival was 85.0% (90% CI 73.9-91.6). Grade 3 or higher adverse event with an incidence 10% of greater were neutropenia (13.2%), anorexia (13.2%), and diarrhea (10.5%). There was no treatment-related death. Conclusions: Adjuvant S-1 showed promising efficacy with manageable safety profile for patients with resectable ESCC after NAC-S, and warrants further evaluation in larger studies. Clinical trial information: UMIN000020204.

286 Background: The standard neoadjuvant chemotherapy (NAC) in Japan for locally advanced esophageal squamous cell carcinoma (ESCC) consists of CDDP + 5-FU (CF). Recently, neoadjuvant DTX plus CF (DCF) therapy has shown promising efficacy for locally advanced ESCC. In addition, immune checkpoint inhibitors (ICIs) have demonstrated a survival benefit in ESCC and potential as an NAC in several cancers, including lung, breast, skin, and bladder. We previously reported that neoadjuvant CF plus nivolumab (Nivo) therapy was tolerable and showed promising efficacy for ESCC; however, the efficacy of ICI with DCF as an NAC and the safety of subsequent surgery for ESCC patients remain unclear. Methods: FRONTiER is a multi-cohort phase I study designed to evaluate the safety and efficacy of ICI combined with NAC in ESCC. The eligibility criteria were histologically proven ESCC with cT1N1-3M0 or cT2-3N0-3M0 (8th-UICC TNM classification), patient age of 20-75 years, PS ≤1, and no prior cancer therapies. The primary endpoint was the incidence of dose-limiting toxicities (DLT) from the initial dose to the 30th postoperative day. This study contained 5 experimental cohorts: cohort C received 3 courses of DTX (70 mg/m2), CDDP (70 mg/m2), and Nivo (360 mg/body) on day 1 and 5-FU (750 mg/m2) on days 1–5 every three weeks. Cohort D received one administration of Nivo (240 mg/body) 2 weeks before chemotherapy followed by the same regimen as cohort C. Results: Twelve patients were enrolled in cohort C (n = 6) and D (n = 6) (median age [range]: 60 [31-74] years, PS 0/1: 11/1, clinical stage I/II/III/IVA: 2/2/7/1). Only one patient in cohort D developed DLT (grade 3 dyspnea and rash); no other DLT were seen. Grade ≥3 adverse events were neutropenia (n = 1), leukopenia (n = 7), anorexia (n = 3), hyponatremia (n = 2), febrile neutropenia (n = 1), nausea (n = 1), and lymphopenia (n = 1) during NAC, and lung infection (n = 2), peritoneal infection (n = 1), anemia (n = 1), lymph leakage (n = 1), vagal reflex (n = 1), dyspnea (n = 1), rash (n = 1), septic shock (n = 1), pleural effusion (n = 1), and pneumatosis intestinalis (n = 1) during the postoperative period. No treatment-related deaths were observed. All patients received 3 courses of NAC, and 11 patients (91.7%) achieved an R0 resection without treatment interruption. One patient (16.7%) in cohort C and 3 patients (50.0%) in cohort D achieved pathological complete responses. Conclusions: Neoadjuvant DCF + Nivo with/without prior Nivo administration followed by surgery for locally advanced ESCC was well tolerated and showed an extremely promising efficacy. Clinical trial information: NCT03914443.

238 Background: Neoadjuvant treatment is the standard care for locally advanced esophageal squamous cell cancer (ESCC). JCOG1109 (UMIN000009482) compared the doublet and triplet of chemotherapy and chemoradiotherapy as neoadjuvant treatment. Methods: Eligible patients (pts) with ESCC of clinical stage IB, II, III (excluding T4) (UICC 7th) from 44 institutions were randomized 1:1:1 to neoadjuvant CF (cisplatin 80 mg/m2 on day1 plus 5-FU 800 mg/m2 on days 1-5 Q3W/2course), DCF (docetaxel 70 mg/m2 on day 1, cisplatin 70 mg/m2 on day1, plus 5-FU 750 mg/m2 on days 1-5 Q3W/3 course), or CF-RT (cisplatin 75 mg/m2 on day 1 plus 5-FU 1000 mg/m2 on days 1-4 Q4W/2course, radiation 41.4 Gy/23 fr). Primary endpoint was overall survival (OS), and secondary endpoints included progression-free survival (PFS), %R0 resection, %objective response by neoadjuvant therapy, %pathological complete response (pCR) and safety. Differences in OS was assessed in the ITT using the stratified log-rank test. The data cutoff date for the analysis was July 20, 2021. Results: Of 601 pts 199 CF, 202 DCF, and 200 CF-RT were enrolled from December 5, 2012 to July 20, 2018, respectively. Among 601 pts, 88.2% were male, median (range) age was 65 (30-75), clinical stage III (nonT4) pts were 62.6%. Median follow-up time (range) was 4.2 years (y) (0-8.5). Median OS in CF, DCF, and CF-RT arm were 4.6 y, not reached (NR), and 6.0y, and 3-year OS was 62.6%, 72.1%, and 68.3%, respectively (stratified log-rank test: p = 0.006 for CF vs. DCF and p = 0.12 for CF vs. CF-RT). By stratified Cox regression analysis for OS, hazard ratios (HR) [95% CI] was 0.68 [0.50–0.92] for CF vs. DCF and 0.84 [0.63–1.12] for CF vs. CF-RT. Median PFS in CF, DCF, and CF-RT arm were 2.7 y, NR, and 5.3 y, and 3-year PFS was 47.7%, 61.8%, and 58.5%, respectively. R0 resection was achieved in 168 (84.4%), 173 (85.6%), and 175 (87.5%), and pCR was 4 (2.1%), 40 (19.8%), and 77 (38.5%), respectively. During neoadjuvant therapy, febrile neutropenia in CF, DCF, and CF-RT arm were 1.0%, 16.3% and 4.7%, and esophagitis (grade>3) were 1.0%, 1.0% and 8.9%, respectively. The treatment-related death was seen in 3 (1.5%), 4 (2.0%), and 2 (1.0%), in CF, DCF, and CF-RT arm, respectively. Conclusions: DCF significantly improved OS over CF as neoadjuvant therapy for locally advanced ESCC, with a manageable toxicity profile. DCF represents a new standard neoadjuvant treatment for ESCC. Clinical trial information: UMIN000009482.

Gastric cancer (GC) poses a burden to patients across the globe as the third leading cause of cancer deaths worldwide. Incidence of GC is particularly high in Asian countries, which is attributed to the prevalence of Helicobacter pylori (H. pylori) infection and has prompted the establishment of unique treatment strategies. D2 gastrectomy, which was established in the 1950s in Japan, has served as a gold standard for locally advanced GC for over half a century. Since the beginning of the 21st century, endoscopic resection (ER) techniques and minimally invasive laparoscopic surgery have greatly changed the treatment of patients with early GC. S-1, which showed a striking survival benefit in a large randomized trial in Japan, has been used as adjuvant therapy for the last decade. Likewise, S-1–based chemotherapy regimens are currently the standard of care for the treatment of unresectable/metastatic GC in Asia. Along with the development of standardized therapy, novel techniques and new drugs have been rapidly brought into clinical practice. State-of-the-art sentinel node (SN) navigation surgery enables clinicians to perform truly minimally invasive surgery for early GC, and appropriate chemotherapy regimens are now determined by a tumor’s molecular expression. New classifications based on gene signatures are proposed and may replace conventional clinical classifications. Such highly individualized treatment has the potential to alter our clinical practice in GC in the near future. The best practice in each geographic region should be shared and integrated, resulting in the best practice without borders.

LBA4001 Background: NIVO demonstrated superior overall survival (OS) vs chemo in previously treated patients (pts) with ESCC (ATTRACTION-3). We report OS and progression-free survival (PFS) from CheckMate 648, the first global phase III study to evaluate both an immuno-oncology (I-O)/chemo combination and an I-O/I-O combination in advanced ESCC. Methods: Adults with previously untreated, unresectable advanced, recurrent or metastatic ESCC were enrolled regardless of tumor cell PD-L1 expression. Pts were randomized to NIVO (240 mg Q2W) + chemo (fluorouracil + cisplatin Q4W), NIVO (3 mg/kg Q2W) + IPI (1 mg/kg Q6W), or chemo alone. Primary endpoints for both comparisons were OS and PFS per blinded independent central review (BICR) in pts with tumor cell PD-L1 ≥ 1%. Hierarchically tested secondary endpoints included OS and PFS in all randomized pts. Results: 970 pts were randomized to NIVO + chemo, NIVO + IPI, and chemo arms (49% with tumor cell PD-L1 ≥ 1%). With 13 months (mo) minimum follow-up, NIVO + chemo and NIVO + IPI led to statistically significant improvement in OS vs chemo in pts with tumor cell PD-L1 ≥ 1% and all randomized pts (Table). Statistically significant PFS benefit was also observed for NIVO + chemo vs chemo (HR 0.65 [98.5% CI 0.46–0.92]; P = 0.0023) in pts with tumor cell PD-L1 ≥ 1%. PFS in NIVO + IPI vs chemo in pts with tumor cell PD-L1 ≥ 1% did not meet the prespecified boundary for significance. The objective response rate (per BICR) was 53% (NIVO + chemo), 35% (NIVO + IPI), and 20% (chemo) in pts with tumor cell PD-L1 ≥ 1% and in all randomized pts was 47%, 28%, and 27%, respectively; longer median (95% CI) duration of response was observed vs chemo for pts with tumor cell PD-L1 ≥ 1%: 8.4 (6.9–12.4), 11.8 (7.1–27.4), and 5.7 (4.4–8.7) mo and for all randomized pts: 8.2 (6.9–9.7), 11.1 (8.3–14.0), and 7.1 (5.7–8.2) mo, respectively. No new safety signals were identified (Table). Conclusions: NIVO plus chemo and NIVO plus IPI both demonstrated superior OS vs chemo, along with durable objective responses and acceptable safety, in pts with advanced ESCC, and each represents a potential new 1L treatment option. Clinical trial information: NCT03143153. [Table: see text]

1026 Background: The Federation of Asian Clinical Oncology (FACO) conducted an international retrospective cohort study of oligometastatic breast cancer (BC) (OLIGO-BC1) (UMIN No.000030047). At ASCO2020, we demonstrated that locoregional and systemic therapy prolonged overall survival (OS) for patients with oligometastatic BC, especially for cases with some type of systemic therapy, younger age, ECOG performance status 0, stage I BC, non-triple negative subtype, fewer metastatic sites, local recurrence and longer disease-free interval from a multivariate analysis (#1025). Although BC is heterogeneous and a retrospective dataset has many kinds of bias, we attempted a subset analysis based on intrinsic subtype and several prognostic factors. Methods: Oligometastatic BC patients diagnosed from 2007 to 2012 were registered from CSCO, KSMO and JSCO. OS period was measured from the diagnosis of oligometastases to the latest follow-up. ER, PgR and HER2 status were determined by immunohistochemistry and/or in situ hybridization. A hazard ratio (HR) of OS was calculated by using a univariate analysis. Results: In 1200 eligible cases, one oligometastatic site was found in 578 cases, two in 289, three in 154, four in 102 and five in 77. Bone metastases were recorded in 301 cases, visceral metastases in 387, locoregional recurrence in 25, local recurrence in 83 and multiple metastatic sites in 404. Luminal subtype was recorded in 526 cases (44%), luminal-HER2 in 189 (16%), HER2 in 154 (13%), triple-negative in 166 (14%) and others in 165 (13%). In any subtype, locoregional and systemic therapy and ECOG performance status 0 were beneficial for OS. Stage I BC, one oligometastatic site and longer disease-free interval were also related to favorable prognosis in luminal and HER2 subtype. However, triple-negative subtype had no survival advantage with these 3 factors. On the other hand, pathological negative or micrometastatic lymph nodes at primary BC and one oligometastatic site of lymph node, lung, liver and bone were favorable prognostic factors. In addition, cases treated locally with surgical resection and conventional radiation therapy were expected to prolong OS. Discussions: Locoregional therapy for oligometastatic BC may be considered in luminal and HER2 subtype with some conditions. As reported in ASCO2020, triple-negative BC should be managed with systemic therapy. Conclusions: Oligometastatic BC is diagnosed by chance, but some cases seem to survive with multidisciplinary treatment. It is worthwhile to consider locoregional therapy in oligometastatic BC after evaluating favorable prognostic factors.

TPS257 Background:Parallel-group controlled trial of esophagectomy versus chemoradiotherapy (JCOG0502) showed chemoradiotherapy (CRT) is considered as a standard treatment option for clinical-T1bN0M0 esophageal squamous cell carcinoma (ESCC) with organ preservation. However, locoregional recurrence after CRT occurs in 20% of patients and requires salvage treatment including salvage surgery, chemo(radio)therapy, and endoscopic resection. Salvage treatments can cause complications and treatment-related death. CRT with elective nodal irradiation (ENI) have been reported to reduce the locoregional recurrence of esophageal cancer. We conduct the clinical trial to evaluate the efficacy of modified CRT with ENI compared with CRT without ENI for the patients with cTbN0M0 ESCC. The primary purpose of this study is to investigate whether modified CRT with ENI reduces the locoregional recurrence that cannot be completely resected by salvage endoscopic resection and preserve esophagus without compromising overall survival. Methods:Eligibility criteria include the followings; thoracic ESCC, adenosquamous, or basaloid cell carcinoma with clinical-T1bN0M0, age 20 or older, performance status 0 or 1, adequate organ function, and patient who do not have a preference to receive a surgical resection as an initial therapy. Patients are randomly assigned to following two arms using the minimization method with a random component to balance institution, tumor length (≤4cm versus > 4cm), and age (≤65 years old versus > 65 years old). Patients in arm A receive CRT consisted of CDDP (70 mg/m2/day, days 1 and 29) and 5-FU (700 mg/m2/day, days 1-4 and 29-32, civ) with concurrent radiotherapy (60 Gy/30 fr without ENI). Patients in arm B receive CRT consisted of CDDP (75 mg/m2/day, days 1 and 29) and 5-FU (1000 mg/m2/day, days 1-4 and 29-32, civ) with concurrent radiotherapy (50.4 Gy/ 28 fr with ENI of 41.4 Gy/23 fr). Primary endpoint is major progression-free survival (MPFS), defined as the time from randomization to the date of death or disease progression except achieving curative resection by salvage endoscopic resection. We assumed 5-year MPFS with arm A to be 70% and expected a 10% improvement for arm B. The planned sample size was calculated as a total of 280 patients (140 patients per arm) with a one-sided alpha of 5%, power of 70%, an accrual period of 4 years, and a follow-up period of 5 years. This trial was activated in July 2020. The study was also registered at the Japan Registry of Clinical Trials (jRCT) as study number jRCTs031200067. Clinical trial information: jRCTs031200067.

207 Background:ATTRACTION-1/ONO-4538-07 (AT-1), an open-label, single-arm, multicenter phase 2 clinical trial conducted in Japan, evaluated the clinical activity and safety of nivolumab in patients with advanced esophageal squamous cell carcinoma (ESCC) refractory/intolerant to fluoropyrimidine-, platinum-, and taxane-based chemotherapy. We previously reported the 2-year follow-up findings of AT-1, in which nivolumab demonstrated antitumor activity with a manageable safety profile for these patients. Here we report the final findings from AT-1 at a minimum follow-up of 5 years. Methods:Patients aged ≥20 years with an Eastern Cooperative Oncology Group Performance Status (ECOG PS) of 0-1 received 3 mg/kg nivolumab intravenously every 2 weeks in 6-week cycles until disease progression or unacceptable toxicity. The primary endpoint was centrally-assessed objective response rate (ORR), defined as the proportion of patients whose best overall response was either a complete or partial response. Secondary endpoints included overall survival (OS), investigator-assessed ORR, progression-free survival (PFS), change in tumor burden, time to response, time to disease progression, and duration of response. Results:Between February 25 and November 14, 2014, a total of 65 patients were enrolled. Sixty-four patients were evaluated for the efficacy, and all patients were evaluated for the safety. At the final database lock on August 6, 2020, 11 (17.2%, 95% confidence interval [CI] 9.9-28.2) of 64 patients had an objective response by central assessment. The median OS was 10.8 months (95% CI, 7.4-13.9), and the estimated 5-year OS rate was 6.3% (95% CI, 2.0-14.0). The median PFS was 1.5 months (95% CI, 1.4-2.8), and the estimated 5-year PFS rate was 6.8% (95% CI, 2.2-15.1). Treatment-related adverse events that occurred with a frequency of > 10% were diarrhea and rash. The presentation will include characteristics of long-term survivors as well as detailed efficacy and safety data of nivolumab. Conclusions:This final assessment represents the longest follow-up of patients with advanced ESCC treated with nivolumab. Nivolumab demonstrated continued long-term efficacy in these patients based on a minimum of 5-year long-term survival update of AT-1. Furthermore, no new safety signals with nivolumab were identified during long-term follow-up. These findings are consistent with those of nivolumab monotherapy for various types of cancer. Clinical trial information: No.142422.

162 Background: We have conducted randomized three-arm phase III trial comparing cisplatin plus 5-FU (CF) versus docetaxel plus CF (DCF) versus radiation with CF (CF-RT) as preoperative therapy for locally advanced esophageal cancer, which is on-follow-up for primary analysis planned in 2023 (JCOG 1109). This study aimed to evaluate the influence of preoperative therapies on perioperative complications and risk factors for perioperative complications after three-arm preoperative therapies. Methods: Patients with potentially resectable advanced thoracic esophageal cancer were randomly assigned to three preoperative therapies and followed by open or thoracoscopic esophagectomy with regional lymphadenectomy. Clinical data, surgical results, and perioperative complications in the patients received DCF and CF-RT were compared with those in the patients received CF. Univariate and multivariate analyses were performed to explore the risk factors of perioperative complications. Results: Between December 2012 and July 2018, 601 patients were randomized (CF/DCF/CF-RT; 199/202/200). Of 589 eligible patients, 546 patients underwent surgery (185/183/178). Patients` characteristics were not different between arms. Median number of harvested lymph node in patients received CF-RT was significantly lower than that in patients received CF (49 vs. 58; P < 0.0001). Incidence of ≥ Grade 2 perioperative complications in patients received DCF was lower than that in patients received CF (44.8% vs. 56.2%; P = 0.036). Incidence of ≥ Grade 2 chylothorax in patients received CF-RT was higher than that in patients received CF (5.1% vs. 1.1%; P = 0.032). Incidence of reoperation and intra-hospital death in patients received DCF and CF-RT did not differ from that in patients received CF. Multivariate analysis showed that operation time (≥ median) and open esophagectomy were independently associated with an increase in ≥ Grade 2 perioperative complications. CF-RT was associated with an increase in occurrence of ≥ Grade 2 chylothorax (Relative Risk 4.84; P = 0.043). Conclusions: Preoperative DCF and CF-RT does not increase the risk of perioperative complications and mortality when compared with standard preoperative CF therapy, but CF-RT increases the risk of chylothorax after esophagectomy for advanced thoracic esophageal cancer.

1025 Background: Systemic therapy is the standard care in metastatic breast cancer (BC). However, retrospective studies demonstrated survival benefits of locoregional and systemic (combination) therapy in oligometastatic BC. To clarify it, the Federation of Asian Clinical Oncology conducted an international retrospective cohort study (OLIGO-BC1) (UMIN No.000030047). Methods: Oligometastatic BC patients diagnosed from 2007 to 2012 were registered. “Oligometastases” is defined as low volume metastatic disease with limited number and size of metastatic lesions up to five and not necessarily in the same organ by the ABC guidelines. Overall survival (OS) from the diagnosis of oligometastases was the primary endpoint and compared between combination and systemic therapy using a log-rank test. Assuming the 5-year OS of 50% and 40%, respectively, 698 patients were required to achieve 80% power to detect the superiority of combination therapy, at a two-sided significance level. A multivariable Cox regression model with stratification by country was performed to estimate hazard ratio (HR) for therapy and other risk factors. Results: While 1,262 cases had been registered from February 2018 to May 2019, 1,200 remained for analysis after exclusion of unavailable cases. Among them, 573, 529 and 98 cases were registered from China, Japan and Korea, respectively. Luminal BC was recorded in 526 cases (44%), luminal-HER2 BC in 189 (16%), HER2 BC in 154 (13%), triple-negative BC in 166 (14%) and others in 165 (13%). One oligometastatic BC was found in 578 cases, 2 in 289, 3 in 154, 4 in 102 and 5 in 77. Bone metastases were recorded in 301 cases, visceral metastases in 387, locoregional recurrence in 25, local recurrence in 83 and multiple metastatic sites in 404. Combination therapy was performed in 595 cases and systemic therapy in 404. At median follow-up of 4.9 years, 5-year OS was 59.6% and 41.9%, respectively (p < 0.01). An adjusted HR was 0.61 (95% CI: 0.51, 0.74). Type of systemic therapy, younger age, ECOG performance status 0, stage I BC, non-triple negative subtype, fewer metastatic sites, local recurrence and longer disease-free interval were significantly favorable prognostic factors. Discussions: Oligometastatic BC under some conditions seems to be curable. Taken together with recent molecular targeted therapy, locoregional therapy will be advantageous to conquer it. Conclusions: Combination therapy is a promising strategy for patients with oligometastatic BC.

4545 Background: Definitive chemoradiotherapy (dCRT) consisting of 5-fluorouracil (5-FU) and cisplatin (CDDP) with 60 Gy radiotherapy (RT) for clinical (c) Stage II/III esophageal carcinoma (EC) is a standard treatment for patients (pts) refusing surgery (S) in Japan based on the previous trial (JCOG9906). However, poor survival, high incidence of late toxicities, and severe complications of salvage S are problems. We conducted a single-arm confirmatory study of CRT modifications including salvage treatment (ST) to reduce CRT toxicities and facilitate ST to improve survival. We reported the 3-year survival at 2018 ASCO Annual Meeting. We report the final data after 5-year follow-up. Methods: EC pts with cStage II/III (UICC 6th, non-T4), PS 0-1, and age 20-75 years were eligible. Chemotherapy (CT) was CDDP (75 mg/m2 on days 1, 29) and 5-FU (1000 mg/m2/d on days 1-4, 29-32). RT was administered to a total dose of 50.4 Gy with elective nodal irradiation of 41.4 Gy. Good responders after dCRT received additional 1-2 cycles of CT. For residual or recurrent disease, salvage endoscopic resection (ER) or S was performed based on the prespecified criteria. Planned sample size was 95, with one-sided alpha of 5% and power of 80%, expected and threshold 3-year overall survival (OS) as 55% and 42%. Key secondary endpoint was ST related toxicity. Final analysis was planned after 5-year follow-up for all pts. Results: From 4/2010 to 8/2014, 96 pts were enrolled, two were ineligible and 94 were included in efficacy analysis (cStage IIA/IIB/III, 22/38/34). Complete response was achieved in 55 pts (59%). Salvage ER and S were performed in 5 (5%) and 27 pts (29%). R0 resection of salvage S was achieved in 23 (85%). With a median follow-up of 5.95 years, 3- and 5-year OS was 74.2% (90% CI 65.9-80.8%) and 64.5% (95% CI 53.9-73.3%). 5-year progression-free survival and esophagectomy-free survival were 48.3% (95% CI 37.9-58.0%) and 54.9% (95% CI 44.3-64.4%). 5-year OS after salvage S was 31.0% and hazard ratio of R1-2 to R0 was 5.635 (95% CI: 1.818-17.467). No complications occurred after salvage ER. Five pts (19%) showed ≥ grade 3 operative complications and 1 treatment related death due to bronchus-pulmonary artery fistula occurred after salvage S. Only 9 pts (9.6%) showed grade 3 late toxicities. And no late operative complications more than grade 3 were observed. Conclusions: This combined modality treatment of dCRT with ST showed acceptable toxicities, favorable 5-year survival, and promising esophageal preservation. Clinical trial information: jRCTs031180110 .

4565 Background: Esophageal squamous cell carcinoma (ESCC) is associated with synchronous or metachronous cancer at other primary sites. Previous studies have suggested that patients (pts) with ESCC are still at a high risk of second primary malignancies after definitive therapies. In particular, early-stage ESCCs result in good prognosis, which is associated with a higher risk of second primary malignancies. Methods: The JCOG0502 was a phase III trial, which included a randomized and a non-randomized part and compared esophagectomy with definitive chemoradiotherapy in clinical T1bN0 ESCC. We additionally reviewed data of all pts enrolled in the JCOG0502 for second primary malignancies. Lugol-voiding lesions (LVLs) were assessed in the noncancerous esophageal mucosa before the treatments. Results: Among 379 enrolled pts, 213 pts received esophagectomy and the remaining received chemoradiotherapy. Patient characteristics of overall cohort were as follows: male, 85%; median age, 63 (range, 41–75) years; upper- /middle- /lower thoracic esophagus, 11/63/27%; alcohol consumption history, 79%; smoking history, 66%; prevalence of no LVLs/several LVLs/many LVLs/unknown, 45/36/8/11%. With a median follow-up of 7.1 years, a total of 118 second malignancies were observed in 99 pts (26%). Cumulative incidences of second malignancies after 3, 5, 10 years were 9, 15, 36%, respectively. Most common primary tumor sites were head and neck (35%), followed by stomach (20%) and lung (14%). In multivariable analyses, several LVLs [hazard ratio (HR): 2.24, 95% confidential interval (CI): 1.32–3.81, vs. no LVLs] and many LVLs (HR: 2.88, 95% CI: 1.27–6.52, vs. no LVLs) were significantly associated with the development of second malignancies. Regarding the three most common types of cancers, 62 out of the 77 cancers (81%) were diagnosed in clinical stage 0–I. Seventeen pts died due to second primary malignancies. There were 4 and 3 deaths from head and neck and lung cancer, respectively. Whereas, mortality caused by stomach cancer was not observed. Conclusions: In the JCOG0502, the incidence of second malignancies was high, indicating that careful follow-up is required for ESCC pts even after treatment completion. The presence of LVLs in esophagus was identified as an independent predictive factor for second primary malignancies, which may be useful for surveillance strategies. Clinical trial information: UMIN000000551 .

314 Background: As surgery for esophageal carcinoma in the elderly people has been increasing, sarcopenia is a severe problem not only in complications, but also in long-term prognosis. However, the relationship between perioperative skeletal muscle loss especially in the early postoperative period and long-term prognosis has not been clarified. Methods: This study retrospectively analyzed 152 patients with thoracic esophageal carcinoma who had underwent radical esophagectomy in our institution from April 2008 to March 2015 (Patients with postoperative hospital stay longer than 6 weeks were excluded). As an index of perioperative sarcopenia, total psoas muscle area (TPA) was measured before surgery (as baseline), at postoperative day (POD) 7 and postoperative month (POM) 6 from CT images. We investigated the correlation between the change of TPA and the postoperative survival. Results: Of 152 patients, 52 (34.2%) showed a TPA decrease from baseline to POD 7, and 98 (64.5%) showed a TPA decrease from baseline to POM 6. At the time of POD 7, overall survival (OS) decreased significantly in a TPA decrease group (P = 0.008, 5-year survival rate: non-decrease group 82.3% / decrease group 56.8%). Recurrence free survival (RFS) was also significantly decreased in a TPA decrease group (P < 0.001, 5-year recurrence free survival rate: non-decrease group 73.7% / decrease group 44.9%). On the other hand, at the time of POM 6, OS and also RFS had no significant difference between decrease and non-decrease groups. In univariate analysis for OS, pStage ≥3 and TPA decrease at POD 7 had poor prognosis. In multivariate analysis for OS, pStage ≥3 (HR:5.516, P < 0.001, 95%CI:2.634-11.551) and TPA decrease at POD 7 (HR:2.036, P = 0.047, 95%CI:1.010-4.103) were also independent poor prognostic factors. In the univariate analysis for RFS, pStage ≥3, TPA decrease at POD 7 and age ≥60 years had poor prognosis. In multivariate analysis, pStaeg ≥3 (HR:3.831, P < 0.001, 95%CI:2.182-6.728) and TPA decrease at POD 7 (HR:1.942, P = 0.021, 95%CI:1.104-3.416) were independent poor prognostic factors. Conclusions: Our findings suggest that the TPA decrease early in a postoperative period has poor prognosis on OS and also RFS.

577 Background: Biliary tract carcinoma (BTC) is a highly malignant tumor, but the detailed pathological mechanism of its development and progression remain ill-defined. Gamma-synuclein (SNCG) promotes invasive behavior in pancreatic cancer and a range of other cancers, and SNCG expression is reportedly a prognostic factor. However, the role of SNCG in BTC remains unknown. Consequently, we investigated the clinicopathological significance and function of SNCG in BTC. Methods: Using surgically resected BTC specimens from 149 patients with adenocarcinoma [extrahepatic cholangiocarcinoma (ECC) (n =98); intrahepatic cholangiocarcinoma (ICC) (n =51)], we immunohistochemically evaluated SNCG positivity and checked for correlations with clinicopathological factors and outcomes. Furthermore, cell lines with high expressions of SNCG were selected from 17 BTC cell lines by using immunohistochemistry and qPCR. Cell proliferation and migration assays were then performed in the presence and absence of SNCG (silenced using). Results: SNCG expression was found in 32 of 149 (21.4%) BTC patients. SNCG expression was significantly correlated with poorly differentiated tumor ( P = 0.001) and lymph node metastases ( P = 0.001). SNCG positivity was also correlated with poorly differentiated tumor in both ECC and ICC ( P = 0.008 and P = 0.03, respectively), but was correlated with lymph node metastases only in ECC ( P = 0.003). Multivariate analyses identified SNCG expression as an independent prognostic factor of poor overall survival ( P = 0.008) and poor recurrence-free survival (RFS) ( P = 0.006). SNCG expression in both ECC and ICC was significantly associated with poor prognosis in univariate analysis, and multivariate analysis identified SNCG as an independent prognostic factor of poor RFS for ECC (P = 0.03). High SNCG expression was found in 3 BTC cell lines (NCC-BD1, NCC-BD3, and NCC-CC6-1). Functional analysis revealed that SNCG silencing by siRNA suppressed cell migration significantly in NCC-BD1 and NCC-CC6-1 and downregulated cell proliferation in NCC-CC6-1. Conclusions: Our results suggest that SNCG may promote tumor cell activity and is potentially a novel prognostic marker in BTC.

402 Background: We first performed thoracoscopic esophagectomy (TE) as a minimally invasive procedure with the left decubitus position in 1996. In 2009 we developed a hybrid of the prone and left lateral decubitus positions for TE with extended LN dissection (Extensive-TE). The patient is fixed with the semi-prone position and we can easily change patient positions from the left lateral decubitus position to the prone position using rotation system of the operation table. The upper mediastinal procedure including lymphadenectomy along the right and left recurrent laryngeal nerve (RLN) is performed with the patient in the left lateral decubitus position, while the middle and lower mediastinal procedures are performed with the patient in the prone position with artificial pneumothorax. Methods: ESCC patients who underwent Extensive-TE between January 2009 and December 2016, were retrospectively reviewed. The patients’ background, surgical outcomes, postoperative complications and recurrence-free survival (RFS) were studied. Results: Primary tumor was located in Cervical esophagus for 2 (1%), the upper-thoracic esophagus for 28 (15%), the mid-thoracic esophagus for 104 (54%) and the lower-thoracic esophagus for 57 (30%). Thenumber of patients classified with pre-treatment clinical stage of 1/2/3/4 was 94(49%)/42(22%)/46(24%)/9(5%), respectively. Eight patients were evaluated as having cM1 disease due to supraclavicular LN metastasis. The number of patients classified with postoperative pathological stage of 0/1/2/3/4 was 5(3%)/70(37%)/48(26%)/49(27%)/19(7%), respectively. The average total operation time was 542.1 and blood loss was 274.2. The incidence of postoperative pneumonia, anastomotic leakage, chylothorax, and recurrent nerve palsy was 17%, 14%, 2%, and 7% respectively. One patient died postoperatively within 90 days after surgery. Three years RFSwith clinical stage of 1/2/3+4 was 91.5%/54.8%/51.9%, respectively. Conclusions: The magnifying effect of thoracoscopy enables us to perform more precise surgery and preserve nerve and vessels. Extensive-TEwith a hybrid position is thought to be feasible and effective methods.

289 Background: Early gastric cancer shows lymph node involvement in about 10-15% of patients. Despite the fact, we perform radical lymphadenectomy for all patients because predicting lymph node metastasis has yet to be successful. In this study, we hypothesize that image analysis using artificial intelligence (AI) technology may help solve the problem. Methods: We retrospectively collected 82 patients with clinical T1N0 and pathological node negative and 82 patients with clinical T1N0 and pathological node positives and then divided the 164 patients into a training:validation set in ratio of 9:1. Endoscopic images of the early tumors were analyzed by transfer learning using AlexNet, a deep neural network containing 5 convolutional layers and 3 fully-connected layers. The model was validated with newly-collected 40 images from 20 clinical T1N0 and pathological node negative and 20 patients with clinical T1N0 and pathological node positives as a test set. For comparison, three methods of prediction were implemented: prediction at random, by logistic regression, and by skilled endoscopists. Results: The AI predicted LNM with accuracy of 80.9% in the validation set and 66.9% in the test set. (48.3% for node negative cancers and 85.4% for node positive cancers) On the other hand, prediction at random, by logistic regression, and by 2 endoscopists resulted in 50.3%, 50.0%, and 47.5%, respectively. Conclusions: Although the accuracy still needs to be improved, image analysis using the AI technology resulted in the best prediction of lymph node metastasis, indicating that AI is a promising technology for the diagnosis of lymph node metastasis in early gastric cancer.

120 Background: Postoperative complications have a negative impact on overall survival after esophagectomy because systemic inflammation may induce residual cancer cell growth. A solution that could suppress micrometastasis is neoadjuvant therapy; however, to date, no study has shown that neoadjuvant therapy suppresses the proliferation of cancer cells due to postoperative complications after esophagectomy. To investigate the preventive effect of neoadjuvant therapy on poor long-term outcomes of postoperative complications in patients with esophageal squamous cell carcinoma. Methods: In total, 509 patients who underwent esophagectomy for primary esophageal squamous cell carcinoma were included in this retrospective cohort study. We investigated the relationship between complications, such as pneumonia, and long-term oncologic outcomes with and without neoadjuvant therapy. Results: Among all the patients, the 3-year overall survival (OS) rate was 68.4% and the disease-free survival (DFS) rate was 58.1%. The patients were categorized into two groups: the neoadjuvant therapy (+) group (n = 227) and the neoadjuvant therapy (−) group (n = 282). Among patients not undergoing neoadjuvant therapy, the patients with pneumonia, atrial fibrillation, pyothorax, or chylothorax had significantly poorer OS and DFS than patients without these complications. However, among patients undergoing neoadjuvant therapy, there were no significant differences in long-term outcomes, regardless of the presence of complications. In multivariate analyses, pneumonia (p = 0.014) and chylothorax (p = 0.006) were identified as predictors of death in the neoadjuvant therapy (−) group. Conclusions: In patients with esophageal squamous cell carcinoma, the negative impact of postoperative complications can be reduced by performing neoadjuvant therapy.

86 Background: S-1 is an oral anticancer drug, containing tegafur (a prodrug of 5-FU), 5-chloro-2,4-dihydroxypyridine (CDHP, inhibitor of dihydoropyrimidine dehydrogenase) and potassium oxonate. Because CDHP is excreted in urine, renal dysfunction increases incidence of severe adverse drug reactions due to higher exposure of 5-FU. As approved dose of S-1 is determined by body surface area (BSA) for patients with normal renal function, dose of S-1 is practically reduced according to renal function of creatinine clearance (CLcr) estimated by the Cockcroft-Gault equation. In a previous pharmacokinetic study (n = 16), we had developed an S-1 dosage formula based on renal function achieving the target area under the concentration-time curve (AUC) of 5-FU: Dose = target AUC x (21.9 + 0.375 x CLcr) x BSA. We conducted a prospective study to evaluate and refine this formula if necessary. Methods: Thirty patients with various renal function received S-1 at dose determined by our developed formula. A series of blood samples were obtained at predefined times after the first dose to calculate the AUC of 5-FU. Predictability of the formula was evaluated by comparison between the observed and the target AUCs. Results: The observed daily AUC was ranged from 712.6 to 2868.7 ng‧h/mL in 30 patients with BSA in the range of 1.14-1.84 m2 and CLcr in the range of 23.8-96.4 mL/min. Eighteen patients of them achieved the target AUC (1447.8 ± 545.4 ng‧h/mL). Since population pharmacokinetic analysis using combined pharmacokinetic data of 30 patients in this study and 16 patients in the previous study demonstrated that clearance of 5-FU is significantly lower in female than in male, the S-1 dosage formula was refined including gender as an additional factor: Dose = target AUC × (14.5 + 8.23 x GENDER [0 for female and 1 for male] + 0.301 × CLcr) × BSA. Revised nomograms showing recommended daily dose of S-1 were proposed for males and females taking into account tablet strengths. Conclusions: The refined formula for determining S-1 dosage on the basis of renal function, BSA and gender can be applied to clinical practice to ensure efficacy and safety for cancer patients treated with S-1. Clinical trial information: UMIN 000023880.

7 Background: Esophagectomy (E) is the standard of care for stage I esophageal squamous cell carcinoma (ESCC), while chemoradiotherapy (CRT) is a treatment option. A parallel-group controlled trial including randomized arms to confirm the non-inferiority of CRT to E for stage IA ESCC was conducted. Methods: Patients (pts) with thoracic ESCC, adenosquamous cell, or basaloid cell carcinoma with stage IA (T1bN0M0), age 20 to 75, performance status 0 to 1, and adequate organ function were eligible. If pts accepted randomization, they were randomly allocated to E with 2-3 field lymph node dissection (arm A) or CRT (arm B). However, if pts had a preference and refused randomization, they were allocated to pts preference arm, E (arm C) or CRT (arm D). CRT consisted of cisplatin and 5-FU, with radiation at the dose of 60 Gy concurrently. The primary endpoint was overall survival (OS) of arm A and B, secondary endpoint included OS of arm C and D using inverse probability weighting with propensity score. The planned sample size in arm A and B was 114 pts in total with one-sided alpha of 10%, power of 75% and non-inferiority margin of HR as 1.78. The sample size in arm C and D was at least 156 pts in each arm with one-sided alpha of 2.5%, power of 85% and non-inferiority margin of HR as 1.78. Results: Between December 2006 and February 2013, 379 (Arm A: 4, B: 7, C: 209 C, D: 159) pts were enrolled. Primary endpoint was not calculated due to small number of randomized arms. Patients characteristics of arm C and D were as follows; median age: 62 and 65, male (%): 82.8 and 88.1, PS 0 (%): 99.5 and D 98.1. All histologic type was SCC except one basaloid cell carcinoma in arm C. The 3- and 5-year OS were 94.7% and 86.5% in arm C, and 93.1% and 85.5% in arm D (adjusted HR 1.05; 95% CI 0.67-1.64 [< 1.78]).Treatment related death were observed in two pts in arm C and none in arm D. Conclusions: Though the accrual of randomized arms was shortened, CRT showed trend toward non-inferiority compared to E in pts preference arms. CRT is considered as a treatment option for stage IA ESCC with organ preservation. Clinical trial information: UMIN000000551.

104 Background: JCOG0909, a single-arm confirmatory trial of definitive chemoradiotherapy (dCRT) including salvage treatment, demonstrated promising efficacy and safety for cStage II/III (UICC-TNM 6th) esophageal cancer (EC) (Ito Y, ASCO 2018). Radiation (RT) fields included the elective regional lymph node during initial 41.4 Gy, such as bilateral supraclavicular fossae and superior mediastinal lymph nodes for upper thoracic (Ut), and mediastinal and perigastric lymph nodes for middle thoracic (Mt)/lower thoracic (Lt) EC. It is unclear whether the safety and efficacy are associated with the tumor location in patients with cStage II/III EC treated with dCRT. Methods: Patients who were enrolled in JCOG0909 and underwent dCRT were analyzed. Patients were categorized into three groups according to primary tumor location (Ut/Mt/Lt). We compared adverse events during dCRT, complete response (CR) rate, progression-free survival (PFS) and overall survival (OS) among groups. Results: Ninety-four patients (Ut/Mt/Lt: 16/59/19) were analyzed. The proportions of cStage IIA/IIB/III were 31%/44%/25% in Ut group, 20%/42%/37% in Mt group, and 21%/32%/47% in Lt group, respectively. The summary of safety and efficacy was listed in Table 1. Grade 3-4 leukopenia, neutropenia and thrombocytopenia were more frequently observed in Mt and Lt groups than in Ut group. CR rate was 63% in Ut, 63% in Mt, and 42% in Lt, respectively. 3-year PFS in Ut/Mt/Lt was 60%/59%/47% and 3-year OS was 73%/78%/58%, respectively. Conclusions: The RT field by the tumor location might be associated with efficacy and safety of dCRT for cStage II/III esophageal cancer. [Table: see text]

4036Background: RAM (a human IgG1 antibody against vascular endothelial growth factor receptor-2) plus paclitaxel has been found to improve overall survival compared with paclitaxel alone as second...

e13111Background: Breast cancer is the most frequent cancer among women both in US and Japan. Triple negative breast cancer (TNBC), a subset of breast cancers that are negative for estrogen recepto...

4042Background: In spite of the recent advancement of palliative chemotherapy for stage IV gastric cancer (GC), its median survival time (MST) still remains around 13 to 16 months. Much attention h...

39 Background: Obese patients with esophageal cancer have been increasing worldwide. However, the impact of obesity on esophagectomy still remains controversial. In this study, we clarified the relations between obesity and postoperative outcomes after esophagectomy in patients with esophageal cancer. Methods: A total of 215 patients who underwent esophagectomy for esophageal cancer between 2008 and 2014 were retrospectively assessed. We qualified the body mass index (BMI), visceral fat area (VFA) and subcutaneous fat area (SFA) using CYNAPSE VINCENT™. Clincopathological factors, postoperative morbidity, and survival after esophagectomy were assessed with these obesity-related factors. Results: The mean BMI, VFA, and SFA were 22.0 kg/m², 83.0cm², and 87.3cm². BMI, VFA and SFA were significantly correlated each other (P < 0.001). The prevalence of diabetes mellitus and hypertension were significantly more in high SFA than normal SFA (20% vs 10%, 30% vs 14%). When the patients were divided to two groups according to BMI (< 25 or ≥25kg/m²), VFA ( < 100 or ≥100cm²) and SFA ( < 100 or ≥100cm²), operative time was significantly longer in high SFA (P = 0.03) than normal SFA, and tended to be longer in high BMI, but blood loss were similar in each groups. Maximum level of postoperative CRP was significantly higher in high BMI (P < 0.001), in high VFA (P < 0.001), and in high SFA (P < 0.001). Length of systemic inflammatory response syndrome was significantly longer in high BMI than normal BMI (1.9 days vs 1.4 days P = 0.04), and length of ventilator support was significantly longer in high SFA than normal SFA (3.0 days vs 2.2 days P = 0.006). The incidence of anastomotic leakage and surgical site infection were significantly higher in high SFA than normal SFA (27% vs 10%, 37% vs 19%). There were no significant difference in overall survival, however, patients in high VFA tended to be longer survival than those in normal VFA (P = 0.078). Conclusions: Anastomotic leakage and surgical site infection occurred more frequently in high SFA. It might be useful to examine SFA preoperatively in obese patients with esophageal cancer for prevention of postoperative complications. Patients in high VFA tended to be longer survival than those in normal VFA.

169 Background: ESD is the one of the options of treatment even for lateral spreading (size is over 5 cm) esophageal squamous cell carcinoma (L-S ESCC). Some patients have to undergo additional treatment because tumor is invaded to submucosal layer or lymphovascular invasion. On the other hand, wide resection by ESD could cause the delay of additional treatment because of the treatment for esophageal stricture after ESD. Thus, treatment strategy for L-S ESCC has to include additional treatment after ESD. Methods: From January 2010 to December 2014, 49 cases of l L-S ESCC were resected by surgery or ESD. Diagnosis, treatment methods and outcomes are evaluated. Results: In 49 cases of L-S ESCC, 32 cases were treated by ESD and 17 case were treated by surgery. Average size of tumor treated by ESD is 59.4 mm(50-85 mm). Accuracy of estimated depth of invasion by endoscopy for ESD cases is 65.7%. Four of 32 cases of ESD underwent additional therapy (3 for surgery and 1 for CRT) because of pT1b (SM2) or lymphvascular invasion, and one case has lymph-node metastasis. Rate of stricture after ESD is 20.0% for sub-circumference ESD and 77.8% for circumference ESD instead of steroid injection. Average time and duration for control of esophageal stricture by Balloon Bougie is 13.5 times and 18 weeks. Average size of tumor treated by surgery is 76.5 mm(50-130 mm). Accuracy of estimated depth of invasion by endoscopy for surgical cases is 47.1%. Seven cases in 17 (41.2%) have lymph node metastasis. One case died by recurrence after surgery and 48 cases were survived without any recurrences. Conclusions: Accuracy of estimated depth of invasion by endoscopy for L-S ESCC is cute low compared to normal superficial ESCC. Most of strictures after sub-circumference ESD could be prevented by steroid injection. However control of strictures after circumference ESD is difficult. Thus, diagnostic ESD should not be performed for circumferencial lesions of L-S ESCC for the patients who will select CRT for additional treatment, and CRT should be selected for first treatment of these cases. Long survival could be obtained by ESD or surgery for the patients of L-S ESCC by our treatment strategy.

168 Background: Laparoscopic wedge resection is widely performed for managing gastric submucosal tumors (SMTs) that measure ≤5 cm in diameter. Laparoscopy and endoscopy cooperative surgery (LECS) is performed to ensure a sufficient margin while avoiding deformation or stenosis. Methods: We conducted a retrospective review of the data of 76 patients who underwent LECS for SMTs or early gastric cancers at our institution between April 2012 and July 2017. While Non-exposed Endoscopic Wall-inversion Surgery (NEWS) was performed for tumors with a major axis diameter of ≤5 cm and minor axis diameter of ≤3 cm, for other tumors, we adopted classical LECS or used the combination of laparoscopic approaches to neoplasia with non-exposure technique (CLEAN-NET). Results: The tumor was located in the upper-/middle-/lower-third of the stomach in 49/20/7 cases. The mean operation time was 201.7 ± 66.2 min, and the mean blood loss was 4.7 ± 12.8 ml. The mean tumor size was 28.6 ± 11.6 mm. The histological tumor types included GIST (n = 38), leiomyoma (n = 14), schwannoma (n = 6), early gastric cancer (n = 8) and others (n = 10). The surgical procedure consisted of classical LECS/CLEAN-NET/NEWS in 17/15/44 cases; all early gastric cancers were treated by NEWS. The mean length of postoperative hospital stay was 8.7 ± 3.9 days. The mean serum levels of C-reactive protein on POD1 and POD3 were 3.2 ± 1.9 and 4.4 ± 3.4 mg/dl, respectively. The postoperative complications were anastomotic stenosis (n = 1), delayed gastric emptying (n = 3), pneumonia (>Grade II by the Clavien-Dindo classification) (n = 1) and exacerbation of GERD (>Grade II by the Clavien-Dindo classification) (n = 3). In the patient group that developed complications, the postoperative resumption of water intake was later, and the postoperative hospital stay was longer as compared to the group without complications (p = 0.04, 0.028). No relationship was observed between the likelihood of development of complications and the tumor location. Conclusions: Use of minimally invasive surgical procedures such as LECS could minimize the extent of gastric resection without deformation, stenosis. We consider that for surgical complications the LECS approach can be employed, regardless of the tumor location.

12 Background: The sentinel node (SN) concept is safely applied and validated in early gastric cancer. Gastric lymph nodes are divided into five basins with the main gastric arteries. The suprapyloric lymph nodes (No. 5) and the anterosuperior lymph nodes with the common hepatic artery (No. 8a) are classified in the right gastric artery (r-GA) basin, and the subpyloric lymph nodes (No. 6) are classified in the right gastroepiploic artery (r-GEA) basin. The aim of this study is to analyze the lymphatic flows between these lymph nodes and basins. Methods: Five hundred and fourty-nine patients with cT1N0 or cT2N0 gastric cancer underwent SN mapping. We used technetium-99 tin colloid solution and blue dye as a tracer. Results: We detected SN No. 5 in 36 (6.6 %) patients. In these patients, we detected SN No. 5 with SNs that belonged to the left gastric artery (l-GA) basin (69 %), r-GA basin (14 %), and r-GEA basin (44 %). No. 6 was detected as SN significantly more frequently with SN No. 5. We detected SN No. 6 in 100 (18.2 %) patients. In these patients, we detected SN No. 6 with SNs that belonged to the l-GA basin (42 %), left gastroepiploic artery (l-GEA) basin (4 %), r-GA basin (28 %), and r-GEA basin (41 %). No. 4d, No. 5 and No. 8a were detected as SNs significantly more frequently with SN No. 6. We detected SN No. 8a in 43 (7.8 %) patients. In these patients, we detected SN No. 8a with SNs that belonged to the l-GA basin (65 %), l-GEA basin (7 %), r-GA basin (21 %), r-GEA basin (58 %), and posterior gastric artery (p-GA) basin (7 %). No. 6 and No. 9 were detected as SNs significantly more frequently with SN No. 8a. SN No. 5 and No. 8a were tend to be detected with not SNs that belonged to the r-GA basin but No. 6, and SN No. 6 was detected significantly more frequently with No. 4d, No. 5 and No. 8a. Conclusions: Our findings show that lymphatic flows exist not only between lymph nodes which are belonged to the same basin, but also between No. 5 and No.6, and No.6 and No. 8a. On the other hand, the lymphatic flows between No. 5 and No. 8a seem to be limited. Our results suggest that the lymph node dissection of No. 6 given lymphatic flows from lesser curvature side and retroperitoneal side is important for surgery of gastric cancer.

23 Background: Some studies suggest anastomotic leakage (AL) may adversely affect long-term survival in patients undergoing radical gastrectomy (RG) for gastric cancer (GC). Therefore, we assume that common risk evaluation system for short-term and long-term prognosis can be devised in patients with GC that have undergone RG. Methods: Five hundred and twenty patients undergone RG against GC during May 2012 to March 2017 were enrolled. Patients’ characteristics, preoperative laboratory data, operative information and pathological factors were analyzed. Also AL with Clavien-Dindo classification Grade≥III (AL≥III), disease free survival (DFS) and disease specific survival (DSS) were examined. Results: By bringing patients under classification of “with or without relapse” and “with or without AL≥III”, we identified preoperative body weight (BW), serum C-reactive protein (CRP) and albumin (Alb) level were independent risk factors of both relapse and AL≥III. Stratifying patients with recurrence, we obtained crude hazard ratio (HR) of 1.04 (95% CI: 1.02-1.06) for 1kg-increase of BW, 1.51 (95% CI: 1.08-1.40) for 1mg/dl-increase of CRP and 1.23 (95% CI: 1.34-1.64) for 1g/dl-decrease of Alb. With these HR and median value of each parameters of none-relapse group, we defined ABC score (ABCs) is calculated by following formula; ABCs = (4.1-Alb)*51+(58.5-BW)*4+(CRP-0.05)*23. AL≥III were observed in 3.1% of the cases and ABCs of patients with AL≥III was significantly low (with AL≥III; -22.4, without AL≥III; 9.03, p = 0.045). We determine a cut off value of ABCs to -7 on the basis of the ROC curve. Patients with ABCs≤-7 had significantly high risk of AL≥III and we found ABCs≤-7 is an independent risk factor of AL≥III (HR 1.775, p = 0.003). Also, we stratified patients as Low group (L), Mid group (M), High group (H), by approximate tertile value of ABCs, -20 and 30. We found significant shortening of DFS and DSS in order of L, M and H (DFS; L vs M p = 0.006, Mid vs High p = 0.03, DSS; L vs M p = 0.002, L vs H p < 0.001). Moreover, HR for disease specific death for ABCs≥-20 was 4.2 (p = 0.003) in multivariate analysis. Conclusions: ABCs can be a risk evaluation system for short-term and long-term prognosis in patients with GC that have undergone RG.

TPS193 Background: Esophageal cancer (EC) is the sixth most common cause of cancer-related death worldwide, with a 5-y survival rate of 5%–8% in pts with metastatic disease. First–line platinum-based doublet chemotherapy (CTX) provides a modest survival benefit in pts with metastatic squamous cell EC (mESCC), with a median OS of 7.6 months. Over 40% of pts with EC have PD-L1+ tumors, which are associated with worse OS outcomes. Nivolumab (NIVO), an anti–PD-1 mAb, demonstrated efficacy and a manageable safety profile in pts with ESCC. In the phase 2 ATTRACTION-1 trial, NIVO 3 mg/kg produced an ORR of 17% and median OS of 10.8 months in heavily pretreated pts with ESCC (Kudo T et al Lancet Oncol 2017). NIVO + ipilimumab (IPI) yielded enhanced responses with a manageable safety profile in multiple tumor types. Similarly, NIVO + platinum-based CTX resulted in encouraging clinical activity and an acceptable safety profile in pts with metastatic gastric cancer and SCC NSCLC. This randomized phase 3 study will evaluate the efficacy and safety of NIVO + IPI or NIVO + fluorouracil (5-FU)/cisplatin (Cis) vs 5-FU/Cis as first-line treatment (Tx) in pts with advanced (adv) or mESCC (CheckMate 648; NCT03143153). Methods: 939 pts aged ≥ 18 y with histologically confirmed unresectable advanced, recurrent or mESCC, not amenable to curative approaches, and no prior systemic therapy for advanced/metastatic disease, with adequate organ function and ECOG PS 0 or 1 will be randomized 1:1:1 to NIVO + IPI, NIVO + 5-FU/Cis or 5-FU/Cis. Pts with autoimmune diseases, serious or uncontrolled medical disorders; active infections or positive for HIV, AIDS, or hepatitis B/C, indicating acute or chronic infection and/or detectable virus; and previous Tx with other therapy targeting T-cell co-stimulation or immune checkpoint pathways are excluded. Primary endpoints are OS and PFS by central assessment in pts with a PD-L1+ tumor (defined as PD-L1 expression on ≥ 1% of tumor cells). Secondary endpoints include ORR, PFS, and OS in all pts, as well as ORR in pts with PD-L1+ tumors. Clinical trial information: NCT03143153.

22 Background: C-reactive protein to albumin (CRP/Alb) ratio is a novel inflammation-based prognostic score. It was originally reported as a prognostic score in septic patients, however, recent reports show CRP/Alb can be a prognostic score in some cancer patients. There are a few reports about CRP/Alb in patients with esophageal cancer. The usefulness of CRP/Alb in patients with esophageal cancer is still unclear. Methods: The aim of this study is to ensure the evidence of CRP/Alb in patients with esophageal cancer as a prognostic score. One hundred and sixty patients with esophageal cancer were reviewed in the current study. CRP/Alb was calculated before the treatment: In neoadjuvant chemotherapy (NAC) patients, CRP/Alb was calculated before NAC, in non-NAC patients, CRP/Alb was calculated before surgery. CRP/Alb was divided into high group and low group, using cut off value 0.035. The number of 0.035 was determined, referred from other reports of CRP/Alb. Results: The distributions of postoperative pathologic stages 0, 1, 2, 3, and 4 were respectively 4, 51, 33, 56, and 16. The number of high CRP/Alb group was 58 (36.3%). That of low CRP/Alb group was 102 (63.7%). In the survival analysis, the patients with high CRP/Alb showed a significantly shorter overall survival. In the multivariate analysis with pathological stage, neoadjuvant treatment, field of lymph node dissection, and CRP/Alb as covariates, CRP/Alb was detected as a significantly independent prognostic score (hazard ratio 2.602; P < 0.001). Conclusions: In this study, CRP/Alb was thought to be a prognostic score with statistical significance in esophageal cancer. Therefore, further investigation of reducing preoperative inflammation and increasing preoperative nutrition in esophageal cancer is needed.

e15592 Background: Molecular heterogeneity represents a significant hurdle in realizing precision medicine for gastric cancer (GC). Large-scale whole exome sequencing projects have identified distinct molecular subtypes to help define the heterogeneity of GC. However, it remains unclear whether the targeted gene panel-based sequencing can provide optimum targeted therapies and clinical utility in GC. The aim of this study is to generate comprehensive genomic profiling data and classify Japanese GC into actionable clusters associated with targeted therapies. Methods: FFPE tumor tissues were obtained from surgical or biopsy specimens of 207 Japanese patients with GC. Extracted DNA was subjected to genomic sequencing for 435 cancer related genes including 69 druggable genes with FDA approved targeted therapies. Somatic mutations, copy number alterations (SCNA), microsatellite instability (MSI) and Epstein-Barr virus (EBV) infection were evaluated using sequencing data. Results: Genomic sequencing identified at least one alteration of 435 genes in 194 pts (94%), and that of 69 druggable genes in 141 pts (68%). The most frequently altered druggable gene was ERBB2 (14%), following BRCA2 (11%) and ATM (10%). We successfully classified 207 tumors into four molecular subtypes, similar to the previously report; EBV (4%), MSI (8%), chromosomal instability (58%) and genomically stable subtype (30%). Frequent alterations of druggable genes ( > 5%) were widely observed through these subtypes. To discover the novel classifications associated with targeted therapies, we classified 207 tumors using mutation rate and hierarchical clustering. We identified a hypermutated group (n = 32), and a remaining non-hypermutated group (n = 175) which were sub-divided into six clusters including five actionable ones; ERBB2 (n = 25), CDKN2A and CDKN2B (n = 10), KRAS (n = 10), BRCA2 (n = 9) and ATM cluster (n = 12). Interestingly, we experienced a case of unresectable GC with a remarkable response for anti-HER2 therapy in the ERBB2 cluster. Conclusions: Genomic sequencing using a 435-gene panel has the potential to provide the information of optimum targeted therapies for upcoming precision medicine in Japanese GC.

e23122 Background: Breast cancer is the leading cancer among women both in US and Japan. Triple negative breast cancer (TNBC) is the subset of breast cancers that are negative for estrogen receptor (ER), progesterone receptor, and HER2. While the selective ER modulator tamoxifen or anti-HER2 therapy for breast cancer patients with ER positive or HER2 protein overexpression are the most successful examples of targeted therapies, only limited therapies are available for patients with TNBCs, which are associated with a poor prognosis. Advance in next-generation sequencing enables us to identify actionable driver mutations that can be potentially treated by targeted therapies in each cancer patient. The aim of this study is to examine actionable driver mutations in TNBCs in Japan by comprehensive genomic sequencing (CGS) with 435 gene panel, and compare the driver events in Japan with TCGA database to validate the utility of CGS. Methods: We examined all exons of 435 known cancer genes in Japanese TNBC patients (N = 53) by CGS and evaluated for concordance among independent data obtained from The Cancer Genome Atlas-TNBC whole exome sequencing database (N = 123). Results: Oncogenic driver mutations were identified in 51 of 53 Japanese patients (96%) with TNBC and 36 of 53 patients (67%) harbored mutations in genes associated with FDA-approved targeted therapies, indicating the potential clinical utility of a large gene panel for evaluating patients with TNBC. Among 80 total genetic alterations, frequently mutated genes ( > 10% patients) were TP53, PIK3CA and PTEN. Overall, the mutation spectrum of the Japanese patients is similar to that of the TCGA population, except amplification of MYC. Conclusions: Use of a CGS panel of 435 genes can reliably identify all of the critical mutations in TNBC patients, which are similar as TCGA data. The information derived from CGS can be used to determine the optimal treatment for TNBC patients.

TPS4131 Background: Expression of the PD-1 ligands PD-L1 and PD-L2 has been reported in ≈ 40% of pts with E/GEJ cancer and is associated with a poor prognosis. In a phase 3 trial, the PD-1 inhibitor nivo demonstrated an OS benefit vs placebo (HR, 0.63; P < 0.0001), resulting in a 37% reduction in the risk of death and double the OS rate at 12 mo (27% vs 11%) in pts with advanced gastric (G)/GEJ cancer refractory to ≥ 2 lines of chemotherapy (Kang YK, et al. J Clin Oncol. 2017;35 (suppl 4S) [abstract 2]). In this study, nivo was well tolerated, with a safety profile comparable with that of the placebo arm. These results indicate that nivo could be a new standard of care (SOC) for pts with heavily pretreated advanced G/GEJ cancer and provide a strong rationale to explore nivo in earlier lines of treatment for G/E/GEJ cancer. Currently, no effective adjuvant SOC is available after chemoradiotherapy (CRT) followed by resection for pts with E/GEJ cancer. This multinational, double-blind, phase 3 trial will evaluate nivo as an adjuvant therapy for pts with resected E/GEJ cancer (CheckMate 577; NCT02743494). Methods: In this study, an estimated 760 pts aged ≥ 18 years with stage II/III E/GEJ cancer are randomized to receive nivo or placebo. Prior to randomization, pts must have completed preoperative CRT followed by surgery and been diagnosed with residual pathologic disease after being surgically rendered free of disease with negative margins following complete resection. Pts with stage 4 resectable disease, cervical esophageal cancer, or those who have not received concurrent CRT prior to surgery are not eligible for study enrollment. Primary endpoints are OS and disease-free survival. Other key endpoints include the OS rate at 1, 2, and 3 years and safety. Clinical trial information: NCT02743494.

e12123 Background: Chemotherapy for breast cancer destroys non–stem cells while sparing cancer stem cells (CSCs). In contrast, anti–HER2 therapy may eliminate resistant cells because HER2 may be a key driver of CSCs. CSC biomarkers have been found to be prognostic of poor outcome and predictive of resistance to therapy. However, there are no comprehensive studies of the impact of anti-HER2 therapies on CSC–related biomarkers. We conducted a prospective biomarker determination study of breast CSCs characterized by CD44v expression and increased aldehyde dehydrogenase 1 (ALDH1) enzymatic activity or expression. Methods: In a prospective trial (ClinicalTrials.gov: NCT01688609), 18 patients with operable primary HER2+ breast cancer (≥T2 excluding inflammatory, any N; median age of 54 yrs) were treated with preoperative anti–HER2 therapy following the NeoALTTO trial dual therapy arm regimen, with a goal of identifying novel predictive biomarkers for pCR. Proportions of tumor cells with CSC characteristics, defined as CD44v+ and ALDH1+, were estimated at baseline, at 6 weeks (after therapy with lapatinib/trastuzumab) and at 18 weeks (after therapy with lapatinib/trastuzumab and paclitaxel) to assess adaptive response. We determined changes in the quantity and characteristics of CSC–related biomarkers during preoperative therapy and correlated them to tumor response. Results: Out of 18 patients, 8 (44%) had a pCR; 5 of these 8 patients (62%) were positive for CD44v staining on tumor cells at baseline and none were positive on the 6–week biopsy. In contrast, 6 of the 10 patients without pCR exhibited persistent levels, or enrichment of CD44v proportion and expression at 6 and 18 weeks (p = 0.0128). ALDH1 expression and other biomarkers were not statistically significant predictors of pCR. Conclusions: Enrichment of CD44v+ tumor cells after double anti–HER2 therapy may predict poor response to dual anti–HER2 therapy with cytotoxic chemotherapy. A second biopsy after the start of preoperative therapy may reflect biological changes useful for the guidance and application of therapeutic strategies for patients with HER2+ breast cancer. Clinical trial information: NCT01688609.

26 Background: Recent meta-analyses have shown acceptable SN detection rates and accuracy of determination of lymph node status. SN mapping may play a key role to obtain individual metastatic information and allows modification of the surgical procedures such as function-preserving gastrectomy for early gastric cancer. Methods: A dual-tracer method that employs radioactive colloids and blue dyes is currently considered the most reliable method for the stable detection of SNs in patients with early gastric cancer. The new technologies such as indocyanine green infrared or fluorescence imaging are also useful for accurate SN mapping in gastric cancer. Results: SN mapping has been performed for 489 patients with clinically T1N0M0 gastric cancer in our institute since 1999. Sensitivity to predict the lymph node metastasis was 90%, and overall accuracy of nodal status based on SN concept was 99 % (484 of 489). Sentinel lymphatic basin dissection is strongly recommended to cover the false-negative results of SN mapping in clinical practices. Based on these results, laparoscopic minimized gastrectomy such as partial resection, proximal gastrectomy, segmental gastrectomy and pylorus-preserving gastrectomy with individualized selective and modified lymphadenectomy for early gastric cancer with negative SN has been performed in our institution. The recent appearance of a new technique, referred to as non-exposed endoscopic wall-inversion surgery (NEWS) is a technique of full thickness resection using endoscopy and laparoscopic surgery without transluminal access mainly designed to treat gastric cancer with partial resection. The combination of NEWS with SN biopsy is expected to become a promising, ideal minimally invasive, function-preserving surgery for cN0 early gastric cancer. We initiated a new multicenter prospective trial to evaluate function-preserving gastrectomy with SN mapping in terms of long-term survival and patients’ QOL. Conclusions: Our results suggested that SN concept for cN0 early gastric cancer could be validated, and function-preserving surgery based on the SN concept which might retain the patients’ quality of life should be established as the next surgical challenge.

31 Background: Wide variety of Predictive factors (PF) of postoperative complications (PC) in patients with gastric cancer(GC) that have undergone laparoscopic gastrectomy(LG) has been reported. The aim of this study is to devise simple PF for PC for LG that can estimate risk of PC preoperatively. Methods: Two hundred and ninety-eight patients that have been underwent curative LG against GC during May 2012 to March 2016 were enrolled. Patients’ characteristics, preoperative (pre-) laboratory data, operative information and pathological factors were analyzed. Also PC with Clavien-Dindo classification Grade (CD)≥II (PC≥II), pancreatic fistula (PF), intraabdominal abscess (IA), pneumonia (PN) and anastomotic leakage (AL) with CD≥III were examined. Results: Patients with pStage I, II, III and IV was 83%, 11%, 5.6% and 0.4%, respectively. Patients with PC≥II, PF or IA, PN and AL was 17.4%, 5%, 1% and 2.3%, respectively. Patients with PC≥II were significantly elder (p=0.01) and had higher pre-serum-C-reactive protein level (CRP) (p=0.02). Also, patients with AL had significantly higher Body Mass Index (BMI) (p=0.04). We devised “ABC score” by Age, BMI and CRP. It is defined as a number of factors in age, BMI and CRP that are higher than cutoff value of 66 years’ old, 23 kg/m2 and 0.04 mg/dL, respectively. Patients with ABCs 0, 1, 2, and 3 were 26%, 33%, 34% and 7%, respectively. Patients with ABCs 3 had significant high risk of PC≥II and AL (HR 2.8, p=0.03 and HR 9.9, p=0.004, respectively). Also we found higher ABCs has higher risk of AL (HR1.2, p=0.07 for ABCs 2 and HR2.2, p=0.04 for ABCs 3). Conclusions: ABCs can be a PF for PC in patients with LG. Especially it may evaluate an incidence ratio of AL.

183 Background: AF is common after thoracic and esophageal surgical procedures. The full spectrum of risk factors, associations, and implications are unclear. Objective:The aim of this study was to assess the impact of new-onset atrial fibrillation (AF) after esophagectomy on short- and long-term outcome. Methods: A retrospective analysis was performed on patients with esophageal cancer who underwent esophagectomy with thoracotomy and laparotomy, with R0 or R1 resection between January 1997 and December 2012. For the 402 patients included in this study, we analyzed the stage of disease, neoadjuvant therapies, surgical approaches, surgical complications, postoperative medical complications, and overall and relapse-free survivals using medical records. Results: The median age was 62 (range 34-82). Most patients were male (90.0%) and had squamous cell carcinoma (90.3%). Pathological staging of the esophageal cancers according to UICC 7th TNM classification system was categorized as follows: 17 (4.2%), 96 (23.9%), 16 (4.0%), 43 (10.7%), 57 (14.2%), 78 (19.4%), 39 (9.7%), 37 (9.2%), and 19 (4.7%) patients were designated as stage 0, ⅠA, ⅠB, ⅡA, ⅡB, ⅢA, ⅢB, ⅢC and Ⅳ respectively. Of the 402 patients studied, 42 (10.6%) had AF, 87 (21.6%) had pneumonia, 80 (20.0%) had anastomotic leakage, and 69 (17.2%) had recurrent laryngeal nerve paralysis. At a median follow-up of 133 months, the median survival was 49 months, and 5-year survival was 48% in the AF group, compared with 138 months and 60% in the non-AF group (p = 0.212). AF was significantly associated with recurrent laryngeal nerve paralysis (p = 0.001), but not with pneumonia or anastomotic leakage. Conclusions: New-onset AF is common, mostly associated with recurrent laryngeal nerve paralysis, but not with pneumonia or anastomotic leakage. AF has no impact on oncologic outcomes.

575 Background: Postoperative adjuvant chemotherapy (CTx) is a standardized treatment for patients with Stage III colorectal cancer (CRC). Since previous studies reported that the benefit of adjuvant CTx is estimated up to only 15% in Stage III patients, it is necessary to establish an appropriate patient selection for adjuvant CTx. Recently, systemic inflammation is reported to play a critical role in the progression of cancer and several studies demonstrate the impact of inflammation-based score, such as mGPS and SIS. There however is a lack of evidence for the influence of inflammation on patient requiring adjuvant CTx. The aim of this study is to evaluate the impact of SIS on patient selection of adjuvant CTx for Stage III CRC. Methods: Between 2004 and 2012 a total of 295 patients with Stage III CRC who underwent curative resection were included. The SIS was developed based on serum albumin and lymphocyte-to-monocyte ratio (LMR) levels (SIS = 0; albumin > 4.0 g/dl and LMR > 4.44, SIS = 1; albumin < 4.0 g/dl or LMR < 4.44, SIS = 2; albumin < 4.0 g/dl and LMR < 4.44). Primary endpoint was defined as disease free survival (DFS). DFS rates were compared using Kaplan-Meier curves and Prognostic factors were assessed by Cox proportional hazard model. Results: The median follow-up period was 5.99 years. 75.9% of patients received adjuvant CTx. The distribution of SIS was SIS = 0; 32.6 %, SIS = 1; 43.3 %, and SIS = 2; 24.1 %. In multivariate analysis for overall patients, SIS was not a significant prognostic factor. Subgroup analysis according to each SIS score demonstrated that adjuvant CTx significantly prevented postoperative recurrence just only in patients with SIS = 0 (5-year DFS rate; adjuvant CTx (+) 85.5 % vs. CTx (-) 33.7 %, p < 0.001 ). In Cox proportional hazard model, protective effect of adjuvant CTx decreased in proportion to the increase of SIS (HR of adjuvant CTx (95 % CI): SIS = 0; 0.187 (0.089-0.393), SIS = 1; 0.601 (0.311–1.162), SIS = 2; 0.747 (0.319–1.752)). Conclusions: The SIS can provide an appropriate patient selection for patients with stage III CRC who can benefit from adjuvant CTx. The results from this study might enhance precision medicine for adjuvant CTx for Stage III CRC.

36 Background: The clinical outcome of esophageal cancer has recently improved. However, the relapse rate in any stages is still high. Prognostic classification in patients with esophageal cancer has not yet established. If the patients with poor prognosis could be extracted before surgery, we can add more intensive preoperative treatment, which might lead more favorable prognosis. We have previously reported clinical utility of fibrinogen (FNG) and Albumin (Alb) Score (FA score) in esophageal cancer. The purpose of this study was to analyze the utility of FA score comparing at each stage as a prognostic factor in esophageal cancer. Methods: Patient characteristics, clinicopathological factors, and preoperative biochemical markers (FNG, Alb, and CRP) were investigated in esophageal cancer patients who underwent transthoracic esophagectomy between January 2004 and November 2013. Correlation between preoperative FNG, CRP, Alb, clnicopathological factors and survival were investigated. We evaluated between recurrence-free survival (RFS), overall survival (OS) and FA score comparing at each stage. The cut-off value was defined according to previous reports; fibrinogen 350mg/dl, albumin 3.8mg/dl. Results: 252 patients were recorded. High FA score were correlated to worse OS and RFS (p < 0.001). Especially in cStageI, high FA score showed remarkably worse 5-year RFS (score0/ score1/ score2, 5-year RFS 84%/ 64%/ 33%; p = 0.038). Conclusions: FA score was significantly associated with postoperative survival in esophageal cancer patients. If the patients with poor prognosis in cStageI could be extracted before surgery, we can add more intensive preoperative treatment, which might lead more favorable prognosis.

582 Background: Our previous study suggested that smoking history could be a risk factor for pulmonary metastasis following rectal cancer surgery. Smoking causes emphysematous change in the lung, however, it remains unclear how much impact of pulmonary emphysema (PE) on the pathogenesis of pulmonary metastasis in rectal cancer. The objective of this study is to clarify the influence of PE on the pulmonary metastasis. Methods: Patients with rectal cancer who underwent curative resection from September 2005 to May 2014 in our institution were enrolled and their medical records were reviewed including the percentage volume of low attenuation area in the whole lung (LAA%), which were calculated from their preoperative chest CT by using AZE virtual place. PE was defined as more than 10% in LAA% and statistically evaluated the correlation with relapse free survival for pulmonary metastasis (RFSP). Results: A total of 286 patients, who consists of 7 in stage0,102 in StageⅠ, 75 in StageⅡ, and 102 in stageⅢ, were enrolled. The median LAA% was 3.16±3.78% and 14 patients (4.9%) were diagnosed PE by LAA%. Postoperative pulmonary metastasis was observed in 22 patients, including 3 patients in PE. Five-year RFSP rate in PE was significantly poorer than non-PE (25.4% vs 7.7%, p=0.02) by log-rank test. Multivariate survival analysis using Cox’s regression model revealed that PE was an independent risk factor for RFSP (Hazard Ratio = 19.14 [3.18 – 115.20], p < 0.001). Conclusions: Pulmonary emphysema may have a robust impact on pulmonary relapse after curative surgical resection of rectal cancer.

TPS212 Background: Chemoradiotherapy (CRT) followed by surgical resection is a current standard of care (SOC) for pts with E/GEJ cancer, with 3- and 5-year survival rates ranging from 30% to 40%. Currently, no effective adjuvant SOC is available following resection. Expression of programmed death-1 (PD-1) ligands 1/2 (PD-L1/L2) has been associated with a poor prognosis in E/GEJ cancer, suggesting that PD-1 inhibition may improve outcomes. Nivo is a fully human IgG4 monoclonal antibody that targets PD-1, with demonstrated survival benefit in multiple tumor types and long-term responses in some patients. A phase 1/2 study of nivo monotherapy in chemotherapy-refractory pts with gastric/E/GEJ cancer demonstrated tumor regression, a median overall survival (OS) of 5 months, and a 12-month OS rate of 36% in pts with PD-L1+ and PD-L1- tumors (Janjigian Y, et al. J Clin Oncol. 2016;34:suppl; abstract 4010). This multinational, double-blind, phase 3 trial will evaluate nivo as an adjuvant therapy for pts with resected E/GEJ cancer (CheckMate 577; NCT02743494). Methods: In this study, an estimated 760 pts aged ≥ 18 years with stage II/III E/GEJ cancer are randomized to receive nivo or placebo. Prior to randomization, pts must have completed preoperative CRT followed by surgery and been diagnosed with residual pathological disease after being surgically rendered free of disease with negative margins following complete resection. Pts with stage 4 resectable disease, cervical esophageal cancer, or those who have not received concurrent CRT prior to surgery are not eligible for study enrollment. Primary endpoints are OS and disease-free survival. The secondary endpoint is OS rate at 1, 2, and 3 years. Clinical trial information: NCT02743494.

4021Background: This multicenter phase II trial assessed the safety and efficacy of docetaxel plus 5-fluorouracil and cisplatin (DCF) induction chemotherapy (IC) and subsequent conversion surgery (...

e22516Background: Although 3-year imatinib adjuvant therapy is recommended by the guidelines, there is a lack of the data in clinical practice regarding use of adjuvant therapy and adherence. A pro...

353 Background: Although surgery plays a key role in the curative treatment for biliary tract cancer (BTC), the prognosis is not satisfactory because of frequent postoperative recurrence. Multimodal treatment has been warranted in order to improve this situation. The aim of this study was to determine the maximum tolerated dose (MTD) and recommended dose (RD) of cisplatin as neoadjuvant chemoradiotherapy with S-1 and external beam radiation for BTC. Methods: Patients with resectable BTC at initial diagnosis were eligible for inclusion. The planned dosages of cisplatin (mg/m2)/S-1 (mg/m2/day) were as follows: 40/80 (level 0); 50/80 (level 2); 60/80 (level 3). S-1 was orally administered twice daily from day 1 to 14 and day 29 to 42. Cisplatin was administered intravenously over 2 hours in each dosage of level at day 1 and 29. Radiation was delivered as a total dose of 50.4 Gy (28 fractions to 1.8 Gy). Dose-limiting toxicity (DLT) was a delay factor of the schedule due to incomplete recovery. (Response rate (RR) was evaluated according to the RECIST version 1.0.). Results: Twelve patients were enrolled, and level 1 was chosen as the starting dose. One of six patients developed DLTs at level 1, and the dose was escalated to level 2. DLTs occurred in two of six patients at level 2. As the MTD were determined as over 60 mg/m2, the RD were confirmed as level 2 for a subsequent phase II study. Conclusions: We determined the MTD and RD of cisplatin with concurrent S-1 and external beam radiation as neoadjuvant chemoradiotherapy for resectable BTC. Although further clinical trials are warranted to evaluate efficacy, this new regimen is potentially expected to become a standard neoadjuvant treatment for resectable BTC. Clinical trial information: 000009028.

29 Background: Venous thromboembolism (VTE) is sometimes found in esophageal cancer patients who underwent surgical treatment. VTE contains deep vein thrombosis, pulmonary embolism and catheter-related thrombosis. Our previous study revealed that pre-therapeutic plasma fibrinogen level, C-reactive protein level, adenocarcinoma histology and neck lymph node dissection are the risk factors for venous thromboembolism in patients[Surg Today, 2015]. However, the correlation between VTE and prognosis is not clear in esophageal cancer. We hypothesized that VTE may have an impact on prognosis of esophageal cancer patients. Methods: One hundred and seventy-two patients who underwent radical esophagectomy from March 2008 to December 2012 in our hospital were reviewed in this study. The existence of VTE was assessed from the neck to the pelvis with computed tomography at the initial visit and after neo-adjuvant chemotherapy (NAC) and 6thpostoperative day. The patient and tumor characteristics, neo-adjuvant chemotherapy were compared between patients with VTE (VTE group) and those without VTE (non-VTE group). The primary outcome is disease-free survival (DFS) and the secondary outcome is overall survival (OS). Results: Twenty-one VTE events among 172 patients (12%) were observed. Six of which occurred preoperatively and were considered to be associated with NAC, 14 were detected postoperatively and one occurred just after inserting a peripherally inserted central catheter preoperatively. The VTE group and the non-VTE group have homogenous characteristics in patients’ backgrounds and tumor features. We found no significant difference in median DFS and OS between two groups. However, in patients with pathologically N 0 or 1(7thedition of UICC TNM classification, n = 157), the median DFS was significantly shorter in VTE group compared with non-VTE group (41 months versus 64 months, p = 0.04). The recurrence risk increased 50.9% in VTE group in comparison with non VTE group (p = 0.048). By using logistic regression analysis, we found that existence of VTE is an independent risk factor with Odd’s ratio 2.964 (p = 0.026). Conclusions: Our study suggests that VTE may be the risk factor for recurrence in esophageal cancer patients.

38 Background: Depression of skeletal muscle mass (sarcopenia) has been linked to postoperative mortality in several carcinomas and various types of surgeries. Total psoas muscle area (TPA) has received attention as one of the indicators of systemic total skeletal muscle volume. The objective of this study is to determine the correlation between the patients who decreased TPA during pre-post operation and prognosis of esophageal cancer treated by surgical resection. Methods: We enrolled 40 esophageal cancer patients who underwent esophagectomy between April 2008 and May 2009. TPA was estimated by measuring the cross-sectional area of the psoas major muscle at the level of the third lumbar vertebra using an image-analysing software, identified on a preoperative CT scan, and a postoperative CT scan, which was taken on post-operative day 6. We divided the patients who decreased TPA between pre-post operation, into “decreasing group”, and the patients who preserved or increased TPA, into “non-decreasing group”. We investigated their backgrounds and prognosis. Results: Eighteen patients (45.0%) were classified as decreasing group, and 22(55.0%) patients were classified as non-decreasing group. Tumor locations, TMN Stage, or rates of neo-adjuvant chemotherapy, did not significantly differ between the two groups. Five-year overall survival rates (54.2% vs 85.7%, P = 0.046) were significantly lower in decreasing group in comparison with non-decreasing group. Recurrence-free survival rates (42.9% vs 85.7%, P = 0.012) were also significantly lower in decreasing group. The incidences of anastomotic leakage, or postoperative pneumonitis, did not differ between the two groups. Conclusions: The patients who decreased TPA during perioperative period associated with poor prognosis of esophageal cancer. The surgical operation for esophageal cancer is highly invasive, therefore active nutrition support and early rehabilitation during perioperative period might be indispensable.