KR20200141166A - Method for providing the information for prediction of lymph node metastasis of gastric cancer - Google Patents

Abstract

The present invention relates to a method of providing information for predicting lymph node metastasis of gastric cancer, and more particularly, to a method of providing information for predicting lymph node metastasis of gastric cancer by combining a maximum standard uptake coefficient of primary cancer (T_SUVmax) and a maximum standard uptake coefficient of lymph nodes (N_SUVmax) with hematological information of a gastric cancer patient in fluoro-2-deoxyglucose (F-18 FDG) positron emission tomography/computed tomography (PET/CT). In the present invention, it was found that the maximum standard uptake coefficient of primary cancer (T_SUVmax) and the maximum standard uptake coefficient of lymph nodes (N_SUVmax) obtained from F-18 FDG PET/CT before surgery and a tumor marker CA19-9 among hematological indicators and albumin are statistically significant related factors in predicting the lymph node metastasis in multivariate logistic regression analysis. Finally, it was found that an area under curve (AUC) of a model created with four parameters T_SUVmax, N_SUVmax, CA19-9, and albumin is 0.780, which is a most suitable model for predicting the lymph node metastasis of the gastric cancer. Accordingly, the above four parameters are used to effectively provide information on whether the gastric cancer has metastasized to the lymph nodes.

Description

Method for providing the information for prediction of lymph node metastasis of gastric cancer

The present invention relates to a method of providing information for predicting lymph node metastasis of gastric cancer, and in more detail, the maximum standard intake of primary cancer in F-18 FDG (fluoro-2-deoxyglucose) PET/CT (positron emission tomography/computed tomography) The present invention relates to a method of providing information for predicting lymph node metastasis of gastric cancer by combining the coefficient (T_SUVmax) and the maximum standard intake factor (N_SUVmax) of the lymph node and hematological information of gastric cancer patients.

Gastric cancer is the cancer with the highest incidence in Korea. As the recent early cancer screening program has been implemented, cases of early diagnosis of gastric cancer are increasing, and if there is no lymph node metastasis among early gastric cancers, endoscopic resection can be used to cure it, but since lymph node metastasis cannot be completely predicted, endoscopic resection is performed. There is a limit to screening the target patients of. In addition, the 5-year survival rate of patients with lymph node metastasis after gastric cancer surgery ranges from 5.9% to 58.1%, depending on the degree of lymph node metastasis, whereas the 5-year survival rate of patients without lymph node metastasis is 86.1%, and lymph node metastasis determines the prognosis of gastric cancer patients. It is also an important element to build.

Various imaging methods are used to determine the preoperative staging of gastric cancer, and recently, F-18 FDG (fluoro-2-deoxyglucose) PET/CT (positron emission tomography/computed tomography) has been used to determine the staging, response evaluation, and It has been widely used for predicting recurrence (Malibari N, et al ., PET Clin. , 2015, 10:311-26; Park S, et al ., J Nucl Med ., 2017, 58:899-904). However, the diagnosis of lymph node metastasis alone, such as F-18 FDG PET/CT or abdominal CT using a contrast agent, has been reported in various ways with a sensitivity of approximately 34.3% to 80.8% and a specificity of 75.0% to 92.2%. Therefore, determining lymph node metastasis of gastric cancer alone is a problem due to its relatively low sensitivity, and development of a new diagnostic method that can accurately predict lymph node metastasis is required.

Accordingly, the present inventors have made diligent efforts to increase the sensitivity and accuracy of predicting lymph node metastasis of gastric cancer by combining data obtained from F-18 FDG PET/CT and clinical information of gastric cancer patients. Maximum standard intake coefficient (T_SUVmax) and maximum standard intake coefficient (N_SUVmax) of lymph nodes obtained from /CT, hemoglobin (Hgb) among hematologic indicators, and the ratio of neutrophils to lymphocytes (Neutrophil to lymphocyte ratio; NLR) ), platelet to lymphocyte ratio (PLR), tumor markers CA19-9 (carbohydrate antigen 19-9) and albumin were statistically significant in predicting lymph node metastasis using univariate logistic regression analysis. It was confirmed that it is a related factor, and finally, through multivariate logistic regression analysis, the area under the curve (AUC) of the model made with four parameters T_SUVmax, N_SUVmax, CA 19-9 and Albumin was 0.780. It was confirmed that it was the most suitable model for predicting lymph node metastasis, and the present invention was completed.

It is an object of the present invention to provide a method of providing information for predicting lymph node metastasis of gastric cancer comprising the step of combining and analyzing F-18 FDG PET/CT data and hematologic indicator information of gastric cancer patients.

To achieve the above object,

The present invention provides prediction of lymph node metastasis of gastric cancer comprising obtaining T_SUVmax and N_SUVmax from F-18 FDG PET/CT of gastric cancer patients and predicting the possibility of lymph node metastasis by measuring CA19-9 and Albumin concentrations in the blood of gastric cancer patients. Provides information provision method for

In a preferred embodiment of the present invention, the possibility of lymph node metastasis (P) can be derived using Equation 1 and Equation 2 below.

[Equation 1]

X = ln(P/1-P) = 0.1066 × (T_SUVmax) + 0.3848 × (N_SUVmax) + 0.0025 × (CA 19-9; U/mL)-0.6632 × (Albumin; g/dL) + 1.5735

[Equation 2]

P = exp^X / (1 + exp^X)

In addition, the present invention (a) after statistical analysis by combining the T_SUVmax and N_SUVmax measured in F-18 FDG PET/CT of gastric cancer patients and CA19-9 and Albumin concentrations in the blood of gastric cancer patients, Deriving grams;

(b) The presence of lymph node metastasis by applying T_SUVmax, N_SUVmax and blood CA19-9 and Albumin concentrations measured in F-18 FDG PET/CT of gastric cancer patients whose lymph node metastasis has not been confirmed to the nomogram of step (b) above. Generating a score associated with the prediction;

(c) providing information that the higher the score is, the higher the probability of lymph node metastasis is. It provides a method of providing information for predicting lymph node metastasis of gastric cancer.

In a preferred embodiment of the present invention, the group of gastric cancer patients in step (a) may be a group that performed F-18 FDG PET/CT imaging and hematological examination before surgical resection.

In another preferred embodiment of the present invention, the score table according to the nomogram of step (b) may be as follows.

The present invention relates to a method of providing information for predicting lymph node metastasis of gastric cancer. In the present invention, T_SUVmax and N_SUVmax obtained from F-18 FDG PET/CT of gastric cancer patients and of hemoglobin (Hgb) and neutrophils among hematologic indicators The ratio of lymphocytes (Neutrophil to lymphocyte ratio; NLR), platelet to lymphocyte ratio (PLR), tumor markers CA19-9, and albumin (Albumin) in predicting lymph node metastasis in univariate logistic regression analysis. It was confirmed that it was a statistically significant association factor, and finally, the area under the curve (AUC) of the model made with the four parameters T_SUVmax, N_SUVmax, CA19-9 and Albumin was 0.780 through multivariate logistic regression analysis. Since it has been confirmed that it is the most suitable model for predicting lymph node metastasis of gastric cancer, it can effectively provide information on whether or not lymph node metastasis of gastric cancer is present.

1 is a graph showing an ROC curve for lymph node metastasis diagnosis using a general N_SUVmax in F-18 FDG PET/CT.
Figure 2 is a graph showing the ROC curve for the lymph node metastasis prediction model made with four parameters T_SUVmax, N_SUVmax, CA19-9, Albumin.
3 is a schematic diagram showing a nomogram for predicting lymph node metastasis of gastric cancer constructed with four parameters of T_SUVmax, N_SUVmax, CA19-9, and Albumin.

Hereinafter, the present invention will be described in detail.

In one aspect, the present invention is to obtain T_SUVmax and N_SUVmax from F-18 FDG PET/CT of a gastric cancer patient and measure the CA19-9 and Albumin concentration in the blood of a gastric cancer patient to predict the possibility of lymph node metastasis. It relates to a method of providing information for predicting lymph node metastasis.

In the present invention, the lymph node metastasis possibility (P) may be characterized in that it is derived using Equations 1 and 2 below.

[Equation 1]

X = ln(P/1-P) = 0.1066 × (T_SUVmax) + 0.3848 × (N_SUVmax) + 0.0025 × (CA19-9; U/mL)-0.6632 × (Albumin; g/dL) + 1.5735

[Equation 2]

P = exp^X / (1 + exp^X)

In the present invention, F-18 FDG PET/CT was performed before surgical resection for gastric cancer, and a group of 538 gastric cancer patients who underwent hematologic examination and radical resection were analyzed, and the clinical information of the patient is shown in Table 1. Shown in. Using the preoperative F-18 FDG PET/CT and hematological information of the gastric cancer patient group, we tried to find clinical indicators showing a significant relationship with lymph node metastasis.

In F-18 FDG PET/CT, T_SUVmax and N_SUVmax were obtained and used for analysis, and as hematological information, leukocyte (WBC), hemoglobin (Hgb), and neutrophil to lymphocyte ratio (NLR) , Platelet to lymphocyte ratio (PLR), tumor markers, carcinoembryonic antigen (CEA), CA19-9, Albumin, and C-reactive protein (CRP) were used for analysis.

As a result of analysis using univariate logistic regression, it was confirmed that Hgb, NLR, PLR, CA19-9, and Albumin concentrations among T_SUVmax, N_SUVmax, and hematological indices were statistically significant predictors of lymph node metastasis. Stepwise backward elimination was performed using the above parameters, and the AIC value (Akaike information criterion; the smaller the value, the more perfect model) of the model made with the four parameters T_SUVmax, N_SUVmax, CA19-9 and Albumin The lowest was confirmed.

There are various ways to select a model, but AIC, one of the model selection methods, is a method in which the model with the least information loss is selected as the most suitable model for data. The AIC evaluates the relative quality of a statistical model for a given data set, and a lower AIC value can be considered a perfect model.

In addition, as shown in Figures 1 and 2, the AUC of the ROC predicting lymph node metastasis with the conventional F-18 FDG PET/CT was 0.640, but the model made with four parameters T_SUVmax, N_SUVmax, CA19-9 and Albumin The AUC of was 0.780, showing a statistically significant improvement in lymph node metastasis diagnosis performance ( P <0.0001).

The ROC curve is one of the widely used methods to compare the accuracy of diagnostic methods (Akobeong, 2007). A graph is drawn with'sensitivity' on the X-axis and '1-specificity' on the Y-axis. In this case, the area under the curve is 0.5 based on the diagonal, and if the ROC curve is above the diagonal, it is predicted as a good model. The AUC has a value between 0.5 and 1, and the closer it is to 1, the wider the curve and the better the predicted model.

That is, it was confirmed that the model made by combining the T_SUVmax and N_SUVmax obtained from F-18 FDG PET/CT and the CA19-9 and Albumin concentrations in the blood as in the present invention is the most suitable model for predicting lymph node metastasis of gastric cancer.

The present invention from another point of view,

(a) Statistical analysis by combining T_SUVmax and N_SUVmax measured by F-18 FDG PET/CT of gastric cancer patients and CA19-9 and Albumin concentrations in the blood of gastric cancer patients, and then deriving a nomogram for predicting lymph node metastasis ;

(b) T_SUVmax and N_SUVmax measured by F-18 FDG PET/CT of gastric cancer patients whose lymph node metastasis was not confirmed, and CA19-9 and Albumin concentrations in the blood of gastric cancer patients were applied to the nomogram of step (a). Generating a score that predicts the likelihood of lymph node metastasis; And

(c) providing information that the higher the score is, the higher the probability of lymph node metastasis is. It provides a method of providing information for predicting lymph node metastasis of gastric cancer.

In the present invention, the gastric cancer patient group in step (a) may be a group that performed F-18 FDG PET/CT imaging and hematological examination before surgical resection.

In the present invention, the score table according to the nomogram of step (a) is as follows.

Nomogram is a graph that predicts the incidence of an individual's disease by scoring each risk factor and calculating the total score to check the corresponding probability. It has the advantage that it is much easier to understand because it is expressed graphically rather than by directly substituting it into a statistical model. The composition of the nomogram consists of a point line that can check the scores of risk factors category, a line of all risk factors that display the scores of each category, a total points line that can check the sum of the scores of each risk factor, and finally, the total points value. It consists of a lymph node metastasis probability (P) line representing the lymph node metastasis probability corresponding to.

In another specific embodiment of the present invention, a nomogram was constructed through multivariate logistic regression, and lymph node metastasis of gastric cancer with four parameters of T_SUVmax, N_SUVmax, CA19-9, and Albumin as shown in FIG. A nomogram score table was constructed for prediction. As a result of the analysis, when the total sum of scores calculated by applying T_SUVmax, N_SUVmax, CA19-9 and Albumin concentration to the nomogram was 52 points, the probability of lymph node metastasis was 95%.

Hereinafter, the present invention will be described in more detail through examples.

These examples are for illustrative purposes only, and it will be apparent to those of ordinary skill in the art that the scope of the present invention is not construed as being limited by these examples.

Patient classification and clinical information

From January 2008 to December 2010, the medical records of 873 patients who underwent surgical treatment for primary gastric cancer at Keimyung University Dongsan Medical Center were retrospectively investigated. Among them, 538 patients underwent F-18 FDG PET/CT and hematologic tests before surgery and underwent surgical radical resection, and the information of these patients was applied to the analysis of the present invention.

In F-18 FDG PET/CT, the maximum standard intake coefficient (T_SUVmax) and the maximum standard intake coefficient (N_SUVmax) of the lymph nodes were obtained and used for analysis. As hematological information, leukocyte (WBC), hemoglobin; Hgb), neutrophil to lymphocyte ratio (NLR), platelet to lymphocyte ratio (PLR), tumor marker CEA (Carcinoembryonic antigen), CA19-9 (carbohydrate antigen 19-9) ), albumin (Albumin) and C-reactive protein (CRP) were used for the analysis. The hematological information was obtained through a venous blood test.

The retrospective study of the present invention was conducted with the approval of the Clinical Research Review Committee of Keimyung University Dongsan Medical Center, and all data were anonymously processed before analysis.

All patients underwent subtotal gastrectomy or total gastrectomy with D2 lymph node dissection (advanced gastric cancer (AGC)) and D1+β or D2 lymph node dissection (early gastric cancer (EGC)).

The patient's clinical pathologic records were examined for sex, age of operation, and pathological T stage. The T and N stages were classified according to the 7th AJCC and TNM stage classification system.

Patient clinical information Lymph node metastasis (-)
( n =330) Lymph node metastasis (+)
( n =208) P Sex 0.290 Male 193 (58.5%) 132 (63.5%) Female 137 (41.5%) 76 (36.5%) Age(yr) 59.3 ± 11.7 60.1 ± 12.5 0.493 WBC 6.3 ± 1.7 6.2 ± 1.7 0.545 Hgb 12.8 ± 6.1 12.1 ± 1.8 0.032 NLR 2.2 ± 1.3 2.5 ± 1.5 0.004 PLR 155.6 ± 64.4 191.6 ± 102.0 <0.001 CEA 4.0 ± 20.6 5.6 ± 16.6 0.307 Albumin 4.1 ± 0.3 3.9 ± 0.3 <0.001 CA19-9 15.0 ± 49.9 74.4 ± 325.2 0.010 T stag <0.001 T1 257 (77.9%) 38 (18.3%) T2 31 (9.4%) 39 (18.8%) T3 26 (7.9%) 43 (20.2%) T4 16 (4.8%) 89 (42.8%) T_SUVmax 2.6 ± 4.1 5.9 ± 5.4 <0.001 N_SUVmax 0.1 ± 0.8 1.4 ± 3.3 <0.001

Data were expressed as mean ± standard deviation, and categorical variables were expressed in order and percentage.

Of the total 538 patients, 210 (38.7%) had lymph node metastasis, the average T_SUVmax and N_SUVmax values of patients without lymph node metastasis were 2.6 ± 4.1 and 0.1 ± 0.8, respectively, and the average T_SUVmax and N_SUVmax values of patients with lymph node metastasis were respectively. It was identified as 5.9 ± 5.4 and 1.4 ± 3.3. There were 325 male patients (60.4%), and the mean age at diagnosis of gastric cancer was 59.6 ± 12.0 years. Of the 538 patients, 295 (54.9%) had pathological T stage 1, 70 (13.0%) had stage 2, 68 (12.6%) had stage 3 and 105 (19.5%) had stage 4.

F-18 FDG PET/CT imaging and image analysis

Before F-18 FDG injection, all patients fasted for at least 6 hours, and blood glucose levels were managed at less than 150 mg/dL. The patient was allowed to rest for 1 hour after F-18 FDG injection, and PET/CT was taken 1 hour after F-18 FDG injection.

F-18 FDG PET/CT imaging was performed using an integrated PET/CT system ((Discovery STe; GE Healthcare, Milwaukee, WI, USA). First, a low-volume CT scan (peak voltage 120 kVp, automated tube current in the range of 60 to 150 mA, thickness 3.75 mm) was obtained for attenuation correction. Immediately after the CT scan, PET images were obtained in 3D mode with an acquisition time of 3 minutes per bed position. PET images were reconstructed using an iterative reconstruction algorithm of an iterative method (Ordered Subset Expectation Maximization, OSEM).

F-18 FDG PET/CT images were analyzed and analyzed by two experienced nuclear medicine doctors at Advantage Workstation 4.3 (GE Healthcare).

First, all F-18 FDG PET/CT images were visually evaluated and classified as positive or negative in relation to F-18 FDG intake in primary cancer or lymph nodes. Primary tumors were classified as benign when ingestion exceeded physiological absorption by the surrounding gastric wall and cancerous lesions corresponding to endoscopy, and F-18 FDG intake did not increase significantly or distinguished from physiological gastric wall absorption. This impossible intake was defined as negative.

In addition, even if the F-18 FDG intake lesions were locally increased, it was evaluated as negative if they were not consistent with the cancer lesions in endoscopy and histopathological findings.

In the case of lymph nodes, lymph nodes showing localized F-18 FDG uptake on F-18 FDG PET/CT were judged positive regardless of the size of the CT.

T_SUVmax and N_SUVmax were then obtained from positive F-18 FDG ingested primary cancer lesions and lymph node lesions for semi-quantitative analysis. The regions of interest obtained were drawn manually over benign primary carcinoma lesions and maximal F-18 FDG uptake lymph node lesions on F-18 FDG PET images.

SUVmax was calculated using Equation 3 below.

[Equation 3]

SUVmax = maximum activity in the region of interest (MBq/g) / [injection dose (MBq)/weight (g)]

Selection of predictors of lymph node metastasis in gastric cancer

In the present invention, the maximum standard intake coefficient (T_SUVmax) of primary cancer and the maximum standard intake coefficient of lymph nodes (N_SUVmax), leukocytes in blood (WBC), hemoglobin (Hgb), neutrophil/lymphocyte ratio (NLR), platelet/lymphocyte ratio (PLR) ), CEA (Carcinoembryonic antigen), CA19-9 (carbohydrate antigen 19-9), albumin (Albumin), and C-reactive protein (CRP) concentrations to select factors predicting lymph node metastasis in gastric cancer using logistic regression analysis The statistical analysis performed in the present invention was MedCalc software (MedCalc for Windows, version 18.6; MedCalc Software, Belgium) and R version 3.4.3 software (R version 3.4.3 software; http://www.r-project. org, R Foundation for Statistical Computing, Austria). All P values less than 0.05 ( P <0.05) were considered statistically significant.

As a result, it was confirmed that T_SUVmax, N_SUVmax, Hgb, NLR, PLR, CA19-9, and Albumin were statistically significant related factors in single-quantity logistic regression analysis.

Using the above parameters, multivariate logistic regression analysis was performed by stepwise backward elimination.

AIC value for each parameter combination Df Deviance AIC T_SUVmax + N_SUVmax + HGB + NLR + PLR + CA19-9 + Albumin HGB One 619.37 633.37 T_SUVmax + N_SUVmax + HGB + NLR + PLR + CA19-9 + Albumin NLR One 619.84 633.84 T_SUVmax + N_SUVmax + HGB + NLR + PLR + CA19-9 + Albumin PLR One 621.23 635.23 T_SUVmax + N_SUVmax + HGB + NLR + PLR + CA19-9 + Albumin <none> 619.29 635.29 T_SUVmax + N_SUVmax + HGB + NLR + PLR + CA19-9 + Albumin Albumin One 622.15 636.15 T_SUVmax + N_SUVmax + HGB + NLR + PLR + CA19-9 + Albumin CA19-9 One 624.70 638.70 T_SUVmax + N_SUVmax + HGB + NLR + PLR + CA19-9 + Albumin N_SUVmax One 635.64 649.64 T_SUVmax + N_SUVmax + HGB + NLR + PLR + CA19-9 + Albumin T_SUVmax One 636.57 650.57 T_SUVmax + N_SUVmax + HGB + NLR + PLR + CA19-9 + Albumin AIC=635.29 T_SUVmax + N_SUVmax + NLR + PLR + CA19-9 + Albumin NLR One 619.92 631.92 T_SUVmax + N_SUVmax + NLR + PLR + CA19-9 + Albumin <none> 619.37 633.37 T_SUVmax + N_SUVmax + NLR + PLR + CA19-9 + Albumin PLR One 621.40 633.40 T_SUVmax + N_SUVmax + NLR + PLR + CA19-9 + Albumin Albumin One 622.43 634.43 T_SUVmax + N_SUVmax + NLR + PLR + CA19-9 + Albumin CA19-9 One 624.76 636.76 T_SUVmax + N_SUVmax + NLR + PLR + CA19-9 + Albumin N_SUVmax One 635.70 647.70 T_SUVmax + N_SUVmax + NLR + PLR + CA19-9 + Albumin T_SUVmax One 636.86 648.86 T_SUVmax + N_SUVmax + NLR + PLR + CA19-9 + Albumin AIC=633.37 T_SUVmax + N_SUVmax + PLR + CA19-9 + Albumin PLR One 621.44 631.44 T_SUVmax + N_SUVmax + PLR + CA19-9 + Albumin <none> 619.92 631.92 T_SUVmax + N_SUVmax + PLR + CA19-9 + Albumin Albumin One 622.95 632.95 T_SUVmax + N_SUVmax + PLR + CA19-9 + Albumin CA19-9 One 625.21 635.21 T_SUVmax + N_SUVmax + PLR + CA19-9 + Albumin N_SUVmax One 636.52 646.52 T_SUVmax + N_SUVmax + PLR + CA19-9 + Albumin T_SUVmax One 637.16 647.16 T_SUVmax + N_SUVmax + PLR + CA19-9 + Albumin AIC=631.92 T_SUVmax + N_SUVmax + CA19-9 + Albumin <none> 621.44 631.44 T_SUVmax + N_SUVmax + CA19-9 + Albumin Albumin One 625.56 633.56 T_SUVmax + N_SUVmax + CA19-9 + Albumin CA19-9 One 626.90 634.90 T_SUVmax + N_SUVmax + CA19-9 + Albumin N_SUVmax One 639.73 647.73 T_SUVmax + N_SUVmax + CA19-9 + Albumin T_SUVmax One 641.22 649.22 T_SUVmax + N_SUVmax + CA19-9 + Albumin AIC=631.44

As a result, it was confirmed that the AIC value (Akaike information criterion; the smaller the value, the perfect model) of the model made with the four parameters T_SUVmax, N_SUVmax, CA19-9 and Albumin was the lowest.

As shown in FIG. 1, the AUC of the ROC predicting lymph node metastasis with the conventional F-18 FDG PET/CT was 0.640, but as shown in FIG. 2, a model made with four parameters T_SUVmax, N_SUVmax, CA19-9 and Albumin The AUC of was 0.780, which showed a statistically significant improvement in diagnostic performance ( P <0.0001).

In addition, as a result of the analysis, the probability of lymph node metastasis (P) calculated by T_SUVmax, N_SUVmax, CA19-9, and Albumin was calculated using Equations 1 and 2 below.

[Equation 1]

X = ln(P/1-P) = 0.1066 × (T_SUVmax) + 0.3848 × (N_SUVmax) + 0.0025 × (CA19-9; U/mL)-0.6632 × (Albumin; g/dL) + 1.5735

[Equation 2]

P = exp^X / (1 + exp^X)

Construction of nomogram to predict lymph node metastasis in gastric cancer

In the present invention, it was attempted to construct a nomogram for predicting lymph node metastasis of gastric cancer using four parameters: T_SUVmax, N_SUVmax, CA19-9, and Albumin.

Nomograms were generated using the Cox proportional hazards model (Shariat SF et al., Clinical Cancer Research , 2006, 12:6663-76; Iasonos A et al , Journal of Clinical Oncology , 2008, 26:1364-70) R version 3.4. .3 It was derived using the'rms package' of the software.

As a result, as shown in FIG. 3, a nomogram score table was designed for predicting lymph node metastasis of gastric cancer using the four parameters T_SUVmax, N_SUVmax, CA19-9, and Albumin. If the total score calculated by applying T_SUVmax, N_SUVmax, CA19-9 and Albumin concentration to the nomogram was 52 points, the probability of lymph node metastasis was 95%.

Claims (5)

The maximum standard intake coefficient (T_SUVmax) and the maximum standard intake coefficient of lymph nodes (N_SUVmax) of primary cancer were obtained from F-18 FDG (fluoro-2-deoxyglucose) PET/CT (positron emission tomography/computed tomography) of gastric cancer patients. A method of providing information for predicting lymph node metastasis of gastric cancer, comprising the step of predicting a possibility of lymph node metastasis by measuring the concentration of CA19-9 and albumin in the patient's blood.
The method of claim 1, wherein the probability of lymph node metastasis is derived using Equations 1 and 2 below.
[Equation 1]
X = ln(P/1-P) = 0.1066 × (T_SUVmax) + 0.3848 × (N_SUVmax) + 0.0025 × (CA19-9; U/mL)-0.6632 × (Albumin; g/dL) + 1.5735

[Equation 2]
Lymph node metastasis probability (P) = exp^X / (1 + exp^X)
(a) The maximum standard intake coefficient (T_SUVmax) and the maximum standard intake coefficient of lymph nodes (N_SUVmax) of primary cancer in F-18 FDG (fluoro-2-deoxyglucose) PET/CT (positron emission tomography/computed tomography) of the gastric cancer patient group. ) By combining the concentrations of CA19-9 and albumin in the blood of gastric cancer patients, deriving a nomogram for predicting lymph node metastasis of gastric cancer;
(b) T_SUVmax and N_SUVmax in F-18 FDG PET/CT of gastric cancer patients whose lymph node metastasis was not confirmed, and CA19-9 and albumin concentrations in blood of gastric cancer patients in the nomogram of step (a) Applying to generate a score associated with prediction of lymph node metastasis; And
(c) providing information that the higher the score is, the higher the probability of lymph node metastasis is.
The method of claim 3, wherein the group of gastric cancer patients in step (a) is a group that performed F-18 FDG PET/CT and hematological examination before surgical resection, providing information for predicting lymph node metastasis of gastric cancer. Way.
The method of claim 3, wherein the score table according to the nomogram in step (a) is the following figure.

Images