Reply: Factors Associated With Recurrence and Survival in N0 Gastric Cancer
We thank Drs Yang, Chen, and Hu for their insightful comments on our study, “Factors associated with recurrence and survival in lymph node-negative gastric adenocarcinoma: A 7-institution study of the US Gastric Cancer Collaborative.”1 They provide several salient points regarding extent of lymph node dissection and accuracy of staging in gastric cancer, and raise the concern that nearly half of the patients in our study group did not receive the recommended staging operation of a D2 lymphadenectomy or having at least 15 lymph nodes harvested.
With regard to this point, we certainly do not dispute that D2 lymphadenectomy is favored over D1 alone for improved survival,2 and that at least 15 lymph nodes should be examined for accurate staging.3,4 This is certainly the goal of most surgical oncologists who care for patients with gastric adenocarcinoma. Our data did show that, incidentally, there was no difference in survival between those who had D2 lymphadenectomy versus D1, or between those who had greater than 15 nodes examined and those who did not, although one must keep in mind that these outcomes exist within a group of nominally “node-negative” patients, and are therefore not entirely comparable with the greater gastric cancer population, off of which recommendations for extent of dissection have been established. These findings in our data are not the primary outcomes of our study, and we certainly do not use these data to extrapolate that patients with resectable gastric cancer should not undergo D2 lymphadenectomy or have at least 15 nodes examined for staging. The result that almost half of our patients failed to undergo the recommended amount of lymph node staging is both illustrative and instructive. The 7 participating institutions of the US Gastric Cancer Collaborative all represent large volume academic institutions, the majority of which are designated as comprehensive cancer centers. Although the rate of compliance did increase throughout the decade plus spanned by this study, many diverse factors may influence the extent of lymph nodes examined as part of a gastrectomy. This includes both the dissection performed by the surgeon and the number of lymph nodes ultimately examined by the pathologist. Although these findings highlight an area in need of improvement, our approach to the present analysis was one of clinical practicality. As reoperation for accurate staging in these patients is not a realistic option, what treatment guidelines should apply to patients who have zero positive lymph nodes if the total number of nodes examined is less than 15 or if they received less than a D2 lymphadenectomy? Our results, that greater depth of invasion, LVI and signet ring histology as associated with decreased survival in these patients, may provide some guidance for adjuvant treatment in a population of patients where existing guidelines may be ambiguous.
We appreciate the attention this important topic brings, and as the US Gastric Cancer Collaborative continues to expand to incorporate greater numbers of patients, we hope to contribute further studies that can more accurately dissect risk factors specific to more accurately subgrouped populations of patients with gastric cancer.