Several prognostic factors have been reported in ampullary cancer after surgical resection, including age, lymph node metastasis, number of lymph node metastases, LNR, pancreatic and duodenal invasion, stage, tumor size, gross appearance, perineural invasion, histological type, and lymphovascular invasion.18–24 This study’s multivariate analysis for OS identified sex and lymph node metastasis as independent prognostic factors (Table 2).
Though sex has not been reported to be a prognostic factor in ampullary cancer, a comparison of the clinicopathological findings between males and females shows that the ratios of pPanc, lymphovascular invasion, and peri-neural invasion are significantly greater in males than in females. In addition, the histological type of poorly differentiated adenocarcinoma was detected only in males. These findings suggest that tumor progression was more advanced in males in our study.
Lymph node metastasis is a well-known prognostic factor in ampullary cancer.21–23 By dividing the patients according to the lymph node metastatic status, the proportion of recurrence in the pN− group was found to be comparatively low compared to the pN+ group (7% vs. 46%, p < 0.01); the OS rate in pN− group was also more favorable than pN+ group (Supplementary Fig. 1). For ampullary cancer, postoperative adjuvant chemotherapy can be omitted in patients without lymph node metastases. Therefore, we focused on the 46 patients with metastatic lymph nodes for further analysis.
The UICC 8th edition advocates classifying lymph node status according to both the presence or absence of lymph node metastasis and the number of metastases (pN1:3 or less, pN2:4 or more). Classifying the 46 patients into pN1 and pN2 groups revealed no significant difference in recurrence-free survival between the two groups; however, the OS in the pN2 group was significantly worse than in the pN1 group (Supplementary Fig. 5). This result indicates that N classification based on the number of metastatic lymph nodes was not useful for predicting recurrence, although it was useful for prognostic stratification. Therefore, we created an ROC curve for the number of lymph node metastases and recurrence and calculated a new cut-off value (≤ 1 and ≥ 2) (Supplementary Fig. 2). However, the two groups had no significant difference in recurrence-free survival based on the new cut-off value (p = 0.073; Supplementary Fig. 3). In patients with only one lymph node metastasis, 29% (5/17) relapsed, and the 5-year recurrence-free survival rate was 68%. These findings suggest that some patients with only one lymph node metastasis are candidates for adjuvant chemotherapy.
As described above, the number of lymph node metastases was not a predictor of ampullary carcinoma recurrence in our study. Previous studies have shown that the LNR can be adopted as a more sensitive prognostic indicator than the number of lymph node metastases in several types of cancers, including colon,25 gastric,26 pancreas,13 esophagus,27 breast,28 and bladder29 cancers. However, the clinical implications of LNRs in ampullary cancer have not been fully discussed. We showed that the LNR was significantly associated with recurrence in patients with lymph node metastases. However, a problem has been identified with using the LNR as a prognostic indicator. The LNR cannot be calculated accurately if the number of dissected lymph nodes is insufficient.
A previous report demonstrated that > 12 lymph nodes should be harvested during pancreaticoduodenectomy for ampullary cancer.23,30 In our cases, the number of dissected lymph nodes was 25.0 ± 16.4 (Table 1) in the whole group and 20.8 ± 16.1 in pN− group (Supplementary Table 1), suggesting adequate lymph node dissection was performed in our study. According to our analyses for metastatic sites, the superior mesenteric artery was the second most frequent metastatic site after the peripancreatic head region. Therefore, calculating an accurate LNR requires lymph node dissection to be performed considering the lymph flow from the metastatic sites. In particular, the dissection of lymph nodes around the superior mesenteric artery is important.
We calculated the optimal cut-off value of LNR, 0.07 (one metastasis per 14 dissected lymph nodes), using an ROC curve (Fig. 1). The LNR ≥ 0.07 group had significantly worse recurrence-free survival and OS than the LNR < 0.07 group (Fig. 2). The LNR may be a significant indicator of recurrence and prognosis in patients with ampullary cancer. Several studies have been reported on the LNR and prognosis of ampullary cancer; however, few studies have been reported on its usefulness as a predictor of recurrence.31–33 Although three reports on the significance of the LNR as a predictor of recurrence are available, these reports included patients without lymph node metastases in their analyses. Because the LNR has clinical significance in cases of positive lymph node metastasis, focusing on cases positive for lymph node metastasis is necessary. Our study is the first to examine the relationship between the LNR and recurrence-free survival and OS in patients with ampullary cancer and lymph node metastases. In addition, no reports have compared the clinical significance of UICC N-classification with the LNR.
Considering that the prognostic cut-off value for ampullary cancer ranges from 0.1 to 0.2,15,16,19,24 the calculated cut-off value of 0.07 is consistent with previous reports. Based on the results of the ASCOT trial,10 S-1 would be a standard treatment for adjuvant chemotherapy after curative surgery; however, in cases with an LNR ≥ 0.07, the introduction of more effective regimens would be required to improve the prognosis.
This study has several limitations. First, the data were collected at a single institute and analyzed retrospectively; thus, care should be taken when interpreting these results because of potential biases. In addition, it took more than 30 years to accumulate 106 cases because of the rarity of ampullary cancer. The content of adjuvant chemotherapies changed during that period, but no regimen for ampullary cancer was established, leading to the patients receiving a variety of adjuvant chemotherapies.
In summary, lymph node metastasis was an independent prognostic factor for patients with ampullary cancer, and the LNR was a significant indicator for recurrence and prognosis. These findings have implications for patient selection and stratification in future clinical trials on adjuvant chemotherapy for ampullary cancer.
Additional information