The precursor for nerve growth factor (proNGF) is not a serum or biopsy-rinse biomarker for thyroid cancer diagnosis

Table 3 Biopsy rinse proNGF levels, stratified by nodule diagnosis

Category	Biopsy-rinse proNGF (dichotomised)		Biopsy-rinse proNGF (continuous)		Multiple logistic regression*
Category	n > 0.05 ng/mL/n in group	p-value	ng/mL Median (IQR)	p-value	OR (95% CI)	p-value
Overall	73/188 (39%)		0 (0–0.12)
By malignancy status		0.41		0.11	1.3 (0.5–3.1)	0.57
-Thyroid cancer	12/26 (46%)		0 (0–0.20)
-Benign nodule	61/162 (38%)		0 (0–0.10)
By follicular lesion		0.01		0.002	3.3 (1.2–8.7)	0.02
-Present	13/20 (65%)		0.12 (0–0.31)
-Absent	60/168 (36%)		0 (0–0.11)
By age		0.61		0.82	1.0 (1.0–1.0)	0.87
-Age < 55	39/96 (41%)		0 (0–0.11)
-Age ≥ 55	34/92 (37%)		0 (0–0.13)
By sex		0.35		0.33	0.7 (0.2–2.3)	0.35
-Female	57/153 (37%)		0 (0–0.11)
-Male	16/35 (46%)		0 (0–0.17)

Categorical variables evaluated with Pearson’s Chi-square, and continuous variables with the Wilcoxon RankSum test. *Binary outcome variable is proNGF > 0.05 ng/mL, adjusting for age (continuous) presence of cancer, presence of follicular lesion and female sex

ISSN: 1472-6823