Open, Laparoscopy-assisted, Robotic-assisted Distal Gastrectomy for Gastric Cancer: Evidence from Randomized Clinical Trials

Comparison of Short-term Oncological Outcomes Between ODG and LADG (Table I)

Eight trials have compared the short-term oncological outcomes between LADG and ODG. In the first trial, Kim et al. evaluated the short- and long-term oncological outcomes of LADG and ODG for 164 early GCs (clinical T1N0M0 and clinical T1N1M0) in a randomized trial (NCT00546468) (7). The primary endpoint was disease-free survival (DFS). The secondary endpoints were quality of life (QOL) and morbidity. QOL was assessed using the European Organization for Research and Treatment of Cancer QLQ-C30 and QLQ-STO22. In 2008, Kim et al. reported the short-term oncological outcomes. All the procedures were performed by a single surgeon using a well-established laparoscopic surgical team. The trial included 164 patients, with 82 registered in the LADG arm and 82 registered in the ODG arm. The patient backgrounds of the ODG and LADG arms were similar. In the analysis of the short-term outcomes of the two arms, QOL after surgery was significantly better in the LADG arm. Both the EORTC QLQ-C30 and QLQ-STO22 scores were significantly better in the LADG arm. In addition, postoperative pain, appetite loss, and sleep quality were better in the LADG arm. As a result, they concluded that LADG improved the QOL outcomes in comparison to ODG.

In the second trial, Sakuramoto et al. compared the postoperative use of analgesics between LADG and ODG in 63 patients with early GC in a randomized trial (UMIN000001513) (8). The primary endpoint was 4-day postoperative analgesic use. The secondary endpoints were postoperative residual pain level (assessed using a visual analog scale) and short-term oncological outcomes (e.g., postoperative complications and postoperative hospital stay). LADG and ODG were performed by a single gastrointestinal surgical team with extensive experience in open and laparoscopic procedures. The trial included 63 patients, with 31 patients registered in the LADG arm and 32 registered in the ODG arm. The patient backgrounds of the ODG and LADG arms were similar. The primary endpoint analysis showed that the ratio of analgesic use in the LADG arm to the ODG arm was 0.68 (95%CI=0.49-0.95, p=0.022). The postoperative use of analgesics in the LADG arm was seven times lower than that in the ODG arm. In addition, the number of epidurals used was lower in the LADG arm, and pentazocine hydrochloride was used much less frequently in the LADG and ODG arms. They concluded that LADG, performed by an expert surgeon, results in less postoperative pain and is associated with fewer postoperative complications.

Third, Takiguchi et al. evaluated the postoperative physical activity and postoperative pain between LADG and ODG in 40 patients with stage IA or IB GC in a randomized trial (UMIN 000002547) (9). The primary endpoint was the comparison of physical activity using the Active Tracer. The secondary endpoints were the duration of pain control, time of pain rescue, and the visual analog pain scale. All laparoscopic and open procedures were performed by a single surgeon who was well-trained in both laparoscopic and open gastric surgeries. The trial included 40 patients, 20 were registered in the LADG arm and 20 in the ODG arm. The patient characteristics (e.g., age, body composition, and clinical and pathological tumor stage) of the ODG and LADG arms were similar. The primary endpoint analysis demonstrated that the recovery of physical activity was two days faster in the LADG arm. When comparing the postoperative recovery between the LADG and ODG arms, 60% recovery relative to the preoperative level was observed at POD3 in the LADG arm and POD5 in the ODG arm. In addition, 80% recovery relative to the preoperative level was observed at POD5 in the LADG arm and at POD7 in the ODG arm. Moreover, postoperative pain (evaluated using a visual analog scale) at rest and walking was significantly lower in the LADG arm. They concluded that the outcomes were favorable, and that the recovery of physical activity occurred earlier in the LADG arm.

In the fourth trial, Kim et al. evaluated the long-term oncological outcomes of ODG and LADG in 1,416 patients with early GC in a randomized phase III trial (NCT00452751) (10). The primary endpoint was 5-year overall survival (OS). In this trial, the participating surgeons were to have conducted at least 50 cases each of LADG and ODG, and each participant’s institution was to conduct at least 80 cases per year for surgical quality control. In 2015, Kim reported the short-term oncological outcomes of the trial. The trial included 1,416 patients, including 686 who were registered in the LADG arm, and 698 registered in the ODG arm. Patient background factors (e.g., age, sex ratio, body mass index, and clinical tumor stage) were similar between the ODG and LADG arms. The incidence rates of postoperative surgical complications in the ODG and LADG arms were 18.9% and 13.7%, respectively (p=0.009). In contrast, the mortality rates were 0.3% in the ODG arm and 0.6% in the LADG arm, which did not amount to a significant difference (p=0.687). The reoperation rates of the ODG and LADG arms were also similar (1.5% vs. 1.2%, p=0.726). They concluded that LADG for patients with clinical stage I GC is safe and has the benefit of a lower occurrence of wound complications in comparison to conventional ODG.

In the fifth trial, Katai et al. evaluated both short- and long-term oncological outcomes between ODG and LADG in 912 patients with early GC in a randomized phase III trial (UMIN000003319) (11). The primary endpoint was relapse-free survival (RFS). The secondary endpoints were OS of all randomized patients, the proportion of LADG procedures that were completed, the proportion of LADG procedures that were converted to open surgery, adverse events, short-term clinical outcomes, and postoperative QOL. In this trial, the surgeons participating in the trial had performed 30 or more cases of LADG and ODG. The trial included 912 patients, with 457 registered in the LADG arm, and 455 registered in the ODG arm. The patient backgrounds of the two arms were similar. In the short-term oncological analysis, the operative time was longer in the LADG arm than that in the ODG arm (median 278 vs. 194 min, p<0.001), while the blood loss was lower (median 38 vs. 115 ml, p<0.001). The proportion of grade ≥3 surgical complications was 3.3% in the LADG arm and 3.7% in the ODG arm. There were no significant differences between the groups. They concluded that LADG was as safe as ODG in terms of adverse events and short-term clinical outcome.

In the sixth trial, Park et al. evaluated the feasibility and safety of LADG with D2 lymph node dissection and ODG in 204 patients with advanced gastric cancer [cT2-4aN0-2M0 (AJCC 7th staging system)] in a randomized phase II trial (INCT01088204) (12). The primary endpoint was the noncompliance rate, defined as the proportion of patients with >1 empty lymph node station. The secondary endpoints were the surgical and oncological outcomes. All surgeons included in the trial had performed at least 30 LADG procedures prior to the start of this study. The trial included 204 patients, 105 were registered in the LADG arm, and 99 registered in the ODG arm. The patient characteristics (e.g., age, body composition, and clinical and pathological tumor stage) of the two arms were similar. The noncompliance rates of the LADG and ODG arms were 47% and 43.2%, respectively. There were no significant differences between the groups (p=0.648). In addition, there was no significant difference in the number of harvested lymph nodes between the LADG and ODG arms (37.0±13.4 vs. 39.7±13.3, p=0.168). In a subgroup analysis, the noncompliance rates of the LADG and ODG arms were 43.5% and 54.5%, respectively, in clinical stage I and 46.2% and 48.9% in clinical stage II (p=0.458 and p=0.788). In contrast, the noncompliance rates in the LADG and ODG arms were 52.0% and 25.0%, respectively. There were significant differences between the groups (p=0.043). In the secondary endpoint analysis, the postoperative surgical complication rates, surgical stress response, and long-term oncological outcomes were similar between the two arms. They concluded that LADG was feasible for AGC treatment based on the non-compliance rate of D2 lymph node dissection.

In the seventh trial, Wang et al. evaluated the short-term oncological outcomes of ODG and LADG in 446 GC patients with cT2-4aN0-3M0 (AJCC 7^th staging system) in a randomized trial (NCT02464215) (13). The primary endpoint was postoperative complications within 30 days after surgery. The secondary endpoint was 3-year RFS. All surgeons involved in this trial specialized in gastric surgery and had already conducted at least 60 ODG and 60 LADG with D2 lymphadenectomy. A total of 223 patients each were registered in the LADG and ODG arms. Patient backgrounds, such as age and clinical tumor stage, were similar between the ODG and LADG arms. The primary endpoint analysis showed that postoperative surgical complication rates were similar between the LADG and ODG arms. The incidence rates of Clavien–Dindo grade II complications in the LADG and ODG arms were 9.5% and 12.7%, respectively (p=0.292). In addition, the incidence rates of grade IIIa complications in the LADG and ODG arms were 0.5% and 0.5%, respectively (p=1.000), while those of grade IIIb complications were 1.8% and 3.6% in the PDG arm (p=0.259). No patients in either group developed grade IV complications and there was no mortality. The time to first flatus [all LADG vs. ODG: 2.8 days vs. 3.1 days (p=0.013)], time to first liquid intake (7.0 days vs. 7.9 days, p=0.003), and postoperative hospital stay (9.9 days vs. 10.9 days, p=0.018) were significantly shorter in the LADG arm. The authors concluded that LADG performed by credentialed surgeons was safe and feasible for patients with advanced GC relative to conventional ODG.

Lastly, Li et al. compared the short-term oncological outcomes between ODG and LADG in 96 patients with locally advanced GC after neoadjuvant chemotherapy in a randomized phase II trial (NCT02404753) (14). The primary endpoint was 3-year RFS. The secondary endpoints were safety, feasibility, and efficacy during the postoperative clinical course. The short-term oncological outcomes of this trial were reported in 2019. Fifty-eight patients were registered in the ODG arm, and 47 were registered in the LADG arm. The patient background factors (e.g., age, dose of neoadjuvant chemotherapy, age-adjusted Charlson Comorbidity Index score, and clinical tumor stage) in the ODG and LADG arms were similar. Although the incidence of Clavien-Dindo grade II complications was significantly higher in the ODG arm (40% vs. 13%, p=0.004), the incidence of Clavien-Dindo classification grade ≥III complications in the ODG and LADG arms were similar (4% vs. 13%, p=0.25). In the postoperative clinical trial, the median visual analog scale score was significantly lower in the LADG arm on postoperative day 2 (2 vs. 3, p=0.04) and 3 (1.5 vs. 3, p=0.04) but not on postoperative day 1 (2 vs. 3, p=0.20). Moreover, patients in the LADG arm were less likely than those in the ODG arm to terminate postoperative adjuvant chemotherapy due to adverse effects (22% vs. 42%, p=0.04). The authors concluded that LADG appears to offer the benefits of better postoperative safety and adjuvant chemotherapy tolerance in comparison to ODG for patients with locally advanced GC who received neoadjuvant chemotherapy.

Comparison of the Long-term Oncological Outcomes of ODG and LADG (Table II)

Six trials have evaluated the long-term oncological outcomes of LADG and ODG. First, Kim et al. evaluated the long-term oncological outcomes between ODG and LADG in 164 patients with early GC (clinical T1N0M or clinical T1N1M0) in a randomized trial (NCT00546468) (15). The primary endpoint was the 5-year DFS. Secondary endpoints were OS and QOL. In this trial, all the procedures were performed by a single surgeon. The trial included 164 patients, 82 were registered in the LADG arm and 82 were registered in the ODG arm. In the primary endpoint analysis, the 5-year DFS rates of the ODG and LADG arms were 97.6% and 98.8%, respectively. The 5-year OS rates of the ODG and LADG arms were also similar (96.3% vs. 97.6%, p=0.91). In addition, the long-term QOL was similar. They concluded that LADG showed similar DFS and OS to ODG in the treatment of early gastric cancer. LADG did not show advantages over ODG in terms of moderate-to-severe complications or long-term QOL.

Second, Kim et al. evaluated the long-term oncological outcomes of ODG and LADG in 1,416 patients with early GC in a randomized phase III trial (NCT00452751) (16). The primary endpoint was 5-year OS. The trial included 1,416 patients, 686 were registered in the LADG arm, and 698 were registered in the ODG arm. The patient background factors (e.g., age, sex ratio, body mass index, and clinical tumor stage) of the ODG and LADG arms were similar. In the primary endpoint analysis, the 5-year OS rates in the ODG and LADG arms were 93.3% and 94.2%, respectively. The 5-year cancer-specific survival rates of the ODG and LADG arms were also similar (97.1% vs. 97.2%, p=0.91). They concluded that LADG is associated with low morbidity and survival, which are comparable to ODG, without compromising long-term oncologic outcomes.

Third, Huang and Yu et al. evaluated the long-term oncological outcomes between ODG and LADG in 1,050 patients with locally advanced GC (clinical stage T2-4a and no nodal metastasis or limited perigastric nodal metastasis) in a randomized phase III trial (NCT01456598) (17, 18). The primary endpoint was 3-year RFS. The secondary objectives were to compare postoperative morbidity and mortality, postoperative recovery, QOL, and 3-year OS between the two groups. To participate in the trial, the surgeons and hospitals had to meet the following criteria: surgeons had to have performed more than 100 gastrectomies for GC (including 50 cases each of laparoscopic and open gastrectomy), and hospitals had to have an annual volume of more than 80 gastrectomies. The trial included 1,039 patients, 519 were registered in the LADG arm and 520 were registered in the ODG arm. The patient backgrounds of the two arms were similar. In the primary endpoint analysis, the 3-year RFS rate was 81.3% (95%CI=77.0-85.0%) in the LADG arm and 80.3% (95%CI=76.0-85.0%) in the ODG arm. A Cox regression analysis after stratification by surgeon revealed an HR of 1.035 (95%CI=0.762-1.406; log-rank p=0.827; p for noninferiority, p=0.039). The 5-year OS rate was 88.9% (95%CI=86.0-91.8%) in the LADG arm and 88.7% (95%CI=85.8-91.6%) in the ODG arm. They concluded that LADG with D2 lymphadenectomy performed by experienced surgeons in high-volume specialized institutions resulted in similar 3-year RFS and 5-year OS rates in comparison to ODG among patients with locally advanced GC.

Fourth, Katai et al. evaluated the long-term oncological outcomes between ODG and LADG in 912 patients with early GC in a randomized phase III trial (UMIN000003319) (19). The primary endpoint was RFS. In this trial, the surgeons participating in the trial were involved in 30 or more cases of LADG and ODG. The trial included 912 patients, 457 were registered in the LADG arm and 455 were registered in the ODG arm. The patient background factors (e.g., age, sex ratio, clinical T and N stage, and body mass index) of the ODG and LADG arms were similar. In the primary endpoint analysis, the 3-year RFS rate was 97.4% (95%CI=95.4-98.5%) in the LADG arm and 95.8% (95%CI=93.5-97.3%) in the ODG arm. The 5-year RFS rates were 95.1% (95%CI=92.7-96.8%) in the LADG arm and 94.0% (95%CI=91.4-95.9%) in the ODG arm. The HR for RFS (ODG vs. LADG) was 0.84 (95%CI=0.56-1.27). The HR for RFS (ODG vs. LADG) was 0.84 (95%CI=0.56-1.27). In addition, the 3-year OS rates were 98.3% (95%CI=96.6-99.1%) in the LADG arm and 97.4% (95%CI=95.4-98.5%) in the ODG arm. The 5-year OS rates were 97.0% (95%CI=94.9-98.2%) in the LADG arm and 95.2% (95%CI=92.7-96.8%) in the ODG arm. They found that LADG was non-inferior to ODG, in terms of RFS, for clinical stage I gastric cancer.

Forth, Hyung et al. and Son et al. evaluated the long-term oncological outcomes of ODG and LADG in 1,039 patients with locally advanced GC (clinical T1-4aN1-3M0) in a randomized phase III trial (NCT01609309) (20, 21). The primary endpoint was the 3-year DFS. The trial included 1,039 patients, 519 were registered in the LADG arm and 520 were registered in the ODG arm. The patient backgrounds of the ODG and LADG arms were similar. In the primary endpoint analysis, the 3-year DFS rate was 76.5% in the LADG arm and 77.8% in the ODG arm. The 5-year OS rates were 72.6% in the LADG arm and 76.3% in the ODG arm. The HR for OS (ODG vs. LADG) was 1.17 (95%CI=0.93-1.48, p=0.19). They concluded that LADG with D2 lymphadenectomy performed by experienced surgeons in high-volume specialized institutions resulted in similar 5-year RFS and OS in comparison to ODG in patients with locally advanced GC. Fifth, Etoh et al. evaluated the long-term oncological outcomes between ODG and LADG in 502 patients with locally advanced GC in a randomized phase III trial (UMIN000003420) (22). The primary endpoint was the 5-year RFS. The secondary endpoints were 5-year OS, mortality, and morbidity. In this RCT, surgeons and institutes were strictly evaluated; all surgeons had Endoscopic Surgical Skill Qualification System certification and had performed a specified number of both laparoscopic and open gastrectomy procedures. Two-hundred fifty-four patients were registered in the ODG arm and 248 patients were registered in the LADG arm. The patient background factors (e.g., age, sex ratio, and clinical tumor stage) and short-term oncological outcomes of the two arms were similar. The incidence rates of postoperative surgical complications in the ODG and LADG arms were 10.7% and 11.5%, respectively. In the primary endpoint analysis, the 5-year RFS was 73.9% (95%CI=68.7-79.5%) in the ODG arm and 75.7% (95%CI=70.5-81.2%) in the LADG arm. The hazard ratio (HR) for RFS in the LADG arm vs. ODG arm was 0.96 (95%CI=0.72-1.26; noninferiority 1-sided p=0.03). In the secondary endpoint analysis, the 5-year OS rates were 79.8% (95%CI=75.0-84.9%) in the ODG arm and 81.7% (95%CI=77.0-86.7%) in the LADG arm, whereas the HR for 5-year OS in the LADG arm vs. ODG arm was 0.83 (95%CI=0.57-1.21; p=0.34). In addition, the recurrence patterns of the ODG and LADG arms were similar. The incidence rate of peritoneal recurrence was 4.3% in the ODG arm and 7.7% in the LADG arm (p=0.13). The incidence rate of lymph node recurrence was 6.3% in the ODG arm and 3.6% in the LADG arm (p=0.22). The incidence rate of liver recurrence was 4.7% in the ODG arm and 2.8% in the LADG arm (p=0.35). They concluded that LADG with D2 lymph node dissection for locally advanced GC, when performed by qualified surgeons, was proven to be non-inferior to ODG.

Comparison of Oncological Outcomes Between LADG and RDG

Two trials comparing LADG and RDG have been conducted, and two trials comparing LADG and RDG are currently ongoing. Lu et al. evaluated the short-term oncological outcomes between LADG and RDG in 300 patients with GC [clinical T 1-4aN0/+M0 (AJCC 8th staging system)] in a randomized trial (NCT03313700) (23). The primary outcome was 3-year DFS. The secondary endpoints were short-term oncological outcomes (e.g., perioperative recovery course, morbidity, mortality, and cost). The short-term oncological outcomes were reported in 2021. Among the 300 patients, 283 (RDG arm, n=141; LDG arm, n=142) were included in the modified intention-to-treat analysis. The patient backgrounds of the RDG and LADG arms were similar. Regarding surgical outcomes, the total surgical time was significantly longer in the RDG arm than in the LADG arm (201 min vs. 181 min, p=0.001), while intraoperative blood loss was greater in the LADG arm than in the RDG arm (55 ml vs. 41 ml, p=0.045). There were no significant differences in the number of harvested lymph nodes between the RDG and LADG arms (40.9±11.2 vs. 39.9±12.2, p=0.452). In terms of the postoperative recovery, the RDG arm showed significantly shorter times to ambulation (2 days vs. 2.5 days, p=0.005), first flatus (3.2 days vs. 3.5 days, p<0.001), and liquid intake (3.5 days vs. 3 days, p=0.001) in comparison to the LADG arm, while reoperation rate (all RDG vs. LADG: 0% vs. 0.7%, p=1.000) and postoperative hospital stay (7.9 days vs. 8.2 days, p=0.062) were similar. The incidence rates of postoperative surgical complications in the RDG and LADG arms were 9.2% and 17.6%, respectively (p=0.039). No patients died during hospitalization in either arm. In the adjuvant chemotherapy course, patients in the RDG arm were more likely to initiate adjuvant chemotherapy earlier than those in the LDG arm (28 days vs. 32 days, p=0.003). The total cost was higher in the RDG arm than that in the LADG arm (USD 13423 vs. USD 10165, p<0.001). They concluded that RDG is associated with a lower morbidity rate and faster recovery and provided evidence for the application of RDG in patients with GC.

Second, Ojima et al. evaluated the short-term oncological outcomes between LADG and RDG in 241 patients with GC (clinical stages I-III) in a randomized phase 3 trial (UMIN00031536) (24). The primary outcome was the rate of postoperative intra-abdominal infections. They estimated that the incidence rates of intra-abdominal infectious complications were 11% in the LADG group and 2% in the RDG group (odds ratio=0.165; 95%CI=0.051-0.512). The trial included 241 patients, 122 were registered in the LADG arm and 119 were registered in the RDG arm. The patient backgrounds of the two arms were similar. In the primary endpoint analysis, the incidence rates of grade ≥II intra-abdominal infectious complications were 8.5% in the LADG arm and 6.2% in the RADG arm (p=0.500). In addition, the incidence rates of grade ≥III intra-abdominal infectious complications were 7.7% in the LADG arm and 4.4% in the RADG arm (p=0.300). They concluded that RDG did not reduce intra-abdominal infectious complications in comparison to LADG. In addition, there are two ongoing trials comparing LADG and RDG (25, 26). Future trials are needed to clarify the role of RDG in GC.

Future Suggestions

According to previous RCTs, the short-term and long-term oncological outcomes of patients who received ODG, LADG, and RDG were similar. However, the following two points need to be clarified in future studies: First, the economic evaluation of each approach was unclear. One RCT compared economic data between RDG and LADG. Lu et al. reported that the total cost in the RDG arm was higher than that in the LADG arm (USD 13,423 vs. USD 10,165, p<0.001) (23). According to the non-RCT data, two studies from Japan demonstrated that there was no significant difference in operation theater costs or total hospitalization costs between LADG and ODG (27). However, a study from China reported that LADG was associated with significantly increased operating theater costs, while there was no significant difference in the total hospitalization costs between LADG and ODG. Previous reports indicate that LADG or RDG may cost more than ODG; however, the total hospital stay may be shorter, and the total cost may be lower in LADG or RDG than in ODG. Second, it is unclear whether the RCT results are generally feasible for GC treatment. Previous RCTs were performed by a single surgeon or by well-trained expert surgeons with their team. In addition, strict patient selection criteria were used. Recently, a nationwide registry database in Japan indicated that LADG was associated with a higher rate of pancreatic leakage than ODG in daily clinical practice (28, 29). Thus, it is necessary to analyze the safety and feasibility of LADG and RDG in daily clinical practice.