In the propensity-matched cohort, we found that LG tended to have more favorable survival than OG in treating GC, without compromising safety. The survival benefit was consistent for both early and advanced stages of diseases.
Several meta-analyses on non-randomized and randomized studies have compared the short-term outcome of LG versus OG for advanced GC [21-23]. Despite the longer operative time, patients undergoing LG tend to be benefit from less blood loss, faster recovery, and less morbidity. In the current study, we further showed the performance of LG in daily practice, in the context of treating mostly advanced stage disease and the elderly. In the propensity score-matched cohort, 62% of patients treated by laparoscopic approach were stage 2 or above, and 51.9% of patients were over 70 years of age (Table 1). The rate of advanced stage of disease and the elderly in the current study were very close to the registry data in Taiwan (68.1% and 48.9%, respectively) . At this point, the rates of overall and specific morbidity for laparoscopic approach are not statistically different from the rates of open approach (Table 2). In addition, there was a trend of better survival in the LG group. Such findings may justify the routine application of LG in daily practice. The practice of applying LG for advanced GC on a large scale had been reported by studies from high-volume centers in Asia, which showed favorable short- and long-term results [24, 25]. Equal or even superior oncologic outcome was possible when performed by experienced surgeons.
Oncologic integrity is the basic requirement for LG to be applied in GC; the efficacy of which is well-accepted in early distal GC [12, 19]. The recent multi-center randomized controlled trials from Korea (COACT 1001) and China (CLASS-01) investigating advanced GC have demonstrated comparable 3-year DFS for LG [16, 17]. Noteworthy, in the COACT 1001 study, the noncompliance rate of D2 lymph node dissection was significantly lower in the laparoscopic arm for clinical stage III patients. Such a finding suggests that the efficacy of radical lymphadenectomy in LG could be more limited when applied to extensive lymph node metastasis, such as suprapancreatic stations. In our practice, the surgeon evaluated resectability from the beginning utilizing preoperative CT scans and diagnostic laparoscopy. The procedure would be converted to open if radical lymphadenectomy was technically difficult by laparoscopic approach, since conversion to open surgery is unlikely to result in inferior long-term outcomes [26, 27]. In the current study, the subgroup analysis for pathologic stage III disease showed that the 5-year DFS was even better for the LG group (39.0 vs. 18.8%, p = 0.06). In addition, the LG group had fewer overall recurrences than the OG group, with the dominant pattern of recurrence being distant metastasis rather than peritoneal carcinomatosis. Such results may help alleviate the concern of peritoneal seeding by pneumoperitoneum. Overall, LG was oncologically safe when resectability was properly evaluated.
The reduced surgical trauma by minimally invasive approach may not only result in faster postoperative recovery but better outcomes overall. One recent randomized controlled trial on patients who underwent neoadjuvant chemotherapy for advanced GC, showed that patients in the LG group were more likely to complete adjuvant chemotherapy and less likely to terminate because of adverse effects . Though the 3-year survival data were pending, one could postulate a better chance of cure with completed courses of adjuvant chemotherapy. In our experience, the better survival in the LG group in the current study could also be attributed to better tolerance of adjuvant therapy. Patients who underwent LG tended to have faster recovery and better performance status postoperatively than those who underwent OG. The better oncologic outcome achieved by laparoscopic surgery had also been observed in colon cancer [29, 30]. The level of serum interleukin-6, which has been shown to be an independent prognostic biomarker for survival in colon cancer, was lower after laparoscopic surgery [31, 32]. The reduced requirement of blood transfusion by laparoscopic approach may explain the better oncologic outcome as well .
Limitations of the current study come from its single-center, retrospective design. Residual biases associated with patient selection might not be fully considered by the propensity score-matching analysis. In addition, the regimen of adjuvant chemotherapy and the course of treatment were not standardized. However, the rate of advanced disease and frequency of elderly patients within this cohort was rather close to the true incidence in the general population. We believe this may provide surgeons with the outcome of LG in real-world practice.
In conclusion, LG can be applied to most resectable GCs, and either an advanced stage of disease or old age should not be contraindications. Patients who undergo LG could not only benefit from faster postoperative recovery but also more favorable oncologic outcomes.