The systemic immune-inflammation index-based model is an effective biomarker on predicting central lymph node metastasis in clinically nodal-negative papillary thyroid carcinoma

Background: A high proportion of papillary thyroid carcinoma (PTC) patients are involved in central lymph node metastasis (CLNM) with preoperative imaging examinations showing clinically nodal-negative (cN0). Meanwhile, many inflammatory biomarkers are also proven as effective factors to predict the outcomes of cancer patients and tumor progression. Thus, the values of these factors are investigated to help detecting CLNM in cN0 PTC patients. Methods: 406 cN0 PTC patients who underwent curative surgery were retrospectively analyzed. CLNM was determined by histopathological examination following the thyroidectomy. Multiple inflammatory biomarkers were comprehensively researched. Results: A total of 406 consecutive patients were eventually included. The univariate and multivariate analyses revealed that age (OR: 0.924, 95% CI: 0.909, 0.940), gender (OR: 1.781, 95% CI: 1.060, 2.993), location of tumors (OR: 2.229, 95% CI: 1.228, 4.046) and level of systemic immune-inflammation index (SII) (OR: 1.005, 95% CI: 1.004, 1.006) were independently associated with CLNM in cN0 PTC patients, and the SII-based predictive model was constructed using these four factors. Receiver operating characteristic (ROC) curves showed significant results of the SII-based predictive model in PTC cohort with area under curve (AUC) as 0.814 (95% CI: 0.771-0.857) and in PTMC subgroup with AUC as 0.803 (95% CI: 0.752-0.854). Conclusions: The SII-based model can effectively help predicting CLNM in both cN0 PTC patients and cN0 PTMC patients.

Automated summary

A high proportion of PTC patients have central lymph node metastasis (CLNM) despite preoperative imaging showing clinically nodal-negative (cN0). Many inflammatory biomarkers can predict cancer outcomes, and a study found age, gender, tumor location, and systemic immune-inflammation index (SII) were independently associated with CLNM in cN0 PTC patients. An SII-based predictive model showed significant results in predicting CLNM.