Highlight
This article critically examines three lymph node-related prognostic indicators for duodenal adenocarcinoma — positive lymph nodes (PLN), lymph node ratio (LNR), and log odds of positive lymph nodes (LODDS) — comparing their predictive performance in Western and Eastern patient cohorts. It establishes LODDS, especially in its categorical classification, as the most powerful prognostic marker and presents a novel nomogram integrating LODDS with key clinical factors for robust individualized survival prediction.
Study Background
Duodenal adenocarcinoma is a rare but aggressive malignancy arising in the small intestine’s duodenum. Prognostic assessment in this cancer heavily relies on lymph node (LN) involvement, which directly influences staging, treatment decisions, and survival outcomes. Commonly used LN staging schemes include the absolute count of positive lymph nodes (PLN), lymph node ratio (LNR, representing the ratio of positive to retrieved nodes), and more recently, log odds of positive lymph nodes (LODDS), a metric combining the positive and negative nodes mathematically to account for varying degrees of LN retrieval and positivity.
Despite these approaches, the comparative effectiveness of these LN-based parameters in predicting prognosis for duodenal adenocarcinoma remains undefined. Moreover, geographic and ethnic variability may affect prognostic model applicability. This study addresses these gaps by retrospectively analyzing large Western and Eastern cohorts to identify the most reliable prognostic parameter for LN status and developing a clinically applicable nomogram for individualized survival estimation.
Study Design
This retrospective study utilized two distinct patient cohorts: 1,693 patients from the Surveillance, Epidemiology, and End Results (SEER) database representing a Western population and 295 patients from Wuhan Union Hospital embodying an Eastern cohort. All patients had surgically treated duodenal adenocarcinoma with available lymph node data.
The study evaluated three LN-related prognostic parameters: positive lymph nodes (PLN), lymph node ratio (LNR), and log odds of positive lymph nodes (LODDS), each assessed both as continuous and categorical variables. Seven established evaluation metrics were applied to gauge prognostic performance, including Harrell’s c-index and area under the receiver operating characteristic curve (AUC) for 1-, 3-, and 5-year overall survival (OS).
Following identification of the optimal LN parameter, a multivariate Cox regression model incorporating this variable with other independent prognostic factors—age, T-stage, number of retrieved lymph nodes, and adjuvant chemotherapy—was constructed to develop a prognostic nomogram. External validation was performed using the Wuhan cohort to assess model generalizability.
Key Findings
Across both cohorts, standardized continuous LODDS consistently outperformed PLN and LNR in prognostic discrimination. Specifically, in the SEER cohort, continuous LODDS showed a c-index of 0.639 and a 5-year AUC of 0.685, while in the Wuhan cohort, it yielded a c-index of 0.679 and 5-year AUC of 0.697. These metrics demonstrate superior predictive accuracy compared with PLN and LNR.
Categorical LODDS classification also showed robust prognostic capacity: c-index 0.619 and 5-year AUC 0.666 in SEER, and c-index 0.637 and 5-year AUC 0.640 in Wuhan, outperforming analogous categories of PLN and LNR.
The developed nomogram included LODDS classification, patient age, T-stage, number of retrieved lymph nodes, positive lymph nodes count, and chemotherapy status. It achieved areas under the curve of 0.735, 0.708, and 0.698 at 1-, 3-, and 5-year OS prediction in SEER data, and 0.734, 0.715, and 0.691 respectively in the Wuhan cohort validation.
Calibration plots revealed excellent concordance between predicted and observed survival probabilities, and decision curve analysis demonstrated clear clinical benefit of nomogram application over traditional staging. Moreover, an accessible online tool has been integrated to facilitate clinical use of this nomogram, promoting personalized patient prognostication.
Expert Commentary
This study offers compelling evidence favoring LODDS classification as the optimal LN-related prognostic marker in duodenal adenocarcinoma over conventional PLN and LNR metrics. LODDS’ mathematical integration of positive and negative nodes stabilizes prognostic assessment against variability in surgical LN retrieval, addressing a key limitation of PLN and LNR.
The inclusion of both Western and Eastern cohorts enhances the representativeness and potential global applicability of findings. The validated nomogram incorporating LODDS and clinicopathological factors advances beyond traditional TNM staging by enabling nuanced, individualized survival predictions that can inform clinical decision making.
Limitations include the retrospective design and potential selection bias inherent in hospital registry and SEER databases, as well as lack of detailed data on adjuvant treatment regimens and molecular tumor characteristics. Future prospective studies should further assess LODDS utility and explore integration with emerging biomarkers.
Conclusion
This comparative analysis confirms the log odds of positive lymph nodes classification as the most powerful lymph node-related prognostic parameter for duodenal adenocarcinoma in diverse populations. The deployed nomogram synthesizes this parameter with clinical and treatment variables to provide a robust, externally validated tool for precise, individualized survival prediction. Its adoption in clinical practice can enhance prognosis accuracy and ultimately support personalized therapeutic strategies in this challenging malignancy.
Funding and ClinicalTrials.gov
The original study did not report specific funding sources or clinical trial registration numbers.
References
1. Jiang Z, Dai J, Li W, Deng S, Gu J, Wang J, Chen M, Liu K, Wu K, Cao Y, Cai K. Comparative performance of 3 lymph node staging schemes in duodenal adenocarcinoma: A retrospective analysis of population-based and real-world data. Surgery. 2026 May 11;195:110201. PMID: 42114234.
2. Strong VE, Karpeh MS. Lymph node ratios in gastrointestinal cancer staging. Curr Opin Oncol. 2010;22(4):415-422.
3. Persiani R, Biondi A, Fico V, et al. Log odds of positive lymph nodes as an independent prognostic factor in colorectal cancer patients during surgical exploration. Surgery. 2012;152(1):16-24.
4. Valsecchi ME, Hasbini SA, Lee JH, et al. Prognostic nomograms in gastrointestinal cancers: Applications and limitations. J Gastrointest Surg. 2020;24(9):2009-2019.
