Background: The combination of paclitaxel (PTX) plus ramucirumab (RAM) has been the standard regimen for advanced gastric cancer (GC) or gastro-esophageal junction cancer (GJC) patients (pts) who were treated with fluoropyrimidine-containing regimen. In a phase 2 trial, it was showen that nab-PTX plus RAM has promising activity (objective response rate [ORR], 55 %; median progression free survival [PFS], 7.6 months) [Bando H, et al. Eur J Cancer 2018] and is, therefore, a treatment option. However, high incidence of myelosuppression led to discontinuation of the nab-PTX. Therefore, we conducted this study to investigate whether schedule modification to biweekly nab-PTX plus RAM therapy is safe and effective. Methods: Pts with GC or GJC treated with fluoropyrimidine-containing regimen were enrolled and received nab-PTX (100 mg/m²) on days 1, 8, and 15 and RAM (8 mg/kg) on days 1 and 15 of a 28-day cycle. The treatment schedule of pts with severe myelosuppression during the first cycle was modified to biweekly therapy from the second cycle. The primary endpoint was PFS in modified pts. Secondary endpoints included ORR, disease control rate (DCR), overall survival (OS), and safety. The exploratory analysis evaluated PFS and OS in pts who continued standard-schedule therapy. Based on the hypothesis that biweekly nab-PTX plus RAM therapy would improve PFS from 4.5 months to 7.0 months, 40 pts were required for a power of 0.8 with a one-sided α of 0.05. Results: Of the 81 pts who were enrolled, 76 pts received standard nab-PTX plus RAM in the first cycle and 47 pts (58%) were modified to the bi-weekly therapy. Pts characteristics in the modified pts were as follows: median age, 70 years; male, 72%; ECOG PS 0, 81%; and poorly differentiated adenocarcinoma or signet ring cell carcinoma, 45%. The median relative dose intensities of the entire period and after modification were 66% and 60%, respectively. In the modified pts, median PFS was 4.7 months (95% Confidence Interval [CI]: 3.7–5.6), median OS was 13.9 months (95% CI: 9.1–16.9), ORR was 25% (95% CI: 11–39), and DCR was 75% (95% CI: 61–89). Tumor shrinkage was observed in 61% of the pts (n = 36) with measurable lesion. In pts who could continue standard-schedule therapy, median PFS was 2.7 months (95% CI: 1.8–4.0) and median OS was 8.0 months (95% CI: 5.6–14.7). Severe (Grade >3) adverse events after modification were neutropenia (55%), anemia (13%), hypertension (17%), and peripheral sensory neuropathy (13%). Conclusions: Our study could not meet the primary endpoint for PFS. However, PFS in pts with standard-schedule therapy was numerically lower than that in the modified pts. The modification to biweekly schedule might be considered to be a treatment option for GC/GJC pts with severe myelosuppression in second-line nab-PTX plus RAM therapy. Clinical trial information: jRCTs031180061.