Methods for evaluating the prognosis of pancreatic cancer

Abstract

The objective is to provide a novel method for evaluating the prognosis of pancreatic cancer. [Solution] A method for evaluating the prognosis of pancreatic cancer, comprising the steps of: measuring the extracellular volume fraction of the non-cancerous pancreatic parenchyma in a CT image of a pancreatic cancer patient; and correlating the measured value of the extracellular volume fraction with the prognosis of pancreatic cancer in the patient, wherein the step of correlating the measured value of the extracellular volume fraction with the prognosis of pancreatic cancer in the patient includes the step of comparing the measured value of the extracellular volume fraction with a cutoff value of the extracellular volume fraction determined by ROC curve analysis.

Description

Technical Field

[0001] The present invention relates to a method for evaluating the prognosis of pancreatic cancer.

Background Art

[0002] Pancreatic cancer is a disease with a very poor prognosis, and early detection and accurate prognosis prediction are important. Conventionally, clinical pathological parameters and imaging findings have been used for prognosis prediction of pancreatic cancer. However, these methods have limitations, and a more accurate prediction biomarker has been demanded.

[0003] By the way, in recent years, the extracellular volume fraction (fECV) obtained from CT images has attracted attention as an index reflecting tissue fibrosis and stromal state. For example, the present inventors reported in Non-Patent Document 1 that fECV is related to pancreatic fibrosis and further related to the presence of pancreatic cancer. In addition, Non-Patent Document 2 reported that a scoring system including fECV is useful for predicting the progression-free survival period of patients with locally advanced pancreatic cancer. Furthermore, Non-Patent Document 3 suggested the possibility that fECV is useful for predicting the response to chemoradiotherapy for pancreatic cancer.

[0004] However, in these prior studies, the conventional ROI (region of interest)-based method was used for measuring fECV, and there were problems in the reproducibility and objectivity of the measurement. In addition, there are few reports on prognosis prediction studies focusing on the fECV of non-cancerous pancreatic parenchyma.

Prior Art Documents

Non-Patent Documents

[0005]

Non-Patent Document 1

Non-Patent Document 2

Non-Patent Document 3

[0006] This invention has been made in view of the problems of the prior art described above, and aims to provide a novel method for evaluating the prognosis of pancreatic cancer. [Means for solving the problem]

[0007] As a result of diligent research to solve the above problems, the inventors of this invention discovered that the prognosis of pancreatic cancer can be predicted with high accuracy by using the extracellular volume fraction (fECV) of non-cancerous pancreatic parenchyma as an indicator, and thus completed the present invention.

[0008] In other words, one aspect of the present invention relates to the following: [1] A method for evaluating the prognosis of pancreatic cancer, A method characterized by using the extracellular volume fraction of the non-cancerous pancreatic parenchyma as an indicator in CT images of pancreatic cancer patients. [2] A method for evaluating the prognosis of pancreatic cancer, (1) A step of measuring the extracellular volume fraction of the non-cancerous pancreatic parenchyma using CT images of a pancreatic cancer patient, and (2) A step of correlating the measured value of the extracellular volume fraction with the prognosis of pancreatic cancer in the patient. Methods that include... [3] The method according to [1] or [2], wherein the step of correlating the measured value of the extracellular volume fraction with the prognosis of pancreatic cancer in the patient includes the step of comparing the measured value of the extracellular volume fraction with a reference value of the extracellular volume fraction. [4] The method according to [3], wherein the reference value of the extracellular volume fraction is the cutoff value of the extracellular volume fraction. [5] The method according to [4], wherein the cutoff value is determined by ROC curve analysis. [6] The method according to any one of [2] to [5], wherein the CT image in step (1) is a preoperative CT image of a pancreatic cancer patient. [7] The method according to any one of [1] to [6], wherein the prognosis is overall survival and / or recurrence-free survival. [8] The method according to [4] or [5], wherein the cutoff value is the following: (a) If the prognosis is overall survival, the cutoff value is 42% or more and 45% or less, and / or (b) If the prognosis is the recurrence-free survival period, the cutoff value is 38% or more and less than 42%. [9] The method according to any one of [1] to [8], wherein the assessment of prognosis is a method that assists in determining overall survival and / or recurrence-free survival. [Effects of the Invention]

[0009] According to the present invention, it becomes possible to predict the prognosis of pancreatic cancer in a non-invasive and highly accurate manner. [Modes for carrying out the invention]

[0010] The following describes in detail some examples of embodiments of the present invention, but the present invention is not limited to these.

[0011] (1. Outline of the present invention) While diligently conducting research with the aim of providing a novel method for evaluating the prognosis of pancreatic cancer, the inventors discovered that the prognosis of pancreatic cancer can be predicted with high accuracy by using the extracellular volume fraction (fECV) of non-cancerous pancreatic parenchyma as an indicator. Specifically, they found the following: Previous studies have primarily focused on fECV in the tumor site, but this invention focuses on fECV in the non-cancerous pancreatic parenchyma. This makes it possible to predict prognosis that reflects the microenvironment surrounding the tumor and the overall state of the pancreas. Statistical analysis of the association between fECV in non-cancerous pancreatic parenchyma and overall survival (OS) and recurrence-free survival (RFS) revealed that fECV is an independent prognostic predictor. · Successfully determined the cut-off value of fECV that is optimal for prognosis prediction statistically using a large amount of patient data. Thus, based on a new idea that was not in the prior art, it is surprising that the prognosis of pancreatic cancer could be predicted with high accuracy by using the extracellular volume fraction (fECV) of the non-cancerous pancreatic parenchyma as an index.

[0012] The prognosis evaluation method of the present invention has the following advantages. · Non-invasive prognosis prediction: By using the fECV obtained from a normal CT examination, prognosis prediction is possible without additional invasive examinations. · Contribution to personalized medicine: Prognosis prediction based on fECV enables risk stratification for each patient and can contribute to the formulation of personalized treatment strategies. · Promotion of early intervention: Early identification of the high-risk group may enable more aggressive treatment intervention and close follow-up observation. · Economic effect: By making an early treatment decision and avoiding unnecessary treatment, medical costs can be reduced.

[0013] (2. Definitions) In this specification, "pancreatic cancer" is intended to mean pancreatic cancer in the clinically general sense, and means a malignant tumor derived from epithelium that has occurred in the pancreas. Also, the degree of progression of pancreatic cancer is not particularly limited, and includes pancreatic cancer at all stages.

[0014] In this specification, "prognosis" means a medical outlook for future disease conditions. In particular, in the present invention, it intends the state of a patient after surgery, treatment effect, survival probability, etc.

[0015] In this specification, "non-cancerous pancreatic parenchyma" means pancreatic tissue located on the opposite side of the tumor and separated by an estimated cutting line.

[0016] As used herein, the "extracellular volume fraction (fECV)" is an index calculated from contrast-enhanced CT images, which is obtained by correcting the changes in the contrast effects of pancreatic tissue and the aortic lumen in the non-contrast phase and the equilibrium phase with the patient's hematocrit value, and represents a value reflecting the proportion of the extracellular space of the tissue.

[0017] As used herein, the "overall survival (OS)" means the period from the start of treatment until the subject dies.

[0018] As used herein, the "recurrence-free survival (RFS)" means the period during which the disease of the treatment target survives without recurrence after the start of treatment.

[0019] (3. Method for evaluating the prognosis of pancreatic cancer) In one embodiment of the present invention, there is provided a method for evaluating the prognosis of pancreatic cancer, which comprises using, as an index, the extracellular volume fraction of the non-cancerous pancreatic parenchyma in the CT image of a pancreatic cancer patient.

[0020] Also, in one embodiment of the present invention, there is provided a method for evaluating the prognosis of pancreatic cancer, which comprises: (1) a step of measuring the extracellular volume fraction of the non-cancerous pancreatic parenchyma using the CT image of a pancreatic cancer patient, and (2) a step of correlating the measured value of the extracellular volume fraction with the prognosis of pancreatic cancer in the patient.

[0021] <Step (1)> In step (1) of the present invention, the extracellular volume fraction of the non-cancerous pancreatic parenchyma is measured using the CT image of a pancreatic cancer patient. By using the extracellular volume fraction of the non-cancerous pancreatic parenchyma as an index, it becomes possible to predict the prognosis of pancreatic cancer with high accuracy.

[0022] The CT image of the pancreatic cancer patient is a preoperative CT image. Specifically, it is a CT image taken 5 to <281> days before surgery, and preferably, it can be a CT image taken 69 to <115> days before surgery.

[0023] The method for measuring the extracellular volume fraction of non-cancerous pancreatic parenchyma is not particularly limited, but for example, the method described in the examples is used. Briefly, fECV measurement is performed using an image workstation (SYNAPSE VINCENT; Fujifilm Medical Co., Ltd.). Subtraction images obtained from unenhanced and equilibrium phase images are automatically segmented into non-cancerous pancreatic parenchyma and pancreatic tumors using the workstation's application. Segmentation excludes local lesions, major vascular branches, and ducts, and includes only the non-cancerous pancreatic parenchyma contralateral to the tumor, demarcated by a presumptive cutting line. Since pancreatic fat infiltration can affect fECV, thresholding is performed by setting a low upper threshold to mitigate the potential impact of fat infiltration on fECV measurement. The mean intensification (Hounsfield units) of non-cancerous pancreatic parenchyma and pancreatic tumors (ΔPancreas and ΔTumor) is measured. To measure the intensification of the aorta (ΔAorta) and portal vein (ΔPortal), the region of interest is drawn as large as possible within the lumen of the aorta and portal vein, excluding the vascular walls. Significant differences in ΔAorta and ΔPortal (±10HU or greater) are considered abnormal values ​​and should be excluded. fECV for non-cancerous pancreatic parenchyma and pancreatic tumors is calculated using the following formula: fECV[%]=(100-hematocrit[%])*ΔPancreas(or ΔTumor) / ΔAorta

[0024] In one embodiment of the present invention, the process may include a step of acquiring a CT image of a pancreatic cancer patient prior to step (1). The method of acquiring the CT image is not particularly limited, and any commonly used method can be used. The acquisition of the CT image may be performed, for example, as described in the embodiment.

[0025] The pancreatic cancer targeted by the evaluation method of the present invention is surgically treatable pancreatic cancer.

[0026] <Process (2)> Step (2) of the present invention correlates the measured value of the extracellular volume fraction obtained in step (1) with the prognosis of pancreatic cancer in the patient.

[0027] In one embodiment of the present invention, the prognosis evaluated by the method of the present invention may be overall survival or recurrence-free survival. The prognosis may be evaluated by overall survival or recurrence-free survival individually, or by a combination of both.

[0028] As a method for correlating the measurement value of extracellular volume fraction with the prognosis of pancreatic cancer in the patient, it is preferable to compare the measurement value of extracellular volume fraction with a reference value for extracellular volume fraction.

[0029] In one embodiment of the present invention, the reference value for the extracellular volume fraction is preferably the cutoff value for the extracellular volume fraction. In one embodiment of the present invention, the cutoff value may be determined based on ROC curve analysis. Since fECV reflects the extracellular lumen and is affected by factors such as fibrosis, the higher the fECV in non-cancerous pancreatic parenchyma, the greater the degree of fibrosis in the patient tends to be. Therefore, the higher the fECV in non-cancerous pancreatic parenchyma, the worse the patient's prognosis can be determined to be.

[0030] In the evaluation method of the present invention, the cutoff value is calculated, for example, by the method described in the examples. Briefly, to identify the optimal cutoff value for non-cancerous pancreatic parenchymal fECV in predicting overall survival (OS) and recurrence-free survival (RFS), a time-dependent receiver operating characteristic (ROC) curve is created, and the cutoff value is determined by maximizing the Youden index. Furthermore, overall survival (OS) and recurrence-free survival (RFS) are compared between patients with fECV values ​​above the cutoff value derived from the optimal ROC and patients with fECV values ​​below it using a Kaplan-Meier analysis with a log-rank test.

[0031] The instruments used for statistical analysis are not particularly limited, but for example, EZR version 1.50 (Saitama Medical Center, Jichi Medical University) could be used.

[0032] In one embodiment of the present invention, the cutoff value for overall survival is, for example, a value of 42% or more and 45% or less, preferably a value of 43% or more and 44% or less, and more preferably 43.6%.

[0033] In one embodiment of the present invention, the cutoff value for recurrence-free survival is, for example, a value of 38% or more and less than 42%, preferably a value of 39% or more and 41%, and more preferably 40.3%.

[0034] In one embodiment of the present invention, the process may include, after step (2), a step of determining a treatment plan based on the prognosis evaluation obtained in step (2).

[0035] In one embodiment of the present invention, the evaluation of prognosis may be a method to assist in determining overall survival and / or recurrence-free survival.

[0036] <Other> In the evaluation method of the present invention, fECV can be used in combination with other factors. In the examples, univariate and multivariate analyses using a Cox proportional hazards model were performed on factors such as age, sex, neoadjuvant therapy and adjuvant therapy regimens, serum levels of CEA, CA 19-9, and DUPAN-2, tumor location, size, TNM stage, and histological type. As a result, fECV, along with adjuvant therapy and histological type (poorly differentiated adenocarcinoma), were identified as significant predictors of overall survival (OS). By combining these factors with fECV, it becomes possible to predict overall survival with greater accuracy. [Examples]

[0037] The present invention will be described in more detail below using examples, but this will not limit the scope of the invention. References made throughout this specification are incorporated entirely by reference.

[0038] (1. Method) The study described below was conducted with the approval of the Institutional Review Board of Osaka University Hospital, National University Corporation (hereinafter referred to as "this hospital"). The need for informed consent was waived.

[0039] <Patient> A total of 541 consecutive patients who underwent pancreatectomy at this hospital from January 2014 to February 2022 were evaluated. Patients who were (1) pathologically diagnosed with pancreatic ductal adenocarcinoma, (2) underwent contrast-enhanced CT (including non-enhanced and equilibrium phases) before surgery, and (3) did not receive preoperative treatment before CT were included in the analysis. As a result, 235 patients were diagnosed with pancreatic cancer (PDAC), but 134 patients who did not undergo non-enhanced and equilibrium contrast-enhanced CT before surgery and 11 patients who received preoperative treatment before CT were excluded. Among the remaining 90 patients, 7 patients with CT images showing pancreatic atrophy with extensive pancreatic duct dilation and 9 patients with evidence of acute pancreatitis on imaging diagnosis or clinical examination were excluded. Three patients with missing survival information were excluded. Finally, 71 patients (35 males and 36 females; mean age 70.3 years; 95% confidence interval, 68.1 - 72.6; SD, 9.8; age range, 45 - 89 years) were evaluated. Among the 71 patients, 34 underwent pancreaticoduodenectomy and 37 underwent distal pancreatectomy. The median interval from pancreatectomy to CT imaging was 93 days (interquartile range (IQR), 69 - 115 range, 5 - 281).

[0040] <CT Imaging> CT scans were performed using a 320-channel (Aquilion One Genesis Edition; Canon Medical Systems), a 160-channel (Aquilion Precision; Canon Medical Systems), or a 64-channel (Discovery CT 750 HD or Discovery CT 750 HD Freedom Edition; GE Healthcare) CT scanner. The scan parameters were as follows: collimation: 0.625 × 64 or 0.5 × 80 mm, pitch: 1.375 or 0.813 mm / revolution; rotation time: 0.4 or 0.5 seconds / revolution; exposure parameters: 120 kV; automatic exposure control; noise index standard deviation: 13.8 or 15; field of view: 345 mm. Image reconstruction was performed using FC03 or the GE standard kernel with a slice thickness of 5 mm. Patient radiation dose was evaluated using the CTDIvol index, with a median of 8.66 mGy. The early arterial phase was obtained 8 seconds after the pre-contrast scan and after achieving 100-HU attenuation of the descending aorta (using the bolus tracking method). The late arterial phase was obtained 7 seconds after the early arterial phase or 20 seconds after 100-HU attenuation of the descending aorta (using the bolus tracking method). The portal venous and equilibrium phases were acquired 30 seconds after the late arterial phase and 120 seconds after the portal venous phase. The contrast agent was administered intravenously at an iodine concentration of 600 mgI / kg body weight with a 30-second fixation. Unweighted and dynamic contrast-enhanced images were obtained, but only unweighted and equilibrium phase images were used. Subtraction images between the unweighted and equilibrium phase images were generated using nonlinear non-rigid registration processing (SURE SUBTRACTION Iodine Mapping (SSIM), Canon Medical Systems).

[0041] <Quantitative evaluation of images> fECV measurements were performed using an image workstation (SYNAPSE VINCENT; Fujifilm Medical Co., Ltd.). Subtraction images obtained from unenhanced and equilibrium phase images were automatically segmented into non-cancerous pancreatic parenchyma and pancreatic tumors using the workstation's application. Segmentation excluded local lesions, major vascular branches, and ducts, and included only the non-cancerous pancreatic parenchyma contralateral to the tumor, demarcated by a presumptive section line. Because pancreatic fat infiltration can affect fECV, thresholding was performed with a low upper threshold to remove pancreatic fat infiltration as much as possible while preserving pancreatic parenchymal tissue, mitigating the potential impact of fat infiltration on fECV measurements. The mean enhancement (in Hounsfield units) for non-cancerous pancreatic parenchyma and pancreatic tumors (ΔPancreas and ΔTumor) was measured. To measure the enhancement of the aorta (ΔAorta) and portal vein (ΔPortal), the region of interest was drawn as large as possible within the lumen of the aorta and portal vein, excluding the vessel walls. ΔAorta and ΔPortal (±10HU or greater) showing significant differences were considered abnormal values ​​and excluded. fECV for non-cancerous pancreatic parenchyma and pancreatic tumors was calculated using the following formula: fECV[%]=(100-hematocrit[%])*ΔPancreas(or ΔTumor) / ΔAorta The hematocrit value closest to the CT scan date was used for the analysis. On average, these measurements were taken 2.87 days after the CT scan (95% CI: 1.74–4.0 days; SD: 4.81 days; range: -14–17 days).

[0042] <Statistical analysis> The Shapiro-Wilk test was performed to determine normality. Univariate and multivariate analyses using Cox proportional hazards models were performed to examine the effects of various factors on overall survival (OS) and recurrence-free survival (RFS). These factors included age, sex, neoadjuvant and adjuvant therapy regimens, serum levels of CEA, CA 19-9, and DUPAN-2, as well as tumor characteristics such as tumor location, size, TNM stage, histological type, and fECV of both tumor and non-cancerous pancreatic parenchyma. For factors with three or more levels, the Wald test was performed to calculate the overall p-value for each factor. To identify optimal cutoff values ​​for tumor fECV and non-cancerous pancreatic parenchymal fECV in predicting RFS and OS, time-dependent receiver operating characteristic (ROC) curves were constructed, and the cutoff values ​​were determined by maximizing the Youden index. For both tumor and non-cancerous pancreatic parenchymal fECV, Kaplan-Meier analysis using the log-rank test was employed to compare RFS and OS between patients with fECV values ​​above and below the optimal ROC-derived cutoff values. A p-value less than 0.05 was considered statistically significant. All statistical analyses were performed using EZR version 1.50 (Saitama Medical Center, Jichi Medical University).

[0043] (2. Evaluation of the prognostic significance of factors including fECV of the tumor and non-cancerous pancreatic parenchyma) This study performed univariate and multivariate Cox hazard analyses to evaluate the prognostic significance of various factors, including fECV (facial extracellular volume fraction) of tumor and non-cancerous pancreatic parenchyma, on overall survival (OS) and recurrence-free survival (RFS) in pancreatic cancer patients. fECV of non-cancerous pancreatic parenchyma was found to be a significant independent predictor of overall survival in both univariate (hazard ratio: 1.05, 95% CI: 1.01-1.09, p=0.018) and multivariate (hazard ratio: 1.05, 95% CI: 1.01-1.10, p=0.019) analyses. Higher fECV values ​​in non-cancerous pancreatic parenchyma were associated with an increased risk of death. This finding suggests that the microenvironment of non-cancerous pancreatic parenchyma may be involved in the progression and prognosis of PDAC.

[0044] On the other hand, tumor fECV was shown not to be a significant predictor of OS or RFS. This finding was in contrast to several previous studies that reported the association between tumor fECV and patient outcomes in PDAC (Fukukura Y et al., Pancreas, 2019, 48:1360-1366; Fukukura Y et al., Eur Radiol, 2020, 30:1679-1689; Fukukura Y et al., European Radiology, 2019, 29:353-361; Wang ZJ et al., AJR Am J Roentgenol, 2020, 215:610-616). Adjuvant therapy was identified as a significant predictor of overall survival in both univariate (p=0.032) and multivariate (p<0.001) analyses. Patients who received gemcitabine (GEM) as adjuvant therapy had a significantly higher risk of death compared to patients who did not receive adjuvant therapy (multivariate hazard ratio: 12.08, 95% CI: 3.11-46.88, p<0.001). However, there was no significant difference in mortality between patients who received S-1 as adjuvant therapy and those who did not (multivariate hazard ratio: 1.19, 95% CI: 0.55-2.60, p=0.656). Poorly differentiated adenocarcinoma was significantly associated with an increased risk of death compared to well-differentiated adenocarcinoma (multivariate hazard ratio: 2.98, 95% CI: 1.03-8.67, p=0.045). TNM stage was a significant predictor of recurrence-free survival in univariate analysis (p=0.06). In the multivariate analysis, only Stage IB showed a significantly higher relapse risk compared to Stage IA (multivariate hazard ratio: 2.94, 95% CI: 1.36-6.39, p=0.006). Stages II and III did not show a significant difference in relapse risk compared to Stage IA in the multivariate analysis. Patients who received neoadjuvant therapy, gemcitabine and S-1 combination therapy (GS), had a significantly lower relapse risk compared to patients who did not receive neoadjuvant therapy (multivariate hazard ratio: 0.33, 95% CI: 0.13-0.83, p=0.019).Other factors, such as age, sex, tumor site, and serum levels of CEA, CA 19-9, and DUPAN-2, did not show a significant association with overall survival or recurrence-free survival in multivariate analysis.

[0045] Time-dependent receiver operating characteristic (ROC) curve analysis was performed to determine the optimal cutoff values ​​for non-cancerous pancreatic parenchymal fECV and tumor fECV to predict OS and RFS. For non-cancerous pancreatic parenchymal fECV, the area under the ROC curve (AUC) was 0.741 for OS and 0.604 for RFS. The optimal cutoff values ​​were 43.65% for OS (sensitivity: 0.677, specificity: 0.758) and 40.32% for RFS (sensitivity: 0.818, specificity: 0.517). For tumor fECV, the AUC was 0.599 for OS and 0.568 for RFS, and the optimal cutoff values ​​were 47.45% for OS (sensitivity: 0.531, specificity: 0.711) and 38.00% for RFS (sensitivity: 0.864, specificity: 0.367).

[0046] Kaplan-Meier analysis using the log-rank test showed that when fECV values ​​in non-cancerous pancreatic parenchyma were below the optimal receiver operational characteristics (ROC) cutoff values ​​(43.65% for OS and 40.32% for RFS), the median OS and RFS were significantly shorter compared to the group where fECV values ​​were below the cutoff values ​​(OS: 584 days vs. 1209 days, p=0.000131; RFS: 267 days vs. 653 days, p=0.0356). On the other hand, there was no statistically significant difference in OS (p=0.1) or RFS (p=0.8) between the group where tumor fECV values ​​were above the optimal ROC cutoff values ​​(OS: 47.45%, RFS: 38.00%) and the group where they were below. These results highlight the potential usefulness of fECV in non-cancerous pancreatic parenchyma as a prognostic marker for risk stratification and individualized treatment strategies for PDAC patients. [Industrial applicability]

[0047] This invention is extremely useful in predicting the prognosis of pancreatic cancer because it allows for non-invasive and highly accurate prognosis assessment.

Claims

1. A method for evaluating the prognosis of pancreatic cancer, A method characterized by using the extracellular volume fraction of the non-cancerous pancreatic parenchyma as an indicator in CT images of pancreatic cancer patients.

2. A method for evaluating the prognosis of pancreatic cancer, (1) A step of measuring the extracellular volume fraction of non-cancerous pancreatic parenchyma using CT images of pancreatic cancer patients, and (2) A step of correlating the measured value of the extracellular volume fraction with the prognosis of pancreatic cancer in the patient. Methods that include...

3. The method according to claim 1 or 2, wherein the step of correlating the measured value of the extracellular volume fraction with the prognosis of pancreatic cancer in the patient includes the step of comparing the measured value of the extracellular volume fraction with a reference value of the extracellular volume fraction.

4. The method according to claim 3, wherein the reference value of the extracellular volume fraction is the cutoff value of the extracellular volume fraction.

5. The method according to claim 4, wherein the cutoff value is determined by ROC curve analysis.

6. The method according to claim 2, wherein the CT image in step (1) is a preoperative CT image of a pancreatic cancer patient.

7. The method according to claim 1 or 2, wherein the prognosis is overall survival and / or recurrence-free survival.

8. The method according to claim 5, wherein the cutoff value is the following value: (a) If the prognosis is overall survival, the cutoff value is 42% or more and 45% or less, and / or (b) If the prognosis is the recurrence-free survival period, the cutoff value is 38% or more and less than 42%.

9. The method according to claim 1 or 2, wherein the prognostic evaluation is a method for assisting in the determination of overall survival and / or recurrence-free survival.