Investigation of the mean maximum elastic modulus as a new diagnostic parameter for prostate cancer

With the advancement of technology and medical skills, there are many methods for clinical diagnosis of PCa, including PSA, MRI, conventional TUE, SWE and needle biopsy^8,9,10. Multiple needle sticks can easily cause various complications, such as rectal bleeding, needle tract obstruction or septicemia^11,12,13. TUE has a wide range of applications in the diagnosis of PCa, but has a high rate of misdiagnosis and cannot clearly distinguish between benign and malignant prostate lesions.¹⁴As one of the embedded functions of ultrasound, SWE can accurately measure tissue stiffness¹⁵. SWE measurement of tissue stiffness and elastic modulus values ​​has received great attention in the diagnosis of prostate cancer in recent years, with the characteristics of real-time imaging and high accuracy¹⁶and it is a non-invasive method for obtaining stiffness information¹⁷.

In order to view the prostate as a whole and obtain clear images during the TSWEUI examination, the operator applies pressure to the rectal wall to compress the prostate. As the pressure changes, the elastic modulus value, which represents the stiffness of the prostate tissue, also changes.¹⁸. In order to avoid excessive pressure on the prostate during the investigation operation of this study, the distance of ≥ 2 mm from the inner wall of the rectum near the probe to the prostate capsule in the ultrasound image is maintained as the pressure standard. Since malignant prostate lesions often have diffuse growth and no clear boundaries with the surrounding tissue¹⁹The detection rate of PCa lesions is not high in grayscale and Doppler ultrasound^20,21. Therefore, in practice, it is not easy to interrogate only the abnormal prostate nodes using SWE to differentiate PCa. In the 1980s, transrectal biopsies under transrectal ultrasound guidance became the gold standard for PCa detection²². Using the sextant technique, biopsy cores were taken from six regions of the prostate: the tip, middle and base of each lobe parasagitally in addition to all hypoechoic lesions²³. Imaging usually begins at the base of the bladder and then by rotating the probe a complete image of the prostate can be obtained (as shown in the attached figure in the cited literature showing the maximum cross-section of the prostate).²⁴. In our study, Emax was measured three times in each quadrant of the maximum cross-sectional area. The m-Emax of the maximum cross-sectional area of ​​the prostate was determined by calculating the mean of 12 Emax measured in the four quadrants.

In our study, the AUC for m-Emax was 0.754 (95% CI 0.642–0.867), indicating high accuracy (Fig. 2). The cutoff value was 64.820 kPa. The risk of prostate cancer increased when m-Emax was ≥ 64.820 kPa. The cutoff value for prostate elastic modulus was different from previous studies, such as Sun et al.²⁵JI et al.²⁶Fu et al.²⁷. This difference may be due to the following reasons: First, the elastic modulus parameter to represent the stiffness of prostate tissue was different from that in other studies. In the study of Sun et al., the cutoff value of Emax = (60.45 kPa for extracapsular expansion and 81.55 kPa for seminal vesicle invasion). A region of interest was placed in the paramedial and lateral regions (5 mm in diameter) at the level of the base, gland center, and apex of the peripheral zone of the prostate to systematically measure the stiffness (Emax) of the whole prostate. In the study of Ji et al., the cutoff value of Emax = 128.48 kPa. In the transverse plane of the prostate, the sampling box was placed on both sides of the base, center, and apex of the prostate, respectively. The elastic modulus value (Emax, et al.) was the average of the measurement results in these planes. In Fu et al. study, cutoff value of Emax = 42 kPa. The elasticity value was measured in the area prepared for systematic biopsy, and suspicious lesions were detected by B-mode or SWE. Second, the quality control of the detection process and calculation method of the elastic modulus value was different from other studies. There is no quality control for the pressure standard in these three literatures.

Because the stress at an elastic point of the model depends not only on the gradient of the deformation but also on the orientation, connection and distribution of its components. The fiber network exhibits a nonlinear stress-strain relationship due to complex interactions that vary from point to point. The progressive increase in deformation will destroy the fibers, which will begin to align in the direction of the load, increasing the stiffness; the stress-strain relationship is approximately linear. Pathologies such as tumors and fibrosis involve changes in consistency, because the structural properties of these abnormalities imply a stiffer region reflecting histological differences in the microstructure of the tissue²⁸. Digital rectal examination revealed that prostate stiffness was higher in PCa patients than in normal subjects. In our study, m-Emax was a strong independent predictor of PCa (OR = 9.95, P= 0.001) (Fig. 3). The RCS model (Fig. 4) showed a linear dose-response relationship between m-Emax and PCa risk (P_{in total}= 0.007 (P_{in total}< 0.05) and P_nonlinear= 0.909 (P_nonlinear> 0.05)). Therefore, m-Emax can be used as an innovative parameter of elastic modulus to represent the stiffness of prostate tissue in TSWEUI. By standardizing the detection process and calculation method of elastic modulus, the data are relatively stable and reliable.

Univariate analysis of the baseline characteristics of the 209 patients showed that PSA (P< 0.001), PV (P< 0.001), platelet count (P= 0.025) and PLR (P= 0.003) were significantly different between the benign group and the malignancy group. Based on the results of the first univariate analysis, a second univariate analysis was performed on the baseline characteristics of 75 patients who completed the TSWEUI, including m-Emax and clinical characteristic factors that correlate with PCa, such as PSA, PV, platelet count, and PLR. The results showed that PSA (P< 0.001), PV (P= 0.008) and m-Emax (P< 0.001) were statistically correlated with PCa. The AUC for PSA was 0.729 (95% CI 0.611–0.846) (Supplementary Figure 1). The cutoff was 7.625 ng/ml. The risk of PCa increased with PSA ≥ 7.625 ng/ml. The specificity of serum PSA tests is relatively low²⁹. The cutoff point for abnormal PSA levels is above 4 ng/ml, indicating the possible presence of PCa³⁰The risk of PCa increases sharply when the PSA level reaches 10 ng/ml.³¹If the PSA level is 4 ≤ ≤ 10 ng/ml, a needle biopsy should be performed to diagnose PCa.³². Given the complications of prostate biopsy, TSWEUI could ideally benefit patients with PSA between 4 and 10 ng/mL, as this group is at the highest risk for undiagnosed malignant cancer. Subjects included in this study were patients with PSA of 4 ng/mL or higher. The AUC for PV was 0.680 (95% CI 0.554–0.805) (Supplementary Figure 2). The cutoff value was 36.875 mL. PV differed significantly between the malignant and benign groups (35.11 vs. 51.3, P= 0.008) (Table 1). The PV in the malignancy group was smaller than in the benignity group. PV is an important factor affecting PSA values³³PV is associated with the incidence of PCa³⁴and with PSA values ​​in the range of 2.0–9.0 ng/ml, a smaller prostate volume is a predictive factor for cancer risk³⁵.

The multivariate logistic regression analysis showed PSA (OR = 4.64, P< 0.05) and PV (OR = 0.09, P< 0.001) as independent predictors of PCa (Fig. 3). Based on the results of multivariate logistic regression analysis, a prediction model including the independent predictors PSA, PV and m-Emax in the form of the nomogram was developed (Supplementary Fig. 3). The specific scoring table of the nomogram is clearly shown in Supplementary Table 2 to facilitate clinical application. This study demonstrated that the nomogram improved the prediction accuracy (the AUC for the nomogram increased to 0.868, 95% CI 0.789–0.948) (Fig. 5). The calibration curve of the nomogram showed a good agreement between the predicted and actual probability (Supplementary Fig. 4).

In our study, a cancer risk prediction model for clinical diagnosis of PCa was developed by using SWE technology in combination with relevant clinical characteristic factors. Due to its high prediction accuracy for patients suspected of PCa, it can reduce the needle biopsy rate in the diagnosis of PCa.

MRI has been used since the 1980s for non-invasive examination of the prostate and surrounding structures.⁷The PI-RADS can provide scoring categories that summarize the level of suspicion or risk of clinically significant prostate cancer that can be used to select patients for biopsies and treatment.⁷. The AUC of 63 patients did not differ statistically between m-Emax (AUC = 0.717) and MRI (AUC = 0.787) in the ROC curve analysis (P= 0.361) (Fig. 6), indicating that there is no difference in discrimination for PCa. MRI is too expensive and time-consuming, and some contraindications (claustrophobia, pacemaker, etc.)³⁶Therefore, this study found that TSWEUI is more cost-effective than MRI in diagnosing PCa.

Study restrictions

The limitation of this study lies in the lack of comprehensive analysis and research on the selection of the elastic modulus value indicator, which represents the stiffness of the prostate tissue, and the influencing factors on the measurement and calculation of the indicator. There is no in-depth study of the effect of pressure changes on the elastic modulus value of the prostate. Targeted work is needed. In this study, a single operator can maintain the stability of the operation, but may produce systematic errors. Due to the small sample size of this study, increasing the sample size could provide further evidence of the relationship between TSWEUI and PCa in the next step research.