Gastric cancer

Inhibition of paclitaxel by α1-acid glycoprotein and biochemical modulation

Author(s)：: Yoshinao Ohbatake¹, Sachio Fushida¹, Tomoya Tsukada¹, Jun Kinoshita¹, Katsunobu Oyama¹, Kouichi Okamoto¹, Keishi Nakamura¹, Hidehiro Tajima¹, Hiroyuki Takamura¹, Itasu Ninomiya¹, Hirohisa Kitagawa¹, Takashi Fujimura¹, Masakazu Yashiro², Kousei Hirakawa², Tetsuo Ohta¹
Affiliation(s)：: 1. Gastroenterologic Surgery, Graduate School of Medical Science, Kanazawa Univ.
2. Department of Surgical Oncology Osaka City Univ. Graduate School of Medicine

Introduction
Peritoneal carcinomatosis (PC) is the most common mode of metastasis in gastric cancer and the prognosis is dismal. Paclitaxel (PTX) is often used for the gastric cancer patient with PC because it can be easily disseminated, and high rate of transition into the peritoneal cavity. But, its effect is not sufficient.
In the peripheral circulation, drugs are present either as unbound fraction, or bound one. In general, unbound fraction has pharmacodynamics activity. A large fraction of PTX bounds to serum protein such as albumin, and α1-acid glycoprotein (AGP) after administration. It is well known that the level of AGP in serum and ascites often increase as cancer progresses. In this context, we assumed elevated AGP in gastric cancer patient decreased unbound fraction of PTX and then inhibited its antitumor effect. Moreover, it was reported that Erythromycin (EM) also bound to AGP. Biochemical modulation is a combination chemotherapy for increasing antitumor effect by adding substance to antitumor agent. We assumed EM was able to compete with PTX to bind AGP, as a result free fraction of PTX might increase and anticancer effect might be restored. We therefore examined whether the level of AGP increased in gastric cancer patient, and AGP inhibited the activity of PTX. We also examined the effect of coadministration of PTX and EM on AGP-mediated suppression of PTX's therapeutic activity.

Methods
We measured the concentration of AGP in serum and in acites of gastric cancer patient with PC. We also measured the level of AGP in serum of healthy volunteers. In addition, to reveal the change of AGP level with cancer progression, we examined the concentration of AGP in acites of xenograft model mouse. Xenograft model mouse was made by incubating OCUM-2MD3 cells in the peritoneal cavity which was highly peritoneal-seeding cell line derived from human scirrhous gastric cancer.
Next, to examine whether AGP inhibits the activity of PTX in vitro, we added PTX (10 nM) and AGP (0 ～ 1200 μg/ml) to the OCUM-2MD3 cells, and cell viability was assessed by MTT assay. We also examined the combined effects of PTX and EM in vitro. To examine the effect of EM on AGP-mediated suppression of PTX's activity, we added EM (0.1 ～ 100 μM) to the culture with AGP (800 μg/ml), PTX (10 nM), and OCUM-2MD3 cells. Cell viability was examined by MTT assay.
We finally examined whether EM reactivates the antitumor effect of PTX in vivo. OCUM-2MD3 cells were injected into peritoneal cavity of the mouse at day 0. The mice were divided into 3 groups : a control group, PTX group , coadministration (PTX and EM) group. PTX was intravenousely administered at 5 mg/kg once per week for two concecutive weeks (day 7 and day 14). EM was administered at 5 mg/kg subcutaneously from day 7 to day 17. The mice were sacrificed at day 17. The weight of the metastatic nodule was examined.
Results
Serum AGP concentration of gastric cancer patients with PC were about two and a half higher than those of healthy volunteers (1524 ± 586 vs. 650 ± 158 μg/ml, P = 0.0049). AGP level in acites of gastric cancer patients with PC was 834 μg/ml, which was higher than serum AGP level of healthy volunteers. The concentration of AGP in acites of xenograft model mouse was increasing from day 10 with cancer progression, and reached 400 μg/ml at day 21.
In vitro, AGP suppressed the cell growth-inhibitory effect of PTX in a dose-dependent manner, and more than 400 μg/ml of AGP suppressed it significantly. The additon of EM significantly reactivated the activity of PTX suppressed by AGP.
In vivo study, the weight of the metastatic nodules in the group receiving coadministration of PTX and EM were significantly reduced but not significantly in the group receiving PTX alone (p=0.011).

Conclusion
In this study, we demonstrated that AGP level in acites of gastric cancer with PC was going up with cancer progression, and elevated AGP inhibited the activity of PTX. We also demonstrated using EM as a chemical modulator, restored suppressed PTX's activity in vitro and in vivo. The combined therapy with PTX and EM might be useful for treatment of PC in gastric cancer.

Randomized phase III trial of irinotecan plus cisplatin versus irinotecan in patients with
metastatic gastric cancer resistant to S-1 monotherapy: Survival analysis

Author(s)：: Shinya Tokunaga¹, Kazuhiro Nishikawa², Yusuke Akamaru³, Yoshio Oka⁴, Masato Kataoka⁵, Hideaki Ando⁶, Shuichi Hironaka⁷, Satoshi Morita⁸, Junichi Sakamoto⁹, Toshimasa Tsujinaka¹⁰
Affiliation(s)：: ¹Department of Clinical Oncology, Osaka City General Hospital, Japan, ²Department of Surgery, Osaka Medical Center, Japan, ³Department of Surgery, Osaka Kosei Nenkin Hospital, Japan, ⁴Department of Surgery, Nishinomiya Municipal Central Hospital, Japan, ⁵Department of Surgery, Nagoya Medical Center, Japan, ⁶Department of Surgery, Nakadori General Hospital, Japan, ⁷Department of Gastroenterology, Chiba Cancer Center, Japan, ⁸Department of Biomedical Statistics and Bioinformatics, Kyoto Univ. Graduate School of Medicine , Japan, ⁹Tokai Central Hospital, Japan, ¹⁰Department of Surgery, Kaizuka City Hospital, Japan

【Background】The standard second-line regimen for the treatment of metastatic gastric cancer remains to be established. We have implemented a phase III trial comparing combination therapy with irinotecan (CPT-11) and cisplatin (CDDP) to monotherapy with CPT-11 (TRICS trial). The results of the safety and a part of efficacy analysis have already been reported. Here we report the results of the survival analysis after the predetermined follow-up period. UMIN-ID: 000002571
【Material and methods】 Patients with metastatic gastric cancer resistant to S-1 monotherapy were randomly allocated to CPT-11 + CDDP arm (60 mg/m2 of CPT-11, 30 mg/m2 of CDDP, Day 1, every 2 weeks) or CPT-11 arm (150 mg/m2, Day 1, every 2 weeks). The main eligibility criteria included disease progression (PD) after receiving S-1 monotherapy or recurrence within 6 months (M) after or during S-1 adjuvant chemotherapy (discontinuation due to adverse events was not included), PS 0 or 1, age over 20 years old. The stratification factors were made according to PS, PD or recurrence, institution and presence or absence of target lesions. The primary endpoint was to compare overall survival (OS). The secondary endpoints are progression free survival (PFS), time to treatment failure (TTF), response rate (RR), and safety. The required sample size was 160 patients assuming 74 events with alpha error 0.05 and beta error 0.2 to detect differences in median OS (CPT-11 arm: 5 M, CPT-11 + CDDP arm: 8 M). The planned sample size was 200, including ineligibility or dropout.
【Results】Between 2007 and 2011, 168 patients were randomly assigned to the CPT-11 + CDDP arm or CPT-11 arm, respectively. The median follow-up was 59.0 M. Median OS was 13.9 M for CPT-11 + CDDP arm and 12.7 M for CPT-11 arm (hazard ratio [HR] 0.834, 95% CI 0.596-1.167, p=0.288). There were no significant differences in the secondary endpoints: PFS (4.6 M vs. 4.1 M, HR 0.860, p=0.376), RR (16.9% vs. 15.4%, p=0.812) and TTF (3.3 M vs. 3.5 M, HR 1.009, p=0.96). The incidence of grade 3-4 neutropenia (35.4% vs 27.2%, p=0.259), anemia (15.9% vs 3.7%, p=0.009) and LDH (5% vs. 0%, p=0.04) was higher in CPT-11 + CDDP arm than in CPT-11 arm. And in any grade adverse events, the incidence of leukopcytes, platelets, creatinine was higher in CPT-11 + CDDP arm, In contrast, diarrhea, constipation and mucositis was higher in CPT-11 arm. An exploratory subgroup analysis revealed that CPT-11 + CDDP arm was significantly more effective in intestinal type gastric adenocarcinoma (median OS 17.3 M vs. 14.0 M, HR 0.561, p=0.021).
【Conclusions】CPT-11 + CDDP combination therapy was not statistically superior to CPT-11 monotherapy for OS. But the combination regimen may be particularly effective for intestinal type tumors.

A feasibility study of docetaxel, cisplatin and S-1-Trastuzumab (DCS-T) chemotherapy
for HER2 positive unresectable gastric cancer

Author(s)：: Yasushi Sato¹, Tetsuji Takayama², Tamotsu Sagawa³, Hiroshi Miyamoto²,
Masahiro Hirakawa¹, Hiroyuki Ohnuma¹, Yasuhiro Sato³, Yasuo Takahashi³,
Shinichi Ktsuki⁴,Tokunori Okuda⁵, Rishu Takimoto¹, Masayoshi Kobune¹,
Takayuki Nobuoka⁶, Koichi Hirata⁶, Junji Kato¹
Affiliation(s)：: ¹Dep. of Medical Oncology and Hematology, Sapporo Med.Univ. School of Med., Japan, ²Univ. of Tokushima Graduate School, Dep. of Gastroenterology and Oncology, Institute of Health Biosciences, Japan, ³Dep. of Gastroenterology, Hokkaido Cancer Center, Japan, ⁴Center of Gastroenterology, Otaru Ekisaikai Hosp., Japan, ⁵Dep. of Gastroenterology, Oji General Hosp., Japan, ⁶Dep. of Surgery, Surgical Oncology and Science, Sapporo Med. Univ. School of Med., Japan

Background: To date, we have developed a triplet-drug combination chemotherapy regimen consisting of docetaxel, CDDP, and S-1 (DCS regimen) and reported that the regimen is associated with a high response rate (Takayma T et al, BJC 2007; Sato Y et al, CCP 2009 and 2013), and have experienced many conversion cases among patients with initially unresectable gastric cancer. Recently the efficacy of trastuzumab (T-mab) for HER2-positive gastric cancer has been reported (Bang YJ et al, Lancet 2010). To increase the efficacy of DCS in patients with HER2-positive unresectable gastric cancer, we carried out a feasibility study for the DCS-T-mab (DCS-T) regimen.
Methods: Eligibility criteria included gastric or esophagogastric junction adenocarcinoma; HER2-positive confirmed by IHC and/or FISH (IHC3+ or IHC2+/FISH+); unresectable or recurrent; measurable or non-measurable disease; no history of chemotherapy or radiotherapy; age 20-80; ECOG PS of 0-2; normal left ventricular ejection fraction (LVEF ≥ 50%), and adequate organ function. Patients received oral S-1 (40 mg/m2 b.i.d.) on days 1-14, intravenous CDDP (60 mg/m2), docetaxel (50 mg/m2) and T-mab (8 mg/kg in the first cycle and 6 mg/kg in the second cycle and thereafter) on day 8 every 3 weeks. This feasibility study was considered to be met when the proportion of patients who had completed three or more cycle of DCS-T regimen exceeded 85 % according to our previous reports.
Results: This study included 15 patients whose clinical characteristics were as follows: median age, 60 years (34 – 76 years), 10 males and 5 females. PS 0/1/2, 9/4/2; differentiated-type / undifferentiated-type histology, 11/4; U/M/L, 6/8/1; HER2 3+, 12; HER2 2+/FISH+, 3; T3/T4a/T4b, 10/4/1; N0/N1/N2/N3, 2/0/3/10; and distant lymph nodes/liver/peritoneum/lungs/bone/ovaries, 10/7/4/2/1/1. Metastasis to 2 or more organs was observed in 8 cases. The completion rate until the third cycle was 100%. According to the RECIST criteria, the objective response rate was 100%, and the median period to achieve a PR was 1 cycle. Most common grade ≥ 3 toxicities were neutropenia (86.7%), leukopenia (73.3%), anemia (13.3%), febrile neutropenia (8.3%). anorexia (26.7%), and diarrhea (26.7%). All of these side effects were well controlled. Non-curative factors disappeared in 8 of 15 cases and R0 resection was carried out in 6 cases (40%) including 2 patients with liver metastases. A pathological response was found in 83 % of 6 resected cases.

Conclusions: Our data suggested the feasibility of this new therapeutic combination in patients with unresectable HER2-positive gastric cancer. The observed response rate and curative resection rate are very promising and warrant further prospective investigations to establish a "conversion" therapeutic strategy for HER2-positive unresectable gastric cancer patients.

A phase II trial of intravenous and intraperitoneal paclitaxel combined with S-1 for treatment of gastric cancer with macroscopic peritoneal metastasis

Author(s)：: Hironori Yamaguchi M.D. Ph.D., Joji Kitayama M.D. Ph.D., Hironori Ishigami M.D., Shigenobu Emoto M.D., and Toshiaki Watanabe M.D. Ph.D.
Affiliation(s)：: Department of Surgical Oncology, the University of Tokyo, Tokyo, Japan

Background: Although the prognosis of gastric cancer patients with peritoneal metastasis is extremely poor, the standard treatment has not been decided yet. Novel treatment strategy for this dismal disease is needed. This phase II study evaluated the benefits and tolerability of weekly intravenous and intraperitoneal paclitaxel (PTX) treatment combined with oral S-1 in gastric cancer patients who had macroscopic peritoneal metastasis using the evaluation system of investigational medical care in Japan.
Methods: Gastric cancer patients who had primary tumors with macroscopic peritoneal metastasis which was confirmed by staging laparoscopy were enrolled. The intraperitoneal access port was implanted in the subcutaneous space on the patient's abdomen at the operation of staging laparoscopy. PTX was administered intravenously at 50 mg/m2 and intraperitoneally at 20 mg/m2 on days 1 and 8, respectively. S-1 was administered at 80 mg/(m²•d) for 14 consecutive days, followed by 7 d of rest. The primary end point was the 1-year overall survival (OS) rate. The secondary end points were the response rate, efficacy against malignant ascites, and safety.
Results: Thirty-five patients were enrolled from Dec 2009 to Nov 2010. The follow-up period was 1.5 years since the last enrollment. The median number of treatment courses was 11 (range, 2–35). The 1-year OS rate was 77.1% (95% confidence interval, 60.5–88.1), the 2-year OS rate was 44.8%, and Median survival time (MST) was 17.6 months. The overall response rate was 71% in 7 patients with target lesions. The 9 patients with Peritoneal Cancer Index (PCI) scores greater than or equal to 20 had a lower survival rate than did the 23 patients with PCI scores of less than 20 (MST, 13.4 months vs. not-reached); however, the difference in survival between the 2 groups was not statistically significant (P = 0.13). Malignant ascites disappeared or decreased in 15 of 22 (68%) patients. The frequent grade 3/4 toxic effects were neutropenia (34%), leukopenia (23%), and anemia (9%). Non-hematological toxic effects were relatively mild. Gastrectomy for cytoredution was performed when peritoneal washing cytology was negative and the second look laparoscopic examination revealed peritoneal metastatic nodules disappeared or decreased in number and size in response to chemotherapy. Gastrectomy with lymph node dissection was performed in 21 patients.
Conclusion: Combination chemotherapy consisting of intravenous and intraperitoneal PTX with S-1 is well tolerated and effective in gastric cancer patients who have macroscopic peritoneal metastasis.

HER2 status of advanced or recurrent gastric cancer - Prospective cohort study
JFMC44-1101

Author(s)：: Toshihiko Tomita 1), Keisho Chin 2), Atsushi Nashimoto 3), Kazuhiro Nishikawa 4), Akira Miki 5), Kazuya Yamaguchi 10), Takaki Yoshikawa 6), Satoshi Morita 7), Atsushi Ochiai 8), Junichi Sakamoto 9), Shigetoyo Saji 9), Kazuhiro Yoshida 10)
Affiliation(s)：: Hyogo College of Medicine 1), Cancer Institute Hospital 2), Niigata Cancer Center Hospital 3), Osaka General Medical Center 4), Kobe City Medical Center General Hospital 5), Kanagawa Cancer Center 6), Yokohama City University Graduate School of Medicine 7), National Cancer Center Hospital East 8), Japanese Foundation for Multidisciplinary Treatment of Cancer 9), Gifu University Graduate School of Medicine 10)

Background: Randomized phase III trial of The Trastuzumab for Gastric Cancer (ToGA) demonstrated that trastuzumab plus chemotherapy was more efficacious than chemotherapy alone in patients with human epidermal growth factor 2 (HER2)-positive advanced/metastatic gastric or gastroesophageal junction cancer. Of patients screened for the ToGA trial, 22.1% of all gastric cancers were HER2-positive, but high positive rate was shown to be about 27% in Japanese patients. However, ToGA trial was the registration study with strict inclusion/exclusion criteria, so the confined patient background might be potential for bias in HER2 positivity rate. Therefore, to determine the true HER2 positivity rate, and to explore the factors associated with HER2-positive in Japanese patients, we planned prospective multicenter observational cohort study by consecutive investigation methods named JFMC44-1101. Methods: Tumors were diagnosed the unresectable advanced or recurrent gastric cancer after August 1, 2011, central HER2 testing was conducted to examine the distribution of HER2 status. Immunohistochemistry (IHC) and fluorescence in situ hybridization (FISH) method were performed in all patients, and we defined IHC 3+ or FISH-positive (HER2 to chromosome 17 centromere ratio of >2) as HER2-positive, according to ToGA's definition. The study endpoints were HER2 positivity rate, correlation between HER2-positive and patients characteristics or sample condition, correlation between FISH-positive and patient characteristics or sample condition in IHC 0 or 1+, and connection between overall survival and patient characteristics or treatment factor. Results: A total of 1461 patients were enrolled from 157 sites between September 2011 and June 2012. HER2 evaluation was conducted in 1427 patients excluding 14 patients of ineligible and 20 patients of FISH undetectable. The patients characteristics were as follows: male/female: 1020/407, median age (min-max): 68 (23-99), advanced/recurrence: 1051/376, biopsy/surgical specimen: 749/678. HER2 positivity rate was 21.2%, the ratio of high levels of HER2 protein (i.e., IHC 3+ or IHC 2+/FISH-positive) was 15.6%. As a result of multivariate logistic regression analysis, HER2-positive was correlated to the intestinal type according to the Lauren's classification, no peritoneal metastasis, and liver metastasis. The incidence of IHC 0 or 1+ with FISH-positive was correlated to age (>65), intestinal type and T1-T3 depth of invasion. Conclusion: HER2 positivity rate and the ratio of high levels of HER2 protein in Japanese unresectable advanced or recurrent gastric cancer was similar to ToGA trial screening population, respectively (JFMC44-1101: 21.2%, 15.6%, ToGA: 22.1%, 16.6%). What the results observed in multivariate analysis is important as the data of the treatment for HER2-positive patients with gastric cancer, and report research results for the promotion of personalized treatment in gastric cancer. We would like to await the findings of future analysis correlation between overall survival and patients characteristics, medical treatment factors.

Analysis of gene mutations in KRAS, NRAS, BRAF and PIK3CA in patients who received systemic chemotherapy with metastatic gastric cancer.

Author(s)：: Masaru Fukahori¹, Yasuhide Yamada¹, Hirokazu Taniguchi², Naoki Takahashi¹, Yusuke Sasaki¹, Hirokazu Shoji¹, Yoshitaka Honma¹, Satoru Iwasa¹, Atsuo Takashima¹, Natsuko Okita¹, Ken Kato¹, Tetsuya Hamaguchi¹, Yasuhiro Shimada¹.
Affiliation(s)：: Division of gastrointestinal oncology¹，
Division of pathology², National Cancer Center Hospital, Tokyo, Japan

Background: Previous studies showed that gene mutations of KRAS, NRAS, BRAF, and PIK3CA are associated with a poor prognosis or resistance of anti-EGFR antibody in metastatic colorectal cancer. On the other hand, the frequency and clinical significance of these gene mutations have not been clarified in metastatic gastric cancer (mGC).
Methods: We gathered formalin-fixed paraffin-embedded tumor samples from 168 patients who were diagnosed gastric cancer and underwent gastrectomy between September 1995 and March 2008. Among 168 patients, we selected the mGC patients who received systemic chemotherapy. We retrospectively evaluated the mutation status of KRAS (exon2, 3, 4), NRAS (exon 2, 3), BRAF (exon 15) and PIK3CA (exon 9, 20) and prognosis in patients who received chemotherapy by gene mutation status.
Results: A total 125 of patients were included in this analysis. Regimens of first line chemotherapy were 5-FU (24.0%), CPT-11/CDDP (31.2%), S-1 (33.6%), 5-FU/MTX (8.0%) and others (3.2%). Mutations of KRAS codon 12/13 (6.4%), PIK3CA exon9 (4%), PIK3CA exon20 (0.8%), and NRAS codon 12/13 (2.4%) were detected by direct-sequence method. Mutations of KRAS codon 61, KRAS codon 146, BRAF V600E and NRAS codon 61 were not detected. There were no significant differences in overall survival (OS) by mutation status of KRAS and PIK3CA. Patients with NRAS codon 12/13 mutation had poor prognosis (MST：15.5 vs 9.4 months, p=0.008; log-rank test). Mutation status of NRAS codon 12/13 are a poor prognostic factor by multivariate analysis (HR: 5.327, 95%CI: 1.566－18.096).
Conclusions: Mutations in KRAS codon 61, KRAS codon 146, BRAF V600E, NRAS codon 61 were not detected, and frequency of mutation in PIK3CA exon 9, PIK3CA exon 20 and NRAS codon 61 were rare in mGC. Although our study has several limitations, mutation in NRAS codon 12/13 may be a poor prognostic factor in mGC who received systemic chemotherapy and further accumulation of cases are needed.

Development and Validation of Gastric Cancer New Nomogram based on Preoperative Clinical Prognostic Factors

Author(s)：: Etsuro Bando¹, Michael W. Kattan², Makoto Ohori³, Masanori Tokunaga¹, Yutaka Tanizawa¹, Taiichi Kawamura¹, Masanori Terashima¹
Affiliation(s)：: Shizuoka Cancer Center, Division of Gastric Surgery ¹
Cleveland Clinic, Department of Quantitative Health Care Sciences²
Tokyo Medical University, Division of Urology³

Background and Objectives:
Nomogram has been developed and shown to be more accurate than the conventional staging systems for predicting prognosis in gastric cancer patients. Preoperative clinical staging is a primary measure with which therapeutic strategies for gastric cancer is determined. While many reports have been published on the relationship between nomogram based on pathologic factors and survival outcomes, there were no reports of developing nomogram based on preoperative clinical factors. The purpose of this study was to firstly build a specific nomogram based on preoperative clinical prognosticators for predicting overall survival (OS) in patients with primary gastric cancer for choosing accurate therapy preoperatively.
Methods:
We included patients with primary 2807 gastric cancer patients, who had not been undergone prior chemotherapy, between 2002 and 2010. The priori chosen pre-operative prognostic covariates were sex, age, tumor site, tumor size, histologic subtype obtained by endoscopy, endoscopic macroscopic appearance, clinical tumor depth evaluated by endoscopy or barium image, numbers of positive nodes and distant metastases on enhanced CT scan, and serum concentration of CEA(carcinoembrionic antigen)/CA19-9 (carbohydrate antigen 19-9). Nomogram validation comprised two activities. First, discrimination was quantified with the Harrell's concordance index. Similar to the area under the receiver operating characteristic (ROC) curve, but appropriate for censored data, the concordance index provides the probability that, in a randomly selected pair of patients in which one patient dies before the other, the patient who died first had the worse predicted outcome from the nomogram. We used bootstrapping to obtain a relatively unbiased estimate. Second, calibration plot was assessed. This was done by grouping patients with respect to their nomogram-predicted probabilities and then comparing the mean of the group with the observed Kaplan-Meier estimate of disease-specific survival. Again, bootstrapping correction was used for this activity.
Results:
At a median follow-up of 57 months, 612 deaths were recorded. Five-year OS and DSS rates were 64.3%. In Cox regression, depth and nodal involvement is the most powerful independent prognostic factors. And type4 is identified negative prognosticators. Sex and serum tumor markers have little effectiveness for survival outcome. The bootstrap-corrected Harrell's C statistics for OS of our nomogram were 0.852, which was higher than that of TNM seventh edition (C=0.826), that of sixth edition (C=0.822) or that of JPN 13th edition (C=0.821). Calibration curves for the nomogram in which X-axis is nomogram predicted survival probability and Y-axis is actual OS, shows the diagonal ideal line.
Conclusions:
New nomogram accurately predicts OS of Japanese gastric cancer patients pre-operatively. Prospective internal validation and worldwide external validation should be necessary for practical use of our nomogram instead of TNM classification.

Randomized phase II study of 6 vs. 12 months of adjuvant imatinib for GIST patients.

Author(s)：: Junya Fujita¹, Yukinori Kurokawa², Toshimasa Tsujinaka³, Kazuya Muguruma⁴, Takuya Nakai⁵, Arimichi Takabayashi⁶, Masako Hiramatsu⁷, Masanori Terashima⁸, Seiichi Hirota⁹, Toshirou Nishida¹⁰
Affiliation(s)：: ¹Department of Surgery, NTT West Osaka Hospital, Japan
²Department of Gastroenterological Surgery, Osaka University Graduate School of Medicine, Japan
³Department of Surgery, Kaizuka City Hospital, Japan
⁴Department of Surgical Oncology, Osaka City University Graduate School of Medicine, Japan
⁵Department of Surgery, Kinki University School of Medicine, Japan
⁶Department of Surgery, Kitano Hospital, Japan
⁷Department of Gastroenterological Surgery, Takatsuki Red Cross Hospital, Japan
⁸Division of Gastric Surgery, Shizuoka Cancer Center, Japan
⁹Department of Surgical Pathology, Hyogo College of Medicine, Japan
¹⁰National Cancer Center Hospital East, Japan

Background:
Z9001 trial demonstrated that postoperative adjuvant imatinib for 12 month improved RFS in patients with GIST ≥3cm, and SSGXVIII trial showed the adjuvant imatinib for 3 years was superior to 1 year in RFS and OS for high risk GIST patients. Postoperative imatinib has gained acceptance as a standard adjuvant therapy, however, the optimal target of patients and the optimal period of treatment is still a matter of debate. We hypothesized that the 6 months of adjuvant imatinib might not be inferior to 12 months of adjuvant imatinib in regard to RFS for intermediate risk or high risk GIST patients, we conducted a multi-institutional randomized controlled phase II trial.
Methods:
Inclusion criteria was as follows; histologically proved primary GIST with completely resection, intermediate risk or high risk according to the Fletcher's criteria, age ≥ 20 and < 80, ECOG performance status 0-1. Familial GIST or GIST with von Recklinghausen disease were excluded. Patients were randomized after resection and assigned to 6 months (6M) or 12 months (12M) treatment of imatinib 400 mg/day. The primary endpoint was RFS and the secondary endpoints were OS, the compliance and safety of adjuvant imatinib. This study was designed as a randomized screening trial to evaluate non-inferiority with margin of hazard ratio 1.67, one-sided alpha 0.2 and power 0.7.
Results: Between Dec 2007 and Aug 2011, from 37 institutions 92 patients were accumulated and randomly assigned to the 6M group (n=45) or the 12M group (n=47), those were well balanced for baseline characteristics. One patient was ineligible due to non-GIST (desmoid) tumor at a pathological central review. The rates of completion of the assigned period of imatinib treatments were 80% in the 6M group and 70% in the 12M group. Adjuvant imatinib was well tolerated, with one patient of Gr. 4 rash and no treatment-related death. The incidences of the adverse events of imatinib were not significantly different between the two groups. The first interim analysis was conducted at Sep 2012 with the median follow-up time of 33 months. The 1- and 2-year RFS were 82% and 65% in the 6M group and 96% and 86% in the 12M group, respectively. Hazard ratio of recurrence (HR) was 1.81 (95%CI: 0.84-3.91), and the 2-sided log-rank p value was 0.12. Because of the lower efficacy of the 6M group than expected, the Data and Safety Monitoring Committee recommended the early release of first interim analysis results. The subgroup analysis regarding to the risk classification, HR between 6M and 12M was 1.15 (95%CI: 0.072-18.2, P=0.92) in the intermediate risk patients (n=35), and 2.15 (95%CI: 0.96-4.81, P=0.056) in the high risk patients (n=53).
Conclusions:
Adjuvant imatinib for 6 months was inferior to 12 months in RFS. Shortening of the adjuvant imatinib duration should not be recommended for high risk GIST patients. As for intermediate risk patients, there is room for discussion of the application of adjuvant therapy and/or the shortening the period of the treatment.

Phase III of regorafenib in pretreated advanced GIST: Japanese sub-analysis

Author(s)：: Toshihiko Doi¹, Yoshito Komatsu², Akira Sawaki³, Tatsuo Kanda⁴, Yasuhide Yamada⁵, Christian Kappeler⁶, Iris Kuss⁶, George D Demetri⁷, Toshirou Nishida⁸
Affiliation(s)：: 1)National Cancer Center Hospital East
2)Hokkaido University Hospital
3)Japan Red Cross Nagoya Daini Hospital
4)Niigata University
5)National Cancer Center Hospital
6)Bayer HealthCare
7)Ludwig Center at Dana-Farber Cancer Institute and Harvard Medical School
8)Osaka Police Hospital

The GRID trial is the international Phase III of regorafenib (REG) compared to placebo (PL) in patients with metastatic gastrointestinal tumor (GIST) following failure of at least imatinib (IM) and sunitinib (SU). The trial met primary endpoint which showed statistically significant longer progression free survival (PFS) in REG compared with PL. We report here the efficacy and safety of REG in Japanese sub-population and additionally summarize overall and Asian population data.
Patients with metastatic and/or unresectable GIST, objective failure of both prior IM and SU were randomized 2:1 to receive best supportive care (BSC) with either REG 160 mg po once daily (3 weeks on/1 week off) or placebo. Progression disease was assessed by centrally, treatment assignment could be unblinded. Patients in PL arm were allowed to crossover to receive open-label REG, and patients in REG arm continued to receive open-label REG both at the discretion of the investigator. Patients visited hospital every 2 weeks, to undergo laboratory test and be checked their health condition, adverse event and drug compliance. Tumors were assessed every 4 weeks for the first 3 months of the trial, every 6 weeks from months 3 to 6, and then every 8 weeks until the end of treatment.
The primary endpoint was PFS by central assessment. Secondary endpoints included overall survival, tumor response and safety.
Seventeen Japanese metastatic GIST patients were randomized to receive regorafenib (REG, n= 12) or placebo (PL, n= 5). Six patients in the REG arm and 4 patients in the PL arm received open-label regorafenib following disease progression. Demographics were generally consistent with the overall study population. In particular, more patients had 2 prior treatment lines in Japanese (75%) and Asan population (69%) compared with overall population (56%). PFS in the Japanese population was significantly longer in REG arm compared with PL arm (HR 0.08, 95% CI, 0.02~0.45; 1-sided p=0.000164), as well as Asian (Hazard ratio 0.30, 95% CI, 0.15–0.62; p=0.0003) and overall population (HR 0.27, 95% CI, 0.19~0.39; 1-sided p<0.000001). The Japanese, Asian and overall population showed similar disease control rate (REG 59%, PL 20 %, p=0.081; REG 41%, PL 7%, p<0.01; REG 53%, PL 9%, p<0.000001, respectively). In Japanese population, commonly observed drug-related adverse events (AEs) in REG arm were hand-foot skin reaction (HFSR) 92%, mucositis oral 58%, hypertension, hoarseness, alopecia, rash maculo-papular, proteinuria and diarrhea 50%. There was no unexpected AE in Japanese population. The subset analysis of Japanese showed higher incidence of AEs compared with overall population. HFSR, alopecia, hoarseness, rash maculo-papular, hypothyroidism, AST increased and proteinurea were occurred with 20% or higher in Japanese compared to that in overall population. Dose modification due to regorafenib related AEs were observed in higher in Japanese population (Japanese: 92%, Asian: 72%, overall: 63%). The incidence of regorafenib associated discontinuation did not show differences (J: 8%, Asian: 3%, overall: 2%).
One Japanese patient developed fatal liver dysfunction. The patient develop symptom acute hepatic failure 6 weeks after the start of regorafenib ; the patient died 2 weeks after entering the hospital
In conclusion, the subset analysis of the Japanese population was considered similar efficacy and acceptable AE profile as well as the results of GRID trial, even in limited number of Japanese population. AEs were manageable by dose modification. Frequent liver function test is recommended, at least first 2 cycles.

Return to Top

Category