Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2022 Aug 29:12:881341.
doi: 10.3389/fonc.2022.881341. eCollection 2022.

DCE-MRI radiomics models predicting the expression of radioresistant-related factors of LRP-1 and survivin in locally advanced rectal cancer

Zhiheng Li  1 , Huizhen Huang  2 , Chuchu Wang  2 , Zhenhua Zhao  1 , Weili Ma  1 , Dandan Wang  1 , Haijia Mao  1 , Fang Liu  3 , Ye Yang  3 , Weihuo Pan  4 , Zengxin Lu  1
Affiliations

DCE-MRI radiomics models predicting the expression of radioresistant-related factors of LRP-1 and survivin in locally advanced rectal cancer

Zhiheng Li et al. Front Oncol. .

Abstract

Objective: Low-density lipoprotein receptor-related protein-1 (LRP-1) and survivin are associated with radiotherapy resistance in patients with locally advanced rectal cancer (LARC). This study aimed to evaluate the value of a radiomics model based on dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) for the preoperative assessment of LRP-1 and survivin expressions in these patients.

Methods: One hundred patients with pathologically confirmed LARC who underwent DCE-MRI before surgery between February 2017 and September 2021 were included in this retrospective study. DCE-MRI perfusion histogram parameters were calculated for the entire lesion using post-processing software (Omni Kinetics, G.E. Healthcare, China), with three quantitative parameter maps. LRP-1 and survivin expressions were assessed by immunohistochemical methods and patients were classified into low- and high-expression groups.

Results: Four radiomics features were selected to construct the LRP-1 discrimination model. The LRP-1 predictive model achieved excellent diagnostic performance, with areas under the receiver operating curve (AUCs) of 0.853 and 0.747 in the training and validation cohorts, respectively. The other four radiomics characteristics were screened to construct the survivin predictive model, with AUCs of 0.780 and 0.800 in the training and validation cohorts, respectively. Decision curve analysis confirmed the clinical usefulness of the radiomics models.

Conclusion: DCE-MRI radiomics models are particularly useful for evaluating LRP-1 and survivin expressions in patients with LARC. Our model has significant potential for the preoperative identification of patients with radiotherapy resistance and can serve as an essential reference for treatment planning.

Keywords: LRP-1; dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI); locally advanced rectal cancer; radiomics models; survivin.

PubMed Disclaimer

Conflict of interest statement

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Figures

Figure 1
Figure 1
Magnetic resonance images of a histopathologically confirmed locally advanced rectal cancer in a 56-year-old woman. (A) Manual region of interest (ROI) placement in a contrast-enhanced sagittal T1-weighted image. (B) Color-coded Ktrans map of the ROI. (C) Color-coded Kep map of the ROI. (D) Color-coded Vp map of the ROI.
Figure 2
Figure 2
Patient inclusion and exclusion details and the patient recruitment flowchart.
Figure 3
Figure 3
Representative immunohistochemical staining of markers. Low-density lipoprotein receptor-related protein-1: (A) low expression, (B) high expression. Survivin: (C) low expression, (D) high expression (magnification: ×10 20).
Figure 4
Figure 4
Radiomics feature selection using sthe least absolute shrinkage and selection operator (LASSO) binary logistic regression model. (A) LASSO coefficient profile, displaying 30 texture features. A coefficient profile plot was produced against the log (lambda) sequence. Each colored line represents the coefficient of an individual feature. (B) Tuning parameter (log lambda) selection in the LASSO model used tenfold cross−validation via 1-SE criteria. Vertical dotted lines were drawn at the selected λ values. (A, C) The error rate curve. (B, D) LASSO coefficient λ graph. Coefficient λ was selected in the LASSO using a tenfold cross-validation. We selected the coefficient λ according to the 1-SE rule.
Figure 5
Figure 5
(A) A comparison of radiomics scores (Rad-scores) between different low-density lipoprotein receptor-related protein-1 expression levels in the training and validation cohorts. (B) A comparison of Rad-scores between different survivin expression levels in the training and validation cohorts.
Figure 6
Figure 6
Receiver operating characteristic curves of the biomarkers for classifying low-density lipoprotein receptor-related protein-1 (A) and survivin (B) expression levels in the training and validation cohorts.
Figure 7
Figure 7
Calibration curves of the radiomics model for predicting low-density lipoprotein receptor-related protein-1 (LRP-1) and survivin expression levels in the training and validation cohorts. (A, B). Calibration curves of the model for LRP-1 in (A) the training and (B) validation cohorts. (C, D) Calibration curves of the model for survivin in (C) the training and (D) validation cohorts. In the calibration plots, the 45° line represents marks the location of the ideal model. The blue line represents the predicted performance of the model, and the red line represents the bias correction in the model.
Figure 8
Figure 8
Decision curve analysis for the model for (A) low-density lipoprotein receptor-related protein-1 and (B) survivin in the test cohorts. The y-axis measures the standardized net benefit. The red curve represents the radiomics model. The gray curve represents the assumption that all patients were treated, and the straight black line at the bottom of the figure represents the assumption that no patient was treated.
Figure 9
Figure 9
Differences in low-density lipoprotein receptor-related protein-1 (LRP-1) and survivin between sensitive and resistant groups. (A) The expression of LRP-1 was higher in the resistant group than in the sensitive group (p = 0.011). (B) The expression of survivin was higher in the resistant group than in the sensitive group (p = 0.006).
See this image and copyright information in PMC

Similar articles

See all similar articles

Cited by

References

    1. Liu Z, Zhang XY, Shi YJ, Wang L, Zhu HT, Tang Z, et al. . Radiomics analysis for evaluation of pathological complete response to neoadjuvant chemoradiotherapy in locally advanced rectal cancer. Clin Cancer Res (2017) 23(23):7253–62. doi: 10.1158/1078-0432.CCR-17-1038 - DOI - PubMed
    1. Wo JY, Anker CJ, Ashman JB, Bhadkamkar NA, Bradfield L, Chang DT, et al. . Radiation therapy for rectal cancer: Executive summary of an astro clinical practice guideline. Pract Radiat Oncol (2021) 11(1):13–25. doi: 10.1016/j.prro.2020.08.004 - DOI - PubMed
    1. Tang Z, Liu L, Liu D, Wu L, Lu K, Zhou N, et al. . Clinical outcomes and safety of different treatment modes for local recurrence of rectal cancer. Cancer Manag Res (2020) 12:12277–86. doi: 10.2147/CMAR.S278427 - DOI - PMC - PubMed
    1. Kirkby CJ, Gala de Pablo J, Tinkler-Hundal E, Wood HM, Evans SD, West NP. Developing a raman spectroscopy-based tool to stratify patient response to pre-operative radiotherapy in rectal cancer. Analyst (2021) 146(2):581–9. doi: 10.1039/d0an01803a - DOI - PubMed
    1. Amatatsu M, Arigami T, Uenosono Y, Yanagita S, Uchikado Y, Kijima Y, et al. . Programmed death-ligand 1 is a promising blood marker for predicting tumor progression and prognosis in patients with gastric cancer. Cancer Sci (2018) 109(3):814–20. doi: 10.1111/cas.13508 - DOI - PMC - PubMed