In 2018, more than one million new cases of GC were diagnosed, and more than 782,000 deaths were recorded worldwide [1]. The highest mortality rates from GC have been reported in East Asia, including Japan, Korea, and China [2], and the lowest rates have been reported in North America. The mortality rate is mainly dependent on the rate of Helicobacter pylori (HP) infection, as HP infection is a known dominant cause of GC [3]. Owing to the improved treatment of HP, the rate of GC has decreased ^[4][5][4,5]. However, the proportion of proximal GCs, which often lead to a poor prognosis, is increasing [6]. The 5-year survival rate for GC is 60% or higher in Japan [7] and Korea [8], where more than half of GC patients are diagnosed at an early stage through well-organized population screening programs ^[9][10][9,10]. In contrast, GC is often diagnosed at an advanced stage in most Western patients. Notably, the biological nature of GC and differences in therapeutic quality between Eastern and Western countries may affect treatment outcomes ^[11][12][13][11,12,13].

In terms of the management of localized operable disease, marked disparities between the East and West regarding surgical procedures [14] and their outcomes result in sizeable geographical variation in the preferred adjuvant treatment for localized GC [15]. For example, adjuvant chemoradiotherapy (CRT) is commonly used following D0/1 surgery for patients with GC in the US [16], and an approach with intensive combination chemotherapy administered before and after surgery (perioperative) is preferable in the UK [17]. In East Asia, adjuvant oral fluoropyrimidine-based chemotherapy following D2 resection is considered the gold standard [18].

The Asian, US, and European guidelines recommend a regimen of oral or intravenous injection of fluoropyrimidine combined with platinum as the palliative first-line chemotherapy for HER2-negative GC ^[19][20][21][27,28,29]. The efficacy of cisplatin + S-1 (SP) or capecitabine (XP) doublet combination therapy has been demonstrated by the results of several phase III trials, including the SPIRITS study and JCOG 9912 study ^[22][23][24][21,34,35]. Among the platinum agents, cisplatin has been mainly used; however, its strong gastrointestinal toxicity, nephrotoxicity, and other side effects have become problematic; thus, other platinum agents have been investigated. In particular, SOX (S-1 + oxaliplatin) and XELOX (oxaliplatin + capecitabine) regimens containing oxaliplatin are recommended as they are easier to administer than SP and XP and do not require hydration. This recommendation is based on evidence obtained from the REAL-2 trial, which revealed that capecitabine and oxaliplatin are as effective as fluorouracil and cisplatin ^[25][36]. In fact, the G-SOX study demonstrated that SOX is as effective as FP and has a favorable safety profile ^[26][37].

Interestingly, in the GO2 phase III trial, the XELOX regimen with a 60% dose reduction from the standard dose was demonstrated to be less toxic and non-inferior in terms of progression-free survival (PFS) in elderly and frail patients ^[27][38]. In addition, the combination of 5-FU/levofolinate (LV) and oxaliplatin (FOLFOX) has demonstrated efficacy and is particularly useful in patients with intestinal obstruction or other difficulties with oral intake ^[28][29][39,40]. S-1 plus docetaxel is conditionally recommended in Japan, as the START trial highlighted its OS superiority over S-1 monotherapy in patients who cannot receive platinum-containing regimens ^[30][41].

There are divided views on the three-drug combination therapy. A phase III V325 trial in the U.S. and Europe revealed the superiority of docetaxel/cisplatin/5-fluorouracil (DCF) triple therapy, in which docetaxel was added to 5FU plus cisplatin infusion ^[31][42]. However, this three-drug combination was not recommended in Asian guidelines owing to its high toxicity, which was not balanced with efficacy and was only recommended for a small number of tolerable patients. Recently, in Japan, the triple combination of docetaxel, cisplatin, and S-1 (DCS) was compared to SP in a phase III trial (JCOG1013) following the results of a promising phase II trial ^[32][33][43,44]; however, no OS benefit was found ^[34][45]. As a result, triple-drug combination therapy, including taxanes, is not currently recommended as a first-line therapy in Japan. There are many criticisms that the study did not employ an appropriate dosage regimen and ultimately did not achieve satisfactory results. Several studies have reported the usefulness of triple-drug combination therapy in cases where conversion therapy is intended ^[35][36][46,47].

In recent years, ICIs have been introduced as a new standard of care for several malignancies, including advanced GC, and have demonstrated good clinical efficacy ^[37][48]. To date, the KEYNOTE-062, ATTRACTION-4, and CheckMate 649 trials serve as the major trials that used ICI as the primary treatment for GC. In the KEYNOTE-062 trial, 763 patients (69% with GC) were randomized to receive pembrolizumab alone, pembrolizumab plus chemotherapy (cisplatin/5-FU or capecitabine), and chemotherapy plus placebo as first-line treatment for HER2-negative and PD-Ll-positive (combined positive score (CPS) ≥ 1) unresectable advanced or recurrent GC and esophagogastric junction cancer ^[38][49]. In patients with PD-L1 CPS ≥ 1, pembrolizumab monotherapy was non-inferior to chemotherapy (median OS 10.6 vs. 11.1 months (HR, 0.91; 99.2% CI, 0.69–1.18; noninferiority margin, 1.2)) but prolonged OS vs. chemotherapy in patients with CPS ≥ 10 (median OS 17.4 vs. 10.8 months, HR 0.69 (95% CI 0.49–0.97)). In particular, in the group of patients with PD-L1 CPS ≥ 1 and MSI-high tumors, the survival benefit was enhanced with pembrolizumab (HR, 0.29; 95% CI, 0.11–0.81) and pembrolizumab plus chemotherapy (HR, 0.37; 95% CI, 0.14–0.97) vs. chemotherapy, demonstrating the benefit of pembrolizumab in MSI-H tumors ^[38][49].

ATTRACTION-4 is a phase II/III trial conducted in Japan, Korea, and Taiwan ^[39][30]. In the open-label section, the superiority of nivolumab plus chemotherapy (SOX or oxaliplatin + capecitabine [CapeOX]) over chemotherapy as a first-line treatment for unresectable advanced or recurrent GC and esophagogastric junction cancer was verified. However, no significant difference was found for the other primary endpoint, OS (median OS: 17.45 months vs. 17.15 months, HR: 0.90, 95% CI: 0.75–1.08, p = 0.257), which is presumably because many patients received subsequent treatments or additional immunotherapy. The CheckMate 649 trial, a large, randomized, phase III study of 1,581 patients (24% Asian, 76% non-Asian, 100% adenocarcinoma) was conducted worldwide to analyze the superiority of nivolumab plus chemotherapy over chemotherapy or nivolumab plus ipilimumab ^[40][41][31,32]. Sixty percent (n = 955) of these patients had a PD-L1 CPS score ≥ 5. The primary endpoints of OS (median OS 14.4 vs. 11.1 months (HR 0.71 (98.4% CI (0.59–0.86)), p < 0.0001) and PFS (median PFS: 7.7 months vs. 6.0 months, HR: 0.68, 98% CI: 0.56–0.85, p < 0.0001) revealed the superiority of nivolumab plus chemotherapy. In addition, for OS, a statistically significant difference was found between CPS ≥ 1 patients (median OS: 14.0 months vs. 11.3 months, HR: 0.77, 99.3% CI: 0.64–0.92, p = 0.0001) and the overall population (median OS: 13.8 months vs. 11.6 months, HR: 0.80,99.3% CI: 0·68–0·94, p = 0·0002).

The PFS results also highlighted the superiority of nivolumab plus chemotherapy for CPS ≥ 1 cases and the overall population. The overall response rate (ORR) of patients with a CPS ≥ 5 was 60% vs. 45% (p < 0.0001), with a significantly higher value obtained in the nivolumab plus chemotherapy arm. Furthermore, an improved ORR was achieved in the nivolumab plus chemotherapy arm across all CPS subgroups, with patients with PD-L1 CPS ≥ 5 and MSI-H achieving a benefit, especially when administered the combination with immunotherapy. The combination of nivolumab and ipilimumab without chemotherapy had no clear OS benefit compared to chemotherapy alone. Based on these results, nivolumab plus chemotherapy in patients with advanced gastric, gastroesophageal junction, and esophageal adenocarcinoma, regardless of PD-L1 CPS status, was approved in the US, Taiwan, and Japan. In Europe, nivolumab plus chemotherapy has been approved for patients with a PD-L1 CPS of ≥ 5.

HER2, also called ERBB2, is a receptor tyrosine protein kinase that is involved in cell proliferation through signaling pathways, such as RAS/RAF/MEK/ERK, PI3K/AKT/mTOR pathways, etc. ^[42][50]. HER2-positive tumors show amplification of the HER2 gene, which is commonly associated with protein overexpression, leading to tumorigenesis in GC ^[43][51]. Although only 15–20% of GC cases are HER2-positive ^[44][52], the clinical efficacy of trastuzumab, a HER2-targeted therapy, has been established for HER2-positive advanced GC. Trastuzumab is a monoclonal antibody that binds to the extracellular domain of the HER2 receptor and inhibits the HER2 signaling pathway. The multicenter phase III ToGA trial revealed that trastuzumab plus cisplatin and fluoropyrimidine significantly improved the survival of patients with HER2-positive advanced gastric and gastroesophageal junction adenocarcinoma compared with that of patients treated with chemotherapy alone (median OS 13.8 vs. 11.1 months, HR 0.74, 95% CI 0.0–0.91; p = 0.0046). In a post-hoc subgroup analysis, the HER-2 overexpression (IHC3+ or IHC2+ and FISH-positive) showed an improvement in OS (median OS 16.0 vs. 11.8 months, HR 0.65, 95% CI 0.51–0.83; p = 0.036) ^[22][21].

Following the success of the ToGA trial, several randomized phase III trials evaluated the efficacy of other HER2-targeted therapies in patients with HER2-positive advanced gastric and gastroesophageal junction cancers ^[45][46][23,53]: first-line treatment with lapatinib + capecitabine + oxaliplatin (TRIO-013/LOGiC) ^[47][54], pertuzumab and trastuzumab + fluoropyrimidine + cisplatin (JACOB) ^[48][55], second-line treatment with lapatinib + paclitaxel (Tytan) ^[49][56], and T-DM1 (GATSBY) ^[50][57]. However, none of these treatments could improve the clinical outcomes of patients with HER2-positive GC. Several mechanisms have been proposed to cause resistance to HER2-targeted therapies, including (1) the intratumor heterogeneity of HER2, (2) aberrant activation of the PIK3CA signaling pathway (a downstream signal of HER2), and (3) simultaneous amplification of EGFR, MET, and CCNE1 ^[45][23].

In a HER2-positive immunocompetent mouse model, anti-PD-1 antibodies have been reported to significantly improve the antitumor activity of trastuzumab by enhancing antibody-dependent cellular cytotoxicity (ADCC) ^[51][64]. A phase II trial evaluating the efficacy of trastuzumab plus pembrolizumab in combination with chemotherapy revealed very promising results, with a median PFS of 13.0 months (95% CI, 8.6–NA) and median OS of 27.2 months (95% CI 18.8–NA) ^[52][65]. The phase III KEYNOTE-811 trial of pembrolizumab plus trastuzumab and chemotherapy revealed a statistically significant increase of 22.7% in the ORR in the pembrolizumab group compared to the placebo group (77.4% vs. 51.9%, p = 0.00006) ^[53][66]. The pembrolizumab group also displayed more profound responses than the placebo group (median change from baseline: 65% vs. 49%; 80% or ≥80% decrease from baseline: 32.3% vs. 14.8%). Further, more complete responses were observed in the pembrolizumab group than in the placebo group (11.3% vs. 3.1%). These interim results for KEYNOTE-811 led to the expedited FDA approval of the addition of pembrolizumab to trastuzumab and chemotherapy as the first-line treatment of patients with HER2-positive advanced GC. The KEYNOTE-811 trial demonstrated that the addition of ICIs to molecularly targeted therapy may be another potentially efficient strategy to overcome HER2 resistance in GC ^[53][66].

Several studies have shown that the administration of taxane or irinotecan results in higher survival rates than best supportive care as a second-line chemotherapy for GC patients with adequate performance status ^{[54][55][56][57]}[67,68,69,70]. In addition, the efficacy of the anti-vascular endothelial growth factor receptor (VEGFR)-2 monoclonal antibody, ramucirumab, was proven in the REGARD and RAINBOW randomized phase III trials ^[58][59][71,72]. In the multicenter randomized phase III REGARD trial ^[59][72], patients with advanced gastric and gastroesophageal junction cancers who progressed after first-line chemotherapy were randomized to receive ramucirumab or a placebo. The median OS was 5.2 months for ramucirumab and 3.8 months for the placebo (p = 0.047). The randomized phase III RAINBOW trial evaluated paclitaxel with or without ramucirumab in patients with advanced gastric and gastroesophageal junction cancers who had progressed after primary chemotherapy ^[58][71]. Patients who received ramucirumab plus paclitaxel had significantly longer median OS (9.63 months) than patients receiving paclitaxel alone (7.36 months, p < 0.0001); the median PFS was 4.4 and 2.86 months, respectively; and the ORR was 6% for paclitaxel alone versus 28% for ramucirumab plus paclitaxel (p = 0.0001).

The KEYNOTE-061 trial comparing pembrolizumab alone with paclitaxel failed to meet its primary endpoint (superior OS for patients with PD-L1 CPS ≥ 1) for patients with advanced gastric and gastroesophageal junction cancers who progressed after first-line chemotherapy. However, pembrolizumab-treated patients with PD-L1 CPS ≥ 10 had superior survival rates compared to patients with PD-L1 CPS ≥ 1 ^[60][73]. Paclitaxel plus ramucirumab is generally recommended based on the results of the RAINBOW study and is currently administered in many cases. As pembrolizumab is highly effective in patients with MSI-H, pembrolizumab must be used appropriately via the performance of the MSI test ^[61][74].

Recently, several new drugs have been developed as treatment options for ineffective secondary chemotherapy. The ATTRACTION-2 trial, a phase III study that sought to evaluate the efficacy of nivolumab in GC after third-line treatment ^[62][75], revealed the superiority of nivolumab over a placebo in OS (nivolumab vs. placebo, median 5.26 months vs. 4.14 months, HR 0.62, 95%CI 0.50–0.75). Notably, the response rate to nivolumab was 11.9%, but the OS in responders was very long (26.68 months). However, the effectiveness of nivolumab in combination with chemotherapy as the first-line treatment for HER2-negative GC has been demonstrated, which may reduce the possibility of using nivolumab as a third-line therapy. A phase III study (TAGS study) was conducted to determine the efficacy of trifluridine/tipiracil (FTD/TPI) as a third-line or later drug and revealed the superiority of FTD/TPI over a placebo in terms of OS (FTD/TPI vs. placebo, median 5.7 months vs. 3.6 months, HR 0.69, 95% CI 0.56–0.85) ^[63][76]. Of note, grade 3 or higher neutropenia was observed in 34% of patients, whereas febrile neutropenia was only observed in 2% of patients. FTD/TPI is an oral drug. As many patients with GC have poor oral intake, it is important to administer FTD/TPI when oral intake is still possible.

As mentioned above, no drug has demonstrated efficacy in HER2-targeted therapy for GC, except trastuzumab, as a first-line therapy. A phase II study (DESTINY-Gastric01 study) was conducted to explore the efficacy of trastuzumab deruxtecan (T-DXd) in HER2-positive GC. T-DXd displayed efficacy for the first time as a HER2-targeted therapy after second-line treatment. The DESTINY-Gastric01 trial randomized patients 2:1 to receive T-DXd or chemotherapy ^[64][77]. The efficacy of T-DXd was very high, with an ORR of 51% for T-DXd and 14% for chemotherapy (p < 0.001). Further, 9% of patients (11/119 patients) had complete response to T-DXd. The OS and PFS results were also promising with T-DXd (T-DXd vs. chemotherapy, median OS 12.5 vs. 8.4 months, HR 0.59, 95% CI 0.39–0.88; median PFS 5.6 vs. 3.5 months, HR 0.47, 95% CI 0.31–0.71). The most frequent grade 3 or higher adverse events were neutropenia (51%), anemia (38%), leukopenia (21%), and anorexia (17%). In addition, febrile neutropenia was observed in six patients and drug-induced pneumonia was observed in 10% (12/125) of patients. Although T-DXd-related interstitial lung disease is an adverse event requiring caution, it is considered a manageable adverse event with nine cases of grade 1/2, two cases of grade 3, and one case of grade 4 recorded. T-DXd is preferred over nivolumab as a third-line therapy for HER2-positive patients owing to its high therapeutic efficacy.