Elastic modulus-reflected liver lesion stiffness relates to worse prognosis in pancreatic cancer patients with liver metastasis

Background Liver stiffness relates to more advanced tumor status and poor outcomes in primary liver cancer, while its prognostic role in pancreatic cancer with liver metastasis is unclear. Therefore, the current study aimed to explore the correlation of elastic modulus (EM)-reflected liver lesion stiffness with clinical characteristics, tumor markers, and survival among pancreatic cancer patients with liver metastasis. Methods Fifty-four pancreatic cancer patients with liver metastasis were enrolled, and the EM of liver metastasis and peripheral liver tissue was measured by two-dimensional shear wave elastography. Relative EM was calculated as the ratio of EM in liver metastasis to that in peripheral liver tissue, which reflected the relative liver lesion stiffness. Results The median relative EM of liver metastasis was 7.8 (interquartile range: 4.8–10.7) folds. Relative EM of liver metastasis was correlated with primary pancreatic cancer location (P = 0.048), the presence of extra lung metastasis (P = 0.040), liver metastasis ≥ 3 cm (P = 0.007), and the absence of extraskeletal metastasis (P = 0.036); but it was not correlated with tumor markers such as CA199, CA125, or CEA (all P > 0.05). Encouragingly, high relative EM of liver metastasis (cut off by median value) was correlated with poor progression-free survival (PFS) (P = 0.032) but not overall survival (OS) (P = 0.285). Multivariable Cox analysis showed that high relative EM of liver metastasis (hazard ratio (HR) = 1.768, P = 0.048) and multiple metastases (HR = 2.262, P = 0.036) independently predicted decreased PFS, but only abnormal CEA independently forecasted decreased OS (HR = 2.390, P = 0.027). Conclusion Elastic modulus reflected liver lesion stiffness may predict a worse prognosis in pancreatic cancer patients with liver metastasis. Supplementary Information The online version contains supplementary material available at 10.1186/s12957-023-03140-4.


Introduction
Pancreatic cancer is a common cancer with high mortality globally and is projected to be the second-highest cause of cancer-related death in the United States by 2030 [1][2][3].Currently, surgical resection followed by adjuvant chemotherapy provides relatively acceptable outcomes among early-stage pancreatic cancer patients [4,5].Unfortunately, a large proportion of pancreatic cancer patients are diagnosed at an advanced stage with distant metastasis [6].The liver is one of the most common metastatic sites in advanced pancreatic cancer patients, and patients with liver metastasis are not eligible for surgery; their prognosis is extremely dismal, with a 5-year overall survival (OS) rate of approximately 2% [7,8].Therefore, exploring potential indicators to reflect prognosis is crucial to improving the management of these patients.
Liver stiffness is able to reflect the degree of liver fibrosis, whose measurement plays an essential clinical role in reflecting the disease risk and prognosis of liver cancer [9][10][11][12][13].For instance, the liver stiffness parameter is able to predict the occurrence of liver cancer among hepatitis C virus patients [11].Elevated liver stiffness is correlated with deteriorated liver injury and decreased OS in advanced liver cancer patients [14].Moreover, early-stage liver cancer patients who have increased liver stiffness after surgery face a higher risk of relapse [15].In addition, liver stiffness can be used to monitor treatment response among colorectal cancer patients with liver metastasis undergoing transarterial chemoembolization [16].In the preclinical setting, it is reported that low stiffness of liver metastasis is associated with survival in mice with pancreatic cancer liver metastasis [17].Moreover, the stiffness of pancreatic cancer is associated with tumor size in mice [18].Taken together, we deduced that liver stiffness might also play a crucial role in pancreatic cancer patients with liver metastasis, while related information is scarce.
Thus, the current study aimed to explore the correlation of liver stiffness (measured via relative elastic modulus (EM)) with clinical characteristics, tumor markers, and survival among pancreatic cancer patients with liver metastasis.

Patients
Between March 2017 and April 2020, 54 pancreatic cancer patients with liver metastasis were serially included in this study.The inclusion criteria were set as follows: (i) diagnosed with primary pancreatic cancer; (ii) confirmed as liver metastasis by image logical examination, such as contrast-enhanced computerized tomography (CT); and (iii) age over 18 years.The exclusion criteria were as follows: (i) with other primary cancers or hematologic malignancies; (ii) with fatty liver, cirrhosis, or other liver diseases; and (iii) pregnant or lactating female patients.This study was approved by the Ethics Committee of Fudan University Shanghai Cancer Center.

Documents
Clinical features of pancreatic cancer patients were obtained after enrollment, including age, sex, tumor number, number of liver metastases, size of liver metastasis, extrahepatic metastasis, tumor markers, and treatment information.Patients were followed up regularly until April 2021.The median follow-up period was 10 months with a range of 3-26 months.Then, progression-free survival (PFS) and overall survival (OS) were imputed.

Evaluation
Liver stiffness of liver metastasis and peripheral liver tissue was measured using two-dimensional shear wave elastography (2D-SWE) by an Aixplorer US imaging system (SuperSonic Imagine, France) with an SC6-1 (frequency of 1-6 MHz) convex probe.For patients with multiple liver metastases, liver stiffness was measured based on tumors 2-5 cm in diameter and without blood vessels or biliary ducts around the tumor.The procedure was the same as in a previous study and was in line with the European Federation of Societies for Ultrasound in Medicine and Biology (EFSUMB) guidelines [19,20].In brief, the probe was positioned into the patients' intercostal spaces of the right lobe of the liver.The 2D-SWE was carried out followed by a real-time B-mode ultrasound scan to ensure the target area of the liver.The size of the sampling box was 4 cm × 3 cm, and the top edge of the box was located approximately 1 cm under Glisson's capsule of the liver.Patients were asked to hold their breath for approximately 5 s.Then, the images were stabilized, and the elastic modulus (EM) was automatically measured and displayed (a representative image is shown in Fig. S1).Each patient received three times of measurements, and the mean value was calculated based on the three measurements.The measurement of the stiffness of liver metastasis and peripheral liver tissue was conducted by two sonographers who had ultrasound professional qualification certificates and experience in sonography for over 10 years.The EM value was calculated as the mean values of the EM values measured by the two sonographers.After liver stiffness measurement, relative EM was calculated, defined as the ratio of EM in liver metastasis to EM in peripheral liver tissue.The median relative EM was used to classify patients into high relative EM and low relative EM groups.

Statistics
SPSS 26.0 (IBM Corp., USA) and GraphPad Prism 7.01 (GraphPad Software Inc., USA) was employed for data analysis and graph construction, respectively.Data are displayed as the mean ± standard deviation (SD) for normally distributed continuous variables, median (interquartile range (IQR)) for skewed-distributed continuous variables, and count (percentage) for categorized variables.The correlation of EM with clinical characteristics was determined using the Wilcoxon rank sum test.PFS and OS were plotted using Kaplan-Meier (KM) curves and determined by the Breslow test.Independent prognostic factors were analyzed using multivariable Cox proportional hazard regression analysis with the backward stepwise method, and all factors were included in the analysis.A P value < 0.05 was considered significant.

Correlation between relative EM and clinical characteristics
Elevated relative EM was correlated with tumors location in the cauda of the pancreas (P = 0.048), lung metastasis (P = 0.040), and liver metastasis ≥ 3 cm (P = 0.007), while decreased relative EM was correlated with skeletal metastasis (P = 0.036).Moreover, no correlation was found in relative EM with other clinical characteristics (all P > 0.05) (Table 2).

Correlation between relative EM and tumor markers
To further explore the association of relative EM with tumor markers, the relative EM was investigated in patients with abnormal as well as normal CA199, CA125, and CEA.However, no correlation was found in relative EM with these tumor markers (all P > 0.05) (Table 3).

Discussion
Over the decades, evaluation of liver stiffness has been helpful for the management of liver cancer [9-11, 14, 15, 21].In the past, liver stiffness was mainly evaluated by liver biopsy, while this invasive method might trigger the risk of complications [22,23].With the development of medical imaging technology, liver stiffness measured by elastography without invasion is widely applied and is viewed as an indicator of the prognosis of liver cancer.
Regarding the correlation of liver stiffness with clinical characteristics in patients with primary liver cancer and liver metastasis, it has been reported that elevated liver stiffness was correlated with a higher tumor stage and fibrosis stage in primary liver cancer [14,15].However, the association of liver stiffness with clinical characteristics in pancreatic cancer patients with liver metastasis still needed exploration.In the current study, relative EM was positively associated with liver metastasis ≥ 3 cm among pancreatic cancer patients with liver metastasis.The possible explanations are that (1) the stiffness of liver metastatic lesions was associated with collagen concentration, which could accelerate the proliferation  of pancreatic cancer cells at the liver metastatic lesion, consequently leading to elevated liver metastasis size [24,25], and (2) the stiffness of peripheral liver tissue could affect epithelial-mesenchymal transition and phosphatidylinositol-3-kinase/protein kinase B (PI3K/AKT) signaling in liver metastasis, which consequently promoted the growth of liver metastatic lesions [1,9,24,26].Combining the above two reasons, plus the fact that relative EM was calculated by the ratio of EM of liver metastatic lesion to EM of peripheral liver tissues, it could be derived that relative EM was positively associated with liver metastasis ≥ 3 cm.Notably, the current study used relative EM value instead of EM value because relative EM value could eliminate the influence of the stiffness of the liver, thus reducing potential deviation.
In terms of the association of liver stiffness with survival in patients with primary liver cancer and liver metastasis, a previous study has reported that elevated liver stiffness was correlated with a decreased 5-year OS, which is also able to predict long-term recurrence among liver cancer patients [27].Moreover, increased liver stiffness is an independent predictive factor for 1-year recurrence among early liver cancer patients after surgery [15].Higher liver stiffness is associated with an elevated 3-year mortality rate among liver cancer patients after liver transplantation [28].A previous study also revealed that reduced liver metastasis stiffness was associated with better treatment response to bevacizumab in patients with metastatic colorectal cancer [29].However, data on the correlation of liver stiffness with survival in liver metastasis patients are limited.In the current study, elevated relative EM was an independent factor for decreased PFS among pancreatic cancer patients with liver metastasis.The possible explanations are that (1) relative EM was correlated with increased liver metastasis volume (abovementioned), which led to declined PFS among patients; (2) increased relative EM indicated abnormal PI3K/AKT pathway and collagen concentration in liver metastasis, which could not only regulate drug resistance but also modulate antitumor immune microenvironment via enrichment of tumor-associated fibroblasts, consequently affecting PFS among patients [24,26,[30][31][32].Therefore, elevated relative EM was linked with decreased PFS among pancreatic cancer patients with liver metastasis.However, no correlation was found in relative EM with OS among pancreatic cancer patients with liver metastasis.The potential reasons might be that (1) the sample size was relatively small, leading to a low statistical power, and (2) OS could be affected by treatment after disease progression among patients; hence, no correlation was found in relative EM with OS among pancreatic cancer patients with liver metastasis.In the current study, there were five patients with oligo metastasis.These patients all received chemotherapy but not surgical resection of metastasis after considering the potential benefit of surgery and their physical status.
To the best of our knowledge, the current study is the first to explore the clinical role of liver stiffness in pancreatic cancer patients with liver metastasis.Meanwhile, the liver stiffness parameter in the current study was a relative value, which was based on the ratio of EM in liver metastasis to EM in peripheral liver tissue.Thus, relative EM in the current study excluded the potential influence of individual differences on the study findings.However, there were several limitations: (1) the sample size of the current study was relatively small, which could be enlarged in the future; (2) the current study only used one approach to measure EM; thus, more approaches could be applied in future studies; and (3) the clinical value of relative EM in liver metastasis from other cancers, such as colorectal cancer and lung cancer, could be explored.

Conclusion
In conclusion, high relative EM correlates with unfavorable PFS in pancreatic cancer patients with liver metastasis, whose monitoring may help promote the management of these patients.

Fig. 1 Fig. 2
Fig. 1 Correlation of relative EM with survival.Association of relative EM with PFS (A) and OS (B)

Table 2
Correlation of relative EM with clinical characteristics EM Elastic modulus, IQR Interquartile range, NR Not reported, GEM Gemcitabine