ORIGINAL ARTICLE Stage Migration Effect on Survival in Gastric Cancer Surgery With Extended Lymphadenectomy The Reappraisal of Positive Lymph Node Ratio as a Proper N-Staging Seong-Ho Kong, MD, MS, Hyuk-Joon Lee, MD, PhD, Hye Seong Ahn, MD, MS, Jong-Won Kim, MD, Woo Ho Kim, MD, PhD,†‡ Kuhn Uk Lee, MD, PhD, FACS, and Han-Kwang Yang, MD, PhD, FACS Objective: The purpose of this study is to analyze the relationship between the number of examined lymph nodes (NexLN) and survival in gastric cancer and to determine whether the metastatic/examined lymph node ratio (LN ratio) system can compensate for the shortcomings of the UICC/AJCC staging. Methods: Prospective data of 8949 primary T1-T4a gastric cancer patients who underwent curative surgery were reviewed. The patients were stratified by T-stage and grouped according to NexLN; 1 to 14 exLN denoted the first group and every subsequent 10 LNs thereafter. Numbers of LN and 5-year survival rates were analyzed according to NexLN. “The NR-staging system” was generated using 0.2 and 0.5 as the cut-off values of LN ratio and then compared with UICC/AJCC stages. Results: The proportion of advanced N-stage increased with NexLN. Survival and the LN ratio were constant regardless of NexLN when combining all N0- N3b patients, however, T2/3 and T4a patients showed an increasing tendency toward survival in N1/2 and N3a as NexLN increased, mainly due to a stage migration effect. The LN ratio system showed better patterns of distribution of the LN stage and survival graph. The power of the differential staging of the LN ratio system was fortified with higher NexLN. Conclusion: The relationship between NexLN and survival is probably af- fected by stage migration in a high-volume gastric cancer center. The LN ratio system could be a better option to compensate for this effect, and the value of the prognosis prediction in this system increases with a higher NexLN. (Ann Surg 2012;255:50–58) L ymph node (LN) dissection is crucial in gastric cancer surgery for treatment and staging. However, the extent of LN dissection represented by D0, D1, D2, and D3 in the Japanese classification is very complicated and has variations in each category. 1 The num- ber of the examined LN (exLN) has instead been used as a simpler indicator of the extent of LN dissection. Smith et al 2 reported a ther- apeutic benefit of a high number of exLN using SEER data, and the UICC/AJCCTNM classification required a minimum of 15 exLN for adequate N staging. 3 Studies using the number of exLN as an indica- tor of the extent of LN dissection have been performed in breast and colon cancer. 4–8 In general, Eastern doctors believe in the therapeutic benefits of extended LN dissection and try to obtain as many LN as possible From the * Department of Surgery; Cancer Research Institute; and Department of Pathology, Seoul National University College of Medicine, Seoul, Korea. Disclosure: We have no commercial sponsorship for this study. All the authors of this study including Seong-Ho Kong, Hyuk-Joon Lee, Hye Seong Ahn, Jong- Won Kim, Woo Ho Kim, Kuhn Uk Lee and Han-Kwang Yang do not have any financial interest or conflict with any industries or parties. Reprints: Hyuk-Joon Lee, MD, PhD, Department of Surgery and Cancer Research Institute Seoul National University College of Medicine, Seoul, Korea, 101 Daehang-Ro, Jongno-Gu, Seoul 110-744, Korea. E-mail: appe98@snu.ac.kr. Copyright C 2011 by Lippincott Williams & Wilkins ISSN: 0003-4932/11/25501-0050 DOI: 10.1097/SLA.0b013e31821d4d75 during surgery. 9,10 However, many Western doctors regard the num- ber of metastatic LN simply as an index of the tumor burden and instead highlight the negative effects of extended LN dissection. 11,12 Too many exLN are also believed to produce a significant stage migration, 13 so it tends to be only small number of LNs are examined in Western institutes. 2,14,15 However, it is still uncertain whether a large number of exLN contributes to stage migration or rather to an accurate assessment of the LN metastatic status; the optimal number of exLN that provides a survival benefit and/or accurate staging is also unknown. On the other hand, the large proportion of patients with poor survival rates and less than 15 exLN suggests that the ratio of the number of metastasized LN to the number of exLN (LN ratio) should be used in Western countries. 14–18 However, the LN ratio system has not been universally accepted, especially by Eastern institutions in which most cases have sufficient number of exLN for the present staging system. A few studies have described the usefulness of the LN ratio system in Korea and Japan, 19,20 and about 13% of the operations in Japan were reported to have less than 15 exLN. 21 It is worthwhile to validate the value of the LN ratio for the patients with sufficient number of exLN in Eastern institutions if the survival rate is greatly affected by the number of exLN. The purpose of this study was to analyze the effect of the number of exLN on survival in gastric cancer patients in an Eastern high-volume gastric cancer center. The LN ratio system was also val- idated regarding the necessity and the appropriateness of LN staging compared with UICC/AJCC staging for the patients with high-yield exLN. PATIENTS AND METHODS A prospectively collected database of 10,492 gastric cancer patients who underwent gastric resection with curative intent at the Seoul National University Hospital (SNUH) from 1991 to 2006 was reviewed. Cases with the following characteristics were excluded: recurrent cancer, residual stomach cancer, concurrent malignancy in other organs, preoperative chemotherapy, R1 or R2 resection, direct invasion of adjacent organs and missing exLN data. In total, 8949 patients were selected (Fig. 1). For the overall survival, the 2007 update of the database of the National Statistical Office of Korea, which records all deaths of Korean people, was accessed. The distribution of the number of patients was shown in Figure 2 when grouped according to every 5 exLN. To avoid small numbers in each group, the first group comprised patients with 1 to 14 exLN and the other groups were composed of every subsequent 10 exLN. The extent of LN dissection was mainly planned by pre- operative assessment of the depth of the tumor to form the basis of the stratification as follows: mucosa and submucosa, muscularis pro- pria and subserosa, and serosa infiltration. These correspond to T1, T2/3 and T4a in the 7th edition of the UICC/AJCC staging man- ual, respectively. 22 The standard surgical strategy employed at the Copyright © 2011 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited. 50 | www.annalsofsurgery.com Annals of Surgery Volume 255, Number 1, January 2012