SUN-342 The Role Of Calcitonin In Predicting Lateral Lymph Node Metastasis In Medullary Thyroid Cancer

Abstract Disclosure: K. Madani: None. M.Y. Roth: None. M. Endo: None. T.S. Tylee: None. Introduction: The 2015 American Thyroid Association (ATA) guidelines suggest using pre-operative serum calcitonin levels to consider prophylactic lateral lymph node dissection (LND) in medullary thyroid cancer (MTC). However, the ATA Task Force did not reach a consensus on this recommendation; hence classified it as a Grade I recommendation. Although no clear cutoff has been mentioned, baseline calcitonin (bCT) cutoffs of 20 and 200 (pg/mL) have been associated with lateral lymph node metastasis (LNM) in a previous study cited in ATA (Machens et al. 2010) and 500 pg/mL as mentioned in revised ATA guidelines (Wells et al. 2015). This study tested whether bCT alone should inform the decision for prophylactic lateral LND. Hypothesis: No baseline calcitonin threshold achieves both 100% sensitivity and a clinically acceptable positive predictive value (PPV) of >80% for lateral LNM. Experimental Design: We performed a retrospective chart review of 62 patients with MTC confirmed on surgical pathology (50% male; 46.8% Caucasian). Baseline calcitonin (bCT) levels were log-transformed to normalize their skewed distribution. We created two groups from this data, one with lateral LNM and the other without lateral LNM based on pathology results. Analytical Methods: Statistical analysis was performed using R Studio. Welch’s two-sample t-test and ROC curve analysis were used to evaluate the effectiveness of baseline calcitonin levels in distinguishing between the two groups. Sensitivity and positive predictive value (PPV) were calculated for bCT thresholds yielding 100% sensitivity (17.7 pg/mL) and for proposed cutoffs of 20, 200, and 500 pg/mL. Results: Patients with lateral LNM had significantly higher mean bCT (2552 vs. 226 pg/mL, p<0.05). The AUC of 0.839 (95% CI: 0.732-0.945) reflects moderate predictive accuracy. However, the bCT threshold yielding 100 % sensitivity (17.7 pg/mL) had a PPV of only 57.1%, meaning nearly half of the patients without metastasis would undergo unnecessary surgery. PPV increases to just 57.1%, 64.6%, and 70.7% at the 20, 200, and 500 pg/mL thresholds, respectively - improvements still insufficient to justify prophylactic lateral LND. Conclusion: Although baseline calcitonin is a good predictor of lateral LNM, its low PPV at 100% sensitivity renders it an inadequate standalone marker for guiding prophylactic lateral neck dissection, as it risks unnecessary operative morbidity. We recommend against prophylactic lateral LND based solely on calcitonin thresholds. Instead, lateral dissection should be reserved for cases with radiologically suspicious nodes. Presentation: Sunday, July 13, 2025