The eligibility criteria for inclusion in the present study were as follows: Age ≥20 years; histologically confirmed human epidermal growth factor receptor type 2-negative unresectable or recurrent gastric cancer; Eastern Cooperative Oncology Group (ECOG) performance status <2; estimated life expectancy ≥3 months; and adequate organ function, as defined by hemoglobin (Hb) ≥8 g/dl, absolute neutrophil count (ANC) ≥1,500/mm³, platelet count ≥100,000/mm³, total bilirubin ≤1.5 mg/dl, serum transaminase level ≤100 U/l and creatinine clearance ≥40 ml/min. The exclusion criteria were as follows: Contraindication to either drug included in the chemotherapy regimen; evidence of prior history of platinum administration; insufficient oral intake; synchronous or previous malignancy other than carcinoma in situ; or severe comorbidities.

This trial was conducted in accordance with the Helsinki Declaration and Good Clinical Practice guidelines, and was approved by the Institutional Review Board of Kobe City Medical Center General Hospital (Kobe, Japan). All the patients were required to provide written informed consent prior to enrolment.

Protocol treatment was defined as chemotherapy consisting of capecitabine and oxaliplatin. The patients received capecitabine 1,000 mg/m² b.i.d. on days 1–14 plus oxaliplatin 130 mg/m² every 3 weeks. The study was designed to determine the recommended dose (RD) of chemotherapy. A total of 6 patients were treated at dose level 1 (capecitabine 1,000 mg/m² b.i.d. on days 1–14 and oxaliplatin 130 mg/m² on day 1). If ≥3 of the 6 patients experienced a dose-limiting toxicity (DLT), an additional 6 patients were accrued at the next lower dose level (level 0; capecitabine 1,000 mg/m² twice daily on days 1–14 and oxaliplatin 100 mg/m² on day 1). The MTD was defined as the dose at which ≥3 of the 6 patients experienced a DLT. Treatment was repeated until disease progression, unacceptable toxicity, or withdrawal of consent. Treatment was delayed if, on the planned day of treatment, the laboratory results included any of the following: ANC <1,500/mm³, platelet count <75,000/mm³, Hb <8 g/dl, serum transaminase >100 U/l, total bilirubin >2.0 mg/dl, or serum creatinine >1.50 mg/dl, or if symptomatic toxicity was present. Patients who could not tolerate oxaliplatin continued to receive capecitabine monotherapy until disease progression or intolerable toxicity. The RD was defined as one dose level below the MTD. If the MTD was not achieved, even at level 1, it was considered as the RD. DLT was defined as any of the following adverse events occurring in the first cycle: i) Grade 4 neutropenia lasting >4 days; ii) grade 4 thrombocytopenia (<25,000/mm³); iii) febrile neutropenia; iv) grade 3 or 4 non-hematological toxicities; v) treatment discontinuation due to adverse events; or vi) treatment-related death. Protocol treatment was administered triweekly until disease progression, unacceptable toxicity, or withdrawal of consent. In patients with pharyngolaryngeal dysesthesia, the duration of oxaliplatin infusion was prolonged from 2–6 h. In the event of grade 4 non-hematological toxicities, treatment was definitively interrupted.

Pretreatment evaluation included a medical history, physical examination, complete blood cell count and serum chemistry tests, esophagogastroduodenoscopy, and chest, abdominal and pelvic computed tomography (CT) scans. Clinical examination and biochemical tests were required before and during each treatment cycle. All images for tumor responses were evaluated according to the Response Evaluation Criteria In Solid Tumors (RECIST), version 1.1 (7). All adverse events during chemotherapy were evaluated using the National Cancer Institute Common Terminology Criteria for Adverse Events, version 4.0 (https://www.eortc.be/services/doc/ctc/CTCAE_4.03_2010-06-14_QuickReference_5x7.pdf).

The primary endpoint in the present study was the MTD and RD of the G-XELOX regimen. Secondary endpoints included toxicities, response rate (RR), progression-free survival (PFS) and overall survival (OS). Safety and efficacy analyses were both conducted in an intention-to-treat population, defined as all patients enrolled in the study who received at least one dose of chemotherapy. All statistical analyses were conducted using the SPSS software package (SPSS 22.0 Inc., IBM Corp., Armonk, NY, USA).

This trial was registered with the University Hospital Medical Information Network (UMIN no 000015951).

Between January and July 2015, 6 patients were enrolled. The characteristics of the enrolled patients are listed in Table I. The median age was 72 years, 67% of the patients had diffuse-type disease, 50% had multiple organ metastases, and all patients were chemo-naïve. One patient had undergone distal gastrectomy for resection of the primary tumor.

A total of 6 patients were enrolled at dose level 1 (capecitabine 1,000 mg/m² b.i.d. on days 1–14 and oxaliplatin 130 mg/m² on day 1). Of the 6 patients administered level 1, 1 patient developed a DLT (grade 3 stomatitis) and, hence, the RD for phase II studies was determined to be capecitabine 1,000 mg/m² b.i.d. on days 1–14 and oxaliplatin 130 mg/m² on day 1.

Toxicity was assessable in all 6 patients. The most severe toxicities throughout the protocol treatment period are listed in Table II. Grade ≥3 thrombocytopenia and febrile neutropenia occurred in 17 and 0% of the patients, respectively. Grade ≥3 non-hematological toxicity (stomatitis) only occurred in 1 patient (17%), but it subsided with conservative treatment. Peripheral neuropathy was observed in all patients, but without functional disorders. The median percentage of relative dose intensity delivered during protocol treatment was 71.3% (range, 12.4–100%) for capecitabine and 92.1% (range, 68.3–100%) for oxaliplatin.

Response was assessable in 5 patients. Of the 6 patients, 3 had measurable lesions according to RECIST; of those 3 patients, 2 had a partial response and 1 had progressive disease, with an RR of 67% and a disease control rate (DCR) of 67%. The median time to the first dose reduction was 2 cycles (range, 2–3 cycles) in 4 of the 6 patients, commonly due to gastrointestinal toxicities or myelosuppression. Of the 6 patients, 4 discontinued the protocol treatment due to disease progression and 1 due to toxicities. One patient underwent curative resection for primary disease after 11 cycles of protocol treatment and remained alive without disease at the time of writing, >2 years after the initiation of the protocol treatment (last follow-up, April 2017). The median PFS and OS in all the patients were 3.6 and 5.7 months, respectively.