OS Benefit From TAS-102 in Gastric Cancer Sustained in Subgroup Analysis


Treatment with TAS-102 led to an improvement in overall survival among patients with metastatic gastric/gastroesophageal junction cancer. The improvement was also seen both in patients who had and had not undergone prior gastrectomy, according to a subgroup analysis from the phase III TAGS study. 

David H. Ilson, MD

Treatment with TAS-102 (trifluridine/tipiracil; FTD/TPI; Lonsurf) led to an improvement in overall survival (OS) among patients with metastatic gastric/gastroesophageal junction (GEJ) cancer. The improvement was also seen both in patients who had and had not undergone prior gastrectomy, according to a subgroup analysis from the phase III TAGS study. 

The analysis showed that patients with prior gastrectomy derived a survival benefit from TAS-102 that was similar to the benefit observed in the overall patient population, David H. Ilson, MD, of Memorial Sloan Kettering Cancer Center, reported at the 2019 Gastrointestinal Cancers Symposium.1

“Hematologic adverse events, such as neutropenia and leukopenia, may have been somewhat more frequent among TAS-102—treated patients with gastrectomy than in the overall population, but this did not result in more treatment discontinuations,” said Ilson.

“Exposure to TAS-102 was similar between patients with gastrectomy and those in the overall population. These results suggest that TAS-102 is an effective treatment option with a manageable safety profile for patients with metastatic gastric cancer, regardless of prior gastrectomy.”

Surgery remains the only treatment option that has curative potential for early-stage gastric cancer. However, as many as half of patients have recurrent disease, and about 40% of those patients have undergone gastrectomy, Ilson noted.

A phase II trial conducted in Japan demonstrated a median OS of 8.7 months in patients who received TAS-102 after failure of standard chemotherapy. Investigation of the combination therapy continued in the phase III, randomized, placebo-controlled, international TAGS trial.

Investigators in 17 countries enrolled patients with unresectable, previously treated metastatic gastric/GEJ cancer. Eligibility criteria included evidence of radiologic progression on 2 or more prior chemotherapy regimens. Patients received either TAS-102 or placebo (2:1 ratio), along with best supportive care. The primary endpoint was OS.

Data analysis included 507 patients, including 221 who had prior gastrectomy. A previous study showed no differences in the pharmacokinetics of FTD or TPI between patients with or without gastrectomy.2The TAGS design included a prespecified subgroup analysis of the primary endpoint in patients with and without gastrectomy.

The primary analysis produced a statistically significant 31% reduction in the survival hazard ratio (95% CI, 0.56-0.85;P= .0003, one sided;P= .0006, 2-sided). The absolute difference in PFS translated into a 43% reduction in the hazard for disease progression or death in favor of the TAS-102 arm (95% CI, 0.47-0.70;&nbsp;P<.0001, 2-sided).

Patients treated with TAS-102 had a median OS of 5.7 months, as compared with 3.6 months for placebo-treated patients. Median progression-free survival also improved slightly from 1.8 to 2.0 months with TAS-102, also a statistically significant difference.

Subgroup analysis showed a consistent survival benefit for patients in the TAS-102 arm, said Ilson. The comparison of patients with or without prior gastrectomy showed that baseline characteristics of the gastrectomy subgroup did not differ substantively from those of the overall study population.

Treatment with TAS-102 was associated with a median OS of 6.0 months in patients with prior gastrectomy, as compared with 3.4 months in the placebo-treated patients with gastrectomy. The difference represented a 43% reduction in the survival hazard (95% CI, 0.41-0.79). The PFS analysis yielded a median value of 2.2 months in the TAS-102 group and 1.8 months in the placebo group, a 52% reduction in the risk of disease progression or death (HR, 0.48; 95% CI, 0.35-0.65).

&ldquo;Prior gastrectomy was not identified as a prognostic or predictive factor in multivariate Cox regression analyses in which all prespecified factors were included,&rdquo; said Ilson. &ldquo;Treatment effect size remained the same after adjusting for all identified potential prognostic factors.&rdquo;

Treatment exposure in the gastrectomy subgroup also was similar to that of the overall study population. Mean weekly dose with TAS-102 in the gastrectomy group was 145 mg/m2versus 159 mg/m2in placebo-treated patients. Respective values in the overall population were 148 and 155 mg/m2. Mean relative dose intensities were 0.83 versus 0.91 in the gastrectomy group and 0.85 versus 0.89 in the overall population. Median number of treatment cycles and mean treatment duration also were similar in patients with gastrectomy and the overall population.

In the overall population and the gastrectomy subgroup, adverse events occurred more often with TAS-102. No grade &ge;3 adverse events occurred in more than 9% of the placebo group.

Across the overall population, rates of hematologic adverse events (all grades) in the TAS-102 and placebo arms were neutropenia/decreased neutrophil count (53% vs 4%), anemia/decreased hemoglobin (45% vs 19%), leukopenia/decreased white count (23% vs 2%), and thrombocytopenia/decreased platelet count (18% vs 5%). The most common grade &ge;3 hematologic events with TAS-102 were neutropenia (34%) and anemia (19%).

Rates of grade &ge;3 gastrointestinal adverse events in the TAS-102 and placebo arms included nausea (3% vs 3%), diarrhea (3% vs 2%), vomiting (4% vs 2%), abdominal pain (4% vs 9%), and constipation (1% vs 2%).

Other reported grade &ge;3 adverse events included decreased appetite (9% vs 7%), fatigue (7% vs 6%), asthenia (5% vs 7%), and back pain (1% vs 2%).

Types and rates of adverse events were similar in the subgroup of patients with prior gastrectomy.


  1. Ilson DH, Prokharau A, Arkenau H-T, et al. Efficacy and safety of trifluridine/tipiracil (FTD/TPI) in patients (pts) with metastatic gastric cancer (mGC) with or without prior gastrectomy: Results from a phase III study (TAGS).J Clin Oncol.2019;37(suppl 4; abstr 3).
  2. Bando H, Doi T, Muro K, et al. A multicenter phase II study of TAS-102 monotherapy in patients with pre-treated advanced gastric cancer (EPOC1201).Eur J Cancer.2016;62:46-53. doi: 10.1016/j.ejca.2016.04.009.
Related Videos
Related Content