Adjuvant S-1 Plus Docetaxel Improves Survival in Stage III Gastric Cancer

Kazuhiro Yoshida, MD, PhD

In the phase 2 JACCRO GC-07 trial, patients with stage III gastric cancer who received the combination of S-1 and docetaxel in the adjuvant setting had a 29% reduction in the risk of relapse, according to updated study results presented during the 2021 Gastrointestinal Cancers Symposium.¹

At a median follow-up of 42.5 months (range, 0.3-85.16), the 3-year relapse-free survival (RFS) rate was 67.74% (95% CI, 63.19-71.84) in the combination arm versus 57.40% in the S-1–alone arm in the intent-to-treat (ITT) population (HR, 0.715; 95% CI, 0.587-0.871; P = .0008), meeting the primary end point of the study.

“The 3-year relapse-free survival and overall survival in the S-1 plus docetaxel group were significantly superior to those in the S-1 group,” lead study author Kazuhiro Yoshida, MD, PhD, FACS, who is director of Gifu University Hospital, and professor and chairman, Department of Surgical Oncology at Gifu University, Graduate School of Medicine, in Gifu, Japan, said in a virtual presentation during the meeting. “Adjuvant S-1 plus docetaxel is recommended for patients with stage III gastric cancer who underwent D2 gastrectomy without neoadjuvant chemotherapy.”

In Asia, S-1 is a standard adjuvant chemotherapy regimen for patients with stage II/III gastric cancer; however, outcomes of those with stage III disease are suboptimal.

The randomized, controlled JACCRO GC-07 trial evaluated adjuvant treatment with S-1/docetaxel compared with S-1 alone after D2 gastrectomy for patients with stage III gastric cancer. Patients received either S-1 alone (n = 459) at 80 mg/m2 on days 1 to 28 in every-6-week cycles, or plus docetaxel (n = 453) at 40 mg/m² on day 1 and S-1 at 80 mg/m² on days 1 to 14 every 3 weeks in cycles 2 to 7 followed by S-1 at 80 mg/m² on days 1 to 28 in cycles 8 and higher, which was every 6 weeks. Treatment was continued in both arms up to 1 year following surgery. Stratification factors included stage (IIIA/IIIB/IIIC), histological type (differentiated/undifferentiated), and institution.

The primary end point was 3-year RFS rate; secondary end points included 3- and 5-year overall survival (OS) rates, 5-year RFS rates, time to treatment failure, and safety.

Patients on the study were aged 20 years to 80 years, had histologically proven gastric cancer that was stage IIIA, stage IIIB, or stage IIIC, had R0 resection with D2 lymph-node dissection, was negative for peritoneal washing cytology, did not receive prior treatment except for initial gastric resection of the primary lesion, had an ECOG performance status of 0 or 1, and were able to start chemotherapy within 42 days of gastrectomy. Patients also needed to have adequate major organ function.

Across both arms, most patients were male (70%), and the median age was 66 years; about one-third of patients in each arm had stage IIIA (32%) stage IIIB (35%), or stage IIIC disease (33%). Five percent of patients had T2 disease, followed by 37.5% and 57.5% with T3 and T4 disease, respectively. One percent of patients in each arm had N0, 10.5% had N1, and 32.5% had N2 disease. More than half (59.5%) of patients had undifferentiated histology, 86.5% had and ECOG performance status of 0, and 59% had a distal gastrectomy. The tumor location varied among patients overall: upper (23.5%), middle (36.5%), lower (37%), and other (3%).

The relative dose intensity was 71.4% in the S-1 alone; in the combination arm, these rates were 61.6% for S-1 and 77.5% for docetaxel.

Prior findings of the trial, which were from a second interim analysis, showed that there was a significant improvement in RFS with the addition of docetaxel, and a recommendation by a Data and Safety Monitoring Committee determined that the study be terminated in September 2017.2

The earlier results showed that the combination led to an estimated 36% reduction in the risk of relapse or death (HR, 0.632; 99.99% CI, 0.400-0.998; P = .0007). The median 3-year RFS rate was 65.9% with the combination versus 49.5% with S-1 alone.

When stratified by disease stage, the RFS benefits were observed in patients with stage IIIA (HR, 0.524; 95% CI, 0.285-0.966; P = .0351), stage IIIB (HR, 0.614; 95% CI, 0.382-0.989; P = .0427) and stage IIIC disease (HR, 0.693; 95% CI, 0.466-1.03; P = .0677). At the time, OS data was immature (P = .1268). The preplanned analysis, set to take place at 3 years following completion of study enrollment, was performed with updated information of the patients.

The updated data also included all stage III disease subsets. In patients with stage IIIA disease, the 3-year RFS rates were 81.46% and 71.25% with S-1/docetaxel and S-1 alone, respectively (HR; 0.617; 95% CI, 0.403-0.944; P = .025). In stage IIIB disease, these rates were 66.14% and 61.61% (HR, 0.881; 95% CI, 0.629-0.234; P = .46); in stage IIIC disease, the 3-year RFS rates were 55.85% with S-1/docetaxel and 39.25% with S-1 alone (HR, 0.640; 95% CI, 0.475-0.863; P = .0032).

The RFS benefit was also observed in the S-1/docetaxel group across most prespecified subgroups, except for N0 stage (HR, 1.291) and N1 stage (HR, 1.154), and those with “other” listed as their extent of gastrectomy (HR, 1.964).

RFS was evaluated by pT and pN stages. In the T2 population (n = 44), the 3-year RFS rates were 100% with S-1/docetaxel and 64.96% with S-1 alone (P = .0079); in the T3 group (n = 342), these rates were 67.62% and 60.88%. respectively (P = .19). In the T4 population (n = 526), the 3-year RFS rates were 65.61% and 54.32% with S-1/docetaxel and S-1 alone, respectively (P = .0062).

For pN stage, however, the 3-year RFS rates were 50.00% and 66.67% with the combination and S-1 alone, respectively, in the N0 population (n = 9; P = .82). In the N1 group (n = 97), these rates were 80.00% and 85.26%, respectively (P = .74); the rates were 77.24% and 67.63% in the N2 subgroup (n = 296). In those with N3 disease (n = 510), the 3-year RFS rates were 60.09% with S-1/docetaxel and 46.82% with S-1 alone.

At a median follow-up of 48.25 months (range, 3.25-85.16) and a data cutoff date of September 9, 2020, the 3-year OS rate was 77.68% with S-1/docetaxel compared with 71.17% with S-1 alone (HR, 0.742; 95% CI, 0.596-0.925; P = .0076).

OS was analyzed across each subset of stage III disease as well. In the stage IIIA subset, the 3-year OS rates were 90.33% and 78.65% with the S-1/docetaxel arm and S-1–alone arm, respectively (HR, 0.523; 95% CI, 0.313-0.873; P = .012). The benefit was least prominent in the stage IIIB subset at 73.09% and 77.23%, respectively (HR, 0.988; 95% CI, 0.681-0.434; P = .095). In the stage IIIC population, the 3-year OS rates were 70.07% with the combination and 57.24% with S-1 monotherapy (HR, 0.682; 95% CI, 0.493-0.943; P = .020).

Similarly, patients with N0 stage disease did not have an OS benefit from S-1/docetaxel (HR, 1.141), and neither did those with “other” extent of gastrectomy (HR, 3.311; 95% CI, 0.299-36.696).

The sites of first relapse varied between arms: local (4.4% with S-1 vs 2.9% with S-1/docetaxel), lymph nodes (15.0% vs 6.4%, respectively), peritoneum (21.4% vs 18.8%), and hematogenous (15.5% vs 9.7%) said Yoshida, noting that some patients relapsed at more than 1 site.

Regarding safety, grade 3/4 adverse events (AEs) occurred in both arms, the most common being neutropenia (39.2% with the combination vs 16.4% with S-1 alone), leukopenia (22.4% vs 2.7%, respectively), febrile neutropenia (5.7% vs 0.4%), and anorexia (13.6% vs 12.0%). There was 1 case of TRD due to respiratory failure in the S-1–alone arm.

When measuring time to treatment failure, Yoshida said treatment compliance was consistent between the 2 groups (P = .86).

_References:

_{1. Yoshida K, Kodera Y, Kochi M, et al. Confirmed 3-year RFS and OS of the randomized trial of adjuvant S-1 versus S-1 plus docetaxel after curative resection of pStage III gastric cancer (JACCRO GC-07). J Clin Oncol. 2021;39(suppl; abstr 159).}

_{2. Yoshida K. Addition of docetaxel to oral fluoropyrimidine improves efficacy in patients with stage III gastric cancer: interim analysis of JACCRO GC-07, a randomized controlled trial. J Clin Oncol.2019;37(15):1296-1304. doi:10.1200/JCO.18.01138}