The prognostic value of the lymphocyte-to-monocyte ratio for high-risk papillary thyroid carcinoma

Figures & data

Figure 1 Receiver operator characteristic curve analysis to predict overall survival for high-risk PTCs using the LMR in the pooled analysis. (A) The preoperative LMR can be used to predict overall survival (AUC=0.654, P=0.004); (B) the preoperative LMR=4.0 is the best cutoff to predict overall survival (AUC=0.637, P=0.010).

Table 1 Relationships between the preoperative LMR and clinical or tumor features in three sets of PTC patients with a high risk of recurrence. A lowser preoperative LMR correlated with a larger tumor size, advanced N stage and M stage, more multiple PTC cases, and more cases with the largest nodule larger than 3 cm in the 3 sets

Factor	Training set			Testing set			Validation set
	Preoperative LMR			Preoperative LMR			Preoperative LMR
	Low (<4)	High (≥4)	P	Low (<4)	High (≥4)	P	Low (<4)	High (≥4)	P
All patients	19	93		19	93		14	34
Age (years)			0.728			0.566			0.476
<45	10	53		12	65		9	18
≥45	9	40		7	28		5	16	0.096
Age (years)			0.188			0.540		33	0.096
<55	18	77		18	84		12	33
≥55	1	16		1	9		2	1
Sex			0.552			0.742			0.205
Female	14	31		14	65		12	23
Male	5	62		5	28		2	11
BMI (kg/m^2)			0.059			0.358			0.195
<24	16	50		13	53		12	22
≥24	3	43		6	40		2	12
Largest tumor size			0.002*			0.014			0.017
<4 cm	11	83		12	87		8	29
≥4 cm	8	10		7	6		6	5
Total tumor size			0.001*			<0.001			0.037
<4 cm	8	74		10	82		7	26
≥4 cm	11	19		9	11		7	8
Intra-thyroid PTC			0.001*			<0.001			0.045
Single	7	69		10	74		8	31
Multiple	12	24		9	19		6	3
T classification			0.258			0.112			0.374
T1–T3	4	32		3	31		6	10
T4	15	61		16	62		8	24
N classification			0.011*			<0.001*			0.048*
N0/N1a	1	27		1	20		0	8
N1b	18	66		18	73		14	26
Distant metastasis			0.003*			0.024*			0.007*
No	11	75		11	76		8	28
Yes	8	18		8	17		6	6
LNM number			0.210			0.009*			0.102
<15	10	63		8	68		3	16
≥15	9	30		11	25		11	18
Largest LNM diameter			0.048*			0.033*			0.023*
<3 cm	11	71		11	75		7	28
≥3 cm	8	22		8	18		7	6

Notes: *P<0.05.

Abbreviations: BMI, body mass index; PTC, papillary thyroid carcinoma; T, tumor invasion depth; N, lymph node involvement; TNM, tumor node metastasis; LNM, lymph node metastasis.

Figure 2 Kaplan-Meier analysis of overall survival in high-risk PTCs according to the LMR in 3 independent sets. (A) Training set (P=0.049); (B) testing set (P=0.039); (C) validation set (P=0.029); (D) combined set (P<0.001).

Figure 3 Kaplan-Meier analysis of PTC-free survival in high-risk PTCs according to the LMR in 3 independent sets. (A) Training set (P=0.042); (B) testing set (P=0.014); (C) validation set (P=0.045); (D) combined set (P<0.001).

Table 2 Univariate Cox regression analyses of overall survival in three sets of high-risk PTC patients. The largest tumor size, total tumor size, distant metastasis, LNM number, and preoperative LMR in the training set were significant. In contrast, only distant metastasis and the preoperative LMR were risk factors for OS in the testing set, and the largest tumor size, total tumor size, distant metastasis, and preoperative LMR were significant in the validation set

Factors	Overall survival
	Training set		Testing set		Validation set
	HR (95% CI)	P	HR (95% CI)	P	HR (95% CI)	P
Age (years)		0.126		0.713		0.182
≤55	1 (reference)		1 (reference)		1 (reference)
>55	2.520 (0.772 to 8.230)		0.683 (0.090 to 5.200)		4.331 (0.503 to 37.260)
Gender		0.204		0.743		0.828
Female	1 (reference)		1 (reference)		1 (reference)
Male	2.658 (0.589 to 11.994)		0.836 (0.286 to 2.445)		0.828 (0.151 to 4.530)
BMI (kg/m2)		0.711		0.965		0.590
<24	1 (reference)		1 (reference)		1 (reference)
≥24	1.229 (0.413 to 3.660)		0.997 (0.348 to 2.746)		0.627 (0.115 to 3.426)
Largest tumor size		0.037*		0.297		0.014*
<4 cm	1 (reference)		1 (reference)		1 (reference)
≥4 cm	2.280 (1.072 to 10.034)		1.962 (0.553 to 6.954)		8.439(1.542 to 46.201)
Total tumor size		0.025*		0.882		0.043*
<4 cm	1 (reference)		1 (reference)		1 (reference)
≥4 cm	3.490 (1.171 to 10.400)		1.100 (0.310 to 3.903)		5.001 (1.915 to 27.338)
Intra-thyroid PTC		0.471		0.192		0.743
Single	1 (reference)		1 (reference)		1 (reference)
Multiple	1.673 (0.561 to 4.992)		1.991 (0.708 to 5.600)		1.688 (0.800 to 5.907)
T classification		0.198		0.329		0.949
T1-T3	1 (reference)		1 (reference)		1 (reference)
T4	2.691 (0.596 to 12.157)		1.878 (0.530 to 6.657)		1.057 (0.193 to5.780)
N classification		0.181		0.590		0.509
N0/N1a	1 (reference)		1 (reference)		1 (reference)
N1b	4.027 (0.523 to 31.033)		1.505 (0.340 to 6.670)		25.964 (0.002 to 40.728)
Distant metastasis		0.039*		<0.001*		0.028*
No	1 (reference)		1 (reference)		1 (reference)
Yes	3.156 (1.5058 to 9.414)		13.537 (4.271 to 42.906)		6.759 (1.236 to 36.964)
LNM number		0.003*		0.166		0.232
<15	1 (reference)		1 (reference)		1 (reference)
≥15	7.292 (2.003 to 26.541)		2.049 (0.742 to 5.655)		3.707 (0.433 to 31.749)
Largest LNM diameter		0.171		0.640		0.663
<3 cm	1 (reference)		1 (reference)		1 (reference)
≥3 cm	2.144 (0.719 to 6.391)		1.314(0.418 to 4.129)		1.459(0.267 to 7.972)
Preoperative LMR		0.031*		0.049*		0.042*
Low	1 (reference)		1 (reference)		1 (reference)
High	0.340 (0.110 to 0.912)		0.340 (0.116 to 0.997)		0.185 (0.034 to 0.912)

Notes: *P<0.05.

Abbreviations: HR, hazard ratio; 95% CI, 95% confidence interval; T, tumor invasion depth; N, lymph node involvement; LNM, lymph node metastasis.

Table 3 Multivariate Cox regression analyses of overall survival in the training, testing, validation, and combined sets. The largest tumor size, distant metastasis, and preoperative LMR were identified as independent prognostic factors associated with overall survival in the 3 sets and in the pooled analysis

Sets Variables		Training Set HR 95% CI P-value			Testing Set HR 95% CI P-value			Validation Set HR 95% CI P-value			Combined Set HR 95% CI P-value
Largest tumor size (<4/≥4 cm)	2.231	1.114–4.251	0.044*	1.688	1.281–2.491	0.032*	2.469	1.071–3.112	0.048*	2.998	1.181–13.606	0.045*
Total tumor size (<4/≥4 cm)	3.513	1.622–19.828	0.045*	3.338	0.810–28.118	0.936	2.929	0.923–4.995	0.299	3.556	0.127–12.427	0.435
Distant metastasis (No/Yes)	3.329	1.027–10.792	0.045*	13.918	4.089–47.378	<0.001*	6.177	1.001–38.122	0.050	5.386	2.631–11.027	<0.001*
LNM number (<15/>15)	6.645	1.770–24.954	0.005*	4.782	0.862–12.339	0.126	3.230	0.898–15.408	0.087	2.234	0.865–4.686	0.033*
Preoperative LMR (High/low)	3.714	1.724–12.441	0.018*	4.457	1.140–14.911	0.019*	2.469	1.071–3.112	0.048*	2.644	1.290–11.432	0.020*

Notes: *P<0.05.

Figure 4 Nomogram for predicting high-risk PTC patient mortality (A) and PTC recurrence (B) for individual high-risk PTC patients. We can calculate the risk of early recurrence as follows. After obtaining the risk factor data, each risk factor can be used to calculate the number of points. Then, the points are added to obtain a total value. On the total-points axis, a line is drawn down to the risk axis. Then, the risk probabilities of mortality (A) and PTC recurrence (B) can be calculated for individual patients.